J UOEH（産業医科大学雑誌）35（ 2 ）: 109－117（2013） 109

［Original］

Adaptation of Mycobacteria on Solid, Egg-Based Media to Anaerobic Conditions and Characterization of Their Diagnostic Phenotypes

Takezo Udou

Laboratory of Clinical Microbiology, Department of Medical Technology II, School of Health Sciences, University of Occupational and Environmental Health, Japan. Yahatanishi-Ku, Kitakyushu 807-8555, Japan

Abstract : Recent evidence indicates that human pulmonary tuberculous granulomas are hypoxic, and a related dormancy model has been established for Mycobacterium tuberculosis using liquid media in tightly sealed tubes. In the present study, we examined the growth capacity of various mycobacterial species on solid, egg-based media under oxygen-limited conditions. During primary anaerobic cultivation on an Ogawa egg-based medium, all of the inoculated mycobacterial strains persisted for 10 weeks without detectable colony formation on the surface of the medium. Because these anaerobic cultures were restored and possessed the capacity to grow within 3 weeks follow- ing transfer to microaerobic conditions (5% or 10% oxygen tension), it was evident that the organisms persisted in a non-replicative dormant state as seen with anaerobic cultures in liquid media. The colonies grown under micro- aerobic conditions were also capable of growth under anaerobic conditions. Anaerobic growth could be observed only when the microaerobically grown colonies were subcultured under anaerobic conditions, suggesting that pre- liminary adaptation to the environment under reduced oxygen tensions is an essential process to enable growth under anaerobic conditions. A similar mode of anaerobic growth on slants of an egg-based medium was also observed with clinical isolates of M. tuberculosis. All of the clones that were adapted to grow in the anaerobic environments retained their species-specific colonial morphologic features, while, among the biochemical characteristics tested, heat-stable catalase activity and niacin accumulation in the medium disappeared during adaptation to anaerobic growth. Based on these results, we conclude that egg-based media actively support the anaerobiosis of mycobacteria exhibiting either a non-replicative dormant state or anaerobic growth, both of which are associated with latency and reactivation of tuberculosis. It was also evident that the use of egg-based media would be of great advantage in the diagnostic characterization of the reactivated clones of mycobacteria.

Keywords : Mycobacterium tuberculosis, hypoxic environment, anaerobiosis, latent infection, endogenous reactivation.

（Received December 5, 2012, accepted April 30, 2013）

Introduction hypothesized that these granulomatous lesions are the sites of dormant mycobacteria that are responsible for Mycobacteria are obligate aerobes, but Mycobac- latent tuberculosis infections. terium tuberculosis encounters hypoxic environments Based on recently established animal models dem- that occur in highly differentiated granulomas during onstrating that pulmonary granulomas are severely the course of infection [1]. Traditionally, it has been hypoxic [2, 3], oxygen depletion has been tested as a

Corresponding Author: Takezo Udou, Higashi-Chikushi Junior College, Department of Food and Nutrition, Shimoitozu, Kokurakita-Ku, Kitakyushu 803-8511, JAPAN. E-mail: [email protected] 110 T Udou

