FEMS Microbiology Ecology 49 (2004) 343–347 www.fems-microbiology.org

Environmental mycobacteria from alpine and subalpine habitats

Marie-Francoise Thorel a, Joseph O. Falkinham, III b,*, R.G. Moreau c

a AFSSA-Alfort, BP 67, 22 rue Pierre Curie, 94703 Maisons-Alfort Cedex, France b Department of Biology, Virginia Polytechnic Institute, State University, Blacksburg, VA 24061-0406, USA c Universite Paris XII, 94000 Creteil, France

Received 29 January 2004; received in revised form 5 April 2004; accepted 9 April 2004

First published online 18 May 2004

Abstract

Mycobacteria were isolated from a variety of materials such as soil, peat, humus, tufa, sphagnum, and wood, collected in alpine and subalpine habitats. Mycobacteria, including kansasii, , , Mycobacterium gordonae, Mycobacterium terrae, , and were recovered from 69 of 81 (85%) samples. All of the isolates were recovered on medium incubated at 20 and 30 °C. None were recovered if the medium was incubated at 37 °C. The isolation of mycobacteria confirms the presence of these opportunistic pathogens in alpine habitats. Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

Keywords: Environmental mycobacteria; Alpine habitats; Growth temperature

1. Introduction tion to environmental extremes. Mycobacteria have been shown to grow over a wide range of pH [12–14], A wide variety of different mycobacterial species have temperature [15,16], and organic matter concentration been recovered from soil, water, vegetables, and dust [1– [15]. Therefore, it would be expected that mycobacteria 4]. In addition to water, soil is a likely reservoir of these could be recovered from extreme environments. Myco- environmental opportunistic mycobacteria [3,5]. Soil have been recovered from acidic forest soils in has yielded almost every mycobacterial species, with the Finland [5] and from coastal acidic swamps of the exception of members of the Mycobacterium tuberculosis eastern coast of the United States [17]. It is also likely complex [1,3–5]. The mycobacteria that occur most that mycobacteria are present in alpine environments frequently in soil are Mycobacterium fortuitum, Myco- and capable of cold-adaptation. For example, a psy- bacterium gordonae, Mycobacterium nonchromogenicum, chrotrophic strain of Rhodococcus has been reported and Mycobacterium terrae [1,3,6–8]. M. gordonae and [18]. Further, a homologue of the mycobacterial nidA Mycobacterium chelonae have been recovered from gene was amplified by PCR from DNA recovered from sphagnum vegetation [9], soils rich in sphagnum (e.g., pristine and hydrocarbon-contaminated alpine soils [19]. peats) [1,5], and from water draining from peats [10]. In fact, based on the presence of the mycobacterial nidA Mycobacterium malmoense has also been recovered from gene, it was postulated that mycobacteria might be soil [11]. stable members of microbial populations in resource- Recovery of the environmental opportunistic myco- limited habitats [19]. In this work, we report the recov- bacteria from the wide diversity of habitats reported in ery and identification of mycobacteria from a variety of the literature suggests that they are capable of adapta- samples collected in the alpine and subalpine region of Europe. The long-range objective of these investigations * Corresponding author. Tel.: +1-540-231-5931; fax: +1-540-231- is to identify the adaptations leading to survival of 9307. mycobacteria in a nutrient-poor, low temperature E-mail address: jofi[email protected] (J.O. Falkinham, III). region.

0168-6496/$22.00 Ó 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.femsec.2004.04.016 344 M.-F. Thorel et al. / FEMS Microbiology Ecology 49 (2004) 343–347

