Enrichment of High-Affinity CO Oxidizers in Maine Forest Soil

doi:10.1128/AEM.67.8.3671-3676.2001

Journal List > Appl Environ Microbiol > v.67(8); Aug 2001

Appl Environ Microbiol. 2001 August; 67(8): 3671–3676.

doi: 10.1128/AEM.67.8.3671-3676.2001.

PMCID: PMC93070

Enrichment of High-Affinity CO Oxidizers in Maine Forest Soil^†

Kathleen R. Hardy and Gary M. King^*

Daring Marine Center, University of Maine, Walpole, Maine 04573

^*Corresponding author. Mailing address: Daring Marine Center, University of Maine, Walpole, ME 04573. Phone: (207) 563-3146, ext. 207. Fax: (207) 563-3119. E-mail: gking/at/maine.edu.

Received January 12, 2001; Accepted May 15, 2001.

This article has been cited by other articles in PMC.

Abstract

Carboxydotrophic activity in forest soils was enriched by incubation in a flowthrough system with elevated concentrations of headspace CO (40 to 400 ppm). CO uptake increased substantially over time, while the apparent K_m (_appK_m) for uptake remained similar to that of unenriched soils (<10 to 20 ppm). Carboxydotrophic activity was transferred to and further enriched in sterile sand and forest soil. The _appK_ms for secondary and tertiary enrichments remained similar to values for unenriched soils. CO uptake by enriched soil and freshly collected forest soil was inhibited at headspace CO concentrations greater than about 1%. A novel isolate, COX1, obtained from the enrichments was inhibited similarly. However, in contrast to extant carboxydotrophs, COX1 consumed CO with an _appK_m of about 15 ppm, a value comparable to that of fresh soils. Phylogenetic analysis based on approximately 1,200 bp of its 16S rRNA gene sequence suggested that the isolate is an α-proteobacterium most closely related to the genera Pseudaminobacter, Aminobacter, and Chelatobacter (98.1 to 98.3% sequence identity).

Carbon monoxide (CO) regulates concentrations of hydroxyl radical (the primary oxidizing agent in the troposphere [33, 39, 40]) and several greenhouse gases (e.g., methane and ozone [16, 27]). Due to its direct and indirect effects on atmospheric chemistry, Daniel and Solomon (19) have suggested that short-term cumulative radiative forcing due to anthropogenic CO emission may be greater than that due to nitrous oxide (see also reference 25).

Although chemical oxidation in the troposphere consumes 75 to 85% of annual CO emissions (7, 8, 16), biological oxidation contributes significantly to CO regulation (10, 36). In particular, soils consume 7 to 25% of net global annual emissions (10, 36, 40, 45, 50). A number of studies have addressed various aspects of soil CO consumption (e.g., 2–4, 6, 12–15, 20, 28, 45), but much remains to be learned about the microorganisms involved.

Certain fungi (31), algae (9), actinomycetes and streptomycetes (4, 26), ammonium oxidizers (5, 32), and methanotrophs (5, 23, 30) oxidize CO. However, apparent half-saturation constants (_appK_m) for many of these organisms (e.g., 465 to 1,110 ppm [10, 11, 15]) substantially exceed values measured for soils (5 to 51 ppm [e.g., references 12 and 35]), with results for a thermophilic streptomycete (88 ppm) being exceptional (26). Accordingly, Conrad et al. (15) have concluded that known CO-oxidizing bacteria (carboxydotrophs) cannot account for observed soil CO consumption. In addition, King (35) has suggested that neither methanotrophs nor ammonia oxidizers are important CO oxidizers in a Maine forest soil.

Since enrichments for carboxydotrophs typically contain headspace CO concentrations of ≥10%, the populations isolated from them may represent taxa that are not representative of those that dominate activity in situ. Thus, in this study forest soil was incubated with a flow of air containing CO at 40 to 400 ppm. Sterile sand and soil inoculated with previously enriched soils were incubated similarly. Results indicated that high-affinity CO oxidizers could be enriched readily from forest soil. These enrichments were inhibited by CO concentrations of >1%. A novel eubacterial isolate consumed CO with an _appK_m of about 15 ppm but was inhibited by CO concentrations of >1%. A similar response to high CO concentrations was observed for fresh forest soils. These observations suggest that CO oxidizers in soils differ from known populations not only with respect to kinetics but also with respect to CO tolerance. Consequently, high CO concentrations appear unsuitable for enriching and isolating bacteria that dominate activity in situ.

MATERIALS AND METHODS

Site description and sampling.