means of inducing dormancy in cultures of M. tuber- anaerobically on egg-based media and compared those culosis [4, 5]. The in vitro-induced dormancy models characteristics with aerobically grown cultures. were established using liquid media in tightly sealed tubes and have been used in a variety of studies, for ex- Materials and Methods ample to characterize the transcriptional dormancy-sur- vival regulator (DosR) the dormancy regulon [6, 7] and Bacterial strains the anaerobic nitrate reductase (Nar) system composed With the exception of M. bovis BCG (Tokyo) and M. of NarK2, NarX and NarGHJI [8, 9]. These mediators avium, strains derived from the American Type Culture are highly inducible under anaerobic conditions that in- Collection (ATCC: Manassas, VA, USA) were used hibit aerobic respiration and prevent bacterial replica- throughout these experiments. Among our stock cul- tion, enabling the organism to survive in a dormant state tures, the following eight strains were selected accord- within the host. ing to Runyonʼs classical groupings, which are based Imaging of pulmonary tuberculosis using computed on the pathophysiological and biochemical character- tomography and examination of surgically resected istics of mycobacteria [19]: M. tuberculosis H37Ra pulmonary lesions have revealed the histological het- (ATCC25177), M. bovis BCG (Tokyo) [20], M. kansasii erogeneity of lesions that can occur within a single in- (ATCC12478), M. scrofulaceum (ATCC15978), M. gor- fected individual [10-13]. Most of these tuberculous donae (ATCC23283), M. avium N-445 [20], M. fortui- lesions, including caseous lesions, cavities and cen- tum (ATCC19542), and M. vaccae (ATCC15483). Re- trilobular opacities, have been thought to harbor the cently isolated clinical strains of M. tuberculosis were pathogen [14]. However, the finding that M. tubercu- used for comparison of their growth characteristics with losis DNA is present not only in macrophages within the type strain during a course of anaerobiosis [20]. If tuberculous lesions, but also in other non-professional necessary, these stock cultures were purified prior to the phagocytic cells in normal lung tissue contradicts the experiments on Brain Heart Infusion Agar plates (Nis- dominant view that latent organisms exist only in old, sui, Co., Ltd, Tokyo) (for rapid growers) or 10% oleic classical tuberculous lesions [15]. In addition, recent acid-albumin-dextrose-catalase (OADC) supplemented work monitoring M. tuberculosis replication indicates Middlebrook 7H10 agar plates (Difco) (for slow grow- that the bacterial population, even in apparently stable ers), as described previously [20]. lesions, continues to undergo replication [16]. It is unlikely that there is a simple binary distribution of Media and culture conditions bacteria in replicating or non-replicating status, and, Colonies of freshly grown mycobacterial strains even within a single granuloma, there are likely to be were each inoculated onto two slants of Ogawa egg- multiple local microenvironments that support unique based medium (Nissui), a modification of Löwenstein- bacterial populations. Jensen medium [21]. The medium contains 1% (w/v) These results strongly suggest the existence of phys- mono-potassium hydrogen phosphate as a neutralizing iologically and/or biochemically distinct subpopula- agent for alkaline-treated sputum from patients with tions of bacteria, including dormant and anaerobically pulmonary tuberculosis. One of the two inoculated and aerobically growing status, within pulmonary tu- cultures of each strain was assembled in a transpar- berculous lesions composed of heterogeneous micro- ent plastic jar (BBL, Baltimore, MD, USA) that also environments. contained a residual oxygen indicator tablet and a In the present study, we aimed to examine the pro- catalyst bag for anaerobic cultivation [Mitsubishi Gas cess responsible for the adaptation of mycobacteria, on Chemical (MGC) Co., Inc., Tokyo]. The other inocu- egg-based media, to oxygen-limited conditions. Be- lated culture was assembled in a second plastic jar and cause bacterial phenotypes vary significantly based on incubated under microaerobic (10% oxygen tension) the conditions in their growth environment [17, 18], we conditions using a catalyst bag for Campylobacter sp. also examined the morphologic, physiologic and bio- (Anaeropac Campylo: MGC, Co.). Clinical isolates chemical characteristics of colonies that were grown of Clostridium tetani and Campylobacter jejuni were Anaerobiosis of Mycobacteria on Egg-based Media 111

