Vol. 58, No. 11
AND ENVIRONMENTAL MICROBIOLOGY, Nov. 1992, p. 3538-3541 0099-2240/92/113538-04$02.00/0 Copyright ©) 1992, American Society for Microbiology
APPLIED
Application of Gas Chromatography-Mass Spectrometry for Rapid Detection of Mycobacterium xenopi in Drinking Water SRINIVAS ALUGUPALLI,l LENNART LARSSON,l* MILAN SLOSAREK,2 AND MARCELA JARESOVA3 Department of Medical Microbiology, University of Lund, Solvegatan 23, 223 62 Lund, Sweden, 1 and National Institute of Public Health, 100 42 Prague 10,2 and Waterworks of Prague, 147 00 Prague,3 Czechoslovakia Received 19 May 1992/Accepted 10 August 1992
Gas chromatography-mass spectrometry (GC-MS) was used to detect 2-docosanol, a secondary alcohol characteristic of Mycobacterium xenopi, in 7 of 10 analyzed drinking water samples culture positive for that species. GC-MS was also used to detect tuberculostearic acid. Both of these chemical markers were analyzed as halogenated derivatives in the negative-ion-chemical-ionization mode. The numbers of CFU of M. xenopi were lowest in the three GC-MS-negative samples. The described GC-MS method is useful for the rapid detection of M. xenopi in drinking water. to another tube and centrifuged at 2,000 x g for 20 min. After removal of the supernatant, the sediment was mixed with 2 ml of sterile distilled water and homogenized by shaking. One milliliter of the homogenate was transferred to a glass ampoule, freeze-dried, and then used for GC-MS analysis; the remaining 1-ml aliquot was used for culturing (see
The presence of mycobacteria in drinking water represents a significant public health problem for persons at risk, including immunocompromised individuals (4, 5). Among the species encountered are the Mycobacterium avium complex, M. terrae, M. fortuitum, M. gastri, M. gordonae, M. chelonae, M. kansasii, M. scrofulaceum, and M. xenopi (3, 10). M. xenopi, the focus of this study, is a slow-growing opportunistic pathogen and one of the mycobacteria most frequently isolated from water, in particular from drinking water (6-8, 12, 13). Hydrolysis of the wax ester mycolates of M. xenopi liberates secondary alcohols, and the most abundant of them is 2-docosanol (9, 11). We recently developed gas chromatography-mass spectrometry (GC-MS) methods for use in detecting trace levels of mycobacterial secondary alcohols (2). In the present study, GC-MS was applied to the detection of 2-docosanol in drinking water; for optimal sensitivity, the pentafluorobenzoyl (PFBO) derivative was analyzed in the negative-ion-chemical-ionization mode. The same water samples were also analyzed for the presence of tuberculostearic acid (TSA), a characteristic branched-chain fatty acid of mycobacteria and related organisms. For comparison, the water samples were cultured for mycobacteria.
below). Culturing. The samples were shaken for 30 min with 3 ml of an aqueous solution containing 3% (wtlvol) sodium lauryl sulfate and 1% (wtlvol) sodium hydroxide. Thereafter, 25 ml of sterile distilled water acidified with 0.05% (vol/vol) aqueous hydrochloric acid was added, and the mixture was centrifuged at 2,000 x g for 20 min. The obtained sediment was mixed with 1 ml of sterile distilled water and shaken. Aliquots (0.25 ml) of the mixture were used to inoculate six Ogawa medium tubes: two undiluted, two after 10-fold dilution, and two after 100-fold dilution. The cultures were incubated at 37°C for 24 h in the horizontal position and then at 42°C (the optimum temperature for the growth of M. xenopi) in the vertical position. Results were recorded after TABLE 1. Detection of M. xenopi in drinking water by GC-MS and culturing
MATERIALS AND METHODS Chemicals. PFBO chloride and pentafluorobenzyl (PFB) bromide were from Janssen Chemica (Beerse, Belgium), tetrabutylammonium hydrogen sulfate was from Fluka AG (Buchs, Switzerland), triethylamine was from Sigma (St. Louis, Mo.), and the solvents acetonitrile, methanol, methylene chloride, and n-hexane were from Lab Scan (Dublin, Ireland). All chemicals used were of analytical grade. Water samples. Tap water was sampled in six residential flats in Prague, Czechoslovakia, with a previously reported occurrence of water-borne M. xenopi (13) and in four neighboring flats. Samples (1 liter) were collected in sterile glass containers and subsequently filtered (pore size, 0.45 ,um; Millipore, Bedford, Mass.). Each filter was cut into strips, which were then placed in 20 ml of sterile distilled water and shaken for 10 min; the resulting suspension was transferred *
Result of: Sample
GC-MS (mg of M. xenopi/500 ml)a
Ml
0.2 0.3 0.9 0.2 1.2 0 1.0 0.4 0 0
M3 M4 MS M6 M7 M8 M10
M11 M12
Culturing ml)b at: (CFU/500
6wk
12wk
0 0 0 0 0 0 0 1 0 0
1,600 1,300 2,700 108 1,800 4 180 80 28 20
a Calculated by comparing the levels of 2-docosanol found in the water samples with the levels found in known amounts of M. xenopi cells. b CFU were calculated by multiplying the number of colonies in the tubes by 2, 20, or 200, according to the corresponding dilution.