Soils were obtained from a mixed deciduous-coniferous forest at the Darling Marine Center in Walpole, Maine. The soils have been described previously as typic haplorthods with organic horizons (O- horizons) varying from 5 to about 10 cm in thickness, with O-horizon pH values from 4 to 4.5, and with organic contents of about 50% (35, 37). Samples of the O-horizon (depths, 0 to 5 cm) were obtained from cores (6.3-cm inner diameter) or in the field by using a trowel after removing the litter layer. Soils were sieved (2-mm mesh size) prior to use, and water contents were adjusted to 65% as necessary.

In situ CO oxidizer enrichments.

A primary enrichment was established using 20-g (fresh weight) (gfw) soil samples that were transferred into 250-ml Erlenmeyer flasks, which were sealed subsequently with silicon stoppers. Triplicate flasks were flushed continuously (30 cm³ min⁻¹) with either sterile air containing 40 ppm CO or sterile ambient air containing 0.15 to 0.25 ppm CO. Gases flowed through 2-μm-pore-size cellulose acetate filters and were bubbled through sterile 0.1 M NaCl before introduction to incubation flasks. A third set of triplicate flasks was opened and exposed to ambient air for brief periods when soil subsamples were removed. At intervals, all flasks were opened and soil samples of 0.5 to 2.0 gfw soil were removed using a sterile spatula. Subsamples were transferred to 250-ml flasks or 160-ml culture bottles that were sealed for CO uptake assays. Initial concentrations of headspace CO were adjusted as necessary and assayed at intervals (see below). Controls containing no soil or sand were assayed for headspace CO to determine losses due to sampling methods. A second set of enrichment flasks was established and monitored similarly. However, it consisted of two 1-liter flasks, each containing 100 gfw of soil. One flask was incubated with a 20-cm³-min⁻¹ flow of air containing CO at concentrations that increased over time from 40 to 400 ppm. The second flask received a comparable flow of ambient air.

Secondary enrichments were initiated by transferring primary enrichments (from the flasks described above) to sterile soil or sand. Organic-free sand (organic C content, 0.002%) and sieved O-horizon soil were autoclaved for three 45-min cycles at 24-h intervals. Fifty grams (dry weight) (gdw) of sand or 25 gfw of autoclaved forest soil was added to sterile 500-ml Erlenmeyer flasks. The sand was supplemented with 5 ml of a 1:1 dilution of a sterile organic carbon-free medium containing mineral salts, a phosphate buffer, ferrous sulfate, a trace element mix, and 0.005% yeast extract (44). The sand was subsequently air dried to a water content of 5%. Forest soil (2 gfw) from the second primary enrichment was mixed gently with the sand, and 1 gfw was mixed with the autoclaved soil. Flasks were sealed and incubated with a 20-cm³-min⁻¹ flow of sterile air containing a CO concentration of 400 ppm.

To allow simultaneous assay of atmospheric-CO uptake and changes in microbial biomass, 80 gdw of autoclaved sand was added to each of six sterile 1-liter flasks containing 12.8 ml of sterile mineral salts medium with 0.05% yeast extract and a water content of 16%. Triplicate flasks received either a 20-cm³-min⁻¹ flow of sterile air containing 400 ppm CO or a comparable flow of ambient air. Flasks were inoculated with 0.5 gfw of sand that had been previously enriched with CO oxidizers (see above). Flasks were removed periodically from the flow lines, flushed with filtered ambient air, and assayed for atmospheric-CO uptake as well as for phospholipid phosphate (PL-P) content.

Response of enriched forest soils and sands to elevated (>1%) CO.

Triplicate 5-gfw samples of air-dried sterile forest soil were transferred to 120-ml culture bottles; sterile deionized water was added to yield a final water content of 25%. Bottles were then inoculated with 0.5 gfw of previously enriched forest soil. Triplicate control soil samples received sterile deionized water only. All samples were incubated with filtered air containing 400 ppm CO flowing through bottle headspaces at 20 cm³ min⁻¹. Gases were bubbled through 0.2 M NaCl; soil water contents were monitored gravimetrically and maintained at 25% by addition of sterile deionized water as needed. When atmospheric-CO uptake rates had increased substantially, CO uptake rates were assayed using initial CO concentrations ranging from atmospheric levels to 8%.

