Method for Detection of Microorganisms That Produce - Applied and ...

Vol. 51, No. 4

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Apr. 1986, p. 776-780

0099-2240186/040776-05$02.00/0 Copyright © 1986, American Society for Microbiology

Method for Detection of Microorganisms That Produce Gaseous Nitrogen Oxides GARY E. JENNEMAN, ANNE D. MONTGOMERY, AND MICHAEL J. McINERNEY* Department of Botany and Microbiology, University of Oklahoma, Norman, Oklahoma 73019 Received 26 August 1985/Accepted 8 January 1986 A method was developed to detect NO- or N20-producing bacteria in solid or liquid medium by their ability to oxidize the redox indicator resazurin from its reduced colorless form to its oxidized pink form. The method was sensitive to as little as 35 nM N20 or 0.5 nM NO. Ninety-one percent of the colonies that oxidized resazurin on plates also produced N20 in slant cultures. Forty-four percent of the colonies that did not oxidize resazurin did produce N20. This percentage was reduced to 15% when colonies in which the coloration was difficult to discern were picked to slants to determine whether they oxidized the slant. The production of N20 preceded the oxidation of resazurin by liquid cultures of Escherichia coli and a sludge isolate. With the denitrifying sewage isolate, the disappearance of N20 was followed by the return of resazurin to its reduced state. WolineUa succinogenes was found to produce small amounts of N20 from N03 , which resulted in a transient oxidation of resazurin.

Nitric oxide (NO) and nitrous oxide (N20) are generally recognized as intermediates in the biological reduction of NO3- to N2 (14). Denitrification is an agriculturally important process, since it leads to the loss of combined nitrogen from soil. Also, N20 has been implicated in chemical reactions involving the loss of ozone in the stratosphere (7). In addition to its production by denitrifiers, N20 is also produced by nitrifying bacteria (24) and some assimilatory nitrate-reducing bacteria, algae, yeasts, and molds (3, 16, 21). Methods for enumerating denitrifiers are based on mostprobable-number methods (8) for detecting visible gas and an increase in pH (20), depletion of NO3- and N02- (10), or depletion of NO3- with the concomitant production of N20 in cultures containing acetylene (26). The inherent problems in enumerating denitrifiers has led to the recommendation that a measure of denitrifying enzyme activity be used in preference to most-probable-number enumerations as an index of the denitrifying activity or catalytic potential of a soil (19). The isolation of denitrifiers is often time consuming, since there is no direct method for detecting denitrifiers on agar medium. We report a method for detecting nitrous oxide-producing bacteria on solid or liquid media based on the ability of these bacteria to oxidize the redox indicator resazurin from its reduced colorless state to its oxidized pink form. This method may be useful as a qualitative indicator of N20 production when pure cultures are used or for the selective isolation of certain N20-producing bacteria from environmental samples.

was obtained from the University of Oklahoma stock culture collection. Wolinella succinogenes VS, Desulfovibrio desulfuricans DD, and Desulfovibrio strain Gll were kindly provided by M. P. Bryant, University of Illinois, Urbana. Media and conditions of cultivation. The basal medium used for the cultivation and detection of bacteria producing gaseous nitrogen oxides was as follows (all in grams per liter): Na2SO4, 2.84; KH2PO4, 0.68; MgCl2 6H20, 0.41; NH4Cl, 0.32; CaCl2 * 2H20, 0.09; and MOPS [3-(Nmorpholine)propanesulfonic acid], 2.09. The pH was adjusted to 7.0 with 10 N NaOH. Resazurin was added at a final concentration of 0.0001% (wt/vol), and cysteine-sulfide reducing solution (13) was added at a final concentration of 2.0% (vol/vol). Trace metals and vitamin solutions described by Balch et al. (1) were added separately at 0.1% (vol/vol) and 0.5% (vol/vol) final concentration, respectively. Solid media contained 1.5% (wt/vol) Bacto-Agar (Difco Laboratories, Detroit, Mich.). Glucose (5 mM) or sodium acetate (15 mM) was used as the carbon and energy source as indicated. NaNO3 was added at the indicated concentrations. Methods for the preparation and use of anaerobic media were those of Hungate as modified by Balch et al. (1) and Bryant (5). The medium was boiled under a stream of 02-free nitrogen (that had been passed over a heated copper column) (5) and dispensed in 10-ml aliquots into anaerobic culture tubes (18 by 150 mm; Bellco Glass, Inc., Vineland, N.J.) that had been purged with 02-free nitrogen; the tubes were then stoppered with black rubber stoppers. Slants were prepared in culture tubes (13 by 100 mm) that were stoppered with black rubber stoppers (no. 00; Bellco.). The medium was autoclaved for 20 min at 121°C in an aluminum press and cooled to approximately 60°C. The tubes containing sterile reduced medium were then placed inside the anaerobic glove box and allowed to cool. Medium used for pour plates was kept at 50°C until used. Environmental samples were 10-fold serially diluted in 10 mM MOPS (pH 7.0) inside the anaerobic chamber. The diluent was sterilized by autoclaving and allowed to preincubate overnight in the anaerobic chamber to reduce the 02 concentration. After the dilutions were made, 0.1 or 1 ml of the appropriate dilution was placed inside a sterile plastic petri dish (also preincubated overnight in the

