IN1 ERNA -I IONA! JOURNAL OF LEPROSY^

Volume 50, Number 2Printed in the 1/..S. . A .

Radiometric Measurement of DifferentialMetabolism of Fatty Acid by Mycobacteria'

Edwaldo E. Camargo, Judith A. Kertcher, Steven M. Larson,Byron S. Tepper, and Henry N. Wagner, Jr.'

A previous report from this laboratory onradiometric studies of Mycobacterium lep-raeuutrium described the differential me-tabolism of [1-' 1 C] fatty acids by this or-ganism ( - 1 ).

Because some radiometric similarities hadbeen found with M. lepraemurium and M.tuberculosi.s. ( 7 •' 3 ), the study of the [1-"C]fatty acid series was extended to includethe latter. The striking differences in theoxidation rates of these substrates ob-served between M. le'praemrriinnn and M.tuberculosis H,Rv led to the hypothesis thatoxidation patterns of fatty acids might pro-vide a basis of species differentiation of thegenus Mycobacterium ( '). To investigatefurther this possibility, two additional or-ganisms (M. bolls BCG and M. tubercu-losis Erdman) were included in this study.

The present report describes the resultsof this investigation.

MATERIALS AND METHODSOne fully susceptible strain of M. tuber-

culosis was obtained from the Johns Hop-kins Hospital Mycobacteriology Labora-tory (F1„,Rv). Another fully susceptiblestrain of Al. tuberculosis (Erdman) and onestrain of M. bort* s (BCG) resistant to 5 /..tgof isoniazid were obtained from Dr. Gard-ner Middlebrook at the University of Mary-land. These organisms were grown in 50 mlof liquid 7H9 medium with 10% ADC en-richment and 0.05% polysorbate 80. Colo-nies were scraped from Lowenstein-Jensen

' Received for publication on 24 April 1981: accept-ed for publication in revised form on 10 November1981.

E. E. Camargo, M.D., Assistant Professor; J. A.Kertcher, Research Technician: S. M. Larson, NI. D.,Professor: B. S. Tepper, Ph.D., Associate Professor:H. N. Wagner, Jr., M.D., Professor, Division of Nu-clear Medicine and the Leonard Wood Memorial Lep-rosy Research Laboratory, The Johns Hopkins Med-ical Institutions, Baltimore, Maryland, U.S.A. S. M.Larson's present address: University of Washington,Seattle, Washington, U.S.A.

slants and transferred to the liquid growthmedium where they were grown for six toeight days. The bacteria were then homog-enized with a Sorvall omni-mixer at 8.5speed scale twice for 30 sec. The numberof bacteria was estimated with MacFarlandbarium sulfate standards C). The final sus-pension was further diluted with sterile 7H9to yield from I x 10 7 to 5 x 10 7 organisms/ml.

Experimental media. The liquid 7H9 me-dium with 10% . ADC enrichment and nopolysorbate 80 was used as the suspendingsolution for the organisms.

Reaction system. The reaction system fordetection of "CO, consisted of 0.8 ml ofmedium in a 5.0 ml multi-dose sterile glassvial (A. H. Thomas Co., Philadelphia,Pennsylvania, U.S.A.) with airtight alumi-num seals (Wheaton Scientific, Millville,New Jersey, U.S.A.) fitted with rubber lin-ers (Johnston Laboratories, Cockeysville,Maryland, U.S.A.), along with 0.1 ml ofbacterial suspension and 0.1 ml (1.0 Ki) of''C-substrate. The following [1-"C] fattyacids (Amersham Corporation, ArlingtonHeights, Illinois, U.S.A.) were used withall organisms: butyric (25 mCi/mM), hex-anoic (23.6 mCi/mM), octanoic (22.5 and31.8 mCi/mM), decanoic (18.7 and 21 mCi/mM), lauric (28.8 mCi/mM), myristic (38and 45 mCi/mM), palmitic (57.9 mCi/mM),stearic (58.4 and 59.7 mCi/mM), oleic(51.8 and 58 mCi/mM), linoleic (56 and61 mCi/mM), and linolenic (60 mCi/mM).Lauric acid was dissolved in methanol; allother fatty acids were dissolved in ethanol."C-Formate (2.83 mCi/mM) was includedas an indicator of the metabolic activity ofthe organisms ("). All vials were preparedat least in duplicate. Control vials were pre-pared in the same way, but with autoclavedbacteria added.

