Low-Concentration Kinetics of Atmospheric CH₄ Oxidation in Soil and Mechanism of NH₄⁺ Inhibition

Abstract

NH₄⁺ inhibition kinetics for CH₄ oxidation were examined at near-atmospheric CH₄ concentrations in three upland forest soils. Whether NH₄⁺-independent salt effects could be neutralized by adding nonammoniacal salts to control samples in lieu of deionized water was also investigated. Because the levels of exchangeable endogenous NH₄⁺ were very low in the three soils, desorption of endogenous NH₄⁺ was not a significant factor in this study. The K_m(app) values for water-treated controls were 9.8, 22, and 57 nM for temperate pine, temperate hardwood, and birch taiga soils, respectively. At CH₄ concentrations of ≤15 μl liter⁻¹, oxidation followed first-order kinetics in the fine-textured taiga soil, whereas the coarse-textured temperate soils exhibited Michaelis-Menten kinetics. Compared to water controls, the K_m(app) values in the temperate soils increased in the presence of NH₄⁺ salts, whereas the V_max(app) values decreased substantially, indicating that there was a mixture of competitive and noncompetitive inhibition mechanisms for whole NH₄⁺ salts. Compared to the corresponding K⁺ salt controls, the K_m(app) values for NH₄⁺ salts increased substantially, whereas the V_max(app) values remained virtually unchanged, indicating that NH₄⁺ acted by competitive inhibition. Nonammoniacal salts caused inhibition to increase with increasing CH₄ concentrations in all three soils. In the birch taiga soil, this trend occurred with both NH₄⁺ and K⁺ salts, and the slope of the increase was not affected by the addition of NH₄⁺. Hence, the increase in inhibition resulted from an NH₄⁺-independent mechanism. These results show that NH₄⁺ inhibition of atmospheric CH₄ oxidation resulted from enzymatic substrate competition and that additional inhibition that was not competitive resulted from a general salt effect that was independent of NH₄⁺.

Atmospheric CH₄ contributes substantially to the greenhouse effect, and the concentration of atmospheric CH₄ has increased dramatically in the past century because of human activity associated with agriculture, land use changes, and industry (34, 35). Bacterial oxidation of atmospheric CH₄ in well-drained soils is an important regulator of atmospheric CH₄ concentration, yet the organisms responsible remain unidentified and the physiology of the process is poorly understood (9, 35, 36). Although soil CH₄ consumption is inhibited by a wide variety of anthropogenic disturbances, such as agriculture, N deposition, and forestry (12, 17, 22, 23, 32, 43, 44), predictable inhibition patterns have failed to emerge, which has made it difficult to predict the effects of disturbance on soil CH₄ flux in various ecosystems. The most commonly reported disturbance effect is that of NH₄⁺ fertilizers, which can suppress soil CH₄ consumption by up to 70% (1, 8, 10, 17, 22, 32, 33, 37, 38, 43). In the field, inhibition may occur immediately following fertilization, may be delayed for months to years, or may never occur despite years of chronic fertilization (9, 17). This variety of responses may stem at least in part from the distribution of physiologically diverse methane oxidizer populations across sites (17, 18, 20).

Of the various NH₄⁺ inhibition patterns, immediate inhibition is the best documented. As in field studies, however, physiological laboratory studies have produced variable results, suggesting that there may be multiple inhibition mechanisms (15, 17, 26–28, 36, 39). Physicochemical similarities between CH₄ and NH₃ may permit these two compounds to compete for enzyme active sites so that fortuitous NH₃ oxidation competitively inhibits CH₄ oxidation (38). Although this mechanism has been demonstrated to occur in pure cultures of methanotrophic bacteria (6) and in a CH₄-producing agricultural soil (15), it has not been demonstrated to occur in well-drained, nonagricultural mineral soils, which comprise the dominant terrestrial sink for atmospheric CH₄ (14, 38, 45), nor has it been demonstrated to occur at near-atmospheric CH₄ concentrations. In many cases, the kinetics of immediate NH₄⁺ inhibition in soil cannot be reconciled easily with substrate competition (15, 16, 26–28, 39). An alternative mechanism has been proposed, whereby the toxicity of NO₂⁻ or NH₂OH produced by fortuitous NH₄⁺ oxidation suppresses methanotrophic activity (26, 27, 39). Hence, multiple inhibition mechanisms may be involved, and these mechanisms may vary with the physiology of different CH₄ oxidizer populations (17).

