INTERNATIONALJOURNAL OF SYSTEMATICBACTERIOLOGY, July 1993, p. 405-413 Vol. 43, No. 3 0020-7713/93/030405-09$02.00/0 Copyright 0 1993, International Union of Microbiological Societies

Mycobacterium brumae sp. nov., a Rapidly Growing, Nonphotochromogenic M. LUQUIN,132*V. AUSINA,1,2 V. VINCENT-LEVY-FREBAULT,3 M. A. LvEELLE,4 F. BELDA,172 M. GARCiA-BARCEL0,172G. PRATS,' AND M. DAFFE4 Servicio de Microbiologia, Hospital de la Santa Cruz y San Pablo, Departamento de Genttica y Microbiolog*a, Universidad Autbnoma de Barcelona, 08025 Barcelona, ' and Servicio de Microbiolog'a, Hospital Germans Trias i Pujol, Departamento de Genetica y Microbiolog'a, Universidad Autonoma de Barcelona, 08915 Badalona, Spain, and Unite de la Tuberculose et des Mycobactkries, Institut Pasteur, 75724 Paris Cedex 15,3 and Departement III du LPTF du Centre National de la Recherche Scientifque et Universitk Paul Sabatier, 31062 Toulouse Ceda, France

Strains of a new species of rapidly growing, nonphotochromogenic mycobacteria, Mycobacterium brumae, have been isolated from water, soil, and human sputum samples in Barcelona, Spain. The inclusion of this organism in the genus Mycobacterium is based on its acid-alcohol fastness, its DNA G+C content, its mycolate pattern, and its mycolate pyrolysis esters. A study of 11 strains showed that they form a homogeneous group with an internal phenotypic similarity value of 94.9 +, 3.7%. The results of a comparison with 39 other mycobacterial species and subspecies are also presented. DNA relatedness studies showed that the M. brumae strains studied form a single DNA hybridization group which is less than 30% related to 15 other species of the genus Mycobacterium. Thin-layer chromatographic analysis showed that only a-mycolates are present. Unlike Mycobacterium fallux and Mycobacterium triviale a-mycolates, M. brumae a-mycolates release only 22-carbon atom esters after pyrolysis. Strain CR-270 is the type strain; a culture of this strain has been deposited in the Collection Nationale de Cultures de Microorganismes de 1'Institut Pasteur, Paris, France, as strain CIP 103465.

From 1983 to 1987, in the course of a search for mycobac- new strains give only docosanoate when they are subjected teria in environmental samples, we isolated 10 strains with to pyrolytical cleavage. A subsequent structural analysis homogeneous properties belonging to the genus Mycobacte- showed that there are great differences in the compositions rium. Later, in 1991, a similar strain was isolated from a of the a-mycolates of these strains and the a-mycolates of human sputum sample. These organisms were rapidly grow- M. fallax and M. triviale. Genotypically, there was clear ing, nonphotochromogenic mycobacteria that differed from evidence for differentiation of these organisms, since the all known species and are considered members of a new levels of DNA-DNA hybridization between the strains of the species. These strains had a colonial morphology similar to new taxon and other related species were less than 30%. In that of Mycobacterium falZax and the same pattern of this paper, we describe the characteristics of these strains mycolates (only a-mycolates) as determined by thin-layer and designate a new species, Mycobacterium brumae. chromatography (TLC) (16). Mycobacterium triviale is the only slowly growing Mycobacterium species with this pat- tern of mycolates (4,16). Mycolic acids are useful taxonomic MATERIALS AND METHODS markers for the differentiation of mycobacteria at both the Bacterial strains. A total of 11 strains were studied. Eight genus level and the species level. Mycobacterial mycolic of these (strains CR-103, CR-104, CR-142, CR-210, CR-267, acids are characteristically 2-branched7 3-hydroxy fatty ac- CR-268, CR-270T [T = type strain], and CR-271) were ids with very long chains (up to 90 carbon atoms long) which isolated from water samples taken from the Llobregat River are major lipid constituents in the cell wall (4, 7). By TLC in Barcelona, Spain; two strains (strains CR-130 and CR-269) they may be separated into seven types according to the were isolated from soil; and one strain (strain CR-148) was content and position of different functional groups in their isolated from a human sputum sample. In addition to these carbon chains (18,19). Mycolic acid methyl esters release by 11 isolates, we studied the following strains of species which pyrolysis straight esters with carbon chain lengths ranging are on the Approved Lists of Bacterial Names (24) or in the from C,, to C,, (4,12,17). Within the genus Mycobacterium, Index of the Bacterial and Yeast Nomenclatural Changes variation in the pyrolysis ester chain lengths is an additional (20): Mycobacterium agn' ATCC 27406=; Mycobacterium differential characteristic that is especially useful for subdi- aichiense ATCC 27280T; Mycobacterium aurum ATCC viding groups of species that produce the same TLC pattern. 23366T; Mycobacterium austroafncanum ATCC 33464T; The strains of the new species could be clearly distinguished subsp. chelonae NCTC 946= and from M. fallax and M triviale by the chain lengths of their CIPT 801159; Mycobacterium chelonae subsp. abscessus pyrolysis esters. Esters containing 22 and 24 carbon atoms ATCC 19977T;Mycobacterium chitae ATCC 19627T; Myco- were released by a-mycolates of M. fallax, and tetra- bacterium chubuense ATCC 27278T; Mycobacterium diem- cosanoate was the only ester liberated by M. triviale a-my- hoferi ATCC 19340T; Mycobacterium duvalii NCTC 358=; colates (14, 16, 17, 22). In contrast, the a-mycolates of the Mycobacterium fallax CIP 8139T, CIPT 1390007, and CIPT 1390014; Mycobacterium jlavescens ATCC 14474T; Myco- bacterium fortuitum ATCC 6841T and ATCC 14467; Myco- * Corresponding author. bacterium gadium ATCC 27726T; Mycobacterium gastri

