Diagnostic HemoglobinopathiesLaboratory Methods and Case Studies

Zia Uddin, PhDSt. John Macomb-Oakland Hospital

Warren, Michigan

Second EditionAugust 2015

Editorial Board       

Diane M. Maennle, MD                                    Chairperson         Kenneth F. Tucker, MD                                    Member         Rita Ellerbrook, PhD                                         Member         Piero C. Giordano, PhD                                    Member         Kimberly R. Russell, MT (ASCP), MBA           Member

I

Contributors and Reviewers

Antonio Amato, MDDirectorCentro Studi Microcitemie Di RomaA.N.M.I. ONLUSVia Galla Placidia 28/3000159 Rome, Rome

Italy

Erol Omer Atalay, MDProfessor, Medical FacultyPamukkale UniversityKinikli, DenizliTurkey

Celeste Bento, PhDLaboratorio de Anemias Congenitas e Hematologia MolecularServico de Hematologia, Hospital PediatricoCentro Hospitalar e Universitario de CoimbraPortugal

Aigars Brants, PhDScientific Affairs ManagerSebia, Inc400-1705 Corporate DriveNorCross, GA 30093USA

Thomas E. Burgess, PhDTechnical Director, Quest DiagnosticsTucker, GeorgiaUSA

Shahina Daar, MD, PhDAssociate ProfessorDepartment of HematologySultan Qaboos University, MuscatSultanate of Oman

II

Angie Duong, MD            Assistant Professor, Hematopathology            Department of Pathology and            Laboratory Medicine            Medical University-South Carolina            Charleston, South Carolina            USA

Rita Ellerbrook, PhDTechnical Director EmeritusHelena Laboratories, USA1530 Lindberg DriveBeaumont, TX 77707 USA

Eitan Fibach, MDProfessor, Department of HematologyHadassah-Hebrew University Medical CenterEin-Kerem, JerusalemIsrael

Bernard G. Forget, MDProfessor Emeritus of Internal MedicineYale School of MedicineNew Haven, CT 06520USA

Piero C. Giordano, PhDHemoglobinopathies LaboratoryHuman and Clinical Genetics DepartmentLeiden University Medical CenterThe Netherlands

Dina N. Greene, PhDScientific Director, ChemistryRegional Laboratories, Northern CaliforniaThe Permanente Medical GroupBerkeley, CA 94710USA

III

Rosline Hassan, PhDProfessor of HematologySchool of Medical SciencesUniversity Sains Malaysia, KelanranMalaysia

David Hockings, PhDFormerly with Isolab, USA andPerkinElmer Corporation, USARaleigh-Durham, North CarolinaUSA

Prasad Rao Koduri, MDDivision of Hematology-OncologyHektoen Institute of Medical ResearchChicago, Illinois 60612USA

John Lazarchick, M.D. Professor, Pathology and Laboratory Medicine Professor, Medicine Director, Hematopathology Director, Hematopathology Fellowship Program Vice Chair, Clinical Pathology Medical University-South Carolina Charleston, SC

Elaine Lyon, PhDAssociate Professor of PathologyUniversity of Utah School of MedicineMedical Director, Molecular GeneticsARUP Laboratories, Salt Lake City, UTUSA

Bushra Moiz, PhDAssociate ProfessorDepartment of Pathology and MicrobiologyThe Agha Khan University Hospital, KarachiPakistan

IV

Herbert L. Muncie, MDProfessor, Department of Family MedicineSchool of Medicine, Louisiana State University1542 Tulane AveNew Orleans, LA 70112USA

Gul M. Mustafa, PhDPost-Doctorate FellowDepartment of PathologyThe University of Texas Medical BranchGalveston, TX 77555USA

Diane M. Maennle, MDAssociate PathologistDepartment of PathologySt. John Macomb-Oakland Hospital

. Warren, MI 48093USA

Jayson Miedema, MDPost-Doctorate FellowDepartment of Pathology and Laboratory MedicineUniversity of North CarolinaChapel Hill, North CarolinaUSA

Christopher R. McCudden, PhDAssistant Professor, Department of Pathologyand Laboratory Medicine, University of OttawaOttawa, OntarioCanada

Michael A. Nardi, MSAssociate ProfessorDepartment of Pediatrics and PathologyNew York University School of MedicineNew York, NY 100016USA

V

John Petersen, PhDProfessor, Department of PathologyThe University of Texas Medical BranchGalveston, TX 77555USA

Joseph M. Quashnock, PhDLaboratory DirectorPerkinElmer Genetics, Inc90 Emerson Lane, Suite 1403P.O. Box 219Bridgeville, PA 15017USA

Semyon A. Risin, MD, PhDProfessor of Pathology & Laboratory MedicineDirector of Laboratory Medicine Restructuring& Strategic Planning ProgramUniversity of Texas Health Science Center-Houston Medical School6431 Fannin Street, MSB, 2.290Houston, TX 77030USA

Maria Cristina Rosatelli, PhDProfessor, Dipartimnto di Scienze Biomediche e Biotecnologie Universit degli Studi di Cagliari09121 Cagliari, SardinaItaly

Donald L Rucknagel, MD, PhDProfessor EmeritusDepartment of Human GeneticsUniversity of Michigan, School of MedicineAnn Arbor, MichiganUSA

Kimberly Russell, MT (ASCP), MBAManager & Operations CoordinatorHematology and Blood BankSt. John Hospital & Medical Centerand affiliated hospitals of St. John Providence Health System, MichiganUSA

VI

Luisella Saba, PhDProfessor, Dipartimnto di Scienze Biomediche e Biotecnologie Universit degli Studi di Cagliari09121 Cagliari, SardinaItaly

Dror Sayar, MD, PhDDepartment of Pediatrics,Hematology-OncologyTel Hashmer Medical CenterRamat GanIsrael

Upendra Srinivas, MDDepartment of HematologyKokilaben Dhirubhai Ambani Hospital& Medical Research InstituteMumbai, MaharashtraIndia

Elizabeth Sykes, MDClinical PathologistWilliam Beaumont HospitalRoyal Oak, MichiganUSA

Ali Taher, MD, PhDProfessor Medicine, Hematology & OncologyAmerican University of Beirut Medical CenterBeirutLebanon

Kenneth F. Tucker, MDDirector, Hematology & Oncology ServicesWebber Cancer CenterSt. John Macomb-Oakland HospitalWarren, MichiganUSA

VII

Zia Uddin, PhDConsultant, Clinical ChemistryDepartment of PathologySt John Macomb-Oakland HospitalWarren, MichiganUSA

Vip Viprakasit, MD, D. PhilProfessorDepartment of Paediatrics & Thalassemia CenterFaculty of Medicine Siriraj Hospital, Mahidol University2 Prannok Road, BangkoiBangkok 10700

Thailand

Dr. Henri WajcmanDirector of Research EmeritusEditor-in-Chief HemoglobinINSERM U955 (Team 11)Hospital Henri Mondor94010 Creteil France

Winfred Wang, MDProfessor of PediatricsUniversity of Tennessee College of MedicinePediatric Hematologist & OncologistSt Jude Children’s Research HospitalMemphis, TennesseeUSA

Andrew N Young, MD, PhDDepartment of Pathology & Laboratory MedicineEmory University School of MedicineAtlanta, GA 30303USA

VIII

Financial Disclosure

I neither had nor will have financial relationship with any of the manufacturers or any otherorganization mentioned in the book.

Similarly all the contributors and reviewersof the book have worked with gratis to furtherthe cause of education.

This book and its translations into severallanguages are provided at no charge.

August 2015 Zia Uddin, PhD

IX

Dedication

This book is dedicated with heartfelt thanks to myprofessors responsible for my PhD level education in Chemistry at the Illinois Institute of Technology, Chicago, Illinois, and post-doctoral education and training in ClinicalChemistry at the University of Illinois Medical Center, Chicago, Illinois.

Illinois Institute of Technology, Chicago, Illinois

Professor Kenneth D. Kopple, PhDProfessor Paul E. Fanta, PhDProfessor Robert Filler, PhDProfessor Sidney I. Miller, PhD

University of Illinois Medical Center, Chicago, Illinois

Professor Newton Ressler, PhD

August 2015 Zia Uddin, PhD

X

Preface

Higher level education is one of the blessings of God. Unfortunately, primarily due to economic and logistic reasons a vast majority of the qualified candidates are denied this opportunity.

Internet has the potential of mass education at an infinitesimal cost. This is the 3rd book launched via Internet by me at no charge.

All the MD/PhD degree holders are most respectfully requested to utilize the Internet as a means of communication to launch books at no charge in their areas of expertise.

Love God

Love People

Serve The World

August 2015 Zia Uddin, PhD

XI

Acknowlegement

During the past three years I contacted worldwide >200 family physicians, clinical chemists, pathologists, hematologists, public health officials and experts in diagnostic hemoglobinopathy for formatting this book. The contribution of all of these individuals is heartfelt and very much appreciated.

I am highly indebted to the following persons for their technical support:

Diane M. Maennle, MDRita Ellerbrook, PhD

Kimberly R. Russell, MT (ASCP), MBAJennifer Randazzo, MS (Information Technology)

The following manufacturers and organizations provided technical support,and facilities for the collection of data for the book:

Helena Laboratories, USASebia, FrancePerkinElmer Corporation, USA

Bio-Rad, USAARUP Laboratories, USAQuest Diagnostics, USACollege of American Pathologists, USASeven Universities and four Newborn Screening Laboratories, USA(names are with held as per their request)

Mr. Mathew Garrin, Biomedical Communications and Graphic Arts Department, Wayne State University, School of Medicine, Detroit has worked on the figures, scans, and layout of the book. I am very grateful to him for his contribution.

Finally, I would like to thank the following persons for facilitating my work:

Adrian J. Christie, MD, Medical Director of LaboratoriesSt. John Macomb-Oakland Hospital, Warren, Michigan, USAAnoop Patel, MD, Assistant Systems Medical DirectorSt John Providence Health System Laboratories, Warren, Michigan, USAMr. Tipton Golias, President & CEO

Helena Laboratories, Beaumont, Texas, USA

August 2015 Zia Uddin, PhD

XII

Table of Contents

Chapter 1 Hemoglobin 1 Thomas E. Burgess, PhD

1.1 Hemoglobin Structure 1.2 Hemoglobin Function 1.3 Hemoglobin Synthesis 1.4 Hemoglobin Variants

Chapter 2 Red Blood Cell Morphology 10 John Lazarchick, MD Angie Duong, MD

Chapter 3 Diagnostic Laboratory Methods

3.1 Basic Concepts 44Jayson Miedema, MDand Christopher R. McCudden, PhD

3.1.1 Unstable Hemoglobins3.1.2 Altered Affinity Hemoglobins3.1.3 Sickle Solubility Test3.1.4 Serum Iron, TIBC, Transferrin, Ferritin 3.1.5 Soluble Transferrin Receptor3.1.6 Hepcidin

3.2 Microcytosis 55Diane Maennle, MDand Kimberly Russell, MT (ASCP), MBA

3.3 Hereditary Persistence of Fetal Hemoglobin 62 Bernard G. Forget, MD

3.3.1 Introduction3.3.2 Deletions Associated with the HPFH Phenotype3.3.3 Non-Deletion Forms of HPFH3.3.4 HPFH Unlinked to the β-Globin Gene Cluster 3.3.5 Conclusion

XIII

3.3.6 Hemoglobin F Quantification

3.4 Flow Cytometry Measurements of Cellular Fetal Hemoglobin, Oxidative Stress and Free Iron in Hemoglobinopathies 75Eitan Fibach, MD

3.4.1 Flow Cytometry of Blood Cells3.4.2 Measurement of Fetal Hemoglobin-Containing

Erythroid Cells 3.4.3 Staining Protocols for F-RBCs and F-Retics (15)

3.4.4 F-Cell Determination for Fetal-Maternal Hemorrhage (FMH) in Pregnant Patients wit β-Thalassemia- A single Case and General Conclusion (16)

3.4.5 Oxidative Stress3.4.6 Staining Protocols for ROS and GSH3.4.7 Intracellular Free Iron3.4.8 Staining Protocol for LIP

3.5 Solid Phase Electrophoretic Separation 95 Rita Ellerbrook, PhD, and Zia Uddin, PhD

3.5.1 Introduction3.5.2 Cellulose Acetate Electrophoresis (alkaline pH)3.5.3 Agarose Gel Electrophoresis (alkaline pH)3.5.4 Agar Electrophoresis (acid pH)3.5.5 Interpretation of Hemoglobin Agarose Gel (pH 8.6)

and Agar Gel (pH 6.2) Electrophoresis3.5.6 Requirements for the Identification of Complex

Hemoglobinopathies

3.6 Capillary Zone Electrophoresis 107 Zia Uddin, PhD

3.6.1 Introduction3.6.2 Basic Principle 3.6.3 Application of CZE in Diagnostic Hemoglobinopathies3.6.4 Interpretation of CZE Results

XIV

3.7 Isoelectric Focusing 117 David Hockings, PhD

3.7.1 Introduction3.7.2 IEF of Normal Adult Hemoglobin: HbA (Adult),

HbF (Fetal), HbA2

3.7.3 IEF of Normal Newborn Hemoglobins: HbF (Fetal) and HbA (Adult)

3.7.4 IEF of Beta-Chain Variant Hemoglobins3.7.5 IEF of Alpha Chain Variant Hemoglobins3.7.6 IEF of Thalassemias

3.8 High Performance Liquid Chromatography 129 Zia Uddin, PhD

3.8.1 Introduction3.8.2 Basic Principle3.8.3 Illustrations

Chapter 4 Globin Chain Analysis

4.1 Solid Phase Electrophoretic Separation 136Zia Uddin, PhD

4.1.1 Cellulose Acetate Electrophoresis (Alkaline and Acid pH)

4.2 Reverse Phase High Performance Liquid 140 Chromatography Zia Uddin, PhD, and Rita Ellerbrook, PhD

4.3 Globin Chain Gene Mutations: DNA Studies 149 Joseph M. Quashnock, PhD 4.3.1 Introduction

4.3.2 Genotyping-PCR Methodology4.3.3 Mutations

XV4.4 Electrospray Ionization-Mass Spectrometry 166

Gul M. Mustafa, PhD and John R. Petersen, PhD

4.5 PCR and Sanger Sequencing 181 Elaine Lyon, PhD

4.5.1 Alpha Globin 4.5.2 Beta Globin 4.5.3 Sequencing

4.5.4 Reporting Sequence variants4.5.5 DNA Sequence Traces4.5.6 Conclusion

Chapter 5 Alpha and Beta Thalassemia 191 Herbert L. Muncie, MD.

5.1 Epidemiology5.2 Pathophysiology5.3 Alpha Thalassemia5.4 Beta Thalassemia5.5 Diagnosis5.6 Treatment5.7 Complications5.8 Other Treatment Issues

5.8.1 Hypersplenism 5.8.2 Endocrinopathies 5.8.3 Pregnancy 5.8.4 Cardiac 5.8.5 Hypercoagulopathy 5.8.6 Psychosocial 5.8.7 Vitamin Deficiencies

5.8.8 Prognosis

XVI

Chapter 6 Neonatal Screening for Hemoglobinopathies 212 Zia Uddin, PhD

6.1 Introduction6.2 Methodologies6.3 Laboratory Reports Format & Interpretation6.4 Examples of Neonatal Screening

6.4.1. Capillary Zone Electrophoresis6.4.2 Isoelectric focusing6.4.3 Isoelectric focusing and High Performance Liquid Chromatography6.4.4 Isoelectric focusing, High Performance Liquid Chromatography and DNA studies

6.5 Genetic Counseling & Screening

Chapter 7 Prenatal Diagnosis of Beta-Thalassemia and Hemoglobinopathies 236 Maria Cristina Rosatelli, PhD, and Luisella Saba, PhD

Chapter 8 Hemoglobin A1c 266 Zia Uddin, PhD

8.1 Introduction

8.2 HbA1c Diagnostic Role in Diabetes Mellitus, and Glycemic Control in Adults

8.3 Measurement of HbA1c 8.4 Factors Affecting the Accuracy of Hb A1c Assay

XVII

Case Studies 278

Introduction

Case # 1 Normal Adult 281

Case # 2 Hemoglobin S trait 286

Case # 3 Hemoglobin S homozygous 292

Case # 4 Hemoglobin S with hereditary persistence of fetal hemoglobin (HPFH) 298

Case # 5 Hemoglobin G-Philadelphia trait 306

Case # 6 Hemoglobin S-G Philadelphia 313

Case # 7 Hemoglobin G-Coushatta trait 321

Case # 8 Hemoglobin C trait 327

Case # 9 Hemoglobin C homozygous 333

Case # 10 Hemoglobin C with hereditary persistence of fetal hemoglobin (HPFH) 340

Case # 11 Hemoglobin S-C disease 346

Case # 12 Hemoglobin D-Los Angeles (D-Punjab) trait 353

Case # 13 Hemoglobin S-D disease 360

XVIII

Case # 14 Hemoglobin E and Associated Disorders 367

Case # 14 a Hemoglobin E trait 373

Case # 14 b Hemoglobin E homozygous 378

Case # 14 c Hemoglobin S-E disorders 384

Case # 15 Hemoglobin S-Korle Bu (G-Accra) 390

Case # 16 Hemoglobin O-Arab trait 396

Case # 17 β-Thalassemia trait 402

Case # 18 Hemoglobin S-β+- thalassemia 408

Case # 19 Hemoglobin C-βo – thalassemia 415

Case # 20 Hemoglobin Hasharon trait 421

Case # 21 Hemoglobin Zurich trait 428

Case # 22 Hemoglobin Lepore trait 434

Case # 23 Hemoglobin J-Oxford trait 442

Case # 24 Hemoglobin J-Baltimore trait 449

Case # 25 Hemoglobin Malmo trait 455

Case # 26 Hemoglobin Koln trait 466

Case # 27 Hemoglobin Q-India trait 475

Case # 28 Hemoglobin Dhofar trait 488

XIX

Chapter 1

HemoglobinThomas E. Burgess, PhD

To attempt a full treatise on hemoglobin in this textbook would be an effort in

futility as the purpose is not to duplicate knowledge already present in the literature.

Rather, this chapter is to provide basic information to the reader which will allow him/her

to properly identify hemoglobin variants in their laboratory. A basic knowledge of the

hemoglobin molecule is absolutely critical to that effort and the sections printed below

are written expressly for that purpose. For a complete treatise on hemoglobin, textbooks

such as Disorders of Hemoglobin 1 edited by Steinberg, Forget, Higgs and Nagel should

be consulted.

1.1 Hemoglobin Structure

Composed of 2 distinct globin chains, the complex protein molecule known as

hemoglobin (“heme” + “globin”) is arguably THE primary component of the red blood cell

in human beings. In “normal” adults, the globin chains are either alpha (α), beta (β),

gamma (ϒ) or delta (δ). In addition, during embryonic life in utero, zeta (ζ) and epsilon

(ε) chains are present in the first several weeks of life, being rapidly converted to alpha,

beta and gamma chains as development occurs.

1

Figure 1. Globin chains concentration changes in embryonic, fetal and post-natal stages of life (Huehns ER, Dance N, Hecht S, Motulsky AG. Human embryonic hemoglobins. Cold Spring Harbor Symp Quant Biol 1969; 29: 327-331). Adopted with permission from Blackwell Publishing (Barbara J. Bain, Haemoglobinopathy Diagnosis, 2nd Edition, 2006).

Each of these globin chains has associated with it a porphyrin molecule

known as heme whose primary function in the red blood cell is the facilitation of

transport of oxygen to the tissues of the human body. The globin portion of the

molecule serves several functions, not the least of which is protection. The internal

pocket of the molecule formed from the convergence of the four globin chains,

2

provides a hydrophobic environment in which the heme molecules reside. This

pocket protects the heme from oxidation and facilitates oxygen transfer to the

tissues of the body. The previously mentioned ζ and ε chain-containing

hemoglobins have very high oxygen affinities, a factor very important in the early

embryonic life of the fetus.

The hemoglobin molecule can be looked at in four different ways; primary,

secondary, tertiary and quaternary structural views. While outside of the scope of

this volume, each of these structures contributes definitive unique properties to the

various hemoglobin molecules from normal hemoglobins to the very rare and

functionally diverse molecules. The primary structure of all hemoglobins is the order

of amino acids found in the globin chains of the molecule. It is this unique sequence

that is the major differentiator of hemoglobin from each other. The secondary

structure of hemoglobin is the arrangement of these amino acid chains into alpha

helices separated by non-helical turns2. The tertiary structure is the 3-dimensional

arrangement of these globin chains forming the “pocket” of hemoglobin that cradles

the iron molecule in its grasp. The quaternary structure is the moving structure of the

molecule that facilitates the oxygenation of the heme molecules in response to the

physiological needs of the human body.

3

Figure 2. Tertiary structure of a β globin chain and the quaternary structure of hemoglobin molecule (Adopted with permission from Blackwell Publishing, Barbara J. Bain, Haemoglobinopathy Diagnosis, 2nd Edition, 2006).

The forthcoming sections will elucidate the effects that these structural

considerations have on the hemoglobin molecule and, more specifically, the

abnormal and atypical hemoglobin variants.

1.2 Hemoglobin Function

As mentioned above, the primary function of hemoglobin is to reversibly

transport oxygen to the tissues of the body. In addition, however, this flexible

molecule can also transport carbon dioxide (CO2) and nitrous oxide (NO). The

transport of CO2 is facilitated by reversible carbamoylation (formation of carbamoyl

moiety, i.e., H2NCO-) of the N-terminal amino acids of the α globin chains and can, 4

via proton scavenging, keep CO2 in the soluble bicarbonate form3. Nitrous oxide is

handled in two different ways by hemoglobin: one as a transporter and the other as

a scavenger. Blood levels of NO are therefore, by definition, a balance between NO

production and NO removal by binding to oxyhemoglobin. Since NO is an extremely

potent vasodilator, hypoxic patients will have lower oxyhemoglobin and therefore

higher amounts of free NO. This free NO can cause significant vasodilation, a

physiological effect that is very desirable in hypoxia.

All hemoglobin molecules, either normal or variant, share the same

functionality in the human body. Therefore, the primary structural differences

mentioned above and in more complete treatises (i.e., amino acid

substitutions/deletions) will be the prime reason for functional differences. It is these

amino acid variances that, along with the secondary, tertiary and quaternary

structural differences, will determine if the variant hemoglobin is either benign or

clinically important.

The bottom line is this – whether the hemoglobin is normal or variant in

nature, the prime reason for determining the hemoglobin phenotype of the patient is

to assess the functionality of the hemoglobin. If the variant is normally functioning in

both the heterozygous and homozygous states, the clinical picture is benign. If,

however, the variant has normal properties in the heterozygous state (i.e., “trait”) but

clinical issues in the homozygous state (i.e., “disease”), the phenotypic analysis and

subsequent interpretation becomes ultimately important to the patient.

5

1.3 Hemoglobin Synthesis

The synthesis of hemoglobin, as mentioned before, is under the control of

gene loci on two chromosomes: chromosome 11 (the beta globin or “non-alpha”

gene) and chromosome 16 (the alpha globin gene). Hemoglobin variants (alpha,

beta, gamma, delta and fusion) are the result of alterations in the nucleotide

sequences of the globin genes and can occur for more than one reason. Mutations

such as point mutations, insertions and deletions can have major, minor or no

influences on hemoglobin function or structure. That being said, the site of the

synthetic variance can in some cases alter the ability of the hemoglobin molecule to

function in a normal manner, i.e., stability, oxygen affinity, solubility or other critical

functions. These alterations truly determine whether the variant hemoglobin is

classified as benign (i.e., no abnormal or pathological effect) or pathological (a

significant physiological effect). The actual nature of the alteration is not of initial

importance to the hemoglobinopathy interpreter. However, once assigned, the

identity of the variant hemoglobin may become of importance when looking at

second generation offspring from the variant carrier, i.e., the pregnant female. For

most hemoglobin variants, the synthetic pathway is of no clinical interest in that the

resulting hemoglobin is benign. It may, however, be of academic interest in that the

identification of the synthetic anomaly can, indeed point to the genetic locus or loci

involved in the alteration, thus giving information to the genetic counselor as to

possible genetic details of the hemoglobinopathy.

6

As mentioned before, the true reason for identifying the abnormal hemoglobin

or hemoglobins in patients is to identify any associated functional anomalies

associated with these hemoglobins. The actual hemoglobin identification in and of

itself is merely of academic interest.

1.4 Hemoglobin Variants

All hemoglobin variants have one thing in common – they all involve the

hemoglobin molecule and its functionality. Whether alpha, beta, gamma, delta,

fusion variant, etc., the variant and its effect are judged not on its migration or

concentration but rather on its functionality. The amino acid variation (e.g., glutamic

acid → valine at position 6 on the beta chain for hemoglobin S) is the prime effector

of the variant’s functional alteration(s) and will in most cases be the causative factor

in any abnormal migration that the variant may have versus the “normal”

hemoglobins (A, F, A2). Most variants therefore will have altered electrophoretic or

chromatographic migrations when compared to the normal variants. Some, such as

hemoglobin Chicago, are not separable by normal electrophoretic techniques and

rely on high performance liquid chromatographic (HPLC) separations to identify its

presence in the blood. As previously mentioned, the presence of variant “traits” (i.e.,

AS, sickle trait) may or may not be of clinical consequence. Where these traits

really are of importance is in the homozygous state (i.e., SS for hemoglobin S). The

clinical picture dramatically changes with significant physiological changes being

directly associated with the homozygous state. This therefore requires the

interpreter to have several pieces of information specific to the patient at hand

7

during the interpretation of the hemoglobinopathy. This data includes, but is not

limited to, pregnancy, transfusion history and ethnicity. All of these pieces of

information can be critical to the proper identification/interpretation of the

hemoglobin variant in the patient’s specimen. For example, an elevation of

hemoglobin F in a female patient with a normal hemogram may be evidence of

hereditary persistence of fetal hemoglobin; whereas, if this female is pregnant, the

elevation may be a normal physiological response to the fetal presence in her body.

These data may not be readily available and may require contact with the ordering

healthcare professional to obtain these facts. However obtained, they are

necessary for the proper identification of the hemoglobin variant or variants in the

patient’s bloodstream and therefore are important in the assignment of a benign or

pathological assessment of the variant hemoglobin.

The variants described in the following chapters all obey the aforementioned

differences, i.e., amino acid substitutions, genetic deletions, sequence modifications,

etc. While not critical, the exact identification of the variant in and of itself is not

normally life-threatening, especially in the heterozygous state, i.e., “trait”. It is

essential that the variant be properly identified as a mis-identification can lead to

other issues. For example, a mis-interpretation of a hemoglobin G trait (AG) as a

sickle trait (AS), while not in and of itself is clinically an issue, presents real

difficulties for a couple expecting a child. If both partners are AS, there is a 1 in 4

chance that a child born to this couple could be homozygous SS or sickle cell

disease. In the case of an AS mother and an AG father (or vice versa), there is a 1

in 4 chance of a child being born with a phenotype of SG. While on the surface this

8

may appear as a problem, the SG phenotype is no more of a clinical issue than a

simple AS trait. Without the exact identification of the AG trait, the interpretation and

action taken by attending clinicians may be very different.

References

1. Steinberg, MH, Forget, BG, Higgs, DR and Nagel, RL., Disorders of Hemoglobin,

Cambridge University Press, 2001.

2. Bain, Barbara J.. in Hemoglobinopathy Diagnosis, 2nd Ed., pg. 4, Blackwell Publishing, 2006.

3. Bain, Barbara J.. in Hemoglobinopathy Diagnosis, 2nd Ed., pg. 1, Blackwell Publishing, 2006.

9

Chapter 2

Red Blood Cell Morphology

John Lazarchick, MDAngie Duong, MD

Knowledge of red blood cell (RBC) morphology is essential for the clinical diagnosis of

hemoglobinopathy. The diameter of RBC, when mature under normal circumstances

is approximately 7-8 microns, and RBC is round, anuclear and biconcave disc-shaped.

A study of RBC morphology includes size, shape, color, inclusions and arrangement. In

this chapter we have presented with pictures of the most commonly encountered RBC

morphologies with legends and few examples of the diseases with abnormal RBC

morphology. In the clinical cases of this book, we have mentioned only the main

features of the peripheral blood smear, therefore a review of this chapter is advised

for a naïve reader for the proper diagnosis of hemoglobinopathy.

The following RBC morphology cases are presented in this chapter:

Size: Macrocyte – large Fig. 1Microcyte –small Fig. 2Normocyte – normal Fig. 3Hemoglobin Content: Hypochromic –low Fig. 4Normochromic – normal Fig. 5Polychromatic – high Fig. 6Shape and Inclusions:Anisocytosis Fig. 7Poikiocytosis Fig. 8Acanthocyte Fig. 9Basophilic Stippling Fig.10Bite Cell Fig.11Blister Cell Fig.12Burr Cell (Ecchinocyte) Fig.13Heinz Body Fig.14

10

Howell-Jolly Body Fig 15Pappenheimer Body Fig.16Schistocyte Fig.17Sickle Cell Fig.18Spherocyte Fig.19Stomatocyte Fig. 20Target Cell Fig. 21Teardrop Cell Fig. 22RBC Agglutination Fig. 23Rouleaux Formation Fig. 24Diseases :Erythroblastosis Fetalis Fig. 25Hemoglobin C Disease Fig. 26Hemoglobin C/beta Thalassemia Fig. 27Hemoglobin S/beta Thalassemia Fig. 28Hemoglobin SC Disease Fig. 29Sickle Cell Disease Fig. 30Fetal-maternal Hemorrhage: Fig. 31Kleinhauer-Betke Stain

11

Fig. 1 – Macrocyte-large

The diameter of RBC >9-14 microns (1.5 to 2 times larger than normal RBC) and the MCV >100 fL is characteristic of macrocyte. Macrocytes are mostly oval in shape.

12

Fig. 2 – Microcyte-small

RBC, when abnormally smaller (< 5 micron) than normacytic RBC (7-8 micron) is defined as microcyte (also called microerythrocyte). The MCV of the microcyte RBC is < 80 fL.

13

Fig. 3 – Normocyte-normal

The diameter of RBC, when mature under circumstances is approximately 7-8 microns, and are round, anuclear, biconcave disc-shaped with an internal volume of 80-100 fL.The term normocyte is used when the size of the RBC is normal.

14

Fig. 4 – Hypochromasia

Hypochromasia is a descriptive term for red blood cells where the central pallor is greater than one third the diameter of the red blood cell (black arrows). This is due to a decrease in the amount of hemoglobin in the cells. Diseases with prominent hypochromasia are iron deficiency anemia, anemia of chronic disease, and sideroblastic anemia. Some cases of myelodysplastic syndrome can also have hypochromatic red blood cells. Hypochromasia is reflected in the complete blood count (CBC) by a decreased mean corpuscular hemo-globin concentration (MCHC).Also present are: target cells/codocytes (red arrow), polychromatic forms (blue arrow), fragmented red blood cells/schistocytes (green arrows), and tear drop forms/dacryocytes (yellow arrows). Overall, this smear shows moderate anisopoikilocytosis.

15

Fig. 5 – Normochromic-normal

This descriptive term is applied to a red blood cell with a normal concentration of hemoglobin. The above figure is a peripheral blood cell smear of a patient treated for iron deficiency anemia. Blue arrow shows normochromic-normal RBC. Black arrow shows hypochromic-microcytic RBC.

16

Fig. 6 – Polychromatic-high

This smear demonstrates polychromasia. Numerous polychromatic forms (black arrows), which are young slightly larger red blood cells with a purple-tinge due to retained RNA, are present. Polychromasia is the bone marrows response to anemia, where the bone marrow releases younger red blood cells. Sometimes, nucleated red blood cells are also released into the peripheral blood. Due to their larger size, when many polychromatic forms are present, the CBC values of mean corpuscular volume (MCV) as well as RDW (red blood cell distribution width) will be increased.In a supravital stains, such as cresyl violet, the retained RNA in the polychromatic forms precipitate out and these cells are called reticulocytes. Thus, sometimes the terms polychromatic form is used interchangeably with reticulocytes.

17

Fig. 7 – Anisocytosis

The term anisocytosis refers to size variation seen among red blood cells. As demonstrated above, there are small red blood cells as well as large red blood cells, some approaching the size of a neutrophil (green arrow). Ansiocytosis is a reactive process where the bone marrow is releasing younger red blood cells, therefore an increased number of polychromatic forms can also be seen (black arrow). In the complete blood count (CBC), anisocytosis is reflected by having an increased red cell distribution width (RDW).

18

Fig. 8 – Poikilocytosis

Poikilocytosis refers to shape variation. In poikilocytosis, the red blood cells have lost their normal discoid appearance. The example shown here has a predominance of ovalocytes/elliptocytes, which are red blood cells that have a length twice their diameter (a few are indicated by blue arrows). Also seen are schistocytes (red arrows), which are fragmented red blood cells. Ovalocytes/Elliptocytes are seen in peripheral blood smear in some conditions, e.g., thalassemia, iron deficiency, etc.Note: When both shape and size variation is seen in the red blood cells, the term anisopoikilocytosis can be used.

19

Fig. 9 – Acanthocyte (Spur Cell)

 These are red blood cells with spike-like projections (arrow) of varying length. They can be seen in both hereditary and acquired hemolytic anemias including alcoholic liver disease, pyruvate kinase deficiency, vitamin E deficiency, Huntington’s disease-like situation and abetalipoproteinemia. In the latter case, malabsorption of fat, neurologic damage and developmental delay are noted.

20

Fig. 10 – Basophilic Stippling

Red blood cells have multiple fine or coarse small basophilic dot-like inclusions which are due to small clumps of ribonucleic acid and mitochondria. These inclusions can be seen in a wide variety of conditions including lead poisoning, hereditary hemoglobinopathies including unstable hemoglobins, thalassemias, sideroblastic anemias, megaloblastic anemia and hereditary pyrimidine 5’- nucleotidase deficiency.

21

Fig. 11 – Bite Cell

Bite cell (arrow) has a semicircular portion of the membrane removed. This morphologic abnormality results from splenic macrophages removing denatured precipitated hemoglobin with Heinz body formation in these cells. The most common cause of this finding is glucose-6-phosphate dehydrogenase deficiency.

22

Fig. 12 – Blister Cell

Red blood cells with cytoplasmic clearing (large arrows) on one side and hemoglobin on the other side in a patient with hemolytic anemia. Multiple polychromatophilic red blood cells (reticulocytes) are noted (small arrow). In addition, a single cell with a Howell-Jolly body inclusion (double arrows) is noted,

23

Fig. 13 – Burr Cell (Echinocyte) These are red blood cells with short round membrane projections with blunt ends (large arrow). Red blood cells with more spike-like projections (small arrow) can also be seen. This finding is often an artifact of slide preparation but is typically seen in patients with uremia and pyruvate kinase deficiency.

24

Fig. 14 - Heinz Body

In a RBC when the hemoglobin is denatured (either by a change of an internal amino acid or glucose-6-phosphatse deficiency, etc.), the heme portion of hemoglobin molecule is dissociated from the globin chain. The globin chain after dissociation from the heme molecule becomes denatured forming a small ball like structure (black arrow) inside the RBC, and thus called Heinz body.

25

Fig. 15 – Howell-Jolly Body

This red blood cell inclusion (arrow) is round basophilic DNA remnant usually noted in the outer third of circulating red blood cells. These inclusions are normally extruded in the bone marrow during normal erythroid maturation. Howell-Jolly bodies can be seen in asplenia, conditions associated with hyposplenia including sickle cell disease and severe hemolytic anemia.

26

Fig. 16 - Pappenheimer Bodies These are small dark irregular staining granules (large arrow) of non-heme iron usually noted on the periphery of red blood cells formed by phagosomes that engulf excess iron. Basophilic stippling is present in the dysplastic nucleated RBC (small arrow) These granules stain positive with Prussian blue stain in both the nucleated RBC and mature red blood cells as shown in the lower image. They can be found in a variety of conditions including sideroblastic anemias, thalassemias and myelodyplastic syndromes.

27

Fig. 17 - Schistocyte (RBC fragments, Helmet Cells) These are red blood cell fragments typically with two pointed ends formed when RBCs are sheared by fibrin strands in clotted blood vessels. Disorders include microangiopathic hemolytic anemia, disseminated intravascular consumption (DIC), thrombotic thrombocytopenic purpura (TTP) and hemolytic uremic syndrome (HUS).

28

Fig. 18 - Sickle Cell

In inherited blood cell disease (change of an amino acid residue in the globin chain) the shape of the RBC is deformed. The deformation of RBC resembles (a waxing crescent) a moon sighted on the first day of lunar month. Since this deformation looks like a sickle (an implement with a semicircular blade attached to a short handle, used for cutting grain), therefore this deformation is called sickle cell.

29

Fig. 19 – Spherocytes

This peripheral blood smear is from a patient with autoimmune hemolytic anemia (AIHA) and is characterized by many spherocytes (blue arrows) and microspherocytes (black arrows). Spherocytes are red blood cells that have no central pallor. As the name implies, microspherocytes are small spherocytes. If the majority of the cells in a peripheral smear are spherocytes, the possibility of hereditary spherocytosis arises. Hereditary spherocytosis is an autosomal dominant disease where one of the genes that code for red blood cell proteins (such as spectrin and ankyrin) become mutated.

30

Fig. 20 – Stomatocyte

 Red blood cells with slit-like central pallor (arrow) caused by a decrease in surface area to volume ratio associated with a membrane permeability disorder. Hereditary stomatocytosis is associated with hemolysis which can be severe. Acquired stomatocytosis can be seen in acute alcohol intoxication, chronic liver disease and as drying artifact in peripheral smear preparation.

31

Fig. 21 - Target Cells

Also known as codocytes, these red blood cells appear to have a bullseye in the center of the red blood cell’s central pallor. This morphologic change is due to a relative excess of cell membrane, due to decreased cell content or increase in the cell’s surface area. Target cells can be seen in liver failure, Hemoglobin C disease, thalessemias (both alpha and beta), and iron deficiency.

32

Fig. 22 - Tear drop cells

Also known as dacryocytes/dacrocytes (red circles), are distorted red blood cells where one end of the cell is drawn into a sharp point. These cells are usually seen in myelophthsic anemias, which is where the normal marrow space is occupied by non-hematopoietic elements, such as fibrosis or metastatic carcinoma. It is hypothesized that the shape of the cells is due to the red blood cells squeezing between fibers or the cells extrinsic to the marrow.

33

Fig. 23 - RBC Agglutination

Clumping of the red blood cells is due to coating of the RBC surface with antibodies. Disorders causing the agglutination may be primary as in cold agglutinin disease or secondary, either clonal as in lymphoproliferative disorders or polyclonal as seen in Mycoplasma pneumonia. The upper left insert is from a slide prepared at room temperature and the upper right insert is a slide after warming the sample to 370 C with clearing of the agglutination in a patient with cold agglutinin disease.

34

Fig. 24 – Rouleaux Formation

Rouleaux formation is seen in peripheral blood smears in association with plasma cell neoplasms, most commonly myeloma. The red cells become stuck together in a “stack of coins” formation, due to the excess immunoglobulin proteins released by malignant plasma cells. Not all cases of plasma cell neoplasms have rouleaux formation. Rouleaux formation is one of the causes of an increased erythrocyte sedimentation rate (ESR).

35

Fig. 25 - Erythroblastosis Fetalis This is an alloimmune hemolytic anemia in the fetus secondary to placental transfer from mother to fetus during pregnancy of anti– A or B or anti-Rh blood group IgG antibodies. These blood groups are present on the fetal RBCs but not on the maternal RBCs which then causes immune hemolysis in the fetal circulation. As noted on the smear, numerous nucleated RBCs (large arrow) and polychromatophilic RBCs (small arrow) are noted.The case shown above was due to antibodies to Rh D blood group. 

 

36

Fig. 26 - Hemoglobin C Disease

In this case of homozygous hemoglobin C disease essentially all of the RBCs are target cells (large arrow). Hemoglobin C crystals are rod shaped inclusions (Washington Monuments—small arrow) in red blood cells in both heterozygous and homozygous hemoglobin C disease as well as hemoglobin SC disease. Upper image shows the crystals at a higher magnification.

37

Fig. 27 - Hemoglobin C/beta Thalassemia

Although most patients with this compound heterozygotic state for hemoglobin C and beta thalassemia are asymptomatic, a mild to moderate hemolytic anemia can be seen. The red blood cells are microcytic and hypochromic. Target cells (double arrow) and C crystals (single arrows) can be seen.

38

Fig. 28 - Hemoglobin S/beta Thalassemia

Hemolytic anemia due to both production of an abnormal hemoglobin (Hemoglobin S) and decreased synthesis of beta globin chains (Beta Thalassemia). Individuals have one abnormal beta chain with substitution of glutamic acid for valine and either decreased synthesis, beta+, or complete absence of the other beta chain, beta0. The peripheral smear shows sickle cells, nucleated red blood cells, polychromasia, microcytosis, hypochromic, target cells and basophilic stippling. Note the sickle cell in the insert and the Howell-Jolly body in the other RBC.

39

Fig. 29 - Hemoglobin SC Disease

This is a representative peripheral blood smear from a patient with hemoglobin SC disease. Hemoglobin SC disease is an inherited hemoglobinopathy where the two normal genes for hemoglobin A have been replaced by one hemoglobin S gene and one hemoglobin C gene. In hemoglobin S, a single nucleotide at position 6 of the gene is substituted by another nucleotide (glutamic acid is substituted by valine). A similar phenomenon occurs in hemoglobin C, where glutamic acid is substituted by lysine. When both hemoglobin S and hemoglobin C is present, the genes are codominant and lead to many interesting peripheral blood findings.

Hemoglobin S produces drepanocytes/sickle cells (black arrows) which are red blood cells that appear as crescent moon shapes or continued next page

40

sickles. Due to the abnormal hemoglobin content, the deoxygenated red blood cells become stuck in this shape, thus causing vascular occlusions which in turn lead to many complications such as pain crisis. Sickle cells are seen when there is no or decreased levels of hemoglobin A (such as hemoglobin SS disease, hemoglobin SC disease, hemoglobin S with thalessemia). In sickle cell trait, where there is one normal hemoglobin A gene and one hemoglobin S gene, sickle cells are not seen and the patients usually have no clinical symptoms.Hemoglobin C manifests in peripheral smears as numerous target cells/codocytes (green arrows). Additionally, in hemoglobin CC disease and in hemoglobin SC disease, hemoglobin C crystals (blue circle) can be seen. These crystals are desicated red blood cells with squared off/blunt edges. In hemoglobin C trait, target cells are seen but hemoglobin C crystals are not.

41

Fig. 30 - Sickle Cell Disease

Sickle cell disease is a hereditary hemolytic anemia caused by a single nucleotide substitution (SNP) of valine for glutamic acid in the beta globin chain of hemoglobin. This results in hemoglobin polymerizing at low oxygen tension with sickle cell formation (small arrow). There is marked polychromasia, target cells and nucleated red blood cells (inset—large arrow) on the peripheral smear.

42

Fig. 31- Fetal-maternal Hemorrhage: Kleihauer-Betke Stain

This test relies on the principle that red blood cells containing fetal hemoglobin (deep red staining RBCs) are less susceptible to acid elution than adult red blood cells. Its use is a means of quantitating fetal-maternal hemorrhage in Rh-negative mothers to determine the dose of Rho (D) immune globulin needed to inhibit formation of Rh antibodies. It can also be used to detect hereditary persistence of fetal hemoglobin (HPFH).

43

Chapter 3 Diagnostic Laboratory Methods

3.1 Basic ConceptsJayson Miedema, MD, and Christopher R. McCudden, PhD

3.1.1 Unstable Hemoglobins

Unstable hemoglobins are characterized by disorders in globin production which

affect the lifespan of the hemoglobin molecule and subsequently the cell leading to

decreased cell stability and increased cell turnover. There are a large number of specific

variants which can result in abnormal hemoglobin production, the most commonly

reported of which is Hb Koln. Many of these abnormal globin chains are a result of

single mutations in the form of deletions (e.g. Hb Gun Hill), insertions (e.g. Hb

Montreal), or substitutions (e.g. Hb Koln) and can result in weakened heme-globin

interactions, subunit interactions, or abnormal folding. These disorders are most

commonly expressed in the heterozygous form, most homozygous situations result in

preterm lethality.

Clinically, these patients often present with symptoms of hemolytic anemia which

can be of varying severity. Symptoms of hemolytic anemia include hyperbilirubinemia,

jaundice, splenomegaly, hyperbilirubinuria or pigmenturia as well as the formation of

Heinz bodies. This pheonotype can present or be exacerbated by infections as well as

certain types of drugs. Specifically sulfonamides, pyridium, and antimalarials are known

to cause exacerbation. Parvovirus can also induce aplastic crisis andHbA2 and HbF

may be increased. The peripheral smear often shows anisocytosis, poikilocytosis,

44

basophilic stippling, polychromasia and, hypochromasia. Since not all unstable

hemoglobins will give abnormal results on HPLC or electrophoresis and/or these results

can be somewhat non-specific, more definitive testing is often performed.

Testing for unstable hemoglobins relies on their decreased stability in heat or

isopropanol alcohol. While normal hemoglobins should be relatively stable in these

conditions, hemoglobins with mutations causing instability tend to be less so and will

precipitate out of solution in these environments. In the context of heat stability testing,

the amount of unstable hemoglobin in a sample is given by the following equation:

(Hb4°C-Hb50°C)/(Hb4°C)x100

Where Hb4°C is the hemoglobin concentration at 4 degrees centigrade and Hb50°C is

the concentration of hemoglobin at 50 degrees centigrade.

False positives may result from samples greater than 1 week in age as well as from

samples with large amounts of fetal hemoglobin. Additional technical and clinical

information on hemoglobinopathies associated with unstable hemoglobin can be

obtained from:

http://medtextfree.wordpress.com/2011/12/30/chapter-48-hemoglobinopathies

3.1.2 Altered Affinity Hemoglobins

Similar to how certain types of mutations can cause instability of the hemoglobin

molecule, other mutations can cause hemoglobins to have altered affinity for oxygen.

These mutations can be single point mutations, insertions, deletions, elongation,

deletion/insertion mutations and are often named after the city in which they were

discovered (Chesapeake, Capetown, Syracuse, etc.). Both alpha-chain variants, e.g. Hb

45

Chesapeake, and beta-chain variants, e.g. Hb Olser, Hiroshima, Andrew-Minneapolis,

etc., are known in the literature for altered affinity for oxygen. Many of these are

probably clinically insignificant but when significant most commonly present

phenotypically as an increase in oxygen affinity often times resulting clinically in

polycythemia (secondary to the bodies perceived lack of oxygen and subsequent

increase in erythropoietin). Measurement of hemoglobin affinity (p50) is critical to the

diagnosis. Conversely and less frequently described, a decreased affinity for oxygen

can lead to clinical cyanosis.

Testing for altered affinity hemoglobins relies on subsequent changes to the

oxygen dissociation curve and the partial pressure of oxygen at which hemoglobin is

50% saturated, the p50. Because most types of altered affinity hemoglobins cause an

increase in oxygen binding, a left shift in the oxygen dissociation curve results.

Automated systems are available for recording the oxygen dissociation curve and rely

on a Clarke electrode to measure oxygen tension while oxyhemoglobin fraction is

measured by dual wavelength spectrophotometer. Abnormal oxygen dissociation curves

are primarily caused by altered affinity hemoglobins but can also be caused by such

factors as pH, temperature, pCO2, and 2,3-diphosphoglycerate (2,3-DPG).

Measurement of pO2, pCO2, pH and SO2 allows for an estimation of p50 to be

calculated.

3.1.3 Sickle Solubility Testing

Sickle cell anemia is a disease resulting in anemia and painful crises, seen

almost exclusively in African Americans. These crises are caused by inappropriate 46

aggregation of deformed blood cells in small blood vessels. Widely believed to have

thrived in the gene pool because of its protective effects against malaria, it affects a

large number of people of African descent in its homozygous and clinically significant

form. An even greater number of people have sickle cell trait (approximately 8-10% of

African Americans), the heterozygous form, which is largely insignificant from a clinical

standpoint.

Sickle cell testing can be performed in a variety of ways and is currently most

commonly tested via hemoglobin electrophoresis when necessary. However, another

form of testing is known as sickle solubility testing which relies on the property of

increased cell fragility as a result of the glutamic acid to valine substitution at the 6 th

position of the beta globin gene, the most common genetic abnormality of sickle cell

anemia. Sickled red blood cells are soluble when oxygenated but upon deoxygenation

tend toward sickling, polymerization, and precipitation. The addition of sodium

metabisulfite reagent to a sample with hemoglobin S promotes deoxygenating and cell

lyses, creating turbidity in the solution. This turbidity makes it difficult to read a card

through the test tube. A negative test is one in which a card can be read through the

tube, a positive test is one in which the card cannot be read.

Several types of hemoglobins can cause false positives (for example some

types of hemoglobin C) so results should be confirmed by electrophoresis; in other

words, when used, solubility testing should be used as a screening test. The test also

fails to differentiate sickle cell trait (a single copy of the sickle cell gene, heterozygous)

from true sickle cell anemia (both copies are sickle cell, homozygous). Samples with low

hemoglobin concentration (<8%) should be doubled as this low concentration can lead

47

to false negatives. False positives can occur in the settings of lipemia or samples with

monoclonal proteins (dysproteinemia). Both positive and negative controls should be

used as results can be somewhat subjective

3.1.4 Serum Iron, TIBC, Transferrin, and Ferritin

Iron is essential for numerous metabolic functions in the body through its

incorporation into proteins involved in oxygen delivery (hemoglobin, myoglobin) and

electron transport and exchange (cytochromes, catalases). While a detailed description

of iron metabolism is beyond the scope of this compendium (interested readers should

seek the references below), it is worth considering the major mechanisms of iron

homeostasis in the context of erythropoiesis. Iron intake in the diet occurs either as free

iron or as heme. Free iron, in the form of Fe3+, requires reduction to Fe2+ by enzymes

and transporters to cross the intestinal mucosa; heme iron is absorbed directly by

mucosal cells where it is split from heme intracellularly. Once absorbed by the GI tract,

iron is either stored in association with ferritin or transported into the circulation in the

ferric (Fe3+) form. Because of the toxicity of ferric iron, it is transported in the circulation

bound to transferrin. The main target of transferrin-bound iron is erythroid tissue, which

takes up iron through receptor-mediated endocytosis. As dietary absorption accounts

for <20% of the daily requirement, iron recycling plays an essential role in maintaining

iron stores. During recycling, senescent red blood cells are phagocytosed by

macrophages in the spleen, liver, and bone marrow. Macrophages store some iron

(bound to ferritin), but most is returned to red cell precursors via transferrin. Unlike

dietary absorption, iron excretion is largely unregulated, where losses occur via

48

epithelial cell sloughing in the skin and GI tract or through menstrual bleeding in

premenopausal women. Accordingly, body stores depend on controlling iron uptake in

the GI tract and recycling.

Disorders of iron homeostasis fall into diseases of excess or deficiency. Iron

deficiency is common, particularly in women, and may result from inadequate intake,

blood loss, and pregnancy; in chronic disease iron deficiency is also common. Iron

excess may occur in hemochromatosis or as a result of repeated transfusions.

Clinically, iron status is assessed by measurement of serum iron, ferritin, transferrin,

and total iron binding capacity (TIBC).

Serum or plasma iron levels can be directly measured using several different

methods. Most commonly, a colorimeteric reaction scheme is used in which iron is

separated from transferrin at low pH (~4) and then reduced to Fe2+ for dye binding; the

color-complex is detected between 530-600 nm spectrophotometrically. Although iron

is typically increased in cases of iron excess and decreased in cases of deficiency,

serum iron measurement by itself is not particularly useful for diagnosis of iron

homeostasis disorders because of the high intra-individual variation in circulating iron

levels.

Total iron binding capacity (TIBC) is another test used to assess iron

homeostasis. TIBC can be measured or calculated. TIBC is measured by adding

excess iron to saturate transferrin (usually transferrin is 30% saturated). Unbound iron

is chelated and removed and then the remaining transferrin-bound iron is measured as

described above yielding the total capacity. This method can be affected by the

presence of non-transferrin iron binding proteins, particularly in cases of

49

hemochromatosis and thalassemias. Alternatively, TIBC may be calculated based on

the stoichiometric relationship between transferrin and iron (2 molecules of iron are

bound to each molecule of transferrin). TIBC is calculated from measured transferrin

using the following equation: TIBC (µg/dL) = 1.43 × transferrin (mg/dL). Conversely, the

concentration of transferrin may be calculated from measured TIBC as follows:

Transferrin (mg/dL) = 0.7 × TIBC (µg/dL). TIBC is increased in iron deficiency and

decreased in chronic anemia of disease and in iron overload (it may be normal or

decreased in thalassemia).

From TIBC and serum iron measurement, it is also possible to calculate the %

transferrin saturation (also known as iron saturation) using a simple formula: %

saturation = serum Fe (µg/dL) / TIBC (µg/dL) ×100. The percent saturation is usually

between 20-50%, supporting an excess capacity for iron binding. In cases of iron

overload, the % saturation increases dramatically. Saturation is moderately increased

in thalassemia and chronic anemia and in iron deficiency the saturation is decreased.

Ferritin is a large ubiquitous protein and the major iron storing protein in the

body. Ferritin serves to store thousands of iron atoms/molecule in a non-toxic form

acting as an iron reserve. Ferritin is found in small amounts in the blood, where it can

be measured as an indication of overall iron reserves (1 ng/mL serum iron approximates

10 mg total storage iron). In the blood, ferritin is generally poor in iron content and is

referred to as apoferritin. Circulating ferritin (or apoferritin) is measured using specific

antibodies, commonly by chemiluminescent immunoassay. Serum or plasma ferritin

levels are produced in proportion to dietary iron absorption; serum ferritin is increased

with iron overload and decreased in iron deficiency. Serum ferritin levels change prior

50

to clinical and morphological manifestations of anemia (e.g. microcytosis) making it a

useful diagnostic marker of iron homeostasis. While considered the most useful of the

currently available tests for non-invasively assessing iron stores, ferritin is also an acute

phase reactant and may be normal or even increased when chronic infection or

inflammation occurs in combination with underlying iron deficiency anemia. In

thalassemias, ferritin is typically elevated reflecting a state of iron overload; in contrast,

ferritin is decreased in iron deficiency making it a useful marker to differentiate causes

of microcytosis.

Transferrin is an iron transporting protein and negative acute phase reactant

produced primarily by the liver. As with ferritin, transferrin is routinely measured by

immunoassay. Most circulating iron is bound to transferrin, binding to Fe3+ with very

high affinity. Transferrin transports iron absorbed in the GI tract to cells containing

specific receptors, in particular erythroid tissue. Transferrin delivers iron to cells via the

ubiquitously distributed transferrin receptor. Clinically, measurement of transferrin is

useful for hypochromic microcytic anemia workups. Transferrin is increased in iron

deficiency anemias, but normal or decreased in chronic anemia of disease, iron

overload, and thalassemias. Transferrin is decreased in cases of liver disease,

nephropathy (or other protein loss or malabsorption), and inflammation.

51

Table 1. Iron Tests in Different Disorders

Disorder Serum

Iron

TIBC %

Saturation

Transferrin Ferritin

Chronic Anemia

of Disease

↓ ↓ ↓ ↔ or ↓ ↔ or ↑

Iron Deficiency ↓ ↑ ↓ ↑ ↓

Thalassemia ↔ or ↑ ↔ ↔ or ↑ ↔ or ↓ ↔ or ↑

Hemochromatosis ↑ ↓ ↑↑ ↔ or ↓ ↑↑

↓decreased; ↔ within reference interval; ↑ increased

3.1.5 Soluble Transferrin Receptor

An additional test that is useful for diagnosis of anemia is the soluble transferrin

receptor (sTfR). The sTfR consists of the N-terminus of the membrane receptor that

can be measured in circulation. Circulating levels reflect the activity of the erythroid

bone marrow, where sTfR levels are decreased in cases of low red cell synthesis (renal

failure and aplastic anemia) and increased in patients with hemoglobinopathies. The

utility of sTfR measurement is that it can differentiate iron deficiency in cases of acute

inflammation because sTfR levels are not affected by inflammatory cytokines. In

thalassemias, sTfR levels are generally increased in proportion to the severity of the

genotype. Despite the apparent advantages, sTfR testing is not widely used and is not

currently standardized.

52

3.1.6 Hepcidin

Discovered in 2000, hepcidin is a hormone involved in iron homeostasis.

Hepcidin is produced by the liver and negatively regulates iron balance by inhibiting

macrophage recycling and decreasing intestinal absorption. Thus, when iron stores are

replete, hepcidin levels are increased and when iron stores are low, hepcidin is

elevated. Similar to ferritin, hepcidin is an acute phase reactant, making interpretation

of circulating levels in patients with inflammation more challenging. At the time of

writing, hepcidin testing was not available commercially. The hepcidin in human iron

stores and its diagnostic implications has been recently reviewed (Kroot JJC, Tjalsma

H, Fleming RE, Swinkels DW. Hepcidin in Human Iron Disorders: Diagnostic

Implications: Clin Chem 2011; 57(12): 1650-1669).

Additional ReadingsFairbanks VF, Klee GG. Biochemical aspects of hematology. In Fundamentals of Clinical Chemistry. Edited by Tietz N. Saunders,1987,789-818.

Guarnone R, Centenara E, Barosi G. Performance characteristics of hemox-analyzer for assessment of the hemoglobin dissociation curve. Haematologica 1995;80:426-430.

Pincus MR and Abraham NZ. Interpreting laboratory results. In: Henry's Clinical Diagnosis and Management by Laboratory Methods (Clinical Diagnosis & Management by Laboratory Methods) Edited by McPherson RA and Pincus MR. 21st Edition.

Higgins T, Beutler E, Doumas BT. Hemoglobin, Iron and Bilirubin. In Tietz textbook of clinical chemistry and molecular diagnostics. Edited by Burtis CA, Ashwood ER, Bruns DE. Elsevier Saunders, 2006,1165-1208.

Marengo-Rowe AJ. Structure-function relations of human hemoglobins. Proc (Bayl Univ Med Cent) 2006;19:239-245.

Mayomedicallaboratories.com/test-catalog. Accessed April 20, 2011.

53

Rees DC, Williams TN, Gladwin MT. Sickle-cell disease. The Lancet. 2010;376:2018-2031.

Steinberg MH. Genetic disorders of hemoglobin oxygen affinity. www.uptodate.com. Accessed April 28, 2011.

Steinberg MH. Unstable hemoglobin variants. www.uptodate.com. Accessed April 28, 2011.

Tietz Textbook of Clinical Chemistry and Molecular Diagnostics. Edited by Burtis CA, Ashwood ER, and Bruns DE. 5th Edition.

Vichinsky EP. Sickle cell trait. www.uptodate.com. Accessed April 28, 2011.

54

Chapter 3Diagnostic Laboratory Methods

3.2 Microcytosis Diane Maennle, MD, and Kimberly Russell, MT (ASCP), MBA

Smaller-than-normal size of Red Blood Cells (RBCs) is defined as microcytosis.

This is quantified by calculating the mean corpuscular volume (MCV) using the following

formula employing the values of hematocrit and RBC count:

MCV = Hematocrit (HCT) X 10 / RBC Count (Million)

In adults, a MCV value of less than 80fL is defined as microcytosis. In pediatric

subjects, the MCV and hemoglobin range distinctly vary with age (Table I).

Table I Age Dependent Mean Hemoglobin and MCV Values1,2,3,4

Age Mean Hemoglobin (g/dL) Mean MCV (fL)

3 to 6 months 11.5 91

6 months to 2 years 12.0 78

2 to 6 years 12.5 81

6 to 12 years 13.5 86

12 to 18 years (female) 14.0 90

12 to 18 years (male) 14.5 88

> 18 years (female) 14.0 90

> 18 years (male) 15.75 90

55

Iron deficiency anemia, α-thalassemia trait, and β-thalassemia trait are the most common

causes of microcytosis. However, other clinical conditions are also associated with microcytosis

(Table II).1,3,5,6 In addition to decreased MCV, the patients with β-thalassemia trait usually have

increased hemoglobin A2. It is pointed out that lower hemoglobin A2 is also observed in patients

with concurrent deficiency of serum iron. Therefore, serum iron deficiency anemia must be ruled

out in order to correctly make the diagnosis of β-thalassemia trait in such patients. Conversely,

patients with β-thalassemia trait may acquire megaloblastic anemia or liver disease, and may

exhibit a normal range for MCV.7

Table II Diagnostic Reasons of Microcytosis (listed in decending order of frequency)

Children and adolescents Menstruating women Men and non-menstruating women

Iron deficiency anemia Iron deficiency anemia Iron deficiency anemia

Thalassemia trait Thalassemia trait Anemia of chronic disease

Other hemoglobinopathies Pregnancy Unexplained anemia

Lead toxicity Anemia of chronic disease Thalassemia trait

Chronic inflammation Sideroblastic anemia

Sideroblastic anemia

Several laboratory tests in addition to the CBC, e.g. serum iron, serum ferritin, total iron-

binding capacity (TIBC), transferrin saturation, hemoglobin electrophoresis, and the examination

of the peripheral blood smears (by a pathologist or hematologist), are employed to provide

insight and etiologies of microcytosis (Table III).3,8

56

Table III Laboratory Tests in the Differential Diagnosis of Microcytosis

Suggested diagnosis

Test Iron deficiency anemia Thalassemia Anemia of chronic disease Sideroblastic anemia

Serum ferritin Decreased Increased Normal to increased Normal to increased

RBC Increased Normal to Normal Increaseddistribution width increased(RDW) Serum iron Decreased Normal to Normal to Normal to increased decreased increased

Total iron- Increased Normal Slightly increased Normalbinding capacity

Transferrin Decreased Normal to Normal to slightly Normal tosaturation increased decreased increased

Van Vranken3 has recently suggested a protocol for diagnosing the cause of microcytosis

(Figure 1). If the cause remains unclear, the diagnosis of hemoglobinopathy by methods besides

electrophoresis alone is imperative. Note: There is a type-setting error in the presentation of

the protocol suggested by Van Vranken.3 We have corrected this error in the figure 1, and

the journal (American Family Physician) editor was also informed.

57

58

Clinical observations of Kenneth F. Tucker, MD, FACP, a practicing hematologist for the last forty years:

Ordinary hemoglobin electrophoresis (cellulose acetate or agarose gel

electrophoresis) was only able to detect the more common types of thalassemias. Although

there were several other types, many of them did not have microcytosis. I had a large

number of patients, who had β-thalassemia minor and a few with probably α-thalassemia,

in which the hemoglobin and hematocrit values were relatively normal. Microcytes may or

may not be present. This diagnosis was suggested by the peripheral smear, and proven by

additional laboratory tests (IFE, globin chain analysis, etc.).

I believe that RDW, which is the average red cell width and reflects standard

deviation of red cell volumes, is a very important test. RDW normal deviation is a bell-

shaped curve. When this value is 2-3% higher, it represents red cells which have varying

widths. This certainly can be seen in patients who are iron deficient with microcytosis, but

have normal or large cells in addition to megaloblastic or dysplastic marrows, elevated

reticulocytes, vitamin B12 or folic acid deficiency, and other conditions. Despite the

availability of automated cell counters, review of the peripheral film is one of the most

informative and rewarding tests that should be done (by pathologist or hematologist) in any

case in which the cause of anemia is not obvious, e.g., bleeding, pure iron deficiency, pure

vitamin B12 deficiency, etc. It is also emphasized that the RDW test is not sensitive or

specific enough to differentiate iron deficiency and β-thalassemia trait.9

A fairly low to extremely low ferritin is an excellent measure of iron deficiency

anemia. In my practice, regardless of what else is going on, any ferritin level of <10 ng/mL,

means there is iron deficiency. As mentioned above (Table III), elevated ferritin levels are

59

seen in refractory anemias, all types of chronic inflammatory conditions, etc. Since this test

is an acute phase reactant (similar to haptoglobin), it must not be used alone, as the ferritin

level may be normal in these clinical conditions.

Women in the second or third trimester are always anemic. This is similar to patients

who are hypervolemic because of renal or cardiac problems. Red cells in these cases are

not microcytes and when the hypervolemia is corrected, the hemoglobin and hematocrit

rises.

Severe anemia in childhood is usually due to the lack of iron in food, since cow’s

milk does not contain iron.

A naïve reader is advised to also review the “Full Color pdf of Complete Blood Count

in Primary Care,” Best Practice Journal, June 2008 (www.bpac.org.nz),

especially the section on Hemoglobin and Red Cell Indices (page 15).

References

1. Richardson M. Microcytic anemia [published correction appear in Pediatr Rev. 2007; 28(7): 275, Pediatric Rev. 2009; 30(5): 181, and Pediatr Rev. 2007; 28(4):151]. Pediatr Rev. 2007; 28(1): 5-14.

2. Beutler E, Waalen J. The definition of anemia: what is the lower limit of normal of the blood hemoglobin concentration? Blood. 2006; 107(5): 1747-1750.

3. Van Vranken ML. Evaluation of Microcytosis. Am Fam Physician. 2010; 80(9): 1117-1122.4. RBC indices calculation and laboratory procedure (2006). St. John Health Laboratories,

Warren, MI 48093.5. Moreno Chulila JA, Romero Colas MS, Gutierrez Martin M. Classification of anemia for

gastroenterologist. World J Gastroenterol. 2009: 15(37):4627-4737.6. Guralnik JM, Eisenstaedt RS, Ferrucci L, Klein HG, Woodman, RC. Prevalence of anemia

in persons 65 years and older in the United States: evidence for a high rate of unexplained anemia. Blood. 2004; 104(8): 2263-2268.

7. Bain BJ. Hemoglobinopathy Diagnosis. 2nd ed. Malden, Mass.: Blackwell Publishing; 2006: 94-106.

60

8. Hematologic diseases. In: Wallach J. Interpretation of Diagnostic Tests. 8th ed. Boston, Mass.: Little Brown and Company; 2006: 385-419.

9. Ntalos G, Chatzinikolaou A, Saouli Z, et al. Discrimination indices as screening tests for beta-thalassemia trait. Ann Hematol. 2007; 86(7): 487-491.

61

Chapter 3 Diagnostic Laboratory Methods

13.3 Hereditary Persistence of Fetal Hemoglobin

Bernard G. Forget, MD

3.3.1 Introduction

Hereditary persistence of fetal hemoglobin or HPFH consists of a group of

genetic disorders characterized by the presence of a substantial elevation of fetal

hemoglobin (Hb F) in RBCs of heterozygotes, as well as of homozygotes and

compound heterozygotes for HPFH and other hemoglobinopathies. Increased levels of

Hb F can ameliorate the clinical course of hemoglobinopathies such as β thalassemia

and sickle cell anemia. HPFH is usually due to deletions of different sizes involving the

β-globin gene cluster, but nondeletion types of disorders have also been identified,

usually due to point mutations in the γ-globin gene promoters (reviewed in refs. 1-3).

Figure 1 diagrammatically illustrates the spatial organization of the β-like globin genes in

the β-gene cluster on chromosome 11. However, as discussed later in this chapter,

certain forms of nondeletion HPFH are clearly not linked to the β-globin gene cluster.

62

Figure 1. Deletions of the β-globin gene cluster associated with fusion proteins and HPFH. The circle 3’ to the β-globin gene indicates the 3’ β-globin gene enhancer. The filled vertical boxes at the 3’ breakpoints of the HPFH-1 and HPFH-6 deletions indicate the locations of DNA sequences with homology to olfactory receptor genes (adopted from reference 2). The references for the individual mutations are cited in references 1, 3 and 6.