Corresponding author. 3538
VOL. 58, 1992
GC-MS FOR DETECTION OF M. XENOPI IN DRINKING WATER
3539
17 9 2 A geX.Ga
190
m/z 520
I1 41
8.906
1~_
7.27 ] eliMmILL la ..0
12.0
14.
a18. a
16 .
2B.. Z
13 35
m/z 297
;FS
'
Min
8.,Qe
--9-.-Qo
la: '00
II: O''o
12. go
.
13.09
..
'A 14.00
FIG. 1. Mass chromatograms of a drinking water sample showing extracted-ion current profiles for the detection of PFBO-derivatized 2-docosanol (top) and PFB-derivatized TSA (bottom).
6 weeks and again after 12 weeks. The isolates were identified by standard biochemical methods (15). All isolates showed microscopically identical colonies, and three colonies were always taken for identification. Preparation of samples for GC-MS. The freeze-dried samples were mixed with 1 ml of 30% (wt/vol) methanolic potassium hydroxide and heated at 80°C for 30 min. The samples were then extracted with hexane; the upper phase was separated and analyzed for the presence of 2-docosanol. The lower phase was acidified (pH < 2) by the addition of a few drops of 25% (vol/vol) aqueous sulfuric acid and subjected to a second hexane extraction; this phase was used for TSA analysis. PFBO derivatives of the alcohols were obtained by adding 35% (vol/vol) PFBO chloride (40 ,ul) and 2% (vol/vol) triethylamine (20 jl), both in acetonitrile, to the samples and then heating the samples at 80°C for 30 min. After the samples were cooled, hexane (1 ml) and water (1 ml) were added to the reaction mixture, which was then shaken; the organic phase was collected, evaporated, and dissolved in hexane.
The amounts of M. xenopi detected by GC-MS were calculated by comparing the levels of 2-docosanol detected in the samples with the levels found in known amounts (dry weight) of cultured M. xenopi cells. For analysis of TSA, the dried hexane phase containing the free acids (see above) was treated with 1 M aqueous sodium hydroxide (0.5 ml)-0.1 M aqueous tetrabutylammonium hydrogen sulfate (0.5 ml)-methylene chloride (1 ml). This mixture was shaken well, and the bottom layer (containing methylene chloride) was removed; to it was added 75 ,ul of 35% (vol/vol) PFB bromide in acetonitrile. The reaction was allowed to proceed at room temperature for 30 min, after which 1 ml of hexane-water (1:1 [vol/vol]) was added and the reaction mixture was shaken for 5 min; the hexane phase was then collected, evaporated, and dissolved in hexane. GC-MS. GC-MS analyses were carried out on a VG (Manchester, United Kingdom) Trio 1-S mass spectrometer coupled to a Hewlett-Packard (Avondale, Pa.) model 5890 gas chromatograph. A fused-silica capillary column (25 m by
3540
APPL. ENvIRON. MICROBIOL.