MATERIALS AND METHODS Sample collection and strains. Anaerobic sewage sludge was collected from a secondary anaerobic digestor at the Norman Municipal Sewage Plant, Norman, Okla., and was immediately brought to the laboratory, where it was stored overnight at room temperature in an anaerobic chamber (Coy Laboratory Products, Ann Arbor, Mich.) under an atmosphere of N2 containing 2 to 10% H2. Escherichia coli 8739 *

Corresponding author. 776

VOL. 51, 1986

anaerobic chamber) and a tube of the molten agar medium was poured into the petri dish and mixed with the dilution. Polyethylene tape was then wrapped around the sides of the petri dishes to reduce water evaporation, and the plates were incubated inside the chamber at the ambient temperature. Slant cultures were also incubated inside the chamber. Strain 4A3 was isolated from anaerobic sewage sludge by using the acetate basal medium. It was maintained on deep agar cultures of this same medium. Strain 4A3 was then subcultured on RF5B medium, which was the basal medium with the following additions: Casamino acids (Difco), 0.005 g/liter; yeast extract (Difco), 0.005 g/liter; NaNO3, 2.0 g/liter; and 5% (vol/vol) clarified rumen fluid. The RF5B medium was dispensed under a 100% gaseous N2 atmosphere into screw-top Hungate tubes (screw-top tubes [11 by 125 mm] fitted with butyl rubber stoppers contained within a plastic screw cap [Bellco]). E. coli was grown in nutrient broth (Difco) containing 0.2% (wt/vol) NaNO3, 0.0001% (wt/vol) resazurin, and 1.0 mM dithiothreitol. The medium was anaerobically prepared and dispensed into screw-top Hungate tubes as described above. W. succinogenes was subcultured at 35°C on the medium as described by Wolin et al. (23), under an atmosphere of 100% N2. A 1-ml inoculum of an early-stationaryphase culture was transferred to 9.0 ml of the same medium without fumarate but with 0.2% NaNO3, 0.0001% resazurin, and 2% (vol/vol) cysteine-sulfide reducing solution. The medium was anaerobically prepared and dispensed as described above. D. desulfuricans and Desulfovibrio strain Gll were grown as described by Mclnerney and Bryant (12) and Mclnerney et al. (13), respectively. A 1.0-ml inoculum from these cultures was transferred to the nitrate-containing medium described above for W. succinogenes, except that 0.1% dl-lactate replaced sodium formate as the electron donor. Detection of gas producers. Culture tubes containing nitrate broth (Difco) and Durham tubes were autoclaved and then preincubated in the anaerobic chamber overnight to reduce the level of 02- Pink and colorless colonies were picked from 3-week-old colonies contained on plates of the glucose basal medium with either 6, 20, or 59 mM NO3 and inoculated into tubes of nitrate broth. The nitrate broth tubes were examined for growth and gas production after 3 weeks of incubation. Nitrate. Nitrate was determined colorimetrically (11) by using the Szechrome NAS reagent (Polysciences, Inc., Warrington, Pa.), and nitrate standards obtained from Hach Chemical Co., Ames, Iowa, were used to prepare a calibration curve. Gas analysis and titration. Gas analysis was performed on a gas chromatograph (no. 427; Packard Instrument Co., Inc., Rockville, Md.) equipped with a Porapak Q column (Supelco, Inc., Bellefonte, Pa.) and a thermal conductivity detector; nitrogen was used as the carrier gas (flow rate, 30 ml per min). The oven, detector, and injector temperatures were 70, 90, and 90°C, respectively. Peak identification was made by comparing peak retention times with external standards prepared from high-purity gases. Nitrous oxide production during the growth of liquid cultures was detected by using a gas chromatograph (no. 3400; Varian Associates, Palo Alto, Calif.) equipped with a Porapak Q column and a 63Ni electron capture detector. The column and injector temperatures were 55°C; the detector temperature was 300°C. The carrier gas was 5% CH4-95% Ar at a flow rate of 15 ml/min. Samples (5,ul) were taken by using a gas-tight syringe. The pressure of each tube was