HPFH is frequently contrasted with δβ thalassemia, which is another genetic

disorder associated with elevated Hb F levels. However, one should probably not

consider the two disorders as being unambiguously separate entities but rather as a

group of disorders with a variety of partially overlapping phenotypes that sometimes

defy classification as one syndrome or the other. The following is a working definition

that is generally applied to the classification of these disorders: δβ thalassemia usually

refers to a group of disorders associated with a mild but definite thalassemia phenotype

of hypochromia and microcytosis together with a modest elevation of Hb F that, in

heterozygotes, is heterogeneously distributed among red cells. In contrast, HPFH

refers to a group of disorders with substantially higher levels of Hb F, and in which there

is usually no associated phenotype of hypochromia and microcytosis. In addition, the

increased Hb F in heterozygotes with the typical forms of HPFH is distributed in a

relatively uniform (pancellular) fashion among all of the red cells rather than being

distributed in a heterogeneous (heterocellular) fashion among a subpopulation of so-

called F cells, as in δβ thalassemia. Homozygotes for both conditions totally lack Hb A

and Hb A2, indicating absence of δ- and β-globin gene expression in cis to the δβ

thalassemia and HPFH determinants. Although the apparent striking qualitative

difference in cellular distribution of Hb F between HPFH and δβ thalassemia may be

63

due in great part to the quantitative differences in the amount of Hb F per cell and the

sensitivity of the methods used to detect Hb F cytologically, nevertheless it would

appear that the increased amount of Hb F in HPFH is caused by a genetically

determined failure to suppress γ-globin gene activity postnatally in all erythroid cells,

rather than being due to selective survival of the normally occurring sub-population of F

cells such as occurs in sickle cell anemia, β+ and βo thalassemia. Nevertheless,

heterocellular forms of HPFH, without a β-thalassemia phenotype, have been clearly

defined and characterized. Therefore, in the final analysis, there is definitely some

overlap between these two sets of syndromes at the level of their clinical and

hematological phenotypes, as well as at the level of their molecular basis.

3.3.2 Deletions Associated with the HPFH Phenotype.

Classic pancellular HPFH, with absence of δ-and β-globin gene expression from

the affected chromosome, is associated with large deletions in the β-globin gene cluster

that remove the δ-and β-globin genes together with variable amounts of their 5’ and 3’

flanking DNA. At least nine different HPFH deletions of this type have been

characterized that vary in size or length from ~13 kb to ~ 85 kb (1-4), some of which are

illustrated in Fig. 1. The mechanisms by which such deletions cause marked elevation

of Hb F are not well understood, but a number of theories have been proposed.

One theory is based on the model of the proposed mechanism for the marked

elevation of Hb F associated with Hb Kenya. Hb Kenya is a structurally abnormal

hemoglobin that, like Hb Lepore, contains a "hybrid" or fused β-like globin chain

resulting from a non-homologous crossing-over event between two globin genes in the

64

β-gene cluster. However, whereas the Lepore crossover occurred between the δ- and

β-globin genes, the Kenya gene resulted from crossover between the Aγ- and β-globin

genes (Fig. 1). The crossover occurred in the second exons of the Aγ and β genes,

between the codons for amino acids 80 to 87, and resulted in deletion of ~24 kb of DNA

between the Aγ to the β gene. Unlike Hb Lepore, that is associated with a β-

thalassemic phenotype, Hb Kenya is associated with a phenotype of pancellular Gγ

HPFH: erythrocytes of affected heterozygotes have normal red cell indices and contain

7-23% Hb Kenya as well as approximately 10% Hb F, all of which is of the Gγ type and

is relatively uniformly distributed among the red cells. A proposed explanation for the

HPFH phenotype associated with Hb Kenya is the influence on the Gγ- and Kenya gene

promoters of a well characterized enhancer element located in the 3' flanking DNA of

the β-globin gene, shown by the filled circle in Fig. 1, that becomes translocated into

close proximity of the γ-globin gene promoters by the crossover/deletion event, resulting

in enhanced activity of these promoters.

Among the HPFH deletions, there is a relatively short deletion, called HPFH-5 or

Italian HPFH, that extends from a point ~3 kb 5' to the δ gene to a point 0.7 kb 3' to the

β gene, deleting the β gene but not its 3' enhancer. The molecular basis of the HPFH

phenotype associated with this deletion is presumably the influence of the translocated

3' β-gene enhancer on the γ-gene promoters, in a manner analogous to that proposed

for the basis of the HPFH phenotype of the Hb Kenya syndrome. In the case of some of

the other larger HPFH deletions, the DNA preserved at or near the 3’ breakpoint of the

deletions has been shown in various types of assays to have enhancer-like activity on

gene expression (2, 5-7). Thus, it has been proposed that the DNA sequences at the

65

HPFH 3' deletion breakpoints, that become juxtaposed to the γ genes as a result of the

deletion events, may influence γ-gene expression, in a manner analogous to the

presumed influence of the 3' β-gene enhancer on γ-gene expression in Hb Kenya and

HPFH-5. Mechanisms by which this could occur include the presence of enhancer-like

sequences in the translocated 3' breakpoint DNA or the presence in this DNA of an

active chromatin configuration that could have a spreading and activation effect on the

expression of the neighboring γ-globin genes.

A second theory for the mechanism of increased γ-gene expression in deletion-

type HPFH is the nature and function of the DNA sequences conserved at the 5’

breakpoint of the deletions. The 5’ breakpoints of the HPFH deletions, as well as many

of the δβ-thalassemia deletions, are located in the DNA between the Αγ and δ genes,

the so-called Αγδ-intergene DNA. It has long been proposed that there may exist

negative regulatory or silencer elements in this region of DNA, deletion of which in

HPFH but not in δβ thalassemia, results in markedly impaired postnatal suppression of

γ-gene activity in all erythroid cells (8). A number of subsequent observations have

been made that support a role for the Aγδ-intergene region in the regulation of γ-gene

expression (reviewed in ref. 9). The Corfu deletion in particular, involving the δ-gene

and ~6 kb of upstream flanking DNA, is associated in homozygotes with a high HbF

phenotype and removes some interesting structural elements, such as a poly-pyrimidine

region that can serve as a binding site for a multi-protein chromatin remodeling complex

containing the transcription factor Ikaros, and a region of DNA that serves as a promoter

for the synthesis of an intergenic RNA transcript preferentially expressed in adult

66

erythroid cells (10). This region of DNA also appears to serve as a boundary region

between fetal and adult domains of the β-globin gene cluster.

The most conclusive evidence for a functional role of the Aγδ-intergene DNA in

the regulation of γ gene expression consists of the observations by Sankaran et al. who

have extensively characterized a negative regulatory transcription factor, called

BCL11A, that down-regulates γ-gene expression in adult erythroid cells and that binds

to the Aγδ-intergene DNA (11-13). BCL11A, originally identified as an important factor in

B-lymphoid cell development, is a component of a multi-protein complex that plays a

negative regulatory role in γ-gene expression. This complex has been shown to contain

GATA1 as well as all components of the nucleosome and histone deacetylase ( NuRD)-

repressive complex (14). Additional studies have shown that this complex physically

interacts with another transcription factor called SOX6 that is thought to be a repressor

of embryonic and fetal globin gene expression (15). Chromatin immunoprecipitation

(ChIP) studies have shown that BCL11A binds to a number of regions in the β-cluster,

including the upstream locus control region (LCR) and the γδ intergenic region, but does

not bind to the γ- or β-gene promoters (4, 14, 15). Sankaran et al. (4) have

characterized two important deletion mutants with nearly identical distal breakpoints but

different upstream breakpoints around the δ-gene and its flanking DNA. One mutant

with a more proximal breakpoint has a δβ-thalassemia phenotype, whereas the longer

deletion removing 3.5 kb of additional upstream DNA is associated with a HPFH

phenotype. The authors propose that this 3.5 kb region of DNA contains a silencer

element, deletion of which can cause HPFH. This hypothesis is strengthened by the fact

that the deleted region contains one of the prominent binding sites of BCL11A detected

67

in their ChIP experiments. These findings provide very strong evidence for a γ-gene

silencer element in the β-gene cluster that associates with a BCL11A-containing

repressor complex and that this interaction is an important factor in the suppression of

γ-gene expression during the perinatal switch from expression of Hb F to Hb A.

3.3.3 Nondeletion Forms of HPFH

In contrast to the deletional types of HPFH syndromes, where both linked Gγ and

Aγ genes are over expressed, only one or the other γ gene is usually over expressed in

the best characterized nondeletional types of HPFH associated with high levels of

pancellular Hb F expression. However, less well characterized nondeletion forms of GγAγ

HPFH have been described that are associated with relatively low levels of

heterocellular expression of both γ genes. Because of the restricted pattern of γ-globin

gene expression in the Gγ and Aγ forms of nondeletion HPFH, it was assumed that the

mutations in these syndromes must be located near the affected gene and molecular

studies focused initially on the DNA sequence analysis of the over expressed γ genes in

these disorders.

68

Table 1 adopted from reference 2. The one patient studied was doubly heterozygous for Hb A and Hb C. About 20% of Hb F (or 8% of the total Hb) was of the Gγ type, and the Gγ gene in cis to the -175 Aγ mutation carried the -158 C→ T change. The references for the individual mutations are cited in references 1 and 3.

The results of these structural analyses revealed a number of different point

mutations in the promoter region of the over expressed γ gene in individuals with

different types of nondeletion HPFH, as listed in Table 1 (reviewed in refs. 1-3). These

point mutations have clustered primarily in three distinct regions of the 5'-flanking DNA

of the affected γ genes. The first region is located approximately 200 base pairs from

69

the "cap site" or site of transcription initiation of the γ genes (at least five different point

mutations involving single nucleotides between residues -195 to -202 from the cap site).

This region of DNA, which had not previously been suspected of playing a role in the

regulation of γ-gene expression, is very G+C rich and its sequence bears homology to

that of known control elements of other genes that contain the binding site for the

ubiquitous trans-acting protein factor called Sp1. Subsequent studies of the γ-gene

promoters have demonstrated that the -200 region is also a binding site for Sp1 and for

at least one other ubiquitous DNA binding protein.

The second region containing a mutation associated with nondeletion HPFH is

located at position -175. A point mutation (T->C) at this position of either the Gγ or Aγ

gene is associated with a phenotype of pancellular HPFH with high levels of Hb F (15-

25%). This region of DNA is noteworthy because it contains an octanucleotide

sequence that is present in the promoter region of a number of genes and is the binding

site of another ubiquitous trans-acting factor called OCT-1. In addition, the octamer

consensus sequence of the γ-gene promoters is flanked on either side by a consensus

sequence for the hematopoietic-specific transcription factor GATA-1. The point

mutation at position -175 affects the one nucleotide that is present in the partially

overlapping binding sites of both OCT-1 and GATA-1.

The third region affected by a point mutation in nondeletion HPFH is in the area

of a well known regulatory element of globin and other genes: the CCAAT box

sequence. In the γ genes, the CCAAT box is duplicated and the mutation associated

with the Greek Aγ type of nondeletion HPFH is a G->A substitution at position -117, 2

bases upstream of the distal CCAAT box of the Aγ-globin gene promoter. The base

70

change disrupts a pentanucleotide sequence, YYTTGA (Y = pyrimidine), that is highly

conserved immediately upstream of the CCAAT sequence in all animal fetal and

embryonic genes. At least two other mutations involving the CCAAT box of one or the

other γ gene have been reported in other cases of HPFH not associated with large

deletions. The CCAAT box region is known to be the binding site of a number of trans-

acting factors, including the ubiquitous factors CCAAT binding protein (CP1) and

CCAAT displacement factor (CDP) as well as the erythroid-specific factor NF-E3.

The unifying model by which these various mutations are thought to affect

hemoglobin switching proposes that these base changes alter the binding of a number

of different trans- acting factors to critical regions of the γ-gene promoters and thereby

prevent the normal postnatal suppression of γ-gene expression (reviewed in refs. 1,2).

The mutations could prevent the binding of negative regulatory factors or enhance the

binding of positive regulatory factors. Either mechanism could be operative with one

mutation or the other.

3.3.4 HPFH Unlinked to the β-Globin Gene Cluster

A number of studies have identified families in which increased levels of Hb F are

inherited due to a genetic determinant that is unlinked to the β-globin gene cluster.

Genome-wide association studies (GWAS), using co-inheritance of single nucleotide

polymorphisms (SNPs) with elevated levels of Hb F, have subsequently demonstrated

the presence of two different quantitative trait loci (QTLs), unlinked to the β-globin gene

cluster on chromosome 11, that are associated with inheritance of mildly elevated levels

of Hb F, similar to the phenotype seen in Swiss-type heterocellular HPFH (see section

71

above on Nondeletion HPFH). These loci are located on chromosome 2 and 6 (16, 17).

The locus on chromosome 2 corresponds to the site of the gene encoding BCL11A and

its identification led to the elegant studies of Sankaran and co-workers demonstrating

the role of BCL11A in the regulation of γ-gene expression. The locus on chromosome 6

is located between the genes encoding HBS1L and MYB. The mechanism by which this

locus causes elevation of Hb F is thus far poorly understood. Finally, mutations in the

gene on chromosome 19 encoding the erythroid-specific transcription factor EKLF1

have been shown to be associated with a form of HPFH (18, 19). The involved

mechanism is probably through the regulation of BCL11A levels, because it has been

demonstrated that EKLF1 binds to the promoter of the BCL11A gene and regulates the

expression of the gene (20).

3.3.5 Conclusion

Significant insights into the normal regulation of expression of the human β-

globin gene cluster have been obtained by a detailed analysis of a group of disorders

called HPFH. On the basis of this information, several important regulatory elements

have been identified for the normal functioning of the individual genes in the cluster

during the developmental switch from fetal to adult hemoglobin gene expression, as well

as for the abnormal persistent expression of the γ-globin genes in adults with HPFH.

These results provide a more sophisticated understanding of the molecular basis of

these syndromes and point to certain strategies for potential future molecular and

cellular therapies for globin gene disorders.

72

3.3.6 Hemoglobin F Quantification

Hb F can be quantified by several methods, and the most commonly used

procedures in a clinical laboratory are a) radial immunodiffusion, b) Elisa method,

c) HPLC, and d) capillary zone electrophoresis.

References

1. Bollekens JA, Forget BG. Delta beta thalassemia and hereditary persistence of fetal hemoglobin. Hematol Oncol Clin North Am 1991;5(3):399-422.2. Forget BG. Molecular basis of hereditary persistence of fetal hemoglobin. Ann N Y Acad Sci 1998; 850:38-44.3 Weatherall DJ, Clegg JB. The Thalassaemia Syndromes. 4th ed. Oxford ; Malden, MA: Blackwell Science; 2001.4. Sankaran VG, Xu J, Byron R, et al. Functional element necessary for fetal hemoglobin silencing. N Engl J Med 2011; 365(9):807-14.5. Feingold, EA, Forget BG. The breakpoint of a large deletion causing hereditary persistence of fetal hemoglobin occurs within an erythroid DNA domain remote from the β-globin gene cluster. Blood 1989; 74: 2178–2186.6. Kosteas T, Palena A, Anagnou NP. Molecular cloning of the breakpoints of the hereditary persistence of fetal hemoglobin type-6 (HPFH-6) deletion and sequence analysis of the novel juxtaposed region from the 3' end of the beta-globin gene cluster. Hum Genet. 1997;100: 441-5.7. Anagnou NP, Perez-Stable C, Gelinas R, et al. Sequences located 3' to the breakpoint of the hereditary persistence of fetal hemoglobin-3 deletion exhibit enhancer activity and can modify the developmental expression of the human fetal A gamma-globin gene in transgenic mice. J. Biol Chem 1995; 270: 10256-63.8. Huisman TH, Schroeder WA, Efremov GD, et al. The present status of the heterogeneity of fetal hemoglobin in beta-thalassemia: an attempt to unify some observations in thalassemia and related conditions. Ann N Y Acad Sci 1974;232(0):107-24.9. Bank A, O'Neill D, Lopez R, et al. Role of intergenic human γ-δ -globin sequences in human hemoglobin switching and reactivation of fetal hemoglobin in adult erythroid cells. Ann N Y Acad Sci 2005;1054:48-54.

73

10. Chakalova L, Osborne CS, Dai YF, et al. The Corfu δβ thalassemia deletion disrupts γ-globin gene silencing and reveals post-transcriptional regulation of HbF expression. Blood 2005;105:2154-60.11. Sankaran VG, Xu J, Orkin SH. Transcriptional silencing of fetal hemoglobin by BCL11A. Ann N Y Acad Sci. 2010;1202:64-8.12. Sankaran VG, Xu J, Ragoczy T, et al. Developmental and species-divergent globin switching are driven by BCL11A. Nature 2009;460(7259):1093-7.13. Sankaran VG, Nathan DG. Reversing the hemoglobin switch. N Engl J Med 2010; 363(23):2258-60.14. Sankaran VG, Menne TF, Xu J, et al. Human fetal hemoglobin expression is regulated by the developmental stage-specific repressor BCL11A. Science 2008; 322(5909):1839-42.15. Xu J, Sankaran VG, Ni M, et al. Transcriptional silencing of γ-globin by BCL11A involves long-range interactions and cooperation with SOX6. Genes Dev 2010; 24:783-98.16. Thein SL, Menzel S, Lathrop M, Garner C. Control of fetal hemoglobin: new insights emerging from genomics and clinical implications. Hum Mol Genet 2009;18(R2):R216-23.17. Galarneau G, Palmer CD, Sankaran VG, Orkin SH, Hirschhorn JN, Lettre G. Fine mapping at three loci known to affect fetal hemoglobin levels explains additional genetic variation. Nat Genet 2010;42(12):1049-51.18. Borg J, Papadopoulos P, Georgitsi M, et al. Haploinsufficiency for the erythroid transcription factor KLF1 causes hereditary persistence of fetal hemoglobin. Nat Genet 2010;42(9):801-5.19. Borg J, Patrinos GP, Felice AE, Philipsen S. Erythroid phenotypes associated with KLF1 mutations. Haematologica 2011; 96:635-8.20. Zhou D, Liu K, Sun CW, Pawlik KM, Townes TM. KLF1 regulates BCL11A expression and γ- to β-globin gene switching. Nat Genet 2010; 42:742-4.

74

Chapter 3Diagnostic Laboratory Methods

3.4 Flow Cytometry Measurements of Cellular Fetal Hemoglobin, Oxidative Stress and Free Iron in Hemoglobinopathies

Eitan Fibach, MD

3.4.1 Flow Cytometry of Blood Cells

Flow cytometry (FC) is a common methodology in clinical diagnostic and

research laboratories. In hematology, it is mainly used for diagnosis, prognosis,

determining therapy efficacy and follow up of patients with leukemia or lymphoma

(1). It is also used for diagnosis of red blood cell (RBC) abnormalities such as in

Paroxysmal Nocturnal Hemoglobinuria (2) and hereditary spherocytosis (3). In

this review, I will summarize FC methodologies for analysis of RBC (and other

blood cells) from patients with hemoglobinopathies with respect to their fetal

hemoglobin (HbF) and free iron (labile iron pool, LIP) contents and parameters of

the oxidative state.

FC analyzes individual cells in a liquid medium. Most techniques use antibodies

directed against internal (following fixation and premeabilization of the

membrane) or surface antigens. The antibodies are labeled with fluorescence

probes (fluochromes) either directly or indirectly (by a secondary antibody). In

addition to antibodies, other fluorescent compounds can be used. For example,

propidium iodide, which binds stochiometrically to nucleic acids, is commonly

used for determining cell viability and their distribution in the cell cycle (4).

Following staining, the cells are analyzed by a flow cytometer; they are first

75

hydro-dynamically focused in a narrow sheath of physiological solution before

being intercepted by one or more laser beams resulting in light scatter and

fluorescence emission. Depending on the number of laser sources and

fluorescence detectors, several parameters (commonly 6, but up to 18) can be

simultaneously detected on each cell: Forward light scattering and side light

scattering provide correlates with regards to size and granularity of the cells,

respectively, and fluorescence light emission by the fluorochromes correlates

with the expression of different antigens as well as other cellular parameters (see

below).

FC is superior to other techniques in several aspects: (I) Technology is widely available

as mentioned above, most hematology and immunology laboratories use FC for both diagnosis

and research purposes. (II) Only cell-associated fluorescence is measured, excluding soluble or

particulate fluorescence. (III) Each cell is analyzed individually, but since measurement is rapid

(msec), a large number of cells can be analyzed (ranging from 0.1-10 x105 cells) within a few

minutes. The results are therefore statistically sound even for very small sub-populations. (IV)

Various sub-populations can be identified and measured simultaneously. (V) The method

produces mean values for each sub-population, and therefore avoids the inaccuracy of

biochemical methods that produce mean value for the whole population. This is of crucial

importance when mixed populations are studied. (VI) The procedure can be automated to permit

high throughput analysis (e.g., for screening of large libraries of compounds for inducers of

HbF). Although the FC data are expressed in arbitrary fluorescence units rather than weight or

molar concentrations, they are useful for comparative purposes.

76

FC is especially fitting for analysis of blood cells: (I) These cells which can be easily

obtained by blood drawing are present as single cells, thus in contrast to cells of solid tissues,

their use does not require harsh procedures for tissue disaggregation (e.g., trypsinization). (II)

They are present as a mixture of various cell types, including numerous subtypes (e.g.,

lymphocytes), with very large (e.g., RBC) to very small (hematopoietic stem cells)

representation. Cells of these sub-types can be identified and "gated" based on differences in

their size (forward light scattering), granularity (side light scattering) and expression of surface

antigens, and can be measured simultaneously. For measurements of various characteristics

(HbF content, oxidative stress parameters and LIP content), the blood sample is stained with

specific probes (as detailed below), and then with fluorescent reagents (usually antibodies)

against surface markers which identify a specific subpopulation. Such markers are glycophorin

A for RBC, CD61 for platelets, CD15 for neutrophils, CD19 for B-lymphocytes and CD3 for T-

lymphocytes. CD45 is particularly useful since it is differentially expressed on various nucleated

blood cells (Fig. 1).

77

Fig. 1. Flow cytometry of blood cells. A dot plot of blood cells with respect CD45 (FL3-H) and side light scatter (SSC-H).

3.4.2 Measurement of Fetal Hemoglobin-Containing Erythroid Cells

Fetal hemoglobin (HbF, α2γ2) is the major hemoglobin (Hb) in the prenatal period

that is largely replaced after birth by the adult Hb (HbA, α2β2) (5). In adults, less than 1%

of the Hb content is HbF which is concentrated in a few RBC, called F-cells (6). High

levels of HbF are frequently seen in hemoglobinopathies (7). Measurement of HbF (as

well as HbA, sickle hemoglobin, HbS, etc.) can assist in diagnosis and in determining

the efficacy of treatment. HbF can be measured by a variety of techniques. Most of the

techniques measure HbF in lysates prepared from RBC. These techniques include

78

spectrofluorometric measurements following treatment with alkaline (to destroy non-fetal

hemoglobins) and staining with benzidine (8), chromatography (ion-exchange HPLC for

hemoglobins and reverse-phase HPLC for globin chains) (9), as well as immunological

techniques, such as Elisa, based on antibodies against HbF (10). However, quantitative

FC measurement of RBC, fluorescently stained with antibodies to HbF (as well as for

the other hemoglobins), has several advantages. For example, in the differential

diagnosis of Hereditary Persistence of Fetal Hemoglobin (11). This condition

encompasses a heterogeneous group of disorders with marked increased levels of HbF.

Based on the cellular distribution of HbF, they are characterized as pan-cellular, where

all RBCs have increased levels of HbF, albeit not always uniformly so; and hetero-

cellular, where nearly all the HbF is confined to a minor, distinct subpopulation of RBCs.

This important distinction is most reliably ascertained by FC.

Epidemiological studies have indicated that high levels of HbF improve the

clinical symptoms of the underlying disease. In sickle cell anemia not only do HbF-

containing cells have a lower concentration of sickle hemoglobin, but HbF inhibits

polymerization of HbS directly, accounting for the lower propensity of such cells to

undergo sickling (12). In β-thalassemia, elevated HbF should compensate partially for

the deficiency of β-globin chains and reduce the excess of α-globin chains. Several

pharmacological agents have been used to stimulate HbF production (13). Hydroxyurea

(HU) is currently the drug of choice (14). When patients are monitored during HU

treatment by measuring HbF in the hemolysate, an increase is usually observed after 2-

3 months (10). HU acts by a still unknown mechanism on the early erythroid precursors

in the bone marrow. It takes several weeks for HbF to accumulate in the peripheral

79

blood to a quantity that allows differences before and after treatment to become

apparent. Measuring differences in F-RBC by FC may be more sensitive, and

measuring F-reticulocytes (retics) may provide early indication of treatment efficacy

(15): Retics have a very short life-span (1-2 days) compared to mature RBC (120 days

in normal subjects) and therefore measuring peripheral blood F-retics more closely

characterizes the current status of HbF production in the bone marrow. Measuring F-

retics can indicate the efficacy of the drug and/or the patient’s compliance several days

after treatment initiation. Such follow up is very important since about 30% of the

patients are non-responders. It is imperative that such patients be identified as early as

possible and the treatment (that is not without potential risks) be discontinued and

replaced by treatment with another drug (e.g., butyroids).

3.4.3 Staining Protocols for F-RBC and F-Retics (15)

Heparinized blood is washed three times in phosphate buffered saline (PBS). For

fixation, 50μl of the packed cells are resuspended in 10 ml of PBS containing 4%

formaldehyde for 15-min at room temperature under constant agitation in polypropylene

tubes. For permeabilization, the cells are centrifuged for 3 min at 1,500 g, and 2 ml

methanol-acetone are added to the pellet, mixed and incubated for 1-min at room

temperature. The cells are then washed three times and resuspended in PBS to a final

volume of 0.5 ml (10% suspension).

Anti-HbF monoclonal antibodies (the amount depends on the Manufacturer’s

instructions or on a pre-performed titration) are added to 5x106 cells (5 μl of the 10%

suspension) and incubated for 1-hr at 370C, after which the cells are washed in PBS. If

80

the antibodies are fluorochrome-conjugated, the cells are resuspended in PBS and

analyzed directly. In the case of unconjugated antibodies, a secondary antibody

(fluorochrome-conjugated rabbit F(ab’)2 anti-mouse IgG) is added for 30-min at room

temperature. For the F-retic count, the blood cells are double labeled with

phycoerythrin-conjugated antibodies to HbF and thiazol orange, a specific nucleic acid

binding green fluorescence dye.

Following staining, the cells are washed and resuspended in PBS and analyzed by FC.

For "acquisition", the threshold is set on forward light scatter to exclude debris and

platelets. Cells are run at about 1000 cells/sec using logarithmic amplification, and data of

104-105 cells are accumulated. RBC are gated based on their forward light scatter and

side light scatter. When the sample is also stained with thiazol orange, RBC are gated based on

their negative staining with thiazol orange, retics - based on their weak staining (because they

contain remnants of RNA) and nucleated cells (including normoblasts) – based on their intense

staining; HbF is then specifically determined for each cell population (Fig. 2).

81

Fig. 2. Flow cytometry analysis of F-RBC and F-Retics. Blood cells stained with thiazol-orange (T.O) and anti-HbF. A. Forward light scatter (FSC) vs. T.O. RBC (negative T.O staining) and retics (intermediate T.O staining) were gated and their HbF determined (B and C), respectively.

3.4.4 F-Cell Determination for Fetal-Maternal Hemorrhage (FMH) in Pregnant Patients with β-Thalassemia – A Single Case and General Conclusion (16)

F-cell analysis is commonly used to detect fetal-maternal hemorrhage (FMH) –

where fetal RBC enter the maternal blood circulation due to fetal or maternal trauma or

a placental defect (17). These RBC of fetal origin can be distinguished from the

maternal adult RBC by their fluorescence following staining with an antibody to HbF.

82

Recently, in order to increase the sensitivity, reproducibility and accuracy of the assay,

another marker was introduced – carbonic anhydrase (CA) (18). The CA isoenzymes

that are mainly represented by CAI and CAII (19) are fully expressed in the RBC only

after birth (20,21). The "Fetal Cell Count kit" manufactured by IQ Products (Groningen,

the Netherlands), which uses a combination of a murine monoclonal antibody directed

to HbF and a polyclonal antibody to the CAII isoform, has significantly improved this

assay (11,18). Most of the RBC of fetal origin do not express CA but highly express HbF

(CA-HbF++), while RBC in adult blood express CA but do not express HbF (CA+HbF-).

Some adult F-cells which express CA and HbF (CA+HbF+) can be differentiated from

fetal F-cells (CA-HbF++) present in FMH based on the extent of HbF and CA expression.

Until recently, β-thalassemia major was lethal. Improvements in treatment, such as the

introduction of blood transfusions and iron chelation, have considerably improved the life

expectancy as well as the quality of the patient’s life, including the ability of thalassemic women

to give birth. Recently, we were confronted with a case of a possible FMH in a β-thalassemic

woman. To establish the usefulness of the CA/HbF procedure, i.e. differentiating between fetal

RBC and the maternal RBC, we screened non-pregnant β-thalassemic patients (men and

women). The results demonstrated, in addition to adult non-F RBC (CA+HbF-) and adult F-RBC

(CA+HbF+), two other sub-populations, CA+HbF++ and CA-HbF++. The presence in these patients

of the latter RBC phenotype, which characterizes fetal cells, precludes the use of the CA/HbF

method for the detection of FMH in thalassemia.

3.4.5 Oxidative Stress

The oxidative status of cells is determined by the balance between pro-oxidants

and antioxidants. The reactive oxygen species (ROS) are pro-oxidants which are 83

generated in most cells mainly during energy production. Although important for various

aspects of normal physiology (e.g., signal transduction), ROS interact with and damage

various cell components when they are in excess. To protect against the deleterious

effects of ROS, cells maintain an effective antioxidant system consisting of water- or

lipid-soluble antioxidants and enzymes that remove ROS by metabolic conversion.

When the oxidant/anti-oxidant balance is tilted in favor of the oxidants, oxidative stress

ensues (22). Although oxidative stress is not the primary etiology of

hemoglobinopathies, it mediates several of their pathologies, including hemolysis which

results in chronic anemia. Hemolysis occurs both in the bone marrow, where developing

erythroid precursors undergo enhanced apoptosis (ineffective erythropoiesis) and in the

peripheral blood, where mature RBC undergo lysis in the blood vessels (intra-vascular

hemolysis). Destruction also occurs in reticuloendothelial tissues, such as the spleen,

where mature RBC undergo phagocytosis by resident macrophages (extra-vascular

hemolysis) (22).

Various factors are responsible for oxidative stress in RBC of patients with hemo-

globinopathies. In β-thalassemia, excess α-globin chains form unstable tetramers that

dissociate into monomers and then are oxidized, first to met-Hb and then to

hemichromes which precipitate intracellularly with time (23). Following the release of

heme and iron, there is deposition of the protein moiety on the plasma membrane. The

outcome of this chain of events is enhanced formation of ROS, catalyzed by free iron,

with a variety of deleterious effects on the membrane lipids and proteins, including

oxidation of the membrane protein band 4.1 and a decrease in spectrin/band3 ratio (24).

In α-thalassemia, the γ- and β-globins, which are produced in excess, do not precipitate

84

right away, but form the soluble tetramers γ4 (Hb Bart’s) and later the β4 (HbH), which

are less stable than HbA and have an increased susceptibility towards oxidation and

hemichrome formation (23). In sickle cell disease, met-HbS is produced at a higher rate

and is less stable than met-HbA resulting in formation of hemichromes, and release of

heme and iron, with resultant denaturation and precipitation as Heinz bodies (25).

Many approaches have been devised to quantify oxidative stress and its damage

as well as the effects of treatment with anti-oxidants (22). Most of these methods assay

the content of body fluids (mainly blood). FC can be utilized for measurements of

oxidative stress parameters in various blood cells. Although the major target of oxidative

stress in hemoglobinopathies is the RBC, other blood cells are affected as well. Thus,

defects in the abilities of polymorphonuclear cells to adhere to, engulf and lyze bacteria

may result in recurrent infections. Chronic activation of platelets may cause

thromboembolic complications (26,27). In order to study the effects of oxidative stress

on the spectrum of symptoms in hemoglobinopathies, all blood cell lineages should be

studied.

FC of oxidative stress parameters utilizes various probes: ROS can be measured

by staining cells with the non-polar compound, 2’-7-dichlorofluorescein diacetate. It

readily diffuses across the membrane and becomes deacetylated by

esterases into a polar derivative that is trapped inside the cells. When it is oxidized by

ROS (mainly peroxides), a green fluorescent product – dichlorofluorescin is produced

(28). The intensity of the fluorescence is proportional to the cellular concentration of

ROS. The applicability of the method was validated by the increased fluorescence

measured following treatment with ROS-generating agents such as hydrogen peroxide

85

and t-butylhydroxyperoxide and with the catalase inhibitor sodium azide, while treatment

with ROS scavengers such as N-acetyl cysteine decreased the fluorescence. ROS can

also be measured by dihydrorhodamine 123, which freely enters into cells, and after

oxidation by ROS to rhodamine 123 emits a bright red fluorescence (29).

Reduced glutathione (GSH), the main cellular antioxidant, can be measured

using mercury orange (26), which forms fluorescent adducts with GSH via the

sulphydryl group, producing an S-glutathionyl derivative that emits red-orange

fluorescence (30). The probe reacts more rapidly with non-protein thiols, such as GSH,

compared with thiol-containing proteins, allowing specificity under controlled staining

conditions (31). The validity of this method was confirmed by demonstrating that N-

ethylmaleimide, which totally blocks thiol groups, decreased the fluorescence in a dose-

dependent manner. To ascertain that non-protein thiols are being stained, cells were

incubated with diethylmaleate, a specific non-protein thiol-depleting agent. This weak

electrophil of the α,β-unsaturated carbonyl group, which reacts with GSH only in the

presence of glutathione transferase, markedly suppressed the mercury orange

fluorescence, suggesting that GSH was the principle thiol which was stained by the dye

(32). Although there is no direct proof that the probe is specific for GSH, the assay

measures predominantly GSH, since it is the main non-protein thiol constituent of the

cellular thiol pool (33).

Other parameters of oxidative stress measured by FC are membrane lipid

peroxidation – by staining with fluor-DHPE (26), and externalization of

phosphatidylserine (PS) moieties, a marker of damage to the membrane lipid, by

fluorochrome-conjugated annexin-V (34).

86

3.4.6 Staining Protocols for ROS and GSH

ROS Assay – Blood cells are incubated with 2'-7'-dichlorofluorescin diacetate, dissolved

in methanol, at a final concentration of 0.4 mM. After incubation at 37°C for 15 min, the cells are

washed and re-suspended in PBS to the original cell concentration.

GSH Assay - Blood cells are washed with PBS and then spun down. The pellet is incubated for

3 min. at room temperature with 40 M (final concentration) of mercury orange. A 100 M stock

solution of mercury orange is made up in acetone and stored at 4°C. In both cases, cells are

then washed and resuspended in PBS, and analyzed by FC.

Fig. 3 shows FC measurements of ROS and GSH in normal and thalassemic RBC. The

results indicate that thalassemic RBC have higher ROS but lower GSH contents than

normal RBC.

87

Fig. 3. Flow cytometry of ROS and GSH in normal and thalassemic RBC. Blood cells derived from a normal donor (A,C) and a thalassemic patient (B,D) were stained for ROS (A,B) and GSH (C,D) following 1-h pre-incubation with (white) or without (pink) 2 mM H2O2. Histograms of RBC are shown.

3.4.7 Intracellular Free Iron

Another contributor to oxidative stress in cells is excess of iron. Iron overload is

generated in thalassemic or sickle RBC as a result of Hb-instability as discussed above.

In addition, iron accumulates in these diseases as a result of increased absorption from

the intestinal mucosa and by a failure to dispose of excess iron acquired by frequent

therapeutic blood transfusions (35). Moreover, iron-containing compounds (Hb or

88

hemin) that are released during hemolysis can add to the iron load and further

aggravate the hemolysis.

Normally, iron is transported in the circulation bound to transferrin and is

transferred into cells through the surface transferrin-receptor (36). Most of the

intracellular iron is firmly bound to various components such as Hb, heme and

cytochrome C; excess is stored in ferritin (37). In iron overload, serum iron which

exceeds the binding capacity of transferrin is present in the form of non-transferrin

bound iron (38). This iron can be taken up through a transferrin-independent pathway,

to form the cellular unbound "labile iron pool" (LIP) (16). The small fraction of LIP was

suggested as a low molecular weight intermediate or transitory pool between

extracellular iron and cellular firmly-bound iron (39). LIP is redox active and it

participates in generation of free radicals by the Fenton and Haber-Weiss reactions and

consequently in cell and tissue damage (40).

Since iron overload plays an important role in the pathology of transfused

patients with β-hemoglobinopathies, the patients are commonly treated with iron

chelators. The three chelators currently in clinical use are deferioxamine, deferiprone

and deferasirox (41). Evaluation of iron overload is important for assessing its severity

and for determining the efficacy of iron chelation therapy. The parameters usually tested

are serum ferritin protein level and transferrin iron saturation. However, serum ferritin is

an acute phase reactant that may increase by iron-independent factors such as

infection, inflammation and liver disease (42). In addition, serum ferritin levels often fail

to predict impending cardiac iron overload and ensuing cardio-myopathies (43). The

advent of non-invasive proton relaxation assays (by NMR R2* or T2*) of organs has

89

provided a significant advance in monitoring iron overload, although, similarly to serum

ferritin, substantial changes in these parameters are seen only weeks to months after

the initiation of chelator treatment. In addition, these techniques require expensive

instrumentation that is not always available.FC quantification of the LIP content in

various blood cell types overcomes many of these problems.

3.4.8 Staining Protocol for LIP

Cells are washed twice with saline and incubated at a density of 1x106 per ml for 15 min

at 37oC with 0.25 μM Calcein Acetoxymethyl Ester (CA-AM). After wash, the cells are treated

with or without Deferiprone (L1, 100 μM). Fig. 4 shows the results of LIP measurements in RBC.

LIP is defined as the difference between histograms of cells treated or untreated with L1.

90

Fig. 4. Flow cytometry of labile iron pool (LIP) in RBC. Blood cells were loaded with calcein, then washed and treated with or without the iron chelator Deferiprone (L1). Distribution fluorescence (FL1-H) histograms are shown. LIP is defined as the difference between the mean fluorescence channels of histograms of cells treated or untreated with L1.

References

1. Virgo PF, Gibbs GJ. Flow cytometry in clinical pathology. Ann Clin Biochem 2012; 49(Pt 1): 17-28.

2. Sutherland DR, Keeney M, Illingworth A. Practical guidelines for the high-sensitivity detection and monitoring of paroxysmal nocturnal hemoglobinuria clones by flow cytometry. Cytometry B Clin Cytom 2012; 82(4): 195-208.

3. Kedar PS, Colah RB, Kulkarni S, Ghosh K, Mohanty D. Experience with eosin-5'-maleimide as a diagnostic tool for red cell membrane cytoskeleton disorders. Clin Lab Haematol 2003; 25(6): 373-6.

4. Krishan A. Rapid flow cytofluorometric analysis of mammalian cell cycle by propidium iodide staining. J Cell Biol 1975; 66(1): 188-93.

91

5. Peterson KR. Hemoglobin switching: new insights. Curr Opin Hematol 2003; 10(2): 123-9.

6. Boyer SH, Belding TK, Margolet L, Noyes AN. Fetal hemoglobin restriction to a few erythrocytes (F cells) in normal human adults. Science 1975; 188(4186): 361-3.

7. Bunn H, Forget B. Hemoglobins: Molecular, Genetic and Clinical Aspects. Philadelphia: WB Saunders Co.; 1986.

8. Fibach E. Measurement of total and fetal hemoglobin in cultured human erythroid cells by a novel micromethod. Hemoglobin 1993; 17(1): 41-53.

9. Huisman TH. Separation of hemoglobins and hemoglobin chains by high-performance liquid chromatography. J Chromatogr 1987; 418: 277-304.

10. Epstein N, Epstein M, Boulet A, Fibach E, Rodgers GP. Monoclonal antibody-based methods for quantitation of hemoglobins: application to evaluating patients with sickle cell anemia treated with hydroxyurea. Eur J Haematol 1996; 57(1): 17-24.

11. Leers MP, Pelikan HM, Salemans TH, Giordano PC, Scharnhorst V. Discriminating fetomaternal hemorrhage from maternal HbF-containing erythrocytes by dual-parameter flow cytometry. Eur J Obstet Gynecol Reprod Biol 2007; 134(1): 127-9.

12. Benesch RE, Edalji R, Benesch R, Kwong S. Solubilization of hemoglobin S by other hemoglobins. Proc Natl Acad Sci U S A 1980; 77(9): 5130-4.

13. Gambari R, Fibach E. Medicinal chemistry of fetal hemoglobin inducers for treatment of beta-thalassemia. Curr Med Chem 2007; 14(2): 199-212.

14. Steinberg MH. Determinants of fetal hemoglobin response to hydroxyurea. Semin Hematol 1997; 34(3 Suppl 3): 8-14.

15. Amoyal I, Fibach E. Flow cytometric analysis of fetal hemoglobin in erythroid precursors of beta-thalassemia. Clin Lab Haematol 2004; 26(3): 187-93.

16. Prus E, Fibach E. Heterogeneity of F-cells in β -thalassemia. Transfusion 2012, in press.17. Sebring ES, Polesky HF. Fetomaternal hemorrhage: incidence, risk factors, time of

occurrence, and clinical effects. Transfusion 1990; 30(4): 344-57.18. Porra V, Bernaud J, Gueret P, Bricca P, Rigal D, Follea G, Blanchard D. Identification

and quantification of fetal red blood cells in maternal blood by a dual-color flow cytometric method: evaluation of the Fetal Cell Count kit. Transfusion 2007; 47(7): 1281-9.

19. Tashian RE. The carbonic anhydrases: widening perspectives on their evolution, expression and function. Bioessays 1989; 10(6): 186-92.

20. Brady HJ, Edwards M, Linch DC, Knott L, Barlow JH, Butterworth PH. Expression of the human carbonic anhydrase I gene is activated late in fetal erythroid development and regulated by stage-specific trans-acting factors. Br J Haematol 1990; 76(1): 135-42.

21. Aliakbar S, Brown PR. Measurement of human erythrocyte CAI and CAII in adult, newborn, and fetal blood. Clin Biochem 1996; 29(2): 157-64.

22. Fibach E, Rachmilewitz EA. The role of antioxidants and iron chelators in the treatment of oxidative stress in thalassemia. Ann N Y Acad Sci 2010; 1202: 10-6.

23. Rachmilewitz EA. Formation of hemichromes from oxidized hemoglobin subunits. Ann N Y Acad Sci 1969; 165(1): 171-84.

24. Advani R, Sorenson S, Shinar E, Lande W, Rachmilewitz E, Schrier SL. Characterization and comparison of the red blood cell membrane damage in severe human alpha- and beta-thalassemia. Blood 1992; 79(4): 1058-63.

92

25. Winterbourn CC. Oxidative denaturation in congenital hemolytic anemias: the unstable hemoglobins. Semin Hematol 1990; 27(1): 41-50.

26. Amer J, Fibach E. Oxidative status of platelets in normal and thalassemic blood. Thromb Haemost 2004; 92(5): 1052-9.

27. Amer J, Fibach E. Chronic oxidative stress reduces the respiratory burst response of neutrophils from beta-thalassaemia patients. Br J Haematol 2005; 129(3): 435-41.

28. Bass DA, Parce JW, Dechatelet LR, Szejda P, Seeds MC, Thomas M. Flow cytometric studies of oxidative product formation by neutrophils: a graded response to membrane stimulation. J Immunol 1983; 130(4): 1910-7.

29. Rothe G, Oser A, Valet G. Dihydrorhodamine 123: a new flow cytometric indicator for respiratory burst activity in neutrophil granulocytes. Naturwissenschaften 1988; 75(7): 354-5.

30. O'Connor JE, Kimler BF, Morgan MC, Tempas KJ. A flow cytometric assay for intracellular nonprotein thiols using mercury orange. Cytometry 1988; 9(6):529-32.

31. Hedley DW, Chow S. Evaluation of methods for measuring cellular glutathione content using flow cytometry. Cytometry 1994; 15(4): 349-58.

32. Plummer JL, Smith BR, Sies H, Bend JR. Chemical depletion of glutathione in vivo. Methods Enzymol 1981; 77: 50-9.

33. Di Simplicio P, Cacace MG, Lusini L, Giannerini F, Giustarini D, Rossi R. Role of protein -SH groups in redox homeostasis--the erythrocyte as a model system. Arch Biochem Biophys 1998; 355(2): 145-52.

34. Freikman I, Amer J, Ringel I, Fibach E. A flow cytometry approach for quantitative analysis of cellular phosphatidylserine distribution and shedding. Anal Biochem 2009; 393(1): 111-6.

35. Rund D, Rachmilewitz E. Beta-thalassemia. N Engl J Med 2005; 353(11): 1135-46.36. Richardson D R, Ponka P. The molecular mechanisms of the metabolism and

transport of iron in normal and neoplastic cells. Biochimica et Biophysica Acta 1997; 1331(1): 1–40.

37. Konijn AM. Iron metabolism in inflammation. Baillieres Clin Haematol 1994; 7(4): 829-49.38. Breuer W, Hershko C, Cabantchik ZI. The importance of non-transferrin bound iron in

disorders of iron metabolism. Transfus Sci 2000; 23(3): 185-92.39. Jacobs A. Low molecular weight intracellular iron transport compounds. Blood 1977;

50(3): 433-9.40. Cabantchik ZI, Kakhlon O, Epsztejn S, Zanninelli G, Breuer W. Intracellular and

extracellular labile iron pools. Advances in Experimental Medicine and Biology 2003; 509: 55–75.

41. Cappellini MD, Piga A. Current status in iron chelation in hemoglobinopathies. Curr Mol Med 2008; 8(7): 663-74.

42. Kalantar-Zadeh K, Kalantar-Zadeh K, Lee GH. The fascinating but deceptive ferritin: to measure it or not to measure it in chronic kidney disease? Clin J Am Soc Nephrol 2006; 1 Suppl 1: S9-18.

43. Wood JC. Cardiac iron across different transfusion-dependent diseases. Blood Rev 2008;22 Suppl 2: S14-21.

93

44. Davis BH, Olsen S, Bigelow NC, Chen JC. Detection of fetal red cells in fetomaternal hemorrhage using a fetal hemoglobin monoclonal antibody by flow cytometry. Transfusion 1998; 38(8): 749-56.

45. Dziegiel MH, Nielsen LK, Berkowicz A. Detecting fetomaternal hemorrhage by flow cytometry. Curr Opin Hematol 2006; 13(6): 490-5.

46. Kleihauer E, Braun H, Betke K. [Demonstration of fetal hemoglobin in erythrocytes of a blood smear]. Klin Wochenschr 1957; 35(12): 637-8.

47. Navenot JM, Merghoub T, Ducrocq R, Muller JY, Krishnamoorthy R, Blanchard D. New method for quantitative determination of fetal hemoglobin-containing red blood cells by flow cytometry: application to sickle-cell disease. Cytometry 1998; 32(3): 186-90.

94

Chapter 3Diagnostic Laboratory Methods

3.5 Solid Phase Electrophoretic SeparationRita Ellerbrook, PhD, and Zia Uddin, PhD

3.5.4 Introduction

Electrophoresis is defined as the movement of charged molecules (e.g. proteins)

under an electrical field, either through a solution (moving boundary electrophoresis) or

through a semi-solid material embedded in a buffer (zone or solid phase

electrophoresis). Historically, the first hemoglobin variant (HbS) identification using

moving boundary electrophoresis was achieved by Professor Linus Pauling1 in 1949 at

the University of Chicago, Chicago, Illinois. Subsequently the moving boundary

electrophoresis due to experimental difficulties was replaced by solid phase

electrophoretic methods, e.g., cellulose acetate, agarose, and agar, etc.

In view of the convoluted three-dimensional structure of the hemoglobin

molecule, even a single genetic mutation, resulting in the substitution of an amino acid

in the globin chain (e.g. the substitution of the amino acid valine for glutamic acid in the

sixth position of the β-chain of hemoglobin molecule) may result in the change of the

secondary/tertiary structure of the hemoglobin molecule/the net charge on the molecule.

This change in the shape/net charge of the hemoglobin molecule is sufficient to modify

its electrophoretic mobility (movement under an electric field), and thus is

advantageously employed for the separation and identification of the hemoglobin

variants. The migration and the identification of hemoglobin variants in solid phase

95

electrophoretic methods are accomplished at alkaline pH (8.6) and acid pH (5.6), and

the commonly used solid phases for this purpose are described here.

3.5.2 Cellulose Acetate Electrophoresis (alkaline pH)

Cellulose upon treatment with acetic anhydride converts into cellulose acetate by

virtue of the acetylation of the hydroxyl groups. The separation characteristic of

cellulose acetate depends on the degree of acetylation reaction and other variables,

e.g., additives used, prewashing procedure utilized by the manufacturer, pore size,

thickness of the membrane, etc. Historically, cellulose acetate electrophoresis (CAE)

was used worldwide in view of the speed of separation, ability to make the membrane

transparent for the quantification of bands by densitometry, ability to store the

transparent membranes for longer periods (plastic backed cellulose acetate plates), no

need for controlled lower temperature for the electrophoresis, low cost, etc. Under the

electrophoretic conditions of pH 8.6, the ionizable groups (e.g. carboxyl group) are

negatively charged thus rendering a negative charge on the hemoglobin molecule. The

relative migration of the hemoglobin towards the anode is dependent on the net

negative charge on the hemoglobin molecule.

CAE laboratory procedure and information about the required hardware and

consumables can be obtained from Helena Laboratories, Beaumont, Texas, USA

(www.helena.com).

96

Fig 1. Computer simulated cellulose acetate electrophoresis of adult hemoglobins (pH 8.6)

In Figure 1, separation of a few hemoglobin variants by CAE is illustrated. This is a

computer simulation of the separation of hemoglobins. Generally in all electrophoretic

separations, a commercially prepared “AFSC” control is used to designate the migration

position of the unknown. Hb S, Hb D, Hb Lepore and Hb G migrate in approximately the

same position, therefore further confirmation of the hemoglobin variant is achieved by

additional laboratory tests, e.g., solubility test and citrate agar electrophoresis at pH 5.6

(see below). In case the hemoglobin variant is not identified by these preliminary

laboratory tests, the laboratory employs other procedures, e.g., HPLC, IEF, and DNA

97

studies. The same procedure is also followed about the co-migration of Hb C, Hb E,

and Hb O-Arab upon CAE.

3.5.3 Agarose Gel Electrophoresis (alkaline pH)

Agar is a gelatinous material prepared from certain marine algae, and is a

mixture of agarose and sulfated polysaccharides contaminants called

agaropectin. The highly purified agar (neutral fraction of agar) that is almost free

of agaropectin (ionizable groups like sulfate and carboxylic) is called agarose.

Agarose gel electrophoresis (AGE) at alkaline pH 8.6 is the widely used clinical

laboratory method for the identification of hemoglobin variants. The reason for the

popularity of AGE is due to the lower affinity of agarose for proteins, ability to exhibit

decreased endosmosis, and also the transparency of the film after drying which allows

quantification of the hemoglobin molecule by densitometry. It is emphasized that

hemoglobinopathy is never determined alone by AGE (alkaline pH 8.6), as is the case

with CAE. The resolution of atypical bands or a band co-migrating at the positions of

commonly encountered bands upon AGE (e.g., HbA2, HbS, etc.) is accomplished by

additional laboratory tests.

Currently the AGE reagents, separation gels, and Peltier cooling device (which

cools the gel during electrophoresis) are supplied by two major manufacturers (Sebia,

France, and Helena Laboratories, USA). Sebia’s hemoglobin AGE kit (Hydragel) is used

in conjunction with their semi-automated HYDRASYS System. Helena Laboratories,

USA is a pioneer in supplying AGE kits for >35 years. The Helena’s QuickGel method

available in manual mode is ideal for smaller volume clinical laboratories, and the same

98

plate form is used in the semi-automated instruments (SPIFE 2000 and SPIFE 3000) for

handling a larger volume of testing. Helena’s fully automated instrument (SPIFE 4000)

utilizes a different plate form than QuickGel. Detailed information about AGE

procedures of these two manufacturers can be obtained from their web site

(www.sebia.com and www.helenalaboratories.com).

In Fig 2 we have presented the computer simulation of the electrophoretic

mobilities of the commonly used “AFSC” control and few hemoglobin bands obtained

from AGE at alkaline pH.

Fig 2. Computer simulation of hemoglobin agarose gel electrophoresis bands

3.5.4 Agar Electrophoresis (acid pH)

Agar electrophoresis (AE) at acid pH (5.6-6.2) for the identification/confirmation

of hemoglobins has been widely used for > 40 years. Agarose and agaropectin are the

two main components of agar. Both the electrophoresis and electroendosmotic flow

99

principles are involved in the separation of hemoglobins by AE. Citrate buffer is usually

used for the electrophoretic purpose (Beckman-Coulter uses maleate buffer in their

Paragon kit), therefore it is also called citrate agar electrophoresis. Commercially the AE

kits (plates, reagents, consumables, etc.) are also available from Sebia, France

(HYDRAGEL ACID HEMOGLOBIN) and Helena Laboratories, USA (Titan Gel and

QuickGel). In both cases, hemoglobin “AFSC” control is used to confirm the

electrophoretic mobility of the unknown (i.e. Hb S, Hb C, Hb E, etc.). Quantification of

the bands is not required and the electrophoregrams are evaluated visually. Laboratory

procedures for AE by Sebia and Helena Laboratories can be obtained from their web

sites (www.sebia.com and www.helenalaboratories.com). In Fig 3, we have presented a

computer simulation of an electrophoregram of the AE.

3.5.5 Interpretation of Hemoglobin Agarose Gel (pH 8.6) and Agar Gel (pH 5.6) Electrophoresis

The commercially available control that consists of a mixture of Hb A, Hb F, Hb

S, and Hb C serves to set the framework upon which the various hemoglobin variant

mobilities are compared. This combination of hemoglobins is run on each

electrophoretic plate and the interpretation is aided by comparing the mobility of the

variant to these hemoglobins in the control material. By assigning the distance from

HbA to HbC an arbitrary distance unit of 10 (under either acid or alkaline conditions), a

relative number may be assigned to any hemoglobin.

Schneider and Barwick2 presented this system of hemoglobin typing and

provided a chart of the relative mobilities of all the hemoglobins fully characterized at

100

that time. This chart provided preciseness to the characterization not before possible.

Ellerbrook and Matthews3 at Helena Laboratories felt that since the process was a

visual one, therefore a quicker way to examine these relative mobilities was to convert

them into a chart as depicted in Appendix I.

It will not be out of place to mention here that for >30 years in the Hemoglobin

Laboratories of Henri Mondor Hospital (Creteil, France), Professor Henri Wajcman and

associates have organized a database of the electrophoretic mobility of > 400 Hb

variants, using a similar format to that proposed by Schneider and Barwick. The

Wajcman group included in their database the results of a) IEF on polyacrylamide gel,

b) electrophoresis on cellulose acetate at alkaline pH, c) citrate agar electrophoresis, d)

electrophoresis of dissociated globin chains in 6M urea at pH 6.0 and 9.0 or in the

presence of Triton X-1004.

An excerpt of eight hemoglobins from the chart developed by Ellerbrook and

Mathews3 is shown in Fig 3 for instructional purposes.

Fig 3. Combined agarose gel (pH 8.6), and citrate agar (pH 5.6) electrophoretic pattern presentation for instructional purposes.

The area labeled “Alkaline” on the left side of this figure depicts the mobility of 101

the named hemoglobins under alkaline conditions. The perpendicular lines

represent the relative mobility of Hb C (-10), Hb S (-5.2), Hb F (-2.6), Hb A (0),

Hb J Toronto (4.5) and Hemoglobin I (8.5). In Fig 3 Hb C is seen to have the

least mobility in an alkaline electric field and is depicted squarely on the line.

Each of the control hemoglobins (e.g. AFSC) will also place squarely on the line.

Hemoglobin I and J are extremely rare and their actual presence on the gels as a

control is not necessary because this grid is all about spacing. The gel will have

hemoglobins A, F, S, and C on it to establish spacing. The distance between Hb

S and Hb A is slightly more than the distance from Hb A to Hb J Toronto, and this

distance is slightly less than the distance from Hb J Toronto to Hb I. While

looking at the actual gel the mobility is not a depiction of the leading edge of the

migration but rather the bulk of that hemoglobin band. Denatured hemoglobins

usually run faster than the native form and therefore the leading edge may be a

function of the age of the sample. The sample application in alkaline conditions is

to the left of Hb C, and most hemoglobins at this pH migrate in the same

direction (left side).

The right side of the figure has the similar approach to the mobility under acidic

conditions. The order of migration is different and the direction is reversed. Here the

sample is applied between Hb S and Hb C. Under acidic conditions Hb F is the fastest

moving hemoglobin. The distance from Hb A to Hb C is assigned a new relative

distance of 10. Hb F is assigned the number -4.4, and Hb S is assigned +5.8. The +/-

sign is relative to the Hb A value of 0 and not due to distance from the application point.

102

There is no allowance for fast hemoglobins under acidic conditions because there are

none. At a pH of 6.2 or less, fast hemoglobins migrate like Hb A.

Looking at the left side of the chart, Hb J Baltimore migrates slower than Hb J

Toronto because the bulk of the hemoglobin has not moved as far as one might expect

Hb J Toronto to move. Acid electrophoresis is of no assistance in this case because

these fast hemoglobins do not migrate differently, and thus all end up lined with Hb A.

The mobility of Hb C Siriraj is not different from Hb C under acidic conditions, but can be

differentiated under alkaline pH. In this case since the alkaline separation would have

been done first the only apparent observation would be the presence of an abnormal

hemoglobin band migrating between Hb F and Hb S. Very few hemoglobins migrate like

hemoglobin S so this second test is very useful in narrowing down the possible identity

of this variant. The chart in Appendix I contains the relative mobilities of 165

hemoglobins. The most common variants were discovered first so this chart should

encompass the relative mobilities of most of the hemoglobins found.

3.5.6 Requirements for the identification of complex hemoglobinopathies

Age, sex, ethnicity, ethnic background of biological parents, blood transfusion

(past three months), CBC with differential, serum iron, TIBC, ferritin, treatment status

(immunotherapy), laboratory results of AGE and AE electrophoresis, capillary zone

electrophoresis, high pressure liquid chromatography, isoelectric focusing, quantitative

results of Hb A2, and Hb F, hemoglobin stability, and globin chain analysis. The

significance of all these parameters shall be obvious from the case studies mentioned

later on in the book.

103

References

1. Pauling L, Itano HA, Singer SH, Wells IC. Sickle cell anemia, a molecular basis disease. Science 1949; 110: 53.2. Barwick RC, Schneider RG. The computer-assisted differentiation of hemoglobin

variants, in Human Hemoglobins and Hemoglobinopathies:A review to1981. Texas Reports on Biology and Medicine 1980-81; 40:143-156

3. Helena Laboratories, Beaumont, Texas, USA4. Wajcman H. Electrophoretic Methods for Study of Hemoglobins. Methods in

Molecular medicine, vol 82: Hemoglobin Disorders: Molecular methods and Protocols, Edited by: Ronald L. Nagel, Humana Press Inc., Totowa, NJ.

104

Appendix 1

Appendix 1 continued next page105

Appendix 1 continued

106

Chapter 3 Diagnostic Laboratory Methods

3.6 Capillary Zone ElectrophoresisZia Uddin, PhD

3.6.1 Introduction

During the last five decades separation science has witnessed unparallel growth.

Chromatography and electrophoresis are the main techniques that are routinely used

worldwide for the separation, identification, and quantification of analytes in clinical

laboratories. Capillary zone electrophoresis (CZE) played a significant role in the

completion of the human genome project. Introduction of CZE instruments by Beckman-

Coulter, Sebia, and Helena Laboratories not only automated but also increased the

sensitivity, specificity and reproducibility of the clinical laboratory procedures (e.g.,

serum protein electrophoresis, immunotyping, hemoglobin variant identification for both

the adult and newborn). Besides references listed at the end of this section, the

interested reader is advised to also review the online literature on CZE (e.g., Righetti,

PG, and Guttman, A. 2012 Capillary Electrophoresis. eLS.)

3.6.2 Basic Principle

In simple terms CZE is a liquid flow electrophoresis in which buffer has replaced

the solid support medium (e.g., agarose gel), and the separation occurs due to the

interaction of the analyte with the pH of the buffer. For this reason initially CZE was also

called “Free Solution Capillary Electrophoresis.” In Figure 1, a pictorial illustration of

CZE principle is presented.

107

Figure 1. Capillary Zone Electrophoresis Principle

In CZE two independent phenomena occur, i.e., a) migration of negatively

charged ions toward the positively charged electrode, and b) interaction of the positive

charges from the buffer and the negative charges from the capillary wall leading to

electro-osmotic flow (EOF) from the anode to the cathode. Both of these two processes

(electrophoretic mobility and EOF) can take place at the same time working in opposite

direction thus providing greater resolution.

108

Automated CZE instrument (Figure 2) consists of the following:

a) Cathodeb) Anodec) Power supply to generate high voltage (10,000 volts)d) Catholyte (buffer solution at the cathode end)e) Anolyte (buffer solution at the anode end)f) Capillary facilitated with a cooling deviceg) Detector (415 nm for hemoglobins)h) Computer for data handling and storage

Figure 2. Capillary Zone Electrophoresis Instrument Components

3.6.3 Application of CZE in Diagnostic Hemoglobinopathies

109

Hemoglobin variants can be separated on CZE as is the case with other proteins.

This method is the most advanced and dedicated alternative to the classic alkaline and

acid electrophoresis and the more sophisticated IEF. Chromatography, the separation

alternative on column, has developed from separation on size, charge and hydrophobic

interaction to the modern dedicated high performance liquid chromatography (HPLC),

as we know today. Both of these dedicated methods (CZE and HPLC) have the

advantage of using minimal amounts of material, of providing a separation in a matter of

minutes, with high reproducibility and sensitivity and above all they are able to measure

virtually all fractions including those present at low levels but essential for the diagnosis

or hemoglobinopathy. In addition these two methods may complement each other up to

a certain extent compensating for specific errors.

3.6.4 Interpretation of CZE Results

The migration time of the hemoglobin variant (since the inception of the injection

of the sample and the moment a specific molecule is detected) is divided into fifteen

(15) zones (Table 1). It is obvious that > 1000 hemoglobin variants cannot be separated

in 15 zones. However, the most common one (e.g., Hb S, C, E, and D) will be putatively

identified by their zone with a specificity >90%. Table 1 shows that there is an overlap of

hemoglobin variants in a particular zone (Z1 – Z15). This limitation of CZE is similar to

that experienced with other techniques employed for the identification of hemoglobin

variants, e.g., HPLC (Szuberski J, Oliveira JL, Hoyer JD. A comprehensive analysis of

hemoglobin variants by high performance liquid chromatography. Int J Lab Hematol

2012; 34: 594-604).

110

In Figure 3 we have presented a CZE scan of the most commonly used “AFSC”

control in the clinical laboratory, which illustrates the position of HbA,

HbF, HbS, and HbC peaks corresponding to their respective zones. Later on in this

book (case studies) we have also presented the CZE scan of the hemoglobin variant for

each case.

One drawback of CZE is the assignment of the migration position of the

hemoglobin bands into Z1-Z15 (Table 1) in cases when the HbA / Hb A2 are absent in

the specimen of interest, e.g. Hb S-C disease. This drawback is due to the shifting of

band positions in the absence of Hb A / Hb A2. This limitation of CZE is avoided by

mixing (1:1 ratio) the specimen devoid of HbA/HbA2 with a specimen containing Hb A,

and performing the CZE test thus achieving the relevant migration position and zoning

(Z1-Z15).