2. Materials and methods all culture, enzymatic, and biochemical tests were per- formed at 20 °C. There were no differences in the reac- 2.1. Environmental samples tions of the culture, enzymatic, or biochemical tests at 20 and 37 °C for the reference strains. However, longer Samples were collected from three Alpine regions, the incubation periods were required (up to 21 days). Jura Forest (peat bogs and meadows), the Vosges For- Comparison of mycolic acid profiles was performed as est, and the Mount White Massif and a subalpine re- described by Goodfellow and Magee [22] and David gion, the Belledone Massif. A total of 81 samples were et al. [23], and carried out on cells that had been grown collected: 44 samples of soil, 8 of peat, 11 of humus, 4 of at 20 °C. For some isolates, identification was confirmed tufa (porous rock), 8 of sphagnum, and 6 of rotting by comparison of patterns of BstEII and HaeII-restric- wood (Table 1). Peat is acidic, partly decomposed plant tion fragments of the PCR amplification product of the remains that accumulate in hollows and wherever the hsp-65 (65-kDa, heat shock protein) gene [24,25]. earth is water-saturated. Humus is organic matter in soils formed from the decomposition of the tissues of dead plants and animals. Samples were collected from 3. Results multiple areas within a site and mixed to produce composite samples. All were collected in the spring 3.1. Recovery of mycobacteria (May) and the ambient air temperature was 15 °C. After collection and preparation of composite samples, all Mycobacteria were recovered from every type of were stored at 4 °C. sample from the different sites and from 69 of the 81 (85%) samples collected. From 81 samples treated with 2.2. Isolation of mycobacteria 0.75% (w/v) HPC without enzymatic treatment, 67 (82.7%) yielded mycobacteria. From the same 81 sam- Sample (2 g) was suspended in 20 ml sterile distilled ples treated with HPC and cellulase, hyaluronidase and water and ground with sterile sand. The particulate amyloglucosidase 69 (85%) yielded mycobacteria. Al- material was allowed to settle on the laboratory bench though addition of enzymes to the isolation regimen for 1 h and the supernatant suspension collected (18 only resulted in a modest increase in the number of ml) to which 20 ml of 0.75% (w/v) hexadecylpyridinium samples yielding mycobacteria (above), the number of chloride (HPC) was added. After incubation at room isolates per sample was significantly (v2, p < 0:05) in- temperature for 18 h, the suspension was centrifuged creased from 74 to 120 isolates from the 81 samples by (5000g for 20 min) and the supernatant liquid discarded. polysaccharide use. This was particularly marked for The pellet was suspended in 1 ml of sterile distilled water soil samples, where polysaccharidase use increased the and distributed on Lowenstein–Jensen (LJ) medium number of isolates from 43 to 74 for the same soil slopes [20]. Because the objective was to isolate myco- samples. bacteria present in alpine and subalpine habitats, the LJ A majority of the isolates (110 of 138, 80%) were slopes were incubated at 20 and 30 °C, in addition to 37 recovered from LJ slopes incubated at 20 °C, whereas °C. To increase the number and variety of mycobacteria, only 20% were recovered at 30 °C (28 of 138). None polysaccharidases were used to release mycobacteria were recovered at 37 °C. The isolates recovered on LJ from particulate and colloidal matter in samples [20,21]. medium incubated at 20 °C failed to grow upon imme- Earlier studies had shown that enzymatic treatment did diate subculture at 37 °C, with the exception of 4 M. not reduce colony counts of cultures or recovery of fortuitum isolates. Mycobacteria were recovered from mycobacteria from soils [20,21]. In fact, the total num- samples whose pH was from 3.1 to 8.3 and whose ber and variety mycobacteria isolated was increased moisture content ranged between 19% and 80%. The [20,21]. Here, it was discovered that hyaluronidase re- highest numbers of isolates were recovered from samples duced the colony forming ability of only one strain of whose pH was below 5 (63%). M. chelonae by 10%. Further, hyaluronidase exposure led to fragmentation of clumps of M. chelonae, but did 3.2. Identification of mycobacteria not disrupt clumps of other mycobacterial species. The identification of 102 of the 120 isolates was de- 2.3. Identification of mycobacteria termined and the isolates were found to belong to eight species of mycobacteria, notwithstanding some differ- Mycobacteria were identified by standard culture, ences observed between the characteristics of the alpine enzymatic, and biochemical tests [22,23], by comparison and subalpine isolates and those of the type strains: 60 to reference strains. Because the majority of isolates M. chelonae,14M. malmoense,11M. gordonae,7M. were recovered at 20 °C and none were isolated at 37 °C, fortuitum,5Mycobacterium szulgai,4Mycobacterium M.-F. Thorel et al. / FEMS Microbiology Ecology 49 (2004) 343–347 345