ALUGUPALLI ET AL.
528.088
100
ZFS-
521.80
231.800 II I* ,**J 2- **---. 88 8.6.8 I-,.*-----
I/z; LOG
150
200
258
35' 300
4i8
502.80 -_ I-
045
508
558
I
688
297.08
JTxFS
298.00
299.88D 181 . as
M/Z
1283
148
168
180
l
-M.
280 288 328 260 2280 240 38B FIG. 2. Mass spectra of PFBO-derivatized 2-docosanol (top) and PFB-derivatized TSA (bottom) recorded from the mass chromatograms shown in Fig. 1.
[inner diameter]) coated with cross-linked OV-1 used. Splitless injections were performed with a Hewlett-Packard model 7673 autosampler; the split valve was opened 1 min after injection. The carrier gas, helium, was used at an inlet pressure of 7 lb/in2. The column temperature was programmed from 120 to 260°C, at 20°C/ min; both the injector and the interface (between GC and MS) temperatures were kept at 260°C. Ionization was performed in the negative-ion-chemical-ionization mode (isobutane) at an ion source temperature of 150°C. 0.25
mm
was
RESULTS AND DISCUSSION
Figure 1 shows extracted-ion current profiles for the detection of 2-docosanol and TSA in one of the studied water samples. The negative-ion-chemical-ionization mass spectra of the PFBO-derivatized alcohol and PFB-derivatized acid exhibited abundant molecular ions of mlz 520 and m/z 297, respectively (Fig. 2). In the illustrated analysis, the signal-
to-noise ratio of the 2-docosanol peak was approximately 700, and that of the TSA peak was 2,300. The results of the culturing and GC-MS are summarized in Table 1. After 6 weeks of incubation, one sample (M10) was clearly culture positive (one colony). In addition, four samples (M5, M7, M8, and Mll) produced microcolonies that were visible under a magnifying glass but whose number and morphology were difficult to determine. All samples were culture positive for M. xenopi after 12 weeks of incubation; no other mycobacterial species were detected. Notably, the three GC-MS-negative samples contained only small numbers of CFU (Table 1). Despite the fact that both live and dead bacteria are detected by GC-MS, the results obtained with this method were in agreement with the culturing results. TSA and 2-docosanol were detected in the same seven samples. 2-Docosanol also is present in M. phlei (1, 14). Culture detection of this rapidly growing mycobacterium in drinking water can be accomplished after only a few days of cultiva-
FOR DETECTION OF M. XENOPI IN DRINKING WATER VC-MS VOL. 58, 1992
tion, whereas with a slow-growing mycobacterium, such as M. xenopi, an incubation time of several weeks is required, as shown in the present study. The reported GC-MS method is useful for the rapid detection of M. xenopi in drinking water, since results can be obtained within 2 days of receipt of the samples. ACKNOWLEDGMENTS This work received financial support from the Swedish National Association against Heart and Chest Diseases and the Faculty of Medicine, University of Lund, Lund, Sweden. REFERENCES 1. Alugupalli, S., and L. Larsson. Unpublished data. 2. Alugupalli, S., Z. Mielniczuk, and L. Larsson. 1992. Gas chromatography-mass spectrometry methods for analysis of secondary alcohols present in the Mycobacterium avium complex. J. Microbiol. Methods 15:229-240. 3. Carson, L. A., N. J. Petersen, M. S. Favero, and S. M. Aguero. 1978. Growth characteristics of atypical mycobacteria in water and their comparative resistance to disinfectants. Appl. Environ. Microbiol. 36:839-846. 4. Collins, C. H., J. M. Grange, and M. D. Yates. 1984. Mycobacteria in water. A review. J. Appl. Bacteriol. 57:193-211. 5. Du Moulin, G. C., and K. D. Stottmeier. 1986. Water-borne mycobacteria: an increasing threat to health. ASM News 52: 525-529. 6. Horak, Z., H. Polakova, and M. Kralova. 1986. Water-borne Mycobacterium xenopi-a possible cause of pulmonary mycobacteriosis in man. J. Hyg. Epidemiol. Microbiol. Immunol.