111

Table 1. Hemoglobin Zones of CZE: Z1 – Z15

N ° zone Hb Variants Library – Software release 8.60

1

Hb Santa Ana free alpha chain, Hb Mizuho (minor peak), Hb delta A'2, Hb alpha A2, Hb T-Cambodia, "Savaria" Hb A2 variant, "Chad" Hb A2 variant, "Arya" Hb A2 variant, "Hasharon" Hb A2 variant, "Fort de France" Hb A2 variant, "Ottawa" Hb A2 variant, "Shimonoseki" Hb A2 variant, "Stanleyville II" Hb A2 variant, "O-Indonesia" Hb A2 variant, "G-Norfolk" Hb A2 variant, "San Antonio" Hb A2 variant, "Handsworth" Hb A2 variant, "Matsue-Oki" Hb A2 variant, "Memphis" Hb A2 variant, "Q-Iran" Hb A2 variant, "G-Waimanalo" Hb A2 variant, "Russ" Hb A2 variant, "Q-India" Hb A2 variant, "Montgomery" Hb A2 variant, "Watts" Hb A2 variant, "G-Pest" Hb A2 variant, "Winnipeg" Hb A2 variant, "Queens" Hb A2 variant, "Inkster" Hb A2 variant, "Chapel Hill" Hb A2 variant, "Q-Thailand" Hb A2 variant, "Park Ridge" Hb A2 variant

2 = Z(C)

Hb C, Hb F-Hull, Hb F-Texas-I, Hb Constant Spring, Hb C-Harlem (C-Georgetown), "Boumerdes" Hb A2 variant, "Bassett" Hb A2 variant, "Tarrant" Hb A2 variant, "Manitoba I" Hb A2 variant, "St. Luke's" Hb A2 variant, "Setif" Hb A2 variant, "Sunshine Seth" Hb A2 variant, "Titusville" Hb A2 variant, "Swan River" Hb A2 variant, "Manitoba II" Hb A2 variant, "Val de Marne" Hb A2 variant

3 = Z(A2)

Hb A2, Hb Chad (E-Keelung), Hb O-Arab, Hb E-Saskatoon, "Dallas" Hb A2 variant*, "Toulon" Hb A2 variant*, "Bonn" Hb A2 variant*, "Chicago" Hb A2 variant*, "Fontainebleau" Hb A2 variant*, "Hekinan" Hb A2 variant*, "Mosella" Hb A2 variant*, "Aztec" Hb A2 variant*, "Frankfurt" Hb A2 variant*, "M-Boston" Hb A2 variant*, "Owari" Hb A2 variant*, "Twin Peaks" Hb A2 variant*, "Conakry" Hb A2 variant*, "Gouda" Hb A2 variant*, "Jura" Hb A2 variant, "Nouakchott" Hb A2 variant

4 = Z(E)Hb E, Hb Seal Rock, Hb Köln (Ube-1), Hb Buenos Aires (minor peak), Hb M-Saskatoon (minor peak), Hb A2-Babinga, Hb G-Siriraj, Hb Agenogi, Hb Sabine, Hb Santa Ana, Hb Savaria, "M-Iwate" Hb A2 variant, "Wayne" Hb A2 variant (peak 1), Denatured Hb C

5 = Z(S)

Hb S, Hb Arya, Hb Hasharon (Sinai), Hb Dhofar (Yukuhashi), Hb Shimonoseki (Hikoshima), Hb O-Indonesia (Buginese-X), Hb Ottawa (Siam), Hb Fort de France, Hb Montgomery, Hb G-Copenhagen, Hb S-Antilles, Hb Handsworth, Hb S-Oman (peak 2), Hb Hamadan, Hb Russ, Hb Stanleyville II, "Lombard" Hb A2 variant, "Tatras" Hb A2 variant, "Cemenelum" Hb A2 variant, "Jackson" Hb A2 variant, "Hopkins-II" Hb A2 variant, "J-Broussais" Hb A2 variant (alpha 2), Denatured Hb O-Arab

6 = Z(D)

Hb D, Hb Memphis, Hb Leiden, Hb Muravera, Hb D-Bushman, Hb G-Norfolk, Hb S-Oman (peak 1), Hb Matsue-Oki, Hb Osu Christiansborg, Hb D-Punjab (D-Los Angeles), Hb G-Waimanalo (Aida), Hb Muskegon, Hb D-Ibadan, Hb Buenos Aires (minor peak), Hb Q-India, Hb Lepore (Lepore-BW), Hb Q-Iran, Hb Summer Hill, Hb G-Philadelphia, Hb D-Ouled Rabah, Hb Yaizu, Hb San Antonio, Hb Watts, Hb Ferrara, Hb Köln (Ube-1), Hb Fort Worth, Hb Korle-Bu (G-Accra), Hb G-Taipei, Hb D-Iran, Hb St. Luke's, Hb G-Coushatta (G-Saskatoon), Hb Inkster, Hb Winnipeg, Hb Zürich, Hb G-Pest, Hb Queens (Ogi), Hb Setif, Hb P-Nilotic, Hb Sunshine Seth, Hb Titusville, "Le Lamentin" Hb A2 variant, "J-Meerut" Hb A2 variant, "J-Rajappen" Hb A2 variant, "J-Anatolia" Hb A2 variant, "J-Oxford" Hb A2 variant, "Ube 2" Hb A2 variant, "J-Broussais" Hb A2 variant (alpha 1), "J-Toronto" Hb A2 variant, "Mexico" Hb A2 variant, "J-Tongariki" Hb A2 variant, "Neuilly-sur-Marne" Hb A2 variant, "J-Paris-I" Hb A2 variant (alpha 2), "J-Habana" Hb A2 variant, "J-Paris-I" Hb A2 variant (alpha 1), "Wayne" Hb A2 variant (peak 2), Denatured Hb E

7 = Z(F) Hb F, Hb Willamette, Hb Alabama, Hb Chapel Hill, Hb Park Ridge, Hb Porto Alegre, Hb Q-Thailand (G-Taichung), Hb Sabine, Hb Bassett, Hb Rampa, Hb G-San José, Hb Barcelona, Hb Geldrop Santa Anna, Hb Richmond, Hb Boumerdes, Hb Swan River, Hb Burke, Hb Tarrant, Hb Presbyterian, Hb Manitoba II,

112

Hb Manitoba I, Hb Port Phillip, "J-Rovigo" Hb A2 variant, Denatured Hb S, Denatured Hb D-Punjab

8Hb Lansing, Hb Hinsdale, Hb Ypsilanti (Ypsi - peak 1), Hb Alberta, Hb Val de Marne (Footscray), Hb Kempsey, Hb Shelby (Hb Leslie), Hb Atlanta, Hb Ypsilanti (Ypsi - peak 2), Hb Rainier, Hb Athens-GA (Waco), Hb Debrousse

9 = Z(A)

Hb A, Hb Olympia, Hb Gorwihl (Hinchingbrooke), Hb Phnom Penh*, Hb Silver Springs*, Hb La Coruna*, Hb Bougardirey-Mali*, Hb Dallas*, Hb Toulon*, Hb Austin*, Hb Bonn*, Hb Buenos Aires (Bryn Mawr, major peak)*, Hb Chicago*, Hb Okayama*, Hb Fontainebleau*, Hb Raleigh*, Hb Hekinan*, Hb Mosella*, Hb Aztec*, Hb Little Rock*, Hb Frankfurt*, Hb Bethesda*, Hb M-Boston (M-Osaka)*, Hb Brisbane (Great Lakes)*, Hb Mizuho*, Hb Grange Blanche*, Hb San Diego*, Hb M-Saskatoon (main peak)*, Hb Malmö*, Hb Minneapolis Laos*, Hb Owari*, Hb Rhode Island (Southwark)*, Hb Twin Peaks*, Hb Wood*, Hb Conakry*, Hb Coimbra (Ingelheim)*, Hb Linköping (Meilahti)*, Hb Templeuve*, Hb Alzette*, Hb Ty Gard*, Hb Gouda*, Hb Syracuse*, Hb Fort Dodge, Hb Camperdown, Hb Jura

10 Hb Nouakchott, Hb Wayne (peak 1), Hb M-Iwate (M-Kankakee), Hb Camden (Tokuchi), Hb Hope

11Hb Vaasa, Hb Providence (X-Asn peak), Hb Tacoma, Hb Corsica, Hb K-Woolwich, Hb Lombard, Hb Andrew Minneapolis, Hb Fannin Lubbock, Hb Kaohsiung (New York), Hb Osler (Fort Gordon), Hb Himeji, Hb Jackson, Hb Tatras, "I (I-Texas)" Hb A2 variant

12

Hb Bart's, Hb Cemenelum, Hb Wayne (peak 2), Hb Hopkins-II, Hb J-Calabria (J-Bari), Hb J-Tongariki, Hb Providence (X-Asp peak), Hb J-Meerut (J-Birmingham), Hb J-Broussais (Tagawa-I - alpha 2), Hb J-Rajappen, Hb Grady (Dakar), Hb Le Lamentin, Hb J-Anatolia, Hb Hikari, Hb J-Broussais (Tagawa-I - alpha 1), Hb J-Chicago, Hb J-Toronto, Hb J-Oxford (I-Interlaken), Hb Ube-2, Hb J-Meinung (J-Bangkok), Hb Neuilly-sur-Marne, Hb Mexico (J-Paris-II), Hb J-Paris-I (J-Aljezur - alpha 1), Hb J-Habana, Hb J-Baltimore (N-New Haven), Hb J-Paris-I (J-Aljezur - alpha 2), Hb K-Ibadan

13 Hb N-Baltimore (Hopkins-I), Hb J-Rovigo, Hb J-Norfolk (Kagoshima), Hb J-Kaohsiung (J-Honolulu)

14 Hb N-Seattle

15 Hb I (I-Texas)

113

Figure 3. CZE scan of “AFSC” control

114

References

1. Borbely N, Phelan L, Szydlo R, Bain B. Capillary zone electrophoresis for haemoglobinopathy diagnosis. J Clin Path 2013; 66: 29-39.

2. Greene DN, Pyle AL, Chang JS, Hoke C, Lorey T. Comparison of Sebia Capillarys Flex Capillary electrophoresis with the BioRad Variant II high pressure liquid chromatography in the evaluation of hemoglobinopathies. Clinica Chimica Acta 2012; 413: 1232-1238

3. Keren DF, Shalhoub R, Gulbranson R, Hedstrom D. Expression of Hemoglobin

Variant Migration by Capillary Electrophoresis Relative to Hemoglobin A2

Improves Precision. Am J Clin Path 2012; 137: 660-664 4. Sae-ung N, Sriyorakun H, Fucharoen G, Yamsri S, Sanchaisuriya K, Fucharoen

S. Phenotypic expression of hemoglobin A2, E and F in various hemoglobin E related disorders. Blood Cells, Molecules, and Diseases 2012; 48: 11-15.

5. Sangkitporn S, Sangkitporn SK, Tanjatham S, Suwannakan B, Rithapirom S, Yodtup C, Yowang A, Duangruang S. Multicenter Validation of Fully Automated Capillary Electrophoresis Method for Diagnosis of Thalassemias and Hemoglobinopathies in Thailand. Southeast Asian J Trop Med Public Health 2011; 6(5):1224-1232.[PubMed]

6. Fucharoen G, Srivorakun H, Singsanan S, Fucharoen S. Presumptive diagnosis of Hemoglobinopathies in Southeast Asia using a capillary electrophoresis system. Int. Jnl. Lab. Hem. 2011; 33: 424-433. 7. Wajcman H, Moradkhani K. Abnormal haemoglobins: detection & characterization. Indian J Med Res 2011;134 (4): 538-546

8. Liao C, Zhou J-Y, Xie X-M, Li J, Li DZ. Detection of Hb Constant Spring by a Capillary Electrophoresis Method. Hemoglobin 2010; 34(2): 175-178.

9. Cotton F, Nalaviolle X, Vertongen F, Gulbis B. Evaluation of an Automated Capillary Electrophoresis System in the Screening for Hemoglobinopathies. Clin Lab 2009; 55: 217-221. 10. Van Delft P, Lenters E, Bakker-Verweij M, De Korte M, Baylan U, Harteveld CL, Giordano PC. Evaluating five dedicated automatic devices for haemoglobinopathy dianostics in multi-ethinic populations. Int Jnl Lab Hem 2009; 31: 484-495 11. Winichagoon P, Svasti S, Munkongdee T, Chaiya W, Boonmongkol P, Chantrakul N, Fucharoen S. Rapid diagnosis of thalassemias and other hemoglobinopathies by capillary electrophoresis system. Translational Research 2008, 152 (4): 178-184 12. Keren DF, Hedstrom D, Gulbransom R, Ou C-N, Bak R. Comparison of Sebia Capillarys Capillary Electrophoresis With the Primus High-Pressure Liquid Chromatography in the Evaluation of Hemoglobinopathies. Am J Clin Pathol 2008, 130: 824-831. 13. Wang J, Zhou S, Huang W, Kiu Y, Cheng C, Lu Xin, Cheng J. CE-based analysis of hemoglobin and its applications in clinical analysis. Electrophoresis 2006; 27: 3108- 3124. 14. Louhabi A, Philippe M, Lali S, Wallenmacq, Maisin D. Evaluation of a new Sebia kit for analysis of hemoglobin fractions and variants on the Capillarys system. Clin Chem Lab Med 2006; 44(3): 340-345. 15. Chang P-L, Kuo I-T, Chiu T-C, Chang H-T. Fast and sensitive diagnosis of thalassemia

115

by capillary electrophoresis. Anal Bioanal Chem 2004; 379: 404-410. 16. Jenkins M, Ratnaike S. Capillary Electrophoresis of Hemoglobin. Clin Chem Lab Med 2003; 41(6): 747-754. 17. Gulbis B, Fontaine B, Vertongen F, Cotton F. The place of capillary electrophoresis techniques in screening for hemoglobinopathies. Ann Clin Biochem 2003; 40: 659-662 18. Gerritsma J, Sinnige D, Drieze C, Sittrop B, Houtsma P, Ulshorst-Jansen NH, Huisman W. Quantitative and qualitative analysis of hemoglobin variants using capillary zone electrophoresis. Ann Clin Biochem 2000; 37 (3): 380-389. 19. Castagnola M, Messana I, Cassiano L, Rabino R, Rossetti DV, Giardina B. The use of

capillary electrophoresis for the determination of hemoglobin variants. Electrophoresis 1995; 16(1): 1492-1498.

20. http://www72.homepage.villanova.edu/frederick.vogt/ppt/2007/Capillary_Electrophoresis.ppt

21. http://chemwiki.ucdavis.edu/Analytical_Chemistry/Instrumental_Analysis/Capillary_Electrophoresis?highlight=capillary+zone+electrophoresis

116

Chapter 3Diagnostic Laboratory Methods

3.7 Isoelectric FocusingDavid R. Hocking, PhD

3.7.1 Introduction

Isoelectric focusing (IEF), also known as electrofocusing and isoelectricfocusing

electrophoresis, is a separation method that resolves complex mixtures of proteins by

their isoelectric points (pI). IEF is a type of electrophoresis that forms a pH gradient

during the run. The technique is capable of extremely high resolution. The formation of

a pH gradient is accomplished by blending a mixture of small molecular weight ‘carrier

ampholytes’ into a support matrix, or gel, usually of purified high-grade agarose. An

anolyte solution (i.e. acetic acid) and a catholyte solution (i.e. ethanolamine) are

saturated onto paper electrode wicks then are placed directly on opposite ends on the

surface of the agarose gel. Proteins (i.e. hemoglobins) that are to be separated are

placed near the cathode wicks using a clear plastic with rectangular wells cut out. The

protein solution (i.e. hemoglobin hemolysate) is then pipetted in the defined wells and

allowed to diffuse into the gel. An electric current is then passed through the medium.

The proteins move through the changing pH gradient until it reaches a point in which the

pH of that molecules pI is reached. At this point the protein no longer has an electric

charge and becomes neutral, or isoelectric (due to the protonation or de-protonation of

the associated functional amino and carboxyl groups) and as such will not proceed any

further within the gel. The proteins become ‘focused’ into sharp stationary bands with

117

each protein positioned at a point within the newly formed pH gradient corresponding to

its pI.

Note: All the IEF figures in this compendium were obtained after agarose gel electrophoresis on the Wallac Resolve Hemoglobin System (Perkin Elmer), and the scans were procured using the Wallac WS-1010 IsoScan Imaging System (Perkin Elmer).

3.7.2 IEF of Normal Adult Hemoglobins: HbA (Adult), HbF (Fetal), HbA2

Normal adult hemoglobins are comprised of α, β, γ and δ globin chains paired as

~96% HbA (α2 β2), ~3% HbA2 (α2 δ2) and <2% HbF (α2 γ2) tetramers (Figure 1). One can

usually find the glycated form of HbA, or HbA1c , anodal to it as shown in the Figure 2.

118

119

Figure 1.

Figure 2.

Aging bands, HbA3, are also anodal to HbA and are the result of post-translational

modifications such as acetylation and glutathione attachment.

It should be noted that beta-chain variants such as HbS, HbE, HbD, etc., will also

display glycated forms anodal to the variant (HbS1c, HbE1c, HbD1c). This observation is

critical to note should the patient exhibit symptoms of diabetes where their blood

glucose values are documented to be high. The percentage can be upwards of 10-20%

in cases of uncontrolled blood glucose levels.

3.7.3 IEF of Normal Newborn Hemoglobins: HbF (Fetal) and HbA (Adult)

Normal newborn hemoglobins are comprised of ~ 60-85% HbF (α2 γ2) and 15-

40% of HbA (α2 β2). It is very rare to see HbA2. About 10% of HbF is partially acetylated

HbFac, which results in higher oxygen affinity, an important property needed for

newborns. Aging hemoglobin bands, or HbF3 are always anodal to HbFac, and are the

result of glutathione (an antioxidant, preventing damage to important cellular

components caused by reactive oxygen species) attachment.

Representative patterns of newborn hemoglobin are shown in Figure 3. Fresh

cord blood is shown in channel 3a. A sample that was collected and stored using a filter

paper is shown in channel 3b. Note the increased levels of HbF3 in the stored blood

collected on filter paper.

120

3.7.4 IEF of Beta-Chain Variant Hemoglobins

It is customary to report hemoglobins in the order of greatest concentrations.

Heterozygotes are usually expressed as HbAX where X represents the beta-chain

variant i.e. Sickle Cell (S), C, D, E or name of the variant. Examples are HbAS, HbAE or

HbAD-Punjab. A few examples are shown (note percentages).

121

Figure 3. IEF of normal newborn hemoglobins: HbF (fetal) and HbA (adult)

122

Figure 4. Hemoglobin A-S Trait

Figure 5. Hemoglobin D-Punjab Trait

Figure 6. Hemoglobin A-E Trait Figure 7. Hemoglobin A-C Trait

Figure 9. Hemoglobin S/ß+-thalFigure 8. Hemoglobin A-O Trait

These examples show the beta-chain mutation along with the Relative Charge

Value (RCV) change as a result of the substitution of the normal amino acid found in

HbA. In each case the HbA is in greater concentration than the beta variant (HbX). This

pattern is true for all positive RCV values.

Background on Hb O-Arab: This rare hemoglobin variant emerged about 2,000

years ago on a singular haplotype, characteristic of the Greek Pomaks. Its frequency

increased as a consequence of high genetic drift within this population, and it was

dispersed throughout the Mediterranean basin and Middle East with minor variations of

its haplotypic pattern. (Haematologica. 2005 Feb;90(2):255-7. HbO-Arab mutation

originated in the Pomak population of Greek Thrace. Papadopoulos V, Dermitzakis E,

Konstantinidou D, Petridis D, Xanthopoulidis G, Loukopoulos D).

The example in Figure 9 shows adult hemoglobin that has more HbS than HbA.

The patient has reduced HbA, an increase in HbA2 and shows >than 3% of HbF. These

findings indicate that the patient has a beta thalassemia (reduced amount of beta-globin

chains from one parent) along with sickle cell hemoglobin (HbS) from the other parent.

Note the aging bands from sample storage.

3.7.5 IEF of Alpha-Chain Variant Hemoglobins

An individual inherits two sets of alpha globin genes, α1 and α2, from each

parent. If one of the alpha genes has a mutation, then one out of the four, or ~25% of

the hemoglobin, will be the variant, not the typical 50% from a beta-chain variant. The

affected alpha globin chain will form dimers with the non-alpha globin chains. HbG-

Philadelphia is a common alpha-chain variant that is shown below (Figure 10).

123

Note that the percentage of HbG-Philadelphia (Figure 10) relative to HbA is less

than what is seen in the beta variant HbD or HbD-Punjab. This is a ‘clue’ to suspecting

an alpha variant. Additionally you should also observe that there should be another

band cathodal to HbA2. This is due to the variant alpha globin chain combining with the

delta chain.

124

Figure 10. Hemoglobin G-Philadelphia Trait

Figure 11. Hemoglobin ASG-Philadelphia

The example in Figure 11 is a rare combination of the beta HbS variant and the

alpha HbG-Philadelphia variant. Note the presence of four prominent bands: HbA, HbG-

Philadelphia, HbS and the hybrid, HbSG-Philadelphia, the tetramer formed by the

dimers of α-GPhiladelphia and βS. Also note the HbA2 variant that resulted from the αG

and δ dimers. It will be seen cathodal (negative electrode) to the hybrid.

3.7.6 IEF of Thalassemias

A typical β+-thalassemia is shown in Figure 12. Note that the percentage of HbA

is reduced (95%) and the amount of circulating HbA2 is increased (>3.5%). Beta

thalassemias occur in persons of Mediterranean origin, and to a lesser extent, Chinese,

other Asians, and African Americans. β+-thalassemia is also known as Thalassemia

Minor and occurs if you receive the defective beta-globin gene from only one parent.

Persons with this form of the disorder are carriers of the disease, Cooley’s anemia or

beta thalassemia major (β0), if their other partner also passes their defective gene to the

baby.

125Figure 12. Hemoglobin ß+-thal

The pattern in Figure 13 is typical of those individuals presenting with a severe form of

Sickle Cell disease. In this example, the patient inherited the HbS from one parent and

is missing the beta globin gene from the other parent. The patient, though missing a

beat globin gene, has compensated for the missing beta-globin gene with the

persistence of making HbF from the gamma-globin gene.

126

Conclusion

IEF can be an important tool in assiting the laboratorian in the dection and

interpretation of hemoglobin variants. The technique offers improved resolution over

traditional electrophoretic methods and is useful for both adult and newborn patients. By

careful observation, one can determine if the variant is either a β or α variant or

combination. One can also correctly interpret β-thalassemias.

References:

1. David R. Hocking. The Separation and Identification of Hemoglobin Variant by Isoelectric Focusing Electrophoresis (May 2004), Catalog # HC-60, Perkin Elmer Life and Analytical Sciences, Wallac Oy, P.O. Box 10, FIN-20101 Turku, Finland. Tel. 358-2-2678111 Fax. 358-2-2678357 Web site: www.perkinelmer.com 2. Additional information about the IEF procedures and instruments can be solicited from:

a) Petra Furu, Ph.D., Global Business Manager, Specialty Diagnostics,Perkin Elmer , Mustionkatu 6 / 20750 Turku / Finland.e-mail: petra.furu@perkinelmer.com Tel. 358 2 267 8497

b) William R. Fisher, Technical Support Specialist, Specialty Diagnostics, Perkin Elmer, 520 South Main Street, Akron, OH 44311, USA e-mail: william.fisher@perkinelmer.com Tel. 330-564-4883

127

Chapter 3 Diagnostic Laboratory Methods

3.8 High Performance Liquid Chromatography Zia Uddin, PhD

3.8.1. Introduction

In 1973 I had the privilege of attending a short course on High Performance

Liquid Chromatography (HPLC), sponsored by the American Chemical Society at

Virginia Polytechnic Institute, Blacksburgh, Virginia, USA. The teachers of this course

were Drs. Lloyd R. Snyder and Joseph J. Kirkland. These two scientists are responsible

for several advancements in HPLC, and their most significant contribution in

collaboration with Dr. John W. Dolan is their latest book (Snyder LR, Kirkland JJ, Dolan

JW, Introduction to Modern Liquid Chromatography, 3rd Edition, John Wiley & Sons,

Hoboken, NJ 20010). Persons interested in HPLC shall find this book very helpful in

understanding the theory and practice of HPLC, and the components of HPLC (solvent

system, pump, injection port, column, stationary phase, detector, computer, etc.).

Additional literature about HPLC can be accessed from the following Internet sites:

http://www.lcresources.com

http://lchromatography.com/hplc find/index.html

http://tech.groups.yaho.com/group/chrom-L/links

http://userpages.umbc.edu/~dfrey1/Freylink

http://www.chromatographyonline.com

128

Note: The name High Pressure Liquid Chromatography was initially used, however now the word “Pressure” is replaced by “Performance.” In this book we shall therefore use High Performance Liquid Chromatography nomenclature.

3.8.2. Basic Principle

Liquid chromatography (LC) consists of a liquid mobile phase

and a stationary phase and the separation is accomplished by the

distribution of the solutes between these phases. Manual LC procedure is slow

and needs additional steps for the identification of the compound of interest. In

HPLC the separation process is expedited by forcing the mobile phase under

high pressure through the column, and almost all the steps of the operation are

automated (Figure 1).

129

Figure 1. Key components of a HPLC system, a) computer, b) detector, c) column, d) injection port, e) pump, f) solvent reservoir. (adopted from Snyder LR, Kirkland JJ, Dolan, JW, Introduction to Modern Liquid Chromatography, 3rd Edition. John Wiley & Sons, Hoboken, NJ 20010).

The identification of a compound of interest in HPLC is ascertained by

matching its retention times (time required to separate a compound after the

injection step) with a standard or control. Several kinds of detectors are

employed in HPLC for detection purposes, e.g., spectrophotometric, flurometric,

electrochemical, etc. Another development in the identification of a compound

after HPLC is coupling it with mass spectrometry (Chapter 3.4). This technique is

very useful when the retention time of the compound is not previously known.

The identification is achieved by the m/z value of the ion associated with the

compound of interest, e.g., globin chain of a hemoglobin variant (Chapter 3.4).

3.8.3 Illustrations

a) Quantification of Hb A2, and Hb A1c :

One of the widely used procedure employing HPLC is the

quantification of Hb A1c and Hb A2 ( Figure 2).

130

Figure 2. Peak at 0.81 (Hb A1c) and at 3.1 (Hb A2). Adopted from the Technical Manual of D-10, Bio-Rad, Hercules, CA.

b) Hb OIndonesia in India: a rare observation

The father is heterozygous for Hb OIndonesia and the mother is normal, however the

daughter has an HPLC pattern similar to her father1 (Figure 3). Although the normal

hemoglobin fractions (Hb F, Hb A, Hb A2)as well as the common variants (Hb S and Hb

C) all have distinct retention times,there are less prevalent variants with similar or

131

identical retention times. In these cases additional laboratory procedures must be

utilized for a conclusive diagnosis.

Figure 3. a) HPLC of daughter, b) HPLC of father, c) HPLC of mother

132

c) Apparent hemoglobinopathies caused by blood transfusions

Any spurious peak in HPLC can cause confusion and lead to

unnecessary additional testing. It is advised that in order to reduce

unwarranted commotion, the patient’s medical record should be examined

for recent blood transfusions. Figure 4 illustrates an example of a patient

with Hb SS disease on hyper-transfusion regimen who received a unit of

blood from a donor heterozygote for Hb O-Arab as demonstrated by a

small, but prominent peak eluting after Hb S.

Figure 4. HPLC of a Hb SS patient transfused with one unit of Hb A-O Arab blood.

133

Cited references:

1. Chopra A, Fisher C, Khunger JM, Pati H. Hemoglobin OIndonesia inIndia: a rare observation. Ann Hematol 2011; 90: 353-354

2. Kozarski TB, Howantiz PJ, Howantiz JH, Lilic N, Chauhan YS.Blood transfusions leading to apparent Hemoglobin C, S, andO-Arab Hemoglobinopathies. Arch Pathol Lab Med 2006; 130: 1830-33.

Additional references:3. Szuberski J., Oliveira JL, Hoyer JD. A comprehensive analysis of hemoglobin

variants by high performance liquid chromatography (HPLC). International Journal of Hematology 2012; 34: 594-604.

4. Ondei LS, Zamaro PJA, Mangonaro PH, Valencio CR, Bonini- Domingos CR. HPLC determination of hemoglobins to establish

reference values with the aid of statistics and informatics. Geneticsand Molecular Research 2007; 6(2): 453-460.

5. Mais DD, Boxer LA, Gulbranson RD, Keren DF. Hemoglobin Ypsilanti.A High-Oxygen-Affinity Hemoglobin Demonstrated by Two AutomatedHigh-Pressure Liquid Chromatography Systems. Am J Clin Path 2007; 128: 850-853.

6. Joutovsky A, Hadzi-Nesic J, Nardi MA. HPLC Retention Time as a Diagnostic Tool for Hemoglobin Variants and Hemoglobinopathies: A Study of 60 000 Samples in a Clinical Diagnostic Laboratory. Clin Chem 2004; 50: 1736-47.

7. Ou C-N, Rognerud CL. Diagnosis of hemoglobinopathies: electrophoresis vs. HPLC. Clin Chim Acta 2001; 313: 187-94.

8. Fucharoen S, Winichagoon P, Wisedpanichkij R, et al. Prenatal and postnatal diagnosis of thalassemias and hemoglobinopathies by HPLC. Clin Chem 1998; 44: 740-748.

9. Riou J, Godart C, Hurtrel D, Mathis M, Bimet C, et al. Cation-exchange HPLC evaluated for presumptive identification of hemoglobin variants.Clin Chem 1997; 43: 34-39.

10. Huisman THJ. Review: Separation of Hemoglobins and Hemoglobin Chains By High-Pressure Liquid Chromatography. J. Chromatogr 1987; 418: 277-304

11. Colah RB, Surve R, Sawant P, D’Souza E, Italia K, Phanasgaonker S, Nadkarni AH, Gorakshaker AC. HPLC studies in hemoglobinopathies. Indian J Pediatr 2007; 74(7): 657-62

12. Sachdey R, Dam AR, Tyagi G. Detection of Hb variants and hemoglobinopathies in Indian population using HPLC: Report of 2600 cases. Indian J Pathol Microbiol 2010; 53: 57-62.13. Rao S, Kar R, Gupta SK, Chopra A, Saxena R. Spectrum of haemoglobinopathies diagnosed by cation exchange-HPLC and modulating effects of nutritional deficiency anemias from north India. Indian J Med Res 2010; 132: 513-519.

134

Chapter 4Globin Chain Analysis

4.1 Solid Phase Electrophoretic Separation Zia Uddin, PhD

In the early stages of the development of the diagnostic hemoglobinopathies,

polyacrylamide gel electrophoresis (in urea, acid and non-ionic detergent Triton-X-100)

and cellulose acetate electrophoresis (alkaline and acid pH) were utilized for globin

chain analysis. These techniques provided information about the globin chains that

contained the substitution. However, due to scientific limitations (selection of known

variants as controls with mobility similar to that of the unknown), these techniques were

abandoned in favor of other methods as described in this chapter. Recently capillary

zone electrophoresis was also used for the separation of globin chains. For historical

reasons we have briefly presented the basic features of cellulose acetate

electrophoresis of globin chains.

4.1.1 Cellulose Acetate Electrophoresis (Alkaline and Acid pH)

First the heme groups and the globin chains are dissociated from the hemoglobin

molecule using 2-mercaptoethanol and urea. Electrophoresis at alkaline pH is

performed using the tris-ethylenediaminetetraacetic acid buffer at pH 8.8-9.5. In Figure

1 the relative mobilities of globin chains at alkaline pH are presented.

There is not much difference in the mobilities of globin chains between the alkaline

(8.8 – 9.5) and acidic (6.0-6.2) pH, and in both cases the alpha chains migrate towards

135

the cathode and the beta chains towards the anode. In Figure 2 the relative mobilities of

globin chains at acidic pH are presented.

Fig 1. Relative Mobilities of Globin Chains (Cellulose Acetate Electrophoresis at pH 8.8-9.5). Adopted from Laboratory Methods for Detecting Hemoglobinopathies, Division of Host Factors, Center for Infectious Diseases, Center for Disease Control, Atlanta, GA (September 1984)

136

Fig 2. Relative Mobilities of Globin Chains (Cellulose Acetate Electrophoresis at pH 6.0-6.2). Adopted from Laboratory Methods for Detecting Hemoglobinopathies, Division of Host Factors, Center for Infectious Diseases, Center for Disease Control, Atlanta, GA (September 1984)

137

References

1. Ueda S, Schneider RG. Rapid identification of polypeptide chains of hemoglobin by cellulose acetate electrophoresis of hemolysates. Blood 1969; 34: 230.

2. Schneider RG. Differentiation of electrophoretically similar hemoglobins- such as S,D,G and P; or A2, C,E, and O- by electrophoresis of the globin chains. Clin Chem 1974; 20(9): 1111-1115.3. Shihabi ZK, Hinsdale ME. Simplified hemoglobin chain detection by

capillary electrophoresis. Electrophoresis 2005; 26: 581-585

138

Chapter 4 Globin Chain Analysis

4.2 Reverse-Phase High Performance Liquid Chromatography

Zia Uddin, PhD and Rita Ellerbrook, PhD

Conventional charge based separation techniques (electrophoresis, ion-exchange liquid

chromatography, and isoelectric focusing) are sometimes ineffective in the separation of

hemoglobins, when the amino acid substitution does not cause a net charge differential.

Several hemoglobin variants migrate upon electrophoresis and elute upon ion-exchange liquid

chromatography in the positions of hemoglobin A, S, D, A2 or C. Further clarification is

necessary for newborn screening or in cases of unexplained clinical disorders. Additional

testing is required to resolve this matter, e.g., DNA studies, reverse-phase chromatography

(RPC), liquid chromatography-mass spectrometry (LC-mass) primarily employing the

electrospray ionization (ESI) technique, and Sanger sequencing.

There are three main chromatographic techniques for the separation of peptides and

proteins, e.g., a) size exclusion, b) ion-exchange, and c) hydrophobic interactions. For a

detailed study of the theory and practice of the liquid chromatography of peptides and proteins

in general and reverse phase high pressure liquid chromatography (RP-HPLC) in particular

(Chapter 13, Section 13.4), the interested reader is advised to review the 3rd edition of

“Introduction to Modern Liquid Chromatography”, by Lloyd L. Snyder, Joseph J. Kirkland and

John W. Dolan (A John Wiley and Sons, Inc. Publication, 2010). Howard and Martin1 first

introduced RPC in 1950, and since then, several improvements in the methodology and

advancements in its application in the separation of peptides and proteins were achieved. The

139

recent literature on RPC can be accessed via the Internet (http://www.lcresources.com) and

the specialized journals in the field.

The separation of globin chains by RP-HPLC is based on the hydrophobicity of the

globin chains, which is defined as a tendency of not combining with water or incapable of

dissolving in water. The RP-HPLC consists of a non-polar column in combination with a polar

mixture of water plus an organic solvent as a mobile phase. In this section, we shall

demonstrate the usefulness of RP-HPLC in the separation of globin chains leading to the

identification of hemoglobin variants. Experimental details of RP-HPLC of globin chains

(hemoglobin specimen preparation, selection of column, solvent system, high pressure liquid

chromatography instrumentation, temperature, retention times, detection system, etc.) were

provided by the work of three research groups in this field in Italy, France and USA2-7.

A few RP-HPLC chromatograms (Fig 1-5) are shown to illustrate the application of this

technique in the separation of globin chains. These chromatograms are either replicated

exactly as cited in the literature (abscissa depicting actual retention times in minutes), or for

comparison, as a normalized scale for the retention times. In the normalized scale, the

retention time for the normal β chain is 10, and for the normal α chain is 20 (Fig 4). The elution

window is of 0.5 units width. It is emphasized that the retention times of RP-HPLC might vary

depending upon the experimental conditions, but the overall shape of the chromatogram is

highly reproducible.

140

Normal Cord Blood: Fetal blood obtained at 18-20 weeks of gestation age, shows the

preponderance of α chains (Fig 1).

Fig 1. RP-HPLC chromatogram of a normal cord blood (Leone L, Monteleone M,Gabutti V, Amione C. Reversed-Phase High Performance Liquid Chromatography of Human Hemoglobin Chains. J Chromatogr. 1985, 321: 407-419)2.

Normal Adult Blood: The first peak (Fig 2) at ≈ 10 minutes is the heme molecule

followed by two major peaks, a) β chain of Hb A (31-35 minutes), and b) α chain

of Hb A (43-48 minutes).

141

Fig 2. RP-HPLC chromatogram of a normal adult blood (Kutlar F, Kutlar A, Huisman THJ. Separation of Normal and Abnormal Hemoglobin Chains by Reverse-Phase High Performance Chromatography. J. Chromatogr 1986, 357: 147-153)3.

Adult hemoglobin A-S trait : In hemoglobin S the variation in the β chain is due to the

substitution of glutamic acid by valine [β6(A3)]. This is shown in Fig 3 by the separation of βA

and βS chains.

142

Fig. 3. RP-HPLC chromatogram of an adult Hb A-S trait (Leone L, Monteleleone M, Gabutti V, Amione C. Reversed-Phase High Performance Liquid Chromatography of Human Hemoglobin Chains. J. Chromatogr 1985, 321: 407-419)2.

Hemoglobin S interacts with Hb D-Punjab [121(GH4) Glu→Gln] causing sickle cell

disease. Hemoglonin S also interacts with Hb Korle-Bu [73(E17) Asp→Asn], but in the

opposite direction, i.e., inhibiting sickling. Both of these hemoglobin variants

(Hb D-Punjab and Hb Korle-Bu) are frequently found in the population sector dominated by Hb

S. The separation of Hb D-Punjab and Hb Korle-Bu is difficult from cellulose acetate

143

electrophoresis and isoelectric focusing, however both the βD-Punjab and βKorle-Bu chains can be

easily separated by RP-HPLC7.

Several electrophoretic separation techniques did not distinguish4 Hb Camperdown

[β104(G6) Arg→Ser] from Hb Sherwood Forest [β104(G6) Arg→Thr]. In this example there is

no change on the charge of the two hemoglobin variants as only the hydrogen atom on serine

is replaced by a methyl group of threonine. The substitution of serine by threonine on the same

position of the β chain changes the hydrophobicity (presumably by altering the

secondary/tertiary structure of the globin chain), thus resulting in their separation by RP-HPLC

(Fig 4).

Fig. 4. Normalized scale of retention times of globin chains on RP-HPLC, a) retention times of βA (10), α (20), Gγ (28) and Aγ (35), b) Hb Campertown (14.1-14.5), and c) Hb Sherwood Forest (16.1-16.5). Adopted from: Wajcman H, Riou J, Yapo AP. Globin Chain Analysis by

144

Reversed Phase High Performance Liquid Chromatography: Recent Developments. Hemoglobin 2002, 26: 272-2844.

Another interesting illustration8 of the usefulness of RP-HPLC was the resolution of

hemoglobinopathy during newborn screening, provided by Hb Sinai-Greenspring [β34(β16)

Val→Ile, GTC>ATC]. IEF showed an abnormal band (slightly anodal to HbA), and HPLC (Fig

5a) was also inconclusive except the broadening of the band due to a hemoglobin variant. RP-

HPLC did indicate a distinct band due to a variant Hb between the β and α chains (Fig 5b).

Substitution of amino acid valine at position 34 of the β-globin chain by isoleucine changed the

hydrophobicity of the protein molecule and thus allowed the separation of two β chains by RP-

HPLC (Fig 5b).

145

Fig 5. Cation exchange HPLC chromatogram (a) of infant with Hb-Sinai-Greenspring,and RP-HPLC chromatogram (b). Adopted from: Dainer E, Wenk RE, Luddy R, Elam D, Holley L, Kutlar A, Kutlar F. Two new hemoglobin variants: Hb Sinai-Greenspring [β34 (β16) Val→Ile, GTC>ATC] and Hb Sinai-Bel Air [β53 (D4) Ala→Asp, GCT>GAT]. Hemoglobin 2008; 32(6): 588-5918

146

Henri Wajcman and associates published the retention times on RP-HPLC of over 200

abnormal globin chains which were also made available on the web7. Additional

chromatographic and electrophoretic information about hemoglobin variants can be obtained

from the database9-11.

References

1. Howard GA, Martin JP. The separation of the C12-C18 Fatty Acids by Reversed-Phase Partition Chromatography. Biochem J 1950; 46: 532-538.

2. Leone L, Monteleone M, Gabutti V, Amione C. Reversed-Phase High Performance Liquid Chromatography of Human Globin Chains. J Chromatogr 1985; 321: 407-419.

3. Kutlar F, Kutlar A, Huisman THJ. Separation of Normal and Abnormal Hemoglobin Chains by Reversed-Phase High-Performance Liquid Chromatography. J Chromatogr 1986; 357:147-153.

4. Wajcman H, Riou J, Yapo AP. Globin Chain Analysis by Reversed Phase High Performance Liquid Chromatography: Recent Developments. Hemoglobin 2002; 26: 271-284.

5. Yapo PA, Datte JY, Yapo A, Wajcman H. Separation of Adult Chains of Abnormal Hemoglobin: Identification by Reversed-Phase High-Performance Liquid Chromatography. J Clin Lab Anal 2004;18: 65-69.

6. Zanella-Cleon I, Becchi M, Lecan P, Giordano PC, Wajcman H, Francina A.Detection of a Thalassemic α-Chain Variant (Hemoglobin Groene Hart) by Reversed-Phase Liquid Chromatography. Clin Chem 2008; 54:1053-1059.

7. Wajcman H, Riou J. Globin chain analysis: An important tool in phenotype study of hemoglobin disorders. Clinical Biochemistry 2009; 42:1802-1806.

8. Dainer E, Wenk RE, Luddy R, Elam D, Holley L, Kutlar A, Kutlar F. Two new hemoglobin variants: Hb Sinai-Greenspring [β34 (β16) Val→Ile, GTC>ATC] and Hb Sinai-Bel Air [β53 (D4) Ala→Asp, GCT>GAT]. Hemoglobin 2008; 32(6): 588-591

9. Hardison RC, Chui DHK, Giardine B, et al. HbVar: a relational database of human hemoglobin variants and thalassemia mutations at the globin gene server. Hum Mutat 2002; 19: 225-33 (http://globin.bx.psu.edu/hbvar/smenu.html).

10. Giardine B, van Baal S, Kaimakis P, et al. HbVar database of human hemoglobin variants and thalassemia mutations: 2007 update. Hum Mutat 2007; 28(2): 206.11. Patrinos GP, Giardine B, Riemer C, et al. Improvements in the Hbvar database of human hemoglobin variants and thalassemia mutations for population and sequence variation studies. Nucleic Acids Res 2004 Jan 1; 32: D537-41(Database issue).

147

Chapter 4Globin Chain Analysis

4.3 Globin Chain Gene Mutations: DNA Studies Joseph M. Quashnock, PhD

14.3.1 Introduction

Hemoglobin A is the designation for the normal hemoglobin that exists after birth.

Hemoglobin A is a tetramer with two alpha chains and two beta chains (á2â2).

Hemoglobin A2 is a minor component of the hemoglobin found in red cells after birth

and consists of two alpha chains and two delta chains (á2ä2). Hemoglobin A2 generally

comprises less than 3% of the total red cell hemoglobin. Hemoglobin F is the

predominant hemoglobin during fetal development. The molecule is a tetramer of two

alpha chains and two gamma chains (á2ã2). Hemoglobinopathies result from amino

acid changes in the alpha or beta globin chains. Most of the mutations are single amino

acid substitutions caused by a single base change, however, other amino acid

mutations can be found due to various base alterations such as:

1. More than one amino acid change e.g. the alpha chain mutation of Hb J

Singapore with Asn>Asp and Ala>Gly, the beta chain mutation of Hb

Poissy with Gly>Arg and Ala>Pro.

2. Elongation of the chain due to frameshifts or insertions such as Hb

Constant Spring or Hb Doha.

3. Shortened chains due to deletions such as Hb Leiden.

4. Hybrids such as the Lepore globin gene that is a crossover of beta and

148

delta globin genes that produces hemoglobin made up of two normal

alpha chains and two Hb Lepore chains.

Additionally, though much rarer, there are also changes in the gamma chains

(Hb F) and delta chains (Hb A2). Over 1,000 hemoglobin mutations have been

described. For a detailed list of the mutations, the reader is directed to the Globin Gene

Server of Pennsylvania State University at:

http://globin.cse.psu.edu/html/ and Department of Microbiology of the

University of Massachusetts at:

http://www.umass.edu/microbio/chime/hemoglob/index.htm.

Mutations that cause diminished production of the globin molecules are termed

Thalassemia. Equal numbers of alpha and beta chains are necessary for normal

hemoglobin synthesis.

4.3.2 Genotyping - PCR Methodology

Determining the genotype requires DNA from the subject and the synthesis of a

primer and probe for the known mutation. The subject’s DNA, a primer, a reporting

probe, DNA bases, and DNA polymerase enzyme are incubated a number of times to

amplify the mutation sufficiently to be detected with a labeled probe. However, the

procedure has limitations; the first is that the mutation must be known so that a unique

primer and probe can be made, secondly, a sufficient amount of sample DNA must be

present to make a sufficient quantity of PCR product (amplicon) which is then detected

and reported by the probe.

149

Methods that have been employed over the years for identifying single mutations are:

1. Restriction Fragment Length Polymorphism (RFLP) detection in which

specific restriction enzyme digested DNA is separated by electrophoresis1.

2. Binding of a labeled Allele-Specific Oligonucleotide (ASO) probe to

amplified DNA2.

3. Allele-Specific PCR (ASP), PCR Amplification of Specific Alleles (PASA), or

Amplification Refractory Mutation System (ARMS), in which the presence or

absence of a normal or mutant sequence is determined by whether the PCR

products generated by specific primers can be detected through a reporting

system such as electrophoresis, or a fluorescent, chemical, colorimetric, or

electric signal. The signals may be read directly by the human eye

(electrophoresis) or detected by instrumentation in which case they may also be

quantitated3.

Some additional methods for multi-mutation detection by PCR assays include:

1. Allele-Specific Primer Extension (ASPE) assays that detect the

incorporation of a labeled nucleotide that binds at a single nucleotide polymorphism (SNP) and is linked to an oligonucleotide that is bound next

to the SNP site3.

2. Binding labeled multiplex ASPE products to mutation specific beads that

can generate identifying signals in solution when separated by laser flow

cytometry as is done by the Luminex®4.

3. Oligonucleotide Ligation Assay (OLA) based on the binding and ligation of 150

an allele-specific probe to a common downstream sequence reporter

probe, which generates a specific fluorescent signal from the completed

ligation products separated on electrophoresis5.

4. Hybridizing PCR amplification products to electrode-bound allele-specific

probes (printed circuit board, microarray, chip-based) to generate electric

signals6.

5. Fluorescence Resonance Energy Transfer (FRET) fluorescent signals

generated by Cleavase® treated PCR products7.

PCR amplification products are produced by incubating extracted DNA from the

specimen with DNA primers, the substrate nucleic acid bases of adenosine, thymidine,

guanosine, and cytidine, DNA polymerase, and a DNA detection probe. The mixture is

repeatedly heated to ~ 95  ̊C and cooled; each heating and cooling cycle doubles the

amount of PCR product produced; most PCR assays use 25-40 cycles.

Rapid cycle PCR is based upon the low heat capacity of air and the ability to

ramp through temperatures at a far greater rate than instruments using thermocyclers

that rely upon heating and cooling an aluminum block. Instruments such as the

LightCycler® from Roche also incorporate the improvement of using glass capillary

tubes to serve as both the reaction vessel and optical cuvette. Detection is by the

Fluorescence Resonance Energy Transfer (FRET) method described below, however,

the time required to complete 25-40 cycles is on the order of 30-40 minutes as opposed

to 3-4 hours for aluminum block thermocyclers.

151

Detection of FRET probes is performed by measuring hybrid stability as modified

by the introduction of base pair mismatch/es. Mismatch destabilization is measured by

observing the melting temperature of the FRET probe as monitored by fluorescence

output. Fluorescence is generated using two fluorophores. The first or “Light-Up”

fluorophore is excited at an appropriate wavelength. The emission of the Light-Up

fluorophore is in turn used to excite the detection fluorophore. The subsequent

emission of the detection fluorophore is monitored. In order for resonance energy

transfer to occur between the Light-Up and detection fluorophores and produce a

florescent signal, the two fluorophores must be in close proximity. Proximity is achieved

by conjugating the fluorophores to oligonucleotides such that when the oligonucleotides

are hybridized to their target in an amplicon, the fluorophores are held in proximity. The

mixture is then heated and a melting curve is generated by the slow thermal denaturing

of the probe-template hybrid. Melting curves are generated by monitoring the loss of

fluorescence over the course of denaturation. Melting peaks are generated by plotting

the inverse derivative of fluorescence verses temperature (-dF/dT) - the bigger the

mismatch between the amplicon and the probe, the lower the melting temperature.

Because most hemoglobinopathies are single amino acid mutations such as

base substitution or base pair insertion or deletion, the ASP method is the commonly

used technology. In this procedure, allele-specific primers for sequences are designed

to bind to and amplify a small region surrounding the site of the known mutation. A

probe of oligonucleotides, which matches the normal or abnormal sequence, binds to

the PCR products. The probes incorporate a label (fluorophore) that produces a signal

to show that binding has taken place and a specific sequence has been detected.

152

4.3.3 Mutations

“Hemoglobin beta” is the name of the hemoglobin gene and is abbreviated HBB.

Sickle cell anemia is the most common mutation and primarily affects African-

Americans with a frequency of 1:400. The defect causes red cells to distort and block

small capillaries. The â-globin gene is located on the small (p) arm of chromosome 11

in the region of 15.5 (HBB; MIM # 141900; 11p15.5). The mutation is the replacement

of an adenosine with a thymidine in the DNA that causes the substitution of valine for

glutamic acid at position 6 in the beta-globin chain. The codon sequence is shown

below, GAG, in the sixth position below, codes for glutamic acid; the replacement of

adenosine (A) with thymidine (T) produces GTG that codes for valine.

1 3 6 9GTG GAC CTG ACT CCT GAG GAG AAG TCT - - - (wildtype)

GluVal Asp Leu Thr Pro Glu Lys Ser

ValGTG GAC CTG ACT CCT GTG GAG AAG TCT - - - (Hb S)

Hemoglobin C is a mutation in the same codon which replaces the first

guanosine with adenosine (GAG becomes AAG) causing the glutamic acid to be

replaced with lysine.

1 3 6 9GTG GAC CTG ACT CCT GAG GAG AAG TCT - - - (wildtype)

GluVal Asp Leu Thr Pro Glu Lys Ser

LysGTG GAC CTG ACT CCT AAG GAG AAG TCT - - - (Hb C)

153

Similar single point mutations cause other variants of hemoglobin. Hemoglobin E

results when glutamic acid is replaced with the amino acid lysine at position 26 in â-

globin (Glu>Lys) due to the same GAG>AAG mutation that causes hemoglobin C at

codon 6. It is the second most common hemoglobin variant. When the hemoglobin E

mutation is present with hemoglobin S, Hb SE disease, the person may have more

severe signs and symptoms associated with sickle cell anemia, such as episodes of

pain, anemia, and abnormal spleen function.

Hemoglobin D-Punjab also known as Hb D-Los Angeles, Hb D-Chicago, Hb D-

North Carolina, Hb D-Portugal, and Hb Oak Ridge is an abnormality due to the

replacement of glutamic acid with glutamine on the hemoglobin beta chain. The

mutation is GAA>CAAÒ at codon 121 (â121 Glu>Gln). Hb D is primarily found in the

Indus River Valley (Punjab) region of Pakistan and Northwestern India but is

widespread, and has been observed in persons from China, England, Holland,

Australia, Greece, Serbia, Bosnia – Herzegovina, Macedonia, Montenegro, and Turkey.

It is the fourth most frequently occurring hemoglobin variant.

Heterozygotes for Hb D are normal. Homozygosity for Hb D is associated with

normal hemoglobin levels, decreased osmotic fragility, and some target cells.

Compound heterozygotes for Hb D and â-Thalassemia have mild anemia and

microcytosis. Hb D has been found in combination with Hb S, Hb C, Hb E, á-

thalassemia, and in the homozygous state. Hemoglobin D has been shown to interact

with the sickle hemoglobin gene S. Individuals who are compound heterozygotes for

154

---------------------------------------------------------------------------Ò See the DNA codon table for degeneracy (redundant) codons.

Hb S and Hb D-Los Angeles (SD) have moderately severe hemolytic anemia and

occasional pain episodes. Populations that have a high frequency of sickle hemoglobin

(SD) disease are those of Asian and Latin American descent.

Hemoglobin O-Arab is an abnormality due to the amino acid substitution of lysine

for glutamic acid at the 121st position in the beta globin gene. The genetic mutation is a

GAA>AAA at this codon (â121 Glu>Lys). The mutation is also known as Hb Egypt and

Hb O-Thrace. The mutation is found mainly in African-Americans, Gypsies, in Pomaks

(a population group in the Balkan countries) and in Arabian, Egyptian, and Black

families of the US and western hemisphere.

Hemoglobin O-Arab is important when found with sickle syndromes. Compound

heterozygotes for Hb S and Hb O-Arab have hemoglobin concentrations in the range of

7-8 g/dL with reticulocytosis, jaundice, splenomegaly, episodes of pain, and many other

complications seen in Hb SS disease. Heterozygote carriers have no clinical

manifestations. Homozygous individuals usually present with mild anemia and

microcytosis. Compound heterozygotes for Hb O-Arab and â-thalassemia have

manifestations similar to thalassemia intermedia.

Thalassemias are named by the chain that is deficient. In â-Thalassemia, there

is an insufficient amount of the beta subunit due to mutations such as -29A>G, -88C>T,

and IVS1+6T>C. The excess alpha subunits precipitate and eventually damage the red

blood cells. In severe á-thalassemia, the â-globin subunits begin to associate into

tetramers due to the reduced concentration of alpha chain. The tetramers of â-globin

do not transport oxygen. No comparable tetramers of á-globin subunits form with

severe á-thalassemia.

155

Below are several melting curves representing the various signals obtained

during an analysis. In allele specific binding assays, it is preferred that the primer and

detection probe are the sequences for the mutation and not the wildtype (“normal”)

sequence. Because the mismatch in base sequences causes the melting temperature

to be lower, the use of the wild type sequence as the detection probe will indeed

demonstrate a lower melting temperature when a mismatch is present, however it will

not be known as to which base/s mismatch (mutation) was present. The use of the

mutation as the template will always result in the specific mutation producing the highest

melting temperature.

The Hemoglobin S templates were used in the analysis of wildtype (“normal”)

hemoglobin in Figure 1 and shows a melting point of 55.5 oC.

Figure 1. Hemoglobin A (WT*) bound to Hemoglobin S probe, melting point is 55.5 ̊C.

156

* WT - Wild Type, the commonly occurring type - no mutation.Figure 2 shows a melting curve for a “carrier”, both hemoglobin sequences were

detected. Hemoglobin S has the higher melting temperature of 62.5 °C while the

wildtype melts at 55.5 °C. A homozygous sickle disease individual would show only one

melting point at 62.5 °C.

Figure 2. Hemoglobin A (WT) and Hemoglobin S (Mutant) bound to Hemoglobin S probe. Melting temperatures: WT - 55.5 °C and Mutant S - 62.5 °C.

As pointed out earlier, Hemoglobin S and Hemoglobin C differ from the wild type

by only one base in the same codon number 6 of the HBB gene. The Hb C mismatch

causes an even lower melting temperature than Hb S or the wildtype. Figure 3 shows

two melting points indicating a Hemoglobin C carrier with Hb C melting at 49.8 °C and

the wildtype again at 55.5 °C. This detection of two mutations is an example of a

157

multiplexed assay. This type of multiplexing is only useful when the bases involved in

the mutation are very close, e.g. ± 3 bases, otherwise the energy transfer would not be

very efficient and no fluorescent signal would be detected.

Figure 3. Hemoglobin C (Mutant) bound to Hemoglobin S probe. Melting temperatures: Mutant C - 49.8 °C and WT - 55.5 °C

158

Figure 4 is an example of a “non-preferred” base sequence for the Hemoglobin E

mutation in which the wildtype probe is used to detect the mismatch at codon 26 of

GAG to AAG. Here the wildtype melts at 70.3 °C and the Hb E mutation melts at 65.2

°C.

Figure 4. Hemoglobin E (mutant) bound to Hemoglobin A (WT) probe. Melting temperatures: Mutant E - 65.2 °C and WT - 70.3 °C

159

Figure 5 illustrates an analytical run in which a “normal”, no Hemoglobin D

present patient is in red, a “carrier” control specimen is in blue, and the green line is the

no DNA control which must not give a signal.

Figure 5. Hemoglobin A (WT) and Hemoglobin D (Mutant) bound to Hemoglobin D Probe. Melting temperatures: WT - 50.95 °C and Mutant D - 63.68°C.

160

Figure 6 is an example of a melting curve for Hemoglobin O. Note that among

the assays shown, there is no correlation of melting temperatures for the wildtype or

mutations. This is because each primer and probe set is different for each specific

mutation.

Figure 6. Hemoglobin A (WT) and Hemoglobin O (Mutant) bound to Hemoglobin D Probe. Melting temperatures: WT - 51.52 °C and Mutant D - 61.91°C .

161

Degeneracy Table

Amino Acid DNA codons

Alanine GCT, GCC, GCA, GCG

Arginine CGT, CGC, CGA, CGG, AGA, AGG

Asparagine AAT, AAC

Aspartic acid GAT, GAC

Cysteine TGT, TGC

Glutamic acid GAA, GAG

Glutamine CAA, CAG

Glycine GGT, GGC, GGA, GGG

Histidine CAT, CAC

Isoleucine ATT, ATC, ATA

Leucine CTT, CTC, CTA, CTG, TTA, TTG

Lysine AAA, AAG

Methionine ATG

Phenylalanine TTT, TTC

Proline CCT, CCC, CCA, CCG

Serine TCT, TCC, TCA, TCG, AGT, AGC

Threonine ACT, ACC, ACA, ACG

Tryptophan TGG

Tyrosine TAT, TAC

Valine GTT, GTC, GTA, GTG

Stop codons TAA, TAG, TGA

162

DNA Codon Table

First T C A G Third

T

TTTPhe

TCT

Ser

TATTyr

TGTCys

T

TTC TCC TAC TGC C

TTALeu

TCA TAAstop

TGA stop A

TTG TCG TAG TGG Trp G

C

CTT

Leu

CCT

Pro

CATHis

CGT

Arg

T

CTC CCC CAC CGC C

CTA CCA CAAGln

CGA A

CTG CCG CAG CGG G

A

ATT

Ile

ACT

Thr

AATAsn

AGTSer

T

ATC ACC AAC AGC C

ATA ACA AAALys

AGAArg

A

ATG Met ACG AAG AGG G

G

GTT

Val

GCT

Ala

GATAsp

GGT

Gly

T

GTC GCC GAC GGC C

GTA GCA GAAGlu

GGA A

GTG GCG GAG GGG G

References

1. 1Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. R. K. Saiki, S. Scharf, F. Faloona, K. B. Mullis, H. A. Erlich, and N. Arnheim, Science (1985) 230:1350–1354.

2. 1Specific Enzymatic Amplification of DNA In Vitro: The Ploymerase Chain Reaction. K. Mullis, F. Faloona, S. Scharf, R. Saiki, G. Horn, H. Erlich, Cold Spring Harbor Symposia on Quantitative Biology (1986) LI:263-273.

3. 1PCR Second Edition - The Basics. M. McPherson and S. Moller, Taylor & Francis Pub., New York (2006).

163

4. Luminex® Corporation, 112212 Technology Boulevard, Austin, TX 78727.5. Automated DNA diagnostics using an ELISA-based oligonucleotide ligation

assay. D. A. Nickerson, R. Kaiser, S. Lappin, J. Stewartt, L. Hood, and U. Landegrent, Proc. Nat. Acad. Sci. USA (1990) 87:8923-8927.

6. 1Design of electrochemical biosensor systems for the detection of specific DNA sequences in PCR-amplified nucleic acids related to the catechol-O-methyltransferase Val108/158Met polymorphism based on intrinsic guanine signal. D. Ozkan-Ariksoysal, B. Tezcanli,B. Kosova, and M.Ozsoz, Anal Chem. (2008) 80(3):588-596.

7. New Cleavase® Fragment Length Polymorphism Method Improves the Mutation Detection Assay, M. C. Oldenburg and M. Siebert, BioTechniques (2000) 28:351-357.

General References

1. C1ompound Heterozygosity Hb S/Hb Hope (β136Gly>Asp): a Pitfall in the Newborn Screening for Sickle Cell Disease. R. Ducrocq, A. Bevier, A. Leneveu, M. Maier-Redelsperger, J. Bardakdian-Michau, C. Badens, and J. Elion, Journal of Med Screening (1998) 5:27-30.

2. Rapid β-globin Genotyping by Multiplexing Probe Melting Temperature and Color. M. Herrmann, S. Dobrowolski, and C. Wittwer, Clinical Chemistry (2000) 46:425-428.

3. Identification of Hb D-Punjab gene: application of DNA amplification in the study of abnormal hemoglobins. Y. T. Zeng., S. Z. Huang, Z. R. Ren, and H. J. Li, Am J Hum Genet. (1989) 44(6):886-9.

4. The inherited diseases of hemoglobin are an emerging global health burden. D. J. Weatherall. Blood (2010) 115:4331.

5. Percentages of abnormal hemoglobins in adults with a heterozygosity for an alpha-chain and/or a beta-chain variant. T. H. Huisman. Am J Hematol (1983) 14:393.

6. http://www.ncbi.nlm.nih.gov/books/NBK1426/ Beta-Thalassemia, GeneReviews [Internet], A. Cao and R. Galanello. (2000 updated 2010).

7. Construction of a Genetic Linkage Map in Man Using Restriction Fragment Length Polymorphisms. D. Botstein, R. L. White, M. Skolnick, and R. W. Davis, Am J Hum Genet (1980) 32:314-331.

8. Specific Enzymatic Amplification of DNA In Vitro: The Ploymerase Chain Reaction. K. Mullis, F. Faloona, S. Scharf, R. Saiki, G. Horn, H. Erlich, Cold Spring Harbor Symposia on Quantitative Biology (1986) LI:263-273.

9. High-throughput SNP genotyping. S. Jenkins and N. Gibson. Comparative and Functional Genomics (2002) 3(1):57-66.

164

Chapter 4Global Chain Analysis

4.4 Electrospray Ionization-Mass Spectrometry Gul M. Mustafa, PhD and John R Petersen, PhD

Mass spectrometry (MS) is an analytical technique that identifies the chemical

composition of a sample on the basis of the mass-to-charge ratio (m/z) of charged ions.

The technique has both qualitative (structure) and quantitative (molecular mass or

concentration) uses. Another way of thinking about mass spectrometry is that it can be

considered as the “world’s most accurate scale”. Mass spectrometers can be divided

into three fundamental parts, namely the ionization source, the analyzer, and the

detector (Figure 1). The molecules of interest are first introduced into the ionization

source of the mass spectrometer, where they are ionized to acquire positive or negative

charges. This is done because ions are far easier to manipulate as compared to

molecules that do not have a charge. The ions then travel through the mass analyzer

and arrive at different parts of the detector according to their mass to charge (m/z) ratio.

After the ions make contact with the detector, useable signals are generated and

recorded via a computer. The computer displays the signals graphically as a mass

spectrum showing the relative abundance of the signals according to their m/z ratio. The

analyzer and detector of the mass spectrometer, and often the ionization source too,

are maintained under high vacuum to allow the ions to travel from one end of the

instrument to the other without colliding with air molecules which decreases the signal.

The entire operation and often the sample introduction process are under complete data

165

system control on modern mass spectrometers.

The method of sample introduction to the ionization source often depends on the

ionization method being used, as well as the type and complexity of the sample. Many

ionization methods are available and each has its own advantages and disadvantages.

The ionization method used depends on the nature and type of sample under

investigation and the mass spectrometer available. Figure 2 shows various ionization

methods of ionization such as Atmospheric Pressure Chemical Ionization (APCI),

Atmospheric Pressure Photo-Ionization (APPI), Electron Impact (EI), and Electrospray

Ionization (ESI). The ionization methods used for the majority of biochemical analyses 166

Figure: 1

are Electrospray Ionization (ESI) and Matrix Assisted Laser Desorption Ionization

(MALDI)

Mass spectrometry using ESI is called electrospray ionization mass spectrometry

(ESI-MS) or, less commonly, electrospray mass spectrometry (ES-MS). Electrospray

ionization mass spectrometry was pioneered by John Bennet Fenn, who shared the

Nobel Prize in Chemistry with Koichi Tanaka in 2002 for his work on the subject

(1). One of the original instruments used by Dr. Fenn is on display at the Chemical

Heritage Foundation in Philadelphia, Pennsylvania. This technique of ionization is

especially useful in producing ions from macromolecules because it overcomes the

propensity of these molecules to fragment when ionized and as such is considered a

167

Figure: 2

soft ionization technique. When analyzing biological molecules of large molecular mass,

ESI-MS is very useful because it does not cause fragmentation of the macromolecules

into smaller charged particles; rather it creates small droplets containing

the macromolecule being ionized and solvent allowing analysis of the molecular weight

of the intact macromolecule. Solvent can then be removed causing the formation of

even smaller droplets, creating protonation of the macromolecules. These protonated

and desolvated molecular ions are then passed through the mass analyzer to the

detector, and the mass of the sample is determined (Figure: 3). This method can be

performed on solid or liquid samples, and allows analysis of nonvolatile or thermally

unstable molecules which means that ionization of proteins, peptides, olgiopeptides,

and some inorganic molecules can be easily performed. The spectrum is shown with

the mass-to-charge (m/z) ratio on the x-axis, and the relative intensity (%) of each peak

shown on the y-axis.  The quantitative analysis is done by considering the mass to

charge ratios of the various peaks in the spectrum.  Calculations to determine the

unknown mass, (Mr) from the spectral data are performed using; p = m/z.

The ionization mechanism first involves the liquid containing the analyte(s) of

interest to be dispersed by electrospray into a fine aerosol. Because the ion formation

involves extensive solvent evaporation, the typical solvents for electrospray ionization

are prepared by mixing water with volatile organic compounds (e.g. methanol,

acetonitrile). To decrease the initial droplet size, compounds that increase the

conductivity (e.g. acetic acid) are customarily added to the solution. Large-flow

electrosprays can benefit from additional nebulization by an inert gas such as nitrogen.

168

There are some clear advantages and disadvantages of using electrospray

ionization mass spectrometry as an analytical method. It is one of the softest ionization

methods available; thus it can not only analyze molecules that have high molecular

masses but also has the ability to analyze biological samples that are defined by non-

covalent interactions. Since the m/z ratio range of a quadrupole instrument is fairly

small, the mass of the sample can be determined with a high amount of

accuracy. Sensitivity of the instrument is also impressive making it useful in both

quantitative and qualitative measurements. The major disadvantage of ESI-MS is that in

the analysis of mixtures the results are unreliable. In addition to the difficulty in handling

mixtures the multiple charges that are attached to the molecular ions can make for 169

Figure: 3

confusing spectral data. The apparatus is also very difficult to clean and has a tendency

to become contaminated with residues from previous experiments.  

In recent years, electrospray ionization (ESI) mass spectrometry has become an

increasingly important method in proteomics not only to analyze peptides but also to

study proteins and protein complexes of increasing size and complexity in structural

biology. The analysis of proteins and protein complexes by mass spectrometry

(macromolecular mass spectrometry) has become possible because of the

development of the relatively gentle ionization procedure related to ESI, which retains

non-covalent interactions. The mass-to-charge (m/z) ratios of these proteins can well be

over 10,000 daltons, and therefore, time-of-flight (TOF) analyzers with orthogonal

injection are the most commonly used analyzers in the field of macromolecules. The

m/z analysis of larger proteins and protein complexes is not a routine technique, since a

careful optimization of the operating conditions is always required. Despite the

theoretically unlimited mass range of TOF analyzers, most instruments have detection

problems when the m/z values exceed 4,000 daltons. It has been shown that a pressure

increase in the first and second vacuum chamber of the mass spectrometer is an

absolute requirement for the analysis of large proteins (2-6). The increased pressure

leads to collisional cooling and focusing of large ions in the ion guides and, therefore,

improved transmission through the ion guides and the TOF (5). In ESI-MS, the ion

signal is proportional to analyte concentration and largely independent of flow rate and

injection volume used for sample introduction. The signal is linear from the limit of

detection (usually pmol/L) to around 10 μmol/L of analyte concentration. For quantitative

measurements, it is important to incorporate an internal standard in the procedure to

170

compensate for losses during sample preparation and variable detection sensitivity of

the MS system. The internal standard should have a structure similar to that of the

analyte and the ideal practice is to synthesize an internal standard by incorporating

stable isotopes on the molecules of interest. When an ideal internal standard is not

available, molecules with similar structure can also be used. Another critical issue in

quantitative ESI-MS is suppression of ionization due to matrix interference. A biological

sample can give significantly lower ionization signals compared to pure standard

solutions with similar analyte concentrations. This phenomenon is the result of high

concentrations of non-volatile materials, such as salts and lipids, present in the spray

with the analyte. To overcome the matrix interference, extensive sample purification

processes are required. However, these elaborate procedures are time-consuming and

can cause poor recovery. A recent development is to use short Liquid Chromatography

(LC) columns (or guard columns) and apply a fast High Pressure Liquid

Chromatography (HPLC) purification (e.g. for 2–5 minutes) prior to MS analysis. The

HPLC serves to separate the non-volatile compounds from the analyte. For HPLC

systems with column-switching capability, the analyte in the biological sample can be

purified and concentrated on separate columns before MS analysis. Unlike many other

techniques which measure one analyte at a time, these techniques can measure

multiple analytes (>40) at one time. In recent years the scope of testing using these

techniques has expanded from toxicological purposes to newborn screening to

hormones, proteins, and enzymes.

171

In recent years a change in the way MS is being used in clinical laboratories has

occurred. In the past MS was commonly used in conjunction with gas chromatograph

(GC). Today it is not uncommon to see MS being coupled to LC in the routine clinical

laboratories. Once considered too expensive and cumbersome to use except in forensic

and reference settings, such systems are now used routinely to generate data for

patient care. Although mass spectrometry has long been recognized as an important

and powerful analytical tool, there were a number of challenges that had to be

overcome to be used in the clinical setting for more than a few special applications. GC-

MS was introduced into the clinical laboratory more than two decades prior to LC-MS.