inoculated onto GAM (Gifu Anaerobic Medium) agar transferred to a colorless, transparent plastic bag that plates (Nissui) and respectively included in the anaero- was tightly sealed and also contained a catalyst bag bic and microaerobic cultivation jars as growth indi- for anaerobic cultivation and an oxygen indicator. The cator strains. The airtight plastic jars were incubated surface of the growth in the plastic bag was then ex- at 35°C in the dark, and growth (colony appearance) posed for 60 min to a bright light (150 W) that was of each strain on the slants of egg-based medium was placed 50 cm from the bag, and incubated at 35°C for observed from outside of the jar. After incubation for 2 to 3 days with daily examination for photo-induced 8 to 10 weeks, the primary anaerobic cultures were pigmentation. transferred to microaerobic environments to confirm Standard procedures for Ziehl-Neelsen acid-fast the possible recovery of their growth. In addition to staining and conventional biochemical assays of niacin the Ogawa egg-based medium, OADC supplemented production, nitrate reduction, and heat stable (68°C) Middlebrook 7H10 or 7H11 agar plates were also used catalase production were employed to characterize the for comparison of their capabilities to support growth anaerobically grown mycobacterial cultures. For the ni- with the egg-based medium under both anaerobic and acin test, benzidine was used instead of aniline to avoid microaerobic conditions. The colonies that appeared plausible positive reactions due to contamination by ca- under microaerobic conditions (see below) were sub- rotenoid pigment in the chromogenic strains. All of the cultured onto a fresh Ogawa egg-based medium or procedures used are routine techniques in our laboratory Middlebrookʼs agar-based medium and then incubated and appear elsewhere in the literature [22, 23]. anaerobically in an airtight jar, as described above. Results Staining and biochemical testing The colonial morphologic features of the mycobac- Table 1 summarizes the tested characteristics of terial strains that grew successfully under anaerobic each strain grown under aerobic and anaerobic condi- conditions (see below) were characterized with regard tions. The results shown in the table are based on the to their pigmentation (scotochromogenicity) and the following evidence observed during characterization roughness or smoothness of the colony surface. For of these cultures. photochromogenicity, non-pigmented cultures were

Table 1. Phenotypic characterization comparing mycobacterial cultures grown under anaerobic conditions with cultures grown aerobically Characteristic M. tuberculosis H37Ra M. bovis BCG (Tokyo) M. kansasii ATCC12478 M. gordonae ATCC23283 aerobic anaerobic aerobic anaerobic aerobic anaerobic aerobic anaerobic Growth + + + + + + + + Colonial morphology† R R R R R R S S Pigmentation‡ N N N N Pc Pc Sc Sc Acid-fastness + + + + + + + + 68°C Catalase +w – – – + – + – Nitrate reduction + +w +w +w + +w – – Niacin production + – +w – – – – –

Characteristic M. scrofulaceum ATCC15978 M. avium N-445 M. fortuitum ATCC19542 M. vaccae ATCC15483 aerobic anaerobic aerobic anaerobic aerobic anaerobic aerobic anaerobic Growth + + + + + + + + Colonial morphology† S S S S R R S S Pigmentation‡ Sc Sc N N N N Sc Sc Acid-fastness + + + + + + + + 68°C Catalase + + +w – + – + – Nitrate reduction +w +w – – +w +w +w +w Niacin production – – – – – – – – +: Positive, +w: Weakly positive, –: Negative, †R: Rough, S: Smooth, ‡N: No pigmentation, Pc: Photochromogenic orange, Sc: Scotochromogenic orange 112 T Udou