DETECTION OF N20 PRODUCERS

777

measured with a pressure transducer (Validyne Corp.). The minimum detectable concentration of N20 was 12.1 Pa when a 5-,u sample was used. The concentration of N20 in the liquid phase was calculated by using the amount of N20 in the gas phase and published values for the Ostwald coefficient (22). For gas titrations, N20 (98% purity; Alltech Associates, Inc., State College, Pa.) and NO (23.6 ,ul/liter in nitrogen; Alltech) were transferred into 160-ml serum bottles with butyl rubber stoppers by evacuating and pressurizing the serum bottle with the respective gas three successive times. The gas was then added via a gas-tight syringe to another stoppered 160-ml serum bottle containing 100 ml of the reduced glucose medium. This gas transfer step was performed inside the anaerobic chamber to eliminate 02 contamination. Each gas (NO or N20) was added until a faint pink endpoint occurred. The gas was equilibrated with the aqueous phase by shaking. The faint pink coloration was visible for several days before being bleached by the reductant. RESULTS Agar medium containing more than 6 mM NaNO3 and inoculated with diluted sewage sludge had colonies with a pink, diffusible coloration after 1 to 3 weeks of incubation inside the anaerobic chamber. This coloration occurred only around a colony and did not occur in medium lacking nitrate or on uninoculated plates incubated under identical conditions. The pink, diffusible coloration around colonies was observed in basal agar medium with 15 mM sodium acetate, 5 mM glucose, or no carbon source added and containing 6 to 59 mM NaNO3. The pink coloration was more distinct at NaNO3 concentrations greater than 36 mM. Medium containing 12 or 24 mM NaNO3 had about the same number of colonies with the diffusible pink coloration, but these colonies were not easily detected unless a clean white background was used. Medium containing less than 12 mM NaNO3 had very few colonies with the diffusible pink coloration. The presence of the pink coloration surrounding a colony suggested that a product of nitrate metabolism was oxidizing resazurin from its reduced (colorless) state to its oxidized (pink) state. It was found that the addition of 38,umol of N20 and 1.0 nmol of NO was required to oxidize 100 ml of the reduced basal medium to a faintly pink endpoint. Published values for the Ostwald coefficients for N20 and NO (22) were used to calculate the concentration of these gases in the liquid phase as about 0.2 mM and 0.5 nM, respectively. Published equilibrium redox potentials for NO and N20 at pH 7.0 (Eo', +1,175 mV for NO/N20 and +1,355 mV for N2O/N2) (18) indicate that they are good oxidizing agents in comparison with nitrate (Eo', +433 mV). Therefore, N20 and NO are capable of oxidizing resazurin (Eo', -42 mV [pH 6.87]) from its reduced (colorless) form to its oxidized (pink) form in the absence of atmospheric oxygen. Uninoculated reduced medium with 59 mM NaNO3 remained colorless for 2 months or more. The addition of 150 mM NaNO2 to reduced medium did not result in the appearance of the pink coloration; instead, the medium turned yellow after being autoclaved (121°C, 15 min). Prolonged incubation of the autoclaved reduced medium containing 150 mM NaNO2 did result in an oxidation of the resazurin indicator. Nitrite is known to undergo chemical denitrification, yielding products such as N20, NO, N2, and NO2 (4, 6). These data show that the addition of NO or N20, but not N03 or N02 , caused a rapid oxidation of the reduced medium.

778

APPL. ENVIRON. MICROBIOL.

JENNEMAN ET AL.

x

o0

1.0

E

c

C

0 0

I

/ -xm

w

0 C:

CDu) -0Ct

0 U,

0 U)

-D

1.0 0

10

20

30

40

50

60

510

Time (hours)

0

x

0.

