INTERNATIONALJOURNAL OF SYSTEMATICBACTERIOLOGY, Apr. 1991, p. 301-305 Vol. 41, No. 2 0020-7713/91/020301-05$02.00/0 Copyright 0 1991, International Union of Microbiological Societies

Contribution of Genome Characteristics to Assessment of of Obligate Methanotrophs J. P. BOWMAN,” L. I. SLY, AND A. C. HAYWARD Department of Microbiology, University of Queensland, Brisbane, Queensland, Australia 4072

The DNAs of a variety of obligate methanotrophic were analyzed for base composition and nucleotide distribution. Genome molecular weights were determined for representative strains. Similarity ‘\, maps attained by plotting DNA base composition versus nucleotide distribution and genome molecular weight showed that related species formed distinct clusters. Group I methanotrophs were found to form three clusters in both nucleotide distribution and genome size analyses. The first cluster consisted of five species, Methylomonas methanica, Methylomonas fodinarum, Methylomonas aurantiaca, “Methylomonas ru- bra,” and “Methylomonas agile.” The other clusters included both Methylomonas species and Methylococcus species, indicating the heterogeneity within these genera. One cluster contained low-G+ C-content Methylo- coccus strains and included Methylococcus whittenburyi, Methylococcus bovis, Methylococcus vinelandii, Methylococcus luteus, and ‘4Methylococcus ucrainicus. ” The type strains of Methylomonas pelagica and “Methylomonas alba” and a marine methanotrophic strain also clustered with the low-G+C-content Methylococcus group rather than with the genus Methylomonas. “Methylomonas gracilis” also appeared to be genetically distinct from the true Methylomonas species and clustered with the high-G+ C-content Methylococ- cus strains. This cluster included Methylococcus capsulatus, Methylococcus thermophilus, and other moderately thermophilic group I methanotrophic strains. The group I1 methanotrophs belonging to the genera “Methy- losinus” and “Methylocystis” formed separate generic clusters according to genome molecular weight data but not according to nucleotide distribution data.

Methanotrophic bacteria are a diverse group of gram- ing the shape of DNA melting profiles. Two measures can be negative bacteria which obligately utilize C, compounds, obtained. The first measure is the DNA melting transition including methane, methanol, and methylamine, as sole width, which is indicative of the nucleotide distribution in a sources of carbon and energy (3, 5, 11, 27). They are unable bacterial genome. The nucleotide distribution (al+ a,) is to utilize as sole energy sources any compounds with car- determined from the time interval required for DNA to bon-carbon bonds. The taxonomy of methanotrophs is based denature one standard deviation to the left (a,)and right (a,) mainly on morphological characters, and the systematics of of the melting temperature and is expressed as G+C content. this group remains confused and incomplete (26). Studies on The second measure is the ratio (a,/a,), which is a measure biochemical pathways, ultrastructure of intracytoplasmic of the skewness or asymmetry of the melting curve. The membranes (3, ll),16s rRNA sequences (25), and fatty acid results of De Ley (7) showed that nucleotide distributions contents (1, 2, 20) have shown that methanotrophs form two within bacterial genera appear to be consistent and thus can major groups (groups I and 11). The group I methanotrophs be used effectively in taxonomic studies. Likewise, the use include the genera Methylomonas and Methy Eococcus and of DNA renaturation techniques has proved to be a useful are characterized by the presence of stacked vesicular disks means to accurately determine genome molecular weight of intracytoplasmic membrane, the ribulose monophosphate (10). These properties of DNA were used in this study in an pathway for C, compound incorporation, and mainly 16:l attempt to observe the inter- and intrageneric relationships and 16:O fatty acids; these organisms appear to belong to the of various representatives of methane-utilizing bacteria and y subdivision of the . The group I1 methano- to assess the taxonomy of these organisms as a guide to trophs (the genera “Methylosinus” and “Methylocystis”) further study. are characterized by intracytoplasmic membranes that are arranged peripherally in the cell parallel to the cytoplasmic MATERIALS AND METHODS membrane, the serine pathway for C, compounds, and mainly 18:l fatty acid and belong to the 01 subdivision of the Bacterial strains. The strains used in this study are listed in proteobacteria. Currently only the group I genera Methy- Table 1. lomonas and Methylococcus have valid taxonomic status Cultivation. Methane-utilizing strains were grown on so- (17, 22). The other genera are not on the Approved Lists of lidified NMS medium (5) under a methane-air-CO, (5:4:1) Bacterial Names (17, 22) and have no standing in bacterial atmosphere. Marine methanotrophic strain A4 was culti- nomenclature. Chemotaxonomic and immunological analy- vated on NMS medium supplemented with 0.5% NaCl and 5 ses (4, 6, 8, 15, 16) have clearly shown that the group I ml of Staley vitamin solution (24) per liter. Methylomonas methanotrophs can be separated into three groups. pelagica was grown on NMS medium prepared with artificial The physicochemical properties of DNA have proved to seawater (derived from Difco marine agar 2216 [catalog no. be useful as a means of characterizing and identifying 0979]), supplemented with 5 ml of vitamin solution per liter, budding and hyphal bacteria (9, 13). DNA compositional and solidified with 1% agarose (type V) (21). Most strains nucleotide distribution studies (7) basically involve measur- were incubated in the dark at 28°C; the exceptions were Methylococcus capsulatus strains, “Methylomonas graci- lis” strains, and Methylococcus sp. strains JB87, JB137, * Corresponding author. JB146, and JB173, which were incubated at 45°C.

