Biomarker development in translational research and commercialisationcgs.hku.hk/portal/files/GRC/Events/Seminars/2012/2012… · · 2012-11-26Biomarker development in translational

Singapore – 22nd & 23rd November 2012

Biomarker development in translational researchand commercialisation

Michael W. Pfaffl

Qiagen Seminar Tour

Singapore, November 2012

Michael W. PfafflPhysiology, TU München Weihenstephan, 85354 Freising, [email protected]


Quantification of specific mRNAs and microRNAs(cattle, sheep, pig, rat, horse, monkey, buffalo, h umans, …. etc.)

Molecular Physiology - Immunology - Endocrinology:• Immuno-modulation and immuno-stimulation of the gastro-intestinal tract of farm animals (cattle, pig & sheep)• Immunology in Mammary Gland (cattle & sheep)• Growth Physiology & Lactation Physiology (cattle & pig) • Transcriptional Biomarker

mRNA and microRNA quantification assays:• competitive RT-PCR (for mRNAs)• real-time RT-qPCR• absolute and relative quantification strategies• qPCR arrays (96-well and 384-well format)• validity testing for reference genes

RNA integrity (Experion, Bioanalyzer 2100, Nano Drop )• Improvement of RNA and microRNA extraction• Total-RNA , mRNA and microRNA integrity measurement• RIN / RQI algorithm development

Software application development:• Relative Expression Software Tool (REST)• BestKeeper• Efficiency calculation (algorithm development)• Kineret• GenEx – multi dimensional modelling (PCA and HCA)

qPCR Quality management:• MIQE Guidelines• QM management in the molecular biology laboratory

Next generation Sequencing• RNA Seq (mRNA and small-RNA) vs. High throughput RT-qPCR


Analytical challenges for the future

• Sensitive assay – on level of single molecules, single-cell or pathogen

• Quick diagnosis – in hours or even less than an hour

• Early detection of pathogen or micro-organism

• Exact identification of the pathogen or micro-organism

• Wide quantification range

• Minimal amount of biological matrix necessary

• Highly standardized and reproducible

• Stable and valid biomarker(s) or biomarker pattern

⇒ qPCR (RT-qPCR) is the method of choice

⇒ Amplification, analysis, and quantification in one tube

⇒ Highly standardized (MIQE compliant)

⇒ Reliable biomarker pattern = Transcriptional biomar kers


1. The central problems in RT-qPCR

2. Method standardisation (= MIQE guidelines)

3. mRNA & microRNA Expression profiling

4. Biomarker discovery using –omics technologies

5. Biomarker validation using PCA and HCA

Outline


Central problem areas in qPCR

Insufficient information about:

1. Sampling technique & Sample quality

2. Extraction quality & extraction efficiency Pre-PCR steps

3. RNA quality (quantity, integrity, and purity)

4. Experimental design & Type of replicates (=> statistics)

5. qPCR protocol (=> unspecific products)

6. Assay validation (specificity, sensitivity, reproducibility, quantification range, LOD, LOQ, … )

7. Raw data handling (=> Background correction)

8. Cq data handling - Handling of outlier

9. Normalization strategy (one RG, multiple RGs, test for optimal RGs) Post-PCR steps

10. Result interpretation => significance of fold change

11. Statistical test => biological relevance


Sources of variability in quantitative RT-PCR work flow

biological variance technical variability

work flow

varia

bilit

y

hu bt sc rn mm cell-culturesolitaire-cells

• genetic variability (n = 1)• inbreeding status• generation interval• environmental impact

Pre-PCR steps:• sampling technique• tissue storage• NA stabilization• NA extraction• long-time NA storage

RTqPCR

normalization

• tissue type• tissue complexity• heterogeneity <-> homogeneity

hardware

model system


⇒ MIQE is a set of guidelines that describe the minim um information necessary for evaluating qPCR experiments

⇒ The goal is:• To provide guidelines for authors, reviewers and editors to measure the technical quality

of submitted manuscripts• To establish a clear framework to conduct quantitative RT-PCR experiments• To support experimental transparency• To increase reproducibility between laboratories worldwide• To promote more consistent, more comparable, and more reliable results• To standardize international qPCR nomenclature

⇒ To increase reliability of results to help to insur e the integrity of scientific work, with major focus on biological relevance


The MIQE checklist:

• 9 headings

• 85 sub-titles

• 57 essential information

• 28 desirable information

• Experimental design

• Sampling

• NA extraction

• RT

• qPCR target information

• qPCR primers

• qPCR protocol

• qPCR assay validation

• Data analysis & statistics

Download iOS, Android and html5 MIQE APPs on MIQE.Gene-Quantification.info


Definitions of „Biomarker“

• Biological markers (biomarkers) have been defined as “cellular, biochemical or molecular alterations that are measurable in biological media such as human tissues, cells, or body fluids” (Hulka, 1990)

• “A biomarker is a characteristic that is objectively measured and evaluated as an indicator of normal biologic or pathogenic processes, or pharmacologic responses to a therapeutic / drug intervention”(definition by “Biomarkers definitions working group” of NIH, 2001)

• More recently the definition has been broadened to include more biological characteristics that can be objectively measured and evaluated as a biological indicator (Mayeux, 2004)

• The application of integrative functional informatics represents the novel direction in such biomarker technology integration (Ilyin et al., 2004)

• Integrative functional informatics = the interfacing and integration of different technologies for data collection and their analysis are pivotal to biomarker identification, characterization, validation, and successful usage.

What is a biomarker ?


DNA => pre-mRNA => mRNA => protein => function↑↑↑↑ ↑↑↑↑ ↑↑↑↑ ↑↑↑↑

methylation transcription splicing translation

↓↓↓↓ ↓↓↓↓ ↓↓↓↓ post-translatory modificationsDNA => pri-miR => pre-miR => miR

genome transcriptome proteome metabolomeepi-genome small-RNA transcriptome

splicome

Biomarkers on various “- ome” layers


Food Quality & Food Safety ?

Food SafetyProtecting the food supply from microbial, chemical, and physical hazards or contamination that may occur during all stages of food production: growing, harvesting, processing, transporting, preparing, distributing, storing, and handling.The goal of food safety monitoring is to keep food wholesome, safe to eat and free of disease causing agents such as:

– infectious agents– toxic chemicals– additives: antibiotics, preservatives, drugs, hormones, …– foreign objects

Food QualityFood quality is the quality characteristics of food that is acceptable to consumers. This includes food appearance (size, shape, colour, gloss, and consistency),texture, flavour, factors such as federal grade standards, and ingredients (chemical, physical, microbial, secondary plant components, pre- or probiotics, synbiotics, …)

Z I E L - Research Center for Nutrition and Food Sc iences


Gynecomastia: development of abnormally large mammary glands

Estrogens => development of primary and secondary sexual characteristics

Prolactin => breast development, milk production


Doping ! ! !


Classical analytical methods fordrug/hormone residue analysis

Use of specific anabolic agents in agriculture is licensed in Canada, USA, Australia, South Africa ... Since 1988 the use of anabolic agents is prohibited in the EU !

Anabolica misuse => sensitive screening and test sy stem are established in EU

– RIA

– ELISA (EIA)

– LC-MS/MS

– GC-MS

– GC-MS/MS

⇒ analysis of specific compounds

⇒ analysis of known anabolics


Aim is the physiological way!

Establishment of a new hormone screening method

Screening of physiological response- in various hormone sensitive tissues- in easy accessible tissues, e.g. body liquids, blood, and hair roots

Regulative impact of hormones on transcriptome

Quantitative RT-PCR => “most sensitive” method(=> 1 molecule detection limit)

Evaluation and analysis of regulated genes mRNA & microRNA transcript candidates

Candidate verification - via statistical methods, e.g. HCA or PCA- via independent methods (NGS)

Specification of “Expressed Biomarkers”

Identification of “Gene Networks”The physiological way: monitoring RNA expression changes as new approach to combat illegal growth promoter application.Riedmaier , Pfaffl & Meyer Drug Test Analysis 2012


Why use transcriptional biomarkers ?

Gene expression biomarker on mRNA or microRNA level :• Activated steroid hormone receptors act as transcription factors:

=> direct influence on gene expression activity• Expression changes are relatively fast – within hours• Various tissues are easily available for biomarker screening:

– body fluids: blood, urine, saliva, sputum, etc ...– non-invasive sampling: hair roots, buccal or nasal swabs & vaginal smear containing

epithelial cells, ...

• Very sensitive RT-qPCR is validated in our lab (single-cell RT-qPCR)

What is the candidate gene approach?• Selection of target genes by screening the actual literature for steroidal effects

in tissues– Published tissues: with high steroid receptor abundance, primary and secondary

sexual organs– Published candidate gene groups: steroid hormone receptors, growth factors,

apoptosis factors, inflammatory factors, transcription factors, ...


Cellular steroid hormone action

Riedmaier I, Becker C, Pfaffl MW, Meyer HHD. - ReviewThe use of -omic technologies for biomarker development to trace functions of anabolic agents. In Journal of Chromatography A, 2009

mRNA

microRNA

cell


Functional target gene groups

• steroid receptors

• steroid hormone metabolism

• proliferation / apoptosis

• angiogenesis

• tissue remodelling

• immune relevant factors


Work flow (1) TBA+E2


RNA extraction RNA integrity control RT-qPCR

∆∆ Cq

data analysisstatistical evaluation

Work flow (2)

gene regulation


mRNA expression profiling resultsfrom heifers ovarian samples

Becker C. et al., HMBCI 2010


microRNA expression profiling resultsmicroRNA qPCR array – 2x 384 miR set

miRNA regulation in bovine liver under the influence of TBA+E2up-regulated genes

miR 378: promotes cell survival, tumor growth, angiogenesis

miR 412: regulates uterine leiomyomas growth, benign uterine smooth muscle tumor

miR 15a: tumorgenesis, unarrested cell cycle


microRNA expression profiling resultsmicroRNA qPCR array – 2x 384 miR set

miRNA regulation in bovine liver under the influence of TBA+E2

down-regulated genes

miR 130a: regulatory functions in the endometriuminduced by steroids

miR 34a: functions as a potential tumor suppressor by inducing apoptosis


� �

15a 0.49 0.023 *

20a 0.49 0.024 *

27b 0.97 0.58 n.s.

29c 1.30 0.066 n.s.

34a 0.67 0.017 *

103 1.30 0.040 *

106a 1.40 0.057 n.s.

138 1.74 0.2 n.s.

181c 0.75 0.038 *

320d 0.72 0.065 n.s.

433 1.15 0.89 n.s.

miR- x-fold regulation (2-∆∆Cq) p-value significance

microRNA expression profilingsingle assays (Qiagen) vs. microRNA qPCR array (Exiqon)

P < 0.05

P < 0.1 (trend of regulation)

single assay PCR

0 1 2 3 4 5 6

PC

R a

rray

0

1

2

3

4

5

correlated regulation single microRNA assay vs. qPCR array

r = 0.662p < 0.001



WBC expression data analysismRNA data

Significantly regulated genes in heiferblood (16 days after treatment)

PCA & HCA


Vaginal epithelial cell expression data analysis

PCA

HCA


Combined WBC & vaginal epithelial cell data analysi s

PCA & HCA

=> integrative functional informatics


Group I control animals (n=2)

Group II treated animals (n=3)

sequencing

data analysis

RNA isolation

TU München

EMBL Gene-Core

Genomatix

Alteration of gene regulation in bovine liver by treatment with anabolic steroids

* alignment / mapping * differential expression analysis

Confirmational study Finding new biomarker candidates

Experimental setup

NGS =>


NGS results - sequence statistics

• number of sequences: > 25 mio.

• sequence length: 76 bases

• successfully mapped bovine genes: 16,947

• successfully mapped transcripts: 45,832

• significantly regulated genes: ~ 9,000 (control vs. treatment group)

• gene regulation at least 2-fold: ~ 300

⇒ screening for significantly regulated genes

⇒ with high relative expression change, more than 2-fold

⇒ and high expression levels


Verification of 20 regulated genes via RT-qPCR (out of 40 selected)

no. acc. no. gene name x-fold NGS sign NGS x-fold qPCR p- value PCR

1 NM_174530 CYP2E1 0,02 1 0,38 0,015

2 NM_001083444 FADS2 0,09 1 0,37 0,004

3 NM_001046551 CXCL10 0,13 1 0,53 0,028

4 NM_001046269 PON1 0,15 1 0,28 0,001

5 NM_001082610 SOD3 0,16 1 0,40 0,007

6 NM_001080354 GSTM1 0,22 1 0,57 0,022

7 NM_173941 MX2 0,28 1 0,47 0,019

8 NM_001037480 APOA4 0,28 1 0,19 0,008

9 NM_174366 ISG15 0,33 1 0,54 0,058

10 NM_001014391 CLDN4 6,08 1 4,52 0,007

11 NM_001102326 SECTM1 6,42 1 3,66 0,037

12 NM_001077933 AK3L1 6,62 1 13,66 0,003

13 NM_173917 HBB 7,04 1 5,57 0,023

14 NM_001104967 PRODH2 7,79 1 3,59 0,000

15 NM_001078026 TNC 8,06 1 8,87 0,000

16 NM_001076400 GDPD1 8,39 1 4,18 0,001

17 NM_174097 HOPX 10,77 1 8,18 0,038

18 NM_001102054 SERPINI2 26,18 1 82,36 0,012

19 XM_602816 CUX2 26,76 1 6,33 0,000

20 NM_174307 EPYC 0,12 1 6,69 0,036

Up-regulated Down-regulateddown-regulated up-regulated


Blue diamonds: control animals

Red triangles: treated animals

PCA - qPCR dataobtained from 11 significantly regulated genes chosen by literature

Becker C. et al, Horm Mol Biol Clin Invest, 2010


PCA and HCA - qPCR dataobtained from 20 significantly regulated genes chosen by NGS

Blue diamonds = control animals

Red triangles = treated animals

Riedmaier et al. Anal Chem (2012)


All 9 untreated control animals vs. one treated animal

no. 5

treated animal no. 5 treated animal no. 8

no. 8


Blue triangles: control animals

Pink diamonds: 1x treatment

Red crosses: 3x treatment

Hormone treatment via „Pour-On Mix“ (collaboration with RIKILT)Estradiolbenzoate, Testosteronedecanoate, Testosteronecypionate

Calves treated with hormone cocktail

PCA with 4 significantly regulated genesCLDN4, CUX2,

SOD3, PON1


Boars treated with Synovex Plus200mg Trenbolon Azetat plus 28mg Östradiol Benzoat

(sample from an EU animal trial from 1998)

Green = control animals

Red = treated animals


BiblioSphere

Cocitation based network of genes regulated at least 2 fold between control and treatment. (and min NE of 0.5 in at least one condition)

4.7

-6.8

0.0

Genes from input list are colored based on ratio between control and treatment. Red – induced , blue –repressed with treatment.

White boxes represent transcription factors that are cocited with at least 10 input genes.Green lines indicate potential binding sites in the promoter region.

Gene network regulated by exogenous anabolics


Conclusion

• Tissues that are directly influenced by steroid hormones with abundant receptor concentrations are more sensitive and show higher gene regulations => hence better for the biomarker identification

• transcriptomics analysis discovered promising expression changes– mRNA served as first biomarker candidates– additional candidates via microRNA profiling– verification of existing candidates via NGS RNA-Seq– new mRNA candidates via NGS RNA-Seq

Integrative biomarker discovery approach on transcr iptome level:– significantly regulated transcripts (mRNA and microRNA)

– in various tissues (WBC, liver, and vaginal smear)

– by different analysis methods (qRT-PCR, qPCR array, and NGS RNA-Seq)

– by various statistical algorithms (PCA and HCA)

⇒ The more integrative biomarker candidates are disco vered the better a prediction of hormone treatment is pos sible


Perspectives

To verify our first biomarker candidates:• Further trials with more animals, different breeds, and various

physiological conditions are needed• Large number of untreated control animals = stable control pool• A panel of anabolic agents / cocktails is needed• Combination of transcriptomics with other “-omics“ technologies, like

proteomics or metabolomics => steroid biology and system biology

New studies in progress:• DFG

• small-RNA profiling

• BISP


Next Generation Thinking in Molecular Diagnostics6th international qPCR & NGS Event - Symposium & Industrial Exhibition & Application Workshops

18th – 22nd March 2013, in Freising-Weihenstephan,Technical University of Munich, Physiology - Weihenstephan, Bavaria, Germany

Symposium Talk and Poster sessions:• Main Topic: Molecular diagnostics• Main Topic: Next Generation Sequencing (NGS)• Main Topic: Transcriptional Biomarkers• High throughput analysis in qPCR• Systems biology• Single-cells diagnostics• MIQE & QM strategies in qPCR • non-coding RNAs - microRNA, siRNA and long non-coding RNAs • Digital PCR & Nano-fluidics • Pre-analytical Steps • qPCR data analysis - BioStatistics & BioInformatics• NGS data analysis - BioStatistics & BioInformatics• Lunch Seminars:

• qBASEplus - data analysis lunch seminar• GenEx - data analysis lunch seminar• NGS data analysis lunch seminars• more to be announced........

symposium.qPCR-NGS-2013.net


Acknowledgements

Documents

Biomarker development in translational research and commercialisationcgs.hku.hk/portal/files/GRC/Events/Seminars/2012/2012… · · 2012-11-26Biomarker development in translational