405 406 LUQUIN ET AL. INT.J. SYST.BACTERIOL.

ATCC 15754=; NCTC 10742=; My- with petroleum ether-diethyl ether (6:4, vol/vol). The spots cobacterium gordonae ATCC 14470T and CIPT 0210008; were visualized by charring after the plates were sprayed Mycobacterium kansasii ATCC 12478T;Mycobacterium ko- with a 10% solution of molybdophosphoric acid in ethanol. mossense ATCC 33013=; ATCC Oxidative cleavage, isolation of oxidation products, and 927=; Mycobacterium moriokaense ATCC 43059=; Myco- structural analysis were performed as previously described bacterium neoaumm ATCC 25795=; Mycobacterium (13). The patterns of nonhydroxylated fatty acids were obuense ATCC 27023T; Mycobacterium paraforhcitum determined by capillary gas chromatography as previously ATCC 19686=; ATCC 117ST; Myco- described (17). bacterium porcinum ATCC 33776=; Mycobacterium Instrumentation. Gas chromatography was conducted poriferae ATCC 35087=; Mycobacterium pulveris ATCC with a Hewlett-Packard model 5890A apparatus equipped 35154=; Mycobacterium rhodesiae ATCC 27024T; Mycobac- with a fused silica capillary column with SPBl as the terium senegalense NCTC 10956=; Mycobacterium smeg- stationary phase. The temperature of the injector was raised matis ATCC 19420T; Mycobacterium sphagni ATCC from 260 to 350°C for pyrolytical conditions. Infrared spectra 33027=; ATCC 15755=; Mycobacte- were recorded with a Perkin-Elmer model FTIR 1600 appa- rium thermoresistibile ATCC 19527T; Mycobacterium to- ratus. ‘H nuclear magnetic resonance spectra were obtained kaiense ATCC 27282T; Mycobacterium triviale ATCC with a 200-MHz Brucker instrument by using CDC1, as the 23292T; Mycobacterium ATCC 27294= and solvent. Polarimetry was conducted with a Perkin-Elmer ATCC 25177; and ATCC 15483=. In model 141 polarimeter. Mass spectra were determined by addition, the frequency distribution of the test results and using an electron impact ion source and a Varian model 311A the results of a lipid analysis for mycobacterial isolates spectrometer . obtained from human host and environmental sources were DNA-DNA hybridization. For DNA extraction and subse- included for comparative purposes. These isolates included 13 M. chelonae subsp. chelonae strains, 6 M. chelonae quent DNA-DNA hybridization, selected strains (see Table subsp. abscessus strains, 8 M. fallax strains, 7 M. jlavescens 4) were cultivated in 1 to 3 liters of nutrient broth (Difco) strains, 12 M. fortuitum strains, 9 M. gastri strains, 28 M. supplemented with a powder base containing 7H9 Middle- gordonae strains, 21 M. kansasii strains, 4 M. marinum brook broth medium (Difco), 1% glycerol, and 0.05% Tween strains, 7 M. smegmatis strains, 13 M. terrae strains, 11 M. 80. Flasks were incubated at 30 or 37°C with gentle agitation triviale strains, and 47 M. tuberculosis strains. We also for 10 to 21 days. DNA was extracted as previously de- included in this study six strains of a recently described new scribed (14, 15). Briefly, the were converted to species, Mycobacterium alvei (1). wall-deficient forms and then lysed by adding sodium dode- Characterization of strains. Colony morphology, the abil- cyl sulfate, as previously described, and proteinase K (Boehr- ity to grow at various temperatures (25,30,37,45, and 52”C), inger, Mannheim, Germany) instead of pronase. The DNA pigment production, and photoreactivity were determined as was then purified by using the usual methods. DNA from the previously described (9, 10, 28-30). The niacin test was type strain of the proposed new mycobacterial taxon, strain performed with test strips (Difco Laboratories, Detroit, CIP 103465, was labeled in vitro by nick translation (nucle- Mich.). Strains were tested for iron uptake by using a otides were obtained from the Radiochemical Centre, Am- modification of the procedure of Szabo and Vandra (26), as ersham, England) as previously described (8), except that we described by Silcox et al. (23). The catalase test was per- modified the concentrations of some of the reagents, as formed by the method of Kubica and Pool (11). The nitrate follows: 3 pl of DNA (800 pg/ml), 40 pl of S1 nuclease (1 reductase and arylsulfatase tests were performed as de- U/ml; Sigma Chemical Co., St. Louis, Mo), and 15 pl of scribed by Vestal (28). The p-glucosidase, urease, penicilli- DNase I (lop8 g/ml). Unlabeled DNAs were sheared by nase, and trehalase tests were performed as recommended sonication in order