10

B

0.01

I

0

40

I

I0 80 120 160 200 240 280

Time (hours)

FIG. 1. Growth and nitrous oxide production by E. coli 8739 (A) and strain 4A3 (B). The cross-hatched the redox indicator resazurin was in its pink (oxidized) state.

Experiments with pure cultures of bacteria showed that the oxidation of resazurin followed the production of N20 by the organism. E. coli cells grown in reduced medium with resazurin and nitrate produced a pink, diffusible coloration when the concentration of N20 reached 34.8 nM (Fig. 1A). The culture of strain 4A3 oxidized resazurin when the N20 concentration reached about 0.26 ,uM (Fig. 1B). E. coli was grown in medium with dithiothreitol rather than cytsteine,uM sulfide as the reductant. This is probably why only about 35 nM N20 was required to oxidize resazurin in the E. coli cultures, since dithiothreitol does not generate as low a redox potential as does cysteine-sulfide. These data did show that N20 production was required before resazurin was oxidized. The pink coloration first developed at the bottom of the tube. Oxygen contamination would result in the development of a pink coloration at the gas-liquid interface at the top of the tube. Strain 4A3 produced N20 and oxidized resazurin during the exponential phase of growth. The pink coloration formed during N20 accumulation by strain 4A3 disappeared (i.e., resazurin returned to the reduced state) as the N20 concentration rapidly declined toward the late-exponential phase of growth. The medium was oxidized by E. coli cultures during the late-stationary phase of growth. The medium remained pink for at least 3 weeks, and the N20 concentration declined more slowly. The fact that the medium remained oxidized after the N20 concentration declined in E. coli cultures may be due to a generalized oxidation of the medium as a result of prolonged incubation and repeated puncturing of the septum during sampling. Bleakley and Tiedje (3) found that various nondenitrifying N20-producing bacteria produced N20 only in stationary phase. Desulfovibrio strain Gil reduced NO3to NH4+ without the production of N20 and did not oxidize the resazurin. About 12% of the initial NO3 was used by

z

area refers to the time at

which

strain Gil in the 2-week incubation period. D. desulfuricans DD grew, but did not reduce any of the added NO3-, and no change was noted in the color of the redox indicator. W. succinogenes did oxidize the growth medium while growing on formate as an energy donor and nitrate as electron acceptor. The pink coloration was noticeable for less than 12 h before resazurin changed back to its colorless reduced state. About 12 Pa of N20 was detected in the gas phase of W. succinogenes during the period when resazurin was oxidized. This is the first report of the production of N20 by W. succinogenes, although Yoshinari (25) reported the reduction of N20 by W. succinogenes. Experiments were then conducted to determine whether colonies that oxidized resazurin also produced N20. Colonies with and without the diffusible pink coloration were picked from glucose basal agar medium plates that contained 6, 20, or 59 mM NO3- and that were inoculated with diluted sewage sludge. Each colony was inoculated into a tube which contained Difco nitrate broth and a Durham tube to determine whether the colonies with diffusible pink coloration produced gas. Fourteen colonies with diffusible pink coloration did not grow in nitrate broth. Of the remaining 35 pink colonies that grew, 74% produced visible amounts of gas (Table 1). The majority of those that did not produce gas (seven of nine) were picked from plates containing a low nitrate concentration (6 mM). The pink coloration is difficult to visualize at low nitrate levels, and thus it is difficult to distinguish colorless colonies from those with diffusible pink coloration. Of the 49 colorless colonies that were picked, 29 produced visible turbidity in nitrate broth within 3 weeks of incubation. None of these colorless colonies produced gas (Table 1). Therefore, the production of a diffusible pink coloration corresponds well to the ability to produce gas, as would be expected if these were colonies of nitrous oxideproducing bacteria. The presence of some false-positives

DETECTION OF N20 PRODUCERS

VOL. 51, 1986

TABLE 1. Gas production by pink and colorless colonies growing anaerobically in nitrate broth No. of tubes withb: Pink colonies

Nitrate concn

(mM)a

Gas +

growth 59 20 6

3 15 8

3 9 2

1 1 7

colonies were analyzed in this manner, the percentage of false-negatives decreased to about 15%.