Growth conditions mogenic strains (M. scrofulaceum, M. gordonae, and When the eight strains of mycobacteria were inoc- M. vaccae) grown anaerobically. ulated onto the slants of the egg-based medium and incubated under anaerobic conditions, none of them Acid-fastness formed detectable colonies during 10 weeks of culti- It is well known that the presence of large amounts vation. However, incubation under microaerobic con- of lipids, such as trehalose-6,6-dimycolate and arabi- ditions (10% oxygen tension) enabled full growth with nogalactan mycolic acids, in the mycobacterial cell colony formation within 3 weeks. Similar growth was wall contributes to the acid-fast staining feature of observed under an atmosphere with 5% oxygen ten- these bacteria. When mycobacterial strains grown in sion, which is the recommended condition for growth an anaerobic environment were stained by the tradi- of Helicobacter sp. (Anaeropac Helico: MGC, Co.). tional Ziehl-Neelsen method, they all maintained their In addition, when the cultures that did not grow dur- acid-fastness, suggesting that no quantitative and/or ing 10 weeks of incubation under anaerobic conditions qualitative modification of the cell wall lipids that were transferred to a microaerobic environment (10% would alter acid-fastness occurred, despite conditions or 5% oxygen tension), they, too, grew and formed that are considered to be inappropriate for mycobacte- colonies within 3 weeks. Thus, the organisms that rial growth. were inoculated onto the egg-based medium remained alive in a non-replicative dormant state throughout the Biochemical characteristics 10 weeks of primary anaerobic incubation. To understand possible metabolic plasticity during The most important observation was that colonies the bacterial adaptation to anaerobiosis, three bio- grown under microaerobic conditions could acquire chemical characteristics known to be especially im- the ability to grow anaerobically; that is, when the portant for differentiating clinical isolates of mycobac- microaerobically grown colonies were subcultured teria were conventionally evaluated in the aerobically onto a fresh egg-based medium and incubated under and anaerobically grown cultures (Table1). anaerobic conditions, every strain examined grew Virtually all mycobacteria produce positive reactions successfully, though the initiation of their growth in semiquantitative catalase tests, but species can be dif- was somewhat delayed. These characteristic growth ferentially characterized in the presence of a heat-stable patterns observed during anaerobiosis could also be (68°C) catalase [22]. Interestingly, with the exception confirmed in another strain of M. tuberculosis H37Rv of M. scrofulaceum (ATCC15978), which is known to (ATCC25618), and in recent clinical isolates of M. tu- produce multiple heat-stable catalases [24], the activity berculosis (data not shown). Interestingly, agar-based of 68°C catalase observed in aerobic cultures was lost media such as Middlebrook 7H10 or 7H11 medium upon their adaptation to anaerobic growth. However, did not support adaptation to anaerobic growth of any semiquantitative catalase tests yielded positive reactions of the strains used. in all anaerobically and aerobically grown cultures (data not shown), suggesting that the enzymes detected by Colonial features this test, but not by heat-stable catalase, are synthesized Mycobacterial colonies exhibit unique features that constitutively, irrespective of the oxygen tension in the are characteristic of their species. Macroscopically, growth environment. In contrast to heat-stable cata- for example, a colony may have a rough or smooth lase, nitrate reductase activity was observed in all an- surface, or may appear dry or wet with brilliance. aerobically grown cultures of strains that also showed Comparison of such features between colonies grown positive reactions when aerobically grown. Converse- anaerobically and aerobically revealed no strain in ly, no strain that was negative for nitrate reductase ac- which the roughness or smoothness was altered dur- tivity when grown aerobically showed induction of ni- ing anaerobic growth. Similarly, another colonial fea- trate reductase during adaptation to anaerobic growth. ture, the ability to produce carotenoid pigments, was It is well known that all mycobacteria have the ability retained in both photo- (M. kansasii) and scotochro- to produce niacin (nicotinic acid) as a component of Anaerobiosis of Mycobacteria on Egg-based Media 113