to get DNA fragments that were 500 to by David and Jahan (5) and David et al. (6). Tests for 800 bp long. The molecular weight of the sonicated DNA resistance to hydroxylamine and sodium chloride, hydroly- was checked by gel electrophoresis by using phage lambda sis of Tween 80, and P-galactosidase were performed as hydrolyzed by restriction enzyme HindIII. The S1 nuclease described by Wayne et al. (29, 30). Tests to determine the method (3) in which the S1 nuclease-trichloroacetic acid use of glucose, inositol, mannitol, and sodium citrate as sole procedure (8)was used was modified as previously described carbon sources in the presence of ammoniacal nitrogen were (14). The temperature (T,) at which 50% of the DNA performed by using the procedure of Silcox et al. (23). Tests became hydrolyzable by S1 nuclease was determined by to determine degradation of salicylate, tolerance to picric using the procedure developed by Crosa et al. in 1973 (3) and acid, and acid formation from carbon sources were per- modified as described previously (14, 15). formed as described by Tsukamura (27). Growth in the presence of different antibiotics was determined by using the Numerical analysis. The data obtained from our method of Canetti et al. (2). study of 59 taxonomic properties were recorded as positive Lipid analysis. To isolate lipid components, all of the or negative, and the matching coefficient (M value) was strains were cultivated on plates containing Middlebrook calculated by using the following equation: M value = (ns x 7H10 agar and incubated at 30 or 37°C in the presence of 5% 100)/(ns + nd), where ns is the number of characters for CO,. Fatty acid esters were prepared as previously de- which two strains gave the same results (both positive or scribed (4). Mycolates were obtained by precipitating them both negative) and nd is the number of characters for which twice with methanol from the methyl ester mixture. We then the strains gave different results (one positive and one checked by TLC that the pellet contained only mycolates. negative) (25). The computer which we used was an IBM The mycolates were fractionated by argentation chromatog- model 3083/XE/VM/SP-HPC computer. The program which raphy by using increasing proportions of diethyl ether in we used is included in the Clustan 2 packet of programs petroleum ether (22). This separation was followed by TLC (Computing Laboratory, University of St. Andrews, St. on silver nitrate-impregnated commercial plates developed Andrews, Scotland). VOL. 43, 1993 MYCOBACTERIUM BRUMAE SP. NOV. 407

FIG. 1. Mature M. brumae CIP 10346ST colonies on Middlebrook 7H10 medium. (A) Micrograph. Magnification, X120. (B) No magnification.

RESULTS AND DISCUSSION 75% with M. moriokaense (Table 2). Characteristics which Cells of the 11 new strains grown on Lijwenstein-Jensen differentiate M. brumae from phenotypically related Myco- agar and on Middlebrook 7H10 agar were rod shaped, gram bacterium species are shown in Table 3. positive, and acid-alcohol fast and were able to form clumps Lipid analysis. Gas chromatographic analysis revealed the and cords. Spores, capsules, and true branching were not presence of hexadecanoate, octadecenoate, and tubercu- observed. At 25, 30, and 37"C, growth occurred within 4 to lostearate (10-methyloctadecanoate) as the major fatty acid 5 days on Lijwenstein-Jensen and Middlebrook 7H10 media methyl esters in lipid extracts of the 11 M. brumae strains and on nutrient agar. Dilute inocula on Middlebrook agar (Fig. 2). All M. brumae strains exhibited a TLC pattern yielded large, eugonic, buff-colored, rough colonies (Fig. 1). consisting of only one spot corresponding to a-mycolate The phenotypic properties of the 11 M. brumae strains are (Fig. 3). When pyrolytical conditions were used, the gas shown in Table 1. These strains constituted a homogeneous chromatographic profile showed an increase of methyl do- group with a level of internal similarity of 94.9 2 3.79%. cosanoate (C22), whereas M. fallax mycolates liberated by Comparisons of M. brumae CIP 103465T with the type pyrolysis do- and tetracosanoate (C22 and CZ4) and M. strains of 39 other mycobacterial species revealed degrees of triviale liberated only tetracosanoate (4, 16, 17, 22) (Fig. 2). similarity of 80% with M. alvei, 75.8% with M. fallax, and The infrared spectrum of the purified mycolates from M. 408 LUQUIN ET AL. INT. J. SYST.BACTERIOL.

TABLE 1. Properties of the M. brumae type strain and TABLE 2. Levels of similarity between M. brumae CR-270T 10 additional strains (= CIP 103465=) and type strains of 39 mycobacterial species and subspecies Type straina Ten Other Characteristic strainsb % Similarity Species or subspecies to strain Morphology CIP 103465= Rods (>1 pm long) 100 M. alvei ...... 80.0 0 Coccobacillary (<1 pm long) M. fallax ...... 75.8 Cord formation 100 M. monokaense ...... 75 .O 0 Smooth colonies M. austroafncanum...... 74.3 Rough colonies 100 M. duvalii 73.0 0 ...... Pigmentation in the dark M. gdvum ...... 73.0 0 Pigmentation in the light M. smegmatis ...... 72.7 Growth at: M. agri ...... 72.7 25°C 100 M. diemhofen ...... 72.5 30°C 100 M. gadium ...... 72.2 30°C in <5 days 100 M. parafortuitum ...... 72.0 100 37°C M. fortuitum ...... 71.2 37°C in <5 days 100 M. chitae ...... 69.7 0 45°C M. neoaurum ...... 69.4 0 52°C M. senegalense...... 68.2 Growth on nutrient agar at: M. aurum ...... 68 .O 30°C 100 M. aichiense ...... 67.5 100 37°C M. rhodesiae ...... 67.5 Growth on MacConkey agar without 0 M. porcinum ...... 66.7 crystal violet M. flavescens ...... 66.7 Iron uptake 80 M. marinum 66.7 0 ...... Niacin M. chelonae subsp. chelonae ...... 66.7 Enzymatic activities M. ponferae ...... 65.9 Arylsulfatase (3 days) 0 M. sphagni ...... 65.9 Catalase (22°C) 100 M. thermoresistibile ...... 65.6 Catalase (68°C) 100" M. kansasii...... 65.2 p-Glucosidase (3 h) 100 M. chubuense ...... 65 .O p-Galactosidase (18 h) 0 M. tokuiense ...... 65.0 Nitrate reductase (2 h) 100 M. pulveris ...... 64.3 Penicillinase (5 h) 100 M. komossense ...... 63.8 Trehalase (5 h) 100 M. chelonae subsp. abscessus ...... 63.6 100 Urease (18 h) M. tnviale ...... 62.1 Tween hydrolysis (10 days) 100 M. phlei 61.5 0 ...... Salicylate degradation (7 days) M. vaccae ...... 60.6 Acid produced from: M. tuberculosis...... 60.6 Glucose 100 M. gastri 57.6 0 ...... L- Arabinose M. terrae ...... 57.6 Dulcitol 0 M. obuense ...... 57.5 Fructose 100 M. gordonae ...... 53.0 meso-Inositol 100 Mannitol 0 L-Rhamnose 0 D-Xylose 20 Trehalose 100 brumae revealed an absorption band at 965 cm-' character- Use of the following compounds as istic of trans double bonds, while absorption due to cyclo- sole carbon sources: propane rings was not observed. The 'H nuclear magnetic Glucose 100 resonance spectrum confirmed this hypothesis by the pres- rneso-Inositol 100 ence of signals assignable to the resonance of olefinic pro- Mannitol 0 tons at 6 = 5.35 ppm and the absence of signals between 6 = Fructose 100 0 and 6 = 0.7 ppm, corresponding to cyclopropane ring Sodium citrate 100 protons resonances. Argentation chromatography fraction- Growth in the presence of: Picric acid (0.2%) 50 ated this mycolate spot into three main compounds; each NaCl (5%) 0 class was subjected to oxidative cleavage, and the oxidation NH20H (500 mglliter) 20 products were analyzed by gas chromatography. These D-Cycloserine (30 mdliter) 40 analyses gave the composition shown in Fig. 4. The I, Ethambutol (2 mg/liter) 0 a-mycolates were defined as diunsaturated esters with two Streptomycin (4 mg/liter) 100 trans double bonds, as suggested by infrared and 'H nuclear Kanamycin (20 mdliter) 0 magnetic resonance spectroscopy (multiplet of the double Capreomycin (40 mg/liter) 0 bond at 6 = 5.35 ppm and additional methyl groups near the Isoniazid (0.1 mg/liter) 100 Isoniazid (1 mg/liter) 80 Isoniazid (10 mg/liter) 0 Rifampin (40 mdliter) 80 p-Aminosalicylic acid (0.5 mg/liter) 100 a- , negative reaction; +, positive reaction; +w, weak reaction. Thiacetazone (2 mg/liter) 100 ' The values are the percentages of strains that are positive. Thiacetazone (10 mg/liter) 100 Weak reactions. TABLE 3. Differentiation of M. brumae CIP 1034565T from other phenotypically related species of mycobacteria