Two physiologically distinct communities of CH₄ oxidizers apparently exist in soil. One group, generally associated with atmospheric CH₄ consumption, exhibits an extremely high affinity for CH₄. Representatives of this group have yet to be cultivated or otherwise identified (9). The second group exhibits a much lower affinity for CH₄ and is generally associated with common methanotrophs, such as those that have been studied in pure culture for many years (2, 9). In upland mineral soils, only high-affinity activity is usually detectable without artificial enrichment with high CH₄ concentrations in the laboratory. However, the only prior study in which kinetic constants for NH₄⁺ inhibition of soil CH₄ oxidation were reported was conducted in a periodically moist, organic matter-rich agricultural soil with demonstrable methanogenesis (15, 16). Such a soil potentially harbors a rich community of CH₄ oxidizers representing a continuum from low-affinity organisms to high-affinity organisms. Although this important investigation demonstrated that NH₄⁺ inhibits CH₄ oxidation via enzymatic substrate competition in an agricultural humisol, it is unclear to what extent its results apply to well-drained mineral soils lacking endogenous CH₄ sources. Physiological studies of soil CH₄ oxidation typically derive kinetic constants from oxidation rates at CH₄ concentrations ranging from atmospheric levels (∼1.7 μl liter⁻¹) to ≫K_m for high-affinity CH₄ oxidizers. Even in soil in which only high-affinity organisms are active, the CH₄-oxidizing enzyme(s) could respond differently to NH₄⁺ at high CH₄ concentrations than at near-atmospheric concentrations (15, 39). Thus, to study NH₄⁺ inhibition of high-affinity CH₄ oxidizers per se, it would be preferable to examine inhibition kinetics at near-atmospheric CH₄ concentrations in a soil with no apparent endogenous CH₄ source.

A common shortcoming of NH₄⁺ inhibition studies, regardless of the organisms involved, has been a lack of attention to nonammoniacal salt effects despite numerous reports of substantial inhibition by such salts (1, 10, 15, 17, 24). King and Schnell (28) examined the effects of several Cl⁻ and SO₄²⁻ salts and concluded that nonammoniacal salts indirectly inhibit CH₄ oxidation by desorbing endogenous NH₄⁺ from cation exchange sites in the soil, which then directly inhibit CH₄ oxidation. Many N-limited soils, however, have extremely low concentrations of exchangeable NH₄⁺, yet are substantially inhibited by nonammoniacal salts (17), suggesting that these salts have NH₄⁺-independent effects on atmospheric CH₄ oxidizers. Additional mechanisms may alter inhibition kinetics, thus hindering the diagnosis of NH₄⁺-specific inhibition.

With the limitations described above in mind, we used a simple steady-state kinetics approach to assess the mechanism of NH₄⁺ inhibition of CH₄ oxidation at near-atmospheric concentrations (1.8 to 15 μl liter⁻¹) in three well-drained, N-limited forest soils that lack known endogenous CH₄ sources. In addition, we examined the effects of nonammoniacal salts in parallel samples to judge the utility of these salts as experimental controls for neutralizing NH₄⁺-independent salt effects.

MATERIALS AND METHODS

Field sites.

We studied soils from two temperate forests and one taiga forest, the major characteristics of which are listed in Table 1. The two temperate soils were from the Harvard Forest Long-Term Ecological Research site in western Massachusetts (29), where fertilizer inhibition of atmospheric CH₄ consumption was first observed (43). The sites and their biogeochemical cycles have been described in detail previously (7, 8, 30). The taiga site is approximately 120 years postburn, and the understory is dominated by Rosa acicularis and Equisetum spp. The mineral soil consists of a uniform layer of silty glacial loess. The pedology, ecology, and biogeochemistry of this site are similar to the pedology, ecology, and biogeochemistry of nearby sites that have been described previously (17). All three sites are well drained and have never been observed to produce CH₄ (7, 8, 19).

TABLE 1.

Sampling sites and characteristics of the soils examined in this study

Ecosystem type (dominant plant)	Location	Soil texture	Concn of extractable NH4+ (mg of N kg of dry soil−1)a	Water-holding capacity (g of H2O g of dry soil−1)
Birch taiga (Betula papyrifera)	University of Alaska, Fairbanks	Fine silt	0.11 (0.014)b	0.62
Temperate hardwood (Quercus velutina)	Harvard Forest, Petersham, Mass.	Sandy loam	6.98 (2.91)c	0.84
Temperate pine (Pinus resinosa)	Harvard Forest, Petersham, Mass.	Sandy loam	7.83 (2.49)c	0.73

^aconcentration of NH₄⁺ N in the upper 10 cm of the mineral horizon, which includes the zone of maximum CH₄ oxidation. The values in parentheses are standard deviations. 

^bData from this study. The concentration of extractable NH₄⁺ was determined as described previously (17). 

^cData from A. Magill (29a). See reference 30 for more information. 

Soil processing and bioassays.

All experiments were performed at the University of Alaska, Fairbanks. At each site, soil was collected in bulk from the upper 10 cm of the mineral soil, which included the zone of maximum CH₄ oxidation, and stored in perforated plastic bags for transport to the laboratory. The soil was homogenized by sieving it through a 4-mm-mesh screen. The water-holding capacity of each soil type was determined as described previously (18), and the moisture was adjusted so that the final water content was 30 to 35% of the water-holding capacity (Table 1) after the final treatment with deionized water or salt solutions. This moisture level was determined previously to be optimal for atmospheric CH₄ consumption in a wide variety of soils (18). Samples were treated with deionized water, K₂SO₄, (NH₄)₂SO₄, Na₂SO₄ (taiga soil only), KCl, or NH₄Cl. Each salt solution was added to a single bulk sample (0.1 ml g of dry soil⁻¹), which was then mixed thoroughly and subdivided into individual samples. The salts were added at a rate of 5.6 μmol of cations per g of dry soil, so that all of the salts were equinormal with respect to cations. The resulting NH₄⁺ additions were equivalent to 75 mg of N per kg of dry soil, which matched the treatments used in previous experiments (17). This amount of NH₄⁺ was intended to overwhelm the endogenous soil N (Table 1) yet remain within the range of soil NH₄⁺ concentrations reported for forest soils with various land use histories (13, 31, 42).