With the advent of relatively small, inexpensive, and user-friendly LC/MS and LC

tandem MS (LC/MS-MS) systems along with advances in column chemistries the door

has been opened to many analyses not possible with GC/MS (7). Although the initial

capital investment for LC ESI-MS equipment is substantial compared to other routine

clinical laboratory analyzers, its operational costs are low. The cost-effectiveness of this

technique comes from the fact that it can measure multiple analytes at the same time.

This technology can be expected to exert an important influence in how analytes, both

large and small, are detected and quantified in the clinical laboratory service.

Since the first report on the successful measurement of large bio-molecules by

ESI-MS, there has been a revolution in the identification of protein molecules in

biochemical research. MS also found its way into the analysis of hemoglobin (Hb)

analysis. In 1981, Wada et al. pioneered the analysis of tryptic peptides of Hb by MS.

The development of the soft ionization techniques (ESI and MALDI) has made it

possible to use MS to study intact globin chains. In 1990, the 1st application of ESI-MS

172

involving intact Hb chains was reported by Falick et al (8). Since then ESI and MALDI-

TOF MS has become more common in routine Hb variant analysis.

Electrospray ionization is efficient in generating cluster ions for structural

elucidation of macromolecules. This has fostered a new and improved approach (vs.

electrophoresis) for identification and quantification of hemoglobin variants. The use of

MS techniques has led to the discovery of more than 60 new mutations and even the

intact Hb tetramer can be analyzed using a nano ESI-MS technique. Furthermore,

MALDI-TOF MS is a highly sensitive method that enables the analysis of Hb chains

from a single red blood cell. Final identification of a variant is achieved either by

molecular biology techniques or by protein sequence analysis, in which MS now also

occupies a key position. In variants with mutation sites close to the termini of the chain

were identified by ESI-MS/MS of the intact Hb chain. With the understanding of

glycohemoglobin (GHb) structure, an IFCC reference method for glycohemoglobin

assay has been established using ESI-MS. It represented a significant advancement for

the standardization of HbA1c in diabetic monitoring. ESI-MS has also become the

preferred technique for a rapid systematic approach to definitive characterization of Hb

variants. In addition, hemoglobin (Hb) variants need to be identified for the investigation

of hemolytic anemia, methemoglobinemia, sickle cell disease and thalassemia.

Occasionally, these variants are detected incidentally because they interfere with the

measurement of GHb.

173

Identification and quantification of hemoglobin variants

Globin chain analysis is an important tool in phenotype study of hemoglobin

disorders. The majority of hemoglobin variants result from changes in the amino acid

sequence of either the α or non-α globin chains of hemoglobin with the majority of these

changes due to a single point mutation in the globin gene. Substitution, insertion,

deletion or the combination of deletion with insertion of a different amino acid than those

normally present, results in changes to the amino acid sequence.

Worldwide, an estimated 150 million people carry Hb variants (9) and

hemoglobinopathies are the most common inherited disorders, constituting a significant

healthcare problem (10). Hemoglobin (Hb) variants lead to inherited disorders with

variable clinical manifestations. Therefore, reliable detection and identification methods

are essential. Among more than 900 hemoglobin (Hb) variants currently described in

the HbVar database of the globin Gene Server, variants with elongated chains are very

rare (11,12). In this database, Hb variants leading to a charge difference are

significantly over represented compared with neutral Hb variants. This result is

surprising, because only 5 of the 20 amino acids contain either a basic (Lys, Arg, His) or

an acidic (Asp, Glu) side chain, whereas the other 15 amino acid side chains are

uncharged. Thirty-six of 141 amino acids in the α-chain and 38 of 146 residues in the β-

chain are charged residues and the rest are neutral so they cannot be detected by

these traditional analytical techniques, such as ion-exchange HPLC and isoelectric

focusing (IEF) on polyacrylamide gel, as these techniques depend on the presence of

charge differences induced by the mutation. Also in the past, definitive characterization

of Hb variants involved tedious and time-consuming analytical procedures requiring

174

days and even months for completion. Recently, a strategy for rapid definitive

characterization of Hb variants to identify a single mutated; inserted or deleted amino

acid residue was reported using ESI-MS. In case of Hb San Martin

[b6(A3)Glu→Val;b85(F1) Phe→Leu], the second mutation leads to an unstable protein

causing chronic hemolytic anemia in the heterozygous carrier (13). Molecular diagnosis,

achieved by DNA analyses, shows the presence of two mutations, but protein or familial

studies was required to prove that the two mutations are carried by the same allele and

not interacting in trans. The identification by MS methods of a new Hb variant: Hb S-

Clichy [b6(A3)Glu→Val;b8(A5) Lys→Thr], which presents a double mutation located on

the same bT-1 tryptic peptide. This new variant adds the amino acid substitution of Hb

Rio Grande[b8(A5) Lys→Thr] (14) on the same b-globin chain, to that of Hb S.

Difficulties encountered in structural determinations are caused by the presence of two

abnormalities in the same polypeptide chain. Variants with two amino acid substitutions

on the same globin chain as in Hb S-Clichy, demonstrated the importance of including

MS studies.

The procedure comprises the following steps:

I. Molecular weight profiling of intact α and β globin chains by direct ESI-MS on a

500-fold dilution of the whole blood sample. The cluster ion spectrum is then

deconvoluted to a true molecular weight scale using computer software that is

usually supplied with the MS analyzer system. This step can detect Hb variants

with molecular weight difference of more than 6 Da when compared with the wild

type globin chains (15).

175

II. Overnight trypsin digestion for investigation of the amino acid substitution on the

Hb variants. ESI-MS on the tryptic digest can identify the specific peptide

harboring the substituted amino acid.

III. ESI-MS/MS of the target peptide can provide the amino acid sequence of the

peptide and thus the position of the substituted amino acid.

These performances can be applied at different steps of the globin variant

analysis process: either as a screening method or as an additional technique to confirm

the results from classical analytical methods. ESI-MS can also identify 95% of the Hb

variants in over 250 samples with a turn-around-time of not more than 2 days for each

sample, making it a powerful tool for Hb analysis.

It must be considered that the 3-dimensional structure of the globins is

determined principally by the residues that form the interhelical and helix-heme

packings (16), and substitutions in these sites may lead to conformational changes in

the proteins. The substitution effect also depends on the 3-dimensional position, viz.

internal or external. For example, the variant Hb Sun Prairie (130Ala3Pro) is silent in

IEF, whereas Hb Fontainebleu (21Ala3Pro) is detectable (12). The substituted amino

acid is internal in Hb Sun Prairie and external in Hb Fontainebleu. As a very simple

model, the calculation of the isoelectric points(pI)-shifts does not consider

conformational changes that might alter the mobility. Therefore, mutations leading to a

distinct conformational change can diverge from the predicted behavior. Furthermore,

the model cannot predict reliably unstable variants. Nevertheless, pI calculations and

the evaluation of the method-specific detect ability allow the prediction of the number of

176

the currently undetected, silent variants. So it is now recommended that other methods

that are not based on electrophoretic or chromatographic mobility should be applied in

Hb variant analysis. In this regard ESI and MALDI-TOF MS are the suggested methods

that enable the detection of variants when the mass difference between the abnormal

and the wild-type globin chains exceeds 6 Da. This limitation in MS determination is due

to the complexity between the normal and mutated globin chains which can be

overcome by using high resolution instruments (FT-ICR, Orbitrap) or by special

precautions on low resolution instruments. For these low mass differences between

normal and variant globin chains, MS analysis of digested peptides is required.

As calculated by various studies, MS method is able to detect 92% of the

undetected variants. Among MS techniques for studying Hb variants, ESI-MS is the

most frequently used and can be associated with peptide sequencing using tandem MS,

but it often gives multiple charged fragment ions. On the other hand, MALDI-TOF MS

gives single-charge peptide ions and has been used for identification of some single

mutation Hb variants. Indeed, with additional MS analysis of lysate samples 3 new

variants, Hb Zurich-Hottingen, Hb Zurich-Langstrasse and Hb Riccarton were detected

by using ESI-MS. Neither variant had a clinical impact. These neutral variants are

exclusively found by MS and are chromatographically silent. Also in an Hb Malay

sample, only the MS analysis revealed the variant chain, as opposed to cation-

exchange HPLC which identified it as a thalassaemia. Recapitulating, 4 out of 2105

samples (0.2%) or 1% of the abnormal samples would be missed without the use of MS

analysis. In ESI-MS, the sample preparation is very simple and requires only the dilution

of the lysate sample. Two important drawbacks of the MS methods are worth

177

mentioning. First, its insufficient resolution prevents the detection of Hb mutations with

small mass differences of the globin chains. The precision of normal low-resolution

mass measurements is insufficient to distinguish the wild-type chain from several chain

variants, such as Hb C, D, or E. Owing to the isotopic pattern, even high-resolution MS

did not separate globin chains that differed only in 1 or 2 Da from the normal chains

(17). Two intact globin chains are not observed as separate entities in MS unless their

masses differ from one another by more than 6 Da. Second, MS as described here is

only a qualitative technique, and in particular, minor Hb fractions such as HbA1C or

HbA2, which are important for diagnosis of diabetes mellitus or thalassemia,

respectively, cannot be quantified. However high resolution MS enables detection of

variants with low mass difference (<2 Da). Also different signals in the spectrum shows

isotopic pattern.

References

1. Fenn, JB. Electrospray Ionization Mass Spectrometry: How It All Began. J. Biomol. Techl. 13:101–118; 2002.

2. Sanglier S, Leize E, Van Dorsselaer A, Zal F. Comparative ESI-MS study of approximately 2.2 M Da native hemocyanins from deep-sea and shore crabs: from protein oligomeric state to biotope. J. Am. Soc. Mass Spectrom.14:419-429; 2003.

3. Schmidt A, Bahr U, Karas M. Influence of pressure in the first pumping stage on analyte desolvation and fragmentation in nano-ESI MS. Anal. Chem.73:6040-6046; 2001.

4. Tahallah N, Pinkse M, Maier CS, Heck A. The effect of the source pressure on the abundance of ions of noncovalent protein assemblies in an electrospray ionization orthogonal time-of-flight instrument. J. Rapid Commun. Mass Spectrom. 15:596-601; 2001.

5. Chernushevich IV, Thomson BA. Collisional cooling of large ions in electrospray mass spectrometry. Anal. Chem. 76:1754-1760; 2004.

6. Krutchinsky AN, Chernushevich IV, Spicer VL, Ens W, Standing KG. A Collisional damping interface for an electrospray ionization time-of-flight mass spectrometry. J. Am. Soc. Mass Spectrom. 9: 569-579; 1998.

178

7. Plumb RS and Balogh MP. The changing face of LC-MS: from experts tousers. Current Trends in Mass Spectrometry November. 14–19; 2008.

8. Falick AM, Shackleton CH, Green BN, Witkowska HE. Tandem mass spectrometry in the clinicalanalysis of variant hemoglobins. Rapid Commun Mass Spectrom. 4:396–400; 1990.

9. Shimizu A, Nakanishi T, Miyazaki A. Detection and characterization of variant and modified structures of proteins in blood and tissues by mass spectrometry. Mass Spectrom Rev. 25: 686–712; 2006.

10.Daniel YA, Turner C, Haynes RM, Hunt BJ, Dalton RN. Rapid and specific detection of clinically significant haemoglobinopathies using electrospray mass spectrometry-mass spectrometry. Br J Haematol. 130:635–43; 2005.

11.Hardison RC, Chui DH, Giardine B, Riemer C, Patrinos GP, Anagnou N. HbVar: a relationaldatabase of human hemoglobin variants and thalassemia mutations at the globin gene server. Hum Mutat. 19:225–33; 2002.

12.HbVar database. http://globin.cse.psu.edu (accessed July 2011).13.Feliu-Torres A, Eandi-Eberle S, Calvo K, et al. Hemoglobin San Martin: A new

unstable variant associated with Hemoglobin S in an Argentinean boy. Proceedings of the 49th American Society of Hematology Meeting, Atlanta, GA, December 8–11, Blood. 110:3806; 2007.

14.Moo-Penn WF, Johnson, MH, Mc Guffey, JE, Jue, DL, Therrell, BL, Jr. Hemoglobin Rio Grande [b8(A5) LYS→THR]: A new variant found in a Mexican-American family. Hemoglobin. 7(1):91–95; 1983.

15.Wild BJ, Green BN, Stephens AD. The potential of electrospray ionization mass spectrometry for the diagnosis of hemoglobin variants found in newborn screening. Blood Cells Mol Dis. 33:308-317; 2004.

16.Lesk AM, Chothia C. How different amino acid sequences determine similar protein structures: the structure and evolutionary dynamics of the globins. J Mol Biol. 136:225–70; 1980.

17.Peter K, Marlis S, Karin Z, Oliver S, Markus S, Bernd R, Silke SD, Leopold U, Thomas K, Claus WH, Hannes F, and Heinz. T. Mass Spectrometry: A Tool for Enhanced Detection of Hemoglobin Variants. Clinical Chemistry. 54(1): 69–76; 2008.

179

Chapter 4Globin Chain Analysis

4.5 PCR and Sanger Sequencing Elaine Lyon, PhD

Molecular methods are commonly employed to identify hemoglobin variants.

Polymerase chain reaction (PCR) exponentially amplifies regions of DNA allowing direct

genotyping or targeted mutation analysis. Detecting common alpha globin deletions is

accomplished by amplifying over deletion breakpoints or using quantitative methods to detect

copy number changes. Sanger sequencing is considered the gold standard for mutation

detection, and can confirm abnormal hemoglobinopathy and thalassemia variants. Molecular

analysis confirms a diagnosis, detects carrier status, and predicts disease prenatally in high-

risk pregnancies. This section will describe general methods, applications and challenges in

PCR and Sanger sequencing for alpha and beta globin molecular analysis.

4.5.1 Alpha Globin

Two alpha globin genes (HBA1 and HBA2) are present on each chromosome 16,

resulting in a normal copy number of four genes (represented by /). One or both alpha

globin genes may be deleted on a single chromosome, with the severity of disease

corresponding to the overall number of deleted alpha globin genes. If two alpha globin genes

are deleted (alpha-thalassemia trait), it is important to determine whether both genes are

deleted on one chromosome (--/), or if each chromosome contains a single gene deletion (-/-

). If both parents carry a chromosome with two deletions (--/), their offspring are at risk for

Hb Bart hydropsfetalis syndrome (--/--). Common deletions include 3.7kB and 4.2kB deletion

180

which eliminate a single gene, while a 20.5kB deletion and the SEA, MED, FIL and THAI gene

rearrangements delete portions of or completely HBA1 and HBA2. In one assay commonly

used in clinical laboratories, PCR primers are designed to flank the breakpoints, amplifying a

product only when the deletion is present. By multiplexing primers, any of the common

deletions can be detected in a single reaction (1). Amplification products are visualized by gel

electrophoresis (figure 1). Other methods to identify deletions include quantitative PCR

analyses such as multiplexed-ligation probe amplification (MLPA) or high resolution (exonic

level) microarray. These methods are capable of detecting known and previously unknown

alpha globin deletions and alpha globin triplications (2,3). Given the frequency of alpha globin

deletions, a molecular work-up for alpha thalassemia often begins with a test for deletions.

Figure 1. Gel electrophoresis for common alpha globin deletions. Each patient is tested with control primers for HBA2 and LIS genes. In a separate reaction, each patient is tested for a common deletion with a multiplex of deletion-specific primers. M: size marker, C: control primers (HBA2 and LIS1), D: deletion primers. Patient 1: no common deletions, patient 2; 3.7 kB heterozygous, patient 3; 3.7kb/SEA compound heterozygous. Note that in patient 2, the control HBA2 band is not present, as this patient has a deletion of this region on both chromsomes.

The alpha globin genes also harbor sequence variants, and full gene sequencing is also

available, although alpha globin sequencing poses challenges. Sequencing is performed on

181

both genes (HBA1 and HBA2) that are highly homologous, and primers are designed to be

specific for each gene. Sequencing the entire coding region (3 exons for each gene),

intron/exon boundaries, proximal promoter regions, 5’ and 3’untranslated regions, and

polyadenylation signals provides a comprehensive sequencing test. Sequencing should be

combined with deletion analysis because deletions are not detected by sequencing, and an

apparent homozygous sequence variant may be one copy of the variant with a deletion on the

opposite chromosome. Samples homozygous for the 3.7kB deletion may not be able to be

amplified for sequencing, resulting in a failed test. However, the common 3.7kB deletion has

a single functional gene, but mutations have also been described in that fusion gene. (4,5) To

be able to identify a mutation in a chromosome with the 3.7kB deletion, primers must be

designed that will amplify over the deletion breakpoints.

4.5.2 Beta Globin

Molecular genotyping assays targeting common beta globin mutations, (e.g. HbS and

Hb C) are available to confirm the mutations for sickle cell or S/C disease. But due to the

number of mutations that have been described, full gene sequencing identifies any sequence

variant and complements other types of globin analysis for hemoglobinopathies and

thalassemias. The mutation spectrum for beta globin is well characterized, and includes

coding region mutations, splice-site mutations, regulatory mutations, and deletions. Therefore

a comprehensive assay consists of sequencing the three exons of the HBB gene, the

intron/exon boundaries, the proximal promoter region, the 5’ and 3’ untranslated regions

(UTR), and known pathogenic deep intronic mutations (e.g. IVS-II-654, IVS-II-705 and IVS-II-

745). Large beta globin deletions or delta-beta fusion genes (e.g. Lepore,) will not be detected

182

by a sequencing assay designed only for HBB, and require a different analysis. Similar to

methods to detect alpha globin deletions, primers can be designed to amplify over the

breakpoints. One example of this is the 619 base pair (bp) deletion found in Indian and other

Asian populations (6). Recently, novel beta globin deletions have been detected by other

methods, such as exonic-level microarrays or MLPA (2,7). Specific large deletions in the

beta globin gene cluster is one of two molecular mechanisms that can result in HPFH. The

other mechanism is point mutations in the promoters of the gamma globin genes (HBG1 and

HBG2).

4.5.3 Sequencing

Sequencing assays begin with PCR for the regions to be interrogated. Primers are

designed to avoid known variants at their 3’ end which would prevent polymerase extension,

resulting in a drop-out of that allele (8). PCR products are treated with ExoSAP (exonuclease

1 from shrimp alkaline phosphatase) to remove excess primers.

PCR primers may be tagged with a M13 sequence which allows sequencing of all amplicons

from the same M13 primers. Alternatively, a second set of sequencing primers internal to the

PCR primers may be used. The sequencing reaction utilizes fluorescently labeled di-

deoxynucleotides (ddATP, ddTTP, ddCTP and ddGTP, collectively refered to as ddNTPs)

which lack a 3’ hydroxyl group on the sugar residue and prevent the newly synthesized product

from extending to the next base when incorporated into the product. Sanger sequencing is

performed as a linear rather than exponential amplification, with separate sequence results

each from the 5’ and 3’ directions (bi-directionally). After the sequencing reactions, the

183

products are again processed to remove excess primers. Sephadex columns are often used

to bind the sequencing products, which are eluted as purified products.

Sequencing products are separated by capillary electrophoresis using a polymer with a single-

base resolution. The last base of each fragment is the ddNTPs with a fluorescent dye which is

incorporated into the product. The sequence is visualized as an electropherogram and aligned

to a reference sequence. The difference between the reference and the patient sequence are

examined to determine the type of mutation present. To accurately identify the mutation, it

should be identified in both the 5’ and 3’ direction.

4.5.4 Reporting Sequence Variants

In standard nomenclature, nucleotides are numbered from the ATG of the start codon.

The protein position is predicted and standard nomenclature is from the methionine of the

translated product, but common or traditional nomenclature (also known as legacy

nomenclature) may be from the mature protein (gene reviews for beta and alpha). For alpha

and beta globin, the traditional nomenclature differs by one amino acid than the standard

nomenclature, representing the cleavage of methionine. Reports should clarify if they are

using the standard or the common (i.e. amino acids numbered from the mature protein).

A variant is described as to whether it is structural (hemoglobinopathy) variant or

quantitative (thalassemia variant). For example, a beta thalassemia mutation is

classified as a B(0) [the absence of beta chains] or B(+) [reduced amount of beta

chains] mutation. One copy of a thalassemia mutation is consistent with beta

thalassemia minor, while two copies on opposite chromosomes are consistent with beta

thalassemia major. On occasion, two mutations may occur on the same chromosome.

184

Sequence analysis can’t determine the phase of two mutations (whether on the same or

opposite chromosomes). HPLC or parental studies may be necessary to evaluate

phase. Over ten complex variants with the Glu6Val variant are listed in the globin gene

server (6). One example is Hb S-Oman, with Glu6Val and Glu121Lys variants on the

same chromosome. The standard nomenclature for the nucleotide changes is HBB:c.

[20A>T;364G>A] (6). These two variants are also seen alone as in Hb S and Hb 0-

Arab. The combination of these two mutations on opposite chromosomes is consistent

with severe sickle cell disease, whereas if they are on the same chromosome, the

individual is a carrier of an HBB mutation.

4.5.5 DNA Sequence Traces:

Several sequencing electropherograms are presented to illustrate the application. The

first shows the common alpha globin variant Constant Spring (Hb CS). This alpha +

thalassemia variant changes the normal termination codon to an elongated 3’ end of the

protein. The second shows a sequence with two beta globin mutations. Information from

family or other laboratory studies could determine if these two variants are on the same or

different chromosomes.

185

Figure 2.Alpha globin sequencing. An apparent homozygous Hb CS variant (c.427T>C, Term>Gln) is detected (arrows). The sequence from the patient (Forward and Reverse) are compared against a reference sequence. Differences between the reference and patient sequences are shown in the middle panes.

186

Figure 3.Beta globin sequencing. A compound heterozygous genotype is detected. The first mutation is Hb S (c.20A>T, Glu6Val), while the second affects a splice site resulting in a beta(0) thalassemia mutation. (c.92+1G>A). The yellow arrow in the electropherogram indicates the exonic region. The sequence from the patient (Forward and Reverse) are compared against a reference sequence. Differences between the patient and reference sequences are shown in the middle panes.

4.5.6 Conclusion

Molecular analysis confirms the familial variant in individuals who are carriers of or

affected with globin gene variants. In prenatal analysis, molecular studies are often the most

direct method to predict the status of a fetus. If molecular testing is used prenatally, the

parents should first be tested to identify the familial mutations. In addition, amniotic fluid,

amniocyte or chorionic villi cell cultures should be tested for contamination from the mother. If

187

the samples show maternal cell contamination, the results may not accurately reflect the fetus’

genotype and a second sample should be obtained.

The alpha and beta loci have complex structures that lead to a variety of molecular

anomalies, such as sequence variants, and large gene rearrangements resulting in deletions

or duplications. Because many mutations in HBA1, HBA2 and HBB are well understood, the

interpretations are typically straight forward. However, because these loci have complex

structures that lead to a variety of molecular anomalies, molecular results should be combined

with clinical, family and other laboratory findings.

188

References

1. Tan AS, Quah TC, Low PS, Chong SS. A rapid and reliable 7-deletion multiplex polymerase chain reaction assay for α-thalassemia. Blood. 2001; 98(1):250–251.

2. Phylipsen M, Chaibunruang A, Vogelaar IP, Balak JR, Schaap RA, Ariyurek Y, Fucharoen S, den Dunnen JT, Giordano PC, Bakker E, Harteveld CL. Fine-tiling array CGH to improve diagnostics for α- and β-thalassemia rearrangements. Hum Mutat. 2012 Jan; 33(1):272-80.

3. Galanello R, Cao A. Alpha-Thalassemia. 2005 Nov 1 [Updated 2011 Jun 7]. In: Pagon RA, Bird TD, Dolan CR, et al., editors. GeneReviews™ [Internet]. Seattle (WA): University of Washington, Seattle; 1993-. Available from: http://www.ncbi.nlm.nih.gov/books/NBK1435/accessed 10-04-12

4. Zhao P, Buller-Burckle AM, Peng M, Anderson A, Han ZJ, Gallivan MV. Secondary mutation (c.94_95delAG) in a -α3.7 allele associated with Hb H disease in two unrelated African American individuals homozygous for the -α(3.7) deletion (-α3.7/-α3.7T). Hemoglobin. 2012; 36(1):103-7.

5. Brennan SO, Chan T, Duncan J. Novel α2 gene deletion (c.349_359 del GAGTTCACCCC) identified in association with the -α3.7 deletion. Hemoglobin. 2012; 36(1):93-7.

6. Hardison RC, Chui DHK, Giardine B, et al. HbVar: a relational database of human hemoglobin variants and thalassemia mutations at the globin gene server. Hum Mutat 2002; 19: 225-33 http://globin.cse.psu.edu/accessed 10-04-2012

7. Mikula M, Buller-Burckle A, Gallivan M, Sun W, Franklin CR, Strom CM.The importance of β globin deletion analysis in the evaluation of patients with β thalassemia.Int J Lab Hematol. 2011 Jun;33(3):310-7

8. Pont-Kingdon G, Gedge F, Wooderchak-Donahue W, Schrijver I, Weck KE, Kant JA, Oglesbee D, Bayrak-Toydemir P, Lyon E; Biochemical and Molecular Genetic Resource Committee of the College of American Pathologists. Design and analytical validation of clinical DNA sequencing assays.Arch Pathol Lab Med. 2012 Jan;136(1):41-6.

189

Chapter 5

Alpha and Beta Thalassemia Herbert L. Muncie, Jr., MD

Alpha (α) and beta (β) thalassemia are hematological disorders that are the

result of a decreased or absent synthesis of a globin chain.1 These genetic alterations

may have been the result of selective pressure from Plasmodium falciparum malaria

from which thalassemia carriers are relatively protected from invasion.2, 3 The altered

globin chain synthesis can be asymptomatic or can cause severe hemolytic anemia and

even death.

5.1 Epidemiology

The thalassemias are prevalent in the tropical and subtropical regions of the

world and affect men and women equally. Alpha thalassemia is found more often in

persons of African or Southeast Asian descent and β-thalassemia occurs more often in

persons of Mediterranean, African or Southeast Asian descent. Thalassemia trait can

be found in 5 to 30 percent of these populations.4 An estimated 1.5% of the global

population are β-thalassemia carriers but only approximately 200,000 people are alive

with β-thalassemia major.5, 6

5.2 Pathophysiology

Hemoglobin has an iron-containing heme ring and four globin chains: normally

two alpha and two nonalpha. The composition of these four globin chains determines

190

the hemoglobin type. The predominant in utero hemoglobin, fetal hemoglobin (Hb F),

has two α and two gamma (γ) chains (α2 / γ2). Adult hemoglobin A (Hb A) has two α and

two β chains (α2/β2) and hemoglobin A2 (HbA2) has two α and two delta (δ) chains

(α2/δ2).

The transition from γ-globin synthesis (Hb F) to β-globin synthesis (Hb A) begins

before birth. Therefore, at birth approximately 20% to 30% of hemoglobin is Hb A and

the remainder is HbF.7 This transition continues and is usually completed from 6 to 24

months of age. At that time normal children will have mostly Hb A (>96%), small

amounts of Hb A2 (2.0 – 3.4%) and very small amounts of Hb F (< 1%).8

5.3 Alpha Thalassemia

Alpha thalassemia occurs when there is reduced or absent α-globin chain

synthesis with subsequent excess β-globin chains.9, 10 Two genes on chromosome 16

control α-globin synthesis (αα/αα). Most defects are due to deletions of one or more of

these genes. Since two genes on each chromosome 16 control the production of α-

globin chains, there are four possible phenotypical presentations. With a single gene

deletion (-α/αα) the result is α-thalassemia silent carrier state which is asymptomatic

with normal hematological indices. With two gene deletions (-α/-α; --/αα) the result is α-

thalassemia trait (minor) which frequently causes microcytosis without anemia. If three

genes are deleted (--/-α) there will be significant amounts of hemoglobin H (Hb H)

consisting of four β-globin chains (β4). The result of significant amounts of Hb H is α-

thalassemia intermedia (Hb H disease), which causes hemolytic anemia, microcytosis

and splenomegaly. While most cases of Hb H disease are deletional, non-deletional

191

forms do occur and are often more symptomatic. Hemoglobin Constant Spring is an α-

globin chain variant that is longer than normal and produced in only small quantities. It

therefore behaves in a similar manner to an alpha gene deletion.11 When Hb Constant

Spring is inherited with a 2 alpha gene deletion, the condition may be referred to as Hb

H / Constant Spring. Finally if all four genes are deleted (--/--) the result will be

significant amounts of hemoglobin Bart’s (Hb Bart’s) with four gamma chains (γ4). With

increased Hb Bart’s and total absence of Hb F, α-thalassemia major results leading to

hydrops fetalis, which is incompatible with life.12

5.4 Beta Thalassemia

Beta thalassemia occurs when there is reduced or absent β-globin chain

synthesis with subsequent excess α-globin chains.3, 6 Most often a mutation is the

genetic defect, with more than 200 reported; a deletion is quite rare. One gene on each

chromosome 11 controls the production of β-globin chains (β,β), therefore, there are

two phenotypical presentations. If a child inherits one normal gene from one parent

(β/β) and a defective gene from the other parent (-/β), the result is β-thalassemia trait

(minor) which causes an asymptomatic mild microcytic anemia. If both genes are

defective, the result depends on the degree they are deficient in β-globin chain

production. If β-globin chain production is severely reduced, the person will have β-

thalassemia major (Cooley anemia). Most individuals with β-thalassemia are

asymptomatic at birth because of the presence of significant amounts of Hb F. As the γ-

globin chain synthesis decreases, infants may experience symptoms starting at six

192

months of age. If the β-globin chain synthesis is only partially reduced, the person will

have β-thalassemia intermedia with less severe symptoms and survival beyond 20

years of age without life-long blood transfusions.

5.5 Diagnosis

Except for α or β thalassemia major, the diagnosis of thalassemia is usually

made incidentally when a patient is found to have microcytosis with or without anemia.

The most common etiologies for microcytic anemia are iron deficiency, thalassemia,

lead toxicity and sideroblastic anemia. The patient’s medical history, mean corpuscular

volume (MCV), red cell count and the red cell distribution width (RDW) can help exclude

many of these etiologies (Table 1). The MCV in β-thalassemia trait is usually lower than

in α-thalassemia trait. In Hb H disease the MCV will be as low as 64 fl.13 Mentzer index

(MCV/red blood cell count) was proposed (which is not true in children) to predict the

likelihood of thalassemia. If the ratio is > 13, iron deficiency is the likely etiology

whereas thalassemia is associated with a value < 13. An exact ratio of 13 would be

uncertain.14

The RDW can be helpful in distinguishing thalassemia from iron deficiency and

sideroblastic anemia. With iron deficiency or sideroblastic anemia the RDW is almost

always elevated while it is elevated in approximately 50% of thalassemia trait patients.15

Therefore, with a microcytic anemia, if the RDW is normal the diagnosis is usually

thalassemia trait. However, if the RDW is elevated additional tests to evaluate for iron

deficiency and sideroblastic anemia will be needed.16

193

A serum ferritin level is the best single test to rule out iron deficiency in the absence of

inflammation.17 Serum iron, total iron binding capacity and transferrin may not be

needed in distinguishing iron deficiency from thalassemia. A peripheral smear or bone

marrow aspirate can rule out sideroblastic anemia. If lead toxicity is suspected, a serum

lead level will be needed. And finally a hemoglobin electrophoresis/ HPLC can evaluate

for hemoglobinopathies and may confirm the diagnosis of thalassemia.

In the past the diagnosis of α-thalassemia in adults was by exclusion. If a patient

had a microcytic (MCV < 80 fl) hypochromic (MCH < 27 pg) anemia, normal iron studies

and a normal hemoglobin electrophoresis and Hb A2 it was assumed the patient had α-

thalassemia trait (minor). Now high-performance liquid chromatography (HPLC) can

often provide an accurate diagnosis in neonates. In infants, if increased amounts of Hb

H or Hb Bart’s are found in cord blood or neonatal blood, the diagnosis of α-thalassemia

is confirmed. Infants who are silent carriers may have a slightly increased amount of Hb

Bart’s (1 – 2%) at birth while infants with α-thalassemia trait have a moderately

increased amount (5 – 6%).10

In adults with β-thalassemia trait (minor) the HPLC or hemoglobin

electrophoresis will show reduced Hb A levels (<96%), elevated Hb A2 levels (>3.5%)

and often elevated Hb F levels (1.0 – 4.0%).3, 4 However, a normal amount of Hb A2

does not exclude thalassemia in some patients. Patients with Iron deficiency often have

lower Hb A2 levels and the Hb A2 quantification may need to be repeated after iron

supplement therapy.18 Genetic coinheritances may reduce Hb A2 production making it

difficult to diagnosis β-thalassemia. If the Hb A2 level is below normal (<2.5%) but with

194

a normal Hb F level and microcytosis, the patient has α-thalassemia intermedia, i.e. Hb

H disease (Table 2)19. Reviews on measuring and interpreting Hb A2 and Hb F levels

are available.8, 20

Beta thalassemia major is diagnosed during the infant’s first year of life. The

infant usually displays growth retardation, pallor, irritability and later jaundice with

abdominal swelling. Children who develop these symptoms after their first birthday will

be diagnosed with β-thalassemia intermedia.

5.6 Treatment

Patients with α or β thalassemia trait (minor) require no treatment or regular long-

term follow-up. While these patients have a microcytic anemia they are not iron deficient

and should not be given iron supplements. If iron deficiency did develop, iron

supplements would be appropriate.4, 21

Beta thalassemia major requires treatment with blood transfusions starting as

early as six months of age. Transfusions correct the anemia, suppress erythropoiesis

and inhibit intestinal iron absorption. Transfusions are initiated either when the

hemoglobin is < 7 g/dl for more than 2 weeks (without another etiology) or if other

factors such as facial changes, poor growth, bony expansion or splenomegaly occur.

Without blood transfusions these patients would not survive into adulthood. They will

need periodic (every 2 - 4 weeks) transfusions (lifelong) to maintain their hemoglobin

higher than 9.5 g/dL.4, 22 The post-transfusion hemoglobin goal is 13 – 14 g/dl. Beta

thalassemia intermedia patients require transfusions only when their reduced

hemoglobin interferes with their quality of life or it effects their growth and development.

195

Transfusions will occasionally be needed for Hb H disease depending on the severity of

the condition.

Patients who receive frequent blood transfusions or who have increased

intestinal iron absorption will eventually develop iron overload since their bodies cannot

actively eliminate excess iron. To treat the iron overload, β-thalassemia major patients

will require iron chelation therapy starting either around age 5, when the serum ferritin

exceeds 1000 ng/ml, or after they have had 10-20 transfusions.23 A liver biopsy is the

gold standard for iron overload diagnosis.24 Beta thalassemia intermedia patients will

begin chelation when the serum ferritin is > 300 mcg/L.3 Deferoxamine either

subcutaneously or intravenously has been the chelation treatment of choice although

long-term compliance is difficult.25 An oral alternative is deferasirox (Exjade®).26 Iron

chelation therapy is relatively benign although it is time consuming and expensive

(Table 3).

The only curative therapy for patients with β-thalassemia major is a bone marrow

transplant. In low-risk patients with no hepatomegaly, no portal fibrosis on liver biopsy

and not receiving regular chelation therapy, hematopoietic stem cell transplantation

usually has excellent results.4

5.7 Complications

Patients with α or β thalassemia trait (minor) have no complications. Patients with

Hb H disease, β-thalassemia major or β-thalassemia intermedia have hemolysis, growth

retardation and skeletal abnormalities as a consequence of the over stimulation of the

196

bone marrow and ineffective erythropoesis.21, 27 Infants with significant amounts of Hb

Bart’s usually die in utero or shortly after birth due to autoimmune hydrops fetalis.

Because of the need for multiple blood transfusions in β-thalassemia major or in some

cases of Hb H disease and the increased intestinal iron absorption with β-thalassemia

intermedia, patients develop iron overload which damages visceral organs (liver,

spleen, endocrine organs) and the heart which is the primary cause of early death.28

Splenomegaly invariably develops in symptomatic thalassemia and can worsen the

anemia. The risk of hepatocellular carcinoma is increased due to iron overload hepatic

damage, longer survival and viral infection with hepatitis B and/or C.29 Gallstones are

more prevalent with β-thalassemia intermedia than with β-thalassemia major.

Beta thalassemia major and intermedia cause a hypercoaguable state.30 This effect

increases the risk of thromboembolic events especially after splenectomy.31

Osteoporosis was found in 51% of patients over age 12 with β-thalassemia major.32

5.8 Other Treatment Issues

5.8.1 Hypersplenism

Splenectomy is required for patients whose splenomegaly causes a marked

increase in their need for blood transfusions, i.e. the annual red cell requirement

exceeds 180 – 200 ml/kg.6 Because of the importance of the spleen in clearing bacteria

and preventing sepsis, the surgery is not done until the patient is at least 4 years old.

One month prior to the surgery the child should be given the pneumococcal

polysaccharide vaccine. They should also receive the pneumococcal conjugate vaccine

197

series if they had not received it during infancy. For the first two years after the surgery

patients should take penicillin 250 mg twice a day. For children the antibiotic prophylaxis

continues until age 16.33 Gallbladder removal should be considered if gallstones are

present34 at the time of splenectomy.

5.8.2 Endocrinopathies

While growth retardation can occur with thalassemia, growth hormone therapy

has limited effectiveness and is not recommended. If hypogonadism develops,

hormonal therapy is effective.35 Bone mineral density has been increased with

alendronate, pamidronate and zolendronate; however, studies evaluating fracture

reduction are needed.36

5.8.3 Pregnancy

Couples from high risk ethnic groups should be encouraged to seek

preconception genetic counseling.37 Individuals with a low MCV (<80fl) and MCH (<27

pg) could be assessed with hemoglobin electrophoresis / HPLC.12 An efficient way to

identify mutations is to study their parent’s hematology and screen them for single

mutations.38

For couples, if both partners have β-thalassemia trait, their child will have a one

in four chance of having β-thalassemia major and a two in four chance of β-thalassemia

trait.(Figure 1) With four genes controlling the expression of α-globin chains, the

inheritance pattern is more complex. If two genes are defective, the phenotype is

influenced by whether the defective genes are on the same chromosome (cis) or

198

different chromosomes (trans), e.g. if one parent is an α-thalassemia silent carrier (-α,

αα) and one parent has α-thalassemia trait [(cis),(--,αα)], they have a one in four chance

their child will have Hb H disease. Whereas, if the α-thalassemia trait parent’s defect is

trans (-α, -α), their children have no risk of Hb H disease (Figure 2).

Once pregnancy occurs, patients should be counseled regarding prenatal

diagnostic testing options. An amniocentesis at approximately 15 weeks gestation or a

chorionic villus sample (CVS) obtained at 10 – 11 weeks gestation can detect point

mutations or deletions utilizing polymerase chain reaction (PCR) testing. Other

diagnostic options include DNA analysis of fetal cells obtained by amniocentesis and in

the future analysis of maternal blood fetal cells.39, 40 If Hb Bart’s is detected, the mother

has an increased risk of pre-eclampsia and postpartum hemorrhage.

For couples using in vitro fertilization, preimplantation genetic testing is available.41

5.8.4 Cardiac

When iron overload occurs, cardiac infiltration and death are significant

concerns. Serum ferritin levels have been used to predict cardiac complications with

improved survival if levels are kept below 2,500 ng per ml (2500 mcg per L).42 Ferritin

levels are unreliable when significant liver disease develops.43

5.8.5 Hypercoagulopathy

While the risk of thromboembolic events in patients with β-thalassemia major or

intermedia is increased, no trials have evaluated prevention of these complications with

anticoagulants. A consensus recommendation for patients with a thrombosis history is

199

prophylactic treatment with low molecular weight heparin especially before surgery and

during pregnancy. Because estrogen containing contraceptives may increase the risk of

thrombosis, an alternative form of contraception should be recommended for these

women during their reproductive years.

5.8.6 Psychosocial

The impact on a patient and their family of a chronic disease such as β-

thalassemia major that requires lifelong treatment is significant. Providing education

regarding the inheritance patterns, the prenatal diagnostics options and the need for

psychological support may help patients better manage their disease. However, based

on the available evidence no specific therapy or combination of therapies can be

recommended.44

5.8.7 Vitamin Deficiencies

With the increase in erythropoesis some patients may develop folic acid

deficiency. For these patients a 1 mg folic acid supplement daily is recommended.34

However, patients receiving frequent transfusions rarely have this problem. While

oxidative stress may contribute to the complications, the use of antioxidants has not

improved the anemia nor reduced the morbidity or mortality of thalassemia.34 If a

transfusion dependent patient has proven vitamin C deficiency, supplements are

recommended.

200

5.8.8 Prognosis

Beta thalassemia major patients live an average of 17 years and usually die by

age 30. With regular blood transfusions and compliance with chelation therapy, their life

can extend beyond age 40.6 Their deaths are commonly caused by cardiac

complications of iron overload.28 Thalassemia trait patients have a normal life

expectancy.

Sources of Additional Information:

Cooley's Anemia Foundation http://www.cooleysanemia.org or

http://www.thalassemia.org

Thalassemia International Federation – www.thalassaemia.org.cy

201

Table 1 – Hematologic Indices for Iron deficiency and thalassemias

Indices Iron Deficiency α-thalassemia β-thalassemia

trait

β-thalassemia

major

MCV (abnormal <80

fl in adults; < 70 fl age

6 months – 6 years; <

75 fl age 7 – 12 years)

Low Low Low Low

RDW (Adult normal -

11.5 – 14.5%)

High Normal Normal,

occasionally

high

Normal,

occasionally high

Ferritin (adult normal

– male 20 – 250

ng/ml; female 10 –

120 ng/ml)

Low Normal Normal Normal

Mentzer Index –

Children (MCV/RBC

count)

> 13 < 13 < 13 < 13

Hb electrophoresis

(Adult normal’s –

HbA - > 96%

HbA2 – 2.5 -3.5%

HbF - < 1%)

Normal (may

have reduced

HbA2 before

iron therapy is

given)

Adults : normal

Newborns:

cord blood may

have HbH or

Hb Bart

Reduced HbA ,

increased

HbA2, and

increased HbF

Reduced HbA,

increased HbA2,

and increased

HbF

202

Table 2 Hb A2 levels with iron deficiency and thalassemia9

Diagnosis HbA2 level

Normal 2.5 – 3.5 %

Iron deficiency 1.6 – 3.2 %

α-thalassemia silent carrier or trait (minor) 2.0 – 3.2 %

HbH disease 1 – 2.4 %

β-thalassemia trait (minor) > 4.0 %

β-thalassemia major > 4.0 %

203

Table 3 – Chelation Therapy Treatment Options

Therapeutic

agent

Route of

administration

Dosage Frequency of therapy

Desferoxamine Subcutaneous

infusion over 8 - 12

hours

Adults - 30 – 50 mg/kg

Children - 20 – 40

mg/kg

5 – 7 days/week

Deferasirox Oral 20 - 30mg/kg/day Once a day

Deferiprone

(only available

in the US

through FDA

Treatment Use

Program)

Oral 75 – 100 mg/kg/day 3 times/day

204

Figure 1 – Beta thalassemia trait genetics

Mother Father (-,β) (-,β)

β-thalassemia trait β-thalassemia trait

Children: (-,-) (-,β) (-,β) (β ,β) β-thalassemia major β-thalassemia trait β-thalassemia trait

Normal or intermedia

Note: Shaded symbols indicate an abnormal β-globin gene on chromosome 11.

205

X

206

207

References

1. Muncie HL Jr, Campbell J. Alpha and beta thalassemia. Am Fam Physician 2009; 80 (4): 339-344.

2. Mantikou E, Arkesteijn SG, Beckhoven JM, Kerkhoffs JL, Harteveld CL, Giordano PC. A brief review of newborn screening methods for hemoglobinopathies and preliminary results selecting beta thalassemia carriers at birth by quantitative estimation of the HbA fraction. Clinical Biochemistry 2009; 42(18): 1780-1785.

3. Cao A, Galanello R. Beta-thalassemia. Genetics in Medicine 2010; 12(2): 61-76.4. Rund D, Rachmilewitz E. Beta-thalassemia. N Engl J Med. 2005; 353(11): 1135-

1144.5. Thalassemia International Federation. Thalassemia. Available at:

http://www.thalassaemia.org.cy/index.html. Accessed 04/10, 2011.6. Galanello R, Origa R. Beta-thalassemia. Orphanet J Rare Dis 2010; 5:11.7. Richardson M. Microcytic anemia. Pediatr Rev 2007; 28 (1): 5-14.8. Mosca A, Paleari R, Ivaldi G, Galanello R, Giordano PC. The role of haemoglobin

A(2) testing in the diagnosis of thalassaemias and related haemoglobinopathies. J Clin Pathol 2009; 62:13-17.

9. Harteveld CL, Higgs DR. Alpha-thalassaemia. Orphanet J Rare Dis 2010; 5:13.10. Galanello R, Cao A. Alpha-thalassemia. Genetics in Medicine 2011; 13 (2): 83-88.11. Chen FE, Ooi C, Ha SY, et al. Genetic and clinical features of hemoglobin H

disease in Chinese patients. N Engl J Med. 2000; 343(8): 544-550.12. Leung TN, Lau TK, Chung TKH. Thalassaemia screening in pregnany. Curr Opin

Obstet Gynecol 2005; 17 (2): 129-134.13. Origa R, Sollaino MC, Giagu N, et al. Clinical and molecular analysis of

haemoglobin H disease in Sardinia: Haematological, obstetric and cardiac aspects in patients with different genotypes. Br J Haematol 2007; 136(2): 326-332.

14. Mentzer WC,Jr. Differentiation of iron deficiency from thalassaemia trait. Lancet 1973; 1(7808): 882.

15. Flynn MM, Reppun TS, Bhagavan NV. Limitations of red blood cell distribution width (RDW) in evaluation of microcytosis. Am J Clin Pathol 1986; 85(4): 445-449.

16. Marsh WL Jr, Bishop JW, Darcy TP. Evaluation of red cell volume distribution width (RDW). Hematol Pathol 1987; 1(2): 117-123.

17. Guyatt GH, Oxman AD, Ali M, Willan A, McIlroy W, Patterson C. Laboratory diagnosis of iron-deficiency anemia: An overview. Journal of General Internal Medicine 1992; 7(2) : 145-153.

18. Kattamis C, Lagos P, Metaxotou-Mavromati A, Matsaniotis N. Serum iron and unsaturated iron-binding capacity in the -thalassaemia trait: their relation to the levels of haemoglobins A, A 2 , and F. J Med Genet 1972; 9(2): 154-159.

19. Van Delft P, Lenters E, Bakker-Verweij M, et al. Evaluating five dedicated automatic devices for haemoglobinopathy diagnostics in multi-ethnic populations. Int J Lab Hematol 2009; 31(5): 484-495.

20. Mosca A, Paleari R, Leone D, Ivaldi G. The relevance of hemoglobin F measurement in the diagnosis of thalassemias and related hemoglobinopathies. Clin Biochem 2009; 42(18): 1797-1801.

208

21. Olivieri NF. The beta-thalassemias. N Engl J Med 1999; 341(2): 99-109.22. Cazzola M, Borgna-Pignatti C, Locatelli F, Ponchio L, Beguin Y, De Stefano P. A

moderate transfusion regimen may reduce iron loading in beta-thalassemia major without producing excessive expansion of erythropoiesis. Transfusion 1997; 37(2): 135-140.

23. Roberts DJ, Brunskill SJ, Doree C, Williams S, Howard J, Hyde CJ. Oral deferiprone for iron chelation in people with thalassaemia. Cochrane Database of Systematic Reviews 2007: 3: CD004839.

24. Angelucci E, Brittenham GM, McLaren CE, et al. Hepatic iron concentration and total body iron stores in thalassemia major. N Engl J Med 2000; 343(5): 327-331.

25. Delea TE, Edelsberg J, Sofrygin O, et al. Consequences and costs of noncompliance with iron chelation therapy in patients with transfusion-dependent thalassemia: a literature review. Transfusion 2007; 47(10): 1919-1929.

26. Deferasirox (exjade): A new iron chelator. Drugs Ther. Med Lett 2006; 48(1233): 35-36.

27. Parano E, Pavone V, Di Gregorio F, Pavone P, Trifiletti RR. Extraordinary intrathecal bone reaction in beta-thalassaemia intermedia. Lancet 1999; 354(9182): 922.

28. Modell B, Khan M, Darlison M. Survival in beta-thalassaemia major in the UK: Data from the UK Thalassaemia Register. Lancet 2000; 355(9220): 2051-2052.

29. Borgna-Pignatti C, Vergine G, Lombardo T, et al. Hepatocellular carcinoma in the thalassaemia syndromes. Br J Haematol 2004; 124(1): 114-117.

30. Eldor A, Rachmilewitz EA. The hypercoagulable state in thalassemia. Blood 2002; 99(1): 36-43.

31. Tso SC, Chan TK, Todd D. Venous thrombosis in haemoglobin H disease after splenectomy. Aust N Z J Med 1982; 126): 635-638.

32. Jensen CE, Tuck SM, Agnew JE, et al. High prevalence of low bone mass in thalassaemia major. Br J Haematol 1998; 103(4): 911-915.

33. Davies JM. Barnes R. Milligan D. British Committee for Standards in Haematology. Working Party of the Haematology/Oncology Task Force. Update of guidelines for the prevention and treatment of infection in patients with an absent or dysfunctional spleen. Clin Med 2002; 2(5): 440-443.

34. Borgna-Pignatti C. Modern treatment of thalassaemia intermedia. Br J Haematol 2007; 138(3): 291-304.

35. De Sanctis V. Growth and puberty and its management in thalassaemia. Horm Res. 2002; 58 Suppl 1: 72-79.

36. Gaudio A, Morabito N, Xourafa A, et al. Bisphosphonates in the treatment of thalassemia-associated osteoporosis. J Endocrinol Invest 2008; 31 (2): 181-184.

37. ACOG Practice Bulletin No. 78: hemoglobinopathies in pregnancy. ACOG Committee on Obstetrics. Obstetrics & Gynecology 2007; 109(1): 229-237.

38. Old JM. Screening and genetic diagnosis of haemoglobin disorders. Blood Rev. 2003;17(1): 43-53.

209

39. Li Y, Di Naro E, Vitucci A, Zimmermann B, Holzgreve W, Hahn S. Detection of paternally inherited fetal point mutations for beta-thalassemia using size-fractionated cell-free DNA in maternal plasma. JAMA 2005; 293(7): 843-849.

40. Papasavva T, Kalakoutis G, Kalikas I, et al. Noninvasive prenatal diagnostic assay for the detection of beta-thalassemia. Ann N Y Acad Sci 2006; 1075: 148-153.

41. Braude P, Pickering S, Flinter F, Ogilvie CM. Preimplantation genetic diagnosis.[erratum appears in nat rev genet. 2003 feb;4(2):157.]. Nature Reviews Genetics. 2002;3:941-953.

42. Hoffbrand AV, Cohen A, Hershko C. Role of deferiprone in chelation therapy for transfusional iron overload. Blood 2003;102(1):17-24.

43. Brittenham GM, Cohen AR, McLaren CE, et al. Hepatic iron stores and plasma ferritin concentration in patients with sickle cell anemia and thalassemia major. Am J Hematol 1993; 42(1): 81-85.

44. Anie KA, Massaglia P. Psychological therapies for thalassaemia. Cochrane Database of Systematic Reviews 2001; 3: CD002890.

210

Chapter 6

Neonatal Screening for HemoglobinopathiesZia Uddin, PhD

With the technical support of Patrick Hopkins, Joseph Quashnock,Aigars Brants, Christine Moore, D’Andra Morin, Rachel Lee, Mahin Azimi, and Bonifacio Dy

6.1 Introduction

A gratifying achievement of my professional career was the organization of an

interdisciplinary conference on “Perinatal Care and Neonatal Screening” in 1978 at

South Macomb Hospital (now a part of St. John Providence Health System), Warren,

Michigan.

CLINICAL CHEMISTRY, VOL 24, No. 7, 1978

“Seven specialists eminent in the respective fields recently presented their research work to a group of obstetricians, gynecologists, pediatricians, pathologists, and clinical chemists from Michigan and Windsor (Canada) at a recent interdisciplinary conference in Warren, Michigan, sponsored by Detroit-Macomb Hospitals Association.

Speakers and their topics were: Dr. Joseph Bieniarz, “Amniocentesis in Perspective: Diagnostic Value of Ultrasonography and Protocol for

Monitoring High Risk Pregnancy.” Dr. Keith H. Marks, “Elective Delivery of the Term Fetus: An Obstetrical Hazard.” Dr. John. L. Kitzmiller,

“Management and Outcome of Pregnancy in Diabetes Mellitus.” Dr. Norman F. Gant, “Supine Hypertension Test and the Clinical Management of Pregnancy-Induced Hypertension.” Dr. Thomas P. Foley, “Neonatal Screening for Congenital Hypothyroidism and Clinical Treatment.” Dr. Samuel Meites, “Clinical Chemistry Laboratory in Neonatology.” Mrs. Ann Bennett, “Public Health and Neonatal Screening.”

For additional information write Zia Uddin, Ph.D., Perinatal Care & Neonatal Screening Conference, South Macomb Hospital, 11800 Twelve Mile Road, Warren, Mich. 48093”.

211

After this conference I started neonatal T4 screening for the four major hospitals

in South Eastern Michigan. Subsequently, with the vision of the then Governor of

Michigan (Honorable Mr. William Milliken), a law was passed for the establishment of a

state of the art laboratory in Lansing, Michigan for the screening of inborn errors of

metabolism and hemoglobinopathies. By statute, each state in the USA performs

newborn screening (NBS), however, the number of tests/neonate and the

methodologies utilized vary from state to state. In 2006, Honorable Senators Edward M.

Kennedy, Barack H. Obama, and Hillary R. Clinton proposed a uniform standard and

protocol of NBS in USA. An integral part of this proposal was to extend this program to

resource-poor countries under the auspices of the United States Agency for

International Development. Unfortunately, due to political events in USA and the death

of Senator Edward M. Kennedy, nothing materialized in this direction.

NBS in America always includes analysis for hemoglobinopathies as described

by the Health Resources and Services Administration (HRSA) Maternal and Child

Health Program of the U. S. Department of Health and Human Services1.

During the last twenty years, I had the opportunity to introduce NBS in a few

developing countries (Kuwait, Iraq, Bahrain, Egypt, and Saudi Arabia) with the

cooperation of Quest Diagnostics, USA and Laboratory Corporation of America, USA.

PerkinElmer Genetics is the most popular private laboratory in USA that provides NBS

services worldwide. Besides the popularity of PerkinElmer Genetics, several countries

in Europe and Asia have instituted liquid chromatography-mass spectrometry (Applied

212

Bioscience), high performance liquid chromatography (Bio-Rad and Trinity Biotech), and

isoelectric focusing instruments (PerkinElmer’s Resolve, and Helena’s SPIFE) for NBS.

6.2 Methodologies

Isoelectric focusing (IEF) and high performance liquid chromatography (HPLC) are

the two most commonly used screening methods for hemoglobinopathies in the

neonate. Recently Sebia (Evry, France) has introduced the capillary zone

electrophoresis (CZE) method (Neonate Hb Fast System) for the newborn screening of

hemoglobinopathies. In order to resolve abnormal results of NBS,the blood of the

biological parents (and sometimes of the siblings) are analyzed by IEF, HPLC, agarose

gel electrophoresis (pH 8.6 and 6.2) and capillary zone electrophoresis to ascertain

genetic inheritance of the abnormality in the neonate. Finally, the diagnosis is confirmed

by means of DNA mutation studies,but for Hb S the final diagnosis can be confirmed by

a “Sickle” test, and complete blood count (CBC) with manual differential. Further

confirmation, if desired, for Hb S in a newborn can be achieved by testing the blood of

the parents. Electrospray ionization-mass spectrometry (Chapter 3.4), PCR and Sanger

sequencing (Chapter 3.5) are also used to confirm DNA mutation. A table of screening

methods by individual states in America can be obtained via internet2.

The principle of the assay by IEF (Chapter 2.7) and HPLC (Chapter 2.8) for the

screening of hemoglobinopathies in the neonate and the adult are similar. However

certain adjustments are required for the neonate specimen (dried blood spot on filter

paper) handling and processing. The “Resolve” kit and instrumentation of the

213

PerkinElmer and “SPIFE 2000 or 3000” instrument of the Helena Laboratories are the

most commonly used methods for the screening of hemoglobinopathies by IEF. HPLC

is most commonly performed employing ion-exchange chromatography by the NBS

instruments manufactured by BioRad, USA., or Trinity Biotech’s Ultra Resolution

System.

The principle of the CZE for the screening of hemoglobinopathies (Chapter 2.6)

in the adults and neonates is identical, except that modifications in the automated

instrument are necessary for the handling of a dried blood sample on filter paper from

the neonate (Figure 1). Sebia is the main supplier of capillary zone electrophoresis

instrument and reagents for the NBS of hemoglobinopathies.

Figure 1. Automated hemolysis, sample dilution and analysis instrument (Sebia)

214

The hemoglobin variants in the neonate by the “Sebia Capillarys Neonat

Fast System” separated into windows or zones (N1-N13) as illustrated in section

5.4. This method has been evaluated 3-5 and found satisfactory, however as with

IEF and HPLC methodologies for NBS, results have to be considered provisional and

confirmatory procedures are always required because many rare hemoglobin variants

migrate or elute on the same position of the common one and because different

homozygous or hemizygous genotypes look identical with these methods.

6.3 Laboratory Reports Format and Interpretation

All the NBS laboratories in America have an elaborate management system for the

procurement, handling and processing of dried blood on filter paper (Guthrie card).

Additional facilities are provided for confirmatory testing and the counseling of the

parents. Advisory consultative services and treatment modalities are also provided by

the medical staff of the NBS facilities in America. The details of all these services are

available online as well as in a printed version.

Normal patterns and abbreviations:

A normal newborn typically displays about 70% Hb F and 20% Hb A and perhaps

traces of Hb A2. The abbreviation used to indicate normal and abnormal patterns by the

laboratories are:

FA Fetal hemoglobin is greater than adult hemoglobin. This isobserved in a newborn < 3 weeks of age.

215

AF Adult hemoglobin is greater than fetal hemoglobin. This is usually observed in a newborn > 3 weeks of age unless transfused within the last eight weeks and anemia is not suspected.

Abnormal patterns and abbreviations:

Fa Lower Hb A levels then expected for the gestational age usuallyindicate a β-thalassemia carrier that could have been born from a couple at risk (50% chance). Reporting these carriers will allowthe couple to consider prospective primary prevention.

BFA? The presence of Hb Bart’s in a further normal pattern will indicate

α-thalassemia (2-5% α+ heterozygous), (5-10% α+ homozygous or α0 heterozygous), (>10% could indicate Hb H disease). Hb Bart’s

in absence of Hb F indicates hydrops foetalis.

FF Absence of Hb A may indicate a delayed appearance of Hb A(early prematurity), hereditary persistence of fetal hemoglobin(HPFH), or β-thalassemia major.

FAS Hb F + Hb A and Hb S indicates heterozygous Hb AS trait (sicklecell trait).

FSS Patterns with only Hb F and Hb S may indicate homozygous Hb Sor Hb S-β-thalassemia (both resulting in sickle cell disease).

FAC Hb F, Hb A and Hb C indicates heterozygous Hb AC trait.

FCC Only Hb F and Hb C indicates either homozygous Hb C orHb C-β-thalassemia.

FSC Hb F + Hb S and Hb C indicates compound heterozygous Hb S andHb C (sickle cell disease).

FAE Hb F, Hb A, and Hb E indicates heterozygous Hb AE trait.

FEE Hb F and Hb E only may indicate mild Hb E homozygosity orsevere Hb E-β-thalassemia.

FSE Hb F, Hb S and Hb E indicates Hb S/E compound heterozygosity(sickle cell disease).

FAD Hb F, Hb A and Hb DPunjab indicates heterozygous Hb AD trait(an asymptomatic condition).

216

FDD Only hb F and hb D indicates Hb D homozygous or Hb D-β-thalassemia (both mild conditions).

FDS Hb F, Hb S and Hb D indicates compound heterozygous Hb SD (sickle cell disease).

FAO Hb F, Hb A, and Hb OArab indicates heterozygous Hb A-OArab trait (an asymptomatic condition).

FSS-Bart’s Homozygous Hb S or Hb S-β-thalassemia (severe sickle cell disease condition).

Earlier, laboratories used the symbol “X” to designate a hemoglobin variant

which could not be identified by the NBS laboratory and further testing was suggested.

This practice of using “X” for a hemoglobin variant was abandoned by some

laboratories and now the symbol “V” is also used for this purpose.

It is emphasized here that some abnormal hemoglobin designated by “V” is

often reported in newborn screening, because the screening laboratories do not have all

the diagnostic methods available. In these situations the physician is advised to have

definitive diagnostic testing done at a specialized laboratory, e.g.

Georgia Health Sciences University Sickle Cell Center, Augusta, Georgia

(http://www.georgiahealth.edu/centers/sicklecell).

The interpretation of NBS in a premature neonate is subject to a possibility of false

positive results6, therefore the blood is retested when the adjusted gestational age is 40

weeks and two months after transfusion if executed.

6.4 Examples of Neonatal Screening

217

In this section, selected cases are presented to illustrate the laboratory data

obtained from NBS from commonly used methods.

6.4.1 Capillary Zone Electrophoresis

Capillary zone electrophoresis (CZE) scans of the most commonly expected

hemoglobin variants are presented in Figures 2-9. These scans were obtained after

analyzing the dried blood spot on the “Sebia Capillarys Neonat Fast System.” The blood

sample was collected by capillary puncture between 2-5 days after birth from neonates

of gestational age > 38 weeks. We have also provided the percentage of major

hemoglobin bands.

218

219

220

221

222

6.4.2 Isoelectric Focusing

Isoelectric focusing is the most widely used NBS method for hemoglobinopathies

in America. Here again, confirmatory testing is desired for accurate diagnosis of any

abnormal hemoglobin. One method of confirmation, if feasible, is to include the testing

of the biological parents. In Figure 10, we have presented the IEF results (Hb SC

disease) of a newborn, along with that of the father and mother of the newborn. The

father has Hb AS trait, and the mother has Hb AC trait, therefore there is a 25% chance

of the genetic inheritance of Hb SC disease in the newborn.

Figure 10. IEF results of newborn (Hb SC disease), father (Hb AS trait), and mother (Hb AC trait).

223

6.4.3 Isoelectric Focusing and High Performance Liquid Chromatography

Generally speaking, it is a common practice in some laboratories in USA to

further evaluate the IEF results for abnormal cases by HPLC and vice versa. In Figures

11-17, we have presented the scans for both of these methods; for the normal and a

few abnormal variants in a newborn. However, absolute certainty is never achieved by

these two methods and DNA sequencing is the method of choice to confirm the variant

and eventually the halotype to define the prognosis, tailor the best treatment and also to

allow primary prevention in case of another pregnancy (Section 5.4.4).

The IEF figures provided in this section were obtained using the RESOLVE

neonatal hemoglobin test kit and testing equipment (PerkinElmer). In all the IEF Figures

(11-17), we have presented at the top the IEF of the traditional laboratory control

“AFSC.” Details about this procedure can be obtained from:

http://www.perkinelmer.com/CMSResources/Images/44-72976FLY_Hemoglob_1244-

9784.pdf

The high performance liquid chromatography scans provided in this section were

obtained using the Trinity Biotech’s Ultra Resolution System. Details about this

procedure and instrumentation can be obtained from:

http://www.trinitybiotech.com/HbA1c_HB/Instruments/Pages/Ultra2Variant.aspx

224

Isoelectric focusing

High performance liquid chromatography

Figure 11 Normal: “FA”IEF of normal phenotype displays three prominent bands, Hb F, Hb A, and acetylated Hb F. Hb F is the prominent band in newborns. Hb A (the middle band) in the IEF pattern often appeared weaker in premature babies compared to full term babies. In all the HPLC separations the prominent Hb peaks (i.e. with significant concentration) eluted at specific retention times.

Figure 12 Hb AS trait “FAS”

Isoelectric focusing

High performance liquid chromatography

Figure 13 Hb AE trait “FAE”

225

Isoelectric focusing

High performance liquid chromatography

Isoelectric focusing

High performance liquid chromatography

Figure 14 Hb AC trait “FAC”

Isoelectric focusing

High performance liquid chromatography

Figure 15 Hb SC disease “FSC”

Isoelectric focusing

High performance liquid chromatography

Figure 16 Hb S disease “FS”

226

Isoelectric focusing

High performance liquid chromatography

Figure 17 Hb Bart’s “FABart’s”Note: A reviewer of this chapter pointed out the possibility that the fastest band on IEF is Hb H and the second is Hb Bart’s, as Hb H affected babies also have fast bands on IEF. Another reviewer suggested that Hb H affected babies have three fast bands on IEF, with a Bart’s result on HPLC exceeds 25%, and usually greater than 30%. This case displayed < 10% Bart’s from HPLC and displayed the typical two-banded Bart’s that is observed on one and two gene deletion alpha thalassemia carriers.

227

6.4.4. Isoelectric focusing, High Performance Liquid Chromatography and DNA Studies

In Figure 18, we present the typical IEF result of a full-term newborn with only Hb

F and Hb S, and no Hb A. This pattern suggests in order of probability the following

diagnostic options:

a) Hb S homozygous (both β genes code for Hb S, genotype associated with severe SCD).b) Hemizygous Hb S / β-thalassemia (one gene codes for Hb S and the other is not active, genotype associated with severe SCD).c) Hemizygous Hb S / deletional HPFH (one β gene codes for Hb S and the other is deleted, associated with mild SCD conditions).d) Double heterozygous Hb S / Hb Lepore (a combination associated with SCD)e) Double heterozygous Hb S-like / β-thalassemia (a combination not associated with SCD).f) Double heterozygous Hb S / Hb S-like (the last migrating like Hb S but causing no SCD).g) Homozygous for the same Hb S-like variant heterozygous for two Hb S-like variants

This means that even the simple SCD newborn pattern comes with different

diagnostic options that have to be sorted out at the DNA level.

228

Figure 18. IEF of newborn

Another example of a complex interpretation is shown in Figure 19. The HPLC

pattern of the newborn shows 54% of Hb F with three additional and significant bands at

a retention time known for a) Hb A at 0.87 minutes (1.3%), b) Hb E/A2 at 1.04 minutes

(8.3%) and c) Hb S at 1.2 minutes (6.3%). It is emphasized that a newborn cannot be

assigned Hb A2 and Hb E was not detected by IEF. Therefore, the band in HPLC at

1.04 minutes is due to a hemoglobin variant to be defined at the molecular level.

229

Figure 19. HPLC of newborn

DNA sequencing revealed Hb S heterozygous mutation at codon 6 and a second

point mutation at codon 43 of the β-globin gene leading to a Glu→Ala amino acid

substitution known as Hb G-Galveston.Therefore the newborn was diagnosed at the

molecular level as compound heterozygous Hb S / Hb G-Galveston, a combination

which is not associated with SCD. It is noteworthy to mention that like Hb G-Galveston 230

elutes on HPLC at the position of the common Hb E, many other variants elute at the

position of Hb S, Hb D, or Hb C and therefore molecular confirmation is always needed.