Radiometric measurement. The vials wereincubated at 37°C. An ion chamber device(Bactec R-301, Johnston Laboratories,

200

50, 1^Camargo, et al.: Fatty Acid Metabolism^201

Cockeysville, Maryland, U.S.A.) was usedto measure the bacterial metabolism ("). Thevials were sampled daily for at least 3 days.The results were expressed as "index units —

(100 units = 0.025 /Xi of ''C activity).Means and standard deviations of the cu-mulative "CO, production for each sub-strate over the entire period of the experi-ment were calculated. Results wereconverted to percent, using the cumulative"CO, production of the best [ I-"C] fattyacid for each organism as 100%.

For M. lepraemurium, the details of thepreparation of organisms, experimental me-dia, reaction system, and radiometric mea-surement have been published previously( 1 ).

Sterility testing. Sterility tests were per-formed on positive samples and consistedof subcultures on chocolate-agar plates, onLowenstein-Jensen medium and radiomet-ric sterility testing with [U-"C] glucose( M.9.1u ).

Assimilation of substrates. At the end ofM. tuberculosis H„,Rv experiments, the in-corporation of ''C-substrates into the bac-teria was measured by liquid scintillationcounting. Details of this procedure havebeen published elsewhere (').

RESULTSTable 1 represents the assimilation and

oxidation patterns of [1-"C] fatty acids byM. tuberculosis H„,Rv. With some of thesaturated fatty acids such as butyric, hex-anoic, and lauric, there was also a directrelationship between incorporation and"CO., production. The amount in auto-claved bacteria increased with the carbonchain length for saturated acids; with lino-leic acid, 98% of the activity was presentin heat-killed organisms and no radioactiv-ity was found in the medium.

Table 2 summarizes the oxidation pat-terns of [1-"C] fatty acids by mycobacteria.Results previously published on M. leprae-minium have been included for comparison(1 ). With this organism, "CO., productionprogressively increased with the carbonchain length, reached a maximum with lau-ric and decreased to a minimum with theunsaturated molecules. In decreasing or-der, "CO, production by this organism wasgreatest with lauric, decanoic, myristic, oc-tanoic, and stearic.

TABLE 1. Assimilation and oxidationpatterns 411-"Cllatty acids by M. tuber-culosis H„ 7/6.."

Substrate Bacteria Medium CumulativeCO,

Formate' 23 75Control' 0 100 0Butyric 36 42Control 0 100 0Hexanoic 20 42 38Control 0 100 0Oclanoic 0 97 3Control 99 0Decanoic 0 98Control 4 96 0Lauric 6 70 24Control 1 98 0Myristic 0 98Control 3 97 0Palmitic 96Control 4 96 0

Stearic 96 1

Control 7 93 0Oleic 1 97 1

Control 7 93 0Linoleic 1 98Control 98 0Linolenic 1 98Control 98 0

Results expressed as percent of the total reactivityinitially added which was found incorporated into thebacteria, left in the medium. and converted to "CO,at the end of the incubation.

'' Containing viable bacilli.Containing the autoclaved bacilli.

With both strains of M. tuberculosis,hexanoic gave the greatest "CO, produc-tion. However, the Fl„,Rv strain also oxi-dized lauric, butyric, and decanoic; where-as Erdman oxidized lauric, myristic andstearic. With M. boris ( BCG) the bestsubstrates were lauric, decanoic, linolenic,and linoleic, in decreasing order. With M.tuberculosis (H„,Rv and Erdman) and M.boils (BCG), however, none of these sub-strates was better than "C-formate, usedas reference substrate.

By using 20% of maximal oxidation asthe lower limit to consider the cumulative"CO, production as positive, the results onTable 2 can be simplified (Table 3). Ac-cording to Table 3, these four organismscan be differentiated by using butyric, hex-anoic, and stearic acids. As expected, "CO,production was not observed in the controlvials and all sterility tests were negative.

202

TABLE 2.

international Journal ul Leprosy^ 1982

Oxidation patterns . of [I-"Clfatty acids by mycobacteria."

Substrate A/. /eproemmitun A/. bolls(13CG)

Al. tuberculosis. M. tuberruh,viA(Erdman)

Formate =-100 >100 >1(X)Butyric 0 4±^2 36 ± I 0I exanoic I 9 ±^2 10(1 ± 6 100 ±^3Octanoic 29 8 ±^3 17 ± 3 to ±^0I)ecanoic 63 36 ±^5 23 ± 1 6 ±^0Laurie 100 100 ± 12 76 ± 4 37 ± 15Myristic 55 4 ±^1 18±11 20 ±^3Palmitic 17 4±^I 9 ± 1 9±^2Stearic 26 5 +^0 7 ± 1 13 ±^0Oleic 13 I ±^0 8 ± 1 9 ±^4Linoleic 19 ±^3 7 ± 1 5 -±^ILinolenic 21 ±^I 4 ± 1 5^I

Cumulative "CO, production as percent of the best I I-"Cl fatty acid for each organism.'' Al. lepraenutrium does not metabolite formate ( 7 ).