For each treatment, 12 subsamples (10 g of dry soil) were placed in 70-ml serum vials sealed with butyl rubber septa and allowed to equilibrate overnight. The following morning the vials were equilibrated with laboratory air (∼1.8 μl of CH₄ liter⁻¹) and sealed, and their headspace CH₄ concentrations were adjusted by injecting appropriate volumes of 1% CH₄ premixed with air (Scott Specialty Gases, Plumsteadville, Pa.); the headspace CH₄ concentrations tested were approximately 1.8 (no CH₄ added), 5, 10, and 15 μl liter⁻¹. Three replicates for each treatment at each CH₄ concentration were prepared. For the temperate soils, a single 2-h CH₄ oxidation assay was carried out with the headspace CH₄ concentration measured at the beginning and the end of the assay. The resulting consumption rate (d[CH₄]/dt) was paired with the corresponding midpoint CH₄ concentration in order to obtain a plot of oxidation rate versus CH₄ concentration (Fig. 1b and c). For the birch taiga soil, a modified procedure was used because oxidation was 2 orders of magnitude slower in this soil than in the temperate soils (Table 2). On the first day of the experiment, a 3.3-h assay was carried out with the headspace CH₄ concentration measured at the beginning and the end of the assay. The samples were kept sealed and were allowed to consume CH₄ overnight, and the 3.3-h assay was repeated on the second day and again on the third day. Identical assays were then repeated every 48 h until either a threshold concentration was established or through the ninth day, whichever occurred first. As with the temperate soils, the rate (d[CH₄]/dt) from each 3.3-h assay was plotted against the corresponding midpoint CH₄ concentration in order to obtain a plot of oxidation rate versus CH₄ concentration (Fig. 1a). CH₄ was analyzed by gas chromatography as described previously (5, 17, 18).

FIG. 1.

CH₄ oxidation kinetics in three upland forest soils, a birch taiga soil (a), a temperate hardwood soil (b), and a temperate pine soil (c). The data for the birch taiga soil are shown with linear regression lines, whereas the data for the temperate soils are shown with nonlinear regression curves fit to the Michaelis-Menten equation. Each point represents a single rate measurement.

TABLE 2.

CH₄ oxidation kinetics in three upland forest soils

Soil	Treatment	Km(app) (μl liter−1)a	Vmax(app) (pmol g−1 h−1)	First-order rate constant (h−1)b		% Inhibition compared toc:		Implied inhibition mechanism	Curve fit (R2)d
				True	Pseudo	H2O	K+
Birch taiga	H2O			0.00711					0.994
	K2SO4			0.00653		8.2		?	0.996
	(NH4)2SO4			0.00526		26*	19*	?	0.992
	KCl			0.00078		89*		?	0.772
	NH4Cl			0.00072		90*	7.7	?	0.771
Temperate hardwood	H2O	15.2 (1.3)	5,807 (329)		0.705				0.998
	K2SO4	9.8 (0.4)	2,662 (584)		0.510	28*		Uncompetitive	0.999
	(NH4)2SO4	22.4 (1.8)	2,870 (157)		0.235	67*	54*	Mixed competitive	0.998
	KCl	7.2 (0.6)	786 (30)		0.200	72*		Uncompetitive	0.993
	NH4Cl	25.6 (4.9)	1,174 (155)		0.085	88*	58*	Mixed competitive	0.991
Temperate pine	H2O	6.7 (0.9)	2,594 (160)		0.715				0.984
	K2SO4	6.1 (0.9)	1,591 (98)		0.485	32*		Noncompetitive	0.978
	(NH4)2SO4	9.8 (1.4)	1,699 (123)		0.320	55*	34*	Mixed competitive	0.986
	KCl	8.8 (4.0)	573 (123)		0.120	83*		Noncompetitive	0.855
	NH4Cl	10.9 (7.2)	515 (176)		0.085	88	29	Mixed competitive	0.769

^aThe values in parentheses are standard errors. 

^bTrue first-order rate constants were calculated by linear regression of CH₄ oxidation rates against midpoint CH₄ concentrations. Pseudo-first-order rate constants were calculated by determining V_max/K_m after K_m values were converted to picomoles of CH₄ per bottle. 

^cAn asterisk indicates that the treatment value was statistically different from the corresponding control value (P ≤ 0.05). 

^dThe regression coefficients are linear for the birch taiga soil and nonlinear (Michaelis-Menten curve fit) for the two temperate soils. 

Because Cl⁻ inhibited CH₄ oxidation much more than did SO₄²⁻, the effects of Cl⁻ and SO₄²⁻ salts on general microbial respiration in the birch taiga soil were examined. The amount of CO₂ that accumulated was determined by measuring headspace CO₂ concentrations, which never exceeded 2%, at the beginning and end of a 1-week incubation period. The amount of CO₂ that accumulated in each salt treatment was compared to the amount of CO₂ in water-treated controls. CO₂ was analyzed by gas chromatography as described previously (5, 18).

Statistical analyses and calculations.