301 302 BOWMAN ET AL. INT. J. SYST.BACTERIOL.

TABLE 1. Physicochemical properties of DNAs isolated from obligate methanotrophs

~ ~-~~~ Nucleotide Laboratory G+C content distribution: Asymmetry: Genome size Strain (source)n no. (mol%) Ul + ur Ubr (lo9 Da) (mol% G+C) 1 Marine methanotrophic strain A4 (M. E. Lidstrom)b 55.4 14.4 1.5 1.6 2 Methylococcus bovis UQM 3504 (IMET 10593) 51.9 14.2 1.4 1.7 3 Methylococcus luteus UQM 3304T (NCIB 11914T)‘ 50.8 13.5 1.3 1.8 4 Methylococcus luteus IMV 3098T (Methylococcus bovis VKM-6T) 49.8 13.6 1.5 ND~ 5 “Methylococcus ucrainicus” UQM 3305* (NCIB 11915T) 48.7 13.9 1.3 1.8 6 Methylococcus vinelandii UQM 3586= (IMV 3030T) 52.8 13.3 1.3 1.5 7 Methylococcus whittenburyi UQM 3310 (NCIB 11128) 53.7 13.1 1.4 1.6 8 Methylococcus capsulatus UQM 3302 (NCIB 11132) 62.5 10.9 1.1 2.8 9 Methylococcus capsulatus JB175 62.3 10.9 1.1 2.4 10 Methylococcus sp. strain JB140 63.4 11.4 1.1 2.7 11 Methylococcus sp. strain JB87 65.1 10.3 1.0 ND 12 Methylococcus sp. strain JB137 65.2 10.7 1.1 3.0 13 Methylococcus sp. strain JB146 64.9 10.2 1.1 2.7 14 Methylococcus sp. strain JB173 65.6 11.5 1.1 ND 15 Methylococcus thermophilus UQM 35ST (IMV 3037T) 60.7 10.2 1.1 2.5 16 “Methylocystis parvus” UQM 3309T (NCIB 11129T) 66.0 9.4 1.2 2.5 17 “Methylocystis minimus” UQM 3497 (IMET 10578) 63.4 8.6 1.1 2.3 18 “Methylocystis echinoides” UQM 3495 (IMET 10491) 62.0 9.7 1.1 2.2 19 “Methylocystis” sp. strain JB170 64.4 9.4 1.1 2.0 20 “Methylocystis” sp. strain JB190 63.6 8.6 1.o ND 21 “Methylocystis” sp. strain JB196 66.9 9.7 1.0 2.5 22 “Methylomonas agile” UQM 3308 (NCIB 11124) 59.6 12.1 1.1 2.6 23 “Methylomonas alba” UQM 3314T (NCIB 11123T) 54.4 14.5 1.4 2.0 24 Methylomonas aurantiaca UQM 3406T 56.2 11.8 1.3 2.3 25 Methylomonas aurantiaca JB104 56.0 11.9 1.3 2.1 26 Methylomonas aurantiaca JB177B 56.9 11.8 1.3 ND 27 Methylomonas fodinarum UQM 3268T 57.8 11.2 1.3 1.8 28 Methylomonas fodinarum JB2 59.0 12.0 1.4 ND 29 Methylomonas fodinarum JB3 58.7 11.6 1.4 1.9 30 Methylomonas fodinarum JB4 58.6 12.0 1.4 ND 31 Methylomonas fodinarum JB6 58.1 11.3 1.3 ND 32 Methylomonas fodinarum JB7 58.2 11.8 1.3 2.1 33 Methylomonas fodinarum JB70 59.1 11.5 1.4 ND 34 “Methylomonas gracilis” UQM 3315T (NCIB 11912T) 60.1 11.1 1.0 2.8 35 “Methylomonas gracilis” JB185 59.5 10.5 1.o ND 36 “Methylomonas gracilis” JB187 59.3 10.8 1.0 ND 37 Methylomonas