Colorless colonies

No gas No growth growthc growthc

Gas + growthd

0 0 0

No gas g

No growth

10 10 9

6 7 7

~~~growth

779

a Nitrate cqncentration of the glucose basal agar medium from which the colonies were picked. b Each tube was inoculated with a different colony. c False-positives. d False-negatives.

may be due to the production of too little N20 or N2 to be detected with Durham tubes. Because of the relatively high solubility of N20 compared with N2, a larger amount of N20 would have to be produced to yield visible gas. Colonies with and without the diffusible pink coloration were picked to slants, and after 2 weeks of incubation the gas phase was analyzed for the presence of N20. It should be noted that NO was not detectable under the gaschromatographic conditions used, but as little as 0.1% N20 in the gas phase was detected by using a thermal conductivity detector. Of the colonies with a diffusible pink coloration, 79% produced detectable levels of N20 (Table 2). An additional 11% of these colonies produced large amounts of a gas that was not H2, CO2, N20, or CH4 and was presumed to be N2. These cultures also oxidized the slant. Thus, 91% of the colonies that were picked from primary isolation medium and that were presumed to produce N20 or to be denitrifiers did produce N20 or N2. However, a large percentage (44%) of the colorless colonies produced N20 or presumably N2 in slants. It was often difficult to distinguish white colonies from those with a faint pink coloration. This was particularly true for crowded plates containing 100 or more colonies. However, when the colony in question was picked to a slant, it was possible to distinguish whether the colony produced N20 by determining whether the slant was oxidized. A great majority of these false-negatives (22 of 28) oxidized the slant (Table 2). All of these colonies produced N20 in slant culture or were presumed to be denitrifiers as a result of the large amounts of gas produced (see above). When questionable

DISCUSSION Our results show that the oxidation of the medium was associated with the accumulation and persistence of N20 and possibly NO. The addition of low levels of either N20 or NO, but not NO3- or NO2-, oxidized the medium. The rereduction of the medium was associated with a decrease in N20 levels (Fig. 1B). Pure cultures of bacteria known to produce N20 (E. coli strains and strain 4A3) also oxidized the medium, while strict anaerobes such as Desulfovibrio spp., which are not known to produce N20, did not oxidize the medium. Nitric oxide was not analyzed, and so it is not known whether this gas was produced and whether it contributed to the oxidation of the medium. Denitrifiers gain energy for growth from the reduction of N20, but dissimilatory nitrate users such as E. coli do not conserve energy through N20 reduction. Instead, N20 is believed to be produced by the reduction of N02 via nitrate reductase in E. coli (15). Therefore, one would expect N20 to accumulate after growth has ceased (Fig. 1B). It also makes sense that strict anaerobes, such as Desulfovibrio spp., which reduce nitrate to NH4' do not accumulate N20, since this would result in an increased redox potential, which could be inhibitory for growth. This method could prove useful in differentiating isolates that denitrify or produce N20 from those that reduce NO3- to NH4+ without N20 production. This method may be particularly useful for the selection of denitrification mutants. This method was useful in showing that W. succinogenes, an organism not previously known to produce N20, transiently produced low levels of N20. This method was tested to determine its effectiveness in directly isolating either NO- or N20-producing organisms. A great majority (91%) of the bacteria that produced pink colonies on the agar medium used for initial isolation also produced N20 in slant cultures (Table 2) if the isolation medium contained at least 20 mM N03 . The pink coloration was difficult to visualize at lower N03 levels, and the number of false-positives increased (Table 1). It was shown previously that NO3- (2, 9), and especially NO, can inhibit the reduction of N20 at high NO3 concentrations, and this could account for the accumulation of N20 at high NO3 concentrations. Also, the presence of free sulfide (used as a reductant in the medium) may contribute to N20 accumulation, since free sulfide is known to inhibit N20 reductase in some denitrifiers (17).

TABLE 2. Nitrous oxide production by sewage sludge isolates No. (%) of coloniesb producing: Colony color"