Table 1 Identification of mycobacteria by sample type and origin Species Soil Humus Tufa Peat Rotting wood Sphagnum Total M. chelonae 42a;b 5a;c 1a 2c 5a;b 5c;d 60 M. fortuitum 5a;d 1a 000 1a 7 M. gordonae 2a 1a 3a 3a;c 02a 11 M. malmoense 12a;d 1a;c 000 1a 14 M. szulgaı03a 01a 01a 5 M. kansasii 0001c 001 M. kansasii, non-pigmented 1d 00 1a 01d 3 M. terrae complex 0 1a 000 0 1 Unidentified mycobacteria 12a;d 1c 001a 4a;d 18 Total 74 13 4 8 6 15 120 a Jura Forest. b Vosges Forest. c Subalpine Forest. d Alpine moor.

kansasii (3 unpigmented), and 1 M. terrae complex 4. Discussion (Table 1). Species assignment of 18 of the 120 isolates (15%) could not be made in keeping with other studies Mycobacteria were isolated from every type of sam- of environmental mycobacteria [5–7,10]. The M. chelo- ple in each of the different alpine and subalpine habitats, nae isolates were all non-pigmented, did not reduce ni- demonstrating that the environmental opportunistic trate, were arylsulfatase- and catalase-positive, and mycobacteria are normal alpine and subalpine inhabit- produced mycolic acids I and II [22,23]. The M. fortui- ants. These data confirm the presence of mycobacteria tum isolates were non-pigmented, reduced nitrate, were in Alpine soils based upon amplification of the myco- arylsulfatase- and catalase-positive and produced my- bacterial nidA gene [19]. The presence of mycobacteria colic acids I and V [22,23]. The M. malmoense isolates in the alpine and subalpine habitats appears to be were slow growing, non-pigmented, catalase-positive, characteristic of a wide variety of rapidly and slowly and nitrate-reductase negative and produced mycolic growing mycobacterial species. The mycobacteria re- acids I, II, and IV [22,23]. Unlike the reference strain, covered include representatives of M. chelonae, M. only two of the 14 M. malmoense isolates hydrolyzed malmoense, M. gordonae, M. fortuitum, M. szulgai, M. Tween 80. All M. malmoense isolates were identified by kansasii, and the M. terrae complex. Thus, adaptation to mycolic acid patterns and the restriction fragments survival and, likely growth, in alpine and subalpine produced from the amplified hsp-65 gene product habitats is not restricted to a few mycobacterial species. [24,25]. The M. gordonae isolates were slow growing, In a previous study, the types of mycobacteria recovered pigmented, catalase-positive, urease-, and nitrate-re- from grazed and non-grazed areas were shown to be the ductase negative, and produced mycolic acids I, III, and same [21]. A comparison of environmental and reference IV [22,23]. None of the 11 M. gordonae isolates hydro- strains of M. malmoense showed that culture, bio- lyzed Tween 80, but their mycolic acids and restriction chemical, enzymatic, and molecular markers could not fragments of the amplified hsp-65 gene were identical distinguish between isolates from the two sources [25]. with M. gordonae. The M. szulgai isolates were pig- Because of the diversity of samples from the four dif- mented, slow growing, nitrate-reductase and catalase- ferent regions, particularly samples collected in the Jura positive, urease- and arylsulfatase-negative, and pro- Forest, mycobacterial species cannot be associated with duced mycolic acids I, III, and IV [22,23]. The M. a particular geographic region, unique habitat or site. kansasii isolates were all slow growing, resistant to Although the identifications were performed on cells thiophen-2-carboxylic acid hydrazide (2 lg/ml), nitrate- grown at 20 °C, the results of cultural, enzymatic, and reductase and catalase-positive, hydrolyzed Tween 80, biochemical tests on representative type strains was no and produced mycolic acids I, III, and IV [22,23]. Only different from the results obtained for the strains grown one of the four M. kansasii isolates was photochromo- at 37 °C. However, it is possible that some