References (Section 6.1 – 6.4)

1. Lin K, Barton M. Screening for Hemoglobinopathies in Newborns. Reaffirmation Update for the U.S. Preventive Services Task Force. Evidence Synthesis No. 52. Rockville, MD: Agency for Healthcare Research and Quality, August 2007. AHRQ Publication No. 07-05104-EF-1. Available at http://www.ahrq.gov/clinic/serfiles.htm#sicklecell

2. http://nnsis.uthscsa.edu/xreports.aspx?xreportID=47&formid=104&fclr=1 3. Giordano PG. Newborn screening for hemoglobinopathies using

capillary electrophoresis. Methods Mol Biol 2013; 919: 131-45.4. Renom G, Mereau C, Maboudov P, Perini JM. Potential of the Sebia Capillarys neonat fast automated system for neonatal screening of sickle cell disease. Clin Chem Lab Med 2009; 47(11): 1423-32.5. Mantikou E, Harteveld CL, Giordano PC. Newborn screening for hemoglobinopathies using capillary electrophoresis technology: Testing the Capillarys Neonate Fast Hb device. Clin Biochem 2010; 43: 1345-1350.6. Hustace T, Fleisher JM, Varela AMS, Podda A, Alvarez O. Increased Prevalence of False Positive Hemoglobinopathy Newborn Screening in Premature Infants. Pediatric Blood Cancer 2011; 57: 1039-1043.

References related to neonatal screening experience for hemoglobinopathies:

● Bouva MJ, Mohrmann K, Brinkman Henri BJM, Kemper-Proper EA, Elvers B, Loeber JG, Verheul Francesco EAM, Giordano PC. Implementing Neonatal screening for haemoglobinopathies in the Netherlands. J Med Screen 2010; 17: 58-65

● Michlitsch J, Azimi M, Hoppe C, Walters MC, Lubin B, Lorey F, Vichinsky E. Newborn Screening for Hemoglobinopathies in California. Pediatr Blood Cancer 2009; 52: 486-490.

● Kafando E, Nacoulma E, Quattara Y, Ayeroue J, Cotton F, Sawadogo, Gulbis B. Neonatal haemoglobinopathy screening in Burkina Faso. J ClinPathol 2009; 62: 39-41.

231

● Streetly A, Latinovic R, Hall K, Henthorn J. Implementation of universal newborn bloodspot screening for sickle cell disease and other clinically significant haemoglobinopathies in England: screening results for 2005-7. J Clin Pathol 2009; 62: 26-30.

● Gulbis B, Cotton F, Ferster A, Ketelslegers O, Dresse MF, Ronge-Collard E, Minon JM, Le PQ, Vertongen F. Neonatal haemoglobinopathy screening in Belgium. J Clin Pathol 2009; 62: 49-52.

● Bardakdjian-Michau J, Bahuau M, Hurtrel D, Godart C, Riou J, Mathis M, Goossens M. Neonatal screening for sickle cell disease in France. J ClinPathol 2009; 62: 31-33.

● Adorno EV, Couto FD, de Moura Neto JP, Menezes JF, Rego M, dos Reis MG, Goncalves MS. Hemoglobinopathies in newborns from Salvador, Bahia, Northeast Brazil. Cad. Saude Publica, Ruio de Janeiro 2005; 21(1): 292-298.

6.5 Genetic Counseling & Screening:

After a careful review of the literature on the worldwide prevalence of

thalassemia and hemoglobinopathies, it is my estimate that by 2050 more than

500 million individuals will be affected by these genetic disorders. During the past two

decades, attempts have been made to provide premarital and prenatal genetic

counseling and screening in both the endemic and non-endemic (in view of migration)

countries, however achieving a thalassemia-and hemoglobinopathy free generation

seems unlikely to me. Although treatment modalities for sickle cell anemia have been

investigated since 1967, including the latest promising treatment with antidepressants in

these individuals by increasing the concentration of Hb F, permanent cure is illusive.

Impediments for the worldwide implementation of a prevention and control

program are: a) financial resources, b) technical personnel, c) religious and social

considerations, d) education of the entire population about the benefits of this program,

and e) poor and resource-lacking population problem.

232

Indeed it is very promising that various religious organizations (Muslims and

Jews) have authorized screening for genetic diseases after taking into consideration the

halachic concerns.

Country and state specific genetic counseling and screening programs (Thailand,

Cyprus, etc.) are steps in the right direction, and let us hope that these initiatives

blossom into an elaborate undertaking.

References

1. Jewish Women’s Health. http://www.jewishwomenshealth.org/article.php?article=32

2. Strauss BS. Genetic Counseling for Thalassemia in the Islamic Republic of Iran. Perspectives in Biology and Medicine 2009; 52(3): 364-376

3. Larijani B, Anaraki FZ. Islamic principles and decision making in bioethics. Nature Genetics 2008; 40(2): 123.

4. Norton ME. Genetic screening and counseling. Current Opinion in Obstetrics and Gynecology 2009, 20: 157-163.

5. Zlotogora J. Population programs for the detection of couples at risk for severe monogenic genetic diseases. Hum Genet 2009; 126: 247-253.

6. Al-Ama JY. Attitudes towards mandatory national premarital screening for hereditary hemolytic disorders. Health Policy 2010; 97: 32-37.

7. Theodoridou S, Alemayehou M, Prappas N, Karakasidou O, Aletra V, Plata E, Tsaftaridis P, Karababa P, Boussiou M, Sinopoulou K, Hatzi A, Voskaridou E, Loutradi A, Manitsa A. Carrier Screening and Prenatal Diagnosis of Hemoglobinopathies. A Study of Indigenous and Immigrant Couples in Northern Greece, over the last 5 years. Hemoglobin 2008; 32(5): 434-439.

8. Koren A, Zalman L, Palmor H, Zamir RB, Levin C, Openheim A, Daniel Spiegel E, Shalev S, Filon D. Sickle Cell Anemia in Northern Israel: Screening and Prevention. IMAJ 2008; 11: 229-234.

9. Yamsri S, Sanchaisuriya K, Fucharoen G, Sae-ung N, Ratanasiri T, Fucharoen S. Prevention of severe thalassemia in northeast Thailand: 16 years of experience at a single university center. Prenat Diagn 2010; 30: 540-546.

10. Tarazi I, Al-Najjar E, Lulu N, Sirdah M. Obligatory premarital tests for thalassemia in the Gaza Strip: evaluation and recommendations. Int Jnl Lab Hem 2007; 29: 111-118.

233

11. Al-Allawi NA, Al-Dousky AA. Frequency of haemoglobinopathies at premarital health screening in Dohuk, Iraq: implications for a regional prevention programme. Eastern Mediterranean Health Journal 2010; 16(4): 381-385.

12. Karimi M, Jamalian N, Yarmohammadi H, Askarnejad A, Afrasiabi A, Hashemi A. Premarital screening for β-thalassemia in Southern Iran: opinions for improving the programme. Journal of Medical Screening 2007; 14(2): 62-66.

13. Al-Sulaiman A, Suliman A, Al-Mishari M, Al-Sawadi A, Owaidah TM. Knowledge And Attitude Toward The Hemoglobinopathies Premarital Screening Program in Saudi Arabia: Population Based Survey. Hemoglobin 2008; 32(6): 531-538.

14. El-Tayeb E-N H, Yaqoob M, Abdur-Rahim K, Gustavson K-H. Prevalence of β-Thalassemia and Sickle Cell Traits in Premarital Screening in Al-Qassim, Saudi Arabia. Genetic Counseling 2008; 19(2): 211-218.

234

Chapter 7

Prenatal Diagnosis of β-Thalassemias and HemoglobinopathiesMaria Christina Rosatelli, PhD and Luisella Saba, PhD

Abstract

Prenatal diagnosis of β-thalassemia was accomplished for the first time in the

1970s by globin chain synthesis analysis on fetal blood obtained by placental aspiration

at 18-22 weeks gestation. Since then, the molecular definition of the β-globin gene

pathology, the development of procedures of DNA analysis, and the introduction of

chorionic villous sampling have dramatically improved prenatal diagnosis of this disease

and of related disorders. Much information is now available about the molecular

mechanisms of the diseases and the molecular testing is widespread. A prenatal

diagnosis has to provide an accurate, safe and early result, an efficient screening of the

population and a rapid molecular characterization of the couple at risk, are necessary

prerequisites. In the last decades earlier and less invasive approaches for prenatal

diagnosis were developed. An overview of the most promising procedure will be done.

Moreover, in order to reduce the choice of interrupting the pregnancy in case of affected

fetus, Preimplantation or Preconceptional Genetic Diagnosis (PGD) has been setting up

for several diseases including thalassemias.

Rosatelli MC, Saba, L. Prenatal Diagnosis of Beta-Thalassemia and Hemoglobinopathies. Mediterr J Hematol Infec. Dis. 2009; 1(1): e200911

This can also be accessed from http://www.mjhid.org/article/view/5079.

235

We acknowledge all those concerned with this publication.

Introduction

Β-thalassemias and hemoglobinopathies are among the most common

autosomal recessive diseases with a high frequency in the population of the

Mediterranean area, the Middle East, the Indian subcontinent, the Far East, Tropical

Africa and the Caribbean [1]. However, in the last decades, the steady migratory flows

have rendered these pathologies much more widespread, thus representing a general

public health problem. In the '70s the set-up of globin chain synthesis analysis for the

detection of a little amount of β-chains in fetal blood during the 18th-22nd week of

gestation [2] has allowed the development of screening programs of the general

population, based on the identification of the couple at risk, and, in addition, the offer of

prenatal diagnosis testing. At that time the thalassemic patients had limited lifespan

and prenatal diagnosis represented the only option for the control of the disease. Such

programs first started in Sardinia, Continental Italy, Cyprus, and Greece [3,4,5,6].

Prenatal diagnosis on fetal blood, even if it represented for couples at-risk an

opportunity to generate healthy sons, was not easily accepted. The late gestational age

in which fetal diagnosis was carried out, the risk of misdiagnosis due to a not clear cut-

off between some heterozygotes and affected fetuses, the high risk of miscarriage due

to the sampling procedures, made indeed the procedure difficult to accept from the

couples.

The continuous advances in the knowledge of the molecular pathology of the

disease, the discovery of restriction fragment length polymorphisms (RFLP) linked to

236

the β-like globin gene, the development of methodologies for mutation detection and the

application of the villocentesis for the recovery of nucleated fetal cells, allowed a fast

improvement both in feasibility and acceptability of prenatal diagnosis. For a short

period, in the eighties, the diagnosis of thalassemia was obtained either indirectly by

linkage analysis using RFLP at the β-globin cluster [7] or directly by oligonucleotide

hybridization on electro-phoretically separated DNA fragments [8] or by enzymatic

digestion of mutated sites. A major impulse has been given by the PCR technology that

allowed the development of a number of procedures, for easier mutation detection, as

well as the development of both PGD and non-invasive prenatal diagnosis procedures.

Nowadays thalassemias are detected directly by the analysis of amplified DNA from

fetal trophoblast or, more rarely, from amniotic fluid cells.

In this review we will delineate current procedures for prenatal and

preimplantation diagnosis of thalassemias as well as the most promising approaches for

non-invasive prenatal diagnosis.

Prenatal Diagnosis

Detection Methods:

Detection of molecular defect in both parents is a prerequisite for prenatal

diagnosis of the disease. The majority of defects affecting the β-globin gene are point

mutations that occur in critical areas for its function, or single/few base addition/deletion

that change the frame in which triplets are translated into protein. Very rarely β-

thalassemia results from gross rearrangement in the β-globin gene cluster. In spite of

the marked molecular heterogeneity, a limited number of molecular defects are 237

prevalent in every at risk population. This may be very useful in practice, because a

panel of most frequent mutations to be searched for can be designed according the

carrier's ethnic origin [9]. Known mutation detection is caried out by a number of PCR-

based techniques. (ARMS, Amplification Refractory Mutation System) and the reverse

oligonucleotide hybridization with specific oligonucleotide probes (RDB, Reverse

Oligonucleotide-probe analysis).

Primer-specific Amplification:

The method is based on the principle that a primer carrying a mismatch in its 3'

region cannot anneal on its template. With this method, the target DNA fragment is

amplified in two separate PCR reactions using a common primer and either of the two

following primers: one complimentary to the mutation to be detected (β-thalassemia

primer) and one complementary, at the same position, to the normal DNA (normal

primer). Normal DNA is amplified only by the normal primer while the DNA from

homozygotes only by the β-thalassemia primer and DNA from heterozygotes by both

primers. A different sized fragment of the β-globin gene is simultaneously co-amplified

as an internal control of the PCR reaction [10]. The method is very simple as it

requires, for each mutation to be searched, only two PCR reactions followed by agarose

gel electrophoresis. A further improvement of the methodology can be obtained by

multiplexing the primers for more than one mutation. In good hands the method is very

safe and particularity useful in fetal DNA analysis to search for mutations previously

detected in the parents.

238

Reverse Oligonucleotide Hybridization:

When the spectrum of mutations to be searched is complex, ARMS is not the

most appropriate method. In this case RDB results can be more informative and

efficient. The method uses membrane-bound allele-specific oligonucleotide probes that

hybridize to the complementary sequence of the PCR product prepared using patient

DNA as starting template [11]. In this format, multiple pairs of normal and mutant allele-

specific oligonucleotides can be placed on a small strip of membrane. Hybridization

with PCR-amplified β-globin gene is able to detect, in a single procedure, any of the

mutations screened. Up to 20-30 mutations have indeed been screened in one single

step and several commercial kits are available to detect the most common beta

thalassemia mutations in Mediterranean population.

Other Known Mutation-detection Procedures:

Several other methods have been developed to search for known mutations, i.e.

oligonucleotide ligation assay [12], restriction enzyme digestion of PCR products [13];

however some of them have been abandoned in routine diagnostics as they are less

informative or more complex.

In recent years a real time PCR assay has been successfully applied to both

carriers screening and prenatal diagnosis [14]. This is a one-step method that is based

on the use of fluorescent hybridization probes followed by a melting curve analysis.

This method, which allows the simultaneous multiple mutation detection, has been

239

successfully applied also to the detection of maternal contamination. In spite of these

advantages its use is still limited as it needs a dedicated apparatus as well as an

accurate population-based design of detection probes.

Technically, we can realistically predict further simplification and full automation

of the procedures for the detection of the β-thalassemia mutations is commercially

available, which are not completely automated and quite expensive. Among them, the

oligonucleotide microchip-based assays have been proposed many times for the large-

scale detection of mutations in genetic diseases, including β-thalassemia [15]. Given

the alternative features of high throughput and automation, the DNA chip has the

potential to become a valuable method in future applications of mutation detection in

medicine. At the moment, the technology developed several years ago is not yet

transferred in the clinical practice, due to the higher costs and to the lower analytical

sensitivity and specificity.

Unknown Mutations Detection:

When carriers escape to the above mutation detection approaches, further

investigations need to be carried out by alternative methods which uncover the

presence of unknown mutations by scanning the whole gene. Denaturing gradient gel

electrophoresis (DGGE) [16,17,18], Denaturing High Pressure Liquid Chromatography

(dHPLC) and Single Strand Conformation Polymorphism (SSCP) [19] are the most

widely used in the last years, followed by direct sequencing analysis [20] which

characterizes the undefined mutation found by these methods. Nowadays, considering

the small size of the β-globin gene (1,8kb), the simplified technologies available and the

240

reduced costs of analysis, direct sequencing, based on cycle sequencing with

fluorescent dye terminators and automated capillary DNA sequencing technology,

seems to be the faster and most useful approach to detect unknown thalassemia

mutations.

If a mutation is not detected by sequence analysis, we search for the presence of

small deletions by polyacrylamide gel electrophoresis of amplicons designed for the

most frequent small deletional defects of the β-globin gene (gap PCR). Furthermore,

the presence of larger deletions of the cluster may be identified by Southern-blotting or

more recently by Multiple Ligation-Dependent Probe Amplification (MLPA) for which a

commercial kit is available (SALSA MLPA KIT P102 HBB-MRC Holland).

In a very limited number of cases, direct sequencing from position -600 to 60 bp

downstream from the β-globin gene and methods for deletion detection, failed to detect

the disease causing defect. In these cases, the molecular defect may reside either in

the locus control region of the β-globin gene cluster, or in one of the genes, outside the

β-globin gene region, encoding for regulatory proteins acting in trans on the function of

the β-globin gene. Very recently it has been proved that the β-thalassemia-like

phenotype could be caused by the coinheritance of a β-globin gene defect and a

duplication of the α-globin gene cluster, which results in an excess of α chain. In these

selected cases, the characterization of these α-globin gene rearrangements (SALSA

MLPA KIT P140-B2 HBA-MRC Holland) can be routinely carried out with success by

MLPA analysis.

241

Genetic Counseling of the Couple at Risk:

Both members of the couple at risk are counseled in a non-directive way. The

nature of the disease, the implications of being carriers and reproductive choices are

analyzed, specifically those concerning birth control, including prenatal or

preimplantation diagnosis and the possibility, in case of affected fetus HLA compatible

to not interrupt pregnancy. As for fetal testing, detailed information is offered regarding

the risk of fetal mortality, the risk of misdiagnosis, and the mortality and morbidity of an

abortion in case of affected fetus.

Fetal DNA Sampling:

Fetal DNA for analysis can be obtained from either amniocytes or chorionic villi.

At present the most widely used procedure is chorionic villi sampling, because of the

clear advantage of being carried out during the first trimester of pregnancy, generally at

the 10th-12th week of gestation [21, 22, and 23]. The risk of fetal mortality associated

with both methods is in the order of 1-2%. Chorionic villi may be obtained

transcervically or transabdominally, the last being most widely used, mainly because it

has a low infection rate and a lower incidence of amniotic fluid leakage. Moreover it is a

simple procedure, largely preferred by pregnant women, which can be carried out also

in late gestational age. Samples obtained by villocentesis need to be accurately

dissected under inverted microscope in order to remove maternal decidua, that

represent the major cause of diagnostic error in prenatal diagnosis of monogenic

diseases.

242

Fetal DNA Analysis:

Fetal DNA is analyzed using the same methods described above for detection of

known mutations during carrier molecular screening. To limit the possibility of

misdiagnosis, we analyze chorionic villous DNA with two different procedures: i.e. RDB

hybridization and primer-specific amplification, using distinct couple primers.

Misdiagnosis may occur for several reasons: failure to amplify one copy of the target

DNA fragment, mispaternity, maternal contamination, and sample exchange.

Misdiagnosis for failure of DNA amplification is obviously limited by the double approach

described above. To avoid misdiagnosis due to maternal contamination as well as

mispaternity and/or sample exchange, a fetal DNA microsatellite analysis is usually

performed to verify the presence of one allele from each parent [9]. In our hands, by the

above mentioned PCR-based procedures, no misdiagnoses have occurred in more than

5000 cases. Figure 1 shows the overall results of the Sardinia prenatal diagnosis

program since the beginning of 1976 up to the end of the past year.

243

Currently, prenatal diagnosis is a widely applied and well-accepted procedure.

Among the patients screened we have found an acceptability of 99.3% for early prenatal

diagnosis by CVS. This data, if compared with previously utilized procedures such as

fetal blood sampling, with an acceptability of 93.2%, and 96.4% by amniocentesis,

demonstrates how the acceptance of the procedure depends on its precocity [22].

The screening program in the Mediterranean countries has proven to be very successful

in reducing the number of thalassemia patients. In Sardinia, thalassemia major was

present in 1 in 250 births, and has declined to 1 in 4000 births (Figure 1). Other

countries in which such thalassemia programs have been introduced also show similar

trends.

244

Preimplantation and Preconceptional Genetic Diagnosis:

The progress in assisted reproduction and molecular genetics techniques,

particularly the advent of PCR that has made possible to analyze the genotype of a

single cell, has paved the way for preimplantation genetic diagnosis (PGD) [24,25].

This technique was introduced as an option for avoiding the decision to terminate an

established pregnancy diagnosed as affected by conventional approaches. The term

preimplantation genetic diagnosis describes those procedures which involve the

removal of one or more nuclei from oocytes (polar bodies) or embryos (blastomeres of

trophectoderm cells) to test for mutation in the target gene or aneuploidy before

transfer.

PGD requires that couples at risk undergo in vitro fertilization (IVF) even if not

infertile and for this reason a multidisciplinary approach including an appropriate genetic

counseling and the referral to both a fertility clinic and to a highly specialized molecular

genetics laboratory is mandatory.

Counseling for couples considering PGD must include additional information

regarding at least the risk associated with IVF procedures and with embryo biopsy, the

technical limitations of DNA analysis, including the risk of failure of the procedure as

well as that of misdiagnosis, and the need of subsequent prenatal diagnosis to confirm

the result. Beyond that, the possibility that no embryos may be transferred and the

dispositions of the embryos not transferred have also to be seriously considered.

245

Cell Biopsy:

Preimplantation may be carried out by either cleavage-stage biopsy of 1-2

blastomeres, from an eight-cell embryo three days after in vitro fertilization carried out

by ICSI (Intracytoplasmatic Sperm Injection), or by the biopsy of polar bodies.

For cleavage-stage biopsy the embryo is grown in vitro until it reaches a six-eight cell

stage which usually occurs on the third day after insemination. Polar bodies diagnosis,

pioneered by Verlinsky and his group in 2006 is based on the analysis of the first polar

body of unfertilized eggs [27], and may lead to distinguish between unfertilized eggs

that carry the defective gene and those without the defect. The successive sampling

and analysis of the second polar body that is extruded from the oocyte after fertilization

and completion of the second meiotic division, is carried out in order to avoid

misdiagnosis due to the high rate of recombination that happens during the first meiosis.

By fertilizing in vitro only the eggs without the defect and replacing them in the mother, a

successful pregnancy with a normal fetus can be obtained. Recently a preconceptional

genetic diagnosis based on the analysis of only the first polar body has been proposed

for countries in which the use of PGD and manipulation of embryos is prohibited [28].

This approach although permitting to avoid the manipulation, cryopreservation and/or

discard of sovranumerary and/or affected embryos, shows several problems: the need

to obtain more than 10-12 oocytes, the increased risk of diagnostic error and the

increased risk of the technical difficulties. Blastocyst biopsy, even if it has the advantage

to provide a higher number of cells, is at present more rarely used because of the

difficulties of the embryos to reach this stage in IVF programs. The cleavage-stage

246

biopsy of blastomeres from an eight-cell embryo is the most frequently used PGD

procedure all over the world.

Detection Methods:

Methods for mutation detection in OGD are always based on multiple steps of

PCR. Mutations are detected in PCR products by various methods that combine speed,

analytical sensitivity and specificity. In particular, a first round of multiplex PCR is

performed to amplify both the β-globin gene region including the mutation and one or

more polymorphic loci. Secondly, two separated nested PCR reactions are performed

to amplify the two or more selected genomic regions. Finally, the polymorphic alleles

are directly detected by capillary electrophoresis of the amplified fragment, while the

presence of β-globin gene mutations are identified by the subsequent mini sequencing

reaction [29]. This approach is expressly designed to detect the presence of the β-

globin gene mutations and to monitor, in the same sample, the presence of

contamination as well as the eventual allele drop-out that represent the most frequent

causes of error in PGD.

Quality Control:

For both techniques a prenatal diagnosis by villocentesis is recommended in

order to avoid diagnostic errors. Successful pregnancies following the transfer of

human embryos in which the β-globin gene defect has been excluded, occur only in 20-

25% of cases and the birth rate of a child is even lower. Due to the low birth rate most

women have to undergo PGD several times in order to give birth to a healthy child [30].

247

Transfer of no more than 1-2 embryos is strongly recommended in order to avoid

multiple pregnancies [31]. Elective Single Embryo Transfer (eSET) is in fact a well-

established procedure which has demonstrated to ensure a better prognosis of IVF

patients [32}.

PD or PGD?

Among clinical geneticists there has been much discussion about the main goal

of PD. Some have argued that the main aim is to avoid the birth of an affected child.

Others have emphasized the reproductive confidence and the purpose of informing the

couples at risk about the status of the fetus. Several studies indicate that if there is no

PD option, a large proportion (up to 50%) of the couples at high risk of an affected child

refrain from pregnancy despite their wish to reproduce. When PD is possible many

more at-risk couples dare to embark on a pregnancy.

Most experts consider PGD as an additional option for couples at risk and not as

a replacement for conventional prenatal diagnosis. PGD is still considered a highly

specialized experimental procedure with limited results, mainly dedicated to couples

against abortion for ethical and religious reasons and to a small proportion of couples

who have experienced repeated abortion, that ask for referral for this procedure.

At present its use in routine monitoring of pregnancies at risk is precluded by the

technical demand for these procedures, the difficulty in organizing the service, and the

high costs.

248

Simplification of preimplantation and preconception genetic diagnosis, together

with an increase in the pregnancy rate may lead to a more extensive use of the

procedure in the future.

Non-Invasive Prenatal Diagnosis (NIPD):

Analysis of Fetal Cells in Maternal Blood:

In the sections below the most significant studies, which have been carried out in

this field of research, are briefly summarized. The mot relevant results have been

grouped in three different sections, according to the different cell type in which they

have been acquired. A separate section is dedicated to NIPD of β-thalassemia.

Trophoblasts:

The first evidence that fetal cells circulate in maternal peripheral blood dates

back to 1893 when George Schmorl observed the presence of placentally derived

trophoblasts in the lungs of 17 autopsied women affected by severe eclampsia [33].

In 1959 Douglas [34] established that migration of trophoblasts is a normal process

during pregnancy and twenty-five years later, Covone et al [35] demonstrated that these

cells could be detected in healthy pregnant women as early as six weeks gestation.

They also found that an increased concentration of trophoblast cells were frequently

presenting in women affected by preeclampsia. Further studies have established that

trophoblasts are entrapped in the maternal lungs and rapidly removed from the

pulmonary circulation [36].

249

Tropoblast-specific cell-surface antigens have not yet been characterized and

several experimental evidences have shown that the H315, initially described as the

specific antigen for trophoblasts, is indeed absorbed in maternal leucocytes [37].

These are some of the reasons why, in recent years, trophoblasts are no longer

considered as the best target cells for non-invasive prenatal diagnosis. Nevertheless,

this line of research has not yet been completely abandoned as the characterization of

trophoblast-specific antigens is one of the objectives of the SAFE (Special Non-Invasive

Advances in Fetal and Neonatal Evaluation) Network (for more information please visit

www.safenoe.org).

Lymphocyte:

Fetal lymphocytes are the second cell type which has been extensively studied

as a possible source of fetal DNA. In 1969 Walknowska et al [38] detected for the first

time 46, XY karyotype cells in maternal peripheral blood of women bearing male

fetuses. Ten years later Herzenberg and colleagues described the use of FACS

(Fluorescent Activated Cell Sorting) as a method for the enrichment of fetal lymphocyte

expressing the HLA-A2 paternal antigen [39]. Detection of Y chromosome was then

obtained in the enriched cells deposited directly onto microscope slides, thus confirming

their fetal origin.

Unfortunately other groups have failed to replicate these results with success,

even when cytogenetic analysis was carried out in fetal cells that were flow sorted on

the basis of several HLA differences and by using monoclonal antibodies.

In the same years further studies demonstrated that lymphocytes were not

removed from maternal circulation after delivery. One of the earliest studies provided

250

the first evidences that fetal lymphocytes persist in maternal circulation one year after

delivery was published in 1974 by Bianchi et al [40]. Several years later Bianchi et al

described the presence of fetal progenitor cells 27 years after delivery [41]. For these

reasons also lymphocytes, as trophoblasts, became an unattractive candidate for non-

invasive prenatal diagnosis.

Erythroblasts:

One of the main advantages to study fetal erythroid cells is that they are

nucleated, terminally differentiated short-lived cells and for this reason they do not

persist in maternal circulation for a long time after delivery. Furthermore, first primitive

erythroblasts appear in the embryonic bloodstream around the four-five week gestations

so they can be detected early during gestation.

Nevertheless, their isolation from maternal peripheral blood is still problematic

because of their rarity and the lack of a fetal specific antibody.

In 1990 Bianchi [41] first described a method for fetal nucleated erythroid cells CD71

transferrin receptor, highly expressed in erythroid cells. Two years later Ganshirt-Ahlert

et al [43] obtained similar results by using a new detection system called MACS

(Magnetic Cell Sorting) which is based on the use of antibodies labeled with magnetic

beads.

Since then, both systems have been extensively improved and used, by several

groups, following different approaches which can consist in the positive selection of

CD71 and/or glycophorin-A fetal cells and/or in the negative depletion of CD45 maternal

cells. Usually, in both cases, a previous density (Ficoll or Histopaque) gradient

centrifugation step is carried out to remove non-nucleated maternal erythrocytes. A

251

schematic workflow resuming one of the strategies used for isolating fetal NRBCs from

maternal peripheral blood is represented in Figure 2. Finally both MACS and FACS

sorted cells are labeled with fluorescent antibodies which recognize embryonic (ε, ζ) or

fetal (γ) hemoglobin chains and are eventually subjected to FISH analysis for

chromosome Y detection. An example of positive labeling with the antibody for gamma-

globin conjugated with FITC is shown in Figure 3. Molecular characterization can

eventually be carried out in positive fluorescent cells isolated by laser microdissection.

Even with the high progress made in the last twenty years in this field, the methods for

erythroblasts enrichment are still limited as they mostly result in the recovery of fetal

samples with low yield (FACS) and scarce purity (MACS), being variably contaminated

by maternal cells.

For these reasons in recent years several studies have been addressed to the

proteomic field with the attempt to characterize novel fetal erythroblast cell-specific

surface markers. For example, bi-dimensional electrophoresis coupled with mass

spectrometry has allowed the identification of 2 proteins, differentially expressed in

sickle erythrocytes in comparison to healthy erythrocytes, and the detection of proteins

up-or downregulated in fetal erythroid cells in comparison to their adult counterparts.

Some of these results have been published as a full-patent application and the data

concerning the new antibodies developed against these new targets expect to be

validated in large samples of maternal blood [44]. In addition, further developments in

fetal cell recovery are expected to be obtained through the application of micro-fluidic

rare-cell capture technologies [45] which are being developed to detect not only fetal but

also cancer as well as other rare cells in biologic fluids.

252

253

Analysis of Fetal Cells in Maternal Blood and Non Invasive Prenatal Diagnosis

(NIPD) of β-Thalassemia:

Despite the difficulties encountered to find the best target cell and the best

method for their enrichment and isolation, several attempts have been made in the last

twenty years, to transfer the results of these researches into clinical practice.

Unfortunately the lack of reproducibility of experiments hardly makes the isolation of

fetal cells from maternal blood as a first choice method of NIPD of monogenic disorders.

Below the most significant results obtained in NIPD of β-thalassemia are briefly

summarized. The first example of non invasive prenatal diagnosis of

hemoglobinopathies was described in 1990 by Camaschella et al [46]. The genetic test

was carried out in three selected couples where the mother was a carrier of β-

thalassemia and the father of the Hb Lepore-Boston trait. The absence/presence of the

paternal trait was successfully detected in PCR amplified samples DNA extracted from

T-cell samples were obtained by incubating Ficoll-separated cells of the mother with the

CD 3-specific MoAb Leu 4 and then separating the positive cells with goat-anti-mouse

immunoglobulin G (1gG)-coated immunomagnetic beads.

In those years most of the studies were addressed to couples carrying different

mutations and only aimed to the exclusion of the paternal allele in the enriched fetal

cells, as most of the times they were contaminated from maternal cells.

In subsequent years, even if the fetal cells enrichment and selection methods

have been greatly improved, other IP diagnosis have been carried out but with

fluctuating results. Below are described three significant examples of NIPD realized,

with different levels of success, by using single or pooled erythroid cells.

254

In 1996 the group of Y.W.Kan47 reported the successful identification of two fetal

genotypes by using fetal nucleated erythroid cells selected by MACS, anti-ζ globin

immunostaining and then isolated by microscopy and cell scraping. The

presence/absence of sickle cell and beta thalassemia mutations of both parents were

finally detected by Reverse Dot Blot in PCR amplified samples constituted by pools of

fetal dissected cells.

A few years later the group of Di Naro [48] replicated these results using a

slightly different procedure for erythroblast enrichment which was carried out by Percoll

and Gastrografin multiple gradient centrifugation. Mutation detection was then obtained

by automated sequencing of single cells amplified by PCR. According to authors, even

if the risk of allele drop out is higher when amplifying single cells, however the possibility

to study several individual, instead of pooled, cells guarantees an accurate diagnosis of

the fetal DNA.

More recently the group of Kolialexi [49] has hardly tried to replicate these

results. In this study, NIPD was performed through magnetic cell sorting (MACS) and

microdissection of single NRBCs with a laser micromanipulation system. Single-cell

genotyping was achieved by nested real-time PCR for genotyping β-globin gene

mutations; a multiplexed minifingerprinting was used to confirm the origin of the isolated

cells and to exclude their possible contamination. A total of 224 cells were isolated but

only half of them were successfully amplified. In the majority (n=80) of these cells

minifingerprinting was not informative because of allele dropout or homozygosity. In the

rest of the samples, 22 cells resulted to be of fetal origin, 26 maternal while 80 were non

informative.

255

Analysis of Fetal DNA in Maternal Plasma and Non Invasive Prenatal Diagnosis

(NIPD) of β Thalassemia:

The existence of cell-free nucleic acids within the human plasma was firstly

reported in 1948 by Mendel and Metais [50] which described their presence both in

normal subjects and in individuals affected by various diseases. Some decades later

other studies have confirmed the presence of circulating DNA as well as of RNA in

several pathological conditions (pancreatitis, inflammatory diseases, cancer, diabetes,

etc) [51].

In 1997 Lo et al discovered for the first time that a fetus may release cell-free

fetal DNA (cffDNA) into maternal plasma, thus providing an alternative to fetal cells for

noninvasive prenatal diagnosis [52].

In recent years more information has been acquired about the concentration, the

origin and the characteristics of the cell-free fetal DNA and several procedures have

been developed in order to use it in prenatal diagnosis.

The cell-free DNA is constantly present in peripheral blood of non pregnant

women and its concentration increases during pregnancy. The cell-free fetal DNA

represents the 3-5% of the DNA present in maternal plasma from which, after delivery, it

is rapidly cleared.

Recent studies carried out by microfluidic digital PCR have revealed that cffDNA

can be present at even higher concentrations which can reach up to 10-20% of total

DNA in maternal plasma [53]. Nevertheless, because of the high background of

maternal DNA, an enrichment step is needed to obtain highly purified fetal DNA

samples suitable for non invasive prenatal diagnosis.

256

Size-fractionation agarose gel electro-phoresis is one of the methods developed

for fetal DNA enrichment and consists in the isolation of short-length DNA fragments

(<300 bp of length) which is the medium length of the cffDNA. This method coupled

with the peptide-nucleic-acid clamp (PNA) PCR, which selectively suppresses the

amplification of maternal alleles, and with the Allele-specific Real-Time PCR for

mutation detection, has been used with success by Li et al [54] to detect mutations of

paternal origin in fetuses at risk for β-thalassemia.

More recently [55] the same group has described a new procedure, still based on

size fractionation method, but coupled with MALDI-TOF mass-spectrometry, a medium-

throughput platform which detects with high sensitivity the presence of known and

unknown point mutations. In this case no suppression of maternal allele was caried out

and the molecular diagnosis was addressed to the exclusion of the paternal mutated

allele. The analysis by MALDI-TOF preceded by size fractionation has given improved

results, in comparison to the absence of enrichment, in the detection of the codon 39 β-

thalassemia paternal allele. Nevertheless, for eventual future diagnostic application the

protocol needs to be validated in larger samples, even if the high cost of the

instrumentation required makes this platform difficult to apply in routine diagnostics and

the size fractionation is considered an enrichment method more susceptible to maternal

contamination.

The use of peptide-nucleic-acid clamping to suppress the amplification of normal

maternal alleles was first described by Cremonesi in 2004 [56]. Peptide nucleic acid is

artificially synthesized polymers similar to nucleic acids and able to hybridize DNA

sequences. The PNA/DNA hybrids are more stable than equivalent DNA/DNA hybrids

257

but less stable in the presence of single-pair mismatches. In that paper their ability to

clamp wild type β-globin sequences was proved in artificial mixture of DNA samples

enriched with increased amounts of wild type alleles, by using a microchip platform to

detect the β-thalassemia mutations.

Four years later [56] the efficacy of PNA was evaluated with success in 41 non

invasive prenatal diagnosis of β-thalassemia and in combination with three different

techniques (microelectronic chip, pyro sequencing and direct sequencing) to detect fetal

DNA mutations in maternal plasma.

Despite its successful application, this strategy, as the other above described

technologies, is still restricted to couples which carry different mutated alleles and

aimed to the detection of mutated paternal alleles.

Another method recently described for NIPD of β-thalassemia is called APEX

namely Arrayed Primer Extension. This is a mutation detection system which is based

on the combined use of the microchip technology and the single nucleotide base

extension method. This system has been recently described by the group of

Papasavva [57] and used to characterize the presence of the paternal β-thalassemia

mutations and associated β-globin gene SNPs, in cffDNA isolated from maternal

plasma. The possibility to study the polymorphisms associated to the mutated alleles

represent a feature of great value since it would give the possibility to extend NPID to

couples which carry the same mutated allele. Prerequisite for its application is to find

informative SNPs associated with parental mutations which can help to discriminate the

paternal mutated allele and to characterize the haplotype inherited from the fetus. The

authors of the paper described the correct application of this methodology in the NIPD

258

of six out of seven couples at risk for β-thalassemia, carried out in the Cypriot

population.

Future Perspective:

As previously reported, one of the major problems which still limits the application

of the described protocols in clinical practice is the impossibility to obtain highly purified

fetal, cellular as well as cffDNA, samples which could allow the detection of parental

alleles, even when they are identical. Few clinical applications of NIPD are actually

restricted to the detection of the Y chromosome, for fetal sex determination, or the

Rhesus D gene, in Rhesus D negative women, or, in general, of genetic loci which are

absent in the maternal genome.

In recent years a great improvement has been obtained in the field of the

technologies which can explore the presence of sequence variations even in single

molecules of DNA. The concept of "Digital PCR" was firstly introduced in 1992 by

Sykes [58] who described a method to determine the number of starting DNA templates

by doing Poisson statistical analysis of PCR results obtained in limiting dilutions. The

more recent development of the emulsion PCR (emPCR) have further enhanced the

possibility to study single molecules of DNA by using a small volume of reactions,

water-oil emulsions and microfluidic as well as high-throughput platforms (for a review

of both methods and application to NIPD please see Zimmermann et al [59].

Recent applications of these technologies in the field of NIPD, and in particular in

the diagnosis of aneuploidies and monogenic disorders, have shown that these

methodologies may find useful application in the near future, even if several drawbacks

259

need to be solved and wider validation studies should be carried out before transferring

their use in routine diagnostics.

References

1. Wetherall DJ, Clegg JB. The Thalassemia Syndromes. Wiley-Blackwell; 2008.2. Kan YW, Golbus MS, Klein P, Dozy AM. Successful application of prenatal diagnosis

in a pregnancy at risk for homozygous β-thalassemia. New Engl J of Med. 1975; 292:1096-9. [PubMed]

3. Cao A. Results of programs for antenatal detection of thalassemia in reducing the incidence of the disorder. Bool Rev. 1987; 1:169-176. [PubMed]

4. Cao A, Rosatelli MC, Galanello R. Control of β-thalassemia by carrier screening, genetic counseling and prenatal diagnosis: the Sardinian experience Variation in the human genome. Chirchester, England: Wiley; 1996. pp 137-155. Ciba Foundation Symposium 197.

5. Loukopulos D. Current states of thalassemia and sickle cell syndromes in Greece. Semin Hematol. 1996; 33: 76-86. [PubMed]

6. Angastiniotis M, Kyriakidou S, Hadjiminas M. The Cyprus Thalassemia Control Program, White Plains. Vol.23. New York: March of Dimes Birth Defects Foundation; 1988. pp.417-432. Original Article Series. [PubMed]

7. Kan YW, Lee KY, Furbetta M, Angius A, Cao A. Polymorphism of DNA sequence in New Engl J of Med. 1980; 302: 185-8. [PubMed]

8. Pirastu M, Kan YW, Cao A, Conner BJ, Teplitz RL, Wallace RB. Prenatal diagnosis of β-thalassemia. Detection of a single nucleotide mutation in DNA. New Engl J of Med. 1983; 309: 284-7. [PubMed]

9. Rosatelli MC, Tuveri T, Scalas MT, Leoni GB, Sardu R, Faa V, Meloni A, Pischedda MA, Demurtas M, Monni G, et al. Molecular screening and fetal diagnosis of β-thalassemia in the Italian population. Hum Genet. 1992; 83: 590-2. [PubMed]

10. Newton CR, Graham A, Hepteinstall Le, Powell SJ, Summers C, Kalsheker N, Smith JC, Markham AF. Analysis of any point mutation in DNA. The amplification refractory mutation system (ARMS) Nucleic Acid Research. 1989; 17:2503-16. [PMC free article] [PubMed}

11. Saiki RK, Walsh PS, Levenson CH, Erlich HA. Genetic analysis of amplified DNA with immobilized sequence-specific oligonucleotide probes. Proc. Natl. Acad. Sci (USA) 1989; 86:6230-4. [PMC free article] [PubMed]

12. Nickerson DA, Kaiser R, Lappin S, Stewart J, Hood L, Landegren U. Automated DNA diagnostic using an Elisa-based oligonucleotide ligation Assay. Proc. Natl. Acad. Sci (USA) 1990; 87: 8923-7. [PMC free article] [PubMed]

260

13. Pirastu M, Ristaldi MS, Cao A. Prenatal diagnosis of β-thalassemia based on restriction endonuclease analysis of amplified fetal DNA. J Med Genet. 1989; 26: 363-7. [PMC free article] [PubMed]

14. Traeger-Synodinos J, Vrettou C, Kanavakis E. Rapid detection of fetal Mendelian disorders: thalassemia and sickle cell syndromes. Methods Mol Biol. 2008; 444: 133-45. [PubMed]

15. Foglieni B, Cremonesi L, Travi M, Ravani A, Giambona A, Rosatelli MC, Perra C, Fortina P, Ferrari M. Beta-thalassemia microelectronic chip: a fast and accurate method for mutation detection. Clin Chem. 2004 Jan; 50(1): 73-9. [PubMed]

16. Myers RM, Fisher SG, Lerman LS, Maniatis T. Nearly all single base substitution in DNA fragments joined to a GS-clamp can be detected by denaturing gradient gel electrophoresis. Nucleic Acids Research. 1985; 13: 3131-5. [PMC free article] [PubMed]

17. Cai SP, Kan YW. Identification of the multiple β-thalassemia mutations by denaturing gradient gel electrophoresis. J Clin Invest. 190; 85: 550-3. [PMC free article] [PubMed]

18. Rosatelli MC, Dozy A, Faa V, Meloni A, Sardu R, Saba L, Kan YW, Cao A. Molecular characterization of β-thalassemia in the Sardinian population. Am J Hum Genet. 1992; 50: 422-6. [PMC free article] [PubMed]

19. Orita M, Iwahana H, Kanazawa H, et al. Detection of polymorphisms of human DNA by gel electrophoresis as single-strand conformation polymorphism. Proc. Natl. Acad. Sci (USA) 1989; 86: 2766-70. [PMC free article] [PubMed]

20. Sanger F, Micklen S, Coulson AR. DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci (USA) 1977; 74: 5463-7. [PMC free article] [PubMed]

21. Hogge WA, Schonberg SA, Golbus MS. Chorionic villous sampling: experience of the first 1000 cases. Am J Obstet Gynaecol. 1986; 154: 1249-52. [PubMed]

22. Cao A, Cossu P, Monni G, Rosatelli C. Chorionic villous sampling and acceptance rate of prenatal diagnosis. Pren Diagno. 1987; 7: 531-3. [PubMed]

23. Brambati B, Lanzani A, Oldrini A. Transabdominal chorionic villous sampling, clinical experience of 1159 cases. Pren Diagn. 1988; 8: 609-13. [PubMed]

24. Monk M, Holding C. Amplification of a β-hemoglobin sequence in individual human oocytes and polar bodies. Lancet. 1990; 325: 985-8. [PubMed]

25. Handyside AH, Kontogianni E, Hardy K, Winston R. Pregnancies from biopsied human preimplantation embryos sexed by Y-specific DNA amplification. Nature. 1990; 344: 768-70. [PubMed]

26. Verlinsky Y, Ginsberg N, Lifchez A, Valle J, Moise J, Strom CM. Analysis of the first polar body: preconception genetic diagnosis. Hum Reprod. 1990 Oct; 5: 826-9. [PubMed]

27. Strom CM, Rechitsky S, Wolf G, Cieslak J, Kuliev A, Verlinsky Y. Preimplantation diagnosis of autosomal dominant retinitis pigmentosum using two simultaneous single cell assays for a point mutation in the rhodopsin gene. Mol Hum Reprod. 1998 Apr; 4: 351-5. [PubMed]

261

28. Fiorentino F, Biricik A, Nuccitelli A, De Palma R, Kahraman S, Sertyel S, Karadayi H, Cottone G, Baldi M, Caserta D, Moscarini M. Rapid protocol for pre-conception genetic diagnosis of single gene mutations by first polar body analysis: a possible solution for the Italian patients. Prenat Diagn. 2008 Jan; 28: 62-4. [PubMed]

29. Monni G, Cau G, Usai V, Perra G, Lai R, Ibba G, Faa V, Incani F, Rosatelli MC. Preimplantation genetic diagnosis for beta-thalassemia: the Sardinian experience. Prenat Diagn. 2004 Dec 15; 24(12): 949-54. [PubMed]

30. Goossens V, Harton G, Moutou C, traeger-Synodinos J, Van Rij M, Harper JC. ESHRE PGD Consortium data collection IX: cycles from January to December 2006 with pregnancy follow-up to October 2007. Hum reprod. 2009 Aug; 24: 1786-810. [PubMed]

31. Cutting R, Morroll D, Roberts SA, Pickering S, Rutherford A, BFS and ACE Elective single embryo transfer: guidelines for practice British Fertility Society and Association of Clinical Embryologists. Hum Fertil (Camb) 2008 Sep; 11: 131-46. [PubMed]

32. Schmorl G. Pathologisch-anatomische Untersuchungen ueber Publer eklampsie. Vogel; Leipzig: 1893.

33. Douglas GW, Thomas L, Carr M, Cullen NM, Morris R. Trophoblasts in the circulating blood during pregnancy. Am J Obstet Gynecol. 1959 Nov; 78: 960-73. [PubMed]

34. Covone AE, Johnson PM, Mutton D, Adinolfi M. Trophoblast cells in peripheral blood of pregnant women. Lancet. 1984; ii: 841-843. [PubMed]

35. Attwood HD, Park WW. Embolism to the lungs by trophoblasts. J. Obstet. Gynaecol. Br. Commonw. 1960; 68: 611-617. [PubMed]

36. Bertero Mt, Camaschella C, Serra a, Bergui L, Caligaris-Cappio F. Circulating, trophoblast cells in pregnancy have maternal genetic markers. Prenat Diagn. 1988; 8: 585-90. [PubMed]

37. Walknoska J, Conte FA, Grumbach MM. Practical and theoretical implication of fetal/maternal lymphocyte transfer. Lancet. 1969; i: 1119-1122. [PubMed]

38. Herzenberg LA, Bianchi DW, Schroder J, Cann HM, Iverson GM. Fetal cells in the blood of pregnant women: detection and enrichment by fluorescence activated cell sorting. Proc. Natl. Acad. Sci USA. 1979; 76: 1453-1455. [PMC free article] [PubMed]

39. Schroder J, Tilikainen A, de la Chapell A. Fetal leucocytes in maternal circulation after delivery. Transplantation. 1974; 17: 346-360. [PubMed]

40. Bianchi DW, Zickwolf GK, Weil GJ, Sylvester S, DeMaria MA. Male fetal progenitor cells persist in maternal blood for as long as 27 years postpartum. Proc. Natl. Acad. Sci. USA. 1996; 93: 705-708. [PMC free article] [PubMed]

41. Bianchi DW, Flint AF, Pizzimenti M, Knoll JH, Latt SA. Isolation of fetal DNA from nucleated erythrocytes in maternal blood. Proc. Natl. Acad. Sci. USA. 1990; 87: 3279-3283. [PMC free article] [PubMed]

42. Ganshirt-Ahlert D, Burschyk M, Garritsen HSP, Helmer L, Miny P, Horst J, Schneider HP, Holzgreve W. Magnetic cell sorting and the transferrin receptor as potential means of prenatal diagnosis from maternal blood. Am. J. Obstet. Gynecol. 1992; 166: 1350-1355. [PubMed]

262

43. Avent ND, Plummer ZE, Madgett TE, Maddocks DG, Soothill PW. Post-genomics studies and their application to non-invasive prenatal diagnosis. Semin Fetal Neonatal Med. 2008; 13:91-8. Review. [PubMed]

44. Huang R, Barber TA, Schmidt MA, Tompkins RG, Toner M, Bianchi DW, Kapur R, Flejter WL. A microfluidics approach for the isolation of nucleated red blood cells (NRBCs) from the peripheral blood of pregnant women. Prenat Diagn. 2008 Oct; 28: 892-9. [PubMed]

45. Camaschella C, Alfarano A, Gottardi E, Travi M, Primignani P, Caligaris Cappio F, Saglio G. Prenatal diagnosis of fetal hemoglobin Lepore-Boston disease on maternal peripheral blood. Blood. 1990; 75: 2102-2106. [PubMed]

46. Cheung MC, Goldberg JD, Kan YW. Prenatal diagnosis of sickle cell anemia and thalassemia by analysis of fetal cells in maternal blood. Nat Genet. 1996 Nov; 14: 264-8. [PubMed]

47. Di Naro E, Ghezzi F, Vitucci A, Tannoia N, Campanale D, D'Addario V, Holzgreve W, Hahn S. Prenatal diagnosis of beta-thalassemia using fetal erythroblasts enriched from maternal blood by a novel gradient. Mol Hum Reprod. 2000 Jun; 6: 571-4. [PubMed]

48. Kolialexi A, Vrettou C, Traeger-Synodinos J, Burgemeister R, Papantoniou N, Kanavakis E, Antsaklis A, Mavrou A. Noninvasive prenatal diagnosis of beta-thalassemia using individual fetal erythroblasts isolated from maternal blood after enrichment. Prenat Diagn. 2007 Dec; 27: 1228-32. [PubMed]

49. Mendel P, Metais P. Les acides nucleiques du plasma sanguine chez l'homme. C. R. Acad. Sci. Paris. 1948; 142: 241-243.

50. Gahan PB, Swaminathan R. Circulating nucleic acids in plasma and serum. Recent developments. Ann N Y Acad. Sci. 2008 Aug; 1137: 1-6. [PubMed]

51. Lo YM, Corbetta N, Chamberlain PF, Rai V, Sargent IL, Redman CW, Wainscoat JS. Presence of fetal DNA in maternal plasma and serum. Lancet. 1997 Aug 16; 350: 485-487. [PubMed]

52. Microfluidics digital PCR reveals a higher than expected fraction of fetal DNA in maternal plasma. Lun FM, Chiu RW, Allen Chan KC, Yeung Leung T, Kin Lau T, Dennis Lo YM. Clin Chem. 2008 Oct; 54: 1664-72. [PubMed]

53. Detection of paternally inherited fetal point mutations for beta-thalassemia using size-fractionated cell-free DNA in maternal plasma. Li Y, Di Naro E, Vittuci A, Zimmerman B, Holzgreve W, Hahn S. JAMA. 2005 Feb 16; 293: 843-9. [PubMed]

54. Li Y, Di Naro E, Vitucci A, Grill S, Zhong XY, Holzgreve W, Hahn S. Size Fractionation of Cell-Free DNA in Maternal Plasma Improves the Detection of a Paternally Inherited beta-Thalassemia Point Mutation by MALDI-TOF Mass Spectrometry. Fetal Diagn Ther. 2009 Jun 5; 2: 246-249. [PubMed]

55. Cremonesi L, Galbiati S, Foglieni B, Smid M, Gambini D, Ferrari A, Viora E, Campogrande M, Pagliano M, Travi M, Piga A, Restagno G, Ferrari M. Feasibility study for a microchip-based approach for noninvasive prenatal diagnosis of genetic diseases. Ann N Y Acad. Sci. 2004 Jun; 1022: 105-12. [PubMed]

263

56. Galbiati S, Foglieni B, Travi M, Curcio C, Restagno G, Sbaiz L, Smid M, Pasi F, Ferrari A, Ferrari M, Cremonesi L. Peptide-nucleic acid-mediated enriched polymerase chain reaction as a key point for noninvasive prenatal diagnosis of beta-thalassemia. Haematologica. 2008 Apr; 93: 610-4. [PubMed]

57. Papasavva T, Kalikas I, Kyrri A, Kleanthous M. Arrayed primer extension for the noninvasive prenatal diagnosis of beta-thalassemia based on detection of single nucleotide polymorphisms. Ann N Y Acad. Sci. 2008 Aug; 1137: 302-8. [PubMed]

58. Sykes PJ, Neoh SH, Brisco MJ, Hughes E, Condon J, Morley AA. () Quantitation of targets for PCR by use of limiting dilution. Biotechniques. 1992; 13: 444-449. [PubMed]

59. Zimmermann BG, Grill S, Holzgreve W, Zhong XY, Jackson LG, Hahn S. Digital PCR: a powerful new tool for noninvasive prenatal diagnosis? Prenat Diagn. 2008 Dec; 28: 1087-93. [PubMed]

264

Chapter 8

Hemoglobin A1c Zia Uddin, PhD

8.1 Introduction

The term glycated hemoglobin refers to the non-enzymatic, irreversible, covalent

bonding of glucose at one or both N-terminal valine residues of the hemoglobin β-chain,

and the N-terminus of the α-chains, or the ε-amino groups of lysine residues (Figure 1).

Figure 1. Non-enzymatic glycation of hemoglobin

The term normal hemoglobin phenotype beyond the neonatal period involves a

major fraction due to Hb A (α2β2), and a minor fraction of Hb A2 (α2δ2). Occasionally a

very minor fraction of Hb F (α2γ2) is also detected. Further chromatographic analysis of

265

Hb A showed that it contains a number of minor hemoglobins, e.g., Hb A1a1, Hb A1a2,

Hb A1b and Hb A1c.These four minor fractions of Hb A were collectively referred as 1)

Hb A1, 2) “fast hemoglobins”, 3) “glycosylated hemoglobins”, 4) “glycated

hemoglobins” , or 5) glycohemoglobins.” To provide more consistency in nomenclature

the Joint Commission on Biochemical Nomenclature of the International Union of Pure

and Applied Chemistry has recommended the term “glycated hemoglobin” instead of

the above mentioned five names used in the literature.

The Hb A1a1 and Hb A1a2 fractions that are covalently bonded to Glucose-6-

phosphate account for ≈ 10% of the glycated hemoglobin. In Hb A1b the N-terminus of

the β-chain is covalently bonded to pyruvic acid instead of glucose molecule, and this

also accounts for ≈ 10% of the glycated hemoglobin.

Hb A1c is a specific species of glycated hemoglobin resulting from covalent

bonding of glucose to the N-terminal valine of the hemoglobin β-chain.1 Hb A1c

accounts for ≈ 80% of the glycated hemoglobin, and more importantly it is the only

portion of the glycated hemoglobin that is elevated in diabetes. Since hemoglobin

remains in the red blood cell during its entire life span (≈ 120 days), the constantly

changing glucose level in the cell will directly effect the formation of Hb A1c. Therefore,

the measurement of Hb A1c is directly proportional to the time averaged glucose levels.

266

The fast moving forms of Hb A were separated in the late 1950s2 and recognized as

being associated with diabetes in the late 1960s3. Since Hb A1a1, Hb A1a2 and Hb A1b

are not elevated in diabetes, the clinical focus has been solely on Hb A1c. For clinical

testing purposes the term Hb A1c analysis is referred to as A1c or the A1c test with the

word hemoglobin omitted as a matter of convenience.

8.2 Hb A1c Diagnostic Role in Diabetes Mellitus, and Glycemic Control in Adults

After three decades of investigation and evaluation of numerous proposals by

various scientific and clinical organizations, Hb A1c found its status as a diagnostic test

for diabetes mellitus. One of the four criteria4 for the diagnosis of diabetes mellitus are:

Hb A1c > or = 6.5% (48 mmol/mL)

Fasting plasma glucose > or = 126 mg/dL (7.0 mmol/L)

2-h plasma glucose > or = 200 mg/dL (11.1 mmol/L) during an Oral Glucose Tolerance Test

Symptoms of hyperglycemia and casual plasma glucose > or = 200 mg/dL (11.1 mmol/L)

Several investigators have recently suggested the combined use of “fasting glucose and

Hb A1c” for the diagnosis of diabetes mellitus.5-7 Beyond diagnosis, modification of medical

treatment for diabetics is now being performed based on the laboratory test results of Hb A1c.

267

One objective would be to get glucose levels to as close to normal as possible with minimal or

no hypoglycemia. American Diabetic Association (ADA) has suggested the lowering of Hb A1c

< 7% for non-pregnant adults for reducing microvascular, and neuropathic complications of the

disease (type I and II).8 Recently a follow up study9 of the ACCORD (Action to Control

Cardiovascular Risk in Diabetes) stipulated that the best target of Hb A1c in middle aged or

older patients with cardiovascular risk factors is between 7.0 and 7.9%.

Hb A1c is widely used to judge the treatment of diabetes and adjustment of the

medication dose when necessary. In chronic glycemia the blood glucose is monitored more

frequently (once a day or more). Since Hb A1c is measured less frequently and in percent, and

is a complicated process to explain to the patient, it is convenient for the physician to relate

the result to glucose concentration (in mg/dL or mmol/L) over the preceding 5-12 weeks. This

derived glucose concentration from Hb A1c value is called Estimated Average Value (eAG).

Patients monitor their blood glucose and their physician can relate that performance to the

eAG. This way the patients can see the effect of their behavior over time on the test outcome.

The only way this feat could be accomplished, if the result for Hb A1c be the same no matter

where the result was run. This simple feat required the cooperation of many government

agencies and all Hb A1c laboratory testing manufacturing facilities and was brought about by

the determination of the Diabetes Control and Complication Trial (DCCT) Research Group and

the American Diabetes Association.

268

The mathematical relationship then developed between HbA1c and eAG is based on

the following linear regression equation.10

eAG (mg/dL) = (28.7 x Hb A1c %) - 46.7

eAG (mmol/L) = (1.59 x Hb A1c %) - 2.59

Table 1 provides the National Glycohemoglobin Standardization Program (NGSP)

Values11 of Hb A1c % and its corresponding eAG.

Table 1.

NGSP (Hb A1c%) eAG(mg/dL) eAG (mmol/L)

5 97 5.4

6 126 7.0

7 154 8.6

8 183 10.2

9 212 11.8

10 240 13.4

11 269 14.9

12 298 16.5

269

8.3 Measurement of Hb A1c

Currently over 100 different methods are available for quantification of Hb A1c.

Most available Hb A1c methods are certified by the NGSP12 and are based on one of

the following techniques:

● Immunoassay● Boronate affinity binding/HPLC● Ion-exchange HPLC● Capillary zone electrophoresis● Enzymatic

Measurement of Hb A1c was recently reviewed (December 12, 2012) by David B.

Sacks.13 (also available on line)

http://care.diabetesjournals.org/content/35/12/2674.full)). It is encouraging to note that

most of the commercial diagnostic manufacturers for Hb A1c test kits are now

attempting to provide an acceptable Hb A1c for eAG calculation.

8.4 Factors Affecting the Accuracy of Hb A1c Assay

In spite of the efficacy of Hb A1c in the diagnosis and the management of

diabetes (type I and II), several factors influence the accuracy of its laboratory results,

e.g., a) hemolytic disease or other conditions with reduced red blood cell survival, b)

recent blood loss, c) iron deficiency anemia, d) patients with renal failure, and e)

hemoglobin variants. All these interferences cannot be easily delineated by the

270

laboratory personnel and the physician. Due to the diluvial of methods, reagents, and

instruments for the assay of Hb A1c , it is impossible for the laboratory to be aware of

the method’s limitation with respect to the presumptive interference by >1000

hemoglobin variants reported so far in the literature (http://globin.cse.psu.edu). In the

case of the most common hemoglobinoathies (AS, AE, AC, AD), Hb A1c can be

accurately measured if the correct method is used. The affect of these hemoglobin

variants (AS, AE, AC, AD) and elevated Hb F in HPFH (not pathological) on the results

of Hb A1c by the most often used methods is presented in Table 2.

271

272

Table 2. Hb A1c methods: Effects of Hemoglobin Variants (Hb C, Hb S, Hb E and Hb D traits) and Elevated Fetal Hemoglobin (Hb F). Updated March 2013. (with the permission of http://www.ngsp.org/interf.asp). The methods are listed in an alphabetic order of manufacturer’s name. The criteria used to determine whether or not a method shows interference that is clinically significant (indicated by “Yes”) is >

+ or – 7% at 6 and/or 9% Hb A1c. In the absence of data for a specific method (designated by “@”), it can generally be assumed that immunoassay methods do not have clinically significant interference from Hb E and Hb D because the E and D substitutions are distant from the N-terminus of the hemoglobin β-chain. In the absence of data for a specific method (designated by “$”), it can generally be assumed that both immunoassay and boronate affinity methods show interference from Hb F levels above ≈ 10-15%.

In situations where Hb A1c cannot be reliably measured, an alternative is the

assay of serum frustosamine. Fructosamine is the generic name for plasma protein

ketoamines and is also known as glycated serum protein (GSP). Frustosamine provides

273

evaluation of glucose status over a short period of time (2-3 weeks rather than months).

Several studies have shown a correlation of Hb A1c with fructosamine and was thus

recommended in patients with hemoglobinopathies.14

References

1. Sacks DB. Carbohydrates. In Burtis CA, Ashwood ER, eds. Tietz Fundamentals of Clinical Chemistry. 5th Ed. St. Louis: W.B. Saunders 2011; 452-457.

2. Allen DW, Schroeder WA, Balog J. Observations on the chromatographic Heterogeneity of Normal Adult and Fetal Human Hemoglobin: A Study of the Effects of Crystallization and Chromatography on the Heterogeneity and Isoleucine Content. Am J Chem Soc 1958; 80: 1628-34.

3. Rahbar S. An abnormal hemoglobin in red cells of diabetics. Clin Chim Acta 1968; 22: 296-98

4. Sacks DB, Arnold M, Bakris GL, Bruns DE, Horvath AR, Kirkman MS, Lernmark A, Metzger BE, Nathan DM. Guidelines and Recommendations for Laboratory Analyzers in the Diagnosis and Management of Diabetes Mellitus. Clin Chem 2011; 57: 793-798.

5. Inzucchi SE. Diagnosis of Diabetes. N Engl J Med 2012; 367: 542-50.6. Hu Y, Kiu W, et al. Combined use of fasting plasma glucose and glycated

hemoglobin A1c in the screening of diabetes and impaired glucose tolerance. Acta Diabetol 2010; 47: 231-36.

7. Heianza Y, Hara S, Arase Y, et al. Hb A1c 5.7-6.4% and impaired fasting plasma glucose for diagnosis of prediabetes and risk of progression to diabetes in Japan (TOPICS 3): a longitudinal study. Lancet 2011; 378: 147-55.

8. American Diabetic Association Clinical Practice Recommendations: Executive Summary: Standard Methods of Care in Diabetes-2010. Diabetes Care 2010: 33, suppl. 1: S4-5.

9. Gerstein HC, Miller ME, Genuth S, et al. ACCORD Study Group. Long term effects of intensive glucose lowering on cardiovascular outcomes. N Engl J Med 2011; 364 (9): 818-828. 10. Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ,

and for the A1c-Derived Average Glucose (ADAG) Study Group. Translating the A1c Assay into Estimated Average Glucose Values. Diabetes Care 2008; 31: 1473-1478.

274

11. Nathan DM, Kuenen J, Borg R, Zheng H, Schoenfeld D, Heine RJ, FOR THE A1c-Derived Average Glucose (ADAG) Study Group. Translating the A1c Assay into Estimated Average Glucose Values.

Diabetes Care 2008; 31: 1-6. 12. List of NGSP Certified Methods-Hb A1c (updated 11/2012).

http:/www.ngsp.org

13. Sacks DB, Measurement of Hemoglobin A1c. A new twist on the path to harmony. Diabetes Care 2012; 35 (12): 2674-2680.

14. http://labtestsonline.org/understanding/analytes/fructosamine/tab/test.

Additional references (not quoted above) concerning hemoglobin variant

interference in the assay of Hb A1c.

i) Sofronescu A-G, Williams LM, Andrews DM, Zhu Y. Unexpected

Hemoglobin A1c Results. Clin Chem 2011; 57:2, 153-157ii) Selvin E, Steffes MW, Ballantyne CM, Hoogeveen RC, Coresh J, Brancati

FL. Racial Differences in Glycemic Markers: A cross-sectional Analysis of Community-Based Data. Ann Inter Med 2011; 154: 303-309

iii) Bergman A-C, Beshara S, Byman I, Karim R, Landin B. A New β-Chain

Variant: Hb Stockholm [β7(A4)Glu→Asp] Causes Falsely Low A1c. Hemoglobin 2009; 33(2): 137-142

iv) Williams JP, Jackson H, Green BN. Hb Belleville [β10(a&)Ala→Thr]

Affects the Determination of HbA1c by Routine Cation Exchange High Performance Liquid Chromatography. Hemoglonin 2009; 33(1): 45-50.

v) Zhu Y, Williams LM. Falsely elevated hemoglobin A1c due to S-beta+-

thalassemia interference in Bio-Rad Variant II Turbo HbA1c assay. Clin Chem Acta 2009; 409(1-2): 18-20.

vi) Thevarajah M, Nadzimah MN, Chew YY. Interference of hemoglobin A1c

(HbA1c) detection using ion-exchange high performance liquid chromatography (HPLC) method by clinically silent hemoglobin variant in University Malaya Medical Center (UMMC)- A case report. Clin Biochem 2009; 42: 430-434.

vii) Mongia SK, Little RR, Rohlfing CL, Hanson S, Roberts RF, Owen WE, D’Costa MA, Reyes CA, Luzzi VI, Roberts WL. Effects of Hemoglobin C and S on the Results of 14 Commercial Glycated Hemoglobin Assays. Am J Clin Pathol 2008; 130: 136-140.

viii) Barakat O, Krishnan STM, Dhatariya K. Falsely low HbA1c value due to a rare variant of hemoglobin J-Baltimore. Primary Care Diabetes 2008; 2: 155-157.

275

ix) Little RR, Rohlfing CL, Hanson S, Connolly S, Higgins T, Weykamp CW, D’Costa M, Luzzi V, Owen WE, Roberts WL. Effects of Hemoglobin (Hb) E

and HbD Traits on Measurements of Glycated Hb (HbA1c) by 23 Methods. Clin Chem 2008; 54:8, 1277-1282.

x) Lee S-T, Weykamp CW, Lee Y-W, Kim J-W, Ki C-S. Effects of 7 Hemoglobin Variants on the Measurement of Glycohemoglobin by 14 Analytical Methods. Clin Chem 2007; 53(12): 2202-2205.

xi) Roberts WL. Hemoglobin Constant Spring can interfere with Glycated Hemoglobin Measurements by boronate Affinity Chromatography. Clin Chem 2007; 53(1): 142-43.

276

Case Studies

Introduction

The following case (# 1-28) studies include laboratory data representing results

from five different hemoglobin separation methods commonly used in the clinical

laboratory. Due to the large number of variants possible the mandate is that

more than one separation method be used in identification. The question is which

two methods would provide discriminative information. The results of the lab

tests for each case are presented in a tabular form to assist in these choices.