Pink White all pink slants white slants

N20

Gas but not N2Or

42/53 (79)

6/56 (11)

26/63 (41) 20/22 (85) 6/41 (15)

2/63 (3) 2/22 (15) 0/41 (0)

a Colonies with or without the diffusable pink coloration were randomly picked from plates of acetate basal medium containing 59 mM N03 and inoculated into slants of the same medium. After 2 weeks of incubation, the gas phase of each slant was analyzed and each slant was checked for coloration and the presence of gas splits in the agar. b Number of positives divided by the total number of colonies picked. All the colonies picked grew in the slants. c N20 and CH4 were not detected. H2 and C02 were present, but at levels similar to other slants without gas. Each slant contained gas pressure as indicated by the displacement of the syringe plunger.

One of the problems in applying this method to the analysis of natural populations was the large number of false-negatives (44%) that were found (Table 2). On plates with 100 or more colonies, it was difficult to distinguish between faintly pink colonies and noncolored colonies. Inoculating these colonies into slants and determining whether the slant became oxidized allowed us to distinguish between these possibilities in some cases and resulted in fewer false-negatives (15%) (Table 2). However, bacteria that transiently produce N20 or produce low levels of N20 would probably not be detected in most cases, particularly if the plates are viewed after long incubation times. Viewing the plates on a more frequent (daily) basis allows one to

N20 NO- or N20-

detect some of the colonies that transiently produce

(unpublished observations).

When this method was used to enumerate producing bacteria in sewage sludge, we found (1.5 + 0.28) x 106 and (3.5 + 0.7) x 106 CFU/ml of sludge when 15 mM

780

APPL. ENVIRON. MICROBIOL.

JENNEMAN ET AL.

acetate or 5 mM glucose was the electron donor, respectively. This method was also used in liquid cultures to determine the most-probable number of H2-using, N20producing bacteria in sludge as 7.4 x 104 per ml of sludge with 95% confidence intervals of 1.7 x 104 to 31 x 104 per ml of sludge. However, it is not known whether the above values for N20-producing bacteria are truly representative of the level of denitrifiers in sewage sludge, since direct comparisons with other methods for the enumeration of denitrifiers were not made. Thus, the use of this method as a quantitative measure of denitrifying populations in an ecosystem should be approached with caution until- factors which affect the detection of the N20- or NO-producing bacteria, such as the redox potential and redox buffer strength of the medium and the rates of N20 or NO production and consumption, are studied in greater detail. However, this method is a convenient way to select for certain, although not necessarily representative, N20-producing bacteria from natural samples. ACKNOWLEDGMENTS This work was supported by U.S. Department of Energy contracts no. DE-AC19-80BC10300 and DE-AS05-83ER1-3053 and by the Energy Resources Institute of the University of Oklahoma. LITERATURE CITED 1. Balch, W. E., G. E. Fox, L. J. Magrum, C. R. Woese, and R. S. Wolfe. 1979. Methanogens: re-evaluation of a unique biological group. Microbiol. Rev. 43:260-296. 2. Blackmer, .A. M., and J. M. Bremner. 1978. Inhibitory effect of nitrate on reduction of N20 to N2 by soil microorganisms. Soil Biol. Biochem. 10:187-191. 3. Bleakley, B. H., and J. M. Tiedje. 1982. Nitrous oxide production by organisms other than nitrifiers and denitrifiers. Appl. Environ. Microbiol. 44:1342-1348. 4. Bollag, J. M., S. Drzymala, and L. T. Kardos. 1973. Biological versus chemical nitrite decomposition in soil. Soil Sci. 116:44-50. 5. Bryant, M. P. 1972. Commentary on the Hungate technique for culture of anaerobic bacteria. Am. J. Clin. Nutr. 25:1324-1328. 6. Bulla, L. A., C. M. Gilmour, and W. B. Boilen. 1970. Nonbiological reduction of nitrite in soil. Nature (London) 225:664. 7. Crutzen, P. J. 1976. Upper limits of stratospheric ozone reductions following increased application of fixed nitrogen to the soil. Geophys. Res. Lett. 3:169-172. 8. Davidson, E. A., M. K. Strand, and L. F. Galloway. 1985. Evaluation of the most probable number metHod for enumeration of denitrifying bacteria. Soil Sci. Soc. Am. J. 49:642-645. 9. Firestone, M. K., M. S. Smith, R. B. Firestone, and J. M. Tiedje. 1979. The influence of nitrate, nitrite, and oxygen on the

10. 11.

12.

13. 14.

15. 16.

17.

18. 19. 20. 21. 22. 23.

24. 25. 26.

composition of the gaseous products of denitrification in soil. Soil Sci. