Mycolic Acid Analysis by High-Performance Liquid Chromatography for Identification of

CLINICAL MICROBIOLOGY REVIEWS,0893-8512/01/$04.00�0 DOI: 10.1128/CMR.14.4.704–726.2001

Oct. 2001, p. 704–726 Vol. 14, No. 4

Copyright © 2001, American Society for Microbiology. All Rights Reserved.

Mycolic Acid Analysis by High-Performance Liquid Chromatographyfor Identification of Mycobacterium Species

W. RAY BUTLER1* AND LINDA S. GUTHERTZ2

Division of AIDS, STD and TB Laboratory Research, National Center for Infectious Diseases, Centers forDisease Control and Prevention, Atlanta, Georgia 30333,1 and Microbial Diseases Laboratory,

California Department of Health Services, Berkeley, California 94704-10222

INTRODUCTION .......................................................................................................................................................704LABORATORY............................................................................................................................................................705MYCOLIC ACIDS ......................................................................................................................................................707HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY..................................................................................709APPLICATION OF HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY FOR

IDENTIFICATION OF MYCOBACTERIA.....................................................................................................709STANDARD METHOD FOR PREPARATION OF MYCOLIC ACIDS .............................................................711DIFFERENTIATION OF MYCOBACTERIUM SPECIES BY UV–HIGH–PERFORMANCE

LIQUID CHROMATOGRAPHY ......................................................................................................................711COMPUTER USE WITH UV–HIGH–PERFORMANCE LIQUID CHROMATOGRAPHY............................716MYCOLIC ACID PATTERN RECOGNITION AND MYCOBACTERIUM SPECIES

IDENTIFICATION .............................................................................................................................................717FLUORESCENT DERIVATIVES FOR MYCOLIC ACID ANALYSIS ...............................................................719UV–HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY AND MOLECULAR METHODS................720INSTRUMENT CERTIFICATION AND INSPECTIONS ....................................................................................721HPLC USERS GROUP..............................................................................................................................................721CONCLUSION............................................................................................................................................................723REFERENCES ............................................................................................................................................................723

INTRODUCTION

Early strategies employed to identify species of the Myco-bacterium genus have included observations of staining prop-erties of bacilli, cultural morphology, biochemical tests, and,rarely, the inoculation of susceptible animals with live bacillifor observation of animal pathogenicity. These tests were de-signed to discriminate among mycobacteria involved in diseaseand were directed toward detecting Mycobacterium tuberculo-sis, Mycobacterium bovis, Mycobacterium avium, or saprophytes(57–59). When other Mycobacterium species were recognizedas infectious agents, it was obvious that additional differenti-ation criteria were needed. A classification system based onpigmentation and growth rate was introduced to define theoccurrence of “atypical” (a term presently in disfavor in my-cobacterial nomenclature) strains and their relationship to theMycobacterium species perceived as pathogenic (80, 95). The“slow growers” were defined as having visible growth in �7days and were categorized in the following groups: group I,photochromogens; group II, scotochromogens; and groupIII, nonphotochromogens. “Rapid growers” were defined ashaving visible growth in �7 days, and they were designatedgroup IV. Although this simple system is not used as exten-sively now, its longevity is demonstrated by references in

publications and frequent communications between myco-bacteriologists. However, these simple designations are notpractical or sufficient for defining species within the Myco-bacterium genus.

In related efforts, members of the International WorkingGroup on Mycobacterial Taxonomy (IWGMT) made signifi-cant contributions to mycobacterial identification and taxonomy.Their collaborative studies evaluated various groups of mycobac-teria and related genera, defined variation in members of a givenspecies, and proposed selected tests for routine species identifi-cation. These extensive studies, involving numeric taxonomy, clar-ified the phenetic integrity of the Mycobacterium genus (42, 106)and provided a practical biochemical identification scheme forclinically important species of Mycobacterium (107, 108). Dur-ing this time, just over 20 of the known 54 species were re-garded as potentially pathogenic, and the recommended testsappeared applicable for identification of these species (55).Eventually, as the number of species increased, the resultingtaxonomical complexity caused ambiguities in the interpretationof biochemical test results due to their reduced discriminationability.

Over the years, the unequivocal identification of M. tuber-culosis and species of clinical interest has dominated the tax-onomy of the mycobacteria (106, 111). This emphasis on iden-tification of the most commonly recovered species by clinicallaboratories prompted a development of genetic probes (Gen-Probe, San Diego, Calif.) for their detection. Laboratories us-ing these genetic tests rarely misidentified species for which thetests were designed. However, tests were not developed for the

* Corresponding author. Mailing address: M/S F08, Centers for Dis-ease Control and Prevention, 1600 Clifton Rd., N.E., Atlanta, GA30333. Phone: (404) 639-2414. Fax: (404) 639-1287. E-mail: [email protected].

704

on April 3, 2019 by guest

http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

majority of the mycobacteria, because they were not consid-ered a serious threat to the public health, inasmuch as themajority of the species were infrequently isolated and werenever transmitted from person to person. On occasion, thesespecies produced serious (even fatal) infections, especially inpatients with a reduced immune response. In these cases, rapidand accurate identification of the clinical species can be ben-eficial for effective medical intervention.

In this report, we will examine the development, introduc-tion, and effectiveness of high-performance liquid chromatog-raphy (HPLC) analysis of the mycolic acids for chemotaxono-mic classification and rapid identification of Mycobacteriumspecies. (For a historical overview of liquid chromatographyleading to the development of HPLC, the reader is referred toreference 49.) It is not within the scope of this review toexamine every method proposed or currently used for theidentification of the mycobacteria.

(Some of the chromatograms shown in this review werepresented at the 96th General Meeting of the American Soci-ety for Microbiology by L. S. Guthertz, P. S. Duffey, and W. R.Butler.)

LABORATORY

HPLC principles and instrumentation originated in the fieldof chemistry during the mid-1960s and found widespread ap-plication in inorganic, organic, and biochemical areas (49).Clinical chemistry sections exploited the versatility of the meth-od for rapid separation and identification of compounds fordrug analysis (26, 38, 49). In 1985, scientists at the Centers forDisease Control and Prevention (CDC) proposed the use ofHPLC as an aid in mycobacterial classification (10). In 1989,the procedure was incorporated into the regiment of tests atthe CDC Mycobacteriology Reference Laboratory, and in1990, it was offered as a standard test for identification ofMycobacterium species (40).

HPLC analysis of mycolic acids for species identificationoffered a reprieve from the traditional time-consuming identi-fication process (55, 84). An isolate submitted on culture could

FIG. 1. Representative HPLC system with dual pumps for gradient elution of solvents.

FIG. 2. Flowchart for isolation and detection of mycolic acids.

705


http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

FIG. 3. Characteristic UV-HPLC chromatograms of Mycobacterium species with late-emerging, simple, single-cluster peak patterns. (A)M. asiaticum ATCC 25276T; (B) M. bovis BCG Pasteur; (C) M. gastri ATCC 15754T; (D) M. gordonae ATCC 14470T, UV-HPLC chromotype I;(E) M. kansasii, ATCC 12478T; (F) Mycobacterium leprae, ‘armadillo’; (G) M. tuberculosis complex (includes M. africanum ATCC 25420T, M. bovisATCC 19210T, “M. canettii” NZM 217/94, M. caprae CIP 105776T, M. microti ATCC 19422T, and M. tuberculosis ATCC 27294T); (H) M. szulgaiATCC 35799T.

706 BUTLER AND GUTHERTZ CLIN. MICROBIOL. REV.


http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

be analyzed in hours by HPLC, compared to weeks or longerfor routine methods. However, the suitability of the methodshould be evaluated prior to incorporation into the laboratory.HPLC is considered a sophisticated procedure compared toother laboratory methods, and a dedicated, highly trained op-erator is required (W. Gross, personal communication). Instru-mentation is costly compared to conventional methods. Labo-ratories that detect acid-fast bacilli only by smears may nothave the capacity to develop a proficiency with the HPLC meth-od. However, many high-throughput laboratories that are pro-ficient in all aspects of mycobacteriology have successfully in-corporated HPLC into their methodologies (14, 21, 29, 45, 52,93, 99). A significant challenge for laboratory personnel liesin developing expertise in the visual interpretation of chro-matographic patterns for species determination. In this report,63 chromatographic patterns representative of 73 known my-cobacteria species are presented and may help with this diffi-culty.

MYCOLIC ACIDS

The analysis of lipid fractions has contributed significantly tothe knowledge of Mycobacterium species. The abundance oflipid constituents in mycobacterial cell walls made them classiccandidates for early chemical investigations. Of intense inter-est to researchers was a difficult-to-purify wax fraction, initiallytermed “unsaponifiable wax” (90). Isolated after prolongedsaponification, it was clarified as an alcohol-insoluble, hydroxyacid of high molecular weight. The saponified wax fraction waschemically stable and was named “mycolic acid” following itsisolation from the original H37 strain of M. tuberculosis. Com-plete saponification required 80 h of reflux in methanolic po-tassium hydroxide and benzene. Chemically, the mycolic acidfraction stained acid fast, contained hydroxy and methoxygroups with approximately 88-carbon chain lengths, and dis-played a pyrolysis product with 26 carbons when heated at 300to 350°C (90). Further structural studies confirmed that thehydroxy group of mycolic acid was in a �-position to the car-boxyl group and thus defined the mycolic acids as high-molec-ular-weight �-hydroxy fatty acids with a long �-side chain (2).Additional studies demonstrated that the structural character

of the mycolic acids was a mycolic acid-arabinogalactan-muco-peptide complex, an essential part of the cell wall whose abun-dance was not adversely affected by culture conditions (53, 61).This consistent production of mycolic acids was noted to be astable phenotypic property for a species (31, 42, 54, 69).

The diagnostic value of the mycolic acids was quickly recog-nized by early chemical investigators who characterize the dif-ferent mycolic acid-containing genera (1, 32, 61). The chemicalcomplexity of members of the different taxa was demonstratedby gas-liquid chromatography, and single-dimension, thin-layerchromatography (1D-TLC) analysis of mycolic acid as methylesters (32, 60). Methanolysates of mycobacterial mycolic acidswere differentiated in 1D-TLC by production of multispot pat-terns, in contrast to single-spot patterns produced by species inrelated taxa (65, 69). Mycobacterium species were also chemi-cally distinguished from those of other genera by the presenceof C22 to C26 products from gas-liquid chromatography py-rolysis of mycolic acid methyl esters (32, 61, 69). Mycobacte-rium species were rapidly distinguished from Nocardia orRhodococcus species by precipitation of mycobacterial mycolicacids from a solution of ether by the addition of alcohol,whereas the mycolic acids from the other genera remainedsoluble (54).

