INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Methane-oxidizing bacteria inhabit the aerobic interfaces of methanogenic environments and reduce the potential methane (CH₄) emissions from these environments (7, 15, 30). Atmospheric CH₄, which has a present mixing ratio of 1.7 parts per million of volume (ppmv), is also oxidized microbially in aerobic upland soils (15). This process represents about 10% of the atmospheric CH₄ sink (10).

The identity of these atmospheric-CH₄ oxidizers is unknown. Whereas soil CH₄ oxidation rates can remain steady for >4 months at 1.7 ppmv of CH₄ (34), calculations based on the kinetic constants of known methanotrophic species suggest that these organisms are incapable of such extended survival (6). Atmospheric CH₄ should not supply sufficient cellular maintenance energy plus reducing power for the methane monooxygenase (MMO) enzyme. Studies with Methylosinus trichosporium and Methylobacter albus (Methylomicrobium album) seem to confirm this (32, 34).

However, the kinetic properties of CH₄ oxidation in upland soils are different from those in pure methanotroph cultures. Apparent half-saturation constants [K_m(app)] of various type I and II methanotrophs range from 0.8 to 66 µM CH₄ (5, 18). While some of these values are overestimated due to diffusion limitation, a lower limit of 0.8 to 2 µM is probable (18). Environmental samples from the aerobic interfaces of methanogenic habitats, such as lake sediments and landfill cover soils, have K_m(app) values in the same range (7, 39). However, the K_m(app) values for upland forest and agricultural soils are 1 to 3 orders of magnitude lower, at 10 to 280 nM CH₄ (1, 3, 7, 8, 13). Although a lower-affinity activity can be induced by enrichment with atmospheres containing 10% CH₄ (1), the methanotrophs normally active in these soils seem to be adapted to reduced CH₄ levels. Either uncharacterized species are involved in atmospheric-CH₄ oxidation or unknown physiological changes are induced in known methanotrophic species living in these soils.

The K_m(app) for CH₄ consumption in an organic soil from Ottawa, Canada, was estimated as 80 to 90 nM (8). This is in the same order of magnitude as values measured in other aerobic upland soils, although slightly higher. Values as low as 10 nM have been measured in soils (1, 3, 7, 13). Here we report on experiments aimed at enriching and characterizing the organisms responsible for the high-affinity activity in this organic soil.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Sampling site. The study site has been described previously (8, 9). It is an organic (60% combustible matter), neutral (pH 6.7 to 7.2) soil located on the Central Experimental Farm of Agriculture and Agri-Food Canada in Ottawa. The soil was sampled from a depth of 5 to 20 cm in August 1993.

Enrichment of soil with <275 ppmv of CH₄. Enrichment cultures were made in nitrate mineral salts medium (NMS) (14) containing 3 nM Cu and 1 mM phosphate buffer at pH 6.0. Deionized distilled or twice-distilled water was used. Initially, 0.15 g of soil was added to 10 ml of NMS in 125-ml serum vials. The vials were capped with autoclaved butyl rubber stoppers, and CH₄ was added at a final gaseous mixing ratio of 75 ppmv. The enrichment cultures were incubated at 25°C. The CH₄ was replaced after declining to below 25 ppmv. After 7 months, and periodically thereafter for 4 years, subsamples of the enrichment culture were transferred into fresh medium. The CH₄ mixing ratio in the vials varied considerably during this enrichment period. It declined to 1 to 50 ppmv of CH₄ before being replaced but never exceeded 275 ppmv.