The alkaline electrophoresis images are of Helena SPIFE Alkaline

Electrophoretic results but identical separation results would have also been

obtained using alkaline cellulose acetate, Helena Biosciences SAS alkaline

hemoglobin gels, Helena Quick Gels or Sebia Hydrasys alkaline hemoglobin

gels.

The acid electrophoretic images are of Helena SPIFE or QUICK Gel Acid

electrophoresis. For Acid electrophoretic separation, two classes of media have

been used with differing separation results. Historically, acid hemoglobin

separation was done on agar using citric acid buffer. Helena SPIFE and Quick

Gels are of this type. Agarose purified from agar has more recently been used by

Beckman, Sebia and Helena BioSciences. The purified nature of the agarose

makes these products easier to produce but historically they lacked easily

available documentation of the differences in mobilities compared to the

277

historically used agar. These differences have been documented in the table

associated with the attached case studies. All acid agarose data were adopted

from (Variant Hemoglobins. a Guide to Identification. 1st edition, by Barbara J.

Bain , Barbara J. Wild , Adrian D. Stephens, Lorraine A. Phelan . Published

2010 by Wiley-Blackwell Publishing Ltd).

All Capillary Zone Electrophoresis (CZE) data were generated using the Sebia

Capillarys System. This CZE system separates hemoglobins into 15 zones and

provides a list of possible variants that migrate in that zone. The operator then

selects the hemoglobin variant they expect that peak to represent. The peaks in

the CZE reports in the case studies have been labeled in such a fashion but a

different assignment could have been made by the operator had they had

information warranting the choice. Details of other vendor results would require

contact with the vendor but the goal again would be to maintain equality as close

as possible and the assumption would be that the order of separation would not

be different.

All isoelectric focusing images are actual or simulated from actual data obtained

with the Helena Isoelectric Focusing Gels either on the SPIFE or the REP

systems. The Perkin Elmer Resolve (formerly IsoLab) isoelectric focusing

systems would obtain the same results, because the pH range of the ampholytes

are the same. These agarose gels contain acrylamide to sharpen the bands.

278

Again the end user is the final discriminator. In this case, proper selection of the

controls determines the degree of discrimination possible instead of the number

of Zones available. If there is Hb S control and the variant migrates anodal to Hb

S, the variant might be Hb D or Hb G but you may not reliably report which, even

though they are both anodal to Hb S. You only know it is not Hb S. If the control

is Hb D or Hb G then you may report based on the migration compared to that

control.

The High Performance Liquid Chromatography (HPLC) separations were all

obtained using BioRad Variant information from several sources.

This data for cases 1-28 is the cooperative effort of many institutions. 

Hemoglobin screening is done on neonates as well as adults.  Sometimes data

from these rarer hemoglobin variants may include Hb F at low levels that is

ignored in the discussion because its presence is to be expected due to the

patient’s age. In this regard, some discrepancy in the data may appear.  The

presence of an alpha chain variant on a newborn can be complicated by this

temporary presence of gamma chains. The gamma chains compete for the

variant alpha chains as well as the normal resulting in two gamma alpha

possibilities.  In neonates the Hb A2 is barely visible because delta chain

production is just beginning. If sufficient delta chains are expressed they also

would show a competition for alpha bands resulting in Hb A2 and a smaller

alpha variant band.  These complications will be discussed in the cases in which

they are encountered.

279

Case # 1 Normal Adult

71 years old male, recent medical examination showed no abnormality.

Laboratory Data:

Hemoglobin 13.5 13.5 -18.5 g/dL Hematocrit 39.6 38.0 - 54.0 %

RBC 4.7 4.6 - 6.2 Mil/mm3

MCV 84.3 80 - 100 fL MCH 28.8 27 - 34 pg MCHC 34.1 31 - 36% RDW 13.5 11.5 - 14.5%

Platelet 200 150 - 400 Th/mm3

Hb A 98.0 94.3 - 98.5%

Hb A2 1.8 1.5 - 3.7% Hb F ≈0.2 0.0 - 2.0%

Peripheral Blood Smear: No abnormality was detected.

Agarose Gel Electrophoresis (pH 8.6)

280

Case # 1 Normal Adult

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

281

Case # 1 Normal Adult

Capillary zone electrophoresis

High performance liquid chromatography

282

Case # 1 Normal Adult

Interpretation & Discussion:

Agarose gel electrophoresis at alkaline pH 8.6 showed a major band

(98%) in the position of Hb A, and a very faint band of Hb A2 (1.8%) in the

position of Hb C. Another very faint band detected cathodal to Hb A2 is due to

the enzyme carbonic anhydrase. This carbonic anhydrase band is mostly

detected in fresh specimens of blood.

Acid electrophoresis at pH 6.2 does not separate Hb A from Hb A2 ,

therefore only one major band is shown in the position of Hb A. A smudged band

cathodal to Hb A includes Hb F, and modified forms of Hb A such as HbA1c.

Isoelectric focusing showed a major band in the position of Hb A, and a

very faint band in the position of Hb A2. A smudged minor band anodal to Hb A

represents modified Hb A such as acetylated Hb F, denatured Hb A, and Hb A1c.

From capillary zone electrophoresis a major peak of Hb A was detected in

window Z9, and a minor peak due to Hb A2 was present in window Z3. No other

peaks were observed.

High performance liquid chromatography showed a major peak at a

retention time of 2.42 (peak value) ascribed to Hb A and a minor peak due to Hb

A2 at a retention time of 3.64 (peak value). There are 2-3 minor peaks before Hb A

and after Hb F, and these peaks represent Hb A1c fraction besides fractions of

other hemoglobins.

283

The term normal hemoglobin phenotype beyond the neonatal period

involves a major band due to Hb A (α2β2), and a minor band of Hb A2 (α2δ2).

Occasionally a very faint band of Hb F (α2γ2) is also detected. By definition the

concentration of both the Hb A2 and Hb F must be in the normal range for that

method regardless of the methodology used.

Reference

Bain BJ. Hemoglobin and the genetics of hemoglobin synthesis: In:Haemoglobinopathy Diagnosis, Blackwell Publishing, second edition, 2006, pp 12-22.

284

Case # 2 Hemoglobin S trait

A 20 year old female African-American pre-nursing student in a local community college was screened for hemoglobinopathy by her family physician. Her physical examination and chemistry profile were normal.

Laboratory Data:

Hemoglobin 13.1 12.0 – 16.0 g/dL Hematocrit 39.6 35.0 - 48.0 %

RBC 4.4 4.0 – 5.5 Mil/mm3

MCV 82.1 79-98 fL MCH 27.3 26-34 pg MCHC 32.1 31-36% RDW 12.6 11.5-14.5%

Platelet 267 150-400 Th/mm3

Hb A 59.2 94.3-98.5%Hb S 38.4

Hb A2 1.8 1.5 -3.7% Hb F 0.6 0.0-2.0%

Peripheral Blood Smear: No abnormality was present.Solubility test for Hb S was positive.

Agarose Gel Electrophoresis (pH 8.6)

285

Case # 2 Hemoglobin S trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

286

Case # 2 Hemoglobin S trait

Capillary zone electrophoresis

High performance liquid chromatography

287

Case # 2 Hemoglobin S trait

Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose Major band (Hb A)

Major band(Hb S)

Minor band

(Hb A2)Acid Agar/Agarose

Major band (Hb A+

Hb A2)

Major band(Hb S)

CZE Major peak(Hb A)Zone 9

Major peak(Hb S)Zone 5

Minor peak

(Hb A2)Zone 3

IEF Major band (Hb A)

Major band(Hb S)

Minor band

(Hb A2)HPLC Major

peak(Hb A) RT=2.34

Major peak(Hb S) RT=4.26

Minor peak(Hb A2) RT=3.65

Since the solubility test was positive and the aberrant band fell between

35 – 40%, a diagnosis of Hb S trait was made. Concentrations of Hb S other than

35 – 40% require consideration of the effect of a transfusion, the possibility of

iron deficiency, a concurrent Hb S-α-thalassemia (Hb S < 33%), a Hb S-β-

thalassemia (Hb S >49%) or the possibility that the fraction may not be Hb S at

all. Mutation at the 6th amino acid position of the β chain [β6 (A3) Glu→Val)

causes the substitution of glutamic acid by valine that results in the formation of

Hb S.

288

Since one negative charge is reduced by this mutation, Hb S migrates

slower than Hb A in alkaline and acid electrophoretic procedures. There are other

Hb variants that migrate in the position of Hb S in alkaline electrophoresis, but

not in acid. Use of Acid electrophoresis eliminates all the other common

hemoglobin variants that migrate in the Hb S alkaline area or by CZE, IEF or

HPLC. Other identification methods do exist.

Individuals with Hb S should be advised that it is almost a benign and

innocuous condition2. However, there are exceptions and in some individuals:

hematuria and aseptic necrosis of bone has been reported. If the hematuria

persists for a long time and is profuse, then the possibility of bladder cancer by

cystoscopy and bladder cancer markers must be evaluated.

Recently,3 a new sickling hemoglobin (Hb S-San Martin) was reported

from an Argentinean family. Besides the usual β-globin chain mutation

associated with sickle cell [β6(A3)Glu→Val, (GAG→GTG)], an additional

mutation on the same β-globin chain [β105 (G7) Leu→Pro (CTC→CCC) ] was

confirmed by the DNA studies. The electrophoretic mobility of Hb S-San Martin

at both the alkaline pH (8.6) and acid pH (6.2) was identical with the Hb S. This is a

rare occurrence and only ten (10) hemoglobin variants out of >1000 variants

discovered so far have double mutation on the same β-globin chain besides the

sickle cell mutation.

289

Case # 2 Hemoglobin S trait

References

1. Bain BJ. Sickle cell haemoglobin and its interactions with other variant haemoglobins and with thalassaemias. In: Bain BJ, Ed. Haemoglobinopathy Diagnosis, 2nd edition, Blackwell Publishing; 2006:141-149.

2. Steinberg MH. Sickle cell trait. In: Steinberg MH, Forget BG, Higgs DR, Nagel RC, eds. Disorders of Hemoglobin: Genetics, Pathophysiology and Clinical Management. Cambridge, England: Cambridge University Press; 2001: 811-830.

3. Feliu-Torres A, Eberle SE, Bragos IM, Sciuccati G, Ojeda MJ, Calvo KL, Voss ME, Pratti AF, Milani AC, Bonduel M, Diaz L, Noguera NI. Hb S-San Martin: A new sickling hemoglobin with two amino acid substitutions [β6(A3)Glu→Val;Β105(G7)Leu→Pro]. Hemoglobin 2010; 34(5): 500-504.

290

Case # 3 Hemoglobin S homozygous

A 21 years old African American female came to the emergency department of a hospital complaining abdominal and joint pain.

Laboratory Data:

Hemoglobin 7.6 12.0 – 16.0 g/dL

RBC 2.6 4.0 - 5.5 Mil/mm3

MCV 80.0 78 - 98 fL RDW 21.0 11.5 -14.5%

Platelet 201 150 - 400 Th/mm3

Hb A ≈4.2 94.3 - 98.5%Hb S 90.0

Hb A2 2.8 1.5 - 3.7% Hb F 3.0 0.0 - 2.0% (Hemoglobin fractions from HPLC)

Peripheral Blood Smear: 2+ poly morphic, 1+ target cells, few Howell-Jolly bodies, sickle cells

Sickle cell solubility test for Hb S: Positive.

Agarose Gel Electrophoresis (pH 8.6)

291

Case # 3 Hemoglobin S homozygous

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

292

Case # 3 Hemoglobin S homozygous

Capillary zone electrophoresis

Note: The original CZE on this specimen showed no presence of Hb A, therefore the analysis was repeated after mixing the specimen 1:1 with a normal blood. This is the standard practice in cases whenever Hb A is not detected.

High performance liquid chromatography

293

Case # 3 Hemoglobin S homozygous

Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band (Hb S)

Minor band

(Hb A2)Acid Agar / Agarose

Major band (Hb S)

CZE Major peak(Hb S) Zone 5

Minor peak

(Hb A2)Zone 3

IEF Major band (Hb S)

Minor band

(Hb A2)HPLC Major

peak(Hb S)RT=4.38

Minor peak(Hb A2)RT=3.6

Minor peak (Hb F)RT=1.04

Agarose gel electrophoresis (alkaline pH 8.6) and citrate agar

electrophoresis (acid pH 6.2) showed only one major band in the position of Hb

S. In view of the positive sickle cell solubility test a diagnosis of homozygous Hb

S disease was apparent.

It must be emphasized that due to co-migrating hemoglobin variants a

confirmatory discriminatory test must be run. The selection of these confirmatory

tests must be done with an eye on the results, for instance CZE would not be a

294

good confirmatory test for the identification of Hb S vs Hb Dhofar following

alkaline electrophoresis, because the migration is not different. Acid

electrophoresis will suffice as confirmation for either of them. Other readily

available tests that can be of use are HPLC and IEF.

If hemoglobinopathy testing is performed within three months of a blood

transfusion, the separation pattern will indicate the presence of Hb A from the

transfused blood and thus complicate the interpretation of the results. Therefore,

it is advised that all the laboratories obtain the blood transfusion history before

interpreting hemoglobin results.

Sometimes it is impossible to know the patient transfusion history,

especially if the patient arrived in the emergency department of the hospital.

About eight years ago, a very unusual case was observed by me in our hospital.

The Hb S diseased patient without insurance and facing sickle cell crisis went to

the emergency department of a large hospital in Detroit. The patient was

transfused with two units of blood and then discharged. He felt a little better after

blood transfusion, but two days later he went to the emergency department of

another large hospital in Detroit and received a second transfusion. Two days

later, this patient was examined in the emergency department of our hospital. In

our laboratory, the hemoglobin assays indicated Hb A (60%), Hb S (34%), Hb A2

(2.5%), and Hb F (3.5%). These results are suggestive of Hb S trait without

knowing the blood transfusion history of the patient. Therefore, in order to make

a correct diagnosis of a hemoglobin variant, it is prudent to know the recent blood

transfusion record.

295

Case # 3 Hemoglobin S homozygous

References

1. Kutlar A. Sickle Cell Disease: A Multigenic Perspective of a Single Gene Disorder. Hemoglobin 2007; 31 (2): 209-224.2. Steinberg MH. Genetic Etiologies for Phenotypic Diversity in Sickle Cell

Anemia. The Scientific World Journal 2009; 9: 46-67.3. Bain BJ. Sickle cell anemia, In: Bain BJ, Ed. Hemoglobinopathy

Diagnosis, 2nd edition, Blackwell Publishing; 2006: 150-164.4. Beutler E. The sickle cell diseases and related disorders. In: Beutler E,

Lichtman MA, Coller BS, Kipps TJ, Seligsohn U, eds. Williams Hematology, 6th ed. New York, NY: McGraw-Hill; 2000: 581-606.

5. Nagel RC, Platt VS. General pathophysiology of sickle cell anemia. In: Steinberg MH, Forget BG, Higgs DR, Nagle RL, eds. Disorders of Hemoglobin: Genetics, Pathophysiology and Clinical Management. Cambridge, England: Cambridge University Press; 2001: 494-526.

296

Case # 4 Hemoglobin S with hereditary persistence of fetal hemoglobin (HPFH)

African-American adult male, apparently healthy and without any previously known major clinical condition visited his family physician for his annual check-up.

Laboratory Data:

Hemoglobin 13.4 13.5 - 18.5 g/dL

RBC 4.78 4.6 - 6.2 Mil/mm3

MCV 80.9 80 - 100 fL MCH 28.0 27 - 34 pg

Hb A 5.7 94.3 - 98.5Hb S 56.0

Hb A2 3.3 1.5 - 3.7% Hb F 35.0 0.0 - 2.0% (Hemoglobin fractions from HPLC)

Peripheral Blood Smear: No abnormality was noticed. Sickle cell solubility test for Hb S: Positive.

Agarose Gel Electrophoresis (pH 8.6)

297

Case # 4 Hemoglobin S with hereditary persistence of fetal hemoglobin (HPFH)

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

298

Case # 4 Hemoglobin S with hereditary persistence of fetal hemoglobin (HPFH)

Capillary zone electrophoresis

High performance liquid chromatography

299

Case # 4 Hemoglobin S with hereditary persistence of fetal hemoglobin (HPFH)

Interpretation & Discussion:

Summary of Results

Method Hb Farea

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band (Hb F)

Major band (Hb S)

Faint band

(Hb A2)Acid Agar / Agarose

Major band (Hb F)

Major band (Hb S)

CZE Major peak(Hb F)Zone 7

Major peak(Hb S ) Zone 5

Minor peak

(Hb A2) Zone 3

IEF Major band(Hb F)

Major band (Hb S)

Faint band

(Hb A2)HPLC Major

peak(Hb F)RT=1.16

Major peak(Hb S) RT=4.38

Minorpeak

(Hb A2)RT=3.6

These laboratory results must have been somewhat of a surprise for an

asymptomatic patient. The few commonly encountered hemoglobins which

migrate in the position of Hb S on alkaline agarose gel electrophoresis are Hb D,

Hb G and Lepore all of which are ruled out by the results of acid agar gel

electrophoresis. In addition the positive Hb S solubility test assures this patient

has only Hb S and High Persistence of Fetal Hemoglobin. Analysis of this

specimen by CZE requires a modification of the procedure because there is no

hemoglobin A present for the software to use as a home base for comparison to

other hemoglobin mobilities. The CZE analysis would be repeated after mixing in 300

1:1 ratio with a normal blood specimen.

From all the five laboratory methods, three abnormalities are evident:

a) Absence of Hb A b) Hb S (≈56%)c) Hb F (≈35%)

The percentage of Hb S and Hb F suggests the following diagnostic possibilities:

i) Homozygous Hb S disease with failure to suppress Hb F productionii) Hb S-β- thalassemia, with failure to suppress Hb F productioniii) Hb S- HPFH due to a deletional mutation in the non S gene

Hb S – HPFH patients are the result of a point mutation on one beta gene

forming Hb S and a deletion of the delta and beta area on the other gene

permitting the production of Hb F to continue. The Hb F expression will be 25 –

35%. This is a pancellar condition so every erythrocyte will contain Hb F as well

as Hb S and the damage caused by Hb S is not seen. Generally speaking

patients with Hb S-HPFH are clinically well, with a benign clinical course, little

evidence of hemolysis and without severe anemia. It is prudent to make a

clinical diagnosis based on all available resources. In this case other laboratory

data showed a positive sickle solubility test, a normal CBC, serum iron and

ferritin, and no other abnormalities except for some sickling. Consultation with the

physician indicated the patient was clinically well and certainly had not been

treated with hydroxyurea. This patient is presumed to be Hb S – HPFH.

Approximately 1% of Homozygous S patients present with 5% or less Hb

F and these patients clinically do better than those without Hb F. Therefore much

301

has been done to increase the production of Hb F in homozygous S patients in

general. Degree of success of hydroxyurea treatment has been very variable for

unknown but at least to some extent genetic reasons. A patient with

homozygous Hb S disease may present with Hb F levels (15% - 30%) following

treatment with Hydroxyurea. Among the symptom ameliorating effects of

hydroxyurea is the apparent interference in the suppression of Hb F manufacture

and the production of nitric oxide. Since Hb F is higher in oxygen affinity than Hb

S and deoxyhemoglobin S polymerizes, its presence protects the cells from

sickling and other but not all symptoms of Sickle Cell Disease. The problem with

this type of fetal persistence is that it is not pancellular. Not all erythrocytes

contain Hb F even though the Hb F is elevated. Those cells without the Hb

F are not protected. That said as a result of several Clincal Trials including BABY

HUG several agencies have recommended use of hydroxyl urea for treatment of

Sickle Cell Disease [McGann PT, Ware RE. Hydroxyurea for sickle cell anemia:

What have we learned and what questions still remain? Curr Opin Hematol 2011;

18(3): 158-165].

In the unlikely circumstance that it was not known if the patient had been

treated with hydroxyl urea there is the possibility that he might have been a

homozygous patient who at the moment his blood was drawn was not very

symptomatic but his Hb F had been chemically altered. The two conditions could

be separated by doing a Kleihuer Betke acid elution test or flow cytometry

(monoclonal antibody agains γ- chains) for the study of pancellular vs

302

heterocellular distribution of the Hb F. An essentially homogeneous distribution

establishes the Hb S-HPFH diagnosis.

Hb S-β-thalassemia is also highly unlikely because of the clinical picture.

Patients with Hb S- β-thalassemia even in the presence of Hb F would have a

thalassemic clinical picture. Hydroxyurea has been used for treatment of beta-

thal patients with some success so the possibility exists that it might be helpful in

a case of Hb S - β-thalassemia. Far less data exists on this treatment even

though it is known that the presence of Hb F lessens the clinical picture.

A few cases of clinical aberrations, e.g. minor joint or abdominal pains,

asceptic necrosis of bone, palpable spleen were reported in persons with Hb S-

HPFH (Fairbanks VF. Hemoglobinopathies and Thalassemias. New York, NY:

Brian C. Decker; 1980:136). Recently Whyte et al (see below reference # 5)

reported massive splenic infarction in an adolescent with Hb S-HPFH. Therefore

the condition is not benign.

References

1. Murray N, Serjeant BE, Serjeant GR. Sickle cell-hereditary persistence of fetal hemoglobin and its differentiation from other sickle cell syndromes. Br J Haemotol 1988; 6: 89-92. (available online since March 2008).2. Hoyer JD, Connie SP, Fairbanks VF, Hanson CA, Katzmann JA. Flow cytometric measurement of hemoglobin F in RBCs: Diagnostic usefulness in the distinction of hereditary persistence of fetal hemoglobin (HPFH) and hemoglobin S-HPFH from other conditions with elevated levels of hemoglobin F. Am J Clin Pathol 2002; 117: 857-863.3. Akinsheye I, Al-Sultan A, Solovieff N, Ngo D, Baldwin CT, Sebastiani P, Chui DH, Steinberg MH. Fetal hemoglobin in sickle cell anemia.Blood 2011; 118: 19-27.

303

4. Ngo D, Aygun B, Akinsheye I, Hanjins JS, Bhan I, Luo HY, Steinberg MH, Chui DH. Fetal haemoglobin levels and haematological characteristics of compound hegterozygotes for haemoglobin S and deletional hereditary persistence of fetal hemoglobin. Br J Haematol 2012; 156(2): 259-64.5. Whyte D, Forget BG, Chui DH, Luo HY, Pashankar F. Massive splenic infarction in an adolescent with hemoglobin S-HPFH. Pediatr Blood Cancer 2013; 60(7): 49-51.6. Chapter 2.3 of this book: Bernard G. Forget, MD. Hereditary Persistence of Fetal Hemoglobin7. Bain BJ. Hereditary persistence of fetal haemoglobin and other inherited causes of an increased proportion of haemoglobin F. In: Haemoglobinopathy Diagnosis, Blackwell Publishing, second edition, 2006, pp119-127.

304

Case # 5 Hemoglonin G-Philadelphia trait

Adult African-American male with no abnormalities.

Laboratory Data:

Hemoglobin 14.5 13.5 -18.5 g/dL

RBC 5.06 4.6 - 6.2 Mil/mm3

MCV 84.0 80 - 100 fL MCH 28.7 27 - 34 pg

Hb A 77.1 94.3 - 98.5%

Hb A2 0.9 1.5 - 3.7% Hb F ≈0.0 0.0 - 2.0%

Hb G 22%(Hemoglobin fractions from HPLC)Peripheral Blood Smear: No abnormality was detected. Sickle cell solubility test: Negative.

Agarose Gel Electrophoresis (pH 8.6)

305

Case # 5 Hemoglobin G-Philadelphia trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

306

Case # 5 Hemoglobin G-Philadelphia trait

Capillary zone electrophoresis

High performance liquid chromatography

307

Case # 5 Hemoglobin G-Philadelphia trait

Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose Major band (Hb A)

Major band(Hb G)

Minor band

(Hb A2)

Minor band close to carbonic anhydrase

Acid Agar/Agarose

Major band(Hb A+

Hb A2 +Hb G+

Hb G2)CZE Major

peak(Hb A)Zone 9

Major peak(Hb G) Zone 6

Minor peak (Hb

A2)Zone 3

Minor peak

(Hb G2)Zone 1

IEF Major band(Hb A)

Major band anodal to Hb S(Hb G)

Minor band

(Hb A2)

Minor band

(Hb G2)as far cathodal to A2 as G is anodal to it.

HPLC* Major peak(Hb A)RT=2.45

Major peak(Hb G) RT=4.04

Minor peak

(Hb A2)RT=3.6

Minor peakRT=4.5-4.6

* Note: HPLC retention time (RT) varies with the type of the instrument used and several other factors, e.g. temperature etc.

Electrophoretic migration bands at less than 1% may be difficult to detect

on alkaline electrophoresis. If the presence of a minor band is expected the

sample amount may be increased. Quantification of the results must not be done

308

on over applied samples because you will have exceeded the quantitative

linearity of the system.

The sickle cell solubility test was negative, ruling out the possibility of Hb

S. The separation table shows the presence of a non Hb S band migrating in the

Hb S area for all methods except Acid Electrophoresis. Common variants found

anodal to Hb S on alkaline electrophoresis that migrate in Hb A position on acid

electrophoretic conditions are Hb D, Hb G-Philadelphia and Lepore. Of these

options only Hb G- Philadelphia is an α-chain variant. If α-chain variant is

expressed in a large enough percentage to compete with normal α-chains for

combination with δ chains a new small modified delta band is created. In

individuals with Hb G-Philadelphia [α68(E17)Asn→Lys], a combination of Hb G-

Philadelphia α-chains with normal δ-chains leads to the formation of about 1% of

a molecule Hb G2 (α2Gδ2). Hb G2 has no clinical significance, but plays an

important role in the distinction between Hb D and Hb G-Philadelphia. Since Hb

G-Philadelphia is entirely innocuous, globin chain electrophoresis and DNA

studies are usually not necessary.

There is a temptation to analyze the available hemoglobin variants by

percentage since Hemoglobin Lepore runs less than 15 % and Hb D runs about

40 while Hb G-Philadelphia trait runs 20-25% in the heterozygote. Differentiation

between Hb D and Hb G-Philadelphia on the basis of the percentage of the

variant is not advised because the percentages of either would be effected by a

concurrent α-thalassemia -2 trait or homozygous α-thalassemia-2 (see below).

309

The single alpha gene deletion resulting in α-thalassemia-2 trait is found in 1/3 of

African Americans therefore this silent mutation could be likely found in

association with Hb G-Philadelphia in this ethnic population.

Of the four alpha genes located on chromosome 16 (two on each

chromosome), alpha gene mutations lead to the following possibilities (adopted

with the permission of College of American pathologists: Hoyer JD and Kroft SH,

eds. Color Atlas of Hemoglobin Disorders. College of American Pathologists,

Northfield, IL, 2003; 67).

1. Hb G trait with no thalassemia, Hb G 20 -25% no hematologic effect

2. Hb G trait, One α gene deleted (α-thalassemia-2 trait), Hb G 25-35% usually no hematologic effect

3. Hb G-trait; Two α genes deleted (homozygous α-thallasemia-2), Hb G 35-45%; microcytosis.

4. Homozygous Hb G; Two α genes deleted (homozygous α- thalassemia-2), Hb G 95%, microcytosis.

References

1. Keren DF. Clinical Evaluation of Hemoglobinopathies: Part II. Structural Changes, Ward Medical Laboratory, Archived Issues 2003; 3: 1-11.Available online (http://www.wardlab.com/14-3.html).

2. Hoyer JD and Kroft SH, eds. Color Atlas of Hemoglobin Disorders. College of American Pathologists, Northfield, IL, 2003; 65.

3. Bain BJ. Hemoglobin G-Philadelphia trait: In: Haemoglobinopathy Diagnosis, Blackwell Publishing, second edition, 2006, pp 212.

4. Milner PF, Huisman TH. Studies of the proportion and synthesis of haemoglobin G-Philadelphia in red cells of heterozygotes, a homozygote, and a heterozygote for both haemoglobin G and alpha thalassemia. Br J Haematol 1976; 34: 207-220. (Available online from July 2008).

310

5. Baine BM, Rucknagel DL, Dublin DA Jr, Adams JG III. Trimodality in the proportion of hemoglobin G-Philadelphia in heterozygotes; evidence of heterogeneity in the number of human alpha chain locations. Proc Natl Acad Sci. 1976; 73: 3633-36.

6. Reider RF, Woodbury DH, Rucknagel DL. The interaction of α- thalassemia and hemoglobin G-Philadelphia. Br. J Haematol. 1976; 32: 159-65.

7. Khalil MSM, Timbs A, Hendrson S, Schuh A, Hussein MRA, Old J. Haemoglobin (Hb) G-Philadelphia, Hb Stanleyville-II, Hb G-Norfolk, Hb Matsue-Oki and Hb Mizushi can form a panel of α-chain variants that overlap in their phenotype: the novel use of StyI to screen for Hb G- Philadelphia. Intl Jnl Lab Hem 2011; 33: 318-325.

311

Case # 6 Hemoglobin S-G Philadelphia

Adult African American female who was asymptomatic.

Laboratory Data:

Hemoglobin 11.7 12.0 -16.0 g/dL

RBC 4.29 4.0 – 5.5 Mil/mm3

MCV 81.6 79 - 98 fL MCH 27.4 27- 34 pg

Hb A 54.0 94.3 – 98.5%Hb S 19.9 Hb G 17.8

Hb A2 1.1 1.5-3.7% Hb F 0.2 0.0-2.0% Hb S-G Hybrid 7.0(Hemoglobin fractions from HPLC)Peripheral Blood Smear: No abnormality.Sickle cell solubility test for hemoglobin S: Positive. Unstable hemoglobin (isopropanol) Test: Negative. No record of blood transfusion during the past six months.

Agarose Gel Electrophoresis (pH 8.6)

312

Case # 6 Hemoglobin S-G Philadelphia

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

313

Case # 6 Hemoglobin S-G Philadelphia

Capillary zone electrophoresis

High performance liquid chromatography

314

Case # 6 Hemoglobin S-G Philadelphia

Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major bandHb A

Major bandHb S + G

Minorband(Hb A2)

Minor band cathodal to A2

Acid Agar / Agarose

Major band(Hb A + G)

Major band(Hb S)

CZE Major peak(Hb A)Zone 9

Major peakZone 6

Majorpeak poorly separated from Zone 6 in Zone 5

Minor Hb A2 peak Zone 3

Major Hb G – S hybrid peakZone 2

MinorHb G – A2

peakZone 1

IEF Major Hb G band Anodal to Hb S

Major band(Hb S)

Medium band (Hb S-G hybrid + A2 )

Minor Hb G2

Band

HPLC* Medium peak(Hb A)RT=2.35

Minor peak(Hb A2)RT=3.58

Medium peak(Hb G)RT=4.0

Medium peak(Hb S)RT=4.24

Medium peak(Hb S-G hybrid)RT=4.8

*Note: HPLC retention time (RT) varies with the type of the instrument used and several other factors, e.g. temperature etc.

Agarose gel electrophoresis (pH 8.6) showed three major bands in the

familiar positions of Hb A (≈ 50%), Hb S (≈ 38%), and Hb A2/C (>10%) and a

barely visible minor band slightly cathodal to the carbonic anhydrase position. It

should be emphasized here that Hb A, Hb S, and Hb C cannot all be

315

manufactured in any single person because there are only 2 beta genes and

these hemoglobins represent three different beta compositions. Either this patient

had a transfusion or one of the hemoglobin variants is not a beta chain variant.

The transfusion could be to a patient with Hb S and C or a transfusion using

blood from an Hb A-S heterozygous donor or an Hb A-C heterozygous donor to a

patient who was a heterozygote of the other type. These unlikely scenarios were

all ruled out as the patient received no blood transfusion.

Beta-thalassemia in conjunction with Hb A-S trait can result in an elevated

Hb A2 which migrates with or near Hb C by most of these methods. In S-β-

thalassemia the Hb A2 is rarely higher than 10% so a >10% band is unlikely

Hb A2.

Secondly in S-β-thalassemia patients Hb F concentration is often

increased especially if the patient is thalassemic to the point that the Hb A2 is

very elevated but in this patient the Hb F was normal (≈ 0.2%).

The identity of the small barely visible minor band is the key to the

identification. The most common alpha chain variant is Hb G-Philadelphia which

would present in the Hb S area at 30 to 35%. This alpha chain variant then

competes with the unmodified alpha chains to combine with the beta and delta

chains available. Since the sickle solubility test was positive we know the band in

the position of Hb S is indeed at least partly due to the S beta gene combined

with normal alpha chains. This Hb S beta gene when combined with a modified

316

Hb G-Philadelphia alpha gene creates a new double hemoglobin variant

combination, Hb S-G Philadelphia hybrid which unfortunately migrates with Hb A2

on alkaline, acid or IEF electrophoresis. This explains the elevated Hb A2. If

half of the alpha chains are modified they would be competing also with the

unmodified alpha chains for delta chains. The unmodified alpha chain delta

combination is Hb A2 seen normally and the modified alpha variant delta

combination is new hemoglobin, Hb G2 which migrates close to the carbonic

anhydrase. The number of different hemoglobin molecules created by a Hb S G-

Philadelphia double mutation is 6. The Hb S-G hybdrid migrates with A2 on

acid, alkaline and IEF electrophoresis.

IEF, CZE and HPLC data support the presence of a heterozygous Hb G-

Philadelphia [α68(E17)Asn→Lys] and Hb S in that two distinct, approximately

equal bands or peaks were seen in the position of Hb S and Hb G. IEF indicated

that the Hb G band is closer to the Hb A band (more anodal) than the Hb S. Two

additional bands in the position of Hb A2 and Hb G2 were also detected from IEF

although the low intensity of the Hb G2 band made it difficult to see.

CZE showed six distinct peaks in the following zones with alleged

hemoglobins indicated in parenthesis:

i) Zone 9 (Hb A)ii) Zone 6 (Hb G-Philadelphia)iii) Zone 5 (Hb S)

iv) Zone 3 (Hb A2)v) Zone 2 (Hb S/G hybrid)

vi) Zone 1 (Hb G2)

317

HPLC showed the following major peaks:

a) Hb F (≈ 0.2%)b) Hb A (54%; RT = 2.35)

c) Hb A2 (1.1%, RT= 3.58)d) Hb G (17.8%, RT= 4.0)e) Hb S (19.9%, RT = 4.24)f) Hb S/G hybrid (7%, RT=4.8)

All the data affirm the presence of a double heterozygous presentation of

an abnormal β chain (Hb S) and an abnormal α chain (Hb G-Philadelphia) in

conjunction with normal α and β chains (αA and βA) found in Hb A. The abnormal

chains end up competing with their normal counterparts creating all the possible

combinations listed below.

Hb A (αA2 βA

2) Hb S (αA2 βS

2)

Hb G (αG2 βA

2) Hb S/G (αG2 βS

2)

HB A2 (αA

2 δ2) Hb G2 (αG

2 δ2)

Had this patient been a newborn the situation would further have been

complicated by the addition of 2 new gamma chain containing forms of HbF.

Hb S-G Philadelphia double heterozygous hemoglobinopathies are

essentially healthy and without anemia.

References

1. Kirk CM, Papadea CN, Lazarchik J. Laboratory Recognition of a Rare

Hemoglobinopathy. Hemoglobin SS and SGPhiladelphia Associated with α-Thalassemia -2. Arch Pathol Lab Med 1999; 123: 963-966.

2. Gu LH, Wilson JB, Molchanova TP, McKie KM, Huisman THJ. Three Sickle Cell Anemia Patients each with a Different α Chain Variant. Diagnostic Complications. Hemoglobin 1993; 17(4): 295-301.

318

3. Kutlar F, Kutlar A, Nuguid E, Prachal J, Huisman. Usefulness of HPLC Methodology for the Characterization of Combinations of the Common β-Chain variants Hb S, C, and O-Arab, and the α Chain variant in G-Philadelphia. Hemoglobin 1993; 17(1):, 55-66.

4. LeCrone CN, Jones JA, Detter JC. Hemoglobin G Trait and S Trait in the Same Patient. Hemotology 1983; 49(3): 165-167.

5. Lawrence C, Hirsch RE, Fataliev NA, Patel S, Fabry ME, Nagel RL. Molecular interactions between Hb alpha-G Philadelphia, Hb C, Hb S: phenotypic implications for SC α-G Philadelphia disease. Blood 1997; 90: 2819-2825.

319

Case # 7 Hemoglobin G-Coushatta trait

A 24 year old male resident of Cheyenne River Indian Reservation, South Dakota, USA. No physical abnormality. Blood sent to a reference laboratory for hemoglobin electrophoresis.

Laboratory Data:

Hemoglobin 12.7 13.5-18.5 g/dL

RBC 4.49 4.6-6.2 Mil/mm3

MCV 81 80-100 fL RDW 13.2 11.5-14.5%

Platelet 243 150-400 Th/mm3

Hb A 56.0 94.3-98.5%

Hb A2 ≈2 1.5-3.7% Hb F ≈1 0.0-2.0%

Hb variant 41.0%

(Hemoglobin fractions from HPLC)Peripheral Blood Smear: No abnormality noticed.Sickle cell solubility test for Hb S: NegativeUnstable hemoglobin (isopropanol) test: Negative.

Agarose Gel Electrophoresis (pH 8.6)

320

Case # 7 Hemoglobin G-Coushatta trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

321

Case # 7 Hemoglobin G-Coushatta trait

Capillary zone electrophoresis

High performance liquid chromatography

322

Case # 7 Hemoglobin G-Coushatta trait

Interpretation & Discussion

Summary of ResultsMethod Hb A

areaHb S area Hb

A2/C area

Alk Agarose

Major band(Hb A)

Major band

Minor band

(Hb A2)Acid Agar / Agarose

Major band(Hb A+

Hb A2 + Hb G)

CZE Major peak(Hb A)Zone 9

Major peak( Hb G )Zone 6

Minorpeak

(Hb A2)Zone 3

IEF Major band(Hb A)

MajorHb G band anodal to S

Minor band(Hb A2)

No G2

band was detected

HPLC* Minor peak(Hb F)RT=1.05

Major peak(Hb A)RT=2.5

Major peak(Hb G + Hb A2)RT=3.6

*Note: HPLC retention time (RT) varies with the type of the instrument used and several other factors, e.g. temperature etc.

Agarose gel electrophoresis (pH 8.6) indicated major bands in the position

of Hb A and at the position of Hb S. Besides a minor band at the position of Hb

A2 and carbonic anhydrase band no other band was detected. Citrate agar

electrophoresis (pH 6.2) showed one major band at the position of Hb A, and a

faint band was also detected in the position of Hb F. CZE showed major peaks

323

in Zone 9 (Hb A), and Zone 6 (Hb variant) and a minor peak in Zone 3 (Hb A2).

IEF indicated that the second major band was in the position of Hb G, or

possibly Hb D but not Hb S, however a Hb G2 (α2Gδ2) band was not detected.

HPLC showed two major peaks at the position of Hb A and Hb A2 rather than

one toward the center of the pattern as seen with all the alkaline electrophoretic

separations (pH 8.6). The tentative identification of the Hb variant (41%

concentration from alkaline agarose gel electrophoresis at pH 8.6) was achieved

by eliminating commonly encountered hemoglobin variants (e.g. Hb S, Hb G-

Philadelphia, Hb Lepore, Hb Hasharon, etc) on the basis of the laboratory

results.

Hb S was also ruled out by a normal sickle cell solubility test. The most

commonly noticed Hb G variant (Hb G Philadelphia) is noticed mostly in African

Americans. The presence of this α-chain variant was ruled out because the minor

Hb G2 (α2Gδ2) band was not detected by IEF or by agarose gel electrophoresis

(pH 8.6) and because alpha chain variants are found in a lower percentage than

β-chain variants. Hb Hasharon and Hb Lepore are also ruled out on the basis of

low concentration. Furthermore Hb Lepore produces a thalassemic picture

including microcytosis, and that was not exhibited in this case. Hemoglobin

variant of 41% is extremely high for Hb Hasharon and Hb Lepore. Hemoglobins

D-Los Angeles and Hb G-β trait are closely migrating variants with no clinical

manifestation. Generally speaking they are found in different ethnic groups.

324

The safest interpretation for this case is that this patient has Hb G trait (β-chain variant ) known as Hb G-Coushatta[ β 22 (β4) Glu→Ala (GAA→GCA)] because of the (American Indian) ethnicity. It is emphasized that Hb G-Coushatta is not limited to American Indian tribes, and this hemoglobinopathy also know as Hb G-Saskatoon, Hb G-Taegu, or Hb G- Hsin Chu, has been reported in Chinese, Korean, Japanese, Thai, Turkish, and Algerian nationals and is harmless.

Homozygous Hb G-Coushatta is very rare and exhibits microcytosis.

Recently a compound heterozygote for Hb E and Hb G-Coushatta was reported

in a Thai family by amplification refractory mutation system-polymerase chain

reaction (ARMS-PCR). It may not be worth the cost to further solidify the identity

of the hemoglobin variant in a situation like this where the variant is functioning

normally.

References

1. Worrawut C, Viprakasit V. Further identification of Hb G-Coushatta [β22(β4) Glu→Ala (GAA→GCA)] in Thailand by the polymerase chain reaction-single-strand conformation polymorphism technique and by amplification refractory mutation system-polymerase chain reaction.Hemoglobin 2007; 31(1): 93-99.

2. Ohba Y, Miyaji T, Hirosaki T, Matsuoka M, Koresawa M, Iuchi I. Occurrence of Hemoglobin G Coushatta in Japan. Hemoglobin 1978; 2(5): 437-441.

3. Wong SC, Tesanovic M, Poon M-C. Detection of two abnormal hemoglobins,

Hb Manitoba and Hb G-Coushatta, during analysis of glycohemoglobin (A1c) by high performance liquid chromatography. Clin Chem 1991; 38(8): 1456-1459.

4. Li J, Wilson D, Plonczynski M, Harrell A, Cook CB, Scheer WD, Zeng Y-T, Coleman MB, Steinberg MH. Genetic studies suggest a multicentric origin for Hb G-Coushatta [β22(β4)Glu→Ala]. Hemoglobin 1999; 23(1): 57-67.

5. Boissel JP, Wajcman H, Labie D, Dahmane M, Benabadji M.[Hemoglobin G-Coushatta (beta 22(β4) glu leads to ala) in Algeria: an homozygous case]. Nouv Rev Fr Hematol 1979; 21:225-230.

6. Dincol G, Dincol K, Erdem S. Hb G-Coushatta or alpha 2 beta 22 (β4) Glu→Ala in a Turkish male. Hemoglobin 1989; 13: 75-77.

325

Case # 8 Hemoglobin C trait

A 28 year old African American male. No physical abnormalities. Participated regularly in basketball and never complained about fatigue.

Laboratory Data:

Hemoglobin 14.8 13.5-18.5 g/dL

RBC 4.91 4.6-6.2 Mil/mm3

MCV 77 80-100 fL RDW 15.1 11.5-14.5%

Platelet 248 150-400 Th/mm3

Hb A 58.0 94.3-98.5%

Hb A2 ≈2 1.5-3.7% Hb F ≈1 0.0-2.0%

Hb variant 39.0%(Hemoglobin fractions from HPLC)

Peripheral Blood Smear: 1+ microcytosis and numerous target cells.Sickle cell solubility for Hb S: NegativeUnstable hemoglobin (isopropanol) test: Negative

Agarose Gel Electrophoresis (pH 8.6)

326

Case # 8 Hemoglobin C trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

327

Case # 8 Hemoglobin C trait

Capillary zone electrophoresis

High performance liquid chromatography

328

Case # 8 Hemoglobin C trait

Interpretation & Discussion

Summary of ResultsMethod Hb A

areaHb S area

Hb A2/C area

Alk Agarose

Major band(Hb A)

Major band

Acid Agar /Agarose

Major band(Hb A+

Hb A2)

Major band

CZE Major peak(Hb A)Zone 9

Minor peak(Hb A2)Zone 3

Major peak(Hb C) Zone 2

IEF Major band(Hb A)

Minor band(Hb A2)

Major band cathodal to A2

(Hb C)

HPLC* Major peak(Hb A)RT=2.45

Minor peak(Hb A2)RT=3.6

MajorPeak(Hb C) RT=5.l0

* Note: HPLC retention time (RT) varies with the type of the instrument used and several other factors, e.g. temperature etc.

Agarose gel electrophoresis (pH 8.6) exhibited a major band in the

position of Hb A, and another intense band (≈40%) in the position of Hb C/Hb E/

Hb O-Arab/ Hb A2. The intense band is not due to Hb A2 only in view of

the fact that the concentration of Hb A2 is never > 10%. Citrate agar

electrophoresis (pH 6.2) indicated two bands. One band was in the position of Hb

A and another band in the position of Hb C. Hb E, Hb O-Arab, and Hb C-Harlem 329

are ruled out on the basis of citrate agar electrophoresis (pH 6.2), as none of

these migrate in the position of Hb C by this method. Combination of alkaline and

acid pH electrophoresis suggested that the Hb variant is most likely Hb C. IEF

also indicated two major bands in the position of Hb A and Hb C. CZE also

indicated two major peaks in Zone 9 (Hb A) and Zone 2 (Hb C). HPLC results

were concordant with above stated observations from IEF and CZE, i.e. one

major peak eluted in the position of Hb A (retention time ≈2.45 minutes) and the

second major peak eluted in the C-window (retention time ≈5.10 minutes).

The peripheral blood smear examination (1+ microcytosis and target cells),

negative for sickle cell solubility and hemoglobin instability tests, and the five laboratory

tests led towards the assignment of the Hb variant as Hb C. In order to be Hb C

trait the percentage of Hb C should be less than Hb A, therefore the diagnosis of

Hb C trait was made.

Hb C is a β-chain variant [β6 (A3) Glu→Lys], caused by the substitution of

glutamic acid by lysine in the sixth position. Hb C trait is prevalent in 2-3% in

African Americans, and rarely found in other ethnic groups. Clinically the Hb C

trait phenotype is insignificant.

330

References

1. Bain BJ. Hemoglobin C trait: In: Haemoglobinopathy Diagnosis, Blackwell Publishing, 2nd edition, 2006, pp 192-195.

2. Wajcman H, Moradkhani K. Abnormal haemoglobins: detection & characterization. Indian J Med Res 2011; 134: 538-546

3. Joutovsky A, Nardi M. Hemoglobin C and Hemoglobin O-Arab variants can be diagnosed using the Bio-Rad Variant II High Performance Liquid Chromatography System without further confirmatory tests. Arch Pathol Lab Med 2004; 128: 435-439.

4. Joutovsky A, Hadzi-Nesic J, Nardi MA. HPLC retention time as a diagnostic tool for hemoglobin variants and hemoglobinopathies: A study of 60 000 samples in a clinical diagnostic laboratory. Clin Chem 2004; 50: 1736-1747.

5. Keren DF, Hedstrom D, Gulbranson R, Ou Ching-Nan, Richard B. Comparison of Sebia Capillary Electrophoresis with the Primus High-Pressure Liquid Chromatography in the evaluation of hemoglobinopathies. Am J Clin Pathol 2008; 130: 824-831

331

Case # 9 Hemoglobin C homozygous

African-American male (22 years old) with no physical complaints.

Laboratory Data:

Hemoglobin 12.1 13.5 - 18.5 g/dL

RBC 4.3 4.6 - 6.2 Mil/mm3

MCV 73 80 -100 fL RDW 13.3 11.5 - 14.5%

Platelet 248 150 - 400 Th/mm3

Hb A Not detected 94.3 - 98.5%

Hb A2 ≈2.5 1.5 - 3.7% Hb F ≈1.6 0.0 - 2.0%

Hb variant 95.9% (Hemoglobin fractions from HPLC)

Peripheral Blood Smear: Target cells, spherocytes, and poikilocytosis.Sickle cell Hb S solubility test: NegativeUnstable hemoglobin (isopropanol) test: Negative

Agarose Gel Electrophoresis (pH 8.6)

332

Case # 9 Hemoglobin C homozygous

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

333

Case # 9 Hemoglobin C homozygous

Capillary zone electrophoresis

High performance liquid chromatography

334

Case # 9 Hemoglobin C homozygous

Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band

Acid Agar /Agarose

Major band

CZE Minor peak(Hb A2)Zone 3

Major peak(Hb C) Zone 2

IEF Minor band

Major band cathodal to A2

(Hb C)

HPLC* Minor peak(Hb A2)RT=3.6

Major peak(Hb C) RT=5.06

*Note: HPLC retention time (RT) varies with the type of the instrument used and several other factors, e.g. temperature etc.

Agarose gel electrophoresis (pH 8.6) showed only one intense and major band in

the position of Hb C/E/O, and Hb A was not detected. Citrate agar electrophoresis (pH

6.2) also showed one intense band in the position of Hb C, therefore at the very outset

the presence of Hb E and O were ruled out. It appeared that the solitary band in the Hb C

position is most likely due to the substitution of amino acid “lysine” with glutamic acid at

the sixth position of β-chain [β 6(A3) Glu→Lys]. Hb C has prevalence of 0.017% among

the African-Americans in the United States, but it has been also reported in persons of

Hispanic and Sicilian ancestry.335

Other laboratory tests (CZE, HPLC, and IEF) also indicated the prominent Hb C

band or peak, however contrary to alkaline and acid electrophoresis (see above) minor

bands or peaks due to Hb F (≈ 1.6%) and Hb A2 (≈ 2.5%) were also detected. Absence

of Hb A by all the five methods in this person suggested either homozygous Hb C or Hb

C/β0-thalassemia and ruled out Hb C/β+-thalassemia (Case # 19).

The clear distinction between homozygous Hb C and Hb C/β0-thalassemia

(double heterozygous state for both Hb C and β0-thalassemia) is problematic,

because the clinical features are similar in both cases. Careful evaluation of

peripheral blood smear, CBC, anemia status, quantitative values of Hb F and Hb

A2, and evaluation of hemoglobinopathy in the biological parents are helpful for

the exactness of the diagnosis.

Fairhurst and Casella reported a diagnosis of homozygous Hb C disease

in a Ghanian child [N Engl J Med 2004; 350(26): e24], with hemoglobin (9.0

g/dL),HCT (24.3), MCV (53.8), RDW (28.8), and an uncorrected reticulocyte

count of 1.6%. The peripheral blood smear (Figure 1) indicated characteristic

features of homozygous Hb C: target cells (arrows), microspherocytes

(arrowheads), rod-shaped cells containing hemoglobin C crystals (asterisk),

anisocytosis, and poikilocytosis. Schwab and Abelson [N Engl J Med 2004;

351(15): 1577] questioned the diagnosis of homozygous Hb C on the basis of

336

extremely low MCV and the clinical status of the child, and suggested the

diagnosis of Hb C/β0-thalassemia.

Figure 1. Peripheral blood smear of the Ghanaian child (adopted with the permission of the N Engl J Med)

The following characteristics are helpful in the differential diagnosis

between the two possibilities:

Test Homozygous Hb C Hb C/β0-thalassemia

Hb A2 3.2 – 3.9% Elevated in most cases

Hb F 0.8 – 1.9% 3 – 10% (generally > 5%)

MCV 68 - 76 55 - 70337

On the basis of Hb A2 (≈ 2.5%), Hb F (≈ 1.6), MCV (73), mild anemia, a

tentative diagnosis of homozygous Hb C is reasonable, however for confirmation,

additional tests in the biological parents are mandatory.

Persons with homozygous Hb C rarely have clinical symptoms and live a

normal life. Symptoms that may develop in these persons include:

● Reduced red blood cell counts during infection or illness● jaundice● Increased risk for gallstones● Enlarged spleen● Episodes of pain

● Increased risk for infection

Hemoglobin C is known to protect individuals against clinical Plasmodium falciparum malaria.

References1. Bunn HF, Forget BG, Hemoglobin: molecular, genetic and clinical aspects. 1st

edition, Philadelphia, PA: WB Saunders Co; 1986: 421-425.2. Nagel RL, Steinberg MH. Hb S/C disease and Hb C disorders. In: Steinberg

MH, Forget BG, Higgs DR, Nagle RL, eds. Disorders of Hemoglobin: Genetics, Pathophysiology and Clinical Management. Cambridge, England: Cambridge University Press; 2001: 756-785.

3. Fairhurst RM, Casella JF. Homozygous hemoglobin C disease. N Engl J Med 2004; 350: e24 (Web only). (Available at www.nejm.org/cgi/content/full/350/26/e24).

4. Schwab JG, Abelson HT. Hemoglobin C. N Engl J Med 2004; 351(15): 1577.5. Weatherall DJ, Clegg JB. The thalassemia syndrome, 4th edition, Oxford,

England: Blackwell Science, 2001: 415-419.6. Modiano D, Luoni G, Sirima BS, et al. Hemoglobin C protects against

clinical Plasmodium falciparum malaria. Nature 2001; 414 (6861): 305-8. [Medline].

7. Rihet P, Flori L, Tall F. Hemoglobin C is associated with reduced Plasmodium falciparum parasitemia and low risk of mild malaria. Hum Mol Genet 2004; 13(1): 1-6.8. Hoyer JD, Kroft SH. Color Atlas of Hemoglobin Disorders. College of American Pathology 2003. Case # 8 (pp 45), Case # 15 (pp 75), Case # 29 (pp 135), Case # 30 (pp 139).

338

Case # 10 Hemoglobin C with hereditary persistence of fetal hemoglobin (HPFH)

A 23 years old white female presented to the Emergency Department of the hospital (2011) complaining of pelvic pain. She was found to have a ruptured right hemorrhagic ovarian cyst which was suspected on CT and ultrasound and then confirmed by laparoscopy. No blood transfusion was executed.

Laboratory Data:

Hemoglobin 11.9 12.0 -16.0 g/dL

RBC 4.8 4.0 - 5.5 Mil/mm3

MCV 74 79 - 98 fL RDW 20.8 11.5 -14.5%

Hb A Not detected 94.3 - 98.5%

Hb A2 ≈2.2 1.5 - 3.7% Hb F ≈29.4 0.0 - 2.0%

Hb variant 68.4% (Hemoglobin fractions from HPLC)Peripheral Blood Smear: Abundant target cellsSickle cell solubility test for hemoglobin S: NegativeFlow cytometry (monoclonal antibody for Hb F) showed a homogeneous distribution of Hb F.

Agarose Gel Electrophoresis (pH 8.6)

339

Case # 10 Hemoglobin C with hereditary persistence of fetal hemoglobin (HPFH)

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

340

Case # 10 Hemoglobin C with hereditary persistence of fetal hemoglobin (HPFH)

Capillary zone electrophoresis

High performance liquid chromatography

341

Case # 10 Hemoglobin C with hereditary persistence of fetal hemoglobin (HPFH)

Interpretation & Discussion

Note: HPLC and hemoglobin electrophoresis tests were performed at three independent laboratories, and all the results were concordant.

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major bandin Hb Farea

Major band

Acid Agar /Agarose

MajorBand in Hb F area

Major band

CZE Major peak Hb FZone 7

Very minor peak(Hb A2)Zone 3

Major peak(Hb C) Zone 2

IEF Major bandinHb F area

Minor band(Hb A2)

Major band cathodal to A2

(Hb C)

HPLC* Major peak(Hb F)RT=1.15

Very minor peak(Hb A2)RT=3.6

Major peak(Hb C) RT=5.14

*Note: HPLC retention time (RT) varies with the type of the instrument and several other factors, e.g. temperature etc.

Agarose gel electrophoresis (pH 8.6) indicated the absence of Hb A and

the presence of two major bands. One major band was detected in the position

of Hb F (≈ 29%) and another major band (≈ 68%) was detected in the position of

Hb C/E/O. Hb E and O were ruled out on the basis of citrate agar electrophoresis

342

(pH 6.2), as only two major bands were detected in the position of Hb C and

Hb F. IEF, CZE, and HPLC also confirmed the presence of only two major

hemoglobins (Hb C and Hb F) in this patient.

This suggested two possibilities, a) heterozygosity for Hb C or b)

heterozygosity for a deletional form of hereditary persistence of fetal hemoglobin

(HPFH). The presence of Hb C > 50% also suggested the presence of HPFH.

Hb C with hereditary persistence of fetal hemoglobin is the diagnosis of

this patient. Generally speaking homozygous Hb C disease (Case # 9) is rare

and is associated with abundant target cells, microcytosis, reticulocytosis, and

minimal hemolytic disease. Contrary to this, Hb C with HPFH is clinically similar

to Hb C trait (Case # 8).

References

1. Bain BJ. Hereditary persistence of fetal hemoglobin and other inherited causes of an increased proportion of hemoglobin F: In: Hemoglobinopathy Diagnosis, Blackwell Publishing, 2nd edition, 2006, pp 119-127.

2. Bollekens JA, Forget BG. δβ thalassemia and hereditary persistence of fetal hemoglobin. Hematol Oncol Clin North Am. 1991; 5: 399-422.

3. Hoyer JD, Penz CS, Fairbanks VF, et al. Flow cytometric measurement of hemoglobin F in RBCs: diagnostic usefulness in the distinction of hereditary persistence of fetal hemoglobin (HPFH) and hemoglobin S-HPFH from other conditions with elevated levels of hemoglobin F. Am J Clin Pathol 2002; 117: 857-863.

4. Weatherall DJ, Legg JB. Hereditary persistence of fetal hemoglobin. In: The thalassemia Syndromes. 4th ed. Oxford: Blackwell Science, 2001: 450-484.

343

5. Wood WB. Hereditary persistence of fetal hemoglobin and δβ thalassemia. In: Steinberg MH, Forget BG, Higgs DR, Nagel RL. Disorders of Hemoglobin: Genetics, Pathophysiology, and Clinical Management, Cambridge, England: Cambridge University Press; 2001: 356-388.

6. Pissard S, M’rad A, Beuzard Y, Romeo PH. A new type of hereditary persistence of fetal hemoglobin (HPFH): HPFH Tunisia beta + (+C-200) G gamma. Br J Haematol 1996; 95(1): 67-72.

7. Martin AW, Lippmann SB, Keeling MM, Lynch JA, Martinez M. Hemoglobin C in association with hereditary persistence of fetal hemoglobin. Postgrad Med 1987; 81(8): 133-37.

Case # 11 Hemoglobin S-C disease344

A 22 year old African American male, who was working at the Chrysler Stamping plant, complained of headache and difficulty in breathing. His supervisor suspected carbon monoxide poisoning and sent him to the Emergency Department.

Laboratory Data:

Hemoglobin 10.8 13.5 - 18.5 g/dL

RBC 3.6 4.6 - 6.2 Mil/mm3

MCV 90.1 80 - 100 fL MCH 30.0 27 - 34 pg

Hb A2 2.4 1.5 - 3.7% Hb F 1.8 0.0 - 2.0%

Hb Variant-1 49.0%Hb Variant-2 46.8%(Hemoglobin fractions from HPLC)

Peripheral Blood Smear: Target cells present. Rare spherocyte seen. Slight anisocytosis and polychromasia.

Sickle cell solubility test for Hb S: Positive.

Agarose Gel Electrophoresis (pH 8.6)

Case # 11 Hemoglobin S-C disease

345

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

346

Case # 11 Hemoglobin S-C disease

Capillary zone electrophoresis

High performance liquid chromatography

Case # 11 Hemoglobin S-C disease

347

Interpretation & Discussion

Summary of Results

Method

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band

Major band

Acid Agar/ Agarose

Major band

Major band

CZE Major peakHb S(Zone 5)

Minor peak(Hb A2) Zone 3

Major peak(Hb C) Zone 2

IEF Major band(Hb S)

Very minor band(Hb A2)

Major band(Hb C)slightly cathodal to A2

HPLC* Major peak(Hb S)RT=4.37

Veryminor peak(Hb A2)RT=3.6

Major peak(Hb C)RT=5.12

*Note: HPLC retention time (RT) varies with the type of the instrument used and several other factors, e.g. temperature etc.

Agarose gel electrophoresis (pH 8.6) indicated the absence of Hb A, but

two intense bands were detected in the position of Hb S and Hb C/E/O/A2.

Citrate agar electrophoresis (pH 6.2) also showed the absence of a band in the

usual position of Hb A, here again two bands were detected in the position of Hb

S and Hb C. The citrate agar electrophoresis (pH 6.2) ruled out the possibility of

Hb S-E since Hb E migrates with Hb A in this system. The possibility of Hb

348

S-O Arab was also ruled out, however with less certainty since Hb O-Arab

migrates between Hb S and Hb A in this system. Since two separate major

bands in the position of Hb S and Hb C were detected, Hb C-Harlem (also called

Hb C-Georgetown) was also ruled out, because this variant migrates with Hb S

upon citrate agar electrophoresis (pH 6.2).

IEF confirmed the above results (absence of Hb A, and two major bands

in the position of Hb S and Hb C). CZE has the advantage of fewer variants with

mobility similar to Hb S, HbC, and Hb E. Since Hb A was absent in the patient,

no zones were detected upon CZE. Therefore, the patient’s blood specimen was

mixed (1:1) with a normal blood specimen, and two major peaks in the patient

were present in Zone 2 (Hb C), and Zone 5 (Hb S). Similarly HPLC showed two

major peaks (besides very minor peaks for Hb F and Hb A2) in the S window

(RT= 4.37 minutes) and C window (RT= 5.12 minutes).

All the above stated tests support the diagnosis of Hb S-C disease in this

patient.

Hb S-C disease is observed in approximately 0.13% of African Americans,

which is approximately half of the homozygous Hb S disease. Most clinical

manifestations of homozygous Hb S disease are also seen in Hb S-C disease,

but in a somewhat milder form.

A characteristic of Hb S-C disease (first pointed out by Professor Virgil F.

Fairbanks, MD, Mayo Clinic, Rochester, MN) is that the concentration

349

of Hb S is always slightly greater than Hb C. In addition, the cellular dehydration

that occurs as a consequence of the presence of Hb C promotes the distortion of

the shape of the red blood cells (Professor James D. Hoyer, MD, Mayo

Clinic, Rochester, MN).

Hemoglobin C-Harlem (also called Hb C-Georgetown) is a rare double β-

chain mutation hemoglobin (β6(A3) Glu→Val; β73(E73) Asp→Asn) and patients

heterozygous for only Hb C-Harlem are asymptomatic. Compound heterozygous

state (e.g. Hb S-C-Harlem) exhibits sickling, and also clinical severity.

The diagnosis of Hb S-C disease and homozygous Hb S disease is

usually straight forward in the appropriate clinical context (e.g. African American

patient).The diagnosis of Hb S-O Arab disease, Hb S-C-Harlem disease requires

the evaluation of a large number of laboratory tests in conjunction with the clinical

status of the patient. Special attention is required if the patient has been recently

transfused.

References:

1. Lionett F, Hammoudi N, Stojanovic KS, Avellino V, Grateau G, Girot R, Haymann J-P. Hemoglobin SC disease complications: a clinical study of 179 cases. Haematologica 2012; 97(8): 1136-1141.

2. O’Keefe EK, Rhodes MM, Woodworth A. A patient with a Previous Diagnosis of Hemoglobin S/C Disease with an unusually Severe Disease Course. Clin Chem 2008; 55(6): 1228-1231.

3. Bain BJ. Sickle cell/hemoglobin C disease: In: Hemoglobinopathy Diagnosis, Blackwell Publishing, 2nd edition, 2006, pp 164-170.

350

4. Joutovsky A, Nardi M. Hemoglobin C and Hemoglobin O-Arab variants can be diagnosed using the Bio-Rad Variant II High-Performance Liquid Chromatography System without further confirmatory tests. Arch Pathol Lab Med 2004; 128: 435-439.5. Nagel RL, Fabry ME, Steinberg MH. The paradox of hemoglobin SC disease. Blood Reviews 2003; 17: 167-178.6. Powars DR. Hiti A, Ramicone E, Johnson C, Chan L. Outcome in Hemoglobin SC disease: A four-decade observational study of clinical, hematologic, and genetic factors. Am J Hematol 2002; 70: 206-215.7. Koduri PR, Agbemadzo B, Nathan S. Hemoglobin S-C disease revisited: Clinical study of 106 adults. Am J Hematol 2001; 68: 298- 300.8. Nagel RL, Steinberg MH, Hb S/C disease and Hb C disorders. In: Steinberg MH, Forget BG, Higgs DR, Nagle RL. Disorders of Hemoglobin: Genetics, Pathophysiology and Clinical Management. Cambridge, England: Cambridge University Press; 2001; 756-785.9. Bunn HF, Forget BG. Hemoglobin: Molecular, Genetic and Clinical

Aspects. 1st ed. Philadelphia, PA: WB Saunders Co; 1986; 533-536.10. Bunn HF, Noguchi CT, Hofrichter J, Schechter GP, Schechter AN, Eaton WA. Molecular and cellular pathogenesis of hemoglobin S/C disease. Proc Natl Acad Sci USA. 1982; 79: 7527-7531.

351

Case # 12 Hemoglobin D-Los Angeles (D-Punjab) trait

First year resident (male, 26 years old) in the Department of Surgery. Originally from India (State of Punjab). Healthy and physically robust.

Laboratory Data:

Hemoglobin 14.7 13.5 - 18.5 g/dL

RBC 4.9 4.6 - 6.2 Mil/mm3

MCV 82 80 - 100 fL MCH 30.2 27 - 34 pg

Platelet 239 150 - 400 Th/mm3

Hb A 58.0 94.3 - 98.5%

Hb A2 1.5 1.5 - 3.7% Hb F 0.3 0.0 - 2.0%

Hb Variant 40.2(Hemoglobin fractions from HPLC)

Peripheral Blood Smear: No abnormalityUnstable hemoglobin (isopropanol) test: NegativeSickle cell solubility test for Hb S: Negative

Agarose Gel Electrophoresis (pH 8.6)

352

Case # 12 Hemoglobin D-Los Angeles (D-Punjab) trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

353

Case # 12 Hemoglobin D-Los Angeles (D-Punjab) trait

Capillary zone electrophoresis

High performance liquid chromatography

354

Case # 12 Hemoglobin D-Los Angeles (D-Punjab) trait

Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band(Hb A)

Major band

Minor band

Acid Agar /Agarose

Major band

CZE Major peak(Hb A)Zone 9

Major peak(Hb D) Zone 6

Minor peak(Hb A2) Zone 3

IEF Major band(Hb A)

Major Hb D band slightly anodic to S

Minor band(Hb A2)

HPLC* Major peak(Hb A)RT=2.42

Minor peak(Hb A2)RT=3.6

Major peak(Hb D)RT=3.99

*Note: HPLC retention time (RT) varies with the type of the instrument and several other factors, e.g. temperature etc.

Agarose gel electrophoresis (pH 8.6) showed two major bands in

approximately equal intensity at the positions of Hb A and Hb S. Citrate agar

electrophoresis (pH 6.2) showed only one major band (≈ 100%) and barely

visible staining in the Hb F position. Several hemoglobin variants migrate in the

position of Hb S upon agarose gel electrophoresis (pH 8.6), and among them,

the most frequently noticed are Hb G, Hb D, and very rarely Hb Korle-Bu (G-

Acra).

355

Hb S was easily ruled out on the basis of the negative sickle solubility test,

Hb G (α-chain variant) was ruled out on the basis of the absence of Hb G2 band

(α2Gδ2) and the observation that the percentage of the abnormal variant

approaches 50%. α-chain variants percentages do not run this high without other

genetic complications (see Case# 5).