nicotinamide adenine dinucleotide (NAD) and nico- for a long period of time in the anaerobic environment. tinamide adenine dinucleotide phosphate (NADP), In addition, survival of M. tuberculosis during anaero- and the M. tuberculosis complex strains produce much bic dormancy requires the availability of nutrients more niacin than other strains [25, 26]. Although M. (e.g., triacylglycerol or fatty acids as carbon sources) tuberculosis and M. bovis BCG (Tokyo) tested posi- that support the DosR regulon-assisted metabolic ho- tive for niacin accumulation when grown aerobically, meostasis that is essential for the maintenance of intra- they tested negative when grown anaerobically. The cellular energy levels and redox balance [28, 29]. It negative niacin test in the anaerobic cultures of these therefore seems likely that the egg-based medium used two strains may have resulted from reduced niacin ac- in the present study was sufficient to provide these nu- cumulation in the medium. tritional requirements. It has been reported that smooth and rough myco- Discussion bacterial colony variation is closely associated with the deletion of species-specific glycolipid antigens on It is known that the abrupt transfer of vigorously their cell surfaces [30]. Our results, which are sum- aerated cultures of M. tuberculosis to anaerobic condi- marized in Table 1, clearly indicate that adaptation to tions results in their rapid death, but gradual depletion anaerobic growth is not necessarily accompanied by of available oxygen permits development of a toler- alterations in colonial morphologic features, including ance to anaerobiosis [27]. In the present study, howev- photo- and scotochromogenic pigmentation and acid- er, primary anaerobic culture on slants of an egg-based fastness. medium enabled mycobacterial cells to persist for long There are two types of mycobacterial catalase: heat- periods of time (～10 weeks) without active replica- labile and heat-stable [22]. Although the precise role tion, as evidenced by the revival of growth upon trans- of intracellular catalase in the overall metabolism of fer to a microaerobic environment (5% or 10% oxygen mycobacteria is unknown, the presence of the enzyme tension). We also found that colonies grown in micro- does not appear to be essential for their growth in vi- aerobic environments acquired the ability to grow un- tro, as isoniazid-resistant strains of M. tuberculosis der anaerobic conditions on slants of egg-based medi- lack both catalase types [31]. In the present study, the um. That is, microaerobic culture was an essential step activity of the heat-stable catalase disappeared during for adaptation of mycobacteria to anaerobic growth, adaptation to anaerobic growth, suggesting that pro- and all microaerobically grown mycobacteria subse- duction of this enzyme is in some way regulated by quently grew well under anaerobic conditions on the oxygen in the growth environment. egg-based medium, without entering a non-replicative On the other hand, the activity of nitrate reductase, dormant state. It is highly interesting to note in these which plays a key role in anaerobic respiration, per- results that, because the observed growth characteris- mitting survival in a non-replicative dormant state tics in type strains could also be confirmed in clinical under hypoxic conditions, is greatly enhanced by the isolates of M. tuberculosis, the dormancy and active hypoxic state, but this enhancement does not reflect an replication of M. tuberculosis during anaerobiosis may increase in expression of the nitrate reductase enzyme reflect the latency and endogenous reactivation of tu- encoded by narGHJI [8]. Notably, in that regard, berculosis in the pulmonary lesions of patients. there was no discernible enhancement of such enzy- OADC-supplemented Middlebrook 7H10 or 7H11 matic activities in any of the anaerobic cultures tested agar-based media did not support anaerobic growth or (Table 1). This finding is consistent with the fact that its revival after transfer to microaerobic conditions: the activity detected by the diagnostic method used is there was no detectable colony formation, indicating that of NarGHJI [32]. We suggest that, because an en- hypoxic death of inoculated organisms on the plates. hancement in nitrate reductase activity under hypoxic These results suggest that the egg-based medium (pos- conditions would likely reflect induction of the nitrate sibly one or more of the egg components) may have a transporter encoded by narK2, and because NarK2 is key stabilizing effect that enabled the cells to survive active only in the absence of oxygen or when aerobic 114 T Udou