Sterile mineral salts medium containing 0.005% yeast extract was added to triplicate 33-gfw samples of fired sand in 1-liter Erlenmeyer flasks to yield a 12% water content. Flasks were inoculated with a secondary sand enrichment (atmospheric-CO uptake rate, 17 ng gdw⁻¹ h⁻¹), sealed with neoprene stoppers, and incubated with a headspace containing 20% CO in air. Control flasks containing uninoculated sand were sealed with 20% CO in the headspace to quantify CO loss from sampling and incubation procedures. After 8 days with no uptake of headspace CO, sample headspaces were flushed with sterile air and CO was added to a final concentration of 1,000 ppm. CO uptake at atmospheric levels and at 1,000 ppm was monitored regularly. Once CO consumption rates at 1,000 ppm had increased 60-fold, each of the triplicate samples was subdivided into two 10-gfw replicates that were incubated in sealed 500-ml Erlenmeyer flasks. Of the six samples thus produced, one set of triplicates was incubated with CO at 1,000 ppm and one set was incubated with CO at 20%. CO uptake rates at the respective incubation CO concentrations (1,000 ppm and 20% CO) were monitored through time as well as at atmospheric levels. Before atmospheric CO uptake assays were begun, flask headspaces were flushed for 1 h with air at 30 cm³ min⁻¹, held 24 h, and then flushed again for 1 h with air. Uninoculated control sands that were incubated with 20% CO were flushed similarly to determine the efficacy of the flushing procedure for reducing residual CO levels before atmospheric-CO uptake assays. Water content was monitored gravimetrically and maintained at 12% by addition of deionized water. Subsamples of sand were removed for PL-P analysis at the termination of the incubations.

Fresh forest soils were collected in December 1999 and prepared as previously described. Two sets of triplicate samples, each with 10 gfw of soil (water content, 80%, from 0- to 2-cm depth intervals) were transferred to 500-ml Erlenmeyer flasks that were sealed with neoprene stoppers. One set of samples was incubated with a static headspace containing 1,000 ppm CO. The other set was incubated with 20% CO in air. Headspace CO levels were maintained by CO addition as needed. CO was monitored by short-term assays of headspace concentrations.

Kinetic analyses.

Kinetic characteristics of enriched forest soils, sands, and control soils were based on uptake rates determined at various initial CO concentrations. A first-order uptake rate constant (k, corrected for CO production as appropriate [35]) was obtained at atmospheric or near-atmospheric CO concentrations, and a maximum rate of metabolism (V_max) was determined at saturating CO concentrations (>50 ppm). _appK_m was obtained from the equation V_max/k = _appK_m (45). In some cases, CO uptake was assayed with multiple initial CO headspace concentrations. _appK_m and V_max were estimated using a Michaelis-Menten model and a nonlinear curve-fitting algorithm (45) with Kaleidagraph software (Adelbeck Software, Inc.).

PL-P assays.

PL-P contents of sand and soils were assayed using a modified Bligh-Dyer procedure (24). Briefly, 2- to 5-gfw samples of sand or soil were added to 50-ml glass screw-cap tubes containing a solution of dichloromethane (DCM), methanol, and deionized water in a ratio of 1:2:0.8. After incubation at ambient temperature for a minimum of 24 h, aqueous and solvent phases were separated by addition of DCM and deionized water to a final ratio of 1:1:0.9. The upper methanolic phase was removed by siphoning. The DCM phase was transferred to 15-ml glass screw-cap tubes and evaporated in a stream of nitrogen. Dried samples were resuspended in 2 ml of DCM and sealed with Teflon-lined caps. Subsamples (100 to 1,000 μl) were transferred to glass ampoules. Solvent was evaporated in a stream of nitrogen, and the residue from each sample was resuspended in 450 μl of saturated potassium persulfate solution (0.185 M K₂S₂O₈ in 0.36 N H₂SO₄). Ampoules were sealed and held at 90°C for 12 h. One hundred microliters of ammonium molybdate solution (2.5% in 5.72 N H₂SO₄) was added to each ampoule after it was opened. After 10 min, 450 μl of malachite green solution (0.111% polyvinyl alcohol dissolved in 80°C water, with 0.011% added malachite green) was added. After 30 min, samples were transferred to 1.5-ml disposable polycarbonate cuvettes and absorbance at 610 nm was assayed spectrophotometrically (Beckman model DU-640 spectrophotometer).

CO analysis.

Gas samples were removed from flasks or bottles with needles and syringes and analyzed immediately. Samples containing CO concentrations less than 5 ppm were assayed with an RGA3 reduction gas analyzer (Trace Analytical, Inc.) or an RGD2 reduction gas detector (Trace Analytical, Inc.) in series with a model 8610C gas chromatograph (SRI Instruments, Inc.). Both instruments were equipped with mercury vapor detectors. Detection limits were approximately 1 ppb. Samples with headspace CO concentrations greater than 5 ppm were assayed using a gas chromatograph (model 3700 or 3400; Varian Instruments, Inc.) equipped with a molecular-sieve 5A column, a flame ionization detector, and a methanizer (SRI Instruments, Inc.). Detection limits were about 2 ppm. These instruments were standardized using certified CO standards (91.9 or 103 ppb CO [National Oceanic and Atmospheric Administration, Boulder, Colo.]; 10.8, 986, and 1,006 ppm CO [SpecAir Specialty Gases, Auburn, Maine]), laboratory dilutions of these standards, and 100% CO in CO-free air.