The effects of the salt treatments and CH₄ concentration on oxidation rates in each soil were analyzed by analysis of covariance by using treatment as the independent factor and the initial CH₄ concentration as a covariate; Bonferroni contrasts were used in multiple comparisons. Because the incubation times were the same for all treatments in a given soil, the treatments with higher oxidation rates consumed more substrate than the treatments with lower oxidation rates. For regression analyses, therefore, the oxidation rate from an individual assay was paired with the corresponding midpoint CH₄ concentration (Fig. 1) rather than the initial concentration. This standard technique normalizes consumption rates for unequal substrate concentrations among treatments and also minimizes the deviation from standard Michaelis kinetics that can result from substrate depletion (41). First-order kinetics were modeled by linear regression, and the rate constants were estimated from the slope of the regression line, with both variables expressed in picomoles. Michaelis constants were obtained from a least-squares nonlinear regression fit of the data to the Michaelis-Menten equation. For treatments exhibiting Michaelis kinetics, pseudo-first-order rate constants were calculated as V_max/K_m, with both constants expressed in picomoles. Relative inhibition was calculated for each treatment as follows: relative inhibition = (1 − k₂/k₁) × 100, where k₁ is the first-order (or pseudo-first-order) rate constant for the control sample and k₂ is the first-order (or pseudo-first-order) rate constant for the treated sample.

Examining the relationship between relative inhibition and CH₄ concentration requires calculating inhibition ratios for two treatments at specific CH₄ concentrations. In doing this, care must be taken not to compare rates derived from substantially different midpoint CH₄ concentrations, even when the initial concentration is the same for all treatments. This problem arises when a faster sample consumes substantially more substrate than a slower sample, resulting in a disparity between midpoint CH₄ concentrations in the two assays. For this reason, the method used to calculate relative inhibition at specific CH₄ concentrations varied according to the relative rates among treatments and the type of kinetics involved for each soil. In the birch taiga soil, which displayed first-order kinetics, inhibition ratios were calculated directly from the rates measured in the experiments on the first day of incubation. Because the oxidation rates were very low in this soil, the differences in midpoint CH₄ concentrations among the treatments were trivial. The inhibition by each salt compared to the deionized water control was calculated for each of the initial CH₄ concentrations (∼1.8, 5, 10, and 15 μl liter⁻¹) and plotted against the midpoint CH₄ concentration occurring in the control (Fig. 2a). In the temperate soils, which displayed Michaelis kinetics, the rates were high, and the midpoint CH₄ concentrations varied among the treatments. Hence, estimated inhibition ratios were calculated by entering four different CH₄ concentrations (1.8, 5, 10, and 15 μl liter⁻¹) into the regression equation obtained for each treatment. The calculated oxidation rates at each concentration were then used to calculate inhibition ratios for each salt compared to the water control. Each ratio was then plotted against the CH₄ concentration from which it was derived (Fig. 2b and c). The slope of the relationship between relative inhibition and CH₄ concentration was estimated by linear regression.

FIG. 2.

Effect of CH₄ concentration on inhibition of CH₄ oxidation. General salt inhibition compared to water controls in the birch taiga soil (a) and in the temperate hardwood soil (b). (c) Specific NH₄⁺ inhibition compared to K⁺ controls in the temperate hardwood soil.

RESULTS

Birch taiga soil.

In the birch taiga soil, the CH₄ oxidation kinetics at concentrations of ≤15 μl liter⁻¹ were approximately first order (R² > 0.99 except for Cl⁻ salts) for all treatments, but the rate constants varied among treatments (Fig. 1a; Table 2). Analyses at higher CH₄ concentrations (10 to 800 μl liter⁻¹) (data not shown) yielded a K_m(app) for oxidation in this soil of 39 μl liter⁻¹ (57 nM in solution), which is typical for upland soils (2, 3, 36, 46, 47). Neither K₂SO₄ nor Na₂SO₄ significantly inhibited CH₄ oxidation compared to deionized water (P = 0.78), and the curves for K₂SO₄ and Na₂SO₄ were indistinguishable (P = 0.91) (Na₂SO₄ data not shown). Specific NH₄⁺ inhibition, calculated using K₂SO₄ as the control, was relatively weak (19%) but was statistically significant (P = 0.04). Both KCl and NH₄Cl inhibited CH₄ oxidation severely (∼90%; slopes were significantly different from zero at P < 0.01) (Fig. 1a). Unlike the comparison of K₂SO₄ and (NH₄)₂SO₄, the effects of KCl and NH₄Cl were indistinguishable (P = 0.84) (Fig. 1a). All four salts caused relative inhibition to increase as CH₄ concentration increased (Fig. 2a). The slopes of the increases were similar for all salts regardless of which cation was added and regardless of the final soil NH₄⁺ concentration. All salts inhibited total microbial respiration, but like CH₄ oxidation, CO₂ production was more sensitive to Cl⁻ salts than to SO₄²⁻ salts; the relative inhibition was ∼18% for both K₂SO₄ and (NH₄)₂SO₄, whereas it was 22 to 25% for KCl and NH₄Cl.

Temperate forest soils.