methanica UQM 3307T (NCIB 11130T) 53.4 10.6 1.2 2.0 38 Methylomonas methanica JB5 51.4 10.5 1.1 ND 39 Methylomonas methanica JB9 51.3 10.7 1.2 ND 40 Methylomonas methanica JB199 53.0 10.5 1.1 ND 41 Methylomonas methanica JB228 51.4 11.0 1.2 2.4 42 Methylomonas pelagica UQM 3505T (NCIMB 2265T) 48.5 13.8 1.4 1.5 43 “Methylomonas rubra” UQM 3303 (NCIB 11913) 52.8 10.6 1.2 2.2 44 “Methylosinus trichosporiurn” UQM 3311T (NCIB 11131T) 62.9 8.8 1.2 1.3 45 “Methylosinus sporiuin” UQM 3306T (NCIB 11126T) 67.1 9.2 1.2 1.5 46 “Methylosinus sporium” JB206 66.4 9.0 1.2 ND 47 “Methylosinus” sp. strain JB120 64.5 9.2 1.1 1.5 48 “Methylosinus” sp. strain HB-D1C 67.0 7.5 1.1 1.4 Abbreviations: UQM, Culture Collection, Department of Microbiology, University of Queensland, Brisbane, Australia; ATCC, American Type Culture Collection, Rockville, Md.; NCIB and NCIMB, National Collection of Industrial and Marine Bacteria, Aberdeen, Scotland; DSM, German Collection of Microorganisms, Braunschweig, Federal Republic of Germany; IMET, National Collection of Microorganisms, Institute for Microbiology, Jena, German Democratic Republic; IMV, Institute of Microbiology and Virology, Academy of Sciences, Kiev, USSR; VKM, All-Union Collection of Microorganisms, Academy of Science, Moscow, USSR; NCTC, National Collection of Type Cultures, Public Health Laboratory Service, London, United Kingdom; HB, Helen Byers, Department of Microbiology, University of Queensland, Brisbane, Australia (environmental isolates); JB, J. Bowman, Department of Microbiology, University of Queensland, Brisbane, Australia (environmental isolates). See reference 14. T = type strain. ND, Not determined.

DNA extraction. Cultures were grown on agar plates for 3 containing 1 mg of lysozyme per ml for 1 h at 37°C before to 5 days. Cell growth was harvested from plates with lysis. All strains were lysed by adding 2% (final concentra- distilled water and washed twice with saline-EDTA (0.1 M tion) sodium dodecyl sulfate at 60°C. DNAs were extracted disodium EDTA, 0.15 M NaC1; pH 8.0) by using centrifuga- and purified from cell lysates by using a modified Marmur tion at 10,000 x g for 10 min. Cells of “Methylosinus” and technique (23). “Methylocystis” strains were pretreated with a solution DNA analyses. DNA samples were analyzed by using a VOL. 41, 1991 METHANOTROPH GENOME CHARACTERISTICS 303