In addition, it was shown that mycobacteria and other my-colic acid-containing taxa could be distinguished using 2D-TLC of mycolic esters as whole-organism methanolysates (61,65). This method was combined with mass spectrometry (MS)and used to identify and chemically characterize the functionalgroups of the mycolic acids, thus demonstrating the species-specific nature of the mycolates. Extensive studies describedmycobacterial mycolic acids as phenotypically stable chemicalcharacters, having a discontinuous distribution with chemotax-onomic potential (68, 69). Because mycolic acids were confinedin related taxa and not widely distributed in nature, they wereideal for discrimination studies (70). The impressive work donewith 2D-TLC and MS advanced the knowledge of the structureand complexity of mycobacterial mycolic acids. Analysis of 50species of mycobacteria by 2D-TLC revealed that the myco-bacteria were separated into 11 structural groups based on thecomposition of their mycolic acids, and the analysis was sug-gested as applicable in clinical laboratories for rapid identifi-

FIG. 3—Continued.

VOL. 14, 2001 HPLC FOR MYCOLIC ACID ANALYSIS IN MYCOBACTERIA 707


http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

FIG. 4. Characteristic UV-HPLC chromatograms of Mycobacterium species with late-emerging, complex, single-cluster peak patterns. (A)Mycobacterium confluentis ATCC 49920T; (B) M. haemophilum ATCC 29548T; (C) M. intermedium ATCC 51848T; (D) M. malmoense ATCC29571T; (E) M. marinum CDC 875; (F) M. thermoresistibile CDC 7305; (G) M. ulcerans ATCC 19423T.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

cation of mycobacteria (69). These comprehensive studiesdefined the value of mycolic acid data in mycobacterial sys-tematics and led to the recommendation that minimal descrip-tions for a new species of Mycobacterium include informationon the chemical composition of mycolic acids (62). Presently,all the mycolic acid-containing genera can be distinguished bythe length of their mycolic acid carbon chains. They includeCorynebacterium, C20 to C38; Dietzia, C34 to C38; Rhodococcus,C34 to C52; Nocardia, C40 to C60; Gordonia, C48 to C66; Wil-liamsia, C50 to C56; Skermania, C50 to C64; Tsukamurella, C64

to C78; and Mycobacterium, C60 to C90.

HIGH-PERFORMANCE LIQUIDCHROMATOGRAPHY

In the mid to late 1970s, the newly developed method wasreferred to as high-pressure liquid chromatography. The pres-sure was produced by pumps which forced a sample suspendedin a liquid through a solid stationary phase, enclosed in a metalcolumn, resulting in the rapid fractionation of sample compo-nents. The separated components were distinguished by a de-tector and depicted as peaks by a recorder (Fig. 1). The peakswere defined by specific elution, emergence, or retentiontimes. Component fractionation was shown to be a complexintermixture of chemical reactions and physical factors oc-curring between the sample, the mobile phase, and the solidstationary phase. (For a review of the basic technique ofliquid chromatography, the reader is referred to reference113.)

HPLC separation methods permitted researchers interestedin the biosynthesis and biological properties of the mycolicacids to fractionate and collect “pure” compounds and con-firmed earlier TLC and nuclear magnetic resonance data ofchemical structure (73). The sensitivity of mycolic acid detec-tion was increased by derivatization to UV-adsorbing p-bro-mophenacyl esters (PBPA), (73, 89). Subsequent studies usedmultiple step methods with TLC, MS, and reverse-phase chro-matography to achieve fractionation of discrete componentsof the mycolic acids (89, 92). These reverse-phase chromatog-raphy procedures utilized high-efficiency C18, microparticle-bonded stationary-phase columns with mixtures of organicsolvents. Chromatographic methods combined reverse-phase

HPLC and MS analysis of cell wall extracts to demonstratedthe chemical nature of mycolic acids from M. tuberculosis strainH37Ra (73, 92, 112). Fractions were separated into chemicallyhomologous mycolic acid series, consisting of 24 fractions fromspecies in the M. tuberculosis complex and Mycobacteriumsmegmatis (92). It was noted that the mycolic acids separated inthe reverse-phase column by a combination of factors, includ-ing chemical functional groups, polarity, and hydrocarbonchain length (89, 91, 92). In prior studies, similar results hadbeen demonstrated with shorter-carbon-chain fatty acids, de-rivatized with chemically analogous phenacyl derivatives whichhad demonstrated an increase in retention times with longercarbon chains but a decrease in retention times with greaterunsaturation (5, 30, 51, 79, 112). The separation properties ofthe mycolic acids were used to resolve an ongoing debateregarding the similarity of Mycobacterium gordonae and Myco-bacterium leprae (67). In an elaborate two-step HPLC proce-dure combined with TLC, the presence of specific methoxy-mycolates was demonstrated in M. gordonae and was absentin M. leprae; thus, the close relationship was not supported.It was reported that reverse-phase HPLC profiles for �-myco-late, methoxymycolate, and ketomycolate chemical functionalgroups demonstrated increasing retention times with increasedcarbon chain length. Moreover, it was noted that the mycolicacids were not completely chemically resolved by HPLC whensubjected only to reverse-phase partition chromatography (73,89). Consequently, chromatographic patterns were publishedof mycolic acids for mycobacteria that demonstrated exten-sive overlapping of the chemical fractions (73, 89, 92). Therecognition of pattern differences between the publishedchromatograms for members of the M. tuberculosis complex(slow-grower) and M. smegmatis (rapid-grower) by one of us(W.R.B.) provided the stimulus for further research with theHPLC instrument as a tool for species detection. These andother chemical studies demonstrated the separation capabili-ties of the various methods and the discriminatory power ofHPLC.

APPLICATION OF HIGH-PERFORMANCE LIQUIDCHROMATOGRAPHY FOR IDENTIFICATION

OF MYCOBACTERIA

Phenacyl esters of fatty acids were separated with a single-step, C18 reverse-phase HPLC method utilizing UV detection(UV-HPLC) (30). The procedure employed a dicyclohexyl-18-crown-6 ether (crown ether) catalyst for a rapid (30-min) der-ivatization of long-chain fatty acids detectable in the nanogramrange. The use of this derivative provided a sensitive detectionmethod for complete separation of mycolic acids from myco-bacteria into structural classes with silica and reverse-phasemodes (73, 92). Moreover, complete chemical separation ofthe mycolic acids was not necessary for species detection, sincethe reverse-phase pattern with gradient elution appeared dis-tinctive for M. smegmatis, BCG, M. bovis, and M. tuberculosis(89).

The reverse-phase UV-HPLC method was used with a mod-ified gradient elution system of chloroform-methanol to differ-entiate 13 Mycobacterium species into chromatographicallyrelated groups (10). Within each chromatographic group,distinctive mycolic acid patterns were used for species classifi-

FIG. 4—Continued.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

cation of mycobacteria. However, the mycolic acid isolationprocedure was cumbersome and involved multiple chloroformextractions. Moreover, the reflux-condensation apparatus usedfor sample derivatization was time-consuming; consequently,few samples could be analyzed in several days.

These early UV-HPLC investigations supported prior stud-ies done with mycolic acids which included the differentiationof Corynebacterium, Rhodococcus, Nocardia, and Mycobac-terium according to the carbon chain lengths of the mycolicacids (25, 69) and separation of Rhodococcus species intotwo groups composed of mycolic acids with either short orlong carbon chains (1, 17, 65). Investigators eventually dem-onstrated that the species with the long carbon chains werea unique group and transferred them to the revived genusGordona (88, 104). Also, in agreement with prior methods, theseparation of mycobacteria from other mycolic acid-contain-ing bacteria by UV-HPLC analysis of PBPA esters of my-colic acids was shown to result from an attraction of thedifferent carbon chain lengths to the column’s nonpolar sta-tionary phase, which resulted in patterns with different reten-tion times (11, 17, 91).

Modifications were made to the UV-HPLC method in stud-ies with species of slow-growing pathogenic mycobacteria andthe nonpathogenic M. gordonae (15). Whole cells of mycobac-teria, grown on Lowenstein-Jensen (L-J) slants, were removedeither by using a transfer loop or by washing the cells with 25%solution of potassium hydroxide in 50% ethanol and analyzedsolely by reverse-phase UV-HPLC. The cells were saponifiedin glass tubes for 18 h at 85°C, and the extracted mycolic acidswere derivatized for 45 min at 85°C to produce the UV-ab-sorbing PBPA esters. The mycolic acids were fractionated witha 5-�m-particle-size, reverse-phase C18 column using a gradi-ent elution system that was equilibrated for 13 min in chloro-form-methanol (10:90, vol/vol), followed by a 1-min adjust-ment to chloroform-methanol (25:75, vol/vol), which rapidlyeluted the more polar sample components, and a final 20-minlinear gradient of chloroform-methanol (70:30, vol/vol) toelute the mycolic acids. Distinctive mycolic acid patterns con-sisted of a single cluster of contiguous peaks for M. tuberculo-sis, Mycobacterium kansasii, and the nonpathogenic M. gordo-nae (Fig. 3). A detection sensitivity for the method was definedwith M. kansasii at 2.5 � 106 CFU. The reproducibility of themethod for the single contiguous peak pattern was demon-strated by examination of the mycolic acid pattern for a con-tinuously growing culture of M. tuberculosis sampled periodi-cally for 3 weeks. Although slight variations in peak heightsand times were noted, the basic pattern did not change. Incontrast to these single peak clusters, Mycobacterium aviumand Mycobacterium intracellulare produced disunited double-clustered peak patterns (see Fig. 8). These early studies re-quired 20 h to identify a species (15).

Additional improvements to the method were made in astudy with rapidly growing mycobacteria of clinical signifi-cance. It was shown that organisms could be killed and fattyacids could be saponified in the autoclave for 1 h at 121°C.Mycolic acids were extracted, derivatized, and processed in 3 h(16). Similar but distinctive chromatographic patterns weredeveloped for Mycobacterium chelonae and Mycobacterium ab-scessus under controlled growth conditions. On the other hand,122 strains of the Mycobacterium fortuitum complex and M.

smegmatis could not be separated under these conditions. Thestability of the mycolic acid patterns demonstrated by UV-HPLC for slow-growing mycobacteria was in contrast to that ofrapidly growing mycobacteria, which were affected by cultureconditions, especially temperature (16). Related studies of rap-idly growing mycobacteria demonstrated either an elongationor a shortening of mycolic acids as an adaptive response tochanges in temperature, although the effects appeared to bespecies specific (3, 96).