The relative inhibition patterns for the various salts in the temperate hardwood and pine forest soils were similar to the patterns in the birch taiga soil, except that CH₄ oxidation conformed well to Michaelis-Menten kinetics (R² > 0.98 in most cases) (Fig. 1b and c; Table 2). With minor differences in magnitude, the pine soil exhibited the same patterns as the hardwood soil. As in the birch taiga soil, the Cl⁻ salts were the most inhibitory salts, followed by (NH₄)₂SO₄ and then K₂SO₄. K₂SO₄ inhibition and specific NH₄⁺ inhibition (relative to K⁺) were stronger in the temperate soils than in the taiga soil; the levels of specific NH₄⁺ inhibition in the temperate hardwood and pine soils were 54 and 34%, respectively (Table 2). As in the birch taiga soil, inhibition of CH₄ oxidation increased with the CH₄ concentration when K⁺ salts were added. Unlike the taiga soil, however, inhibition in the temperate soils decreased as CH₄ concentration increased when NH₄⁺ salts were added (Fig. 2b) (pine forest results not shown). When specific NH₄⁺ inhibition was calculated using K⁺ salts as controls, inhibition decreased sharply from ∼50% in the presence of 1.8 μl of CH₄ liter⁻¹ to ∼20 to 30% in the presence of 15 μl of CH₄ liter⁻¹ in the temperate hardwood soil (Fig. 2c).

The Michaelis parameters K_m and V_max exhibited similar patterns of responses to the various treatments in the two temperate soils (Table 2). In deionized water controls, the K_m(app) values were 15 and 6.7 μl liter⁻¹ (22 and 9.8 nM in solution) in the hardwood and pine soils, respectively. In the hardwood soil, the values of both K_m and V_max were about double the corresponding values in the pine soil, so the pseudo-first-order rate constants (V_max/K_m) were similar in the two soils (Table 2). K⁺ salts (irrespective of the anions) either decreased or had no effect on K_m(app) values compared to water controls, whereas NH₄⁺ salts always increased the K_m(app). In contrast, V_max(app) values decreased similarly in the presence of both K⁺ and NH₄⁺ salts. Compared to K⁺ salts, however, NH₄⁺ salts increased K_m(app) but had no effect on V_max(app).

DISCUSSION

Often, soil CH₄ oxidation at near-atmospheric CH₄ concentrations follows first-order reaction kinetics (3, 39, 46), as was the case in the birch taiga soil in this study (Fig. 1a). In fine-textured soils, first-order kinetics at lower CH₄ concentrations may result from restricted gas diffusion from the atmosphere into the soil, a purely first-order process (14, 37, 45). The birch taiga soil studied is a fine silt soil and therefore strongly limits gas diffusion from the atmosphere to the CH₄ oxidizers (14, 37), thus possibly increasing the K_m(app) and creating a problem for studying low-concentration CH₄ oxidation kinetics in this soil (Fig. 3). By contrast, the two temperate forest soils studied have a coarse sandy texture, which enhances CH₄ diffusion, allowing uptake kinetics to reflect enzyme activity more closely and permitting standard kinetic analyses of NH₄⁺ and salt inhibition of CH₄ oxidation at near-atmospheric CH₄ concentrations. The maximum CH₄ concentration used in our experiments (15 μl liter⁻¹) was similar to the K_m(app) values in the temperate forest soils. The regression curves resulting from the kinetic analyses provided very good fits to the actual data (generally, R² > 0.98) (Fig. 1; Table 2), indicating that the kinetic models which we used accurately described the process as measured in this study.

FIG. 3.

Potential effect of soil texture on CH₄ oxidation kinetics in soil. Because gas transport by diffusion is purely first order, a fine-textured soil may exhibit first-order kinetics at lower CH₄ concentrations, whereas a coarse-textured soil containing the same CH₄-oxidizing enzyme or a similar enzyme may exhibit Michaelis-Menten kinetics at the same concentrations.

Although there have been numerous reports of steady-state kinetic constants for soil CH₄ oxidation (2–4, 15, 36, 46, 47), only one previous study reported kinetic parameters for NH₄⁺ inhibition (15). The soil studied previously was an agricultural humisol with an organic matter content of ∼70% and demonstrable methanogenic activity (15, 16). These conditions probably supported a much different CH₄ oxidizer community than the community expected in upland mineral soils that lack an endogenous CH₄ source. Indeed, the K_m(app) values for our temperate forest soils (Table 2) were substantially lower than the K_m(app) values reported in the previous study (15), suggesting that there were physiological differences in the CH₄ oxidizer communities in the upland mineral soils and the agricultural humisol. In the present study we focused specifically on high-affinity CH₄ oxidation in three non-CH₄-producing upland soils from two North American biomes, subarctic taiga forest and northeastern temperate forest.

Salt effects.

Interpreting inhibition mechanisms based on kinetic parameters in a system that is as biologically and chemically complex as soil requires careful consideration of how ions added to the system may affect the process of interest, both directly and indirectly (15, 28). All of the ions used in this study potentially could affect CH₄ oxidation in three basic ways. First, they could change the soil osmotic potential and impose water stress on the microbial community (18, 40); second, they could affect ion exchange, thereby altering NH₄⁺ availability (28); and third, they could affect the CH₄ oxidizers directly in a number of ways (11, 15, 21). Any of these factors could alter CH₄ oxidation rates and kinetics and thus affect the interpretation of the specific NH₄⁺ inhibition mechanism.