3541
30:405-409. 7. Jenkdns, P. A. 1991. Mycobacteria in the environment. J. Appl. Bacteriol. (Symp. Suppl.) 70:137S-141S. 8. Kubin, M. 1984. Distribution and ecology of mycobacteria in non-living reservoirs: opportunists and pathogens, p. 13131338. In G. P. Kubica and L. G. Wayne (ed.), The mycobacteria-a sourcebook. Part B. Marcel Dekker, Inc., New York. 9. Larsson, L., J. Jimenez, P. L. Valero-Guillen, F. Martin-Luengo, and M. Kubin. 1989. Establishment of 2-docosanol as a cellular marker compound in the identification of Mycobacterium xenopi. J. Clin. Microbiol. 27:2388-2390. 10. Lockwood, W. W., C. Friedman, N. Bus, C. Pierson, and R. Gaynes. 1989. An outbreak of Mycobacterium terrae in clinical specimens associated with a hospital potable water supply. Am. Rev. Respir. Dis. 140:1614-1617. 11. Luquin, M., F. Lopez, and V. Ausina. 1989. Capillary gas chromatographic analysis of mycolic acid cleavage products, cellular fatty acids, and alcohols of Mycobacterium xenopi. J. Clin. Microbiol. 27:1403-1406. 12. McSwiggan, D. A., and C. H. Collins. 1974. The isolation of M. kansasii and M. xenopi from water systems. Tubercle 55:291297. 13. glos6rek, M., M. Kubin, and M. Jaresova. Water-borne household infections due to Mycobacterium xenopi. J. Hyg. Epidemiol. Microbiol. Immunol., in press. 14. Toriyama, S., S. Imaizumi, I. Tomiyasu, M. Masui, and I. Yano. 1982. Incorporation of 180 into long-chain, secondary alcohols derived from ester mycolic acids in Mycobacterium phlei. Biochim. Biophys. Acta 712:427-429. 15. Vestal, A. L. 1975. Procedures for the isolation and identification of mycobacteria. Public Health Service publication no. 75-8230. Center for Disease Control, Atlanta.
AND ENVIRONMENTAL MICROBIOLOGY, Nov. 1992, p. 3538-3541 0099-2240/92/113538-04$02.00/0 Copyright ©) 1992, American Society for Microbiology
APPLIED
Application of Gas Chromatography-Mass Spectrometry for Rapid Detection of Mycobacterium xenopi in Drinking Water SRINIVAS ALUGUPALLI,l LENNART LARSSON,l* MILAN SLOSAREK,2 AND MARCELA JARESOVA3 Department of Medical Microbiology, University of Lund, Solvegatan 23, 223 62 Lund, Sweden, 1 and National Institute of Public Health, 100 42 Prague 10,2 and Waterworks of Prague, 147 00 Prague,3 Czechoslovakia Received 19 May 1992/Accepted 10 August 1992
Gas chromatography-mass spectrometry (GC-MS) was used to detect 2-docosanol, a secondary alcohol characteristic of Mycobacterium xenopi, in 7 of 10 analyzed drinking water samples culture positive for that species. GC-MS was also used to detect tuberculostearic acid. Both of these chemical markers were analyzed as halogenated derivatives in the negative-ion-chemical-ionization mode. The numbers of CFU of M. xenopi were lowest in the three GC-MS-negative samples. The described GC-MS method is useful for the rapid detection of M. xenopi in drinking water. to another tube and centrifuged at 2,000 x g for 20 min. After removal of the supernatant, the sediment was mixed with 2 ml of sterile distilled water and homogenized by shaking. One milliliter of the homogenate was transferred to a glass ampoule, freeze-dried, and then used for GC-MS analysis; the remaining 1-ml aliquot was used for culturing (see
The presence of mycobacteria in drinking water represents a significant public health problem for persons at risk, including immunocompromised individuals (4, 5). Among the species encountered are the Mycobacterium avium complex, M. terrae, M. fortuitum, M. gastri, M. gordonae, M. chelonae, M. kansasii, M. scrofulaceum, and M. xenopi (3, 10). M. xenopi, the focus of this study, is a slow-growing opportunistic pathogen and one of the mycobacteria most frequently isolated from water, in particular from drinking water (6-8, 12, 13). Hydrolysis of the wax ester mycolates of M. xenopi liberates secondary alcohols, and the most abundant of them is 2-docosanol (9, 11). We recently developed gas chromatography-mass spectrometry (GC-MS) methods for use in detecting trace levels of mycobacterial secondary alcohols (2). In the present study, GC-MS was applied to the detection of 2-docosanol in drinking water; for optimal sensitivity, the pentafluorobenzoyl (PFBO) derivative was analyzed in the negative-ion-chemical-ionization mode. The same water samples were also analyzed for the presence of tuberculostearic acid (TSA), a characteristic branched-chain fatty acid of mycobacteria and related organisms. For comparison, the water samples were cultured for mycobacteria.