Kinetic experiments. Half-liter batches of the culture were grown on <200 ppmv of CH₄ for 1 to 4 months. Since our aim was to study substrate affinity rather than specific activity, cell counts were not done. However, from the dilution series described in the previous section, methanotroph densities of <10⁷ cells ml¹ could be expected after such incubation times. Aliquots (2 to 8 ml, depending on the experiment) of the culture were transferred to 14-ml serum vials, and chloramphenicol at a final concentration of 50 µg liter¹ was added to prevent further enzyme production. The vials were capped with butyl rubber stoppers and injected with CH₄ at mixing ratios ranging from 5 to 1,000 ppmv. The vials were incubated at 25°C and rotated around the short axis at 21 rotations per min. At 1- to 2-day intervals for 3 to 10 days, CH₄ was measured by injection of 0.3-ml gas samples into a Carlo Erba gas chromatograph equipped with a flame ionization detector (oven temperature, 100°C; injector temperature, 140°C; 3-m by 3-mm Porapak Q column).

Transfer of the enrichment culture into CH₄ at higher mixing ratios. The enrichment culture was inoculated into 50 ml of NMS in 125-ml serum vials and grown on CH₄ at initial mixing ratios of 1,000 ppmv and 1%. The CH₄ was replaced after declining to <20% of these initial values. The CH₄ was replaced five times for the 1,000-ppmv treatment and six times for the 1% treatment. Kinetic experiments were then performed as described above at a range of 20 to 3,000 ppmv of CH₄ (for the 1,000-ppmv enrichment culture, 1-ml aliquots were diluted to 2 ml and measured at 2-day intervals for 6 days; for the 1% enrichment culture, 1-ml aliquots were diluted to 6 ml total and measured at 1 or 2 intervals of 3 h).

Isolations. Members of the low-CH₄ enrichment culture (<275 ppmv of CH₄) were isolated by performing a decimal dilution series in liquid NMS medium and making spread plates of each dilution onto NMS medium solidified with 15 g of Bacto Agar (Difco Laboratories, Detroit, Mich.) liter¹. The plates were incubated at 25°C in closed chambers containing a gaseous mixing ratio of 3% CH₄. Colonies which formed on plates of the highest two dilutions showing growth (10⁵ to 10⁶ times the initial inoculum) were restreaked for isolation (i) onto NMS solidified with Noble agar (Difco) and incubated under 3% CH₄ and (ii) onto plates of a general medium for culturing nonfastidious organisms, R2A agar (Difco), and incubated under air. Because we failed to isolate a methanotroph in this way (see Results), the procedure was repeated with a sample of the enrichment culture grown on 1% CH₄ (see the previous section).

DNA extraction. Extraction of DNA from the low-CH₄ enrichment culture (<275 ppmv of CH₄) and from the various isolates was adapted from the procedure of Moré et al. (28). Approximately 1 g of sterilized (170°C for 4 h) zirconia-silica beads of 0.1-mm diameter (Biospec Products Inc., Bartlesville, Okla.) was added to cell pellets in 2-ml screw-cap tubes. The cells and beads were suspended by vortexing them in 800 µl of NaH₂PO₄-Na₂HPO₄ buffer (120 mM; pH 8) and 260 µl of sodium dodecyl sulfate solution (10% [wt/vol] sodium dodecyl sulfate, 0.1 M NaCl, 0.5 M Tris-HCl, pH 8). The cells were lysed in a bead beater (Fastprep FP120; Savant Instruments Inc., Farmingdale, N.Y.) at a setting of 6.5 m s¹ for 45 s. The tubes were centrifuged for 3 min at 12,000 × g, and the supernatant was collected. The beads were then resuspended in 700 µl of NaH₂PO₄-Na₂HPO₄ buffer, and the DNA extraction was repeated. Proteins and debris were precipitated from the supernatant by adding 2.5 volumes of 7.5 M ammonium acetate, incubating the solution for 5 min on ice, and centrifuging it at 12,000 × g for 3 min. The DNA was precipitated by centrifugation (4°C; 45 min; 12,000 × g) with 70% (vol/vol) isopropanol added. The DNA pellet was washed with 70% (vol/vol) ethanol at 4°C, dried, and resuspended in 200 µl of elution buffer (Bio-Rad, Munich, Germany).