During the study with rapid-growing mycobacteria, a propri-etary high-molecular-weight standard synthesized by Ribi Im-munoChem (now Corixa Corp.), Hamilton, Mont., was usedfor the first time as an internal standard to determine repro-ducible relative retention times (RRT) with a standard devia-tion of �0.01 to 0.09 min for individual peaks (16). This stan-dard was used to devise an identification scheme to comparepeak height ratios, calculated for peaks from different chro-matograms with identical RRTs. The identification schemerequired manual calculation of values and was time-consumingbut correctly identified 36 isolates of M. chelonae and 24 of 25isolates of M. abscessus (16). A subsequent 10-step dichoto-mous peak height differentiations scheme was reported forslow-growing species and tested at two different laboratories.At one laboratory, 129 isolates of the following species werecorrectly identified: Mycobacterium asiaticum, M. bovis, BCGattenuated variants of M. bovis, M. tuberculosis, Mycobacteriumgastri, M. gordonae, M. kansasii, Mycobacterium marinum, andMycobacterium szulgai. The other laboratory identified 661 of670 different strains of these same species. The misidentifica-tions occurred with two species, M. kansasii and M. gordonae.Overall, 790 strains (98.6%) were correctly identified by bothlaboratories (14).

Final modifications of the chromatographic conditions wereevaluated by two laboratories with closely related, slow-grow-ing Mycobacterium species (19). The robustness of the dichot-omous identification scheme and UV-HPLC were challengedwith mycolic acid extracts from M. avium, M. intracellulare,Mycobacterium scrofulaceum, Mycobacterium xenopi, and somephenotypically confusing M. gordonae. A gradient system ofmethanol-methylene chloride with a constant solvent flow of1.5 ml/min was programmed to change from polar to nonpolaras follows. Initially, the column was equilibrated at 98%-2%(vol/vol), and then it was changed linearly after injection for1 min to 80%-20% (vol/vol); thereafter, the solvents werechanged linearly for 9 min to a final composition of 35%-65%(vol/vol). Final column reequilibration was carried out for 2min at the beginning solvent concentration, for an analysistime of 12 min. The by-products of the derivatization proce-dure and the cellular short-carbon-chain fatty acids eluted rap-idly at the beginning of the separation and did not interferewith the elution of the mycolic acids. Identifications with theflow chart scheme compared with results from control bio-chemical tests for M. avium, M. intracellulare, and M. scrofula-ceum were 97.9, 97.5, and 89.2%, respectively. All strains ofM. xenopi and M. gordonae were correctly identified. Theserefinements produced a standard method and permitted asingle Mycobacterium isolate received by the laboratory asa fully grown culture to be identified in approximately 2 h(19).



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

STANDARD METHOD FOR PREPARATIONOF MYCOLIC ACIDS

The sample preparation protocol consisted of cell harvest-ing, saponification, extraction, derivatization, and a cleanup orclarification step to generate mycolic acids amenable to detec-tion by UV-HPLC (Fig. 2). The procedure involved a whole-cell saponification, acidification, and methanolic extraction ofall cellular fatty acids, including the mycolic acids. Mycobac-teria were removed from a Lowenstein-Jensen slant with asterile polyester swab and placed in 2 ml of saponificationreagent composed of 25% potassium hydroxide in a water-methanol (1:1, vol/vol) mixture. The suspension was vigorouslymixed and autoclaved for 1 h at 121°C. After the mixture hadcooled to room temperature, 2 ml of chloroform was added,and then 1.5 ml of acidification reagent, a mixture of water andconcentrated hydrochloric acid (1:1, vol/vol) was added slowly.This solution was mixed, and the layers were separated. Thelower, organic layer was transferred with a Pasteur pipette to anew tube and evaporated in a heat block at 85 to 105°C with astream of air. The commercial derivatization reagent was sus-pended in acetonitrile and consisted of PBPA and a crownether catalyst (30). A mycolic acid sample was derivatized toPBPA esters by adding 20 mg of potassium bicarbonate, 50 �lof derivatization reagent, and 1.0 ml of chloroform. The tubeswere secured with Teflon-lined screw caps and heated at 85 to

105°C for 20 min. Samples were cooled and then clarified by achoice of methods, including either filtration through a 0.45-�m-pore-size nylon 66 membrane filter (10, 15), solid-phaseextraction (27), or liquid partitioning with an acidified aqueousmethanolic solution (19). Derivatization of the mycolates toUV-absorbing PBPA esters resulted in a stable compoundwhich could be stored for several months in the dark at 4°C.Specific details of the method are presented in standardizationmanuals (12, 13).

DIFFERENTIATION OF MYCOBACTERIUM SPECIESBY UV–HIGH-PERFORMANCE LIQUID

CHROMATOGRAPHY

The reproducibility of mycolic acid patterns from mycobac-teria has been established by laboratory studies conductedunder the auspices of the IWGMT. The first study employedmolecular, biochemical, serologic, and chemotaxonomic meth-ods in an attempt to identify closely related species suspectedof being M. avium, M. intracellulare, or M. scrofulaceum (110).In fact, the taxonomical complexity of the study group wasdemonstrated when all the methods were combined and aconsensus could not be reached for identification of four of theisolates. Overall, UV-HPLC results agreed with results fromT-catalase serology for 86% of the isolates but were not asgood as the 94% agreement between nucleic acid tests andT-catalase serology. The UV-HPLC result was comparable tothe 84% result for laboratories using panels of sera that cor-responded to the serovars represented in the study (110). An-other IWGMT study with problematic phenotypic clusters ofslowly growing mycobacteria demonstrated species-specificchromatographic patterns for Mycobacterium malmoense, My-cobacterium interjectum, and Mycobacterium simiae. However,some M. avium complex (MAC) species appeared to be misi-dentified when chromatographic results were compared withthe molecular data. It was suggested that when chromato-graphic techniques were used for identifying MAC species, theresults be supplemented with biochemical tests (109).

FIG. 5. Characteristic UV-HPLC chromatograms of Mycobacte-rium species with late-emerging, simple, single-cluster peak patternswith unresolved shoulder peaks. (A) Mycobacterium brumae ATCC51384T; (B) M. fallax ATCC 35219T; (C) M. triviale ATCC 23292T.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

FIG. 6. Characteristic UV-HPLC chromatograms of Mycobacterium species with two-peak clusters that emerge late and close together. (A)Mycobacterium agri ATCC 27406T; (B) M. chelonae ATCC 35751 (includes M. abscessus ATCC 19977T); (C) Mycobacterium chitae ATCC 19629;(D) Mycobacterium farcinogenes ATCC 35753T; (E) M. fortuitum complex (includes M. goodii ATCC 700504T, M. peregrinum ATCC 14467T,M. smegmatis ATCC 14468, and M. wolinskyi ATCC 700010T); (F) M. porcinum, ATCC 33775T; (G) M. senegalense ATCC 35796.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

This relationship between UV-HPLC and standard bio-chemical tests was compared in a year-long study with 502cultures of Mycobacterium and demonstrated a 97.2% agree-ment (45). Additionally, chromatographic patterns were com-pared with DNA probe results for 111 cultures that were des-ignated MAC strains, and they demonstrated an agreement of98.2%. Also, in this study, a positive reaction with a MAC ge-netic probe for a single isolate was clarified as a mixed cultureby UV-HPLC. In another study UV-HPLC, biochemicals, andgenetic probe results were compared in routine use for 18months (93). UV-HPLC identified 96.1% of the 1,103 strainstested. Biochemical and/or DNA probes identified 98.9% ofthe strains. The chromatographic method accurately identi-fied M. chelonae, M. xenopi, Mycobacterium terrae complex,M. simiae, and the rarely encountered species Mycobacteriumhaemophilum, M. malmoense, Mycobacterium shimoidei, andMycobacterium fallax. UV-HPLC results were compared to ge-netic probe results and were 100% specific and 98.9% sensitivefor M. tuberculosis complex. UV-HPLC identified 250 isolatesas MAC, which was in 98.7% agreement with results obtainedwith genetic probes. Thirty-one strains were identified by UV-HPLC as M. scrofulaceum, and 90.3% of these strains werenegative for MAC by genetic probes. Moreover, the accuracyof the genetic probe tests for M. tuberculosis complex andMAC species was confirmed for discrepant strains that pro-duced negative results by detection of other species of nontu-berculous mycobacteria (NTM) with UV-HPLC (4, 83).

Detection of unknown mycobacteria by UV-HPLC supportsthe belief that many species of mycobacteria remain unchar-acterized. Unusual mycobacteria have been recovered fromAIDS patients in liquid media from a BACTEC (Becton Dick-inson Diagnostic Instrument Systems, Sparks, Md.) instrument(22, 85). The identity of the isolates was problematic becausethey failed to grow when subcultured on conventional solidmedium used for mycobacteria. Dysgonic growth obtained inMiddlebrook 7H11 medium supplemented with mycobactin Jproduced a mycolic acid pattern with three separate peak clus-ters (22). Similar mycolic acids were also detected in Bactec12B vials by adding oleate-albumin-dextrose-catalase and byincreasing the incubation 3 to 5 days beyond the final 999growth index determination (21, 48, 85). In both studies, atriple-clustered mycolic acid pattern similar to that for M.

simiae was reported but the chromatographic data were notdefinitive for a known species. Investigators performing ge-netic sequencing analysis clarified the mycobacteria as a novelspecies, Mycobacterium genavense (22, 78). Another uniquechromatographic pattern was reported from an outbreak ofperitonitis involving the use of dialysis machines (105). Thebiochemical reactions resembled those of M. chelonae; conse-quently, the isolates were referred to as M. chelonae-like or-ganism. However, the UV-HPLC pattern and drug suscepti-bility characterizations were distinctive, which implied a novelspecies. Ultimately, the M. chelonae-like organism group wasdescribed as a new species, Mycobacterium mucogenicum (87,105). Other characterization studies of new species have in-cluded mycolic acid patterns from UV-HPLC analysis in theirinitial description and include Mycobacterium celatum (18),Mycobacterium heidelbergense (46), Mycobacterium goodii (8),Mycobacterium septicum (82, 105), and Mycobacterium wolinsky(8). Also, Mycobacterium triplex and Mycobacterium tusciaewere initially recognized by their UV-HPLC mycolic acid pat-terns (34, 101). Novel mycolic acid patterns are easily detectedby UV-HPLC. However, when mycolic acid patterns from un-defined isolates resemble those of known species, differentia-tion may be difficult (74). This was demonstrated in a charac-terization study with a suspected unknown environmentalsaprophyte with a mycolic acid pattern that was similar to thoseof several known rapid growers; this organism was later shownto be a variant of Mycobacterium austroafricanum by molecularmethods (7).

Infrequent isolation of fastidious or uncommon mycobacte-ria does not offer laboratory personnel the opportunity to de-velop the expertise needed for their identification. Develop-ment of mycolic acid patterns can provide the results necessaryto make a final decision. Such was the case for unusual isolatesfrom cutaneous lesions, a lymph node and eye of a patient withAIDS, and a cervical lymph node of a 3-year-old child, whichwere suspected to be M. haemophilum (94). This initial iden-tification was based on low growth temperature and a require-ment for iron. However, the identification was suspect becausethe organism was rarely isolated from patients and the inci-dence of infection was unknown for this laboratory. The spe-cies was confirmed by comparison of the isolate’s mycolic acidpattern to reference patterns by UV-HPLC similarity.