differences genic, but it shared all other characteristics in common between the characteristics of the alpine and subalpine with the other isolates, with the exception that it failed isolates and those of type strains (e.g., Tween 80 to hydrolyze Tween 80. The M. terrae isolate was slow hydrolysis) could be due to the lower temperature of growing, non-pigmented, nitrate-reductase negative, incubation. catalase-positive, hydrolyzed Tween 80, and produced The data also confirm the utility of enzymatic treat- mycolic acids I and VI [22,23]. ment to release mycobacteria from environmental 346 M.-F. Thorel et al. / FEMS Microbiology Ecology 49 (2004) 343–347 samples that contain particulates [20,21]. Mycobacteria [6] Jenkins, P.A. (1991) Mycobacteria in the environment. J. Appl. are hydrophobic and would be expected to adhere to Bacteriol. Symp. Suppl. 70, 137S–141S. surfaces where the mycobacterial surface macromole- [7] Katila, M.-L., Iivanainen, E., Torkko, P., Kauppinen, J., Mar- tikainen, P. and Vaananen, P. (1995) Isolation of potentially cules (e.g., polysaccharides) could form bridges with pathogenic mycobacteria in the Finnish environment. Scand. J. either clay minerals or cellulose [26]. Infect. Dis. Suppl. 98, 9–11. Recovery of mycobacteria from samples whose pH [8] Tsukamura, M. (1984) The non-pathogenic species of ranged from 3.1 to 8.3 is consistent with data on the mycobacteria, their distribution and ecology in non-living pH range of mycobacteria growth in laboratory media reservoirs. In: The Mycobacteria (Kubica, G.P. and Wayne, L.G., Eds.), pp. 1339–1359. Marcel Dekker, New York, [12–14]. The fact that mycobacteria were recovered Basel. from samples of very low moisture content (i.e., Jura [9] Kazda, J., Muller,€ K. and Irgens, L.M. (1979) Cultivable Forest soil, Table 1), indicates that mycobacteria can mycobacteria in sphagnum vegetation of moors in South Sweden survive in habitats of low moisture, possibly within and coastal Norway. Acta Pathol. Microbiol. Scand. Sect. B 87, hydrated microenvironments. The effect of water ac- 97–101. [10] Iivanainen, E., Sallantaus, T., Katila, M.-L. and Martikainen, P.J. tivity on mycobacterial growth is unknown. It is also (1999) Mycobacteria in runoff-water from natural and drained possible that mycobacteria are capable of entering a peatlands. J. Environ. Qual. 28, 1226–1234. state in which they are desiccation-resistant. The high [11] Saito, H., Tomioka, H., Sato, K., Tasaka, H. and Dekio, S. (1994) frequency of recovery of mycobacteria from humus Mycobacterium malmoense isolated from soil. Microbiol. Immu- and peats reported here is consistent with other reports nol. 38, 313–315. [12] Chapman, J.S. and Bernard, J.S. (1962) The tolerances of [1,4,5,9,10] and the observation that humic and fulvic unclassified mycobacteria. I. Limits of pH tolerance. Am. Rev. acids stimulate the growth of M. avium and M. intra- Respir. Dis. 86, 582–583. cellulare [27]. [13] Portaels, F. and Pattyn, S.R. (1982) Growth of mycobacteria in It is of particular interest to find that 80% of the relation to the pH of the medium. Ann. Microbiol. (Inst. Pasteur.) isolates were recovered on media incubated at 20 °Cand 133, 213–221. [14] George, K.L. and Falkinham III, J.O. (1986) Selective medium for none at 37 °C. Not only does this demonstrate the ne- the isolation and enumeration of Mycobacterium avium, M. cessity of considering temperature of incubation when intracellulare, and M. scrofulaceum. Canad. J. Microbiol. 32, 10– attempting to isolate mycobacteria from environmental 14. samples, but, further, it indicates the ability of myco- [15] George, K.L., Parker, B.C., Gruft, H. and Falkinham III, J.O. bacteria to survive and possibly grow at temperatures (1980) Epidemiology of infection by nontuberculous mycobacte- ria. II. Growth and survival in natural waters. Am. Rev. Respir. substantially lower than that of mammals. However, Dis. 122, 89–94. these mycobacteria may not be capable of infecting [16] Schulze-Robbecke,€ R. and Buchholtz, K. (1992) Heat suscepti- mammals and causing disease. We have not yet deter- bility of aquatic mycobacteria. Appl. Environ. Microbiol. 