With the salt additions used in this study, the water potential in the birch taiga soil was approximately −0.2 MPa, which is the optimum water potential for atmospheric CH₄ oxidation in a wide variety of upland soils (18, 40). Because the birch taiga soil had the lowest water-holding capacity of the soils used in this study (i.e., the lowest water/salt ratio [Table 1]), its osmotic potential should have been the most sensitive to the salt additions. Thus, salt-related inhibition of atmospheric CH₄ oxidation in this study did not appear to be related to water stress.

It is unlikely that K⁺ salts indirectly produced the inhibition observed in this study by desorbing NH₄⁺ from cation exchange sites, as proposed elsewhere (28). This mechanism requires that an untreated soil contain sufficient exchangeable NH₄⁺ to account for the inhibition observed with nonammoniacal salts, yet the soils we studied had very low concentrations of exchangeable NH₄⁺ relative to our NH₄⁺ additions (Table 1). Cl⁻ salts consistently inhibit soil CH₄ consumption to a far greater extent than do SO₄²⁻ salts (28; this study). King and Schnell (28) attributed this phenomenon to greater NH₄⁺ adsorption to cation exchange sites in the presence of SO₄²⁻ than in the presence of Cl⁻. In the present study, however, it was impossible for the KCl treatments to produce free NH₄⁺ concentrations approaching those of the (NH₄)₂SO₄ treatments, because we added 1 to 3 orders of magnitude more NH₄⁺ than was potentially available in the untreated soils (Table 1). Even so, KCl inhibition was far greater than (NH₄)₂SO₄ inhibition in all three soils (Fig. 1; Table 2). KCl and NH₄Cl produced similar levels of inhibition in each of the soils, despite the fact that the NH₄Cl treatments necessarily resulted in much higher free NH₄⁺ concentrations. Similarly, the results of King and Schnell show that NaCl caused inhibition equal to or greater than the inhibition that equinormal NH₄Cl caused in another temperate forest soil (28). Again, this result could not have been dependent on NH₄⁺ concentrations. Hence, desorption of endogenous NH₄⁺ cannot account for the extremely inhibitory effects of Cl⁻ salts in a variety of soils, and it is clear that Cl⁻ salts should be avoided in NH₄⁺ inhibition studies, unless it can be demonstrated that Cl⁻ is not toxic to CH₄ oxidizers in a particular soil.

Unlike KCl, K₂SO₄ inhibited CH₄ oxidation less than (NH₄)₂SO₄ inhibited CH₄ oxidation, raising the possibility that there is indirect inhibition by cation exchange when SO₄²⁻ salts are used. Compared to water, K₂SO₄ inhibition was 32 to 58% of (NH₄)₂SO₄ inhibition in the three soils (Table 2). Assuming that the desorption of endogenous soil NH₄⁺ was 100%, which is unlikely, the K₂SO₄ treatments would have produced maximum NH₄⁺ concentrations that were between ∼0.1 and 10% of the amount added in the (NH₄)₂SO₄ treatment (Table 1). Hence, compared to the (NH₄)₂SO₄ treatments, the ratios of relative inhibition to potential NH₄⁺ concentration obtained with the K₂SO₄ treatments seem unlikely. More importantly, as discussed below, steady-state kinetic parameters indicate that K₂SO₄ and (NH₄)₂SO₄ inhibited CH₄ oxidation via different physiological mechanisms, which would not be the case if K⁺ acted indirectly via NH₄⁺ desorption.

The ubiquity of NH₄⁺-independent salt effects and the variety of salts that induce similar responses suggest that a fundamental physiological process is involved. Roslev et al. (36) found that high-affinity CH₄ oxidizers in a temperate forest soil efficiently incorporated ¹⁴CH₄-C into biomass. Adding NH₄Cl to the soil not only decreased CH₄ oxidation rates but also reduced the C assimilation efficiency and dramatically increased the proportion of ¹⁴C oxidized to CO₂. It is impossible to know whether this response was to NH₄⁺ or Cl⁻ or both, as the experiments did not include parallel salt controls. However, because we found that Cl⁻ overwhelmingly dominated NH₄Cl inhibition in all of our soils, the response that Roslev et al. observed may also have been predominantly due to Cl⁻. Killham (25) reported that NaCl additions had the same effect on microbial assimilation and respiration of [¹⁴C]glucose in soil and found that an increase in the ratio of respired C to assimilated C was a sensitive index of physiological stress within a soil heterotroph community. Shifts in the ratio were attributed to an increase in the maintenance energy required for the cells to cope with the imposed stress. If active transport of ions out of the cell or some other energy-intensive coping strategy were required by energy-limited CH₄ oxidizers exposed to a salt, then cellular reductant might be diverted to this process, making less reductant available for growth and potentially to the CH₄-oxidizing enzyme, thus decreasing the CH₄ oxidation rates. This scenario is plausible for an extremely energy-limited population and is reconcilable with the inhibition kinetics reported here, as diverting reductant away from the CH₄-oxidizing enzymes should reduce the catalytic efficiency of the extant enzyme pool, thereby potentially altering K_m(app) and V_max(app) as described below. Gulledge et al. (17) observed an apparent growth response of the atmospheric CH₄ oxidizer community in samples obtained from depths of 20 to 40 cm in another forest soil. The in situ CH₄ concentrations at depths below 20 cm were chronically <0.5 μl liter⁻¹. After 14 days of exposure to ambient atmospheric CH₄ in the laboratory, the CH₄ consumption rates in water-treated samples increased severalfold compared to the rates measured after only 5 days of exposure. In K₂SO₄-treated samples a less pronounced increase occurred, and in (NH₄)₂SO₄-treated samples no increase occurred, suggesting that the effects of NH₄⁺ and salt were synergistic. These results also are consistent with a cellular stress response by an energy-limited population and may illustrate why atmospheric CH₄ oxidizers have a limited capacity to recover from soil fertilization (32, 33, 39).