respiration is inhibited [33], a precise quantitative as- say (rather than conventional diagnostic testing) might Acknowledgements be more useful for assessing the possible induction of nitrate reductase activity under anoxic conditions. The author thanks Tomoko Hiraoka and Satoko Because nitrate added to anaerobic M. tuberculosis Shiota for their excellent technical assistance and Prof. cultures had no effect on nitrate reductase activity and Haruaki Tomioka of Shimane University School of did not enable anaerobic respiration [34], the presence Medicine for helpful advice and encouragement dur- of electron-transport bypass mechanisms that use ter- ing the course of this research. This article is dedi- minal electron acceptors other than nitrate cannot be cated to my wife with love and gratitude. excluded [35]. One recent report demonstrated that expression of ferredoxins and a hydrogenase provides References a potential conduit for disposing of and transferring electrons in the absence of exogenous electron ac- 1 . Russell DG, Barry III CE & Flynn JL (2010): Tuber- ceptors [36], which would rationalize the remarkable culosis: What we donʼt know can, and does, hurt us. metabolic plasticity of mycobacterial adaptation to an- Science 328: 852-856 aerobiosis. 2 . Aly S, Wagner K, Keller C, Malm S, Malzan A, Bran- The biochemically functional forms of niacin (nico- dau S, Bange FC & Ehlers S (2006): Oxygen status tinic acid) are components of the nicotinamide nucleo- of lung granulomas in Mycobacterium tuberculosis- tide coenzymes (NAD and NADP). In both aerobic infected mice. J Pathol 210: 298-305 and anaerobic respiration, these coenzymes, especially 3 . Via LE, Lin PL, Ray SM et al (2008): Tuberculous NAD, play an important role in the oxidation-reduc- granulomas are hypoxic in guinea pigs, rabbits, and tion systems coupled to the energy-yielding reactions nonhuman primates. Infect Immun 76: 2333-2340 of the cell [37]. We suggest that the negative niacin 4 . Rustad TR, Harrell MI, Liao R & Sherman DR (2008): production test results from the anaerobic cultures of The enduring hypoxic response of Mycobacterium tu- M. tuberculosis and M. bovis BCG (Tokyo) reflecting berculosis. PLoS One 3: e1502 a reduced accumulation of niacin. Therefore, a quanti- 5 . Wayne LG & Hayes LG (1996): An in vitro model for tative assay for accumulated niacin will be required to sequential study of shiftdown of Mycobacterium tu- test this hypothesis. berculosis through two stages of nonreplicating persis- In conclusion, the present study demonstrates that tence. Infect Immun 64: 2062-2069 all of the mycobacterial strains examined, including 6 . Leistikow RL, Morton RA, Bartek IL, Frimpong I, clinical isolates of M. tuberculosis, are capable of Wagner K & Voskuil MI (2010): The Mycobacterium growth or persistence in a non-replicative dormant tuberculosis DosR regulon assists in metabolic homeo- state under anaerobic conditions on slants of an egg- stasis and enables rapid recovery from nonrespiring based medium. This is the first report indicating that dormancy. J Bacteriol 192: 1662-1670 mycobacteria exhibit two modes of adaptation pat- 7 . Sherman DR, Voskuil M, Schnappinger D, Liao R, terns, active replication and non-replicative persis- Harrell MI & Schoolnik GK (2001): Regulation of the tence, during their anaerobiosis on solid, egg-based Mycobacterium tuberculosis hypoxic response gene media. Although the clinical significance of the po- encoding alpha-crystallin. Proc Natl Acad Sci USA tential plasticity of mycobacteria may be limited to 98: 7534-7539 the isolates recovered from patients with chronic and 8 . Sohaskey CD & Wayne LG (2003): Role of narK2X persistent infections, rapid identification followed by and narGHJI in hypoxic upregulation of nitrate reduc- characterization of the reactivated pathogens should tion by Mycobacterium tuberculosis. J Bacteriol 185: be required for appropriate diagnosis in clinical mi- 7247-7256 crobiology laboratories. Accordingly, detailed char- 9 . Wayne LG & Hayes LG (1998): Nitrate reduction as acterization of the phenotypes of anaerobically grown a marker for hypoxic shiftdown of Mycobacterium tu- cultures is currently ongoing. berculosis. Tuberc Lung Dis 79: 127-132 Anaerobiosis of Mycobacteria on Egg-based Media 115

10 . Long R, Maycher B, Dhar A, Manfreda J, Hershfield American Society for Microbiology. Washington, DC E & Anthonisen N (1998): Pulmonary tuberculosis pp 400-437 treated with directly observed therapy: Serial changes 23 . Roberts GD (1994): Mycobacteria. In: Diagnostic mi- in lung structure and function. Chest 113: 933-943 crobiology, 9th (Baron EJ, Peterson LR & Finegold 11 . Poey C, Verhaegen F, Giron J, Lavayssiere J, Fajadet SM, ed ). Mosby, London pp 590-633 P & Duparc B (1997): High resolution chest CT in tu- 24 . Mayer BK & Falkinham III JO (1986): Catalase activ- berculosis: evolutive patterns and signs of activity. J ity and its heat inactivation for differentiation of Myco- Comput Assist Tomogr 21: 601-607 bacterium avium, Mycobacterium intracellulare, and 12 . Kaplan G, Post FA, Moreira AL et al (2003): Mycobac- Mycobacterium scrofulaceum. Int J Syst Bacteriol 36: terium tuberculosis growth at the cavity surface: A mi- 207-212 croenvironment with failed immunity. Infect Immun 25 . Konno K, Oizumi K & Oka S (1965): Biosynthesis of 71: 7099-7108 niacin ribonucleotide from quinolinic acid by myco- 13 . Ulrichs T, Kosmiadi GA, Jorg S, Pradl L, Titukhina M, bacteria. Nature 205: 874-875 Mishenko V, Gushina N & Kaufmann S H (2005): Dif- 26 . Konno K, Oizumi K, Shimizu Y & Oka S (1965): Nia- ferential organization of the local immune response in cin metabolism in mycobacteria: Quinolinic acid as a patients with active cavitary tuberculosis or with non- precursor of niacin ribonucleotide and the differences progressive tuberculoma. J Infect Dis 192: 89-97 in niacin biosynthesis among various species of myco- 14 . Loring WE & Vandiviere HM (1956): The treated pul- bacteria. Am Rev Respir Dis 91: 383-390 monary lesion and its tubercle bacillus. I. Pathology 27 . Wayne LG & Lin K-Y (1982): Glyoxylate metabolism and pathogenesis. Am J Med Sci 232: 20-29 and adaptation of Mycobacterium tuberculosis to sur- 15 . Hernandez-Pando R, Jeyanathan M, Mengistu G, vival under anaerobic conditions. Infect Immun 37: Aguilar D, Orozco H, Harboe M, Rook GAW & Bjune 1042-1049 G (2000): Persistence of DNA from Mycobacterium 28 . Daniel J, Deb C, Dubey VS, Sirakova TD, Abomoelak tuberculosis in superficially normal lung tissue during B, Morbidoni HR & Kolattukudy P (2004): Induction latent infection. Lancet 356: 2133-2138 of a novel class of diacylglycerol acyltransferases and 16 . Gill WP, Harik NS, Whiddon MR, Liao RP, Mittler JE triacylglycerol accumulation in Mycobacterium tuber- & Sherman DR (2009): A replication clock for Myco- culosis as it goes into a dormancy-like state in culture. bacterium tuberculosis. Nat Med 15: 211-214 J Bacteriol 186: 5017-5030 17 . Runyon EH (1970): Identification of mycobacterial 29 . McKinney JD, Honerzu BK, Munoz-Elias EJ, Miczak pathogens utilizing colony characteristics. Am J Clin A, Chen B, Chan WT, Swenson D, Sacchettini JC, Ja- Pathol 54: 578-586 cobs Jr WR & Russell DG (2000): Persistence of My- 18 . American Thoracic Society (2000): Diagnostic stan- cobacterium tuberculosis in macrophages and mice dard and classification of tuberculosis in adults and requires the glyoxylate shunt enzyme isocitrate lyase. children. Am J Respir Crit Care Med 161: 1376-1395 Nature 406: 735-738 19 . Runyon EH (1959): Anonymous mycobacteria in pul- 30 . Belisle JT & Brennan PJ (1989): Chemical basis of monary disease. Med Clin North Am 43: 273-290 rough and smooth variation in mycobacteria. J Bacte- 20 . Udou T (2006): Comparative evaluation of the in vitro riol 171: 3465-3470 antimycobacterial activities of six aminoglycoside an- 31 . Middlebrook G (1954): Isoniazid resistance and cata- tibiotics using an agar dilution method. J Chemother lase activity of tubercle bacilli: a preliminary report. 18: 610-616 Am Rev Tuberc 69: 471-472 21 . Collins CH, Grange JM & Yates MD (1985): Organiza- 32 . Stermann M, Bohrssen A, Diephaus C, Maass S & tion and practice in tuberculosis bacteriology. Butter- Bange F-C (2003): Polymorphic nucleotide within the worths, London pp 104-110 promoter of nitrate reductase (NarGHJI) is specific 22 . Nolte FS & Metchock B (1995): Mycobacterium. In: for Mycobacterium tuberculosis. J Clin Microbiol 41: Manual of clinical microbiology. 6th (Murray PR, 3252-3259 Baron EJ, Pfaller MA, Tenover FC & Yolken RH, ed). 33 . Sohaskey CD (2005): Regulation of nitrate reductase 116 T Udou