In a restricted study, the etiologic agents for six cases ofchronic tenosynovitis of the hand were all identified as Myco-bacterium nonchromogenicum, a member of the M. terrae com-plex, by UV-HPLC (77). Standard biochemical tests were re-ported unreliable for routine discrimination of members of thiscomplex (77, 97). The final identification was made by a con-currence of the mycolic acid pattern with a susceptibility to thedrug ofloxacin (77, 87). Interestingly, mycolic acid patternswere recently correlated with drug susceptibility results forstrains of M. tuberculosis (36).

Initially, representative chromatographic patterns of knownMycobacterium species were determined by examination oftype strains from American Type Culture Collection or Col-lection de I’Institut Pasteur grown with standardized growthconditions. However, chromatographic pattern variation wasdemonstrated when multiple strains of a species were exam-ined. Normally, this pattern variation was in the form of minorpeak height differences and did not affect the overall appear-

FIG. 6—Continued.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

FIG. 7. Characteristic UV-HPLC chromatograms of Mycobacterium species that have widely separated, double-peak clusters with prominentpeaks in the early cluster that emerge prior to 5.0 min. (A) Mycobacterium aichiense ATCC 27280T; (B) Mycobacterium chlorophenolicum ATCC49826T; (C) Mycobacterium diernhoferi ATCC 19340T; (D) M. flavescens ATCC 14474T (underlined peaks are not present in all strains); (E)Mycobacterium gadium ATCC 27726T; (F) Mycobacterium gilvum ATCC 43909; (G) M. hiberniae ATCC 49874T; (H) Mycobacterium komossenseATCC 33013T; (I) Mycobacterium madagascariense ATCC 49865T; (J) M. mucogenicum ATCC 49651; (K) M. neoaurum ATCC 25795T; (L)Mycobacterium phlei ATCC 11758T; (M) Mycobacterium rhodesiae ATCC 27024T; (N) Mycobacterium sphagni ATCC 33027.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

FIG. 7—Continued.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

ance of the chromatogram (15). Infrequently, known speciesthat normally produced multiple-cluster mycolic acid patternswould develop a pattern with a cluster of peaks dramaticallyreduced in height or altogether absent (13). For example, M.interjectum strains produced patterns with a single cluster ofpeaks that differed from the biphasic pattern for which thespecies was known (64, 100). It was speculated that the dis-crepant patterns resulted from analyzing variants of the samespecies but with different mycolic acid patterns. A similar phe-nomenon had been reported for 14 of 63 isolates of M. gordo-nae, except that the variant pattern was biphasic compared tothe normal single-cluster pattern (20). The frequency of my-colic acid patterns with dramatic differences from the standardspecies pattern are uncommon. However, when they occur,they appear stable for the isolate and may reflect a biologicaldiversity within the species. In most cases, the UV-HPLC re-sult can help avoid an incorrect identification on the basis of agross pattern incompatibility with the reference pattern (78,98, 99). This diversity of mycolic acid patterns demonstrated aneed to develop a comprehensive Mycobacterium library ofmycolic acid patterns for use with this method. A study of 23species frequently encountered in clinical samples by five lab-oratories, using 350 well-characterized strains, established rep-resentative patterns for each species (12, 13).

COMPUTER USE WITH UV–HIGH-PERFORMANCELIQUID CHROMATOGRAPHY

To improve on what has been perceived as an intrinsic sub-jectiveness of the manual identification schemes, automatedchemometric methods of analysis using pattern recognitionsoftware were evaluated (75, 76). A multivariate method ofanalysis that simultaneously compiled the test data was usedto compare the routinely analyzed results to a control set ofmycolic acid samples with a preassigned species category. Thebasis of the pattern recognition method was a measure of sim-ilarity based on the K Nearest Neighbors (KNN) algorithm,part of Pirouette, a commercially available software package(Informetrix, Seattle, Wash.). The algorithm found to have thegreatest accuracy for species prediction was a four-level, 15KNN decision model. When tested with a validation test set of

549 cultures representing 27 species, the multilevel KNN ap-proach accurately identified 92% of the species. The resultswere improved to 97% by combining the MAC-M. scrofula-ceum prediction results into a complex. The performance ofthe model for the 27 species was as follows: 14 species wereidentified at 100%, 4 species were identified at 90% to 95%, 5species were identified at 70 to 80%, and 4 species were iden-tified at less than 70%. In general, the individual accuracy ofthe test was promising for some species but the initial objectiveof obtaining greater accuracy than that obtained with themanual method was not achieved. Mistakes by the softwareappeared to result from subsets of heterogenous species orintraspecies variation in a species, which had not been repre-sented in the control set due to an insufficient number of testspecimens. Support for this assumption was provided whenincorrectly identified specimens from a validation set werereincorporated into a preassigned category, so that the com-puter model would then recognize the variation. Upon repeat-ing the analysis, the computer model recognized the samplescorrectly (75). Further verification was demonstrated in ananalysis which combined the commercial software with an in-dependently generated “library” of mycolic acid patterns from45 species of mycobacteria that were used to define the pre-assigned species categories (39). In an effort to increase theaccuracy of the test, individual peak identification tables de-signed to compensate for variable ranges in the elution of theinternal standard were included. For 1,155 strains represent-ing 18 species of slowly growing mycobacteria, 97% were cor-rectly identified. For 168 rapidly growing mycobacteria, 93%were correctly identified. It was verified that incorrect iden-tifications occurred when an unknown sample’s profile rep-resented a chromatographic pattern variation for the specieswhich was not incorporated in the software’s preassignedcategory.

A related approach for computer pattern recognition wasdesigned with commercially available software spread sheetsand macro commands (52). The method used a calibration al-gorithm with a peak-naming table and pattern recognition fordetection of different Mycobacterium complexes, includingM. tuberculosis complex, MAC, and a user-defined MAC-M. scrofulaceum complex. Raw chromatographic data were

FIG. 7—Continued.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

imported into the spreadsheet and subjected to automatedroutine calculations. Calibrated peak retention times and peakheight ratios were calculated for an unknown, and the resultswere compared to files in the Mycobacterium control set. Thiscreative method standardized the elution time of the peaksfrom unknown patterns to both an internal high-molecular-weight standard and an external standard of pooled, derivat-ized mycolic acids extracted from the type strain of M. avium,ATCC 19705. The program demonstrated a sensitivity andspecificity of 94.3 and 100%, respectively, for the MAC com-plex and 99 and 100%, respectively, for the M. tuberculosiscomplex (52).

The use of software classification models with the UV-HPLC method for automated prediction of mycobacteria spe-cies has limitations. Reliance on software alone will result inmisidentification due to the inadequacy of the library database,which exists as a static entity and thus forces identification ofan unknown sample. This is complicated by an instability ofpeak retention times caused by the coemergence of multiplepeaks, which is a problem for the software’s rigid identifi-cation times for peak location. For software to be used foridentification, the system must be constantly updated asvariants are recognized and new species are characterized.Presently, it is recommended that an identification be madeby manual examination of the chromatogram and visualcomparison of the pattern to those of reference chromato-grams (12, 13).

MYCOLIC ACID PATTERN RECOGNITIONAND MYCOBACTERIUM SPECIES

IDENTIFICATION

Chromatographic pattern reproducibility and visual patterninterpretation are related factors, and a laboratory’s expertisein species identification will be consistent with the experienceof the chromatographer. A prerequisite for routine UV-HPLCanalysis is adherence to defined chromatographic principles,including verifying the stability of the system by analysis ofmycolic acid control samples for confirmation of analysis times(12, 13).

The visual pattern recognition method employs only chro-matographic criteria, although when available, other identifi-cation test results should be included in the decision-makingprocesses. The initial step for identifying a species is determin-ing the overall complexity and number of mycolic acid peakclusters. These clusters may consist of a few peaks or manypeaks and are further defined as single-, double-, and triple- ormultiple-peak clusters. The range of time of elution betweenmultiple-peak groups and the positions of the peaks are deter-mined as RRTs to an internal standard. The RRT for a peakmay also be adjusted by comparison to an external mycobac-terial mycolic acid peak (13, 52). Usually, species with grosspattern differences are quickly eliminated, and the pattern ofthe unknown is compared to the remaining reference patterns(Fig. 3 to 10). When patterns are similar, the relationship ofpeak heights of major diagnostic peaks must be determined(16, 19, 52). If a reference pattern matches the unknown pat-tern, the species is identified; if not, the strain is reported asan “unidentified NTM.” Although this designation is not a

specific identification, it reflects the detection of mycobacterialmycolic acids and could be of clinical value.

Recently, the occurrence of NTM species reported by statelaboratories to the CDC Public Health Laboratory Informa-tion System (PHLIS) was tabulated and summarized for 1993through 1996. The report can be viewed at the World WideWeb site http://www.cdc.gov/ncidod/dastlr/TB/TBarchive.htM.It should be noted that CDC personnel did not verify theidentification methods or the identity of the species. ThePHLIS report reflected the data as they were received by CDC,but the incidence of species was consistent with prior reports(59, 106). However, sampling errors associated with these datawere recently demonstrated in an examination of the reportsfor M. malmoense (9).

For the PHLIS analysis, a total of 82,475 NTM speciesreports were examined, of which 11,161 were reported as un-known or were not identified to the species level. For thosespecifically identified, the species reported and the number ofreports were compared with a representative mycolic acidchromatogram with RRT’s and were as follows: M. avium,34,633 (Fig. 8A); M. gordonae, 18,440 (Fig. 3D); M. fortuitum,5,601 (Fig. 6E); M. kansasii, 2,784 (Fig. 3E); M. terrae, 2,053(Fig. 8H); M. chelonae, 1,789 (Fig. 6B); M. xenopi, 1,218 (Fig.8I); M. intracellulare, 1,081 (Fig. 8F); Mycobacterium flavescens,843 (Fig. 7D); M. simiae, 760 (Fig. 9C); M. marinum, 635 (Fig.4E); M. scrofulaceum, 616 (Fig. 8G); M. szulgai, 226 (Fig. 3H);M. asiaticum, 107 (Fig. 3A); M. malmoense, 101 (Fig. 4D);M. abscessus, 90 (Fig. 6B); M. smegmatis, 49 (Fig. 6E); BCGvariant of M. bovis, 36 (Fig. 3B); Mycobacterium vaccae, 32(Fig. 10K); M. gastri, 31 (Fig. 3C); M. mucogenicum, 26 (Fig.7J); M. haemophilum, 25 (Fig. 4B); M. nonchromogenicum, 20(Fig. 8H); Mycobacterium peregrinum, 14 (Fig. 6E); M. interjec-tum, 9 (Fig. 8E); M. fallax, 7 (Fig. 5B); M. genavense, 4 (Fig.9A); Mycobacterium neoaurum, 4 (Fig. 7K); M. celatum 3 (Fig.8C); Mycobacterium ulcerans 3 (Fig. 4G); Mycobacterium trivi-ale, 2 (Fig. 5C); and M. shimoidei, 2 (Fig. 10I). Species reportedonce during the reporting period were Mycobacterium aurum(Fig. 10B), Mycobacterium farcinogenes (Fig. 6D), Mycobacte-rium hiberniae (Fig. 7G), Mycobacterium intermedium (Fig.4C), Mycobacterium senegalense (Fig. 6G), Mycobacterium ther-moresistibile (Fig. 4F), and Mycobacterium tokaiense (Fig.10J). The remaining chromatograms shown in Fig. 3 to 10represent known species which were not reported by statehealth departments to CDC. However, the epidemiologicpicture of NTM infection is reported to be changing, and itis possible that the etiologic significance of these species willalso change (71).

The PHLIS report excluded members of the M. tuberculosiscomplex because the Division of Tuberculosis Elimination,CDC, annually reports the verified cases of tuberculosis as partof the tuberculosis surveillance system. However, the M. tuber-culosis complex chromatographic pattern is illustrated by theproduction of a stable, late-emerging, simple, mycolic acidpeak cluster (Fig. 3G). It has been reported that BCG atten-uated strains of M. bovis can be differentiated from the com-plex by this method (35). However, recently a confirmed isolateof a drug-resistant M. tuberculosis isolate from a foreign patientproduced a mycolic acid pattern that was confused with theBCG pattern. A study is being conducted to examine this dis-crepancy.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

FIG. 8. (Continued on following page.)



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

FLUORESCENT DERIVATIVES FORMYCOLIC ACID ANALYSIS

The most commonly used derivatization reagents for long-carbon-chain fatty acids are UV-adsorbing derivatives; how-ever, the detection sensitivity can be improved by the use offluorescence-labeling compounds. Theoretically, fluorescencedetection could increase the sensitivity 10 to 1,000 times. Apossible application for liquid chromatography separations wasproposed when short-carbon-chain fatty acids were labeledwith 4-bromomethyl-7-methoxycoumarin (Br-Mmc) and usedfor TLC analysis for detection of picomolar amounts (28).However, the production of 7-methoxy derivatives was tedious,and in gradient liquid chromatography the fluorescence yieldvaried and was solvent dependent (63). Moreover, carboxylicacid Br-Mmc derivatives had a lower detection sensitivity, andthe fluorescence intensity of the label was affected by the typeof carboxylic acid. Based on the assumption that the smallerthe molecular size of the labeling reagent, the better the sep-aration of the labeled carboxylic acids, an alternate fluores-cence-labeling reagent, 4-bromomethyl-7-acetoxycoumarin (Br-Mac), was suggested for femtomolar detection (103). However,Br-Mmc derivatives were demonstrated to be comparable inyield and sensitivity to Br-Mmc derivatives. Subsequent deri-

vatization of C1 to C5 short-carbon-chain fatty acids with 4-bro-momethyl-6,7-dimethoxycoumarin (coumarin) was shown tobe suitable for gradient elution with a reverse-phase column(33). In actual application, in two different studies the mycolicacid coumarin derivative produced 20 and 80-fold increases insensitivity compared to the UV-absorbing PBPA derivative(47, 86). An advantage of using the fluorescence detection wasthat only a portion of a single colony was required, comparedto a transfer loop full of cells for the UV method. Other fluo-rimetric reagents used to produce mycolic acid derivatives foridentification of clinical mycobacteria were 3-bromomethyl-7-methoxy-1,4-benzoxazin-2-1 and 4-bromomethyl-7-acetoxycou-marin (47). They were shown to have 16 to 26.6 times greatersensitivity, respectively, for detection of mycobacterial my-colic acids than did the UV-adsorbing PBPA reagent. All threefluorophores produced mycolic acid patterns similar to thepatterns for PBPA derivatives and were suggested for optimi-zation of the method for clinical mycobacterial species identi-fication.

An early demonstration of the use of a coumarin derivativewith mycobacterial mycolic acids and reverse-phase HPLCwas the detection of M. tuberculosis and M. paratuberculosis,the causative agent of Johne’s disease (23). C48 to C76 my-colic acids were separated with a gradient mobile phase ofchloroform and acetyl nitrile. The fluorescent derivative dis-played an excitation at 285 nm and emission at 340 nm, withcharacteristic, coemerging homologous mycolic acid peaks.The author speculated that the mycolic acid pattern for M.paratuberculosis would be useful in detecting this bacterium in

FIG. 8. Characteristic UV-HPLC chromatograms of Mycobacte-rium species that have widely separated, double-peak clusters withprominent peaks in the early cluster that emerge after 5.0 min. (A)M. avium ATCC 19698T; (B) Mycobacterium branderi ATCC 51789T;(C) M. celatum ATCC 51131T; (D) M. gordonae ATCC 35759 (UV-HPLC chromotype II); (E) M. interjectum ATCC 51457T (underlinedpeaks are not present in all strains); (F) M. intracellulare ATCC13950T; (G) M. scrofulaceum ATCC 19981T; (H) M. terrae ATCC15755T; (I) M. xenopi ATCC 19250.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

the feces of infected cattle. In another multiple-step chromato-graphic study, normal and reverse-phase HPLC produced high-ly characteristic profiles for M. tuberculosis with mycolic acidsas anthrylmethy esters from sputum samples (66).

The sensitivity of automated identification of M. tuberculosiscomplex and MAC from BACTEC 12B cultures and fromfluorochrome-stained smear-positive sputum specimens hasbeen described (52). The attractiveness of this method to rec-ognize mycobacteria sooner than conventional methods wasdemonstrated by a definitive identification of M. tuberculosiscomplex within 24 h of receiving specimens from individualswho were shedding a large number of bacilli. The fluorescencepatterns were noted to be comparable with those from thestandard UV-HPLC method.

The advantage of an increased sensitivity of detection forfluorescence detection was recently demonstrated by chro-matographic separation criteria for some rapidly growing my-cobacteria associated with wound infections, which had notbeen reported with UV-HPLC (8). However, the increased

sensitivity of fluorescence detection requires careful manipu-lation of samples to prevent cross-contamination, and car-ryover of mycolic acids from previously analyzed samples mustbe cautiously monitored. It may also be necessary to use solid-phase extraction to remove undesirable by-products and con-taminants to improve the signal-to-noise ratio (27). However,the potential for concomitantly reducing the cell mass andincreasing the sensitivity, thus reducing the time requiredfor identification, could substantially improve the chromato-graphic identification method.

UV–HIGH-PERFORMANCELIQUID CHROMATOGRAPHY AND

MOLECULAR METHODS

The conventional tests used in the clinical laboratory cannotidentify many of the newly described NTM species, and newer,rapid methods must be employed (6, 50). A problem with somenew identification methods is they may not be affordable or

FIG. 9. Characteristic UV-HPLC images of Mycobacterium species that have triple-peak clusters, close together that emerge later than 6.0 min.(A) M. genavense CDC 892120; (B) Mycobacterium lentiflavum 2186/92; (C) M. simiae ATCC 25275T; (D) M. triplex ATCC 700071T.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

amenable to routine laboratory schedules. Even so, chemotax-onomic and molecular identification methods reduce the turn-around times and are more accurate for discriminating species.The intrinsic nature of genetic methods for the analysis ofnucleic material has resulted in a variety of methods that mayinvolve combinations of PCR amplification, oligonucleotidehybridization probes (specific or array), restriction enzymedigestion, and sequence analysis. The molecular method ofchoice for detecting M. tuberculosis is a single, specific geneticprobe. Several of the methods mentioned above have beenproposed for molecular identification of NTM, but a consensusmethod is not in use. Here, a comparison is made to the mo-lecular method frequently reported for classifying NTM, i.e.,PCR amplification and direct DNA sequencing of portions ofthe 16S rRNA gene (6, 56).

DNA sequence analysis of the 16S rRNA gene can be per-formed manually or with automated sequencers. Manualsequencing is inexpensive to perform, but the data are sub-ject to base-calling errors, and depending on the method,radioactive by-products may be produced. For routine use,the automated method is preferred, but it is expensive. How-ever, the data from automated sequencers are virtually freeof base-calling errors and the method does not producehazardous by-products. UV-HPLC produces hazardous by-products consisting of mixed organic solvents, includingchloroform, methanol, and methylene chloride. UV-HPLCsample processing is rapid and easy, but for reproducibilityof chromatographic patterns, standardized conditions ofgrowth should be used. Molecular methods of analysis donot require standardized growth conditions but generally dorequire more hands-on time for sample processing. Althoughnone of these methods requires viable cells, UV-HPLC re-quires more cell biomass than the molecular methods do.Molecular methods have a greater detection sensitivity thanthe UV-HPLC method. Both UV-HPLC and automated se-quencing methods require expensive, sophisticated instrumen-tation. Although automated sequencing instrumentation isabout twice as costly to install and maintain as chromato-graphic instruments. A single sample can be processed in 2 h byUV-HPLC, and automated sequencing requires �8 h. UV-HPLC and manual sequencing are inexpensive to run on adaily basis, while automated sequencing is expensive. The re-ported cost of UV-HPLC per sample is approximately $3.00(45, 52, 93). Identifications of species using sequence analysisare made by comparison to international sequence databasesor in-house references. Identifications of species with mycolicacids are made by comparison to in-house databases of refer-ence patterns.

The support system for analyzing sequencing data is exten-sive and impressive, with easy-to-use commercial software andexhaustive World Wide Web sites. The only interactive Website for support of the chromatographic method specifically foridentification of mycobacteria is maintained by the HPLC Us-ers Group at http://hplc.cjb.net. Both methods represent po-tentially rapid and reliable techniques for recognition of knownand unknown mycobacteria and complement each other foridentification of Mycobacterium species (8, 18, 34, 46, 81, 82,101, 102).

INSTRUMENT CERTIFICATIONAND INSPECTIONS

Chromatograph manufacturers are intensely concerned withthe quality and performance of their products. Vendors con-stantly incorporate recent technological advances into themanufacture and operation of their instruments. In the 11years since the UV-HPLC method was first described, instru-ment and column reliability has improved substantially. A ma-jor improvement was the replacement of manual controllers bycomputers which provided a visual, easy-to-use platform fordirect control of all of the chromatographic features for solventcomposition, delivery, and program development. Any com-puterized system capable of producing a controlled solventgradient can be programmed with the UV-HPLC method. Thecomponents of an instrument have a finite life, and in high-throughput laboratories it would be prudent to maintain acommercial service contract. However, instrument dependabil-ity is excellent, and routine maintenance is normally all that isrequired. Since the quality of the chromatographic pattern isaffected by the performance of the instrument, routine qualitycontrol is essential. When components are replaced, a refer-ence standard, usually a designated species of Mycobacterium,must be routinely used to verify the reference conditions foranalysis (12, 13).

In the United States, laboratories considering using HPLCshould have the proficiency of a Level II or higher mycobac-teriology laboratory before instituting the method in order tosatisfy the Clinical Laboratory Improvement Act (CLIA) of1988 (55). It should be noted that CLIA inspectors considerUV-HPLC to be a high complex diagnostic procedure, and theoperator is accountable for records of method validation, qual-ity control, and routine use of the system.

HPLC USERS GROUP

The HPLC Users Group was formed during the 1992 Gen-eral Meeting of the American Society for Microbiology inNew Orleans, La., by a group of mycobacteriologists whowere using UV-HPLC and were concerned with the differentlaboratory techniques employed to analyze mycolic acids foridentification of mycobacteria (L. S. Guthertz, Letter, ASMNews 58:467, 1992). Afterwards, a survey was conducted forlaboratories using UV-HPLC, and it demonstrated a varietyof instruments and methods in use and a definite need forstandardization. However, even with a lack of standardiza-tion, it was found that all laboratories produced comparableUV-HPLC patterns for similar species, attesting to the ro-bustness of the method. After an extensive standardizationstudy with five laboratories, the HPLC Users Group, in co-operation with CDC, electronically published two manuals,Standardized Method for HPLC Identification of Mycobacte-ria and the Mycolic Acid Pattern Standards for HPLC Iden-tification of Mycobacteria. These manuals are available atthe CDC Web page http://www.cdc.gov/ncidod/dastlr/TB/TB_HPLC.htm.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

FIG. 10. (Continued on following page.)



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

CONCLUSION

The identification of Mycobacterium species by UV-HPLChas been documented as a rapid, reliable method by numer-ous laboratories. The versatility of this method for detection

of M. tuberculosis, a wide range of known or unknown NTMspecies, and other mycolic acid-containing genera by a singletest offers a significant advantage for the laboratory.

REFERENCES

1. Alshamaony, L., M. Goodfellow, and D. E. Minnikin. 1976. Free mycolicacids as criteria in the classification of Nocardia and the “rhodochrous”complex. J. Gen. Microbiol. 92:188–199.

2. Asselineau, J., and E. Lederer. 1950. Structure of the mycolic acids ofmycobacteria. Nature (London) 166:782–783.

3. Baba, T., K. Kaneda, E. Kusunose, M. Kusunose, and I. Yano. 1989.Thermally adaptive changes of mycolic acids in Mycobacterium smegmatis.J. Biochem. 106:81–86.

4. Body, B. A., N. G. Warren, A. Spicer, D. Henderson, and M. Chery. 1990.Use of Gen-Probe and Bactec for rapid isolation and identification ofmycobacteria: correlation of probe results with growth index. Am. J. Clin.Pathol. 93:415–420.

5. Borch, R. F. 1975. Separation of long chain fatty acids as phenacyl esters by

FIG. 10. Characteristic UV-HPLC chromatograms of Mycobacte-rium species that have triple-peak clusters with peaks in the earlycluster that emerge before 5.0 min. (A) Mycobacterium alvei ATCC51304T; (B) Mycobacterium aurum ATCC 25793; (C) M. austroafrica-num ATCC 33464T; (D) Mycobacterium chubuense ATCC 27278T; (E)Mycobacterium cookii ATCC 49103T; (F) Mycobacterium duvalii ATCC43910T; (G) Mycobacterium obuense ATCC 27023T; (H) Mycobacte-rium parafortuitum ATCC 19686T; (I) M. shimoidei ATCC 27964; (J)M. tokaiense ATCC 27282T; (K) M. vaccae ATCC 15483T.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

high pressure liquid chromatography. Anal. Chem. 47:2437–2439.6. Bottger, E. C. 1996. Approaches for identification of microorganisms: de-

spite longer experience with fatty acid profiles, DNA-based analysis offersseveral advantages. ASM News 62:247–250.

7. Bottger, E. C., P. Kirschner, B. Springer, and W. Zumft. 1997. Mycobac-teria degrading polycyclic aromatic hydrocarbons. Int. J. Sys. Bacteriol. 47:247.

8. Brown, B. A., B. Springer, V. A. Steingrube, R. W. Wilson, G. E. Pfyffer,M. J. Garcia, M. C. Menendez, B. Rodriguez-Salgado, K. C. Jost, Jr., S. H.Chiu, G. O. Onyi, E. C. Bottger, and R. J. Wallace, Jr. 1999. Mycobacteriumwolinskyi sp. nov. and Mycobacterium goodii sp. nov., two new rapidly grow-ing species related to Mycobacterium smegmatis and associated with humanwound infections: a cooperative study from the International WorkingGroup on Mycobacterial Taxonomy. Int. J. Syst. Bacteriol. 49:1493–1511.

9. Buchholz, U. T., M. M. McNeil, L. E. Keyes, and R. C. Good. 1998. Myco-bacterium malmoense infections in the United States, January 1993 throughJune 1995. Clin. Infect. Dis. 27:551–558.

10. Butler, W. R. 1985. Reverse phase high performance liquid chromatographyseparation of mycolic acids as para-bromophenacyl esters as an aid inmycobacterial classification. Masters Thesis, Georgia State University, At-lanta.

11. Butler, W. R., D. G. Ahearn, and J. O. Kilburn. 1986. High-performanceliquid chromatography of mycolic acids as a tool in the identification ofCorynebacterium, Nocardia, Rhodococcus, and Mycobacterium species.J. Clin. Microbiol. 23:182–185.

12. Butler, W. R., M. M. Floyd, V. Silcox, G. Cage, E. Desmond, P. S. Duffey,L. S. Guthertz, W. M. Gross, K. C. Jost, Jr., L. S. Ramos, L. Thibert, andN. Warren. 1996. Standardized method for HPLC identification of myco-bacteria. Centers for Disease Control and Prevention, Atlanta, Ga.

13. Butler, W. R., M. M. Floyd, V. Silcox, G. Cage, E. Desmond, P. S. Duffey,L. S. Guthertz, W. M. Gross, K. C. Jost, Jr., L. S. Ramos, L. Thibert, andN. Warren. 1999. Mycolic acid pattern standards for HPLC identification ofmycobacteria. Centers for Disease Control and Prevention, Atlanta, Ga.

14. Butler, W. R., K. C. Jost, Jr., and J. O. Kilburn. 1991. Identification ofmycobacteria by high-performance liquid chromatography. J. Clin. Micro-biol. 29:2468–2472.

15. Butler, W. R., and J. O. Kilburn. 1988. Identification of major slowlygrowing pathogenic mycobacteria and Mycobacterium gordonae by high-performance liquid chromatography of their mycolic acids. J. Clin. Micro-biol. 26:50–53.

16. Butler, W. R., and J. O. Kilburn. 1990 High-performance liquid chroma-tography patterns of mycolic acids as criteria for identification of Mycobac-terium chelonae, Mycobacterium fortuitum, and Mycobacterium smegmatis.J. Clin. Microbiol. 29:2094–2098.

17. Butler, W. R., J. O. Kilburn, and G. P. Kubica. 1987. High-performanceliquid chromatography analysis of mycolic acids as an aid in laboratoryidentification of Rhodococcus and Nocardia species. J. Clin. Microbiol. 25:2126–2131.

18. Butler, W. R., S. P. O’connor, M. A. Yakrus, R. W. Smithwick, B. B.Plikaytis, C. W. Moss, M. M. Floyd, C. L. Woodley, J. O. Kilburn, F. S.Vadney, and W. M. Gross. 1993. Mycobacterium celatum sp. nov. Int. J. Syst.Bacteriol. 43:539–548.

19. Butler, W. R., L. Thibert, and J. O. Kilburn. 1992. Identification of Myco-bacterium avium complex strains and some similar species by high-perfor-mance liquid chromatography. J. Clin. Microbiol. 30:2698–2704

20. Cage, G. D. 1992. High-performance liquid chromatography patterns ofMycobacterium gordonae mycolic acids. J. Clin. Microbiol. 30:2402–2407.

21. Cage, G. D. 1994. Direct identification of Mycobacterium species in BACTEC7H12B medium by use of high-performance liquid chromatography. J. Clin.Microbiol. 32:521–524.

22. Coyle, M. B., L. C. Carlson, C. K. Wallis, R. B. Leonard, V. A. Raisys, J. O.Kilburn, M. Samadpour, and E. C. Bottger. 1992. Laboratory aspects of“Mycobacterium genavense,” a proposed species isolated from AIDS pa-tients. J. Clin. Microbiol. 30:3206–3212.

23. Craig, M. A., and R. G. Burdett. 1983. Separation by high-performanceliquid chromatography of fatty acids from Mycobacterium paratuberculosis,p. 2–8. In R. S. Merkal (ed.), Proceedings of the International Colloquiumon research in paratuberculosis. The National Animal Disease Center,Ames, Iowa.

24. Crawford, J. T. 1994. Development of rapid techniques for identification ofM. avium infections. Res. Microbiol. 145:13–17.

25. De Briel, D., F. Couderc, P. Riegel, F. Jehl, and R. Minck. 1992. High-performance liquid chromatography of corynomycolic acids as a tool inidentification of Corynebacterium species and related organisms. J. Clin.Microbiol. 30:1407–1417.

26. Dixon, P. F., M. S. Stoll, and C. K. Lim. 1976. High pressure liquidchromatography in clinical chemistry. Ann. Clin. Biochem. 13:409–432.

27. Duffey, P. S., L. S. Guthertz, and G. C. Evans. 1996. Improved rapididentification of mycobacteria by combining solid-phase extraction withhigh-performance liquid chromatography analysis of BACTEC cultures.J. Clin. Microbiol. 34:1939–1943.

28. Dunges, W. 1977. 4-Bromomethyl-7-methoxycoumarin as a new fluores-

cence label for fatty acids. Anal. Chem. 49:442–445.29. Dupree, W. G., and H. B. Bradford, Jr., 1992. New developments in my-

cobacteria identification: public health laboratory modernization. J. La.State Med. Soc. 144:379–382.

30. Durst, H. D., M. Milano, E. J. Kikta, Jr., S. A. Connelly, and E. Grushka.1975. Phenacyl esters of fatty acids via crown ether catalysts for enhancedultraviolet detection in liquid chromatography. Anal. Chem. 47:1797–1801.

31. Embley, T. M., and E. Stackebrandt. 1994. The molecular phylogeny andsystematics of the actinomycetes. Annu. Rev. Microbiol. 48:257–289.

32. Etemadi, A.-H. 1967. The use of pyrolysis gas chromatography and massspectroscopy in the study of the structure of mycolic acids. J. Gas Chro-matogr. 5:447–456.

33. Farinotti, R., P. Siard, J. Bourson, S. Kirkiacharian, B. Valeur, and G.Mahuzier. 1983. 4-Bromomethyl-6,7-dimethoxycoumarin as a fluorescentlabel for carboxylic acids in chromatographic detection. J. Chromatogr. 269:81–90.

34. Floyd, M. M., L. S. Guthertz, V. A. Silcox, P. S. Duffey, Y. Jang, E. P.Desmond, J. T. Crawford, and W. R. Butler. 1996. Characterization of aSAV organism and proposal of Mycobacterium triplex sp. nov. J. Clin.Microbiol. 34:2963–2967.

35. Floyd, M. M., V. A. Silcox, W. D. Jones, Jr., W. R. Butler, and J. O. Kilburn.1992. Separation of Mycobacterium bovis BCG from Mycobacterium tuber-culosis and Mycobacterium bovis by using high-performance liquid chroma-tography of mycolic acids. J. Clin. Microbiol. 30:1327–1330.

36. Garza-Gonzalez, E., M. Guerrero-Olazaran, R. Tijerina-Menchaca, andJ. M. Viader-Salvado. 1997. Determination of drug susceptibility of Myco-bacterium tuberculosis through mycolic acid analysis. J. Clin. Microbiol. 35:1287–1289.

37. Garza-Gonzalez, E., M. Guerrero-Olazaran, R. Tijerina-Menchaca, andJ. M. Viader-Salvado. 1998. Identification of mycobacteria by mycolic acidpattern. Arch. Med. Res. 29:303–306.

38. Gerson, B. 1981. HPLC monitoring improves drug therapy. Lab World 1:43–45.

39. Glickman, S. E., J. O. Kilburn, W. R. Butler, and L. S. Ramos. 1994. Rapididentification of mycolic acid patterns of mycobacteria by high-performanceliquid chromatography using pattern recognition software and a Mycobac-terium library. J. Clin. Microbiol. 32:740–745.

40. Good, R. 1990. New culture identification procedure initiated in CDCMycobacteriology Lab. TB notes, Summer, p. 13. Division of TB Control,Centers for Disease Control, Atlanta, Ga.

41. Good, R. C. 1992. The genus Mycobacterium—medical, p. 1238–1270. In A.Balows, H. G. Truper, M. Dworkin, W. Harder, and K.-H. Schleifer (ed.),The prokaryotes—a handbook on the biology of bacteria: ecophysiology,isolation, identification, applications, vol. II. Springer-Verlag, New York,N.Y.

42. Goodfellow, M., and D. E. Minnikin. 1984. Circumscription of the genus, p.1–24. In G. P. Kubica and L. G. Wayne (ed.), The mycobacteria: a source-book. Marcel Dekker, Inc., New York, N.Y.

43. Reference deleted.44. Reference deleted.45. Guthertz, L. S., S. D. Lim, Y. Jang, and P. S. Duffey. 1993. Curvilinear-

gradient high-performance liquid chromatography for identification of my-cobacteria. J. Clin. Microbiol. 31:1876–1881.

46. Haas, W. H. W. R. Butler, P. Kirschner, B. B. Plikaytis, M. B. Coyle, B.Amthor, A. G. Steigerwalt, D. J. Brenner, M. Salfinger, J. T. Crawford,E. C. Bottger, and H. J. Bremer. 1997. A new agent of mycobacteriallymphadenitis in children: Mycobacterium heidelbergense sp. nov. J. Clin.Microbiol. 35:3203–3209.

47. Hagen, S. R., and J. D. Thompson. 1995. Analysis of mycolic acids byhigh-performance liquid chromatography and fluorimetric detection impli-cation for the identification of mycobacteria in clinical samples. J. Chro-matogr. 692:167–172.

48. Heifets, L. B., and R. C. Good. 1994. Current laboratory methods for thediagnosis of tuberculosis, p. 100. In B. R. Bloom (ed.), Tuberculosis: patho-genesis, protection, and control. American Society for Microbiology, Wash-ington, D.C.

49. Horvath, C. 1980. High-performance liquid chromatography: advances andperspectives. Academic Press, Inc., New York, N.Y.

50. Huebner, R. E., R. C. Good, and J. I. Tokars. 1993. Current practices inmycobacteriology: results of a survey of state public health laboratories.J. Clin. Microbiol. 31:771–775.

51. Jordi, H. C. 1978. Separation of long and short chain fatty acids as naph-thacyl and substituted phenacyl esters by high performance liquid chroma-tography. J. Liquid Chromatogr. 1:215–230.

52. Jost, K. C., D. F. Dunbar, S. S. Barth, V. L. Headley, and L. B. Elliot. 1995.Identification of Mycobacterium tuberculosis and M. avium complex directlyfrom smear-positive sputum specimens and Bactec 12B cultures by high-performance liquid chromatography with fluorescence detection and com-puter-driven pattern recognition models. J. Clin. Microbiol. 33:1270–1277.

53. Kanetsuna, F. 1968. Chemical analyses of mycobacterial cell walls. Bio-chim. Biophys. Acta 158:130–143.

54. Kanetsuna, F., and A. Bartoli. 1972. A simple chemical method to differ-



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

entiate Mycobacterium from Nocardia. J. Gen. Microbiol. 70:209–212.55. Kent, P. T., and G. P. Kubica. 1985. Public health mycobacteriology: a

guide for the level III laboratory. Centers for Disease Control, U.S. De-partment of Health and Human Services, Atlanta, Ga.

56. Kirschner, P., B. Springer, U. Vogel, A. Meier, A. Wrede, M. Kiekenbeck,F.-C. Bange, and E. C. Bottger. 1993. Geotypic identification of mycobac-teria by nucleic acid sequence determination: report of a 2-year experiencein a clinical microbiology. J. Clin. Microbiol. 31:2882–2889.

57. Kubica, G. P. 1973. Differential identification of mycobacteria. VII. Keyfeatures for identification of clinically significant mycobacteria. Am. Rev.Respir. Dis. 107:9–21.

58. Kubica, G. P. 1978. Classification and nomenclature of the mycobacteria.Ann. Microbiol. (Inst. Pasteur) 129:7–12.

59. Kubica, G. P. 1984. Clinical microbiology, p. 133–175. In G. P. Kubica andL. G. Wayne (ed.), The mycobacteria: a sourcebook. Marcel Dekker, Inc,New York, N.Y.

60. Laneelle, G. 1963. Nature des acides mycoliques de Mycobacterium paratu-berculosis: application de la chromatographie sur couche mince a leurfractionnement. C. R. Hebd. Seances Acad. Sci. Paris 257:781–783.

61. Lechevalier, M. P., A. C. Horan, and H. Lechevalier. 1971. Lipid compo-sition in the classification of nocardiae and mycobacteria. J. Bacteriol. 105:313–318.

62. Levy-Frebault, V. V., and F. Portaels. 1992. Proposed minimal standards forthe genus Mycobacterium and for description of new slowly growing Myco-bacterium species. Int. J. Syst. Bacteriol. 42:315–323.

63. Lloyd, J. B. F. 1980. Phenanthrimidazoles as fluorescent derivatives in theanalysis of fatty acids by high-performance liquid chromatography. J. Chro-matogr. 189:359–373.

64. Lumb, R., A. Goodwin, R. Ratcliff, R. Stapledon, A. Holland, and I. Bastian.1997. Phenotypic and molecular characterization of three clinical isolates ofMycobacterium interjectum. J. Clin. Microbiol. 35:2782–2785.

65. Minnikin, D. E., L. Alshamaony, and M. Goodfellow. 1975. Differentiationof Mycobacterium, Nocardia, and related taxa by thin-layer chromato-graphic analysis of whole-organism methanolysates. J. Gen. Microbiol. 88:200–204.

66. Minnikin, D. E., R. C. Bolton, S. Hartmann, G. S. Besra, P. A. Jenkins, A. I.Mallet, E. Wilkins, A. M. Lawson, and M. Ridell. 1993. An integratedprocedure for the direct detection of characteristic lipids in tuberculosispatients. Ann. Soc. Belg. Med. Trop. 73:13–24.

67. Minnikin, D. E., G. Dobson, M. Goodfellow, P. Draper, and M. Magnusson.1985. Quantitative comparison of the mycolic and fatty acid compositions ofMycobacterium leprae and Mycobacterium gordonae. J. Gen. Microbiol. 131:2013–2021.

68. Minnikin, D. E., G. Dobson, J. H. Parlet, A. K. Datta, S. M. Minnikin, andM. Goodfellow. 1986. Analysis of mycobacteria mycolic acids, p. 139–143. InR. Klein and B. Schmitz (ed.), Topics in lipid research: from structuralelucidation to biological function. Royal Society of Chemistry, London,United Kingdom.

69. Minnikin, D. E., and M. Goodfellow. 1980. Lipid composition in the clas-sification and identification of acid-fast bacteria, p. 189–256. In M. Good-fellow and R. G. Board (ed.), Microbiological classification and identifica-tion. Academic Press, Ltd., London, United Kingdom.

70. Minnikin, D. E., S. M. Minnikin, J. H. Parlett, M. Goodfellow, and M. Mag-nusson. 1984. Mycolic acid patterns of some species of Mycobacterium.Arch. Microbiol. 139:225–231.

71. O’Brien J. R., J. L. Geiter, and D. E. Snider, Jr. 1987. The epidemiology ofnontuberculous mycobacterial diseases in the United States: results from anational survey. Am. Rev. Respir. Dis. 135:1007–1014.

72. Pei, P. T., W. C. Kossa, S. Ramachandran, and R. S. Henly. 1976. Highpressure reverse phase liquid chromatography of fatty acid p-bromo-phenacyl esters. Lipids 11:814–816.

73. Qureshi, N., K. Takayama, H. C. Jordi, and H. K. Schnoes. 1978. Charac-terization of the purified components of a new homologous series of �-my-colic acids from Mycobacterium tuberculosis H37Ra. J. Biol. Chem. 253:5411–5417.

74. Rafii, F., W. R. Butler, and C. E. Cerniglia. 1992. Differentiation of arapidly growing, scotochromogenic, polycyclic-aromatic-hydrocarbon-me-tabolizing strain of Mycobacterium sp. from other known Mycobacteriumspecies. Arch. Microbiol. 157:512–520.

75. Ramos, L. S. 1992. Classification of mycobacteria by HPLC and patternrecognition. Am.Biotechnol. Lab. 10:27–32.

76. Ramos, L. S. 1994. Characterization of mycobacteria species by HPLC andpattern recognition. J. Chromatogr. Sci. 32:219–227.

77. Ridderhof, J. C., R. J. Wallace, Jr., J. O. Kilburn, W. R. Butler, N. G.Warren, M. Tsukamura, L. C. Steele, and E. S. Wong. 1991. Chronictenosynovitis of the hand due to Mycobacterium nonchromogenicum: use ofhigh-perofrmance liquid chromatography for identification of isolates. Rev.Infect. Dis. 13:857–864.