below). Culturing. The samples were shaken for 30 min with 3 ml of an aqueous solution containing 3% (wtlvol) sodium lauryl sulfate and 1% (wtlvol) sodium hydroxide. Thereafter, 25 ml of sterile distilled water acidified with 0.05% (vol/vol) aqueous hydrochloric acid was added, and the mixture was centrifuged at 2,000 x g for 20 min. The obtained sediment was mixed with 1 ml of sterile distilled water and shaken. Aliquots (0.25 ml) of the mixture were used to inoculate six Ogawa medium tubes: two undiluted, two after 10-fold dilution, and two after 100-fold dilution. The cultures were incubated at 37°C for 24 h in the horizontal position and then at 42°C (the optimum temperature for the growth of M. xenopi) in the vertical position. Results were recorded after TABLE 1. Detection of M. xenopi in drinking water by GC-MS and culturing
MATERIALS AND METHODS Chemicals. PFBO chloride and pentafluorobenzyl (PFB) bromide were from Janssen Chemica (Beerse, Belgium), tetrabutylammonium hydrogen sulfate was from Fluka AG (Buchs, Switzerland), triethylamine was from Sigma (St. Louis, Mo.), and the solvents acetonitrile, methanol, methylene chloride, and n-hexane were from Lab Scan (Dublin, Ireland). All chemicals used were of analytical grade. Water samples. Tap water was sampled in six residential flats in Prague, Czechoslovakia, with a previously reported occurrence of water-borne M. xenopi (13) and in four neighboring flats. Samples (1 liter) were collected in sterile glass containers and subsequently filtered (pore size, 0.45 ,um; Millipore, Bedford, Mass.). Each filter was cut into strips, which were then placed in 20 ml of sterile distilled water and shaken for 10 min; the resulting suspension was transferred *
Result of: Sample
GC-MS (mg of M. xenopi/500 ml)a
Ml
0.2 0.3 0.9 0.2 1.2 0 1.0 0.4 0 0
M3 M4 MS M6 M7 M8 M10
M11 M12
Culturing ml)b at: (CFU/500
6wk
12wk
0 0 0 0 0 0 0 1 0 0
1,600 1,300 2,700 108 1,800 4 180 80 28 20
a Calculated by comparing the levels of 2-docosanol found in the water samples with the levels found in known amounts of M. xenopi cells. b CFU were calculated by multiplying the number of colonies in the tubes by 2, 20, or 200, according to the corresponding dilution.
Corresponding author. 3538
VOL. 58, 1992
GC-MS FOR DETECTION OF M. XENOPI IN DRINKING WATER
3539
17 9 2 A geX.Ga
190
m/z 520
I1 41
8.906
1~_
7.27 ] eliMmILL la ..0
12.0
14.
a18. a
16 .
2B.. Z
13 35
m/z 297
;FS
'
Min
8.,Qe
--9-.-Qo
la: '00
II: O''o
12. go
.
13.09
..
'A 14.00
FIG. 1. Mass chromatograms of a drinking water sample showing extracted-ion current profiles for the detection of PFBO-derivatized 2-docosanol (top) and PFB-derivatized TSA (bottom).