CO oxidizer isolation and initial characterization.

Tertiary sand enrichments (0.25 gfw) were transferred to 15-ml disposable centrifuge tubes containing 5 ml of phosphate buffer, 0.18% sodium pyrophosphate, and 20 μl of Tween 80. Tubes were mixed by vortexing them for 10 min and then centrifuged for 5 min at 200 × g. The supernatant (0.25 ml) was transferred to 160-ml serum bottles containing 10 ml of the previously described mineral salts medium with yeast extract (0.05%). Headspace CO concentrations were adjusted to 1,000 ppm and maintained by periodic additions of CO. Mixed cultures were incubated at 30°C with shaking at 125 rpm on a rotary shaker. After an increase in CO consumption rates and turbidity, cells adhering to the walls of a growth flask were collected for purification through a series of dilutions and transfers in liquid and solid media with CO until a morphologically uniform culture was obtained.

Gram reaction, oxygen requirements, fermentative capacity, motility, and morphology were determined using standard methods (47) and cultures that had been grown in a medium containing 25 mM sodium pyruvate, 0.005% yeast extract, and mineral salts as previously described (44). Pyruvate-grown cells were also harvested for analysis of their 16S rRNA gene sequences. Cells were lysed using a freeze-thaw cycle (three times at −80 to 65°C) and a mini-bead-beating procedure followed by DNA purification using standard methods (1). The 16S rRNA gene was amplified by PCR using previously published eubacterial primers (27F and 1492R [38]) and standard reaction conditions in a 50-μl volume. The PCR product (about 1,500 bp) was purified using agarose gel electrophoresis (1.0% agarose) and submitted to the University of Maine Sequencing Facility for a complete double-stranded sequence using previously published internal primers (357F, 926F, 519R, and 907R [38]). The resulting sequence was compared to other bacterial sequences following a BLAST search of the GenBank database. Closely related sequences were aligned. using CLUSTAL W (49). Gaps were removed (resulting in an ~1,200-bp sequence) prior to application of a maximum-likelihood algorithm (DNAml) with bootstrap analysis (100 replicates from Seqboot) using the PHYLIP package (version 3.5c; http://evolution.genetics.washington.edu/phylip.html [22]). TREEVIEW (version 1.6.5; http://taxonomy.zoology.gla.ac.uk/rod/treeview.html [43]) was used for visualizing the consensus tree derived from the Consense routine in PHYLIP.

CO utilization by the isolate was assayed for pyruvate-grown cells that were harvested in mid-log and stationary phases and resuspended to an optical density (A₆₀₀) of 1.2 to 1.9 in 10 ml of organic-free mineral salts medium contained in 120-ml serum bottles. Bottles were sealed with neoprene stoppers, and headspace CO concentrations were adjusted from 200 ppb to about 800 ppm for assays of CO uptake. Kinetic parameters were estimated using the Michaelis-Menten equation and a nonlinear-curve-fitting algorithm for the paired uptake rate and substrate concentration data or were calculated from V_max and first-order rate constants.

Nucleotide sequence accession number.

The sequence of the 16S rRNA gene isolated has been deposited in GenBank under accession number AF377867.

RESULTS

CO oxidizer enrichments.

Over a 56-day interval, atmospheric-CO uptake rates in primary enrichments established with 40 ppm CO in air increased 2.5 times relative to rates of controls incubated with ambient air (data not shown). Atmospheric-CO uptake by control soil with ambient air and soil incubated in sealed flasks with a static air headspace remained stable throughout the enrichment. The average atmospheric-CO uptake rate in the sealed flasks was approximately 60% of that for flasks receiving a flow of air with ambient CO levels.

The _appK_m of the initial CO-enriched soils (15 ± 4 ppm [mean ± standard error]) was lower than but not statistically different from (P > 0.1) values for control soils (23 ± 8 ppm) and similar to values for fresh forest soil (_appK_m, 17 ± 2 ppm [35]). The V_max of the enriched soil (15.4 ± 0.9 μg gdw⁻¹ h⁻¹) was 1.5 and 1.4 times greater than and statistically different from (P < 0.05) values for ambient controls (10.1 ± 0.9 μg gdw⁻¹ h⁻¹) and fresh forest soil (11.3 ± 0.4 μg gdw⁻¹ h⁻¹ [35]), respectively.

During a second primary enrichment, atmospheric-CO uptake rates increased over a period of 150 days, with a maximum rate 55 times that of a control with a flow of ambient air only. This was followed by a gradual decline (Fig. 1). Kinetic assays performed during the incubation revealed _appK_m values from 5 to 20 ppm (mean ± standard error, 10.2 ± 3.2 ppm; n = 4), which were comparable to values for controls (_appK_ms ranged from 9 to 26 ppm).

FIG. 1

Atmospheric-CO oxidation rates for forest soil incubated with a 20-cm³-min⁻¹ flow of CO at concentrations from 40 to 400 ppm ( [open circle]

) or ambient air (□) over time. ●, CO concentrations in flask inlets. d, day.

Atmospheric-CO uptake capacity in secondary enrichments based on sterile sand or forest soil also increased over time, followed by a gradual decline (data not shown). Kinetic assays for forest soils in these enrichments revealed _appK_m values from about 3 to 12 ppm. Values for sands were somewhat higher (21 ± 11 ppm; n = 4) but not statistically different from those for forest soils.

For tertiary enrichments based on sterile sand, atmospheric-CO uptake rates were 120-fold greater for treatments with a flow of 400 ppm CO in air than for treatments with a flow of ambient air (Fig. 2). Atmospheric-CO uptake rates decreased gradually when samples were switched from a flow of 400 ppm CO in air to ambient air, but uptake remained 60-fold greater than that of controls after 40 days. PL-P levels increased in the CO-enriched sands to maximum values approximately 12 times those of ambient controls. PL-P content remained stable for 40 days after CO-enriched sand was shifted to a flow of ambient air (Fig. 2).

FIG. 2

Atmospheric-CO oxidation rates (filled symbols) and phospholipid contents (open symbols) for autoclaved sand inoculated with previously enriched sand and incubated with a 20-cm³-min⁻¹ flow of 400 ppm CO (circles) or ambient air (filled squares) (more ...)

Responses to elevated (>1%) CO concentrations.

Atmospheric-CO uptake rates for a secondary enrichment based on autoclaved forest soil inoculated with previously enriched forest soil increased during incubation with a flow of 400 ppm CO in air. After 120 days of incubation, kinetic assays were conducted using initial CO headspace concentrations from 0.2 ppm to 8 to 10% CO. V_max and _appK_m values were calculated using data from headspace concentrations of <200 ppm, as rates obtained within this range varied according to a Michaelis-Menten model (_appK_ms, 7 ± 3 ppm; n = 3) (Fig. 3A). At initial headspace CO concentrations above 500 ppm, CO uptake increased, followed by a dramatic decrease for CO concentrations of >2% (Fig. 3B).

FIG. 3

(A) Representative response of CO oxidation rates for secondary forest soil enrichments as a function of initial headspace CO concentration. The curve fit represents a Michaelis-Menten model. Results are from one replicate of triplicate enrichments: (more ...)

A secondary enrichment based on sterile sand inoculated with enriched forest soil and incubated with 20% CO did not consume CO during an 8-day interval (data not shown). After reducing headspace concentrations to 1,000 ppm, CO uptake rates increased over time. In contrast, uptake rates for a parallel incubation with 20% CO decreased during the same interval (data not shown).

CO uptake by freshly collected forest soils incubated for 2 months with a headspace CO concentration of 1,000 ppm was about 4.5-fold greater than uptake by parallel fresh soil incubated with 20% CO. Values for the two treatments were 9.1 ± 1.8 and 2.0 ± 0.4 μg gdw⁻¹ h⁻¹, respectively.

Isolate characterization.

A CO-utilizing gram-negative nonmotile rod was obtained from a secondary sand enrichment. The isolate was obligately aerobic and nonfermentative. Pyruvate-grown cells that were harvested by centrifugation, washed, and transferred to a mineral salts medium containing 0.01% yeast extract consumed CO at rates that varied according to a Michaelis-Menten model over a CO concentration range from 200 ppb to 800 ppm. For cells that had entered stationary phase after growth on 25 mM pyruvate, the _appK_m was 14.8 ± 1.6 ppm (n = 3) and the V_max was 126 ± 4 nmol of CO mg of protein⁻¹ h⁻¹. Similar results were obtained for logarithmically growing cells, although the V_max was lower than in stationary phase.

A BLAST analysis using approximately 1,500 bp of the 16S rRNA gene sequence indicated that the isolate's closest relatives were within the Agrobacterium-Rhizobium subgroup of the α-proteobacteria. Results from maximum-likelihood analysis indicated that the isolate was similar to but distinct from Chelatobacter, Aminobacter, and Pseudaminobacter spp., with which it shares a 98.1 to 98.3% 16S rRNA gene sequence identity (Fig. 4).

FIG. 4

Phylogenetic analysis based on the 16S rRNA gene sequence of a CO-utilizing isolate, COX1, from a tertiary sand enrichment. Results are derived from maximum-likelihood analysis of a bootstrapped data set (resampled 100 times; branching is indicated at (more ...)