The differentiation between Hb D-Los Angeles and other kinds of

heterozygous D hemoglobins (which are also β-chain variants) or heterozygous

Hb Korle-Bu (G-Accra) on the basis of electrophoretic tests (alkaline, acid, IEF,

CZE) was not possible with certainty due to identical mobilities. HPLC

differentiated Hb D-Los Angeles from Hb Korle-Bu. We have summarized the

HPLC retention times from three separate studies for Hb D-Los Angeles and Hb

Korle-Bu:

Nardi et al* Nardi-2013** Hoyer et alπ

Hb Korle-Bu 3.92 + 0.050 3.9+ 0.034 3.88+ 0.08

Hb D- Los Angeles 4.18+ 0.007 4.11+ 0.078 4.08+ 0.08

* Bio-Rad Variant II (Clin Chem 2004; 50: 1736-1747)

** Bio-Rad Variant II (personal communication)

Π Bio-Rad Variant Classic (Intl J Lab Hematol 2012; 34: 594-604)

It is the observation of Professor Michael A. Nardi (personal communication) that

Hb Korle-Bu rarely separates from Hb A2 (due to the closeness of their retention times),

while Hb D-Los Angeles always separates from Hb A2.

356

In view of the laboratory tests, the diagnosis of Hb D-Los Angeles trait

was most likely. Since the patient had a clinically silent and harmless

condition, it was not advised to perform globin chain analysis and DNA studies.

Hb D-Los Angeles results from a substitution of glutamic acid by glutamine

on position 121 of the β-chain [β121(GH4)Glu→Gln.GAA>CAA] and is a

harmless condition. Hb D-Los Angeles has been found double heterozygotes for

other variants (e.g., Hb S, Hb C, Hb E). Hb D-Los Angeles in combination with

Hb S causes a severe sickling disorder (Case # 13).

Homozygous Hb D-Los Angeles patients exhibit normal hematologic

indices (e.g. hemoglobin, RBC), and no evidence of hemolysis. However,

patients with Homozygous Hb D-Los Angeles and βo-Thalassemia do have a

mild anemia and mild hemolysis.

References

1. Pandey S, Mishra RM, Pandey S, Saxena R. Homozygous hemoglobin D with alpha thalassemia: case report. Open Journal of Hematology 2011; 2: 1-4.

2. Basmanj MT, Karimpoor M, Amirian A, Jafrinejad M, Katouzian L, Valei A, Bayat F, Kordafshari A, Zeinali S. Co-inheritance of Hemoglobin D and β-thalassemia Traits in Three Iranian Families: Clinical Relevance.Archives of Iranian Medicine 2011;14(1): 61-63.

3. Srinivas U, Pati HP, Saxena R. Hemoglobin D-Punjab syndromes in India: a single center experience on cation-exchange high performance liquid chromatography. Hematology 2010; 15 (3): 178-181.

4. Yavarian M, Karimi M, Paran F, Neven C, Harteveld CL, Giordano PC. Multi Centric Origin of Hemoglobin D-Punjab [β121(GH4)Glu→GLN, GAA>CAA]. Hemoglobin 2005; 29 (4): 307-310.

357

5. Atalay EO, Koyuncu H, Turgut B, Atalay A, Yildiz S, Bahadir A, Koseler A. High incidence of Hb D-Los Angeles [β121(GH4)Glu→Gln] in Denizli Province, Aegean Region of Turkey. Hemoglobin 2005; 29(4): 307-310.

6. Owaidah TM, Al-Saleh MM, Al-Hellani AM. Hemoglobin D/β-thallasemia and β-thalassemia major in a Saudi family. Saudi Med J 2005; 26(4): 674-677.

7. Thornburg CD, Zimmerman SA, Schultz WH, Ware RE. An infant with Homozygous D-Iran. Journal of Pediatric Hematology/Oncology 2001; 23(1): 67-68.

8. El-Kalla S, Mathews AR. Hb D-Punjab in the United Arab Emirates. Hemoglobin 1997; 21(4): 369-375.

9. Zago MA, Costa FF. Hb D-Los Angeles in Brazil: Simple Heterozygotes and Associations with β-Thalassemia and with Hb S. Hemoglobin 1988; 12(4): 399-403.

10. Harano T, Harano K, Ueda S, Nakaya K. Hb D-Los Angeles [β121 Glu→Gln] in Japan. Hemoglobin 1987; 11(2): 177-180.

11. Li HJ, Liu DX, Li L, Liu ZG, Lo SL, Zhao J, Han XP, Yu WZ. A Note About The Incidence And Origin of Hb D-Punjab in Xinjiang, People’s Republic of China. Hemoglobin 1986; 10(6): 667-671.

12. Husquinet H, Parent MT, Galacteros F. Hemoglobin D-Los Angeles [β121 (GH4)Glu→Gln] in the Province of Liege, Belgium. Hemoglobin 1986; 10(6): 587-592.

13. Baiget M, del Rio E, Gimferrer E. Hemoglobin D-Punjab (β121 Glu→Gln) in a Spanish Family. Hemoglobin 1982; 6(2):193-198.

14. Ramot B, Rotem J, Rahbar S, Jacobs AS, Udem L, Ranney HM. Hemoglobin D-Punjab in a Bulgarian Jewish Family. Israel J. Med. Sci. 1969; 5(5):1066-1070.

358

Case # 13 Hemoglobin S-D disease

17 years old male patient. No other information provided due to the privacy requested by the patient. No record of blood transfusion during the past three months.

Laboratory Data:

Hemoglobin 10.2 13.5 - 18.5 g/dL

RBC 3.2 4.6 - 6.2 Mil/mm3

MCV 90.7 80 - 100 fL MCH 30.9 27 - 34 pg

Platelet 229 150 - 400 Th/mm3

Hb A (mostly Hb A1c) ≈6.0

Hb A2 2.7 1.5 - 3.7% Hb F 2.5 0.0 - 2.0%

Hb Variant-1 49.3Hb Variant-2 39.5(Hemoglobin fractions from HPLC)

Peripheral Blood Smear: Moderate sickle cells. Target cells and polychromasia.Sickle cell solubility test for Hb S: Positive.Hemoglobin instability (isopropanol) test: Negative.

Agarose Gel Electrophoresis (pH 8.6)

359

Case # 13 Hemoglobin S-D disease

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

360

Case # 13 Hemoglobin S-D disease

Capillary zone electrophoresis

High performance liquid chromatography

361

Case # 13 Hemoglobin S-D disease

Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band

Minor band

Acid Agar /Agarose

Major band(Hb D)

Major band(Hb S)

CZE Major*

peak (Hb D) Zone 6

Major*

peak(Hb S) Zone 5

Minor peak

(Hb A2) Zone 3

IEF Major band(Hb S)

Major band anodal to S position(Hb D)

Minor band

(Hb A2)

HPLCп Minor peak(Hb F)RT=1.05

Major peak (Hb D)RT=4.0

Major peak(Hb S)RT=4.3

Minor peak

(Hb A2)RT=3.6

* Overlap of the two peaks (Zone 5-6) due to approximately equal and higher concentration of Hb S and Hb D-Los Angeles.

П Note: HPLC retention time (RT) varies with the type of instrument used and several other factors, e.g. temperature etc.

Agarose gel electrophoresis (pH 8.6) showed a major and very intense

band in the position of Hb S. Another band of faint intensity was detected in

the Hb F position. A faint band in the position of Hb C/E/O/A2 was also noticed.

362

No band was detected in the position of Hb A. Citrate agar electrophoresis (pH

6.2) presented with two major bands in approximately equal intensity in the

position of Hb A and Hb S. A faint band was also detected in the position of

Hb F. Since the sickle cell solubility test was positive, therefore the band in the

Hb S position upon citrate agar electrophoresis (pH 6.2) suggested the presence

of a β-chain variant (Hb S). The migration of a band of equal intensity in the

position of Hb A upon citrate agar electrophoresis (pH 6.2) suggested the

presence another hemoglobin variant (since Hb A was absent upon alkaline

agarose gel electrophoresis). Several hemoglobin variants (e.g. Hb G, Hb D-Los

Angeles, Hb Korle-Bu, etc) exhibit this kind of migration pattern, therefore

assignment of this hemoglobin variant was deferred.

IEF confirmed the presence of Hb S, however another band in between

the customary position of Hb G and Hb S was also prominent. The presence of

Hb G from IEF was ruled out positively as no band in the position of Hb G2

(α2Gδ2) was detected. Since Hb D-Los Angeles and Hb Korle-Bu have similar

mobilities upon IEF, therefore a distinction could not be made between these two

possibilities. HPLC was helpful in differentiating between the Hb D-Los Angeles

and Hb Korle-Bu variants, as Hb D-Los Angeles has a longer retention time (4.0

minutes) as compared to Hb Korle-Bu (3.75 minutes). Hb S eluted at retention

time of 4.3 minutes, thus the two major bands in this case were separated nicely

upon HPLC.

363

In a heterozygous situation upon CZE, Hb S migrates in Zone 5 and Hb D-

Los Angeles in Zone 6. In this case since the concentration of the two variants is

intense (≈ 40-49% from HPLC), thus clearly separated peaks were not detected

but the scan positively showed two overlapping peaks in the position of Zones 5-

6. Distinct peaks for Hb F and Hb A2 from CZE were noticed in Zone 7 and and

Zone 3 respectively.

The specimens of father and mother of this person were not available for

additional studies. Furthermore globin chain and DNA studies were also not done

on the blood of this person. On the basis of the available laboratory data a

tentative diagnosis of a double heterozygosity of Hb S [β6 (A3) Glu→Val] and

Hb D-Los Angeles [β121(GH4)Glu→Gln.GAA>CAA] was advised to the

physician.

Hb D-Los Angeles in both the heterozygous (Case # 12) and homozygous

state is clinically silent and harmless. However patients with homozygous Hb D-

Los Angeles and βo-thalassemia do have mild anemia and also exhibit mild

hemolysis. Hb D-Los Angeles is not itself a sickling hemoglobin, but compound

heterozygosity (Hb S + Hb D-Los Angeles) produces a severe sickle cell anemia

because Hb D-Los Angeles enhances Hb S polymerization by forming an

additional contact stabilizing the Hb S polymer.

364

References

1. Adekile A, Mullah-Ali A, Akar NA. Does Elevated Hemoglobin F Modulate the Phenotype of Hb SD-Los Angeles?. Acta Haematol 2010; 123: 135-139.

2. Isoa EM. Current Trends in the Management of Sickle Cell Disease: An Overview. Benin J Postgraduate Med 2009; 11:50-73.

3. Mukherjee MB, Surve RR, Gangakhedkar RR, Mohanty D, Colah RB. Hemoglobin sickle D Punjab-a case report. Indian J Hum Genetics 2005; 11(3): 154-155.

4. Jiskoot PMC, Halsey C, Rivers R, Bain BJ, Wilkins BS. Unusual splenic sinusoidal iron overlaod in sickle cell/haemoglobin D-Punjab disease. J Clin Pathol 2004; 57: 539-540.

5. Athanasiou-Metaxa M, Economou M, Tstra I, Pratsidou P, Tsantali C. Co-Inheritance of Hemoglobin D-Punjab and Hemoglobin S: Case Report. J Ped Hematology/Oncology 2002; 24(5): 421.

6. Perea FJ, Casas-Castaneda M, Villalobos-Arambula AR, Barajas H, Alverez F, Camacho A, Hermosillo RM, Ibrarra B. Hb D-Los Angeles Associated with Hb S or β-Thalassemia in Four Mexican Mestizo Families. Hemoglobin 1999; 23(3): 231-237.

7. Dash S. Haemoglobin S-D Disease in a Bahraini Child. Bahrain Med Bulletin 1995; 17(4): 154-56.

8. Samperi P, Dibenedetto SP, Cataldo AD, Mancuso GR, Schiliro G. Unusual Sickle Cell Disease observed for the First Time in Italy. Haematologica 1990; 75: 464-66.

9. McCurdy PR, Lorkin PA, Casey R, Lehmann H, Uddin DE, Dickson LG. Hemoglobin S-G (S-D) Syndrome. The American J of Med 1974; 57: 665-670.

10. Barton LL, Stark AR, Zarkowsky HS,.Hemoglobin S-D disease in a Negro Child. The Journal of Pediatrics 1973; 82(1): 164-165.

11. Ozsoylu S. Haemoglobin S-D Disease in a Turkish Family. Scand. J Haematol 1969; 6: 10-14.

12. Cawein MJ, Lappat EJ, Brangle RW, Farley CH. Hemoglobin S-D Disease. Annals of Internal Medicine 1966; 64(1): 62-70.

365

Case # 14 Hemoglobin E and Associated Disorders

The contents of this section are presented from Hoyer JD, Kroft SH, eds. Color Atlas of Hemoglobin Disorders: A Compendium Based on Proficiency Testing. Northfield, IL: College of American Pathologists; 2003 (Reproduced with Permission).

In addition to Hb E, several other disorders of hemoglobin are prevalent in the

Southeast Asian population. Therefore, Hb E may be encountered in conjunction with

another abnormality. A description of the various Hb E-associated disorders is provided

below.

1. Hb E trait. A harmless condition characterized by mild microcytosis and

often by erythrocytosis. No icterus, no splenomegaly, no anemia. MCV

about 75 fL (adult). Electrophoresis: Hb E 30-35%, Hb A 65-70%, Hb F

<2%.

2. Homozygous Hb E. A harmless condition characterized only by mild

microcytosis and erythrocytosis. No icterus, no splenomegaly, no anemia

(hemoglobin concentration >11 g/dL in females, >14 g/dL in males). MCV

about 67 fL (adults). Electrophoresis: Hb E about 99%, the rest Hb F.

3. Hb E trait/α-thalassemia. This combination results in microcytosis, but

usually no other adverse effects (no anemia, no splenomegaly, no

icterus). Serum ferritin assay is required to differentiate this condition from

Hb E trait/iron deficiency. Electrophoresis (1 α gene deletion): Hb E 25-

30%; remainder Hb A; Hb F normal. Electrophoresis (2 α gene deletion):

Hb E 20-25%; remainder Hb A; Hb F normal. Since Hb E and Hb A2 co-

migrate in all electrophoresis media and co-elute from chromatography

366

columns, a common laboratory error is to ascribe the electrophoresis

findings to β-thalassemia trait. However, in the latter, Hb A2 is always

<10%.

4. Hb E trait/Hb H disease. In this disorder, Hb E trait is inherited in

conjunction with a three locus α gene deletion. This is a moderately

severe thalassemic disorder with features identical to Hb H disease.

However, electrophoresis does not reveal Hb H. Instead, Hb E represents

about 10-15% of hemoglobin; most of the remainder is Hb A. This

paradox is due to reduced total synthesis of β globin chains. As a result,

not enough surplus β chains are present to form β tetramers (Hb H).

Instead, Hb Bart's is present. (Thus, this condition has also been called

"Hb A + E + Bart's Disease").

5. Homozygous Hb E/Hb H disease. This disorder has the same features as

Hb H disease. However, electrophoresis reveals mostly Hb E (about

95%) and a small proportion of Hb F. It is believed that in this condition,

the βE tetramers co-migrate with Hb E in all electrophoresis media.

6. Hb E trait/α-thalassemia/Hb Constant Spring. Features are the same as 4

and 5 above, except for faint additional hemoglobin bands (as many as

five) between the positions of Hb E and the site of application. These

additional faint bands represent Hb Constant Spring.

7. Hb E trait/iron deficiency. A benign condition characterized by

microcytosis, often erythrocytosis, and anemia. The anemia is due to iron

367

deficiency and thus may be minimal to severe. There is no icterus and no

splenomegaly. Electrophoresis shows the same pattern as Hb E trait/α-

thalassemia. The combination should be suspected in an anemic patient

with an "Hb A2" concentration of 10-20%. The diagnosis is confirmed by a

serum ferritin assay. Following treatment, repeat electrophoresis will

show Hb E representing 30-35% of total (unless the patient also has Hb E

trait/α-thalassemia).

8. Hb E/β°-thalassemia. This is a serious thalassemic disorder due to

compound heterozygosity for both Hb E trait and β-thalassemia trait.

Characteristics are severe anemia, icterus, marked splenomegaly, and

microcytosis. Affected children suffer all the problems of β-thalassemia

major. Most require frequent transfusions and should also receive iron

chelation therapy. This is the most common severe thalassemia of

Southeast Asians. Neurologic manifestations are often reported that are

due to brain or spinal cord compression by extramedullary hematopoietic

tumors, which may cause paraplegia. The tumors respond to

radiotherapy. Electrophoresis: Hb E is 40-90% total; the rest is Hb F.

(Note: Because these patients usually require transfusion, Hb A may be

present from donor blood). It should be pointed out that it is not necessary

to document elevated Hb A2 levels to establish a diagnosis of Hb E/β°-

thalassemia. The diagnosis is easily established on the basis of an Hb E

368

concentration >40% with the remainder representing Hb F (usually 30-

60%) and an absence of Hb A.

9. Hb E/β°-thalassemia, post-splenectomy. Same condition as # 8 (see

above), but often confusing in laboratories. Splenectomy is a common

treatment in Hb E/β°-thalassemia and is reputed to be beneficial for those

with severe anemia. The post-splenectomy blood picture is characterized

by marked normoblastemia and a positive solubility test for sickling

hemoglobin. The latter is due to the large number of normoblast nuclei

causing strong persistent turbidity. Pulmonary artery occlusion is a

common complication in splenectomized patients with Hb E/β°-

thalassemia. Prophylactic therapy with daily doses of aspirin or

dipyridamole is indicated for all patients with this disorder who have been

splenectomized.

Note: It will not be out of place to mention here that another disorder “Hb S-E heterozygous” has been also diagnosed in persons of Southeast Asian origin (Case # 14 C).

References

1. Fucharoen S, Weatherall DJ. The Hemoglobin E Thalassemias. Cold Spring Harb Perspect Med 2012; 2: a011734.

2. Sae-ung N, Srivorakun H, Fucharoen G, Yamsri S, Sanchaisuriya K,

Fucharoen S. Phenotypic expression of hemoglobins A2 , E and F in various hemoglobin E related disorders. Blood Cells, Molecules, and Diseases 2012; 48: 11-16.

3. Tatu T, Kasinrerk W. A novel test tube method of screening for hemoglobin E. Int. Lab. Hem 2012; 34: 59-64.

4. Moiz B, Hashmi MR, Nasir A, Rashid A, Moatter T. Hemoglobin E syndromes in Pakistani population. NMC Blood Disorders 2012; 12: 1-6.

369

5. Khan MR, Aziz MA, Shah MSU, Imam H. Hemoglobin E trait- in Rajshahi, Bangladesh. Bangladesh Med ResCounc Bull 2012; 38: 72-73.

6. Tamminga RYJ, Doombos ME. Muskiet FAJ, Koetse HA. Rhabdomyolysis in a child with Hb SE. Pediatric Hematology-Oncology 2012; 29(3): 267-269.

7. Edison ES, Shaji RV, Chandy M, Srivasta A. Interaction of Hemoglobin E with Other Abnormal Hmoglobins. Acta Haematol 2011; 126: 246-248.

8. Tay SH, Teng GG, Poon M, Lee VKM, Lim AYN. A Case of Hemoglobin SE Presenting with Sickle Cell Crisis: Case Report and Histological Correlation. Annl Acad Med 2011; 40 (12): 552-553.

9. Colah R, Gorakshakar A, Nadkarni A. Global burden, distribution and prevention of β-thalassemias and hemoglobin E disorders. Expert Review of Hematology 2010; 3: 103-117.

10.Patel J, Patel A, Patel J, Kaur A, Patel V. Prevalence of Haemoglobinopathies in Gujrat, India: A Cross-Sectional Study. The Internet Journal of Hematology 2009; 5 (1): DOI: 10.5589/1764.

11. Intorasoot S, Thongpung R, Tragoolpua K, Chottayaporn M. Hemoglobin E Detection Using PCR with Confronting Two-Pair Primers. J Med Assoc Thai 2008; 91: 1677-1680.

12.Masiello D., Heeney MM, Adewoye AH, Eung SH, Luo Hong-Yuan, Dteinberg MH, Chui D HK. Hemoglobin S-E Disease- A Concise Review. Am H Hematol 2007; 82: 643-649.

13.Jetsrisuparb A, Sanchaisuriya K, Fucharoen G, Fucharoen S, Wiangnon S, Jetsrisuparb C, Sirijirachai J, Chansoong K. Development of Severe Anemia During Fever Episodes in Patients with Hemoglobin E trait and Hemoglobin H Disease Combinations. J Pediatr Hematol Oncol 2006; 28 (4): 249-253.

14. Bain BJ. Hemoglobin E. Other significant hemoglobinopathies. In: Hemoglobinopathy Diagnosis. 2nd Ed, 2006, pg 201-209, Blackwell Publishing, London.

15.Edison ES, Shaji RV, Srivastava A, Chandy M. Compound Heterozygosity for Hb E and Hb Lepore-Hillandia in India: First report and potential diagnostic pitfalls. Hemoglobin 2005; 29(3): 221-224.

16.Andino L, Risin SA. Pathologic Quiz case. A 24-Year-Old Woman With Abnormal Hemoglobin and Thrombocytopenia. Arch Pathol Lab Med 2005; 129: 257-258.

17.Mishra P, Pati HP, Chatterjee T, Dixit A, Choudhary DR, Srinivas MV, Mahapatra M, Choudhary VP. HB SE Disease: a clinico-hematological profile. Ann Hematol 2005; 84: 667-670.

18.Sirichotiyakul S. Tongprasert F, Tonsong T. Screening for hemoglobin E trait in pregnant women. Intl J Gyn & Obstet 2004; 86: 390-392.

19.Fucharoen S, Sanchaisuriya K, Fucharoen G, Panyasai S, Devenish R, Luv L. Interaction of hemoglobin E and several forms of α-thalassemia in Cambodian families. Haematologica 2003; 88: 1092-1098.

20.Piplani S. Hemoglobin E Disorders in the North East India. JAPI 2000; 48(11): 1082-1084.

370

21.Fucharoen S. Hemoglobin E disorders. In: Steinberg MH, Forget BG, Higgs DR, Nagel RL, eds. Disorders of Hemoglobin: Genetics, Pathophysiology and Clinical Management. Cambridge, England: Cambridge University Press; 2001: 1139-1154.

22.Gupta R, Jarvis M, Yardumian A. Compound Heterozygosity for hemoglobin S and hemoglobin E. Br J Haematol 2000, 108: 463.

23.Joseph VJ, Sunny AO, Pandit N, Yeshwanth M. Double Heterozygosity for Hemoglobin S and E. Indian Pediatrics 1992; 29: 895-897.

24.Fairbanks VF, Gilchrist GS, Brimhali B, Jereb JA, Goldston EC. Hemoglobin E Trait Rexamined: A Case of Microcytosis and Erythrocytosis. Blood 1979; 53(1): 109-115.

371

Case # 14 a Hemoglobin E trait

A 56 year old male of Southeast Asian origin, migrated to America in 1972 with his parents. Physical examination showed no abnormalities.

Laboratory Data:

Hemoglobin 14.8 13.5-18.5 g/dL

RBC 5.7 4.6-6.2 Mil/mm3

MCV 74 80-100 fL RDW 13.5 11.5-14.5%

Platelet 240 150-400 Th/mm3

Hb A 64.0 94.3-98.5%

Hb A2 (CZE) ≈2.0 1.5-3.7% Hb variant 34.0

Peripheral Blood Smear: Slight microcytosis, and occasional target cellsSickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test: Positive

Agarose Gel Electrophoresis (pH 8.6)

372

Case # 14 a Hemoglobin E trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

373

Case # 14 a Hemoglobin E trait

Capillary zone electrophoresis

High performance liquid chromatography

374

Case # 14 a Hemoglobin E trait

Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose Major band (Hb A)

Majorband(Hb E +

Hb A2)Acid Agar/Agarose

Major band (Hb A+Hb E+

Hb A2)CZE Major

peak(Hb A)Zone 9

Majorpeak(Hb E)Zone 4

Minor peak

(Hb A2)Zone 3

IEF Major band (Hb A)

Major band(Hb E)Slightly anodal to Hb A2

Very minor band

(Hb A2)

HPLC* Veryfaint peak(Hb F)RT=1.07

Major peak(Hb A) RT=2.42

Major peak (Hb E + Hb

A2) RT=3.63

* Note: HPLC retention time (RT) varies with the type of the instrument used and several other factors, e.g. temperature etc.

Alkaline agarose electrophoresis (pH 8.6) showed two major bands in the

position of Hb A (≈65%), and Hb C/E/O/A2 (≈35%). Citrate agar electrophoresis

showed only one band in the position of Hb A. This kind of electrophoretic

migration pattern (pH 8.6 and 6.2) ruled out the possibility of Hb C

375

and Hb O, and suggested the possibility of Hb E, as Hb A2 is never > 10%. IEF

also showed a major band in the position of Hb A and another band slightly

anodal to Hb A2 suggesting the presence of Hb E variant. Hb E and Hb A2 co-

eluted upon HPLC, therefore their quantification was not feasible. However,

CZE presented three distinct peaks in the zones for Hb A, Hb E and Hb A2 and

also provided quantification of the peaks.

Hb E is a β-chain variant (α2β26Glu-Lys) and is the second most prevalent

hemoglobin variant in the world after Hb S. It is prevalent in sixteen Southeast

Asian countries, however it is also encountered in Europe and North America.

A diagnosis of Hb E trait was made in view of the electrophoretic results and the

following characteristics:

Microcytosis, hypochromia, target cells, irregularly contracted cells, basophilic stippling or any combination of these features of the peripheral blood film.

Negative sickle cell solubility test. Positive isopropanol test for unstable hemoglobin. Per se harmless condition and not associated with anemia.

Hb A2 and Hb F in the normal range.

Hb E in the range of 30-39%. Hb E is a β-chain variant, however the βE chain is

synthesized in Hb E trait at a reduced rate in comparison with βA. In view of this

slower ribosomal synthesis of βE chain, a mild thalassemic blood picture is also witnessed.

Case # 14 b Hemoglobin E homozygous376

A 25 year old female from Laos-Northern Vietnam region, asymptomatic; physical examination showed no abnormalities. No icterus, no splenomegaly.

Laboratory Data:

Hemoglobin 14.1 12.0-16.0 g/dL

RBC 5.7 4.0-5.5 Mil/mm3

MCV 68 80-100 fL MCH 24.9 26-34 pg

Platelet 223 150-400 Th/mm3

Hb A Not Detected 94.3-98.5%

Hb A2 (CZE) ≈0.5 1.5-3.7% Hb F (HPLC) ≈0.8 0.0-2.0% Hb variant ≈98.7%

Peripheral Blood Smear: Significant microcytosis, hypochromia, target cells and occasional basophilic stippling of erythrocytes.Sickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test: Positive

Agarose Gel Electrophoresis (pH 8.6)

Case # 14 b Hemoglobin E homozygous

377

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

Case # 14 b Hemoglobin E homozygous

378

Capillary zone electrophoresis

High performance liquid chromatography

Case # 14 b Hemoglobin E homozygous

379

Interpretation & Discussion

Summary of Results

Method

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band(Hb E +

Hb A2)

Acid Agar / Agarose

Major band(Hb E +

Hb A2)

CZE Major peak(Hb E)Zone 4

Very minor peak(Hb A2)Zone 3

IEF Major band(Hb E)slightly anodal to Hb A2

HPLC* Major peak(Hb E + Hb A2)RT=3.65

* In addition to Hb F peak at RT= 1.05 minutes, there are two additional minor peaks at RT= 1.75 minutes and RT= 2.1 minutes. The peak at RT=2.1 minutes (A0 window) must not be construed as Hb A. Similar peaks were detected upon HPLC (Bio-Rad Variant II), and were alleged to post-translationally modified Hb E (Other Significant Hemoglobinopathies. In: Hemoglobinopathy Diagnosis, Bain, Barbara J., 2nd edition, pg 206, Blackwell Publishing, 2006).

380

Alkaline agarose electrophoresis (pH 8.6) showed only one band in the

position of Hb C/E/O/Hb A2, therefore Hb A was not present. Citrate agar

electrophoresis (pH 6.2) indicated only one major band in the position of Hb A,

thus the possibility of Hb C and Hb O was ruled out. IEF also indicated the

absence of Hb A and only one major band slightly anodal to Hb A2 was detected.

Absence of Hb A was also shown by HPLC, and only one major peak eluted at

RT = 3.65. CZE clarified the ambiguity about the solitary band /peak in

electrophoretic methods and HPLC, and two peaks were detected at Zone 4

(major peak presumably of Hb E) and Zone 3 (minor peak of Hb A2).

A diagnosis of Hb E homozygous was made in view of the electrophoretic,

and HPLC results and the following characteristics:

Absence of anemia and hemolysis. Spleen was not enlarged. Negative sickle cell solubility test. Isopropanol test positive for unstable hemoglobin. Increased red cell count, normal hemoglobin, decreased MCV and MCH. Significant microcytosis, hypochromia, and variable number of target cells. Harmless condition.

Hb A2 and Hb F in the normal range. Hb E concentration >95 %.

Homozygous Hb E is a clinically benign condition. Unfortunately, it is

prevalent in the population areas (e.g. Cambodia and Northeastern India) that

also have the higher frequency of β-thalassemia minor. Therefore genetic

counseling is advised to prevent the occurrence of severe thalassemic

381

(Hb E/βo – thalassemia) disorders in their children.

On the basis of hematological studies alone, homozygous Hb E may be

confused with iron deficiency and β-thalassemia. The following characteristic

features can distinguish between Hb E/β-thalassemia and homozygous Hb E.

In Hb E/β-thalassemia the concentration of Hb E varies between 40-70%, and the Hb F concentration is found in the range of 30-60%.

In homozygous Hb E the Hb F concentration is in the normal range, and Hb E concentration is >95%.

Case # 14 c Hemoglobin S-E disorder

382

A 29 year old female nurse (Southeast Asian descent) complained of knee joint pain and weakness of the lower extremities. Hemoglobin electrophoresis was ordered after lower MCV was found from CBC. Clinical and laboratory data of her parents were not available to the physician.

Laboratory Data:

Hemoglobin 12.2 12.0-16.0 g/dL

RBC 4.6 4.0-5.5 Mil/mm3

MCV 72 80-100 fL MCH 26 26-34 pg

Platelet 212 150-400 Th/mm3

Hb A Not Detected 94.3-98.5%

Hb A2 (CZE) ≈0.5 1.5-3.7% Hb F (HPLC) ≈0.5 0.0-2.0% Hb S 55% Hb E 44%

Peripheral Blood Smear: Hypochromia, microcytosis, irregular contracted cells, occasional target cells, polychromatic cells.Hemoglobin instability (isopropanol) test: PositiveSickle cell solubility test for Hb S: Positive

Agarose Gel Electrophoresis (pH 8.6)

Case # 14 c Hemoglobin S-E heterozygous disorder

383

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

Case # 14 c Hemoglobin S-E heterozygous disorder

384

Capillary zone electrophoresis

High performance liquid chromatography

Case # 14 c Hemoglobin S-E heterozygous disorder

385

Interpretation & Discussion

Summary of Results

Method

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band(Hb S)

Major band(Hb E)

Acid Agar/ Agarose

Major band(Hb E)

Major band(Hb S)

CZE Major peak(Hb S)Zone 5

Major peak(Hb E)Zone 4

Minor peak

(Hb A2)Zone 3

IEF Major band(Hb S)

Major band(Hb E)slightly anodal to

Hb A2

HPLC* Major peak(Hb S)RT=4.48

Major peak(Hb E +

Hb A2)RT=3.65

*Note: HPLC retention time (RT) varies with the type of the instrument used and several other factors, e.g. temperature etc.

Alkaline agarose gel electrophoresis (pH 8.6) showed two major bands in

the position of Hb S (≈55%) and Hb C/E/O/A2 (≈45%). Citrate agar

386

electrophoresis (pH 6.2) confirmed the presence of a major band due to Hb S and

presence of another band at the position of Hb A ruled out the possibility of Hb C

and Hb O. IEF indicated a major band in the position of S and another major

band in the position of Hb A2. These three electrophoretic methods suggested

the presence of double heterozygous Hb S and Hb E. HPLC was not helpful as

both Hb E and Hb A2 co-eluted at RT = 3.65. CZE provided a clear separation of

the three hemoglobin entities, i.e. Hb S (Zone 5), Hb E (Zone 4) and Hb A2 (Zone

3), therefore a presumptive diagnosis of Hb S-E double heterozygous disorder

was made.

Hb S [β6 (A3) Glu→Val) and Hb E [β26 (β8) Glu→Lys] are the two most

prevalent hemoglobin variants in the world. However, due to their existence in

different ethnic groups and continents, compound heterozygosity for Hb S and

Hb E is extremely rare. As of 2011 only thirty (30) cases were reported, therefore

hematological parameters are too scant to provide a module for diagnostic

purposes.

Majority of Hb S-E subjects have mild or absent anemia, microcytic

indices, and some target cells. Contrary to some earlier reports, a severe sickle

cell crisis was recently reported in a 66-year-old Bangladeshi woman in

Singapore (Ann Acad Med 2011; 40: 552-553).

A recent review of Hb S-E double heterozygosity by Masiello et al (Am J

Hematol 2007; 82: 643-649) mentioned that patients aged 18 and younger are

387

usually well. Sickling-related complications, including potentially life threatening

acute chest syndrome was developed in a majority of cases. Generally these

patients have Hb S concentration in the range of 60-65%, which is also similar to

the patients with Hb S-β+-thalassemia, and therefore hematological features and

clinical course of these patients appeared to parallel those of Hb S-β+-

thalassemia.

Coincidental complication per se not related to hemoglobinopathy was

also reported, e.g. idiopathic thrombocytopenic purpura in a 28-year-old woman

with Hb S-E heterozygosity (Arch Pathol Lab Med 2005, 129: 257-58).

388

Case # 15 Hemoglobin S-Korle Bu (G-Accra)

A 23 year old female administrative assistant of Ghanaian decent at the Embassy of Ghana in Washington, DC was found to have abnormal hemoglobinopathy. Physical examination was unremarkable.

Laboratory Data:

Hemoglobin 13.1 12.0 – 16.0 g/dL

RBC 4.4 4.0 – 5.5 Mil/mm3

MCV 82.1 79-98 fL MCH 27.3 26-34 pg MCHC 32.1 31-36% RDW 12.6 11.5-14.5%

Platelet 267 150-400 Th/mm3

Hb A Not Detected 94.3-98.5%Hb S (HPLC) 53.4 %

Hb Korle-Bu ≈45%

Hb A2 (CZE) ≈1.6 1.5 -3.7% Hb F Not Detected 0.0-2.0%

Peripheral Blood Smear: No abnormality was present.Sickle cell solubility test for Hb S: PositiveHemoglobin instability (isopropanol) test: Negative

Agarose Gel Electrophoresis (pH 8.6)

389

Case # 15 Hemoglobin S-Korle Bu (G-Accra)

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

390

Case # 15 Hemoglobin S-Korle Bu (G-Accra)

Capillary zone electrophoresis

High performance liquid chromatography

391

Case #15 Hemoglobin S–Korle Bu (G-Accra)

Interpretation & DiscussionSummary of ResultsMethod Hb A

areaHb S area

Hb A2/C area

Alk Agarose Broad Major band starting anodic to S(Hb Korle-Bu + Hb S)

Very faint band

(Hb A2)

Acid Agar Major band slightly anodic to Hb A position(Hb Korle-Bu)

Major band(Hb S)

Acid Agarose Major band in Hb A position(Hb Korle-Bu)

Major band(Hb S)

CZE Small degradation peaksZone 7

Major peakZone 6(Hb Korle-Bu)

Major peak Zone 5(Hb S)

Minor peak Zone 3

(Hb A2)

IEF Major band slightly anodal to Hb S band (Hb-Korle-Bu)

Major band (Hb S)

Very faintband

(Hb A2)

HPLC Major peakRT=4.48(Hb S)

Major peak(Hb A2 + Hb Korle Bu) RT=3.75

392

Alkaline agarose electrophoresis (pH 8.6) showed a major band in the

area of Hb S and a very faint band in the Hb A2 area, but Hb A was not

detected. Citrate agar electrophoresis (pH 6.2) showed two variants in the

position of Hb S and very slightly anodic to the Hb A band position.

IEF also indicated the presence of two variants, one in the position of Hb S and

another band anodal to Hb S in the migration position of Hb D-Los Angeles/G-

Philadelphia and few other variants. The presence of Hb S as one of the variants

was also confirmed by HPLC, CZE and the positive solubility test. It is obvious

that the three electrophoretic methods (i.e. alkaline agarose, acid agar, and IEF)

could not identify the second major band with certainty. Hb G-Philadelphia was

ruled out due to the absence of G2 variant (αG2δ2) in both the alkaline agarose

electrophoresis and IEF (Case # 5). In situations like this the patient’s history and

clinical status may indicate the likelihood of the hemoglobin variant. Both of these

hemoglobin variants (Hb D-Los Angeles and Hb Korle-Bu) are found in the

population sector dominated by Hb S. HPLC is helpful in the separation of Hb D-

Los Angeles from Hb Korle-Bu, since Hb D-Los Angeles has a longer elution time

(Case # 12) as compared to Hb Korle-Bu. The βD-Los Angeles chain can be easily

separated from βKorle-Bu chain by reverse phase chromatography, but all these

additional tests are not necessary. Hb S interacts with Hb D-Los Angeles causing

sickle cell disease. Hb S also interacts with Hb Korle-Bu, but in opposite

direction, i.e. inhibiting sickling. This patient does not have an abnormal blood

picture so the second variant in this case is most likely Hb Korle-Bu (G-Accra).

393

Korle-Bu means “valley of the Korle lagoon”, and this hemoglobin was

named after its discovery at Korle-Bu Hospital, Accra, Ghana. Initially it was

called Hb G, and since several other hemoglobins with mobility similar to Hb G

were discovered, its name was changed to Hb Accra. The reason for changing its

name from Hb Accra to Hb Korle-Bu is not known to me. This mutation has been

reported from the Ivory Coast, Costa Rica, and Mexico, so it is not highly

prevalent but is widely spread.

Hb Korle-Bu is the result of mutation GAT→ AAT at codon 73 of the β

chain [(E73)Asp→Asn]. Both the heterozygote and homozygote of Hb Korle-Bu

are clinically normal.

References

1. AKL PS, Kutlar F, Patel N, Salisbury CL, Lane P, Young AN. Compound Heterozygosity for Hemoglobin S [β6(A3)Glu6Val] and Hemoglobin Korle-Bu [β73(E17)ASP73Asn]. Lab Hematol 2009; 15: 19-23.

2. Chico A, Padros A, Novials A. The Korle-Bu Hb Variant in Caucasian Women With Type I Diabetes. A pitfall in the assessment of diabetes control. Diabetes Care 2004; 27(9): 2280-2281.

3. Vukelja SJ. Hemoglobin Korle-Bu (G-Accra) in Combination with Hemoglobin C. Am J Hematol 1993; 42(4): 412.

4. Nagel RL, Lin MJ, Witkowska HE, Fabry ME, Bestak M, Hirsch RE. Compound Heterozygosity for Hemoglobin and Korle-Bu: Moderate Microcytic Hemolytic Anemia and Acceleration of Crystal Formation. Blood 1993; 82(6): 1907-1912.

5. Konotey-Ahulu FID, Gallo E, Lehman H, Ringelhann B. Hemoglobin Korle-Bu (β73 aspartic acid→asparagine). J Med Genet 1968; 5: 107-111.

394

Case # 16 Hemoglobin O-Arab trait

A 23 year old male student of Northern African descent at Michigan State University, East Lansing, Michigan, USA.

Laboratory Data:

Hemoglobin 14.8 12.0 – 16.0 g/dL

RBC 4.9 4.0 – 5.5 Mil/mm3

MCV 86 79-98 fL MCH 29.3 26-34 pg MCHC 32.6 31-36% RDW 12.5 11.5-14.5%

Platelet 273 150-400 Th/mm3

Hb A 56.5% 94.3-98.5Hb O-Arab 41%

Hb A2 ≈2% 1.5 -3.7 Hb F ≈0.5% 0.0-2.0 (Hemoglobin fractions from HPLC)

Peripheral Blood Smear: No abnormality was present.Sickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test: Negative

Agarose Gel Electrophoresis (pH 8.6)

395

Case # 16 Hemoglobin O-Arab trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

396

Case # 16 Hemoglobin O-Arab trait

Capillary zone electrophoresis

High performance liquid chromatography

397

Case # 16 Hemoglobin O-Arab trait

Interpretation & Discussion

Summary of ResultsMethod

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band(Hb A)

Major band(Hb O-Arab)

Actually on Hb S side of Hb C

Acid Agar

Actually on Hb S side of Hb A

Major band

May appear as a broader Hb A band

Major band(Hb O-Arab)

Actually migrates cathodal to Hb S)

Acid Agarose

Major band(Hb A)

Major band(Hb O-Arab)

Actually on Hb C side of Hb S

CZE Smallpeak(Hb F)Zone 7

Major peak(Hb A)Zone 9

Minor peak(Hb O-Arab)Zone 3

IEF Major band(Hb A)

Major band(Hb O-Arab)

HPLC Small peak(Hb F) RT=1.08

Major peak(Hb A)RT=2.38

Minor peak(Hb A2)RT=3.64

Major peak (Hb O-Arab)RT=4.3

Note: Separation under acidic conditions has traditionally been done with agar instead of agarose because all the early descriptions of hemoglobin variants contained data collected in this manner. As the separation quality of agar has deteriorated many vendors have chosen to switch to use of the purified agarose in order to maintain a more constant product. Hemoglobin O-Arab migrates close to hemoglobin A historically and with well selected current agars. In the heterozygous state one broad band is seen starting with Hb A but extending on toward Hb S somewhat. When agarose is substituted Hb O-Arab migrates close but on the Hb C side of the Hb S location. The user is advised to take note of the separation media used for acid electrophoresis in the interpretation of the results.

398

Since the concentration of the band that migrated at or near Hb C/E/O/A2

position on alkaline electrophoresis was significantly > 10%, Hb A2

was ruled out because Hb A2 virtually never has such an increase. Citrate agar

electrophoresis (pH 6.2) eliminates the possibility of Hb C or Hb C-Harlem.

Incidentally the migration of Hb C-Harlem, Hb O-Arab and Hb E is virtually

identical upon IEF, therefore it was not helpful in the differentiation of these

variants. HPLC and CZE show characteristic elution times (RT) and migration

mobilities respectively for Hb O-Arab. In view of the characteristic laboratory tests

and normal peripheral blood smear a tentative diagnosis of Hb O-Arab trait was

made.

Hemoglobin O-Arab was first discovered (in association with Hb S) in an

Arabic-speaking Israeli village (Giser-A-Zarke), and thus got its name as Hb O-

Arab. It is the same village Sayar reported the homozygous Hb O-Arab

(reference 2). It is emphasized here that Hb O-Arab is not prevalent in Israel or

the Jewish population. However, three homozygous Hb O-Arab cases from

progeny of parents who originally came from South Sudan were recently reported

[Sayar D. Clinical and Hematological Features of Homozygous Hemoglobin O-

Arab (Beta 121 Glu→Lys). Pediatr Blood Cancer 2013; 60: 506-507]. Hb O-Arab

has been found in Northern Africa (Tunisia), African-American, Saudi Arabia,

Bulgaria and the Mediterranean littoral. Hemoglobin O-Arab is a β-chain variant

(β121 Glu→Lys), and exhibits no evidence of hemolysis and anemia in the

heterozygous state. Persons with Hb O-Arab trait are clinically well. Homozygous

399

Hb O-Arab exhibits a mild anemia. Hb O-Arab interacts with Hb S (double

heterozygous) and produces a disorder similar to homozygous Hb S disease with

all of its characteristic features. Hb O-Arab also interacts with β-thalassemia, and

these individuals exhibit moderately severe hemolytic anemia and splenomegaly.

References

1. Bain BJ. Hemoglobin O-Arab, other significant hemoglobins. In: Hemoglobinopathy Diagnosis, 2066, 2nd edition, Blackwell Publishing, pg 213-15.

2. Sayar D. Clinical and Hematological Features of Homozygous Hemoglobin-O Arab [Beta 121 Glu→Lys]. Pediatr Blood Cancer 2013; 60: 506-507.

3. Zimmerman SA, O’Branski EE, Rosse WF, Ware RE. Hemoglobin S/OArab : Thirteen New Cases and Review of the Literature. Am J Hematol 1999; 60: 279-284.

4. Sangore A, Sanogo I, Meite M, Ambofo Y, Abe Sopie V, Segbena A, Tolo A. Hemoglobin O Arab Disease in Ivory Coast and West Africa. Medicine Tropicale 1992; 52(2): 163-167.

5. Altay C, Gurgey A, Huisman Titus TJ. Homozygosity For Hemoglobin O-Arab

(α2β2121 Glu→Lys) Hb O-Arab Disease. The Turkish Journal of Pediatrics 1986;

28: 67-72.6. Rachmilewitz EA, Tamari H, Liff F, Ueda Y, Nagel RL. The interaction of

hemoglobin O Arab with Hb S and β+ thalassemia among Israeli Arabs. Hum Genet 1985; 70: 119-125.

7. Ballas SK, Atwater J, Burka ER. Hemoglonin S-O Arab-α-Thalassemia: Globin Biosynthesis and Clinical Picture. Hemoglobin 1977; 1(7): 651-662.

8. Milner PF, Miller C, Grey R, Seakins M, DeJong WW, Went LN. Hemoglobin O Arab in four negro families and its interaction with hemoglobin S and hemoglobin G. N Eng J Med 1970; 283(26): 1417-1425.

400

Case # 17 β-Thalassemia trait

A 27 year old Caucasian female.

Laboratory Data:

Hemoglobin 12.5 12.0 – 16.0 g/dL

RBC 5.13 4.0 – 5.5 Mil/mm3

MCV 63.8 79-98 fL MCH 20.5 26-34 pg RDW 12.1 11.5-14.5%

Platelet 267 150-400 Th/mm3

Serum Iron 110 30-160 ug/dLFerritin 75 8-120 ng/mLHb A 93.5

Hb A2 6.0 1.5 -3.7% Hb F 0.5 0.0-2.0% (Hemoglobin fractions from HPLC)

Peripheral Blood Smear: Hypochromasia, microcytosis, target cells, basophilic stippling

Sickle solubility test for hemoglobin S: NegativeUnstable hemoglobin test (isopropanol): Negative

Agarose Gel Electrophoresis (pH 8.6)

401

Case # 17 β-Thalassemia trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

402

Case # 17 β-Thalassemia trait

Capillary zone electrophoresis

High performance liquid chromatography

403

Case #17 β-Thalassemia traitInterpretation & Discussion

Summary of ResultsMethod

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band(Hb A)

Minor band

(Hb A2)Acid Agar/ Agarose

Major band(Hb A +

Hb A2)CZE Small

peak(Hb F)Zone 7

Majorpeak(Hb A)Zone 9

Minorpeak

(Hb A2)Zone 3

IEF Major band(Hb A)

Minor band

(Hb A2)

HPLC* Smallpeak(Hb F) RT=1.05

Major peak(Hb A)RT=2.42

Minorpeak

(Hb A2)RT=3.64

*Note: HPLC retention time (RT) varies with the type of instrument used and several other factors, e.g. temperature etc.

Alkaline agarose electrophoresis (pH 8.6) showed no abnormality except

that the staining of the band at the Hb A2 position was relatively denser than the

normal adult. No abnormal band was detected from citrate agar

electrophoresis (pH 6.2). Two bands in the migration position of Hb A (major

band) and Hb A2 (minor band but more intense than a normal adult) were

indicated by IEF. CZE and HPLC results were concordant suggesting

404

an increased concentration of Hb A2 and no other abnormal peaks.

Hemoglobinopathies can be classified as a manufacture of a modified

globin chain or a failure or decrease in the ability to manufacture a particular

globin chain. This latter set of conditions is referred to as a thalassemia. A

decreased ability to manufacture beta chains is called β-thalassemia and results

in small erythrocytes (microcytosis) and a decreased amount of hemoglobin per

erythrocytes and thinness of the cell (hypochromasia). Due to insufficient beta

chains there is a surplus of alpha chains which bind to the red blood cell

membranes causing damage and an occasional clump of alpha chains is the

center of the “Target Cells”. The delta chains compete with the beta chains

present with the delta chains getting a larger proportion in this beta chain

deprived environment and this accounts for an elevated Hb A2.

The hematological and morphological parameters along with elevated

Hb A2 suggested the diagnosis of β-thalassemia trait. In the presence of serum

iron deficiency Hb A2 can be falsely lower, therefore quantification should be

done again after the correction of the iron deficiency. β-thalassemia trait is

clinically a benign condition most often found in persons of the Mediterranean,

Chinese, African American and other Asian ethnic groups. However problems

arise when a thalassemic gene is inherited from both parents, e.g. causing

Cooley’s anemia (thalassemia major). Incidentally this disease was first

discovered in an Italian population by Dr. Denton Cooley in Detroit, Michigan,

USA.

405

A thorough review of the articles mentioned in the references is strongly

advised to make a correct diagnosis of various kinds of thalassemias (minor,

intermedia, and major) and also its interactions with several other hemoglobin

variants. Molecular characterization is necessary for genetic counseling when

both parents are carriers of β-thalassemia minor or other hemoglobinopathies.

References

1. Galanello R, Origa R. Review: Beta-thalassemia. Orphanet J Rare Diseases 2010; 5(11): 1-15. http://www.ojrd.com/content/5/1/11

2. Cao A, Galanello R. Review: Beta-thalassemia. Genetics in Medicine 2010;12(2): 61-76.

3. Colah R, Gorakshakar A, Nadkarni A. Global burden, distribution and prevention of β-thalassemias and Hemoglobin F disorders. Expert Review of Hematology 2010; 3(1): 103-116.

4. El Rassi F, Cappellini D, Inati A, Taher A. Beta-thalassemia Intermedia: An Overview. Pediatric Annals 2008; 37(5): 302-328.

5. Bain BJ. β-thalassemia trait. The α, β, δ and γ-thalassemia and related conditions. In: Hemoglobinopathy Diagnosis, 2nd Edition, Blackwell Publishing; 2006: 95-105.

6. Oliveri N, Weatherall DJ. Clinical Aspects of β thalassemia. In: Steinberg MH, Forget BG, Higgs DR, Nagel RL, eds. Disorders of Hemoglobin: Genetics, Pathophysiology, and Clinical Management. Cambridge, England: Cambriadge University Press; 2001: 277-341.

7. Weatherall DJ, Clegg JB. The βthalassemias. In: The Thalassemia Syndromes, 4th ed. Oxford, England: Blackwell Science, Ltd; 2001: 287-356.

8. Qari MH, Wali Y, Albagshi MH, Aishahrani M, Alzahrani A, Alhijji IA, Almomen A, Aljefri A, Al-Saeed HH, Abdullah S, Al-Rustamani A, Mahour K, Mousa SA. Regional consensus opinion for the management of beta thalassemia major in the Arab Gulf Area. Orphanet J Rare Diseases 2013; 8: 143. Available fromhttp://www.ojrd.com/content/8/1/143

406

Case # 18 Hemoglobin S-β+- thalassemia

A 37 year old African American female.

Laboratory Data:

Hemoglobin 10.3 12.0 – 16.0 g/dL

RBC 4.28 4.0 – 5.5 Mil/mm3

MCV 74.8 79-98 fL MCH 23.9 26-34 pg RDW 16.2 11.5-14.5%

Platelet 203 150-400 Th/mm3

Hb A 22.1 94.3-98.5%

Hb A2 6.4 1.5 -3.7% Hb F 3.4 0.0-2.0% Hb S 68.1

(Hemoglobin fractions from HPLC)

Peripheral Blood Smear: microcytosis, rare target cells, moderate poikilocytosis

Sickle cell solubility test for Hb S: PositiveHemoglobin instability (isopropanol) test: NegativeNo record of blood transfusion during the past seven months.

Agarose Gel Electrophoresis (pH 8.6)

407

Case # 18 Hemoglobin S-β+-thalassemia

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

408

Case # 18 Hemoglobin S-β+-thalassemia

Capillary zone electrophoresis

High performance liquid chromatography

409

Case # 18 Hemoglobin S-β+-thalassemia

Interpretation & Discussion

Summary of Results

Method

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Very weakband(Hb F)

Small band(Hb A)

Major band(Hb S)

Minor band

(Hb A2)

Acid Agar/ Agarose

Weakband(Hb F)

Small band(Hb A+

Hb A2)

Major band(Hb S)

CZE Smallpeak(Hb F)Zone 7

Small peak(Hb A)Zone 9

Major peak(Hb S) Zone 5

Minor peak

(Hb A2)Zone 3

IEF Verysmall band(Hb F)

Verysmall band(Hb A)

Major band(Hb S)

Minor band

(Hb A2)

HPLC Smallpeak(Hb F) RT=1.05

Small peak(Hb A)RT=2.40

Major peak(Hb S)RT=4.32

Minor peak

(Hb A2)RT=3.6

Agarose gel electrophoresis (pH 8.6) showed one major band in the

position of Hb S, minor band (less in intensity than Hb S band) in the migration

position of Hb A, and another band in the position of Hb C/E/O/A2 with intensity

greater than a normal adult. Citrate agar electrophoresis (pH 6.2) indicated

increased Hb F than a normal adult, and a major band in the position of Hb S,

410

major band in the position of Hb A but less in intensity than Hb S. IEF, CZE, and

HPLC also provided concordant results and the evidence for the following four

bands:

Hb S Major band-positive sickle cell test Hb A Major band-concentration less than Hb S

Hb A2 Minor band-concentration greater than a normal adultHb F Minor band –concentration greater than normal adult

Quantitatively increased Hb A2 (6.4% by HPLC) suggested the presence of β-

thalassemia in conjunction with Hb S. Two types of Hb S-β-thalassemia are

found in African Americans:

Hb S-β+-thalassemia: Type 1 with Hb A concentration 5-15%

Hb S-β+-thalassemia Type 2 with Hb A concentration 20-30%

This case represents Hb S-β+-thalassemia Type 2. It is emphasized here that

precaution is warranted in the interpretation at any time Hb A is less than Hb S.

This kind of situation can be encountered in homozygous sickle cell disease with

recent blood transfusion.

Hb S-β-thalassemia in African Americans is present in two clinically

significant conditions. If the Hb A is completely absent then it is termed Hb S-

β0-thalassemia, and is clinically similar to homozygous sickle cell disease.

If Hb A is present (Type 1 or Type 2) the person will have a milder

clinical course, and elevation of Hb F is also characteristic feature of Hb S-

β+-thalassemia.

411

Appendix:

I looked for a case of Hb S-β0-thalassemia for > three years but no luck. Just yesterday a 23 year female came to the Emergency Department with a severe sickle cell crisis. I immediately contacted the attending physician and my associates involved in this book to include the data of this patient for the benefit of readers for understanding the distinction between the two types of Hb S-β-thalassemias.

CBC

Hemoglobin 5.3 12.0 – 16.0 g/dLHematocrit 15.1 35.0 - 48.0%

RBC 2.13 4.0 - 5.5 Mil/mm3

MCV 70.7 79 - 98 fLMCH 24.8 26 - 34 pgRDW 24.3 11.5 - 14,5 %

Platelet 407 150 - 400 Th/mm3

Reticulocyte Count 13.5 0.5 - 1.5%

Alkaline agarose (pH 8.6), Citrate agar (pH 6.2), IEF, HPLC and CZE indicated the absence of Hb A. The concentration of hemoglobin fractions from CZE were:

Hb A Not Detected

Hb A2 4.5 ↑Hb F 8.6 ↑Hb S 86.9

On the basis of CBC and laboratory results a diagnosis of Hb S-β0-thalassemiais most likely.

References

1. Bain BJ. Sickle cell/β thalassemia, Sickle cell hemoglobin and its interactions with other variant hemoglobins and with thalassemia. In: Hemoglobinopathy Diagnosis, 2006, 2nd edition. Blackwell Publishing, England, pg 170-173.

2. Steinberg MH. Compound heterozygous and other sickle hemoglobinopathies. In: Steinberg MH, Forget BG, Higgs DR, Nagle RL. Disorders of Hemoglobin: Genetics, Pathophsiology and Clinical Management. Cambridge , England: Cambridge University Press; 2001: 786-792.

412

3. Sunna EI, Gharaibeh NS, Knapp DD, Bashir NA. Prevalence of Hemoglobins S and β-Thalassemia in Northern Jordan. J Obstet Gynecol Res 1996; 22(1): 17-20.4. Gonzalez-Redondo JM, Kutlar A, Kutlar F, McKie VC, Mckie KM, Baysai E, Huisman THJ. Molecular Characterization of Hb S (C) β-Thalassemia in American Blacks. Am J Hematol 1991; 38: 9-14.5. Serjeant GR, Ashcroft Mt, Serjeant BE, Milner PF. The clinical features of

sickle cell/β-thalassemia in Jamaica. Br J Hematol 1973; 24: 19-30.

413

Case # 19 Hemoglobin C-β0-thalassemia

A 17 year old female student from Turkey (most likely of Eti Turkish descent) visiting a prestigious high school in Michigan to brush up her English. She was asymptomatic.She declined to participate in athletic activities because she felt fatigue upon physical activity. Low hemoglobin and MCV triggered hemoglobin electrophoresis by the physician.

Laboratory Data:

Hemoglobin 10.3 12.0 – 16.0 g/dL

RBC 5.4 4.0 – 5.5 Mil/mm3

MCV 66.5 79-98 fL MCH 20.9 26-34 pg

Hb A (all methods) Not Detected 94.3-98.5%

Hb A2 5.9 1.5 -3.7% Hb F 1.4 0.0-2.0% Hb C 87.0

Hb A1c + other fractions 5.7Peripheral Blood Smear: Hypochomosia, microcytosis, target cells,

anisocytosis and poikilocytosisSickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test: Negative

Agarose Gel Electrophoresis (pH 8.6)

414

Case # 19 Hemoglobin C-β0-thalassemia

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

415

Case # 19 Hemoglobin C-β0-thalassemia

Capillary zone electrophoresis

High performance liquid chromatography

416

Case # 19 Hemoglobin C-βo-thalassemia Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Faintband(Hb F)

Major band(Hb C +

Hb A2)

Acid Agar /Agarose

Faint band(Hb F)

Major band(Hb C +

Hb A2)

CZE Minorpeak(Hb F)Zone 7

Minor peak(Hb A2)Zone 3

Major peak(Hb C)Zone 2

IEF Minor band(Hb A2)

Major Hb C band cathodal to A2

HPLC Minor peak(Hb F)RT=1.05

Minor peak(Hb A2)RT=3.6

Major peak(Hb C)RT=5.09

Note: Hb A was not detected by any of the six methods.

Alkaline agarose electrophoresis (pH 8.6) showed a single band in the

position of Hb C/E/O/Hb A2, thus indicating the absence of Hb A. Citrate agar

electrophoresis (pH 6.2) also indicated the absence of Hb A, and a major band

(87%) was indicated in the position of Hb C.

417

IEF showed a major band in the position of Hb C (cathodal to

Hb A2), minor band in the position of Hb A2 and a faint band in the position of Hb

F. CZE also indicated three peaks:

Hb C major peak in Zone 2Hb F minor peak in Zone 7Hb A2 minor peak in Zone 3 but obscured by

the larger peak in Zone 2 (Hb C)

HPLC separated the hemoglobins into three peaks, i.e. Hb C, Hb A2 and Hb F

and also provided quantitative results.

Increased Hb A2 (5.9%), absence of Hb A, microcytosis, target cells and a

major Hb C peak (87%) from HPLC suggested the presence of compound

heterozygosity for Hb C and β-thalassemia. It is emphasized here that a

distinction between homozygous Hb C and Hb C-β0-thalassemia is not feasible

from alkaline agarose gel electrophoresis (pH 8.6) alone due to lack of the

correct quantitative value of Hb A2 because of the overlap of Hb C and Hb

A2 bands. Absence of Hb A as in this case rules out the possibility of Hb C-β+-

thalassemia.

Due to similarity in clinical features it is sometimes not possible to

differentiate with certainty between homozygous Hb C and Hb C-β0-thalassemia.

Similar clinical features of homozygous Hb C and Hb C-β0-thalassemia are:

Mild to moderate chronic hemolytic anemia, with hemoglobin levels in the range of 8-12 g/dL and splenomegaly. The blood film shows large number of target

418

cells, folded cells, scattered spherocytes, hypochromia, microcytosis and polychromasia.

The two parameters that lead to the putative diagnosis of Hb C-β0-

thalassemia in this case are MCV (55-70) and increased Hb A2.

However, the Hb A2 fraction could be overestimated and HbC/beta0

syndromes are usually characterized by some hemolysis and

spelenomegaly while this patient is asymptomatic. Microcytosis could be

caused by alpha thalassemia and the genotype should be confirmed by

direct sequencing of the beta genes.

References

1. Kumar S, Rana M, Handoo A, Saxena R, Verma IC, Bhargava M, Sood SK. Case report of Hb C/β0-thalassemia from India. Int Jnl Lab Hem 2007; 29: 381-385.

2. Nagel RL, Steinberg MH, Hb S/C disease and Hb C disorders. In: Steinberg MH, Forget BG, Higgs DR, Nagel RL. Disorders of Hemoglobin: Genetics, Pathohysiology and Clinical Management.Cambridge, England: Cambridge University Press; 2001; 756-785.

3. Fattoum S, Guemira F, Abdennebi M, Ben Abdeladhim A. [Hbc/beta-thalassemia association. Eleven cases observed in Tunisia]. Ann Pediatr (Paris) 1993; 40(1): 45-8.

4. Maberrry MC, Mason RA, Cunningham G, Pritchard JA. Pregnancy Complicated by Hemoglobin CC and C-β-Thalassemia Disease. Obstet Gynecol 1990; 76: 324-327.

5. Ozsoylu S, Sipahioglu H, Altay F. Hemoglobin C-beta (O) thalassemia. Isr J Med Sci 1989; 25: 410-412.

419

Case # 20 Hemoglobin Hasharon trait

A 55 year old male computer programmer with age onset diabetes mellitus, was screened for hemoglobinopathy since one of his family member was anemic, and another having a hemoglobin variant. His parents (Ashkenazi Jews) migrated from Poland to Detroit, Michigan.

Laboratory Data:

Hemoglobin 14.8 12.0 – 18.0 g/dL

RBC 5.1 4.6 – 6.2 Mil/mm3

MCV 88.0 80 - 100 fL MCH 28.3 27 - 34 pg RDW 12.3 11.5 -14.5%

Platelet 203 15 - 400 Th/mm3

Hb A 77.3 94.3 - 98.5%

Hb A2 (CZE) 1.6 1.5 - 3.7% Hb F (CZE) 0.8 0.0 - 2.0% Hb variant (CZE) 20.3

Peripheral Blood Smear: No abnormality was detected.Sickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test was positivebut heat stability test was negative.

Agarose Gel Electrophoresis (pH 8.6)

420

Case # 20 Hemoglobin Hasharon trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

421

Case # 20 Hemoglobin Hasharon trait

Capillary zone electrophoresis

High performance liquid chromatography

422

Case # 20 Hemoglobin Hasharon trait

Interpretation and Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band(Hb A)

Major bandslightly toward C(Hb Hasharon)

Minor band(Hb A2)

Acid Agar

Major band(Hb A)

Major bandslightly toward A(Hb Hasharon)

Acid Agarose

Major band(Hb A)

Major banddirectly in S position(Hb Hasharon)

CZE Smallpeak(Hb F)Zone 7

Major peak(Hb A)Zone 9

Major peak(Hb Hasharon)Zone 5

Minor peakZone 3(Hb A2)

Very minor peakZone 1(Hb-“Hash

aron-A2)

IEF Major band(Hb A)

Major band slightly cathodal of Hb S.(Hb Hasharon)

Minor band(Hb A2)

A very faint band cathodal to Hb C

Hb A2

(α2Hδ2)

HPLC Very smallpeak (Hb F)RT=1.05

Majorpeak(Hb A)RT=2.40

Major peakelutedbetween S and C(Hb Hasharon)RT=4.74

Minor peak(Hb A2)RT=3.58

423

Alkaline agarose electrophoresis (pH 8.6) showed a major band in the

Hb A region, and another major band of lesser intensity cathodal to Hb S. A very

faint band was present in Hb A2 position. Citrate agar electrophoresis (pH 6.2)

also revealed two major bands with intensities equivocal to that described for

agarose gel electrophoresis (pH 8.6) in the respective Hb A and Hb S positions.

The sickle cell test was negative, ruling out the presence of Hb S.

IEF showed four bands: two intense bands and two faint bands. One

intense band in the position of Hb A, and a second band slightly cathodal to

Hb S. Additionally, there was a very faint band migrating in the Hb A2 position

and a second faint band in the delta chain variant position (cathodal to Hb C).

Hb A2 variants are due to the presence of an abnormal α-chain as seen in Hb G-

Philadelphia trait (Case # 5) or due to the presence of an abnormal delta chain.

Since this specimen also has an abnormal Hb A, the Hb A2 variant is likely due

to an alpha mutation.

CZE showed a major peak in the Hb A zone (Zone 9), and a lesser

intense peak than Hb A in Zone 5. Two minor peaks in the position of Hb F

(Zone 7) and Hb A2 (Zone 3) were also detected as well as a very small peak

in Zone 1.

HPLC showed a major peak for Hb A and two minor peaks for

Hb A2 and Hb F. Another major peak was detected between the Hb S and Hb

C window.

424

A narrative report was communicated to the attending physician with a

request for consultation with him to identify the exact hemoglobinopathy. The

physician communicated that the patient is an orthodox Ashkenazic Jew of Polish

origin.

Consistently typical migration patterns by the four electrophoretic

methods, elution retention times upon HPLC and the Ashkenazic Jewish ethnic

origin of the patient suggested the possibility of Hb Hasharon trait. Hb Hasharon

was first discovered in Hasharon Hospital, Israel in an Ashkanezic Jew, whose

father was from Poland and mother from Romania. It is α-chain variant caused

by a mutation on condom 47 that results in the substitution of aspartic acid by

histidine (α47 Asp→His).

The presence of Hb Hasharon is typically found in Ashkanezic Jews (who

have also migrated to several countries after World War II), and Italians from

Ferrara district of Italy. Hb Hasharon has not been recognized in Sephardic

Jews. No consistent clinical and hematological abnormalities are associated with

Hb Hasharon. It is innocuous hemoglobinopathy, however some patients have

indicated drug-Induced (sulfonamide, dapsone) hemolytic anemia.

The percentage of Hb Hasharon varies between Ashkenezic Jews and the

subjects of the Ferrara district of Italy. The Hb Hasharon concentration in Italians

of Ferrara district origin is usually in the range of 30-35%. Contrary to this the

Ashkanezic Jews have the Hb Hasharon concentration in the range of 15-20%.

425

The DNA studies have determined that this difference is because of an

underlying α-thalassemia (α-thalassemia-2) in Italians of Ferrara area. Thus,

these individuals have both an alpha chain mutation and an alpha deletion. The

Ashkanezic Jews have no evidence of the presence of α-thalassemia-2 trait.

References

1. Unstable hemoglobin variants, Martin H. Steinberg, MD, www.uptodate.com © 2013 UpTodate. http://www.uptodate.com/contents/unstable-hemoglobin-variants

2. Zur B, Ludwig M, Stoffel-Wagner B. Hemoglobin Hasharon and Hemoglobin NYU in subjects of German origin. Biochemia Medica 2011; 21: 321-25.

3. Chinelato-Fernandes AR, Mendiburu CF, Bonini-Domingos CR. Utilization of different methodologies for the characterization of Hb Hasharon heterozygotes. Genet Mol Res 2006; 5: 1-6.