M. bmmae M. alvei M. mon'okaeme M. austro- M. duvalii M. gilvum M. smeg- M. agri M.gadium ~~~~~ M. fortui- Characteristic afncanum NCTC NCTC matisATCC ATCC ATCC ATCC tumATCC CIP 103465T CIP 103464T CIP 8139T ATCC 43059T ATcc 33464T 35gT 10742T 19420T 27406T gF$ 27726T 19686~ 6841T Morphology Rods (> 1 km long) + + + + + + Coccobacillary (< 1 p,m long) + ------Cord formation + ND ND ND Smooth colonies + + + + Rough colonies + + - Pigmentation in the dark + + + + Pigmentation in the light + + + + Growth at: - 37°C in <5 days + + + + + + + + + + + 45°C - - - - - + + - - - - Growth on MacConkey agar ND ND ND ND - - ND + without cristal violet Iron uptake + ND ND ND ND + + + + rn Enzymatic activities s Arylsulfatase (3 days) + - - + Catalase (68°C) + + + + P-Galactosidase (18 h) + + - Urease (18 h) + + + Tween hydrolysis (10 days) - - + Salicylate degradation (7 days) + Use of the following compounds as sole carbon sources: mesa-Inositol + + - + + + Mannitol + + + + + + Sodium citrate - + - - - + Growth in the presence of Picric acid (0.2%) + + ND ND + + + + + NaCl(5%) - - + + ND ND + + + - + NH,OH (500 mg/liter) + + ------+ Mycolate contents a-Mycolates + + + + + + + + + + + + + Oxygenated mycolates - + - + + + + + + + + + +

ff ff -, negative reaction;+, positive reaction;+w, weak reaction; ND, not determined. 410 LUQUIN ET AL. INT. J. SYST.BACTERIOL.

M. brunrac

Retention time (niin) 2 4 6 8 10 12 14 16 FIG. 2. Gas chromatogram of fatty acid methyl esters and mycolic acid cleavage products of M. brurnue CIP 103465= (A), M. fullax CIPT 1390005=(B), and M. triviale ATCC 23292= (C). The numbers above the peaks indicate the number of carbon atoms, followed by the number of double bonds. TBS, tuberculostearate (10-methyloctadecanoate); FID, flame ionization detector. double bond at 6 = 0.94 ppm) (13) and by the negative value existence of double bonds in these mycolates indicate that of the molecular rotation of the intact mycolate (13). The they are related to M. fallax mycolates (22). However, the main mycolic acid of this series contained 76 carbon atoms. differences in the number of double bonds (tetraunsaturated The I, a-mycolates consisted of a mixture of compounds, a-mycolates have been described in M. fallax) and in the each bearing one cis double bond and one trans double bond. chain length of the pyrolysis esters (16, 22) suggest that the The major mycolic acid of this series contained 75 carbon new strains belong to a species different from M. fallax. atoms. The I, a-mycolates contained only cis diunsaturated DNA relatedness. The ratios of absorbance for DNA esters. 'H nuclear magnetic resonance of this fraction solutions ranged from 1.9 to 2.1 and from 1.7 to 1.8 for the showed a well-defined triplet at 6 = 5.35 ppm (instead of a A2,&4,,, and A26,J!230ratios, respectively. The results of multiplet) assignable to the resonance of olefinic protons. DNA-DNA hybridization experiments are shown in Table 4. These a-mycolates consisted of mycolic acid homologs We selected the type strains of species which produced the containing 68 to 76 carbon atoms (the main component was same mycolate pattern (only a-mycolates) as M. brumae (M. an acid with 70 carbon atoms). The chain length and the fallax and M. triviaZe) for DNA-DNA hybridization experi- VOL.43, 1993 MYCOBACTERIUM BRUMAE SP. NOV. 411

homologous reaction and the T,,, of the heterologous reac- tion). The percentages of relative binding of strain CIP 103465Twith all other strains were less than 30%. The AT,,, values determined for the higher levels of DNA complemen- tarity were all more than 12.1"C. The guanine-plus-cytosine (G+C) contents of DNAs were calculated from the melting temperatures by using the equation of Owen and Pitcher (21). The G+C content of strain CIP 103465=was 70 mol%. This value was an underestimation of the actual G+C content as sonicated DNAs were used to determine the melting temperatures. Description of type strain CR-270 (= CIP 103465) of Mycobacterium brumue sp. nov. Mycobacterium brumae (bru' mae. L. gen. n. brumae, of winter, referring to the time of year at which the first strains were isolated). The cells of M. brumae are gram-positive rods, 2.0 to 2.5 pm long and 0.3 to 0.5 pm wide; they are mostly strongly acid fast, except for a small number (less than 10%) of cyanophil forms. Smears prepared from Youmans medium (without Tween 80) and Middlebrook 7H12 medium contain clumps or cords. Colo- FIG. 3. Thin-layer chromatogram of methylmycolates from M. nies on Lijwenstein-Jensen medium and on Middlebrook tuberculosis ATCC 27294T lane l),M. fallax CIPT 1390005T(lane 2), M. trivide ATCC 23292I (lane 3), M. brumae CIP 103465T(lane 7H10 agar are eugonic, rough, and nonpigmented in the dark 4), M. fortuitum ATCC 6841T (lane 5), and M. chelonae NCTC 94(jT or after exposure to light. Growth occurs within 5 days at 30 (lane 6). The mycolate types are indicated as follows: I, a-mycolate; and 37°C but not at 45°C. The type strain produces thermo- 11, a'-mycolate; 111, methoxymycolate, IV, ketomycolate; and V, stable catalase and is positive for P-glucosidase, nitrate epoxymycolate. The elution system consisted of two runs with reductase, penicillinase, trehalase, urease, and iron uptake. petroleum ether-diethyl ether (85:15, voltvol) as the eluent. Tween 80 is hydrolyzed after 10 days; niacin is not pro- duced, and salicylate is not degraded to catechol. No growth occurs on MacConkey agar without crystal violet. Glucose, ments, as well as other rapidly growing and slowly growing meso-inositol, fructose, and sodium citrate are utilized as mycobacterial species. The use of the hybridization and S1 sole carbon sources in the presence of ammoniacal nitrogen. nuclease treatment conditions mentioned above resulted in Acid is produced from glucose, fructose, meso-inositol, and levels of nonspecific hybridization (in control tubes contain- trehalose but not from L-arabinose, dulcitol, mannitol, L- ing herring DNA and labeled M. brumae CIP 103465TDNA) rhamnose, and D-xylose. Growth is inhibited by sodium ranging from 7.6 to 9.4%. Four strains of M. brumae were 95 chloride (5%), hydroxylamine (500 mdliter), D-cycloserine to 100% related to strain CIP 103465T, with ATm values less (30 mg/liter), ethambutol (2 mdliter), kanamycin (20 mg/ than 2.5"C (ATm is the difference between the Tm of the liter), capreomycin (40 mg/liter), and isoniazid (10 mdliter).

a-mycolatcs I,,

OH dOCH, x=lS y=13 ~17.19 a-mycolatcs I,

a-mycolatcs E

FIG. 4. Molecular structures of a-mycolates of M. bmmae CIP 103465T. For double bonds, E indicates trans, and Z indicates cis. 412 LUQUIN ET AL. INT. J. SYST.BACTERIOL.

TABLE 4. Levels of DNA relatedness between strain CR 270T 5. David, H. L., and M. T. Jahan. 1977. p-Glucosidase activity in (= CIP 103465T)and reference strains belonging to selected mycobacteria. J. Clin. Microbiol. 136-9. mycobacterial species 6. David, H. L., I. Traore, and A. Feuillet. 1981. Differential identification of and Mycobacterium Source of DNA" % of relative chelonei. J. Clin. Microbiol. 5383-384. Other binding at 7. Goodfellow, M. E., and D. E. Minnikin. 1984. Circumscription Species Strain designation 75°C of the genus, p. 1-24. In G. P. Kubica and L. G. Wayne (ed.), The mycobacteria. A sourcebook. Part A. Marcel Dekker, Inc., M. brumae CR-270T 100 New York. (= CIP 103465T) 8. Grimont, P. A. D., M. Y. Popoff, F. Grimont, C. Coynault, and CR-130 100 M. Lemelin. 1980. Reproducibility and correlation study of three CR-267 100 deoxyribonucleic acid hybridization procedures. Curr. Micro- CR-269 100 biol. 4:325-330. CR-142 95 9. Kubica, G. P., and H. L. David. 1980. The mycobacteria, p. M. agri CIPT 1320001T ATCC 27406T 20 1693-1730. In A. C. Sonnenwirth and L. Jarett (ed.), Grad- M. alvei CIP 103464T 7 wohl's clinical laboratory methods and diagnosis. The C. V. M. chelonae CIPT 0420003T NCTC 946T 7 Mosby Co., St. Louis. CIPT 801159 4 10. Kubica, G. P., and R. C. Good. 1981. The genus Mycobacte- M. chitae CIPT 1160001T ATCC 19627T 12 rium, p. 1962-1984. In M. P. Starr, H. Stolp, H. G. Truper, A. M. fallax CIPT 139000ST CIP 8139T 21 Balows, and H. G. Schlegel (ed.), The prokaryotes: a handbook CIPT 1390007 24 on habitats, isolation and identification of bacteria. Springer CIPT 1390014 30 Verlag, Berlin. M. fortuitum CIPT 0410001T ATCC 6841T 7 11. Kubica, G. P., and G. L. Pool. 1960. Studies on the catalase CIPT 0410020 ATCC 14467 11 activity of acid fast bacilli. I. An attempt to subgroup these M. gordonae CIPT 0210008 6 organisms on the basis of their catalase activities at different M. moriokaense CIPT 1470001T ATCC 43059T 5 temperatures and pH. Am. Rev. Respir. Dis. 81:387-391. M. phlei CIPT 1300001T ATCC 1175gT 12 12. Lambed, M. A., C. W. Moss, V. A. Silcox, and R. C. Good. M. porcinum CIPT 1460001T ATCC 33776T 10 1986. Analysis of mycolic acid cleavage products and cellular M. pulveris CIPT 1480001T ATCC 35154T 11 fatty acids of Mycobacterium species by capillary gas chroma- M. senegalense CIPT 1350002T NCTC 10956T 12 tography. J. Clin. Microbiol. 23:731-736. M. smegmatis CIPT 1330010T ATCC 19420T 14 13. LanCelle, M. A., C. Lacave, M. DaffC, and G. LanCelle. 1988. M. triviale CIPT 0330001T ATCC 23292T 12 Mycolic acids of Mycobacterium aurum. Structure and bioge- M. tuberculosis CIPT 0010002 ATCC 25177 21 netic implications. Eur. J. Biochem. 177:631-635. 14. Uvy-FrCbault, V., F. Grimont, P. A. D. Grimont, and H. L. a CR, Collection of the Department of Tuberculosis, Hospital de la Santa David. 1984. Deoxyribonucleic acid relatedness study of Myco- Cruz y San Pablo, Barcelona, Spain; CIP, Collection Institut Pasteur, Paris, France; CIPT, Collection lnstitut Pasteur Tuberculose, Paris, France; ATCC, bacterium fallax. Int. J. Syst. Bacteriol. 34:423-425. American Type Culture Collection, Rockville, Md.; NCTC, National Collec- 15. Uvy-Frkbault, V., F. Grimont, P. A. D. Grimont, and H. L. tion of Type Cultures, London, England. David. 1986. Deoxyribonucleic acid relatedness study of the Mycobacterium fortuitum-Mycobacten'um chelonae complex. Int. J. Syst. Bacteriol. 3fk458-460. Tuberculostearate is detected by gas chromatography; do- 16. Uvy-FrCbault, V., J. Rafidinarivo, J. C. Prom&,J. Grandry, H, Boisvert, and H. L. David. 1983. Mycobacterium fallax sp. nov. cosanoate is a mycolic acid cleavage product. Secondary Int. J. Syst. Bacteriol. 33:333-346. alcohols are not detected. M. brumae contains only a-my- 17. Luquin, M., V. Ausina, F. Lhpez-Calahorra, F. Belda, M. colates (68 to 76 carbon atoms long), which release only Garcia-BarceM, C. Celma, and G. Prats. 1991. Evaluation of docosanoate after pyrolysis. practical chromatografic procedures for identification of clinical isolates of mycobacteria. J. Clin. Microbiol. 29:120-130. ACKNOWLEDGMENTS 18. Luquin, M., J. Roussel, F. Lhpez-Calahorra, G. LanCelle, V. Ausina, and M. A. LanCelle. 1990. A novel mycolic acid in a This work was supported by grant 91/0400 from the Fondo de Mycobacterium sp. from the environment. Eur. J. Biochem. Investigaciones Sanitarias and by grant EE90/2-4.434 from the 192:753-759. Comissio Interdepartamen tal de Recerca i Innovacio Tecnologica. 19. Minnikin, D. E., I. A. Hutchinson, A. B. Caldicott, and M. We gratefully acknowledge E. Ageron and P. A. D. Grimont for Goodfellow. 1980. Thin-layer chromatography of methanoly- their support in the carrying out of the DNA-DNA hybridizations sates of mycolic acid containing bacteria. J. Chromatogr. 188: and G. Loren for his assistance with the taxonomic study. We are 221-233. also grateful to H. L. David for his helpful comments and sugges- 20. Moore, W. E. C., and L. V. H. Moore. 1989. Index of the tions throughout this study. bacterial and yeast nomenclature changes, p. 41. American Society for Microbiology, Washington, D.C. REFERENCES 21. Owen, R. J., and D. Pitcher. 1985. Current methods for estimat- Ausina, V., M. Luquin, M. Garcia-Barcel6, M. A. LanCelle, V. ing DNA base composition and levels of DNA-DNA hybridiza- Uvy-FrCbault, F. Belda, and G. Prats. 1992. Mycobacterium tion, p. 67-93. In M. Goodfellow and D. E. Minnikin (ed.), alvei sp. nov. Int. J. Syst. Bacteriol. 42529-535. Chemical methods in bacterial systematics. Academic Press, Canetti, G., N. Rist, and J. Grosset. 1963. Mesure de la Inc. (London), Ltd., London. sensibilitk du bacille tuberculeux aux drogues antibacillaires par 22. Rafidinarivo, E., J. C. PromC, and V. Uvy-FrCbault. 1985. New la mkthode des proportions. Rev. Tuberc. Pneumol. 27:217-272. kinds of unsaturated mycolic acids from Mycobacterium fallax Crosa, J. H., D. J. Brenner, and S. Falkow. 1973. Use of a sp. nov. Chem. Phys. Lipids 35:215-228. single-strand-specific nuclease for analysis of bacterial and 23. Silcox, V. A., R. C. Good, and M. M. Floyd. 1981. Identification plasmid deoxyribonucleic acid homo- and heteroduplexes. J. of clinically significant Mycobacterium fortuitum complex iso- Bacteriol. 115:904-911. lates. J. Clin. Microbiol. 14:686-691. DaffC, M., M. A. LanCelle, C. Asselineau, V. Uvy-FrCbault, and 24. Skerman, V. B. D., V. McGowan, and P. H. A. Sneath (ed.). H. David. 1983. IntCrSt taxonomique des acides gras des myco- 1980. Approved lists of bacterial names. Int. J. Syst. Bacteriol. bacteries: proposition d'une methode d'analyse. Ann. Inst. 30:225420. Pasteur Microbiol. 134B:241-256. 25. Sokal, R. R., and P. H. A. Sneath. 1963. Principles of numerical VOL. 43, 1993 MYCOBACTERIUM BRUME SP. NOV. 413