70

60 -

55 -

50 I

Low GtC Methylococcus spp 45 1 1 I I I I I 7 8 9 10 11 12 13 14 15

DNA MELTING TRANSITION WIDTH (MOL% GtC) FIG. 1. Similarity map of obligately methanotrophic bacteria based on DNA base compositions and nucleotide distributions. The numbers are the laboratory numbers for strains given in Table 1. microprocessor-controlled Gilford model 2600 spectropho- base composition (G+C content) (Fig. 1) and genome mo- tometer equipped with a thermoprogrammer. The Timethod lecular weight versus DNA base composition (G+C content) (23) was used to determine G+C contents; Escherichia coli (Fig. 2) were created from the DNA analysis data. ATCC 11775 (G+C content, 51.7 mol%) was used as the Nucleotide distributions. The group I methanotrophs internal standard. The nucleotide distributions of DNA formed three clusters on the nucleotide distribution-DNA samples were determined from the thermal denaturation base composition map. Strains of Methylococcus spp. pro- profiles generated by using the method of De Ley (7). duced two separate clusters. The first cluster contained Genome molecular weights were determined by using a Methylococcus capsulatus, Methylococcus thermophilus, DNA renaturation procedure in 5 x SSC (0.75 M NaCl, 0.075 and other moderately thermophilic strains with DNAs hav- M trisodium citrate; pH 7.0) containing 25% formamide ing high G+C contents (59 to 66 mol%) and transition widths E. coli (Sigma Chemical Co.) (13). The molecular weight of of 10.2 to 11.5 mol% G+C. Strains of “Methylomonas UQM 884 (= 10538) K-12 strain NCTC reference DNA was gracilis” taken to be 2.7 x lo9 (12). All procedures were repeated at also clustered with these strains. The second least four times. cluster consisted of low-G+C-content (48 to 55 mol%) mesophilic strains which had DNAs with broader melting transition widths (13.1 to 14.5 mol% G+C). This cluster RESULTS AND DISCUSSION included strains of Methylococcus whittenburyi, Methylo- The DNA base composition, nucleotide distribution, and coccus bovis , Methylococcus vinelandii, Methylococcus lu- genome molecular weight for each of the methanotrophs teus, and “Methylococcus ucrainicus.” The low-G+C-con- which we studied are shown in Table 1. Similarity maps tent Methylococcus cluster also included “Methylomonas constructed by plotting nucleotide distribution versus DNA alba ,” Methylomonas pelagica (21), and marine methan- 304 BOWMAN ET AL. INT. J. SYST.BACTERIOL.

70

Me thy losinus SPP.

Methylomonas SPP.