DISCUSSION

Although bacterial CO consumption has been well documented, the emphasis of most microbiological studies has been on utilization of CO at concentrations about 10⁶-fold greater than those in the atmosphere (11). The carboxydotrophs that use such high CO concentrations cannot account for activities observed in soil (15). Soil CO consumption involves a high-affinity uptake system that allows efficient CO utilization at levels from <0.1 to 0.3 ppm (e.g., references 12 and 35).

Previous efforts to enrich, isolate, and characterize high-affinity carboxydotrophs from soil have produced mixed results. Bartholomew and Alexander (3) have concluded that soil CO uptake involves a cometabolism not coupled to growth. In contrast, Spratt and Hubbard (48) have reported an increase in uptake over time for soils incubated with 200 ppm CO, which supported their conclusion that CO contributed to growth. Conrad and Seiler (13) have observed an increase over time in CO uptake by soil suspensions incubated for 80 days with a 0.5- to 1-ppm flow of CO in air. They also concluded that high CO levels could not enhance CO uptake by oligotrophic soil bacteria.

In the study reported here, Maine forest soil incubated with a flow of 40 to 400 ppm CO led to significant increases in rates of atmospheric CO uptake relative to those of controls receiving air only (Fig. 1). In secondary enrichments based on these soils, atmospheric-CO uptake rates increased during extended incubations with 400 ppm CO (Fig. 2). Simultaneous increases in atmospheric-CO uptake rates and PL-P levels demonstrate that CO uptake was likely coupled to growth since no such increases occurred in the absence of added CO (Fig. 2). Kinetic assays revealed a high-affinity CO uptake system in all enrichments (e.g., Fig. 3), with _appK_m values within the range for unenriched soils (5 to 50 ppm) (11, 12, 35).

The kinetics of the CO oxidizers enriched during this study differ substantially from those of extant carboxydotrophs, for which _appK_m values of 465 to 1,110 ppm have been reported (e.g., references 4, 11, 12, and 15). Notably, the soil and sand enrichments in this study consumed CO with _appK_m values comparable to those of unenriched soils (e.g., references 12 and 35) (Fig. 3A), in spite of incubations with CO concentrations more than 2 orders of magnitude greater than atmospheric levels. An isolate obtained from tertiary sand enrichments also expresses a high-affinity CO uptake system with an _appK_m (14.8 ± 1.6 ppm) similar to that of fresh soils. These results indicate that enrichments with low-to-moderate CO concentrations provide a useful strategy for obtaining isolates with characteristics similar to those expressed in situ. An analogous approach has been used with mixed success for enriching methanotrophs capable of growing with near-atmospheric methane levels (21, 46).

Both enriched soils and an isolate from them exhibited another trait found in fresh soils, but not in extant carboxydotrophs. In contrast to the ability of carboxydotrophs to tolerate CO concentrations up to 90% (17, 18, 34, 41, 42), fresh forest soils, soil enrichments, and a CO-utilizing isolate were inhibited by CO concentrations of approximately 1% (e.g., Fig. 3B). Hendrickson and Kubiseski (29) have also reported evidence consistent with inhibition of CO oxidation by high CO levels. In their study, net CO consumption ceased when forest soils were incubated with headspace CO concentrations of >16% and activity decreased in general at concentrations of >2%.

Although responses to elevated CO may vary among soils, results from this study suggest that incubation with high concentrations (>1%) of CO inhibits at least some populations that are important for atmospheric-CO oxidation and may favor enrichment of fast-growing, low-affinity carboxydotrophs. Enrichment of carboxydotrophs in a forest soil using a flow of 40 to 400 ppm CO facilitated isolation of a novel gram-negative α-proteobacterium (Fig. 4) that consumes CO with an _appK_m substantially lower than those of any previously isolated carboxydotrophs. Similar approaches applied to diverse soils may lead to additional novel CO oxidizers and new insights about the role of carboxydotrophs in the atmospheric-CO budget.

ACKNOWLEDGMENT

This work was supported in part by funds from the National Science Foundation (DEB-9728363).

Footnotes

^†Contribution 367 from the Darling Marine Center.

REFERENCES

Ausubel, F M; Brent, R; Kingston, R E; Moore, D D; Seidman, J G; Smith, J A; Struhl, K. Short protocols in molecular biology. 2nd edition. New York, N.Y: John Wiley and Sons, Inc.; 1992.

Badr, O; Probert, S D. Sinks and environmental impacts for atmospheric carbon monoxide. Appl Energy. 1995;50:339–372.

Bartholomew, G W; Alexander, M. Microbial metabolism of carbon monoxide in culture and in soil. Appl Environ Microbiol. 1979;37:932–937. [PubMed]

Bartholomew, G W; Alexander, M. Microorganisms responsible for the oxidation of carbon monoxide in soil. Environ Sci Technol. 1982;16:300–301.