taxonomy. W. H. Freeman and Co., San Francisco. techniques for use in systematic bacteriology in the genus 26. Szabo, I., and E. Vandra. 1963. Mycobactenurn minetti (Penso Mycobactenurn: tests for pigment, urease, resistance to sodium et. al., 1952): bacteriological and epidemiological observations. chloride, hydrolysis of Tween 80, and p-gaiactosidase. Int. J. Acta Microbiol. Acad. Sci. Hung. 10215-223. Syst. Bacteriol. 24:412-419. 27. Tsukamura, M. 1967. Identification of mycobacteria. Tubercle 30. Wayne, L. G., H. W. B. Engel, C. Grassi, W. Gross, J. Hawkins, 48:311-338. 28. Vestal, A. 1975. Procedures for the isolation and identification of P. A. Jenkins, W. Kiippler, H. H. Kleeberg, I. Krasnow, E. E. mycobacteria. Publication (CDC) 76-8230. Centers for Disease Nel, S. R. Pattyn, P. A. Richards, S. Showalter, M. Slosarek, I. Control, Atlanta. Szabo, I. Tarnok, M. Tsukamura, B. Vergmann, and E. Wolin- 29. Wayne, L. G., H. C. Engbaek, H. W. B. Engel, S. Froman, W. sky. 1976. Highly reproducible techniques for use in systematic Gross, J. Hawkins, W. Kappler, A. G. Karlson, H. H. Kleeberg, bacteriology in the genus Mycobactenurn: tests for niacin and I. Krasnow, G. P. Kubica, C. McDurmont, E. E. Nel, S. R. catalase and for resistance to isoniazid, thiophene-2-carboxylic Pattyn, K. H. Schriider, S. Showalter, I. Tarnok, M. Tsukamura, acid hydrazide, hydroxylamine, and p-nitrobenzoate. Int. J. B. Vergmann, and E. Wolinsky. 1974. Highly reproducible Syst . Bacteriol. 26:3 11-3 18.