Comparative 16S rRNA gene sequence analysis. PCR-mediated amplification of the 16S rRNA genes from positions 28 to 1491 (numbering according to the International Union of Biochemistry nomenclature for Escherichia coli 16S rRNA) and sequencing analyses were done as previously described (22). Phylogenetic placement of strain LR1 was performed with the ARB program package (37). The 16S ribosomal DNA (rDNA) sequence of strain LR1 was integrated into a database of about 6,000 complete or partial bacterial 16S rRNA sequences (24, 31, 38) with the automatic alignment tool of the ARB program package. This procedure showed strain LR1 to be a member of the alpha subclass of the class Proteobacteria. The phylogenetic position of strain LR1 was determined in more detail by comparing its 16S rDNA gene sequence with alpha proteobacterial reference sequences. The tree topology was evaluated by a distance matrix analysis. For phylogenetic inference, the only nucleotide sequence positions considered were those which contained identical nucleotides in at least 50% of a representative selection of 16S rRNA sequences from the major lineages of the alpha subclass of proteobacteria (1,353 nucleotide sequence positions). Evolutionary distance values between pairs of microorganisms were calculated with the Felsenstein correction (ARB) (11). The tree was constructed by the neighbor-joining algorithm (33). The statistical significance values of interior branch points were tested in a bootstrap analysis by the neighbor-joining method (ARB; 1,000 data resamplings).

Comparative analysis of pmoA and mxaF gene fragments. (i) PCR amplification. Primers targeting gene fragments of pmoA (A189f and A682r) (16), coding for a subunit of particulate MMO (pMMO), and of mxaF (mxaF f1003 and mxaF r1561) (25, 27), coding for the large subunit of methanol dehydrogenase, were used for PCR amplification. A GC clamp (CCCCCCCCCCCCCGCCCCCCGCCCCCCGCCCCCGCCGCCC) was attached to the 5' end of each forward primer. PCR was performed in 50-µl reaction mixtures with an Eppendorf (Hamburg, Germany) gradient cycler. Taq polymerase (Perkin-Elmer Applied Biosystems, Branchburg, N.J.) was added to 0.5 µM concentrations of each primer and a PCR premix (Epicentre Technologies, Madison, Wis.). The mxaF fragment was amplified as described previously (27). For the pmoA amplification a touchdown program was developed, consisting of an initial denaturing step (94°C; 3 min) followed by 20 touchdown cycles (62 to 52°C), eight further cycles (52°C for 1 min followed by 72°C for 45 s), and a final extension (72°C; 5 min). The PCR products were analyzed on 3% agarose gels stained with ethidium bromide.

(ii) DGGE. PCR products were separated on 1-mm-thick polyacrylamide gels (6.5% [wt/vol] 37.5:5 acrylamide-bisacrylamide) (Bio-Rad) poured on GelBond support medium (FMC Bio Products, Rockland, Maine). They were prepared and run in Tris-acetic acid-EDTA buffer (0.04 M Tris-base, 0.02 M sodium acetate, 1 mM EDTA) at pH 7.4 and 60°C. The Dcode electrophoresis system (Bio-Rad) was used for separation. Denaturing gradient gel electrophoresis (DGGE) conditions for the various PCR products were optimized by perpendicular DGGE. For the pmoA fragments, a gradient of 35 to 80% denaturant (80% corresponded to 6.5% [vol/vol] acrylamide, 5.6 M urea, and 32% [vol/vol] deionized formamide) at a constant voltage of 200 V for 6 h was used. For the mxaF fragments, a gradient of 20 to 70% denaturant at a constant voltage of 150 V for 5 h was used. The gels were stained with 1:50,000 (vol/vol) SYBR-green I (Biozym, Hessisch-Oldendorf, Germany) for 30 min and then scanned on a PhosphorImager (Molecular Dynamics, Sunnyvale, Calif.).