78. Ritter, D., L. C. Carlson, B. K. Logan, L. S. Ramos, J. O. Kilburn, andM. B. Coyle. Differentiation of Mycobacterium genavense and Mycobacte-rium simiae by automated mycolic acid analysis with high-performanceliquid chromatography. J. Clin. Microbiol. 34:2004–2006.

79. Roggero, J. P., and S. V. Coen. 1981. Isocratic separation of fatty acidderivatives by reversed phase liquid chromatography. Influence of the sol-vent on selectivity and rules for elution order. J. Liquid Chromatogr. 4:1817–1829.

80. Runyon, E. H. 1959. Anonymous mycobacteria in pulmonary disease. Med.Clin. North Am. 43:273–290.

81. Rynkiewicz, D. L., G. D. Cage, W. R. Butler, and N. M. Ampel. 1997.Clinical and microbiological assessment of Mycobacterium simiae isolatesfrom a single laboratory in southern Arizona. Clin. Infect. Dis. 26:625–630.

82. Schinsky, M. F., M. M. McNeil, A. M. Whitney, A. G. Steigerwalt, B. A.Lasker, M. M. Floyd, G. G. Hogg, D. J. Brenner, and J. M. Brown. 2000.Mycobacterium septicum sp. nov., a new rapidly growing species associatedwith catheter-related bacteraemia. Int. J. Syst. Evol. Microbiol. 50:575–581.

83. Sherman, I., N. Harrrington, A. Rothrock, and H. George. 1989. Use of acutoff range in identifying mycobacteria by the gen-probe rapid diagnosticsystem. J. Clin Microbiol. 27:241–244.

84. Shinnick, T. M., and R. C. Good. 1995. Diagnostic mycobacteriology lab-oratory practices. Clin. Infect. Dis. 21:291–299.

85. Siddiqi, S. H., A. Laszlo, W. R. Butler, and J. O. Kilburn. 1993. Bacterio-logic investigations of unusual mycobacteria isolated from immunocompro-mised patients. Diagn. Microbiol Infect Dis. 16:321–323.

86. Simon, V. A., Y. Hale, and A. Taylor. 1993. Identification of Mycobacteriumtuberculosis and related organisms by HPLC with fluorescence detection.LC-GC 11:444–448.

87. Springer, B., E. C. Bottger, P. Kirschner, and R. J. Wallace, Jr. 1995.Phylogeny of the Mycobacterium chelonae-like organism based on partialsequencing of the 16S rRNA gene and proposal of Mycobacterium muco-genicum sp. nov. Int. J. Syst. Bacteriol. 45:262–267.

88. Stackebrandt, E., J. Smida, and M. D. Collins. 1988. Evidence of phyloge-netic heterogeneity within the genus Rhodococcus; revival of the genusGordona (Tsukamura). J. Gen. Appl. Microbiol. 34:341–348.

89. Steck, P. A., B. A. Schwartz, M. S. Rosendahl, and G. R. Gray. 1978.Mycolic acids: a reinvestigation. J. Biol. Chem. 253:5625–5629.

90. Stodola, F. H., A. Lesuk, and R. J. Anderson. 1938. The chemistry of thelipids of tubercle bacilli. LIV. The isolation and properties of mycolic acid.J. Biol Chem. 126:505–513.

91. Takayama, K., H. C. Jordi, and F. Benson. 1980. Separation of fatty acidsas their p-bromophenacyl esters on a C30-bonded silica column by highperformance liquid chromatography. J. Liq. Chromatogr. 3:61–69.

92. Takayama, K., N. Qureshi, H. C. Jordi, and H. K. Schnoes. 1979. Separa-tion of homologous series of mycolic acids from Mycobacterium tuberculosisH37Ra by high performance liquid chromatography. Chromatogr. Sci. 10:91–101.

93. Thibert, L., and S. LaPierre. 1993. Routine application of high-perfor-mance liquid chromatography for identification of mycobacteria. J. Clin.Microbiol. 31:1759–1763.

94. Thibert, L., F. Lebel, and B. Martineau. 1990. Two cases of Mycobacteriumhaemophilum infection in Canada. J. Clin. Microbiol. 28:621–623.

95. Timpe, A., and E. H. Runyon. 1954. The relationship of “atypical” acid-fastbacteria to human disease. J. Lab. Clin. Med. 44:202–209.

96. Toriyama, S., I. Yano, M. Masui, E. Kusunose, M. Kusunose, and N.Akimori. 1980. Regulation of cell wall mycolic acid biosynthesis in acid-fastbacteria. I. Temperature-induced changes in mycolic acid molecular speciesand related compounds in Mycobacterium phlei. J. Biochem. 88:211–221.

97. Torkko, P., M. Suutari, S. Suomalainen, L. Paulin, L. Larsson, and M. L.Katila. 1998. Separation among species of Mycobacterium terrae complex bylipid analyses: comparison with biochemical tests and 16S rRNA sequenc-ing. J. Clin. Microbiol. 36:499–505.

98. Tortoli, E., and A. Bartoloni. 1996. High-performance liquid chromatogra-phy and identification of mycobacteria. Rev. Med. Microbiol. 7:207–219.

99. Tortoli, E., A. Bartoloni, C. Burrini, A. Mantella, and M. T. Simonetti.1995. Utility of high-performance liquid chromatography for identificationof mycobacterial species rarely encountered in clinical laboratories. Eur.J. Clin. Microbiol. Infect. Dis. 14:240–243.

100. Tortoli, E., P. Kirschner, A. Bartoloni, C. Burrini, V. Manfrin, A. Mantella,M. Scagnelli, C. Scarparo, M. T. Simonetti, and E. C. Bottger. 1996.Isolation of an unusual Mycobacterium from an AIDS patient. J. Clin.Microbiol. 34:2316–2319.

101. Tortoli, E., R. M. Kroppenstedt, A. Bartoloni, G. Caroli, I. Jan, J. Pawlow-ski, and S. Emler. 1999. Mycobacterium tusciae sp. nov. Int. J. Syst. Bacte-riol. 49:1839–1844.

102. Tortoli, E., C. Piersimoni, P. Kirschner, A. Bartoloni, C. Burrini, C. Lac-chini. A. Mantella, G. Muzzi, C. P. Tosi, V. Penati, C. Scarparo, M. T.Simonetti, and E. C. Bottger. 1997. Characterization of mycobacterial iso-lates phylogenetically related to, but different from, Mycobacterium simiae.J. Clin. Microbiol. 35:697–702.

103. Tsuchiya, H., T. Hayashi, H. Naruse, and N. Takagi. 1982. High-perfor-mance liquid chromatography of carboxylic acids using 4-bromomethyl-7-acetoxycoumarin as fluorescence reagent. J. Chromatogr. 234:121–130.

104. Tsukamura, M. 1971. Proposal of a new genus, Gordona, for slightly acid-fast organisms occurring in sputa of patients with pulmonary disease and insoil. J. Gen. Microbiol. 68:15–26.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

105. Wallace, R. J., Jr., V. A. Silcox, M. Tsukamura, B. A. Brown, J. O. Kilburn,W. R. Butler, and G. Onyi. 1993. Clinical significance, biochemical features,and susceptibility patterns of sporadic isolates of the Mycobacterium chelo-nae-like organism. J. Clin. Microbiol. 31:3231–3239.

106. Wayne, L. G. 1984. Mycobacterial speciation, p. 25–65. In G. P. Kubica andL. G. Wayne (ed.), The mycobacteria: a sourcebook. Marcel Dekker, Inc.,New York, N.Y.

107. Wayne, L. G., H. C. Engbaek, H. W. B. Engel, S. Froman, W. Gross, J.Hawkins, W. Kappler, A. G. Karlson, H. H. Kleeberg, I. Krasnow, G. P.Kubica, C. McDurmont, E. E. Nel, S. R. Pattyn, K. H. Schroder, S. Sho-walter, I. Tarnok, M. Tsukamura, B. Vergmann, and E. Wolinsky. 1974.Highly reproducible techniques for use in systematic bacteriology in thegenus Mycobacterium: tests for pigment, urease, resistance to sodium chlo-ride, hydrolysis of Tween 80 and �-galactosidase. Int. J. Syst. Bacteriol. 24:412–419.

108. Wayne, L. G., H. W. B. Engel, C. Grassi, W. Gross, J. Hawkins, P. A.Jenkins, W. Kappler, W. H. Kleeberg, I. Krasnow, E. E. Nel, S. R. Pattyn,P. A. Richards, S. Showalter, M. Slosarek, I. Szabo, I. Tarnok, M. Tsuka-mura, B. Vergmann, and E. Wolinsky. 1976. Highly reproducible tech-niques for use in systematic bacteriology in the genus Mycobacterium: testsfor niacin and catalase and for resistance to isoziazid, thiophene 2-carbox-ylic hydrazide, hydroxylamine and p-nitrobenzoate. Int. J. Syst. Bacteriol.26:311–318.

109. Wayne, L. G., R. C. Good, E. C. Bottger, R. Butler, M. Dorsch, T. Ezaki, W.Gross, V. Jonas, J. Kilburn, P. Kirschner, M. I. Krichevsky, M. Ridell,T. M. Shinnick, B. Springer, E. Stackebrandt, I. Tarnok, Z. Tarnok,H. Tasaka, V. Vincent, N. G. Warren, C. A. Knott, and R. Johnson. 1996.Sematide- and chemotaxonomy-based analyses of some problematic phe-notypic clusters of slowly growing mycobacteria: a cooperative study of theInternational Working Group on Mycobacterial Taxonomy. Int. J. Syst.Bacteriol. 46:280–297.

110. Wayne, L. G., R. C. Good, A. Tsang, R. Butler, D. Dawson, D. Groothuis, W.Gross, J. Hawkins, J. Kilburn, M. Kubin, K. H. Schroder, V. A. Silcox, C.Smith, M.-F. Thorel, C. Woodly, and M. A. Yakrus. 1993. Serovar deter-mination and molecular taxonomic correlation in Mycobacterium avium,Mycobacterium intracellulare, and Mycobacterium scrofulaceum: a coopera-tive study of the International Working Group on Mycobacterial Taxon-omy. Int. J. Syst. Bacteriol. 43:482–489.

111. Wolinsky, E. 1984. Nontuberculous mycobacteria and associated diseases,p. 1141–1207. In G. P. Kuciba and L. G. Wayne (ed.), The mycobacteria: asourcebook. Marcel Dekker, Inc., New York, N.Y.

112. Wong, M. Y. H., P. A. Steck, and G. R. Gray. 1979. The major mycolic acidsof Mycobacterium smegmatis: characterization of their homologous series.J. Biol. Chem. 254:5734–5740.

113. Yost, R. W., L. S. Ettre, and R. D. Conlon. 1980. Practical liquid chroma-tography: an introduction. Perkin-Elmer Corp., Norwalk, Conn.



http://cmr.asm

.org/D

ownloaded from

http://cmr.asm.org/

Documents

Mycolic Acid Analysis by High-Performance Liquid Chromatography for Identification of