Bedard, C; Knowles, R. Physiology, biochemistry, and specific inhibitors of CH₄, NH₄⁺, and CO oxidation by methanotrophs and nitrifiers. Microbiol Rev. 1989;53:68–84. [PubMed]

Bender, M; Conrad, R. Microbial oxidation of methane, ammonium and carbon monoxide, and turnover of nitrous oxide and nitric oxide in soils. Biogeochemistry. 1994;27:97–112.

Bergamaschi, P; Hein, R; Heimann, M; Crutzen, P J. Inverse modeling of the global CO cycle. 1. Inversion of CO mixing ratios. J Geophys Res. 2000;105D:1909–1927.

Bergamaschi, P; Hein, R; Brenninkmeijer, C A M; Crutzen, P J. Inverse modeling of the global CO cycle. 2. Inversion of ¹³C/¹²C and ¹⁸O/¹⁶O isotope ratios. J Geophys Res. 2000;105D:1929–1945.

Chappelle, E W. Carbon monoxide oxidation by algae. Biochim Biophys Acta. 1962;62:45–62. [PubMed]

10.

Conrad, R. Biogeochemistry and ecophysiology of atmospheric CO and H₂. Adv Microb Ecol. 1988;10:231–283.

11.

Conrad, R. Soil microorganisms as controllers of atmospheric trace gases (H₂, CO, CH₄, N₂O, and NO). Microbiol Rev. 1996;60:609–640. [PubMed]

12.

Conrad, R; Seiler, W G. Role of microorganisms in the consumption and production of atmospheric carbon monoxide by soil. Appl Environ Microbiol. 1980;40:437–445.

13.

Conrad, R; Seiler, W. Utilization of traces of carbon monoxide by aerobic oligotrophic microorganisms in ocean, lake and soil. Arch Microbiol. 1982;132:41–46.

14.

Conrad, R; Seiler, W. Characteristics of abiological carbon monoxide formation from soil organic matter, humic acids, and phenolic compounds. Environ Sci Technol. 1985;19:1165–1169.

15.

Conrad, R; Meyer, O; Seiler, W. Role of carboxydobacteria in consumption of atmospheric carbon monoxide by soil. Appl Environ Microbiol. 1981;42:211–215.

16.

Crutzen, P J; Gidel, L T. A two-dimensional photochemical model of the atmosphere, 2: the tropospheric budgets of the anthropogenic chlorocarbons CO, CH₄, CH₃Cl and the effect of various NO_x sources on tropospheric ozone. J Geophys Res. 1983;88:6641–6661.

17.

Cypionka, H; Meyer, O. Influence of carbon monoxide on growth and respiration of carboxydobacteria and other aerobic organisms. FEMS Microbiol Lett. 1982;15:209–214.

18.

Cypionka, H; Meyer, O. Carbon-monoxide insensitive respiratory chain of Pseudomonas carboxydovorans. J Bacteriol. 1983;156:1178–1187. [PubMed]

19.

Daniel, J S; Solomon, S. On the climate forcing of carbon monoxide. J Geophys Res. 1998;103:13249–13260.

20.

Duggin, J A; Cataldo, D A. The rapid oxidation of atmospheric CO to CO₂ by soils. Soil Biol Biochem. 1985;17:469–474.

21.

Dunfield, P F; Liesack, W; Henckel, T; Knowles, R; Conrad, R. High-affinity methane oxidation by a soil enrichment culture containing a type II methanotroph. Appl Environ Microbiol. 1999;65:1009–1014. [PubMed]

22.

Felsenstein, J. PHYLIP—phylogeny inference package (version 3.2). Cladistics. 1989;5:164–166.

23.

Ferenci, T; Strøm, T; Quayle, J R. Oxidation of carbon monoxide by Pseudomonas methanica. J Gen Microbiol. 1975;91:79–91. [PubMed]

24.

Findlay, R H; King, G M; Watling, L. Efficacy of phospholipid analysis in determining microbial biomass in sediments. Appl Environ Microbiol. 1989;55:2888–2893.

25.

Fuglestvedt, J S; Isaksen, I S A; Wang, W-C. Estimates of indirect global warming potentials for CH₄, CO and NO_x. Climatic Change. 1996;34:405–437.

26.

Gadkari, D; Schricker, K; Acker, G; Kroppenstedt, R N; Meyer, O. Streptomyces thermoautotrophicus sp. nov., a thermophilic CO- and H₂-oxidizing obligate chemolithoautotroph. Appl Environ Microbiol. 1990;56:3727–3734.

27.