DISCUSSIONWe have previously reported and dis-

cussed the oxidation of the fatty acid seriesby M. lepraemu•ium ('). Most of our resultswere found to be in agreement with thestudies of Kusaka who described fatty acidsynthesizing enzyme activity in extracts ofthis organism ('').

There are several reports on the fatty acidsynthesizing enzyme activity in extracts ofM. tuberculosis and other mycobacteria

' 7 ). Briefly, these studies haveshown that several long chain fatty acidsare formed, from 12 to 26 carbons, by ad-dition of two carbons to the acceptor fattyacid's carboxyl end. Acetate, malonate, andNADH are essential for these reactions.Also, there is evidence that M. tuberculosisaccumulates lipids during exponentialgrowth (I), with a synthesis of mycolic acidsthat parallels the growth rate ( 3). With M.

tubercutoNis H„Rv there was utilization ofmost of the substrates as both carbonsources and energy sources (Table 1). Theexceptions were formate and lauric, usedmainly for energy purposes. The oxidationof formate was probably due to the pres-ence of a formate dehydrogenase alreadyidentified in M. pith , / ("). Whether the pref-erential oxidation of lauric over its use ascarbon source may have the implicationsdiscussed with M. lepraemurium is un-known at present. The considerations onthe possible mechanisms of elimination oftoxic substances by Al. lepraemurium can-not be extended to M. tuberculosis and M.boils, since no radiochromatographic anal-ysis of the media was done for the lattertwo. Although adsorption may have influ-enced assimilation of saturated long chainfatty acids (Table 1), the enzyme systemsinvolved with oxidation of fatty acids in both

TABLE 3. Oxidation patterns of selected [I-''Cljatty acids by mycobacteria."

Al. lepraenurrium(Hawaiian)

M. bolls(BCG)

Al. tuberculosis(R,Ftv)

Al. tuberculosis(Erdman)

Butyric 0 0 0Hexanoic 0 ()Octanoic 0 0 0Decanoic 0 0LaurieMyristic 0 () 0Stearic 0 0 (1

" + = >20(i oxidation of the best I 1-"C] fatty acid for each organism. (1 = <20rii oxidation of the best11-"CI fatty acid for each organism.

50, 2^ Camari,,o, et al.: Fatty Acid Metabolism^ 203

strains of M. tuberculosis showed prefer-ence for the six-carbon chain (hexanoicacid). On the other hand, the enzyme sys-tems of M. boas preferred the 12-carbonchain (lauric acid), as did M. lepraemu-rium.

The enzyme systems of the various or-ganisms used in this study did not oxidatethe entire I I-"C] fatty acid series at the samerates. Moreover, the different oxidationrates for the various fatty acids also variedfrom one organism to another. Differentialoxidation patterns could therefore be rec-ognized. By using a 20% threshold of max-imal oxidation rate to consider the cumu-lative ''CO production as positive (Table3), it becomes apparent that by selectingthree convenient fatty acids it is possible todifferentiate the various organisms used inthis study. This simpler approach only in-cludes butyric, hexanoic, and stearic acids.Only stearic is positive with M. lepraenut-that!: butyric and hexanoic are positive withM. tuberculosis 1-1,,,Rv: only hexanoic ispositive with M. tuberculosis Erdman, andnone of them are positive with M. boris.

There is evidence that the uptake of iso-niazid by tubercle bacilli, an enzyme-de-pendent phenomenon is followed by afall in the incorporation of(I-"CI acetateinto fatty acids (' 2 ). This suggests some re-lationship between the action of the drugand the metabolism of fatty acids. It is con-ceivable that isoniazid-resistant and isoni-azid-susceptible organisms of the samestrain of M. tuberculosis may show differ-ent patterns of fatty acid oxidation. In thiscase, a comparison of the oxidation pat-terns of selected fatty acids by these organ-isms would bring new information on themetabolic pathways involved with themechanism of susceptibility and resistanceto isoniazid.

SUMMARYAn assay system has been developed

based on automated radiometric quantifi-cation of ' 'CO, produced through oxidationof [I-HCI fatty acids by mycobacteria.

Two strains of M. tuberculosis (F1,,,Rv andErdman) and one of M. boris (BCG) in 7H9medium (ADC) with I .0 ,u.Ci of one of thefatty acids (butyric, hexanoic, octanoic,decanoic, lauric, myristic, palmitic, stearic,oleic, linoleic, and linolenic) were studied.