(iii) DNA sequencing and construction of pmoA- and mxaF-based trees. Distinct bands were excised from the SYBR-green-stained gels with sterile pipette tips and suspended in 200 µl of H₂O. The bands were reamplified and rerun on DGGE to ensure purity.

Comparative sequence analysis of mmoX. A gene fragment of mmoX coding for the alpha subunit of soluble MMO (sMMO) was amplified from both the low-CH₄ enrichment culture (<275 ppmv of CH₄) and the methanotrophic isolate LR1 with primers (mmoX f882 and mmoX r1403) under PCR conditions described previously (25). The resulting PCR products were checked for size and purity on a 1.5% agarose gel and purified with the Prep-A-Gene system (Bio-Rad). Sequencing was performed as described above.

Nucleotide sequence accession numbers. The nucleotide sequences of the nearly complete 16S rRNA gene and of partial mmoX, mxaF, and pmoA genes and gene fragments of strain LR1 have been deposited in the EMBL, GenBank, and DDBJ nucleotide sequence databases under accession no. Y18442, Y18440, Y18441, and Y18443, respectively.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Soil enrichment cultures grown on <275 ppmv of CH₄. Soil enrichment cultures grown on 10 to 20% CH₄ display low-affinity kinetics [K_m(app)> 1 µM (Table 1)]. We therefore chose a lower CH₄ mixing ratio in an attempt to enrich for higher-affinity methanotrophic activity.

View larger version (14K): [in this window] [in a new window]   FIG. 1.   Oxidation of CH4 at mixing ratios of <200 ppmv () in closed vials (125 ml) containing 50 ml of NMS inoculated with the enrichment culture at day 0. Methane was added to the vials at 34 and 46 days. The data are the means of results with six vials ± 1 standard error of the mean. , mean of results with three uninoculated vials.

Kinetics. A typical kinetic curve from the low-CH₄ enrichment culture is shown in Fig. 2. The measured K_m(app) in four trials varied from 56 to 186 nM CH₄ (40- to 132-ppmv mixing ratio) (Table 1). This range overlaps the K_m(app) estimated for the organic soil itself (Table 1).

View larger version (15K): [in this window] [in a new window]   FIG. 2.   Kinetic curve of CH4 oxidation in a soil enrichment culture which was grown continuously on CH4 at a gaseous mixing ratio of <275 ppmv. Each symbol represents the mean of results with two samples.

View larger version (15K): [in this window] [in a new window]   FIG. 3.   Kinetic curve of CH4 oxidation in a soil enrichment culture initially grown on CH4 at a gaseous mixing ratio of <275 ppmv and later on 1% CH4. Each symbol represents the mean of results with two samples. , samples diluted 1:1 with water.

Members of the methanotrophic enrichment culture. When samples of the low-CH₄ enrichment culture (<275 ppmv of CH₄) were plated onto NMS agar, various types of colonies grew. These had sizes and morphologies typical of heterotrophic organisms which grow on trace contaminants in the medium (14). Four separate organisms which grew when transferred to complex medium (R2A agar) were isolated. These were identified based on partial 16S rRNA gene sequences. The closest relatives of these isolates were "Pseudomonas pavonacea" IAM1155 (similarity, 99.0% based on an analyzed stretch corresponding to E. coli 16S rRNA numbering positions 104 to 1293), Variovorax paradoxus (similarity, 99.2%; numbering positions 158 to 1192), Bradyrhizobium elkanii USDA76 (similarity, 98.5%; numbering positions 28 to 819), and Hyphomicrobium vulgare MC-750 (similarity, 96.9%; numbering positions 28 to 908).

View larger version (22K): [in this window] [in a new window]   FIG. 4.   Phylogenetic tree, based on 16S rRNA gene sequences, showing the relationship of isolate LR1 to other type II methanotrophs of the Methylocystis/Methylosinus group and to representative members of the alpha subclass of the class Proteobacteria. The 16S rRNA gene sequence from Methylococcus capsulatus was used as an outgroup reference. The numbers indicate the bootstrap values (percentage of outcome) for the respective interior branch points from a neighbor-joining test. Only values above a threshold of 50% are shown. The scale bar represents the estimated number of base changes per nucleotide sequence position.

View larger version (13K): [in this window] [in a new window]   FIG. 5.   Unrooted phylogenetic tree, based on derived amino acid sequences of pmoA gene fragments, showing the identity of the pmoA product from isolate LR1 and that from the high-affinity enrichment culture (grown on <275 ppmv of CH4) and their relationship to other methanotrophic species. The distance bar represents percent dissimilarity.

View larger version (19K): [in this window] [in a new window]   FIG. 6.   Unrooted phylogenetic tree, based on derived amino acid sequences of mxaF gene fragments, showing the identity of the mxaF product from isolate LR1 and that from the high-affinity enrichment culture (grown on <275 ppmv of CH4) and their relationship to other methylotrophic species. The distance bar represents percent dissimilarity.

DGGE analyses. The results of the DGGE analyses suggested that there was a single methanotrophic species present in the low-CH₄ enrichment culture (<275 ppmv of CH₄). The pmoA primer system, which should detect all known methanotrophic species, produced a single band on the DGGE for the enrichment (Fig. 7). Although methanotrophs appear to contain two copies of pmoA (36), these are not always distinguishable on DGGE. Two pmoA bands were occasionally evident from methanotroph isolates when DGGE gels were silver stained, but only a single band was evident from the enrichment culture and from isolate LR1 (data not shown).

View larger version (59K): [in this window] [in a new window]   FIG. 7.   DGGE patterns of DNA, amplified with primers for mxaF and pmoA, from selected methanotrophic strains (first four lanes), the high-affinity enrichment culture (grown on <275 ppmv of CH4), the methanotrophic isolate LR1, a coculture of LR1 with a Variovorax sp., and a Hyphomicrobium sp. isolated from the enrichment culture.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Methanotrophs in soils exposed to only atmospheric CH₄ possess a higher affinity for CH₄ than do pure cultures of methanotrophic bacteria, which are routinely grown on CH₄ at mixing ratios of 10 to 20% (Table 1). There is no conclusive evidence indicating whether this high-affinity activity is carried out by novel organisms or simply by known methanotrophs behaving in an unknown manner. Recently it was shown that the K_m(app) (CH₄) in M. albus varied depending on whether the organisms were grown on CH₄ or methanol as an energy source (4). This demonstrates that the K_m(app) in known methanotrophic bacteria changes with culture conditions, although neither reported value was near the low-nanomolar K_m(app) values measured in upland soil (both were >1 µM) (4). The K_m(app) of soils enriched with CH₄ at high mixing ratios has also been shown to vary with the CH₄ supply, but again only in the K_m(app) range of >1 µM (19).

Our data suggest two major points: that a type II methanotrophic species in mixed culture can exhibit K_m(app) values which vary depending on the CH₄ supply and that the K_m(app) of such a culture can approach that measured in upland soils when low CH₄ mixing ratios (<275 ppmv) are used for growth. We produced a methanotrophic enrichment culture which, based on its pmoA, mmoX, and mxaF products, contained a single detectable methanotroph. This organism (LR1) was a type II species of the Methylosinus/Methylocystis group. In the enrichment culture, and in a coculture containing only LR1 and a Variovorax sp., growth on <275 ppmv of CH₄ resulted in K_m(app) values of 56 to 188 nM. However, when the CH₄ mixing ratio was raised to 1%, the K_m(app) shifted upward to a value typical for methanotrophic isolates (>1 µM [Table 1]).