6 weeks and again after 12 weeks. The isolates were identified by standard biochemical methods (15). All isolates showed microscopically identical colonies, and three colonies were always taken for identification. Preparation of samples for GC-MS. The freeze-dried samples were mixed with 1 ml of 30% (wt/vol) methanolic potassium hydroxide and heated at 80°C for 30 min. The samples were then extracted with hexane; the upper phase was separated and analyzed for the presence of 2-docosanol. The lower phase was acidified (pH < 2) by the addition of a few drops of 25% (vol/vol) aqueous sulfuric acid and subjected to a second hexane extraction; this phase was used for TSA analysis. PFBO derivatives of the alcohols were obtained by adding 35% (vol/vol) PFBO chloride (40 ,ul) and 2% (vol/vol) triethylamine (20 jl), both in acetonitrile, to the samples and then heating the samples at 80°C for 30 min. After the samples were cooled, hexane (1 ml) and water (1 ml) were added to the reaction mixture, which was then shaken; the organic phase was collected, evaporated, and dissolved in hexane.
The amounts of M. xenopi detected by GC-MS were calculated by comparing the levels of 2-docosanol detected in the samples with the levels found in known amounts (dry weight) of cultured M. xenopi cells. For analysis of TSA, the dried hexane phase containing the free acids (see above) was treated with 1 M aqueous sodium hydroxide (0.5 ml)-0.1 M aqueous tetrabutylammonium hydrogen sulfate (0.5 ml)-methylene chloride (1 ml). This mixture was shaken well, and the bottom layer (containing methylene chloride) was removed; to it was added 75 ,ul of 35% (vol/vol) PFB bromide in acetonitrile. The reaction was allowed to proceed at room temperature for 30 min, after which 1 ml of hexane-water (1:1 [vol/vol]) was added and the reaction mixture was shaken for 5 min; the hexane phase was then collected, evaporated, and dissolved in hexane. GC-MS. GC-MS analyses were carried out on a VG (Manchester, United Kingdom) Trio 1-S mass spectrometer coupled to a Hewlett-Packard (Avondale, Pa.) model 5890 gas chromatograph. A fused-silica capillary column (25 m by
3540
APPL. ENvIRON. MICROBIOL.
ALUGUPALLI ET AL.
528.088
100
ZFS-
521.80
231.800 II I* ,**J 2- **---. 88 8.6.8 I-,.*-----
I/z; LOG
150
200
258
35' 300
4i8
502.80 -_ I-
045
508
558
I
688
297.08
JTxFS
298.00
299.88D 181 . as
M/Z
1283
148
168
180
l
-M.
280 288 328 260 2280 240 38B FIG. 2. Mass spectra of PFBO-derivatized 2-docosanol (top) and PFB-derivatized TSA (bottom) recorded from the mass chromatograms shown in Fig. 1.
[inner diameter]) coated with cross-linked OV-1 used. Splitless injections were performed with a Hewlett-Packard model 7673 autosampler; the split valve was opened 1 min after injection. The carrier gas, helium, was used at an inlet pressure of 7 lb/in2. The column temperature was programmed from 120 to 260°C, at 20°C/ min; both the injector and the interface (between GC and MS) temperatures were kept at 260°C. Ionization was performed in the negative-ion-chemical-ionization mode (isobutane) at an ion source temperature of 150°C. 0.25
mm
was
RESULTS AND DISCUSSION
Figure 1 shows extracted-ion current profiles for the detection of 2-docosanol and TSA in one of the studied water samples. The negative-ion-chemical-ionization mass spectra of the PFBO-derivatized alcohol and PFB-derivatized acid exhibited abundant molecular ions of mlz 520 and m/z 297, respectively (Fig. 2). In the illustrated analysis, the signal-