If salts generally inhibit soil CH₄ oxidation by an NH₄⁺-independent mechanism, then it seems appropriate to quantify specific NH₄⁺ inhibition based on a parallel salt control rather than a deionized water control. This approach has been challenged by the view that other cations may have unique inhibition mechanisms that make them ineffective as experimental controls (28). Although this hypothesis is plausible, no differential toxicity of potential control cations, such as Na⁺ and K⁺, has been reported for soil CH₄ oxidation. King and Schnell (28) found that KCl inhibited CH₄ uptake by pure cultures of Methylosinus trichosporium more than did NaCl, but they observed no difference in soil CH₄ consumption in the presence of these two salts. Similarly, we observed no difference in the effects of K₂SO₄ and Na₂SO₄ in the birch taiga soil in the present study (the temperate soils were not tested with Na⁺). Moreover, it is equally plausible that similar cations, such as NH₄⁺, K⁺, and Na⁺, exert equivalent nonspecific effects that, in conjunction with counteranion effects, account for the nonammoniacal inhibition observed with salts in general. Since K⁺ and Na⁺ salts inhibit soil CH₄ oxidation to the same extent (28; this study), this hypothesis appears to be sound. Our view, therefore, is that parallel salt controls must be employed when NH₄⁺ inhibition is examined, because there is no other way to account for the nonammoniacal effects that salts clearly have on soil CH₄ consumption. In some cases, salt effects can be substantial compared to specific NH₄⁺ inhibition and therefore probably interfere with kinetic analysis of the NH₄⁺ inhibition mechanism. In the present study we used both deionized water and nonammoniacal salt controls in order to examine the relative efficacies of the two approaches for elucidating the mechanism of NH₄⁺ inhibition.

Specific NH₄⁺ inhibition.

Determining the physiological mechanism of specific, immediate NH₄⁺ inhibition has proven to be difficult (6, 15, 17, 26–28). Dunfield and Knowles (15) demonstrated that NH₄⁺ inhibited CH₄ oxidation by enzymatic substrate competition in an agricultural humisol assayed at high CH₄ concentrations. The kinetics varied between samples, however, indicating that an additional mechanism may have been involved. King and Schnell (27, 39) examined NH₄Cl inhibition at low CH₄ concentrations and found that relative inhibition increased with CH₄ concentration. They concluded that this phenomenon resulted from the fortuitous oxidation of NH₄⁺ to toxic NO₂⁻ or NH₂OH, which in turn reduced the activity of the methanotroph population (39). They did not examine Michaelis constants or comparable kinetic parameters. We observed similar increases in inhibition with increasing CH₄ concentrations in all three of the soils we examined. In the taiga soil, this phenomenon occurred with nonammoniacal salts as well as NH₄⁺ salts, and the slope of the increase was not affected by the NH₄⁺ concentration (Fig. 2a). In the temperate soils, K⁺ salts caused inhibition to increase, whereas NH₄⁺ salts caused inhibition to decrease as the CH₄ concentration increased (Fig. 2b and c). The same pattern occurred whether Cl⁻ or SO₄²⁻ salts were added, indicating that it was not specific to a particular counterion (Fig. 2a and b). These results indicate that the increase in inhibition did not result from NH₄⁺ or its by-products. Thus, although NO₂⁻ undoubtedly inhibits atmospheric CH₄ oxidation when it is added directly to soil (19, 26), an increase in NH₄⁺ salt inhibition when the CH₄ concentration increases more likely results from a general salt effect than from by-products of fortuitous NH₄⁺ oxidation.

A net increase in inhibition with an increase in the CH₄ concentration in response to NH₄⁺ salts may actually indicate that specific NH₄⁺ inhibition is weak or absent. For instance, in the birch taiga soil, in which NH₄⁺ inhibition was relatively weak (Table 2), both NH₄⁺ and K⁺ salts caused similar increases in inhibition as the CH₄ concentration increased (Fig. 2a). However, in the temperate hardwood soil, in which NH₄⁺ inhibition was relatively strong (Table 2), only K⁺ salts caused inhibition to increase, whereas NH₄⁺ salts caused inhibition to decrease as the concentration of CH₄ increased (Fig. 2b). Specific NH₄⁺ inhibition, isolated by using K⁺ salts as controls, declined precipitously as the CH₄ concentration increased (Fig. 2c). Hence, salts generally caused increases in inhibition, whereas NH₄⁺ caused decreases in inhibition as the CH₄ concentration increased, indicating that there are separate inhibition mechanisms for NH₄⁺ specifically and salts generally. In our soils, the relative strengths of these two mechanisms were apparent from the slopes of the plots of (NH₄)₂SO₄ inhibition (relative to deionized water) versus CH₄ concentration; a positive slope indicated a stronger salt effect, whereas a negative slope indicated a stronger NH₄⁺ effect.