4. Eliakim R, Rachmilewitz EA. Hemoglobinopathise in Israel. Hemoglobin 1983; 7: 479-85.

5. Mavilio F, Marinucci A, Fontanarosa PP, Tentori L, Cappellozza G.

Hemoglobin Hasharon [α2 47(CD5) Asp→Hisβ2] linked to α-Thalassemia in Northern Italian carriers. Acta Haemat. 1980; 63: 305-311.

6. del Senno L, Bernardi F, Marchetti G, et al. Organization of α globin genes

and mRNA translation in subjects carrying hemoglobin Hasharon (α47 Asp replaced by His) from the Ferrara region (Northern Italy). Eur J Biochem 1980; 111(1): 125-130.

7. Alberti R, Mariuzzi GM, Marinucci M, Bruni E, Tentori L. Hemoglobin Hasharon in a north Italian community. J Med Genet 1975; 12: 294-98.

Case # 21 Hemoglobin Zurich trait

426

A 18 year old white female student from Grand Rapids, Michigan. Parents migrated from Europe, but no information available about their ethnicity and country of origin.Her physical examination was normal.

Laboratory Data:

Hemoglobin 11.6 12.0 – 16.0 g/dL

RBC 4.4 4.0 – 5.5 Mil/mm3

MCV 102.0 79 - 98 fL MCH 28.4 26 - 34 pg RDW 12.3 11.5 -14.5%

Platelet 228 15 - 400 Th/mm3

Hb A 66.0 94.3 - 98.5%

Hb A2 (HPLC) 1.6 1.5 - 3.7% Hb F (HPLC) 0.8 0.0 - 2.0% Hb variant 31.6%

Peripheral Blood Smear: Macrocytic red blood cells. Serum iron and ferritin were normal; very mild anemia with slight reticulocytosis.Sickle cell solubility test for Hb S: Negative Hemoglobin instability (isopropanol) test: Positive No congenital deficiency of glucose-6-phosphate dehydrogenase.

Agarose Gel Electrophoresis (pH 8.6)

Case # 21 Hemoglobin Zurich trait

427

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

428

Case # 21 Hemoglobin Zurich trait

Capillary zone electrophoresis

High performance liquid chromatography

Case # 21 Hemoglobin Zurich trait

429

Interpretation & Discussion

Summary of Results

Method

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band(Hb A)

Major band

Minor band(Hb A2)

Acid Agar

Major band(Hb A +

Hb A2+ Hb

variant)

CZE MajorpeakZone 9(Hb A)

Majorpeak Zone 6

Minor peakZone 3(Hb A2)

IEF Major band(Hb A)

Major bandSlightly cathodal to Hb S

Minor band(Hb A2)

HPLCп Very minorpeak(Hb F)RT=1.06

Major peak(Hb A)RT=2.34

Major

peak*RT=3.55

*The hemoglobin variant eluted with Hb A2, and Hb A2 was in the normal range from CZE.

П Note: HPLC retention time (RT) varies with the type of instrument used and several other factors, e.g. temperature etc. Please note that we do not have data of acid agarose electrophoresis separation at this time.

430

Agarose gel electrophoresis (pH 8.6) showed two major bands; one at

the migration position of Hb A, and one at the Hb S position. In addition,

a very weak band was noticed in the position of Hb C/E/O/Hb A2. Citrate

agar electrophoreis (pH 6.2) showed only one band in the position of Hb A and

a very weak, smudged band in the position of Hb F.

In view of the negative sickle cell test and the migration patterns of the

alkaline and acid electrophoresis, the presence of Hb S was ruled out. Similarly,

the possibility of Hb G-Philadelphia and Hb D- Los Angeles was eliminated on

the basis of the positive (isopropanol) instability test, the absence of G2 band of

Hb G-Philadelphia and a lower percentage of the variant as compared to Hb D-

Los Angeles. Hemoglobin electrophoresis on this patient was performed about

seven years ago in our laboratory. At that time, neither the HPLC nor the CZE

testing instruments were available in our laboratory. After consultation with the

attending physician, the specimen was sent to a reference laboratory for globin

chain analysis by reverse phase HPLC and DNA studies. Note: The IEF, CZE

and HPLC scans inserted here in this case are adopted from other sources for

educational purposes.

The globin chain analysis and DNA studies provided the correct

identification of the hemoglobin variant (≈31.6% from alkaline agarose

electrophoresis) as Hb Zurich. Hb Zurich is an unstable hemoglobin and found

only in the heterozygous state. It is caused by the substitution of histidine by

431

arginine at the 63rd position of the β-chain [α2β263 (E7) His→Arg]. Hb Zurich

was initially found in Europeans of Swiss descent, but later this variant

was reported in Japanese, and Brazilian citizens of non-Swiss ancestry.

Physicians should be advised of possible induced or exacerbated

hemolysis in subjects with Hb Zurich by exposure to oxidant drugs, e.g.

sulfonamides, sulfones, phenacitin-like analgesics, and most of the local

anesthetics.

References

1. Unstable hemoglobin variants. Steinberg MH. www.uptodate.com ©2013 UpTodate. August 2013. http://www.uptodate.com/contents/unstable-hemoglobin-variant

2. Aguinaga MdP, Wright CJ, Roa PD, Terrel F, Turner EA, Houston M. Molecular Diagnosis and Characterization of Hb Zurich [β63(E7)His→Arg] Carriers in a Kentucky Family. Hemoglobin 1998; 22 (5 & 6): 509-515.

3. Harano T, Harano K, Nagasaka I, Yamasaki S. Hb Zurich [β63(E7)His→Arg] Found in a Japanese Woman. Hemoglobin 1996; 20 (4): 429-434.

4. Miranda SRP, Kimura EM, Saad STO, Costa FF. Identification of Hb Zurich

[α2β263(E7)His→Arg] by DNA Analysis in a Brazilian Family. Hemoglobin 1994; 18 (4 & 5): 337-341.

5. Zinkham WH, Winslow RM. Unstable Hemoglobins: Influence of Environment on Phenotypic Expression of a Genetic Disorder. Medicine 1989; 68(5): 309-320.

6. Zinkham WH, Houtchens RA, Caughey WS. Relation between variations in the phenotypic expression of an unstable hemoglobin disorder (hemoglobin Zurich) and carboxyhemoglobin levels. Am J Med 1983; 74: 23-29.

7. Murata K, Yamamoto S, Hirano Y, Omine Mitsuhiro O, Tsuchiya J, Ohba Y, Miyaji T. First Japanese Family with the Unstable Hemoglobin Zurich [β63(E7)His→Arg]. Jap J Med 1982; 21 (1): 40-45.

8. Dickerman JD, Holtzman NA, Zinkman WH. Hemoglobin Zurich. A Third Family Presenting with Hemolytic Reactions to Sulfonamides. Am J Med 1973; 55: 638-642.

432

Case # 22 Hemoglobin Lepore trait

A 41 year old male employee of General Motors, Detroit, Michigan. Parents migrated from Italy.

Laboratory Data:

Hemoglobin 14.3 13.5 -18.5 g/dL

RBC 5.72 4.6 - 6.2 Mil/mm3

MCV 69 80 -100 fLMCH 22.4 27 – 34 pg

RDW 13.2 11.5 -14.5%

Platelet 243 150 - 400 Th/mm3

Hb A 80.7 94.3 - 98.5%

Hb A2 2.1 1.5 - 3.7% Hb F 5.4 0.0 - 2.0%

Hb variant (CZE) 11.8

Peripheral Blood Smear: Microcytosis, hypochromasia, target cells, poikilocytosis.Sickle cell solubility test for Hb S: NegativePatient was not transfused during the past four months.

Agarose Gel Electrophoresis (pH 8.6)

433

Case # 22 Hemoglobin Lepore trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

434

Case # 22 Hemoglobin Lepore trait

Capillary zone electrophoresis

High performance liquid chromatography

435

Case # 22 Hemoglobin Lepore trait

Interpretation & Discussion

Summary of Results

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose Major band(Hb A)

Major band(Hb Lepore)

Minor band(Hb A2)

Acid Agar/Agarose

Major band(Hb A+ Hb Lepore + Hb A2)

CZE Smallpeak(Hb F)Zone 7

Major peak(Hb A)Zone 9

Major peak(Hb Lepore)Zone 6

Minor peak(Hb A2) Zone 3

IEF Major band(Hb A)

Major band in Hb G position (Hb Lepore)

Minor band(Hb A2)

HPLC Smallpeak(Hb F) RT=1.05

Major peak(Hb A)RT=2.45

Major peakHb A2 & Hb Lepore(RT=3.5)

.

Since the sickle cell test was negative and also no band was observed in

the area of Hb S upon acid electrophoresis, the presence of Hb S was ruled out.

The possibility of other commonly encountered variants (e.g. Hb D, Hb G, Hb

436

Russ) that exhibit alkaline and acid electrophoresis pattern similar to this case

were ruled out. Hb D has a concentration of approximately 40% in the

heterozygous state, and this variant quantified at 11.8% by CZE. Hb G was ruled

out due to the absence of the δ-chain variant (αG2 δ2) band/peak by

electrophoretic methods (alkaline agarose, IEF and CZE). This variant was

associated with microcytosis, while Hb Russ is not.

With IEF the hemoglobin variant migrated exactly in the position of Hb G,

even though the presence of Hb G was ruled out on the basis of the absence of

the delta chain Hb G band (see above). Similarly differential diagnosis of the

hemoglobin variant with CZE was not helpful due to the overlap of several

hemoglobins in zone 6.

HPLC showed increased intensity of the Hb A2 peak (RT=3.5), which was

inconsistent with other electrophoretic methods. This suggested that the

hemoglobin variant eluted with Hb A2. Another thing the HPLC ruled out was the

presence of the other common variants exhibiting electrophoretic patterns similar

to this case (Hb D, Hb G, Hb Russ), because none of these variants elute with

Hb A2.

In view of the thalassemic peripheral blood picture, low concentration of

the variant (11.8%), and the separation data a diagnosis of Hb Lepore trait was

made.

437

Hb Lepore is hybrid (fused globin chains) hemoglobin consisting of two

α-globin chains and two δ-β hybrid chains. In the δ-β hybrid chain the N-terminal

position of the δ-chain joins at the C-terminal of the β-chain. There are three

genetically controlled δ-β chains fusions (hybrid formation) that are

characterized by their fusion points of the amino acids in the chains. This

characteristic fusion of δ and β chains leads to the following variants of Hb

Lepore:

i) Hb Lepore-Boston [δ(1-87) β(115-146)]

In this case of Hb Lepore, the hybrid δ-β chain consists of the first 87 amino acids of the δ-chain and the last 32 amino acids of the β-chain. This variant is also called Hb Lepore-Washington.

ii) Hb Lepore-Baltimore [δ(1-50) β(86-146)]

iii) Hb Lepore-Hollandia [δ(1-22) β(50-146)]

Among these three variants, Hb Lepore-Boston is found with some

frequency in people of Mediterranean descent.

Lepore- Boston, (Lepore-Washington) migrates in the same position as

Hb S in alkaline conditions. However Lepore –Hollandia and Lepore –Baltimore

migrate slightly slower than Hb S in alkaline conditions (Bain, BJ, Wild BJ,

Stephens AD and Phelan L. Variant Hemoglobins: A Guide to identification;

2010: Wiley-Blackwell Publishing).

438

To the best of our knowledge both CZE and HPLC do not differentiate

among these variants.

All the Lepore traits have the same clinical symptoms. Hb Lepore trait is a

stable hemoglobin but exhibits features of thalassemia minor due to the

decreased production of β-chains. β-thalassemia intermedia or major type

disorder is exhibited by homozygous Hb Lepore. Hb Lepore interacts with Hb S

to give Hb S/Hb Lepore, however very few cases (<18) were reported in the

literature. Hb S/Hb Lepore patients exhibited mild microcytic anemia, and

clinically were either asymptomatic or with complications generally associated

with Hb S disease. A case of Hb E interaction with Hb Lepore-Hollandia was

described in the literature but without any significant clinical condition except

microcytic anemia without the need for transfusion.

References

1. McKeown SM, Carmichael H, Markowitz RB, Kutlar A, Holley L, Kutlar F. Rare Occurrence of Hb Lepore-Baltimore in African Americans: Molecular Characteristics and Variations of Hb Lepores. Ann Hematol 2009; 88: 545-548.

2. Pasanga J, George E, Nagaratnam M. Haemoglobin Lepore in a Malay family: a case report. Malasian J Pathol 2005; 27(1): 33-37.

3. Shaji RV, Edison ES, Krishamoorthy R, Chandy M, Srivatava A. Hb Lepore in the Indian Population. Hemoglobin 2003; 27: 7-14.

4. Viprakasit V, Pung-Amritt P, Suwanthon L, Clark K, Tanphachitr VS. Complex interactions of [delta] [beta] hybrid haemoglobin (Hb Lepore Hollandia) Hb E([beta]26G>A) and [alpha]+ thalassemia in a Thai Family. Eu J Haematol 2002; 68-107-12.

439

5. Goncalves I, Henriques A, Raimundo A, Picanco I, Reis A, Correia Jr E, Santos E, Nogueria P, Osorio-Almedia L. Fetal Hemoglobin Elevation in Hb Lepore Heterozygotes and its correlation with β Globin Cluster Linked Determinants. Am J Hematol 2002; 69: 95-102.

6. Forget BG. Molecular mechanism of beta-thalassemia. In: Steinberg MH, Forget BG, Higgs DR, Nagel RC, eds. Disorders of Hemoglobin: Genetics, Pathophysiology and Clinical Management. Cambridge, England: Cambridge University Press; 2001: 264-265.

7. Ropero P, Gonzalez FA, Sanchez J, Anguta E, Asenjo S, Arco AD, Murga MJ, Ramos R, Fernandez C, Villegas A. Identification of the Hb Lepore phenotype by HPLC. Haematologica 1999; 84: 1081-1084.

8. Romana M, Diara JP, Merghoub T, Leclard L, Saint-Martin C, Berchel C, Merault G. Hemoglobin Sickle-Lepore: An Unusual Case of Sickle Cell Disease. Acta Haematologica 1997; 98: 170-71.

9. Fairbanks VF, McCormick DJ, Kubik KS, Rezuke WN, Black D, Ochaney MS. Hb S/Hb Lepore with Mild Sickling Symptoms: A Hemoglobin Variant with Mostly Delta-Chain Sequences Ameliorates Sickle-Cell Disease. Am J Hematol 1997; 54: 164-165.

10. Waye JS, Eng B, Patterson M, Chui DHK, Chang LS, Coglonis B, Poon AO, Oliveri NF. Hb E/Hb LeporeHollandia in a Family From Bangladesh. Am J Hematol 1994; 47: 262-265.

11. Hemoglobin Sickle-Lepore: Report: Report of Two Siblings and Review of the Literature. Am J Hematol 1993; 44: 192-195.

440

Case # 23 Hemoglobin J-Oxford trait

A 55 year old male farmer from Saginaw, Michigan. His mother belonged to a French settlement in Newfoundland, Canada. Ancestors from father side immigrated from Norway. No abnormality was found from an annual medical examination except a slight elevation of serum cholesterol.

Laboratory Data

Hemoglobin 14.8 13.5 - 18.5 g/dL

RBC 5.1 4.6 - 6.2 Mil/mm3

MCV 90.7 80 - 100 fL MCH 29.9 27 - 34 pg

Platelet 279 150 - 400 Th/mm3

Hb A2 2.2 1.5 - 3.7% Hb F 0.8 0.0 - 2.0%

Hb A 72.0 94.3 – 98.5%Hb Variant 25.0 %

Peripheral Blood Smear: No abnormalitySickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test: Negative

Agarose Gel Electrophoresis (pH 8.6)

441

Case # 23 Hemoglobin J-Oxford trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

442

Case # 23 Hemoglobin J-Oxford trait

Capillary zone electrophoresis

High performance liquid chromatography

443

Case # 23 Hemoglobin J-Oxford trait

Interpretation & Discussion

Summary of ResultsMethod

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band as much anodal to Hb A as Hb S is cathodal to Hb A(Hb J)

Major band(Hb A)

See note below

Very faint band

(Hb A2)

Acid Agar/ Agarose

Major band(Hb A +

Hb A2+

Hb J)CZE Major

peak(Hb J)Zone 12

Majorpeak(Hb A)Zone 9

Barely visible peak(Hb "J-Oxford-A2") Zone 6

Very minorpeak

(Hb A2)Zone 3

IEF Major band anodal to Hb A (Hb J)

Major band(Hb A)

Very minor band

(Hb A2)

HPLC Very smallpeak(Hb F) RT=1.05

Major peak(Hb A)RT=2.44

Major peak(Hb J)RT=1.62

Barely visiblepeak

(Hb A2)RT=3.64

Note: A faint band due to Hb "A2-J" was detected in the position of Hb S, when a concentrated hemolysate was used for heavier application in alkaline agarose electrophoresis (pH 8.6).

444

In the later stages of our investigation, when Hb J was considered in this

case, the alkaline agarose gel electrophoresis (pH 8.6) was repeated with a

heavier application, and a faint Hb "A2-J" band was detected in the area of Hb S.

The reason for this additional test was to rule out a beta chain variant form of Hb

J. Because the intensity of the fast band is far less than that of the Hb A band

this variant is most likely an α-chain variant.

During my discussion with the attending physician the following points

were brought to his attention:

i) We suspect an Hb J (α-chain variant) trait, probably Hb J-Oxford

trait [α15(A13) Gly→Asp].

ii) There are >50 Hb J variants that are known in the literature. In

addition to that there are > 24 Hb variants which are not designated

as Hb J variant but exhibit electrophoretic mobilities akin to Hb J.

Most of these Hb J variants are entirely without any clinical or

hematological manifestations.

Note: As of today 57 hemoglobins are designated Hb J by

electrophoretic mobility and they are roughly divided equally

between α and β chain variants. Six of these are unstable

and one has increased oxygen affinity.

iii) There are three Hb J variants reported in the literature as

associated with clinical disorders:

445

a) Hb J-Altgeld (unstable hemoglobin hemolytic anemia)

b) Hb J-Cape Town (erythrocytosis due to high oxygen affinity)

c) Hb J-Buda (erythrocytosis resulting from interaction with

Hb G-Pest in persons doubly heterozygous for Hb J-Buda

and Hb G-Pest)

iv) The exact identification of the Hb J trait variant in the patient is not

necessary since the patient is clinically normal. Further testing to

designate the type of Hb J (DNA sequencing, LC-Mass) may be

deferred indefinitely due to exorbitantly associated cost.

At the time this patient was analyzed, both the CZE and HPLC testing

facilities were not available in our laboratory. For instructional purposes, we have

illustrated the CZE and HPLC scans of another established Hb J-Oxford trait

patient. Both the CZE (major peak in zone 12) and HPLC (minor peak at RT=

1.62) provided concurrent evidence for Hb J trait (α-chain variant).

Hb J was first discovered in 1956 (Thorup OA, Itano HH, Wheyby M,

Leavll BS. Hemoglobin J. Science 1956; 123: 889-90), and the 57 variants found

so far are roughly divided equally between α-chain and β-chain variants. Hb J

variants are rarely found, but have been reported from the USA, northern

European countries, China and Japan. The only Hb J variant which is

encountered with any notable frequency, is Hb J-Baltimore (Case # 24).

446

Hb H, J, I, N, K, Camden and Hope are designated "fast hemoglobins" in

view of their faster mobility on agarose gel electrophoresis (pH 8.6). Additional

cases of other "fast hemoglobins" will be included in the 2nd edition of the book.

References

1. Caruso D, Crestani M, Riva LD, Mitro N, Giavarini F, Mozzi R, Franzini C. Mass spectrometry and DNA sequencing are complementary techniques for

characterizing hemoglobin variants: the example of hemoglobin J-Oxford. Haematologica 2004; 89(5): 608-609.

2. Joutousky A, Hadzi-Nesic J, Nardi MA. HPLC Retention Time as a Diagnostic Tool for Hemoglobin Variants and Hemoglobinopathies: A study of 60 000 Samples in a Clinical Diagnostic Laboratory. Clin Chem 2004; 50(10): 1736-1747.

3. Harano K, Harano T, Shibata S, Mori H, Ueda S, Imai K, Ohba Y, Irimajiri K. Hb J-Oxford [α15(A13) Gly---Asp] in Japan. Hemoglobin 1984; 8(2): 197-198.

447

Case # 24 Hemoglobin J-Baltimore trait

A 33 year old male, residing in Windsor, Canada, whose ancestors migrated from Europe. While donating blood, his hemoglobin was found to be low, therefore his family physician ordered hemoglobin electrophoresis.

Laboratory Data

Hemoglobin 12.8 13.5 - 18.5 g/dL

RBC 4.9 4.6 - 6.2 Mil/mm3

MCV 86.0 80 - 100 fL MCH 28.3 27 - 34 pg

Platelet 232 150 - 400 Th/mm3

Hb A2 2.3 1.5 - 3.7% Hb F 3.8 0.0 - 2.0%

Hb A 55.2 94.3 – 98.5%Hb Variant (HPLC) 38.7 %

Peripheral Blood Smear: No abnormalitySickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test: Negative

Agarose Gel Electrophoresis (pH 8.6)

448

Case # 24 Hemoglobin J-Baltimore trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

449

Case # 24 Hemoglobin J-Baltimore trait

Capillary zone electrophoresis

High performance liquid chromatography

450

Case # 24 Hemoglobin J-Baltimore trait

Interpretation & Discussion

Summary of Results

Method

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band as much anodal to Hb A as Hb S is cathodal to Hb A(Hb J)

Major band(Hb A)

Note: Both major bands of equal intensity

Very faint band

(Hb A2)

Acid Agar/ Agarose

Major band(Hb A +

Hb A2+

Hb J)CZE Major

peak(Hb J)Zone 12

Majorpeak(Hb A)Zone 9

Minorpeak

(Hb A2)Zone 3

IEF Major band anodal to Hb A (Hb J)

Major band(Hb A)

Note:Both major bands of equal intensity

Minor band

(Hb A2)

HPLC Smallpeak(Hb F) RT=1.05

Major peak(Hb A)RT=2.4

Major peak(Hb J)RT=1.8

Very smallpeak

(Hb A2)RT=3.6

Note: Since Hb J-Baltimore is a β-chain variant, therefore the major bands (Hb A and Hb J-Baltimore) are approximately in equal concentration.

451

Agarose gel electrophoresis (pH 8.6) showed two major bands. One band

was in the position of Hb A. Another major band was approximately as much

anodal to Hb A as Hb S is cathodal to Hb A. Visually, the intensity of both of

these two bands was similar. No other band was detected besides a minor band

for Hb A2. There are several fast moving hemoglobin variants. Among them are

Hb H and Hb I, which migrate much faster towards anode than Hb J (see Case #

23). Similarly, Hb N also migrates slightly faster than Hb J. There are several α-

chain variants that exhibit similar migration patterns as this patient. These α-

chain variants are usually present in a 1:3 ratio relative to Hb A. Since the

intensity of the two major bands on alkaline agarose electrophoresis was similar,

it was suggestive of a β-chain variant in this case.

Citrate agar electrophoresis (pH 6.2) showed only one major band in the

position of Hb A. IEF also showed two major bands, i.e. one in the position of Hb

A and another more anodal to it in the position of Hb J. The attending physician

was consulted about the clinical condition of the patient. Since no abnormality

was noted by the physician except a slightly lower hemoglobin, a report was

submitted advising the presence of a harmless Hb J-β trait without any

hematological or clinical consequences..

For instructional purposes, we have included the CZE and HPLC scans of

Hb J-Baltimore (β-chain variant), which is the most prevalent β-chain variant

among the category of Hb J variants (both α and β chains).

452

CZE showed a major peak (approximately 40%) in zone 12 (Hb J-

Baltimore), and HPLC also showed a major peak at a retention time of 1.8

minutes (Hb J-Baltimore). We suspect that this case was most likely a

representative of Hb J-Baltimore. Confirmation by DNA studies, globin chain

analysis, and LC-Mass spectrometry would be necessary for definitive diagnosis.

However, for financial reasons, all these additional tests are not required in view

of the benign status of the hemoglobinopathy in the patient.

Hb J-Baltimore (also called J-Trinidad, J-Ireland, J-Georgia) is a β-chain

variant [β 16(A13) Gly→Asp] and is encountered rarely in Afro Americans, and

very rarely in Europeans.

References

1. Arribalzaga K, Ricard MP, Carreno DL, Sanchez J, Gonzalez A, Ropero P,

Villegas A. Hb J-Baltimore [β16(A13)Gly→Asp] Associated with β+-Thalassemia in a Spanish Family. Hemoglobin 1996; 20(1): 79-84.

2. Landin B, Jeppsson J-O. Rare β-Chain Hemoglobin Variants Found in

Swedish Patients During Hb A1c Analysis. Hemoglobin 1993; 17(4): 303-318.3. Vandenesch F, Baklouti F, Francina A, Vianey-Liaud C, Bertrand A, Le

Devehat C, Delaunay J. Hemoglobin J-Baltimore [β16(A13)Gly→Asp]:

Interference with the assay of Hb A1c. Clin Chim Acta 1987; 168(2): 121-28.4. Musumeci S, Schiliro G, Fisher A, Musco A, Marinucci M, Mavilio F,

Fontanarosa PP, Tentori L. Hb J-Baltimore [β16(A13)Gly→Asp] in Association with β-Thalassemia in a Sicilian Family. Hemoglobin 1979; 3(6): 459-464.

453

Case # 25 Hemoglobin Malmo trait

An 18 year old high school student from Warren, Michigan was hurt during football practice and brought to the Emergency Department of the hospital. He had a ruddy face and complained of pain in the lower extremities.

Laboratory Data:

Hemoglobin 20.1 13.5 -18.5 g/dL

RBC 6.8 4.6 - 6.2 Mil/mm3

MCV 88.0 80 - 100 fL MCH 29.7 27 - 34 pg

Platelets 270.0 150 – 400 Th/mm3

Hb A (HPLC) 56.0 94.3 - 98.5%

Hb A2 2.0 1.5 - 3.7% Hb F ≈4.0 0.0 - 2.0%

Hb Variant (HPLC) ≈38%

Peripheral Blood Smear: Crowding of erythrocytes. Sickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test: Negative

Agarose Gel Electrophoresis (pH 8.6)

454

Case # 25 Hemoglobin Malmo trait

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

455

Case # 25 Hemoglobin Malmo trait

Capillary zone electrophoresis

High performance liquid chromatography

456

Case # 25 Hemoglobin Malmo trait

Interpretation & Discussion:

Note: Acid agarose data was not available for Hb Malmo.

Method Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band (Hb A + Hb Variant)

Very faint band

(Hb A2)

Acid Agar Minor Hb F band detected

Major band (Hb A + Hb Variant)

CZE Major peak (Hb A + Hb Malmo)Zone 9

Minor peak(Hb A2)Zone 3

IEF Faint bandinHb F area

Major band slightly anodal to Hb A band

Major band in Hb A position

Faint band inHb A2 area

HPLC* Minor

peak(Hb F)RT=1.03

Major peak (Hb Malmo)RT=1.66

Major peak (Hb A) RT=2.44

Minor peak(Hb A2)RT=3.6

*Note: HPLC retention times varies with the type of the instrument used and several other factors, e.g. temperature, etc.

457

CZE was not available in our laboratory when we encountered this case,

therefore the CZE scan provided here is of another Hb Malmo patient. However

the CZE scan of a Hb Malmo trait was not helpful in the identification of the

variant since both the Hb A and Hb Malmo migrated together in zone 9.

IEF did separate Hb Malmo from Hb A in a pattern identical with a proven

case of Malmo hemoglobinopathy reported in the literature. Hb Malmo migrated

slightly anodal to Hb A, and the migration mobility was much less than some “fast

hemoglobins” (e.g. Hb J-Oxford, Hb J-Baltimore, Hb N, Hb I, etc).

HPLC provided two clearly separated major peaks and three small peaks

of Hb F, Hb A1c and Hb A2. One major peak was due to Hb A (RT= 2.43

minutes), and another with a faster elution time (RT=1.66), was due to the

hemoglobin variant of this case. In summary, Hb Malmo elutes with Hb A or

before, depending on the chromatographic system used. In our system it eluted

before Hb A.

We were aware that occasionally erythrocytosis has been found to be

associated with high-oxygen affinity hemoglobins, but we did not have the

capability to determine a hemoglobin-oxygen dissociation curve and its p50 ( the

point on the curve where the hemoglobin molecule is half-saturated with oxygen).

458

Normally, the hemoglobin-oxygen dissociation curve is sigmoid-shaped.

High affinity hemoglobins, e.g. Hb Malmo, show a markedly leftward shifted

curve (p50 of about 13 torr compared to normal values of 26-30 torr) resulting in

a hyperbolic shape. The oxygen delivery to the tissues is impaired whenever the

oxygen affinity is high (low p50). Erythropoietin production is stimulated, which in

turn increases the red cell mass, resulting in erythrocytosis.

After consultation with the attending physician, a narrative report was

submitted stating that a hemoglobin variant is present and in view of marked

erythrocytosis a possibility of a high affinity hemoglobin cannot be ruled out.

Fortunately, the parents of the patient agreed to provide their blood for

analysis. The mother was found to have a normal CBC and hemoglobin pattern.

The father, who had immigrated from Sweden to USA belonged to a family with

known erythrocytosis. Some years ago, when he complained of fatigue,

headaches and lethargy a diagnosis of Hb Malmo was made in Sweden. In order

to relieve his symptoms, phelebotomy was performed. The electrophoretic

(alkaline, acid, IEF) results and HPLC curve were identical for both the father and

son.

459

In view of the ancestral background and the laboratory results on both the

patient and the parents, a putative diagnosis of Hb Malmo was made. The

attending physician and the family were advised that the hemoglobin disorder

was essentially benign. However, the patient should refrain from smoking and

should be followed periodically for any signs of fatigue, headaches or light-

headedness.

More than 100 high oxygen affinity hemoglobin variants are reported in

the literature. Hb Malmo is a member of this class and is the result of the

substitution of glutamine for a histidine amino acid at the 97th amino acid of the β

chain [β97(FG4)His→Gln]. This mutation is in the area of the peptide chain that

moves during the oxygenation – deoxygenation process. The substitution inhibits

movement in such a manner that deoxygenation becomes more difficult and

deters transfer to the tissue to the point that the patient would become anemic if

the body did not compensate by making excess erythrocytes. The amino acids

from position 94 through 103 constitute a nonhelical section of the beta chain and

mutations effecting ionicity of those positions effects the spacing between the

alpha and beta chains near the point of oxygenation. This area of the globin

chain is called the FG segment or FG corner and a list of these variants is

found in Table 1 (courtesy of Hoyer & Kraft, College of American Pathologists).

460

The fit between the alpha and beta chains is critical because the gap becomes

narrower when oxygen is attached to the ferrous iron and expands as oxygen is

released. A second region of the beta chain (amino acids 143 through 146 on

the Carboxy end of the molecule) also effects this spatial control. Several

hemoglobin variants have been identified as possessing mutations in this area

and thus assisting in understanding the synchronous action involved in the

oxygenation / deoxygenation process. Table 2 is a list of these variants

(also courtesy of Hoyer & Kraft, College of American Pathologists).

In all the cases on this list except for Heathrow and Brigham the

mutation effects the shape of the globin chains such that the electrophoretic

mobility is altered so the mutations are not silent (personal communication, Rita

Ellerbrook, PhD, Helena Laboratories, USA).

Hb Malmo only exists in the heterozygous state. The homozygous state

has not been reported and is thus most probably incompatible with life.

Reference

1. Bain BJ. High-affinity Hemoglobins. In: Hemoglobinopathies Diagnosis, Blackwell Publishing, Oxford, United Kingdom. 2006, 224-226.

2. Steinberg MH. Genetic disorders of hemoglobin oxygen affinity.www.uotodate.com ©2013 UpToDate

3. Fernandez FAG, Villegas A, Ropero P, Carreno MD, Anguita E, Polo M, Pascual A, Henandez A. Hemoglobinopathies with high oxygen affinity. Experience of Erythropathology Cooperative Spanish Group. Ann Hematol 2009; 88: 235-238.

461

4. Giordano PC, Harteveld, Brand A, Willems LNA, Kluin-Nelemans HC, Plug RJ, Batelaan DN, Bernini LF. Hb Malmo[β-97(FG-4) His→Gln] leading to polycythema in a Dutch family. Ann Hematol 1996; 73: 183-188.

5. Landin B, Berglund S, Wallman K. Two Different Mutations in Codon 97 of the β-Globin Gene Cause Hb Malmo in Sweden. Am J Hematol 1996; 51: 32-36.

6. Girino M, Riccardi A, Mosca A, Paleari R, Bonomo P. Double Heterozygosity for Hemoglobin Malmo [β97 (FG4) His→Gln] and β-Thalassemia Traits. Haematologica 1989; 74: 187-90.

7. Boyer SH, Charache S, Fairbanks VF, Maldonado JE, Noyes A, Gayle EE. Hemoglobin Malmo β-97 (FG-4) Histidine→Glutamine: A Cause of Polycythemia. J Clin Invest 1972; 51: 666-676.

462

Table 1. β Chain Variants in FG “Corner” and G Helix

Position Helical # Substitution Name Effect

94 FG1 Asp→His Barcelona polycythemiaAsp→Asn Bunbury normal

95 FG2 Lys→Asn Detroit normalLys→Glu N-Baltimore normal

97 FG4 His→Gln Malmo polycythemiaHis→Leu Wood polycythemia

98 FG5 Val→Met Koln hemolysis, ↑ O2

AffinityVal→Gly Nottingham hemolysisVal→Ala Djelfa (?)

99 G1 Asp→Asn Kempsey polycythemiaAsp→His Yakima polycythemiaAsp→Ala Radcliffe polycythemiaAsp→Tyr Ypsilanti polycythemiaAsp→Gly Hotel Dieu polycythemiaAsp→Val Chemilly polycythemia

100 G2 Pro→Leu Brigham polycythemia

101 G3 Glu→Gly Alberta polycythemiaGlu→Gln Rush hemolysisGlu→Asp Potomac polycythemiaGlu→Lys British polycythemia

Columbia

102 G4 Asn→Lys Richmond normalAsn→Thr Kansas cyanosisAsn→Ser Beth Israel cyanosisAsn→Tyr Saint Mande cyanosis

103 G5 Phe→Leu Heathrow polycythemia

463

Table 2. β Chain Variants Near the C-Terminus

Position Helical # Substitution Name Effect

143 H21 His→Arg Abbruzzo polycythemiaHis→Gln Little Rock polycythemiaHis→Pro Syracuse polycythemiaHis→Asp Rancho (?)

Mirage

His→Tyr Old normal CBC

Dominion ↑ O2 affinity

144 HC1 Lys→Asn Andrew- polycythemiaMinneapolis

Lys→Glu Mito polycythemia

145 HC2 Tyr→His Bethesda polycythemiaTyr→Cys Rainier polycythemiaTyr→Asn Osler polycythemiaTyr→Stop McKees polycythemia

Rocks

146 HC3 His→Asp Hiroshima polycythemiaHis→Pro York polycythemiaHis→Arg Cochin-Port (?)

RoyalHis→Leu Cowtown polycythemiaHis→Gln Kodaira polycythemia

The contents of these tables are presented from Hoyer JD, Kroft SH, eds. Color Atlas of Hemoglobin Disorders: A Compendium Based on Proficiency Testing. Northfield, IL: College of American Pathologists: 2003 (Reproduced with Permission)

Case # 26 Hemoglobin Koln trait464

A 18 year old female student. No ancestral information was available. Physical examination revealed scleral icterus and spleen palpable 4 cm below left costal margin.

Laboratory Data:

Hemoglobin 10.7 12.0 -16.0 g/dL

RBC 3.9 4.0 - 5.5 Mil/mm3

MCV 106 79 - 98 fL RDW 13.6 11.5 -14.5%

Platelet 235 150 – 400 Th/mm3

WBC 7.1 4.0 – 11.0 Th/mm3

Reticulocyte 2.9 0.7 - 1.8 %Serum Iron 39 30 – 160 ug/dLTotal Bilirubin 2.8 0.0 – 1.5 mg/dLIndirect Bilirubin 2.4 0.0 – 0.4 mg/dLHb A (HPLC) 74.8%

Hb A2 (HPLC) ≈2.2 1.5 - 3.7%

Peripheral Blood Smear: Mild macrocytic anemia with slight hypochromasia, polychromasia, and occasional target cells.Sickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test: Positive

Agarose Gel Electrophoresis (pH 8.6)

Case # 26 Hemoglobin Koln trait

465

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

Case # 26 Hemoglobin Koln trait

466

Capillary zone electrophoresis

High performance liquid chromatography

Case # 26 Hemoglobin Koln trait

467

Interpretation & Discussion

Summary of Results

Method

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major bandHb A

Broad smudge from Hb S through Hb C

No discrete

band

detected

Acid Agar / Agarose

Major band(All Hbs)

CZE Major peak(Hb A + Hb Köln)Zone 9

Minor peak(Hb Koln) Zone 6

Minor peak(Hb Koln) Zone 4

Minor peak(Hb A2 + Hb Koln) Zone 3

IEF Major band(Hb A)

Broad smudge of Hb Koln from Hb

A-Hb A2

Barely visible minor band

Hb A2

HPLC#

Major peak(Hb A)RT=2.39

Two barely visible minor peaks (Hb Koln) RT=4.5-4.8

Minor peakHb A2

RT=3.6

Minor peak(Hb Koln)RT=4.95

# Due to the instability of Hb Koln, several minor bands were noticed in HPLC. A similar phenomenon was observed in other electrophoretic methods except in the acid agar/agarose electrophoresis, where only one major band was detected.

Alkaline agarose gel electrophoresis (pH 8.6) showed a major band in the

position of Hb A and a smudged band migrating on both the anodal and cathodal

sides of Hb S extending further towards the area of Hb C. Citrate agar

468

electrophoresis (pH 6.2) showed only one band in the position of Hb A. IEF

showed an intense band in the position of Hb A, and a broad smear between Hb

A and Hb A2 (similar to a fresh painting with a brush). CZE showed a major peak

in zone 9 (Hb A area) and three minor peaks in zones 3, 4 and 6. HPLC showed

a major peak at a retention time (RT) of 2.39 minutes (Hb A), and minor peaks

from a RT of 3.5 – 5.0 minutes.

The laboratory tests indicated that either the hemoglobin variant did

not separate from Hb A, or it formed discrete peaks, or it formed blurred bands.

Since the hemoglobin instability test was positive, the attending physician was

advised of the possibility of an unstable hemoglobin variant.

In August 2013, Professor Steinberg reviewed (reference # 1) the current

clinical and hematological characteristics of unstable hemoglobins. According to

this recent review, approximately 140 of the1028 known mutations of hemoglobin

were found to be unstable. To date, it is not feasible to identify an unstable

hemoglobin variant (either alpha-, beta-, gamma, and delta-globin chains

abnormalities) by the commonly used laboratory methods.

In most worldwide laboratories, Hb E, Hb H, Hb Hasharon and Hb

Koln are the most frequently reported unstable hemoglobin variants. Three of

these variants (Hb E, Hb H, and Hb Hasharon) were excluded in our patient on

the basis of their electrophoretic mobilities and retention times on HPLC. Hb E

(Case # 14a) migrates in the position of Hb A2/E/O/C. Hb H is a fast migrating

469

hemoglobin variant (more anodal to Hb A) on alkaline agarose gel

electrophoresis (pH 8.6). Hb Hasharon (Case # 20) migrates in the Hb S area on

both the alkaline and acid electrophoresis.

Hemoglobin Koln, a prevalent unstable hemoglobin, has a rather

atypical electrophoretic migration pattern, which is helpful in its identification

in conjunction with the associated clinical and hematological manifestations of

the patient.

In Hb Koln, a valine amino acid at the 98th position of the β-chain

(β98Val→Met) is substituted by the amino acid methionine. Since both the valine

and methionine amino acids are neutral amino acids, there would be no net

change in the charge. However Hb Koln does separate from Hb A on agarose

gel electrophoresis (pH 8.6). This anomaly is explained by the modification at the

site of β-chain contact with the heme molecule, which causes a quaternary

structure change of the hemoglobin molecule. As a consequence of this,

especially during alkaline electrophoresis, Hb Koln loses heme groups from the

abnormal β-chain and thus loses negative charge. The hemoglobin migrating in

the Hb S position on alkaline electrophoresis is essentially des-heme hemoglobin

Koln.

The diagnosis of Hb Koln was substantiated by the following observations:

i) Unstable hemoglobinii) Negative sickle cell solubility test for Hb Siii) Minimal or no anemia

470

iv) Splenomegaly and regenerative erythrocyte changes even in the absence of anemia

v) Hypochromasia and macrocytosis are usually evidentvi) Increased oxygen affinityvii) Atypical migration pattern on alkaline agarose gel electrophoresis

(pH 8.6), IEF, CZE and multiple peaks on HPLC.

In view of the above considerations, the physician was advised of the

possibility of Hb Koln. Due to the autosomal dominant mode of familial

transmission, the parents of the patient should also be analyzed for the

confirmation of Hb Koln.

Generally, patients with Hb Koln are asymptomatic unless complications

develop, e.g. blow to the left side of the abdomen by contact sports, exacerbation

of hemolysis during infection (e.g. upper respiratory tract infection), or after use

of some medications, e.g. sulfonamides. Hb Koln exists only in the heterozygous

state; life is not compatible with homozygous Hb Koln.

Hb Koln and Hb Hasharon (Case # 20) are two of the many unstable

hemoglobins that have been identified. Unstable hemoglobins may be suspected

in patients presenting with the symptoms of congenital non-sherocytic anemia

with splenomegaly and pigmented gallstones, or with hemolytic anemia in which

the red cells contain Heinz bodies and are sensitive to oxidant drugs as

sulfonamides, or with mild anemia in which the reticulocyte count is elevated

compared to the amount of hemoglobin, or with a peripheral blood smear

containing target cells, basophilic stippling, a few anisocytes and

hypochromic red cells. If the condition was present in a newborn, the cause

471

might be a gamma chain variant whose effect would be eliminated as beta chain

production increased. Likewise, an asymptomatic newborn, in which clinical

condition began to develop at 4 to 6 months of age might be a beta chain variant.

A suspicious peripheral blood smear would lead to a Heinz body test and

hemoglobin instability (isopropanol) test, but the later two tests should not be run

in the presence of Hb F levels >5%.

References

1. Steinberg MH. Unstable hemoglobin variants. www.uptodate.com © UpToDate. Literature review current through August 2013.

2. Chang YH, Hur M, Lee DS, Park SS, Kim BK, Park S, Ohba Y, Hattori Y, Cho HI. The first case of Hb Koln [β98(FG5)Val→Met] in Korea. Hemoglobin 1999; 23(3): 287-289.

3. Chang J-G, Yang T-Y, Perng L-I, Wang J-C, Tsan K-W. Hb Koln [β98(FG5)Val→Met] : The first case found in a Chinese family. Hemoglobin 1998; 22 (5&6): 535-536.

4. Landin B, Frostad B, Brune M, Ljung R. Haemoglobin Koln as de novo mutations in Sweden: Diagnosis by PCR and specific enzymatic cleavage. Eur J Haematol 1994; 52:156-61.5. Indrak K, Brabec V, Wilson JB, Webber BB, Huisman THJ. Hb Koln or

α2β298(FG5)VAL→Met in a Czechoslovakian family. Hemoglobin 1991; 15(1& 2): 133-135.6. Ohba Y. Unstable hemoglobins. Hemoglobin 1990; 14: 353-388.7. Bird AR, Karabus CD, Hartley PS, Lehman H. Haemoglobin Koln in Cape Town. A case reprt. S Afr Med J 1987; 72: 154-156.8. Gurgey A, Altay C. Hemoglobin Koln [β 98(FG5) Val→Met] in a

Turkish child. The Turkish Journal of Pediatrics 1982; 24: 271-73.9. Ricco G, Ravazzolo R, Rege-Cambrin G, Capaldi A, Trento M, Leechi M, Sartori ML, Furlani C, Rietto GB, Rabino-Massa E. Koln haemoglobinopathy in Italy. Pan. Med 1981; 23:227-233.10. Stirling M. Koln Haemoglobinopathy in a Second Scotish Family. Scott Med J 1980; 25: 121-125.11. Egan EL, Fairbanks VF. Postsplenectomy Erythrocytosis in Hemoglobin Koln Disease. N Eng J Med 1973; 288: 929-931.

12. Hallen J, Charlesworth D, Lehmann H. Haemoglobin Koln in a Jewish

472

Family. Acta Med. Scand. 1972; 191: 177-180.13. Luan Eng L-I, Lopez CG, Eapen JS, Eravelly J, Wiltshire BG, Lehmann H. Unstable Haemoglobin Koln Disease in Members of a Malay Family. J Med Genetics 1972; 9: 340-43.

Case # 27 Hemoglobin Q-India trait473

27 year old male from India. No abnormality detected. Physical examination was unremarkable.

Laboratory Data

Hemoglobin 14.8 13.5 - 18.5 g/dL

RBC 5.4 4.6 - 6.2 Mil/mm3

MCV 86.0 80 - 100 fL MCH 28.3 27 - 34 pg

Platelet 232 150 - 400 Th/mm3

Hb A2 ≈1.1 1.5 - 3.7% Hb F ≈0.8 0.0 - 2.0%

Hb A (HPLC) 78.8 94.3 - 98.5%Hb Variant (HPLC) 19.3 %

Peripheral Blood Smear: No abnormalitySickle cell solubility test for Hb S: NegativeHemoglobin instability (isopropanol) test: Negative

Agarose Gel Electrophoresis (pH 8.6)

Case # 27 Hemoglobin-Q India trait

474

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

475

Case # 27 Hemoglobin-Q India trait

Capillary zone electrophoresis

High performance liquid chromatography

Case # 27 Hemoglobin Q-India trait

Interpretation & Discussion476

Summary of Results

Method

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band(Hb A)

Major band slightly towards Hb A side of Hb S(Hb Q-India)

Faint band(Hb A2)

Acid Agar

Major band between Hb A and Hb S (Hb A + Hb Q-India)

Hb A and Hb Q-India combine as a broader band

Slightly anodal towards Hb S

Acid Agarose

Major band (Hb A + Hb Q-India)

Exactly in the position of Hb A

CZE Major peak(Hb A)Zone 9

Major peak(Q-India) Zone 6

Minor peak(Hb A2) Zone 3

Minor peak (Hb Q-India + Hb A2 )Zone 1

IEF Major band (Hb A)

Major band (Hb Q-India)

Slightly anodal towards Hb A

Faint band (Hb A2)

HPLC Major peak(Hb A)RT=2.39

Major peak(Hb Q-India)RT=4.7

Faint peak (Hb A2)RT=3.6

477

The diagnosis of Hb Q-India trait is cumbersome on alkaline

electrophoresis and may be mistaken for Hb S/D/ Lepore because the band in

the Hb S area is close to Hb S. Sickling or sickle solubility tests should be

employed to rule out Hb S. These and other variants are characterized through

additional laboratory tests (HPLC and IEF). Molecular analysis is required for

establishing a definitive diagnosis. This can be achieved through ARMS-PCR,

RFLP-PCR and mass spectrometry. However, first two of these are fraught with

technical pitfalls and should be done with inclusion of appropriate controls to

acquire correct results. LC/ESI-MS offers a rapid and unambiguous

characterization of individual chains. DNA sequence is currently the most

accurate way of identifying Hb Q-India in a blood sample.

Review of Hemoglobin Q Bushra Moiz, PhD

Introduction

Hb Q was first reported in 1958 in a Chinese patient (1). Since then a number of

cases have been described in Asians. It is a rare hemoglobinopathy resulting from a

single point mutation (GAC→CAC) of α-1 globin gene present on chromosome 16. The

resulting hemoglobin is modified structurally at α polypeptide chain replacing aspartic

acid by histidine. Depending on the implicated codon, three variants have been

described namely Hb Q-India (α 64 Asp→His), Hb Q-Thailand (α 74 Asp→His), and Hb

Q-Iran (α 75 Asp→His). Using computerized models for protein structure, it was

478

observed that there is no difference between the predicted secondary structures of

normal α-globin and that of Hb Q-India (2). In contrast, Hb Q-Iran carries an extra helix

while Hb Q-Thailand carries two extra helices. The predicted results of tertiary structure

also support these findings (2). Since the residue and hence charge changes involve

the surface of the hemoglobin tetramer, the properties of the hemoglobin molecule are

not affected (3).

Hb Q-Thailand [α74(EF3) Asp→His] is often found in Thailand, China, and other

Southeast Asian countries (4). It has several synonyms including Hb Mahidol, Q-

Chinese, G-Taichung, Kurashiki, and Asabara (5). The alpha-Q-Thailand gene is

strongly linked to α gene deletion and has important implications in the identification and

diagnosis of hemoglobinopathies and thalassemias. Subjects with Hb Q-Thailand

invariably show microcytosis as the variant is invariably linked to (-α4.2). However,

individuals who are doubly mutated for Hb Q-Thailand and αo thalassemia may be more

severely anemic (6,7). More complex interaction of Hb Q-Thailand with Hb E, Constant

Spring and hereditary persistence of fetal hemoglobin has been described in the

literature (8-10).

Hb Q-Iran was first described in 1970 by Lorkin et al (11) and later by Rahimi in

an Iranian individual (12).

Hb Q-India was first reported in 1972 by Sukumaran in a Sindhi family (13). Later

reports were published by Dash (14), Abraham (3) and Desai (15); their observations

came from Sindhi and Punjabi families. Hb Q-India usually occurs in the heterozygous

state (αQ-India α/ αα and β/β), however double heterozygotes with both α (-αQ-India / αα

479

and β/β) and β-thalassemia (αQ-India α/ αα and β/βo) were reported (3, 14). A novel

Hb D-Punjab / Hb Q-India was recently reported in an Indian diabetic (16). No report of

a homozygous state has ever been published.

Clinical manifestation

The presence of Hb Q does not impart any functional deficit since the

hemoglobin is not altered structurally at its tertiary level (8). Hb Q is a stable

molecule and has normal oxygen affinity. Therefore, Hb Q is clinically silent

in a heterozygous individual. In contrast, subjects who are compound

heterozygotes with other hemoglobinopathies exhibit a thalassemic phenotype.

For example, co-inheritance of Hb Q-India and β-thalassemia results in a mild

anemia (17). Similarly, Hb Q-H disease caused by the co-inheritance of Hb Q-

Thailand and αo-thalassemia [--/-αo] presents with a chronic, hemolytic anemia

with associated jaundice and splenomegaly (18). The severity of anemia may

warrant blood transfusions and or splenectomy (18). Subjects with a single copy

of Hb Q-Iran do not show any distinctive clinical manifestation (19). Interestingly,

a report from Turkey described a subject with homozygous Hb Q-Iran who was

clinically asymptomatic (20).

Diagnostic laboratory tests and interpretation

CBC and peripheral blood smear

An individual heterozygote with Hb Q-Thailand usually shows a

slight microcytosis. A thalassemic blood picture similar to Hb H disease is 480

observed in Hb Q-H disease (18). The peripheral blood smear in this

disorder showed anisocytosis, poikilocytosis, nucleated red cells and

target cells (6). Intracellular crystals were also observed in red cells in

brilliant cresyl blue preparations (6) similar to Hb H disease.

Hb Q-India usually demonstrates normochromic normocytic red cell

indices with normal or near normal hemoglobin (3). However, a few cases

with mild anemia and microcytosis have been reported in the literature

(10). Hb Q-India with co-inherited β-thalassemic trait usually shows

hypochromic microcytic red cell indices with mild anemia (14,17).

Hb Q-Iran shows normal red cell indices with normal hemoglobin

irrespective of zygosity (12,20).

Alkaline agarose gel, citrate agar and acid agarose electrophoresis

Hb Q migrates in the position of Hb S/D/Lepore on agarose gel

and cellulose acetate electrophoresis (pH 8.6). Presence of αo chains

lead to the appearance of accessory bands corresponding to abnormal

hemoglobin and hence double bands of Hb A2 can be observed (14).

Thus, Hb Q can easily be misinterpreted as Hb S if confirmatory testing

such as sickling or sickle solubility test are not performed. On citrate

agar (pH 6.2), it migrates between Hb A and Hb S. Upon electrophoretic

migration in agarose medium at pH 6.2, Hb Q migrates exactly in the

position of Hb A.

481

HPLC

.The separation of hemoglobin variants depends on their retention

times (21). Hb Q-India, Iran and Thailand exhibited retention times similar

to that of other α-chain variants in the range of 4.76 to 4.78 minutes (3). In

a heterozygote, Hb Q-India and Iran usually eluted as 17-19% of the total

hemoglobin. In contrast, Hb Q-Thailand represented 30-35% of the total

hemoglobin in the heterozygote state, because of the α gene deletion

accompanying it (5). Hb Q % may be further decreased with

concomitant iron deficiency or with co-inheritance of β-thalassemia

trait (14,17).

IEF

Hb Q-India focused in the position of Hb S. However, both the Hb

Q-Iran and Hb Q-Thailand migrated slightly anodal to Hb S, i.e. towards

Hb A.

Mass spectrometry

Liquid chromatography electrospray ionization mass spectrometry

(LC/ESI-MS) has the advantage of providing molecular information in

individual polypeptide chains α and β of the hemoglobin molecule (23).

LC/ESI-MS has been described in recent years to evaluate unknown Hb

variants including Hb Q-India (24). It detects the presence of a mutant α-

chain differing in mass from a normal α-chain by 22 DA. The later is

482

assigned to a mutation of an aspartic acid residue to a histidine residue

thus identifying Hb Q. The site can be identified by tandem-mass analysis

of a tryptic digested fragment encompassing residues αV62-K90 of

hemoglobin α-chain. Sequencing these fragments can establish the

diagnosis of Hb Q-India.

ARMS-PCR

This technique can be used for the successful detection of various

hemoglobin variants including Hb Q-India (3). This technique is based on

the amplification of allele specific primers because of 3’-terminal matches

and mismatches. The methodology is simple, rapid and inexpensive;

however, it is non-specific since either sub-optimal amplification or

deteriorating primers can lead to false positive results (25).

RFLP-PCR

Recently, a restriction enzyme digestion assay was employed for

the diagnosis of Hb Q-India (22). Restriction enzyme EaeI was utilized in

RFLP-PCR since Hb Q-India abolishes the recognition site of this enzyme.

It can be used as a simple, robust and alternative method to ARMS-PCR

for DNA diagnosis of Hb Q-India. However, any other rare variant that

abolishes the same EaeI restriction site would also be detected. Hence,

RFLP-PCR can be used as an adjuvant test after HPLC and or IEF for

primary diagnosis of Hb Q-India.

483

Gene sequencing

This is the most definitive technique for identifying a hemoglobin

variant. Recently, Bhat described DNA sequencing in a patient with Hb Q-

India (26). This methodology of sequence electrophoretogram clearly

demonstrates the specific location of the mutation of Hb Q-India.

Not only did it show that the codon of GAC encoding aspartic acid was

mutated to the codon CAC encoding for histidine, but it also depicted the

zygosity of the patient (26).

References

1. Vella F, Wells RH, Ager JA, Lehmann H. A haemoglobinopathy involving haemoglobin H and a new (Q) haemoglobin. Br Med J 1958; 1: 752-755.

2. Yadav AK. Comparative analysis of protein structure of common Hb Q variants. Indian J Pathol Microbiol 2010; 53: 696-698.

3. Abraham R, Thomas M, Britt R, Fischer C, Old J. Hb Q-India: an uncommon variant diagnosed in three Punjabi patients with diabetes is identified by a novel DNA analysis test. J Clin Pathol 2003; 56: 296-299.

4. Higgs DR, Hunt DM, Drysdale HC, Clegg JB, Pressley L, Weatherall DJ. The genetic basis of Hb Q-H disease. Br J Haematol 1980; 46: 387-400.

5. Hoyer JD, Kroft HS, editors. Color Atlas of Hemoglobin Disorders. A Compendium Based on Proficiency Testing, 159 pp, Northfield, Illinois, College of American Pathologists, 2003.

6. Lieinjo LE, Pillay RP, Thuraisingham V, Further Cases of Hb Q-H disease (Hb Q-alpha thalassemia). Blood 1966, 28: 830-839.

7. Beris P, Huber P, Miescher PA, Wilson JB, Kutlar A, Chen SS, Huisman TH. Hb Q-Thailand –Hb H disease in a Chinese living in Geneva, Switzerland: Characterization of the variant and identification of the two alpha-thalassemic chromosomes. Am J Hematol 1987; 24: 395-400.

8. Sanchaisuriya K, Chunpanich S, Fucharoen S, Fucharoen G, Sanchaisuriya P, Changtrakun Y. Association of Hb Q-Thailand with homozygous Hb E and heterozygous Hb Constant Spring in pregnancy. Eur J Haematol 2005; 74: 221-227.

9. Li D, Liao C, Li J, Xie X, Zhong H. Association of Hb Q-Thailand with heterozygous Hb E in a Chinese patient. Hemoglobin 2008; 32: 319-321.

484

10. Zheng W, Liu Y, Chen D, Rong K, Ge Y, Gong C, Chen H. Complex interaction of Hb Q-Thailand and Hb E with alpha (0)-thalassemia and hereditary persistence of fetal hemoglobin in a Chinese family. Ann Hematol 2010; 89: 883-888.

11. Lorkin PA, Charlesworth D, Lehmann H, Rahbar S, Tuchinda S, Eng Li. Two haemoglobins Q, alpha-74 (EF3) and alpha-75(EF4) aspartic acid to histidine. Br J Haematol 1970; 19: 117-125.

12. Rahimi Z, Aktamipour R, Vaisi-Raygani A, Nagel RL, Muniz A. An Iranian child with Hb Q-Iran [alpha75(EF4)Asp→His]/-alpha3.7 kb/IVSII.1 G→A]: first report. J Pediatr Hematol Oncol 2007; 29: 649-651.

13. Sukumaran PK, Merchant SM, Desai MP, Wiltshire BG, Lehmann H. Haemoglobin Q India [alpha64(E13) aspartic acid→ histidine] associated with beta-thalassemia observed in three Sindhi families. J Med Genet 1972; 9: 436-442.

14. Dash S, Huisman TH. Hemoglobin Q-India [64(E13) Asp→His] and beta thalassemia: a case report from Punjab (North India). Eur J Haematol 1988; 40: 281.

15. Desai DV, Dhanani H, Kapoor AK, Yeluri SV. Hb Q-India in a Sindhi family: an uncommon hemoglobin variant. Lab Hematol 2004; 10: 212-214.

16. Higgins T, Schnabl K, Savoy M, Rowe P, Flamini M, Bananda S. A novel double heterozygous Hb D-Pubjab / Hb Q-India hemoglobinopathy. Clin Biochem 2012; 45: 264-266.

17. Moiz B, Moatter T, Hashmi MR, Hashmi N, Kauser T, Nasir A, Khurshid M. Identification of Hemoglobin Q India (alpha 1-64 Asp→His) through ARMS-PCR. First report from Pakistan. Ann Hematol 2008; 87: 385-389.

18. Leung KF, ma ES, Chan AY, Chan LC. Clinical phenotype of haemoglobin Q-H disease. J Clin Pathol 2004; 57: 81-82.

19. Rahimi Z, Rezei M, Nagel RL, Muniz A. Molecular and hematological analysis of hemoglobin Q-Iran and hemoglobin Setif in Iranian families. Arch Iran Med 2008; 11: 382-386.

20. Ozdag H YI, Akar N. First observation of homozygote Hb Q-Iran [alpha 75(EF4) Asp→His]. Turk J Hematol 2008; 25: 48-50.

21. Joutovsky A, Hadzi-Nesic, Nardi MA. HPLC retention time as a diagnostic tool for hemoglobin variants and hemoglobinopathies: a study of 60 000 samples in a clinical diagnostic laboratory. Clin Chem 2004; 50: 1736-1747.

22. Khalil MS, Henderson S, Schuh A, Hussein MR, Old J. The first use of Eael restriction enzyme in DNA diagnosis of Hb Q-India. Intl J Lab Hematol 2011; 33: 492-497.

23. Wild BJ, Green BN,, Cooper EK, Lalloz MR, Erten S, Stephens AD, Layton DM. Rapid identification of hemoglobin variants by electrospray ionization mass spectrometry. Blood Cells Mol Dis 2001; 27: 691-704.

24. Mandal AK, Bisht S, Bhat VS, Krishnaswamy PR, Balaram P. Electrospray mass spectrometric characterization of hemoglobin Q (Hb Q-India) and a double mutant hemoglobin S/D in clinical sampes. Clin Biochem 2008; 41: 75-81.

485

25. Old JM, Khan SN, Verma I, Fucharoen S, Kleanthous M, Ioannou P, Kotea N, Fisher C, Riazuddin S, Saxena R, Winichagoon P, Kyriacou K, Al-Qudbaili F, Khan B. A multi-center study in order to further define the molecular basis of beta-thalassemia in Thailand, Pakistan, Sri Lanka, Mauritius, Syria, and India, and to develop a simple molecular diagnostic strategy by amplification refractory mutation system-polymerase chain reaction. Hemoglobin 2001; 25: 397-407.

26. Bhat VS, Dewan KK, Krishnaswamy PR, Mandal AK, Balaram P. Characterization of a hemoglobin variant: Hb Q-India / IVS 1-1 [G>T]-beta –thalassemia. Indian J Clin Biochem 2010; 25: 99-104.

Case # 28 Hemoglobin Dhofar trait

486

A 24 year old male, belonging to Qara tribes from the Dhofar region of the Sultanate of Oman. The patient had not been transfused during the past six months.

Laboratory Data: Hemoglobin 13.0 13.5 -18.5 g/dL

RBC 5.1 4.6 - 6.2 Mil/mm3

MCV 68 80 -100 fLMCH 22.4 27 – 34 pg

RDW 13.2 11.5 -14.5%

Platelet 243 150 - 400 Th/mm3

Hb A 81.5 94.3 - 98.5%

Hb A2 (HPLC) 4.1 1.5 - 3.7% Hb F 0.8 0.0 - 2.0%

Hb Dhofar (HPLC) 13.6

Peripheral Blood Smear: Microcytosis, hypochromasia, target cells.Hemoglobin instability (isopropanol) test: Negative Sickle cell solubility test for Hb S: Negative

Note: In HPLC, Hb A2 is slightly under estimated due to overlap with Hb Dhofar peak,

but in heterozygotes , Hb A2 is found to be raised if quantified by capillary zone electrophoresis or elution after cellulose acetate electrophoresis.

Agarose Gel Electrophoresis (pH 8.6)

Case # 28 Hemoglobin Dhofar trait

487

Citrate Agar Electrophoresis (pH 6.2)

Isoelectric focusing

Case # 28 Hemoglobin Dhofar trait

488

Capillary zone electrophoresis

High performance liquid chromatography

Case # 28 Hemoglobin Dhofar trait

489

Interpretation & Discussion

Summary of Results

Method

Hb A area

Hb S area

Hb A2/C area

Alk Agarose

Major band(Hb A)

Medium size band(Hb Dhofar)

Minor band(Hb A2)

Acid Agar

Major band( Hb A + Hb Dhofar)

Hb A and Hb Dhofar combine as a broader band

Acid Agarose

Major band (Hb A + Hb Dhofar)

Major band broadened by Hb Dhofar

CZE Major peak(Hb A)Zone 9

Medium size peak (Hb Dhofar) Zone 5

Minorpeak (Hb A2) Zone 3

IEF Major band (Hb A)

Medium size band (Hb Dhofar)

Minor band (Hb A2)

HPLC* Major peak(Hb A)RT=2.35

Medium size peak (Hb Dhofar) RT=4.04

Minorpeak (Hb A2)RT=3.6

*Note: HPLC retention time (RT) varies with the type of the instrument used and several other factors, e.g. temperature etc.

490

Alkaline agarose gel electrophoresis (pH 8.6) showed a major band in the

position of Hb A and a medium size band in the area of Hb S/G/D/Lepore/Korle

Bu, and a few other variants. Citrate agar electrophoresis (pH 6.2) showed one

major band in the area of Hb A and no other major band was detected. Hb S

was ruled out due to the negative sickle cell solubility test and absence of a band

in the Hb S area on acid electrophoresis. Hb Korle Bu was also ruled out due to

the absence of a band in the Hb G region on IEF. Hb G-Philadelphia was ruled out due

to the absence of a G2 band on alkaline agarose electrophoresis and IEF. Both Hb D

and Hb Lepore migrate in the area of Hb G on IEF, therefore the possibility of these two

variants was also ruled out; Hb Dhofar migrates in the area of Hb S on IEF.

It is emphasized here that the identification of Hb Dhofar by alkaline and

acid electrophoretic methods alone is not prudent. Secondly, the range of Hb

Dhofar (26 – 59% in homozygous and compound heterozygous and 8.8 – 21.5.%

in heterozygous) can be a confounding factor in its differentiation with Hb D-Los

Angeles trait (Case # 12).

HPLC was informative since Hb Dhofar eluted at a retention time slightly

longer than Hb A2 and not in the Hb S window. Here again, the retention time

(4.04 minutes) of Hb Dhofar was in the ‘D’ window, thus not providing conclusive

evidence for its differentiation.

491

CZE scan of the patient indicated that Hb Dhofar peak migrated in

zone 5 (Hb S zone), thus other possibilities (migration peak assigned in zone 6

for Hb D-Los Angeles, Hb G-Philadelphia, Hb Lepore, etc) were ruled out.

Hb Dhofar [β29 (GGC-GGT) Gly-Gly β58 (CCT-CGT) Pro→Arg ] exists

predominantly in the Sultanate of Oman and with a thalassemic phenotype.

References

1. Daar S, Gravell D, Hussein HM, Pathare AV, Wali Y, Krishnamoorthy R. Haematological and clinical features of β-thalassemia associated with Hb Dhofar. Eur J Haematol 2008; 80: 67-70.

2. Williamson D, Brown KP, Langdown JV, Baglin TP. Haemoglobin Dhofar is

linked to the codon 29 C-T(IVSI nt-3) splice mutation which causes beta+ thalassemia. Br J Haematol 1995; 90: 229-31.

3. Marengo-Rowe AJ, Lorkin PA, Gallo E, Lehmann H. Haemoglobin Dhofar- a new variant from Southern Arabia. Biochim Biophys Acta 1968; 168: 58-63.

4. Haemoglobin Dhofar-β58 (Pro→Arg) heterozygote. In: Variant Haemoglobins: A Guide to Identification. Bain BJ, Wild BJ, Stephens AD, Phelan L. pp 188. Wiley-Blackwell, United Kingdom, 2010.

5. Tony S, Daar S, Zachariah N, Wali Y. Prepubertal Hypertransfusion in Thalassemia Intermedia: Sustained Positive Effects on Growth, Splenic Function and Endocrine Parameters. Oman Med J 2012; 27(6). Available from http://wwwomjournal.org/fultext_pdf.aspx?DetailsID=321&type=fultext

6. Qari MH, Wali Y, Albagshi MH, Aishahrani M, Alzahrani A, Alhijji IA, Almomen A, Aljefri A, Al-Saeed HH, Abdullah S, Al-Rustamani A, Mahour K, Mousa SA. Regional consensus opinion for the management of beta thalassemia major in the Arab Gulf Area. Orphanet J Rare Diseases 2013; 8: 143. Available fromhttp://www.ojrd.com/content/8/1/143

492