to-noise ratio of the 2-docosanol peak was approximately 700, and that of the TSA peak was 2,300. The results of the culturing and GC-MS are summarized in Table 1. After 6 weeks of incubation, one sample (M10) was clearly culture positive (one colony). In addition, four samples (M5, M7, M8, and Mll) produced microcolonies that were visible under a magnifying glass but whose number and morphology were difficult to determine. All samples were culture positive for M. xenopi after 12 weeks of incubation; no other mycobacterial species were detected. Notably, the three GC-MS-negative samples contained only small numbers of CFU (Table 1). Despite the fact that both live and dead bacteria are detected by GC-MS, the results obtained with this method were in agreement with the culturing results. TSA and 2-docosanol were detected in the same seven samples. 2-Docosanol also is present in M. phlei (1, 14). Culture detection of this rapidly growing mycobacterium in drinking water can be accomplished after only a few days of cultiva-
FOR DETECTION OF M. XENOPI IN DRINKING WATER VC-MS VOL. 58, 1992
tion, whereas with a slow-growing mycobacterium, such as M. xenopi, an incubation time of several weeks is required, as shown in the present study. The reported GC-MS method is useful for the rapid detection of M. xenopi in drinking water, since results can be obtained within 2 days of receipt of the samples. ACKNOWLEDGMENTS This work received financial support from the Swedish National Association against Heart and Chest Diseases and the Faculty of Medicine, University of Lund, Lund, Sweden. REFERENCES 1. Alugupalli, S., and L. Larsson. Unpublished data. 2. Alugupalli, S., Z. Mielniczuk, and L. Larsson. 1992. Gas chromatography-mass spectrometry methods for analysis of secondary alcohols present in the Mycobacterium avium complex. J. Microbiol. Methods 15:229-240. 3. Carson, L. A., N. J. Petersen, M. S. Favero, and S. M. Aguero. 1978. Growth characteristics of atypical mycobacteria in water and their comparative resistance to disinfectants. Appl. Environ. Microbiol. 36:839-846. 4. Collins, C. H., J. M. Grange, and M. D. Yates. 1984. Mycobacteria in water. A review. J. Appl. Bacteriol. 57:193-211. 5. Du Moulin, G. C., and K. D. Stottmeier. 1986. Water-borne mycobacteria: an increasing threat to health. ASM News 52: 525-529. 6. Horak, Z., H. Polakova, and M. Kralova. 1986. Water-borne Mycobacterium xenopi-a possible cause of pulmonary mycobacteriosis in man. J. Hyg. Epidemiol. Microbiol. Immunol.
3541
30:405-409. 7. Jenkdns, P. A. 1991. Mycobacteria in the environment. J. Appl. Bacteriol. (Symp. Suppl.) 70:137S-141S. 8. Kubin, M. 1984. Distribution and ecology of mycobacteria in non-living reservoirs: opportunists and pathogens, p. 13131338. In G. P. Kubica and L. G. Wayne (ed.), The mycobacteria-a sourcebook. Part B. Marcel Dekker, Inc., New York. 9. Larsson, L., J. Jimenez, P. L. Valero-Guillen, F. Martin-Luengo, and M. Kubin. 1989. Establishment of 2-docosanol as a cellular marker compound in the identification of Mycobacterium xenopi. J. Clin. Microbiol. 27:2388-2390. 10. Lockwood, W. W., C. Friedman, N. Bus, C. Pierson, and R. Gaynes. 1989. An outbreak of Mycobacterium terrae in clinical specimens associated with a hospital potable water supply. Am. Rev. Respir. Dis. 140:1614-1617. 11. Luquin, M., F. Lopez, and V. Ausina. 1989. Capillary gas chromatographic analysis of mycolic acid cleavage products, cellular fatty acids, and alcohols of Mycobacterium xenopi. J. Clin. Microbiol. 27:1403-1406. 12. McSwiggan, D. A., and C. H. Collins. 1974. The isolation of M. kansasii and M. xenopi from water systems. Tubercle 55:291297. 13. glos6rek, M., M. Kubin, and M. Jaresova. Water-borne household infections due to Mycobacterium xenopi. J. Hyg. Epidemiol. Microbiol. Immunol., in press. 14. Toriyama, S., S. Imaizumi, I. Tomiyasu, M. Masui, and I. Yano. 1982. Incorporation of 180 into long-chain, secondary alcohols derived from ester mycolic acids in Mycobacterium phlei. Biochim. Biophys. Acta 712:427-429. 15. Vestal, A. L. 1975. Procedures for the isolation and identification of mycobacteria. Public Health Service publication no. 75-8230. Center for Disease Control, Atlanta.