Results previously published on M. leprae-miium (Hawaiian) were also included forcomparison.

Both strains of M. tuberculosis had max-imum ''CO production from hexanoic acid.Oxidation of butyric and avid oxidation oflauric acids were also found with the H„ 7 Rvstrain but not with Erdman. In contrast,''CO production by M. boris was greatestfrom lauric and somewhat less from deca-noic acid. M. lepraemurium showed in-creasing oxidation rates from myristic, dec-anoic and lauric acids. Assimilation studiesof M. tuberculosis FI,,Rv confirmed thatmost of the oxidized substrates were con-verted into by-products with no change inthose from which no oxidation was found.

These data suggest that the radiometricmeasurement of differential fatty acid me-tabolism may provide a basis of strain iden-tification of the genus Mycobacterium.

RESUMENSe desarrolki un ensayo para la cuantificaciOn ra-

diometrica del "CO, producido por Ia oxidaciOn mi-cohacteriana de acidos grasos marcados en elC-1 ("C-1).

Sc cstudiaron 2 cepas de Al. tuberculosis (H„ 7 Rv yErdman) y una de M. boils (BCG) en medio 7H9 (ADC)con 1.0 pCi de uno de los siguientes acidos grasos:butirico, hexanoico, octanoico, decanoico, laurico,miristico, palmitico, estearico, oleico, linoleico ylaic°. Para comparacitin, tambien se incluyerongunos resultados previamente publicados obtenidos conel Mycobacterium lepraemurium, cepa Hawaii.

Las dos cepas de Al. tuberculosis tuvieron Ia max-ima producitin de "CO, cuando se use el acido hex-anoico. Con la cepa 11„,Rv, pero no con Ia Erdman,tambien se observ6 la oxidackin de los acidos but iricoy laurico. En contraste, la producciOn de "CO 2 por elAI. boils fue maxima con el acid° laurico y algo menorcon el acid° decanoico. El Al. lepraettutrium mostroun increment° en so capacidad de oxidaciOn que foe,del acid() miristico al acid° decanoico, hasty el acid()laurico. Los estudios de asimilackin con el M. tuber-culosis H„,Rv confirmaron que Ia mayoria de los sub-stratus oxidados fueron convertidos en biproductos entanto que no hobo camhio en aquellos que no fueronoxidados.

Estos datos sugieren que Ia medicitin radiometricadiferencial del metabolism° de los acidos grasos puedeproporcionar una base para Ia identificaciOn de cepasdel genero Mycobacterium.

RESUME

On a mis au point one methode experimentaledevaluation bases clue la quantification radiometrique

204^ International Journal of Leprosy^ 1982

automatisee du "CO, produit a Ia suite de l'oxydation

par les mycobactéries des acides-gras 1-"C.

Deux souches de M. tuberculosis (11„,Rv et Erd-

man), ainsi qu'une souche de M. boric (BCG), ont etcétudiees en milieu 7H9 (ADC), august avait ("Me :Owe

1,0 p.Ci de run des acides-gras suivants: butyrique,hexanoique, decanoique, laurique, myristique, palmi-

tique, stearique, oleique, linoleïque, et linolcnique. Des

resultats publics précedemment stir A/. lepraemurimn

(Hawaii) out également eté repris a des fins de com-

paraison. Lune et l'autre souche de A/. tuberculosis

prcsentaient une production maximale de "CO„ avec

l'acide hexanoique. On a egalenrent observe uneoxydation de l'acide butyrique (avid oxidation) des ac-

ides lauriques, avec la souche Fl„,Rv, mail non avecla souche Erdman. Au contraire, la production de

"CO, par M. boric etait la plus prononcée avec l'acidelaurique, et tin peu [twins prononcee pour lacide

decanolque. M. /epraemurium temoigne d'une aug-mentation des taux d'oxydation avec les acides myris-

tiques, décanoiques et lauriques. Les etudes d'assim-dation menses stir la souche F1 3 ,1(v de M. tuberculosis

out conlirmé que Ia plupart des substrats oxydes étaienttransformes en sous-produits, alors qu'aucune trans-

formation n'etait relevée avec les produits qui ne té-moignaient pas d'oxydation.

Ces données suggerent que les mesures radiome-triques des differences metaboliques des acicles-gras

pourraient fournir une base pour ['identification des

souches du genre Mycobacterium.

Acknowledgments. The authors are indebted to Dr.