The soils which we examined were relatively acidic (pH ∼3.5 to 4.5). Because NH₃, rather than NH₄⁺, is probably the competitive inhibitor of CH₄ oxidation, salt effects may be more prevalent in acidic soils, whereas competitive inhibition may be relatively more important in neutral to alkaline soils, such as the agricultural humisol investigated by Dunfield and Knowles (15). Despite the intuitive appeal of this hypothesis, there is no obvious relationship between pH and the degree of NH₄⁺ inhibition in soils with different pH values, suggesting that other cross-site variables are generally more important (17). Moreover, it is clear from the results obtained with the temperate hardwood soil, in which NH₄⁺ accounted for 58% of the total inhibition, that NH₄⁺ inhibition can be dominant in acidic soils. Perhaps the intracellular pH, which should be near neutral regardless of the soil pH, is the relevant control on NH₃/NH₄⁺ ratios at the enzyme level.

Compared to the K_m in the water controls, the K_m(app) either decreased (hardwood soil) or remained unchanged (pine soil) when K⁺ salts were added, but it always increased when NH₄⁺ salts were added (Table 2). Again, this pattern supports the hypothesis that there are different inhibition mechanisms for NH₄⁺ and salts in general and also eliminates the possibility that K⁺ salts acted indirectly by desorbing soil-bound NH₄⁺ into solution, as concluded previously for another temperate forest soil (28). If K⁺ ions acted indirectly via NH₄⁺, then K⁺ and NH₄⁺ salts should have produced similar inhibition kinetics, yet they had different effects on K_m(app) compared to deionized water. In contrast to K_m(app), V_max(app) decreased in response to both K⁺ and NH₄⁺ salts (Table 2). Thus, the kinetic constants suggest that there is a partial mixed-type inhibition for NH₄⁺ salts, with both a competitive component (increasing K_m) and a noncompetitive or uncompetitive component (decreasing V_max) that is independent of NH₄⁺ (41) (Table 2). If general salt effects account for the additional inhibition, then using K₂SO₄ as a control rather than deionized water should isolate the specific NH₄⁺ effect. Indeed, compared to K⁺ salts, NH₄⁺ salts caused K_m(app) to increase substantially, whereas they had no effect on V_max(app) (Table 2). The Cl⁻-salt pair produced the same kinetic pattern as the SO₄²⁻ pair despite the greater inhibition by Cl⁻. These consistent results strongly indicate that NH₄⁺ inhibited atmospheric CH₄ oxidation in the two temperate forest soils via simple enzyme substrate competition.

Summary and conclusions.

Our approach using K⁺ salts as controls, and the resulting interpretation, provided a plausible NH₄⁺ inhibition mechanism that is consistent with the data presented here and can also account for the contrasting results of previous studies (15, 27, 39). Whereas Dunfield and Knowles (15) observed competitive inhibition kinetics in their agricultural humisol, Schnell and King observed increasing inhibition with increasing CH₄ concentrations in a temperate forest soil, a result that, by itself, is inconsistent with competitive inhibition. Both phenomena occurred simultaneously in our temperate forest soils and could be explained by a mixed- type inhibition resulting from at least two independent mechanisms, enzymatic substrate competition by NH₄⁺ and one or more noncompetitive or uncompetitive mechanisms common to salts in general. Although the inhibition mechanism in the birch taiga soil could not be determined directly because it displayed first-order kinetics (Fig. 1a), the relative inhibition pattern for the various treatments was consistent with the patterns obtained with the two temperate soils, so that all three soils may have shared the same mechanisms. Moreover, it is notable that despite very different soil characteristics, we found essentially the same inhibition mechanism that Dunfield and Knowles (15) found in an agricultural humisol. This convergence of physiological responses in ecologically diverse environments suggests that enzymatic substrate competition is an important NH₄⁺ inhibition mechanism in a wide variety of soils.

Although the results readily explain immediate inhibition of soil CH₄ oxidation, delayed inhibition, which has been observed in both field and laboratory studies (17), remains enigmatic. Delayed inhibition probably results from shrinkage of the CH₄ oxidizer population over time rather than from decreases in the specific activities of individual CH₄ oxidizers (17). NH₄⁺ and salt effects may act synergistically to impose whole-cell stress that increases maintenance energy requirements, thereby diverting reductant from growth, even if sufficient reductant for the CH₄-oxidizing enzyme remains available. This scenario might diminish a population’s ability to replace dying biomass, yet might not slow the oxidation of CH₄ until the population begins to shrink, resulting in a delayed inhibition response (17). Hence, multiple physiological mechanisms may contribute synergistically to both immediate and delayed NH₄⁺ fertilizer inhibition of atmospheric CH₄ consumption in soil. Moreover, nonammoniacal salts in the environment, especially KCl and NaCl (both of which are used heavily in agriculture and industry), may be as problematic as NH₄⁺ fertilizers for soil CH₄ consumption.