Guthrie, P D. The CH₄ 000000000000 000000000000 000000000000 000000000000 111111111111 000000000000 000000000000 000000000000 000000000000 COOH conundrum: a simple analytical approach. Global Biogeochem Cycles. 1989;3:287–298.

28.

Heichel, G H. Removal of carbon monoxide by field and forest soils. J Environ Qual. 1973;2:419–423.

29.

Hendrickson, O Q; Kubiseski, T. Soil microbial activity at high levels of carbon monoxide. J Environ Qual. 1991;20:675–678.

30.

Hubley, J H; Mitton, J R; Wilkinson, J F. The oxidation of carbon monoxide by methane oxidizing bacteria. Arch Mikrobiol. 1974;95:365–368. [PubMed]

31.

Inman, R E; Ingersoll, R B. Uptake of carbon monoxide by soil fungi. J Air Pollut Control Assoc. 1971;21:646–647.

32.

Jones, R D; Morita, R Y. Carbon monoxide oxidation by chemolithotrophic ammonium oxidizers. Can J Microbiol. 1983;29:1545–1551.

33.

Khalil, M A K; Rasmussen, R A. The global cycle of carbon monoxide: trends and mass balances. Chemosphere. 1990;20:227–242.

34.

Kiessling, M; Meyer, O. Profitable oxidation of carbon monoxide or hydrogen during heterotrophic growth of Pseudomonas carboxydoflava. FEMS Microbiol Lett. 1982;13:333–338.

35.

King, G M. Attributes of atmospheric carbon monoxide oxidation by Maine forest soils. Appl Environ Microbiol. 1999;65:5257–5264. [PubMed]

36.

King, G M. Characteristics and significance of atmospheric carbon monoxide consumption by soils. Chemosph Global Change Sci. 1999;1:53–63.

37.

King, G M. Impacts of land use on atmospheric carbon monoxide consumption by soils. Global Biogeochem Cycles. 2000;14:1161–1172.

38.

Lane, D J. 16S/23S rRNA sequencing. In: Stackebrandt E, Goodfellow M. , editors; Stackebrandt E, Goodfellow M. , editors. Nucleic acid techniques in bacterial systematics. New York, N.Y: John Wiley and Sons, Inc.; 1991. pp. 115–175.

39.

Logan, J A; Prather, M J; Wofsy, S C; McElroy, M B. Tropospheric chemistry: a global perspective. J Geophys Res. 1981;86:7210–7254.

40.

Lu, Y; Khalil, M A K. Methane and carbon monoxide in OH chemistry: the effects of feedbacks and reservoirs generated by reactive products. Chemosphere. 1993;26:641–655.

41.

Meyer, O; Schlegel, H G. Reisolation of the carbon monoxide utilizing hydrogen bacterium Pseudomonas carboxydovorans (Kistner) comb. nov. Arch Microbiol. 1978;118:35–43. [PubMed]

42.

Mörsdorf, G; Frunzke, K; Gadkari, D; Meyer, O. Microbial growth on carbon monoxide. Biodegradation. 1992;3:61–82.

43.

Page, R D M. TREEVIEW: an application to display phylogenetic trees on personal computers. Comput Appl Biosci. 1996;12:357–358. [PubMed]

44.

Rich, J; King, G M. Carbon monoxide oxidation by bacteria associated with the roots of aquatic macrophytes. Appl Environ Microbiol. 1998;64:4939–4943. [PubMed]

45.

Sanhueza, E; Dong, Y; Scharffe, D; Lobert, J M; Crutzen, P J. Carbon monoxide uptake by temperate forest soils: the effects of leaves and humus layers. Tellus. 1998;50B:51–58.

46.

Schnell, S; King, G M. Stability of methane oxidation capacity to variations in methane and nutrient concentrations. FEMS Microbiol Ecol. 1995;17:285–294.

47.

Smibert, R M; Kreig, N R. Phenotypic characterization. In: Gerhardt P, Murray R G E, Wood W A, Kreig N R. , editors; Gerhardt P, Murray R G E, Wood W A, Kreig N R. , editors. Methods for general and molecular bacteriology. Washington, D.C.: ASM Press; 1994. pp. 607–654.

48.

Spratt, H G; Hubbard, J S. Carbon monoxide metabolism in roadside soils. Appl Environ Microbiol. 1981;41:1192–1201.

49.

Thompson, J D; Higgins, D G; Gibson, T J. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994;22:4673–4680. [PubMed]

50.

Yonemura, S; Kawashima, S; Tsuruta, H. Carbon monoxide, hydrogen and methane uptake in soils in a temperate arable field and a forest. J Geophys Res. 2000;105(D):14347–14362.

Articles from Applied and Environmental Microbiology are provided here courtesy of
American Society for Microbiology (ASM)