Gardner Middlebrook for supplying strains of M. tu-

berculosis Erdman and M. boris BCG and to Dr. Pa-

tricia Charache for supplying a susceptible strain ofM. tuberculosis (H„.,Rv) maintained in the Johns Hop-kins Hospital Mycobacteriology Laboratory.

REFERENCESI. ANTOINE, A. D. and TEPPER, B. S. Environmen-

tal control of glycogen and lipid content of My-

cobacterium tuberculosis . . J. Bacteriol. 100 (1969)

538-539.

2. BAILEY, W. R. and SCOTT, E. G. Diagnostic Mi-

crobiology, 2nd ed. St. Louis, Missouri: C. V.Mosby, 1970, p. 368.

3. BENNET, P. and ASSELINEAU, J. Influence de ragesur Ia teneur en acides gras a chaine ramifiee dubacile tuherculeux. Ann. Inst. Pasteur 118 (1970)324-329.

4. CANIARGO, E. E., KERTCIIER, J. A., LARSON, S.

TEPpER, B. S. and WAGNER, H. N., JR. Ra-

diometric measurement of differential metabolismof fatty acids by Mycobacterium lepraemurium.

Int. J. Lepr. 47 (1979) 126-132.

5. CAntARGo, E. E., LARSON, S. M., CHARActiE, P.,TEPPER, B. S. and WAGNER, H. N., JR. Currentstatus of radiometric detection of Al. tuberculosisand M. lepraemurium. J. Nucl. Med. 16 (1975)518.

6. CANIARGO, E. E., LARSON, S. M., TEPPER, B. S.and WAGNER, H. N., JR. Radiometric measure-ment of metabolic activity of Al. lepraemuriton.Appl. Microbiol. 28 (1974) 452-455.

7. CANIARGo, E. E., LARsoN, S. NI.. TErpER, B. S.and WAGNER, H. N., JR. Radiometric studies ofMycobacterium lepraemurium. Int. J. Lepr. 44(1976) 294-300.

8. DE131.,N NC, H. J., JR., ClIARACIIE, P. and WAG-

NER, II. N., JR. Automated radiometric mea-

surement of antibiotic effect on bacterial growth.Antimicrob. Agents Chemother. 2 (1972) 360-366.

9. DEBLANc, H. J., JR., DELAND, F. H. and WAG-

NER, H. N., JR. Automated radiometric detectionof bacteria in 2,967 blood cultures. Appl. Micro-biol. 22 (1971) 846-849.

10. DELAND, F. H. and WAGNER, H. N., JR. Auto-mated radiometric detection of bacterial growth inblood cultures. J. Lab. Clin. Med. 75 (1970) 529-534.

11. DIAHLE, R. R. and BARTON, L. L. Nicotinamideadenine dinucleotide-independent formate deity-drogenase in Mycobacterium phlei. Can. J. Mi-crobiol. 23 (1977) 125-130.

12. EIIINA, T., MUNAKATA, K. and MtyromiYA, M.Effect de l'isonizide stir ('incorporation de^ace-tate de sodium 1- 11C dans les acides gras de Iamycohacterie. C.R. Soc. Biol., Paris 155 (1961)1190-1192.

13. KERTCHER, J. A., CHEN, M. F., CilARAcilE, P.,HwANGito, C. C., CANIARGO, E. E., NIcINTYRE,P. A. and WAGNER, H. N., JR. Rapid radiometricsusceptibility testing of Mycobacterium tubercu-

losis. Am. Rev. Respir. Dis. 117 (1978) 631-637.14. KUSAKA, T. Fatty acid synthesizing enzyme ac-

tivity of cultured Mycobacterium lepraemurium.

Int. J. Lepr. 45 (1977) 132-138.15. PIERARD, A. and GorpmAN, D. S. Enzyme sys-

tems in the mycohacteria. 14. Fatty acid synthesisin cell-free extracts of Mycobacterium tubercu-

losis. Arch. Biochem. Biophys. 100 (1963) 56-65.16. WANG, L., KUSAKA, T. and GOLDMAN, D. S.

Elongation of fatty acids in Mycobacterium tu-

berculosis. J. Bacteriol. 101 (1970) 781-785.17. WINDER, F. G., BRENNAN, P. and RATLEDGE, C.

Synthesis of fatty acids by extracts of mycobac-teria and the absence of inhibition by isoniazid.Biochem. J. 93 (1964) 635-640.

18. WIMPENNY, J. W. T. 'lite uptake and fate of iso-niazid in Mycobacterium tuberculosis var boricBCG. J. Gen. Microhiol. 47 (1967) 389-403.