1.o 1.5 2.0 2.5 3.0 3.5 GENOME MOLECULAR WEIGHT ( 10 ’daltons) FIG. 2. Similarity map of obligately methanotrophic bacteria based on DNA base compositions and genome molecular weights. The numbers are the laboratory numbers for strains given in Table 1. otrophic strain A4 of Lidstrom (14). The species Methylomo- philus and Methylococcus capsulatus (Table l), with which nas methanica, Methylomonas fodinarum, Methylomonas it shares phenotypic characteristics that are common to aurantiaca, “Methylomonas rubra,” and “Methylomonas other moderately thermophilic group I methanotrophs (19, agile” exhibited DNA melting transition widths of 10.5 to 20) and are not present in most mesophilic group I metha- 12.1 mol% G+C and formed a single cluster. The DNA base notrophs. These characteristics include the ability to fix compositions of the members of this cluster ranged from 51 nitrogen and the ability to fix CO, via Calvin-Benson cycle to 60 mol% G+C. enzymes in the presence of methane (18, 19, 26). Methy- The group I1 methanotrophs (members of the genera lomonas pelagica also appeared to be unrelated to Methy- “Methylosinus” and “Methylocystis”) exhibited similar nu- lomonas methanica and clustered with the low-G+C-con- cleotide distributions and DNA base compositions and tent Methylococcus strains because of its significantly formed two overlapping clusters (Fig. 1). For these strains smaller genome size and the broad DNA melting transition the G+C contents ranged from 62 to 67 mol% and the widths of these organisms. The genome size of “Methylomo- transition widths ranged from 7.5 to 9.7 mol% G+C. nas alba” overlapped the genome sizes determined for Genome molecular weights. The genome molecular weights members of the Methylomonas cluster, but this organism of the methanotrophs which we studied were useful for had a significantly higher transition width, which was similar demonstrating relationships between taxa; these relation- to the transition widths determined for the low-G +C-content ships were consistent with the relationships determined by Methylococcus strains. These observations indicate that the using nucleotide distribution data. description given previously for the genus Methylomonas The Methylomonas species again fell into separate clus- (19) may be imprecise; this description has led to the ters. The genome sizes of Methylomonas methanica strains assignment of morphologically similar but genotypically were similar to those of Methylomonas fodinarum, Methy- unrelated species to the genus. lomonas aurantiaca, “Methylomonas rubra ,’ ’ and “Methy- Methylococcus capsulatus , Methylococcus thermophilus, lomonas agile” strains, varying from 1.8 X lo9 to 2.6 x lo9 and moderately thermophilic strains JB137, JB140, and daltons. On the other hand, “Methylomonas gracilis” ap- JB146 have genome sizes (2.4 x lo9 to 3.0 x lo9 daltons) that peared to be genotypically similar to Methylococcus thermo- are considerably larger than the genome sizes of the other VOL.41, 1991 METHANOTROPH GENOME CHARACTERISTICS 305 species of the genus Methylococcus (19), which also have rates. Eur. J. Biochem. 12:143-153. lower G+C contents (55 mol% or less). The latter species 11. Hou, C.-T. (ed.). 1984. Methylotrophy: microbiology, biochem- have genome sizes that range from 1.6 x lo9 to 2.0 x lo9 istry and genetics. CRC Press, Inc., New York. daltons. The genus Methylococcus was defined on the basis 12. Klotz, L. C., and B. H. Zimm. 1972. Size of DNA determined by viscoelastic measurements: results on bacteriophages, Bacillus of purely phenotypic characteristics (19) and is clearly subtilis and Escherichia coli. J. Mol. Biol. 72:779-800. genotypically heterogeneous. The results of several limited 13. Kolbel-Boelke, J., R. Gebers, and P. Hirsch. 1985. Genome size studies of both phenotypic and chemotaxonomic properties determinations for 33 strains of budding bacteria. Int. J. Syst. (1, 2, 6, 8, 20) support this view. Bacteriol. 35270-273. The group I1 methanotrophs (the genera “Methylosinus” 14. Lidstrom, M. E. 1988. Isolation and characterization of marine and “Methylocystis”) can be distinguished by their different methanotrophs. Antonie van Leeuwenhoek 54:189-199. genome molecular weights. “Methylocystis” genome sizes 15. Lysenko, A. M., V. F. Gal’chenko, and N. A. Chernykh. 1988. range from 2.0 x lo9 to 2.5 x lo9 daltons and are on the Taxonomic study of obligate methanotrophic bacteria using the average 50 to 100% larger than the genome sizes of “Meth- DNA:DNA hybridization technique. Mikrobiologiya 57553- ylosinus” 1.3 x x 658. strains (range, lo9 to 1.5 lo9 daltons). 