The organic soil from which our enrichment cultures were made consumes atmospheric CH₄ over much of the year (9). We previously measured K_m(app) values of 60 to 280 nM CH₄ in this soil, estimating the value in the absence of inhibitory NH₄⁺ as 80 to 90 nM (8). Because the K_m(app) values in our cultures grown on <275 ppmv of CH₄ were in the same range as these values, species such as LR1 could be the active methane oxidizers in this soil. It is therefore unnecessary here to postulate novel organisms as the agents of atmospheric-methane oxidation. The 16S rRNA gene sequence, as well as sequences for the functional genes pmoA and mxaF, which are also useful phylogenetic markers for methanotrophic bacteria (26, 27), all showed that LR1 clusters with other type II species and does not belong to a novel lineage of methanotrophs.

The K_m(app) values in our cultures grown on <275 ppmv of CH₄ were lower than the lowest previously reported value for methanotrophic isolates. However, the values in our cultures, and in the organic soil itself, are higher than those measured in many other upland soils. These can be as low as 10 nM (1, 3, 13). This organic soil is periodically methanogenic (9) and possibly supports a methanotrophic flora different from that of other soils. Although organisms such as LR1 may account for atmospheric-CH₄ oxidation in this organic soil, other species may dominate in other soils. On the other hand, the fact that the K_m(app) in our cultures was variable suggests that even lower values may be inducible under more oligotrophic conditions.

The CH₄ concentration was a key factor in determining K_m(app), but this could be mediated through the MMO enzyme, the methanotroph, or the bacterial community as a whole. Only apparent kinetic coefficients are measured in a study with a mixed microbial community, and it cannot be assumed that these represent true enzyme properties. A variety of reasons could therefore account for the variable K_m(app). We propose three possibilities. (i) A modified form of MMO is responsible for high-affinity activity. (ii) The measured K_m(app) is diffusion limited and affected by physiological properties of methanotrophs which change with growth conditions. Although the specific CH₄ uptake rate in methanotrophs (10¹⁵ mol of CH₄ cell¹ h¹) (15) is 100 to 1,000 times lower than the maximum rate of CH₄ diffusion through the cell boundary layer (calculated with the van Smoluchowski equation [21]), diffusion limitation could occur if cells aggregate or if the cell envelope presents a diffusion barrier (21). In a diffusion-limited system the measured K_m(app) values are affected by the aggregation, the specific activity, or the envelope structure of the cells. (iii) It is difficult to apply simple Michaelis-Menton kinetics to a complex terreactant enzyme such as MMO. The K_m(app) for a substrate is a function of both an equilibrium constant (binding of substrate to enzyme) and a reaction rate constant (decomposition to product). These may be altered by the availability of cosubstrates. For example, it might be expected that the cosubstrate NADH is more limiting in methanotrophic cultures grown on <275 ppmv of CH₄ than in those grown on 1% CH₄. If NADH limitation slowed the decomposition of the MMO-CH₄ complex (although this does not agree with the present model of the sMMO reaction sequence [23]), a lower K_m(app) could result (35).

Methane oxidation rates in soil have been stimulated by incubation on >1,000 ppmv of CH₄ (1, 2, 29, 34). Transient or minor increases may follow incubation with less CH₄ (3, 12), but until now no study had produced a long-term methanotrophic enrichment with <300 ppmv of CH₄ (2, 4, 20, 34). It has been suggested that periodic methanogenic events (9) or alternate substrates, such as methanol (4, 17), are necessary for growth and maintenance of the methanotrophs which consume atmospheric CH₄ in soils. The results of the present study demonstrate that methanotrophs are more oligotrophic than previously believed but they do not contradict these theories. Our data suggest that the K_m(app) value of known methanotrophic species is dependent on environmental conditions, especially the CH₄ supply, and therefore that the low K_m(app) values measured in upland soils do not necessarily imply that novel species of "atmospheric-methane oxidizers" exist.

In summary, we propose that organisms closely related to known type II methanotrophic species contribute to high-affinity atmospheric CH₄ consumption in an organic soil and that the K_m(app) value of methanotrophic organisms is determined by the conditions under which they are cultured, especially by the CH₄ supply.