activity in Mycobacterium tuberculosis by oxygen and 36 . Berney M & Cook GM (2010): Unique flexibility in nitric oxide. Microbiology 151: 3803-3810 energy metabolism allows mycobacteria to combat 34 . Sohaskey CD (2008): Nitrate enhances the survival of starvation and hypoxia. PLoS One 5: e8614 Mycobacterium tuberculosis during inhibition of respi- 37 . Boshoff HI, Xu X, Tahlan K et al (2008): Biosynthesis ration. J Bacteriol 190: 2981-2986 and recycling of nicotinamide cofactors in Mycobacte- 35 . Ramakrishnan T, Murthy PS & Gopinathan KP (1972): rium tuberculosis. An essential role for NAD in non- Intermediary metabolism of mycobacteria. Bacteriol replicating bacilli. J Biol Chem 283: 19329-19341 Rev 36: 65-108

Anaerobiosis of Mycobacteria on Egg-based Media 117

小川培地上での抗酸菌の嫌気的培養環境への適応とその性状

有働 武三

産業医科大学 産業保健学部 第2生体情報学

要 旨：結核菌Mycobacterium tuberculosisは経気道感染後に肺胞マクロファージに取り込まれたのち，そこに 形成される低酸素環境の肉芽腫病巣内で非増殖性の休眠状態で数十年にわたって生き続けることが知られている． 液体培地を用いて確立されたin vitro休眠モデルを通じてそのような持続感染を可能にする分子機構も明らかにさ れつつある．本研究は液体培地に代えて加熱凝固卵を基調とした小川培地上での抗酸菌標準菌株を中心とした低酸 素環境への適応化を試みた．直接的な嫌気培養では10週の培養期間を通じて集落の形成を認めず，それを5％ある いは10％酸素分圧の微好気性環境に移すと3週以内での集落の形成がみられ，この初代嫌気培養を通じて非増殖性 の休眠状態で生残することがわかった．微好気性環境での発育に適応したクローンを新鮮な培地上に継代接種し嫌 気的環境で培養を続けると3週以内での集落の形成がみられ，固型培地上での抗酸菌の嫌気的発育現象が観察され た．同様の嫌気的環境への適応現象（休眠化および嫌気的発育）は結核菌の臨床分離株でも確認されたことから，卵 培地上で観察されたこれらの現象は結核の持続感染，再発症にいたる様式として興味深い結果である. また嫌気的 条件下で発育したクローンは生化学的性状試験において耐熱性カタラーゼ活性，ナイアシン産生に陰性化がみられ， 鑑別・同定上の有益な指標となった．

キーワード：結核菌，低酸素環境，嫌気的適応，持続感染，内因性再燃．

J UOEH（産業医大誌）35（2）：109－117（2013）