16. Meyer, J., R. Haubold, J. Heyer, and W. Bockel. 1986. Contri- This study of the physicochemical properties of genomic bution to the taxonomy of methanotrophic bacteria: correlation DNAs provided information that is helpful for assessing the between membrane-type and GC-value. J. Basic Microbiol. taxonomic relationships among the genera and species of the 26:155-160. obligately methanotrophic bacteria; in particular, it demon- 17. Moore, W. E. C., and L. V. H. Moore. 1989. Index of the strated the genetic heterogeneity in the genera Methylomo- bacterial and yeast nomenclatural changes published in the nus and Methylococcus. This approach was undertaken as International Journal of Systematic Bacteriology since the 1980 the first stage of an extensive taxonomic study of these Approved Lists of Bacterial Names (1January 1980 to 1January organisms, and the results highlight the areas that need 1989). American Society for Microbiology, Washington, D.C. 18. Romanovskaya, V. A. 1979. Diagnostic characters of methane- nomenclatural revision. oxidizing bacteria. Mikrobiol. Zh. (Kiev) 41587-592. 19. Romanovskaya, V. A., Y. R. Malashenko, and V. N. REFERENCES Bogachenko. 1978. Corrected diagnosis of the genera and spe- 1. Andreev, L. V., and V. F. Gal’chenko. 1978. Fatty acid compo- cies of methane-utilizing bacteria. Mikrobiologiya 47:96-103. sition and identification of methanotrophic bacteria. Dokl. 20. Romanovskaya, V. A., Y. R. Malashenko, and N. I. Grish- Akad. Nauk. SSSR 269:1645-1653. chenko. 1981. Diagnosis of methane-oxidizing bacteria by nu- 2. Andreev, L. V., and V. F. Gal’chenko. 1983. Phospholipid merical methods based on cell fatty acid composition. Mikrobi- composition and differentiation of methanotrophic bacteria. J. ologiy a 49: 762-764. Liq. Chromatogr. 6:2699-2707. 21. Sieburth, J. M., P. W. Johnson, M. A. Eberhardt, M. E. 3. Anthony, C. (ed.). 1982. The biochemistry of methylotrophs. Sieracki, M. E. Lidstrom, and D. Laux. 1987. The first methane- Academic Press, Inc. (London), Ltd., London. oxidizing bacterium from the upper mixing layer of the deep 4. Bezrukova, L. V., Y. I. Nikolenko, A. I. Nesterov, V. F. ocean: Methylomonas pelagica sp. nov. Curr. Microbiol. 14: Gal’chenko, and M. V. Ivanov. 1983. Comparative serological 285-293. analysis of methanotrophic bacteria. Mikrobiologiya 52:800- 22. Skerman, V. B. D., V. McGowan, and P. H. A. Sneath (ed.). 805. 1980, Approved lists of bacterial names. Int. J. Syst. Bacteriol. 5. Bowman, J. P., L. I. Sly, J. M. Cox, and A. C. Hayward. 1990. 30: 225420. Methylomonas fodinarum sp. nov. and Methylomonas auranti- 23. Sly, L. I., L. L. Blackall, P. C. Kraat, T. Tian-Shen, and V. aca sp. nov.: two closely related type I obligate methanotrophs. Sangkhobol. 1986. The use of second derivative plots for the Syst. Appl. Microbiol. 13:278-286. determination of mol% guanine plus cytosine of DNA by the 6. Collins, M. D., and P. N. Green. 1985. Isolation and character- thermal denaturation method. J. Microbiol. Methods 5139-156. ization of a novel enzyme Q from some methane-utilizing 24. Staley, J. T. 1968. Prosthecomicrobium and Ancalomicrobium: bacteria. Biochem. Biophys. Res. Commun. 133:1125-1131. new freshwater prosthecate bacteria. J. Bacteriol. 951921- 7. De Ley, J. 1969. Compositional nucleoticie distribution and the 1942. theoretical prediction of homology in bacterial DNA. J. Theor. 25. Tsuji, K., H. C. Tsien, R. S. Hanson, S. R. DePalma, R. Scholtz, Biol. 101:738-754. and S. LaRoche. 1990. 16s ribosomal RNA sequence analysis 8. Gal’chenko, V. F., and A. I. Nesterov. 1981. Numerical analysis for determination of phylogenetic relationships among methylo- of protein electrophoretograms of obligate methanotrophic bac- trophs. J. Gen. Microbiol. 136:l-10. teria. Mikrobiologiya 50:973-979. 26. Whittenbury, R., and N. R. Krieg. 1984. Family IV. Methylo- 9. Gebers, R., U. Wehmeyer, T. Roggentin, H. Schlesner, J. coccaceae, p. 256-261. In N. R. Krieg and J. G. Holt (ed.), Kolbel-Boelke, and P. Hirsch. 1985. Deoxyribonucleic acid base Bergey’s manual of systematic bacteriology, vol. 1. The compositions and nucleotide distributions of 65 strains of bud- Williams & Wilkins Co., Baltimore. ding bacteria. Int. J. Syst. Bacteriol. 35260-269. 27. Whittenbury, R., K. S. Phillips, and J. F. Wilkinson. 1970. 10. Gillis, M., J. De Ley, and M. De Cleene. 1970. The determination Enrichment, isolation and some properties of methane-utilizing of molecular weight of bacterial genome DNA from renaturation bacteria. J. Gen. Microbiol. 61:205-218.