58, mined whether the isolates could be induced to grow at 1869–1873. 37 °C and thus become potential pathogens. If the iso- [17] Kirschner Jr., R.A., Parker, B.C. and Falkinham III, J.O. (1992) Epidemiology of infection by nontuberculous mycobacteria. X. lates are unable to grow at 37 °C, it is unlikely they pose Mycobacterium avium, M. intracellulare, and M. scrofulaceum in a threat to indigenous mammals or birds. Current efforts acid, brown-water swamps of the southeastern United States and are aimed at identifying the adaptations leading to their association with environmental variables. Am. Rev. Respir. survival and possibly growth of mycobacteria at low Dis. 145, 271–275. temperatures. [18] Whyte, L.G., Hawari, J., Zhou, E., Bourbonniere, L., Inniss, W.E. and Greer, C.W. (1998) Biodegradation of variable chain-length alkanes at low temperature by a psychrotrophic Rhodococcus sp. Appl. Environ. Microbiol. 64, 2578–2584. References [19] Margesin, R., Labbe, D., Schinner, F., Greer, C.W. and Whyte, L.G. (2003) Characterization of hydrocarbon-degrading microbial [1] Kazda, J.K. (1983) The principles of the ecology of mycobac- populations in contaminated and pristine Alpine soils. Appl. teria. In: The Biology of the Mycobacteria (Ratledge, C. and Environ. Microbiol. 69, 3085–3092. Stanford, J., Eds.), Vol. 2, pp. 323–341. Academic Press, [20] Thorel, M.F., Moreau, R., Charvin, M. and Ebiou, D. (1991) London. Debusquement enzymatique des mycobacteries dans les milieux [2] Kubin, M. (1984) Distribution and ecology of non-living reser- naturels. C. R. Soc. Biol. 185, 331–337. voirs, opportunists and pathogens. In: The Mycobacteria (Kubica, [21] Thorel, M.F. (1998) Application du debusquement enzymatique a G.P. and Wayne, L.G., Eds.), pp. 1313–1338. Marcel Dekker, l’isolement des mycobacteries de l’environnement. C. R. Acad. New York. Agric. Fr. 84, 89–100. [3] Falkinham III, J.O. (1996) Epidemiology of infection by [22] Goodfellow, M. and Magee, J.G. (1998) of mycobac- nontuberculous mycobacteria. Clin. Microbiol. Rev. 9, 177– teria. In: Mycobacteria. I. Basic Aspects (Gangadharam, P.R.J. 315. and Jenkins, P.A., Eds.), pp. 1–71. Chapman & Hall, New York, [4] Kazda, J. (2000) The Ecology of Mycobacteria. Kluwer Academic NY. Publishers, Dordrecht, The Netherlands. [23] David, H., Levy-Frebault,, V. and Thorel, M.F. (1989) In: [5] Iivanainen, E.K., Martikainen, P.J., Rais€ anen,€ M.L. and Katila, Methodes de laboratoire pour Mycobacteriologie clinique M.-L. (1997) Mycobacteria in boreal coniferous forest soils. Commission des laboratoires de reference et d’expertise de FEMS Microbiol. Ecol. 23, 325–332. l’Institut Pasteur de Paris, Vol. 1. Institut Pasteur, Paris. M.-F. Thorel et al. / FEMS Microbiology Ecology 49 (2004) 343–347 347

[24] Telenti, A., Marchesi, F., Balz, M., Bally, F., Bottger,€ E.C. and and of 16S and ITS1 rDNA. Res. Microbiol. 151, 629– Bodmer, T. (1993) Rapid identification of mycobacteria to the 638. species level by polymerase chain reaction and restriction enzyme [26] Brisou, J.F. (1986) Adherence et debusquement enzymatique des analysis. J. Clin. Microbiol. 31, 175–178. microorganismes fixes sur des supports inertes et vivants. Rev. [25] Leclerc, M.C., Haddad, N., Moreau, R. and Thorel, Inst. Pasteur Lyon. 19, 1–6. M.F. (2000) Molecular characterisation of environmental [27] Kirschner, R.A., Parker, B.C. and Falkinham III, J.O. (1999) Mycobacterium strains by PCR-restriction fragment length Humic and fulvic acids stimulate the growth of Mycobacterium polymorphism of hsp 65 and by sequencing of hsp 65, avium. FEMS Microbiol. Ecol. 30, 327–332.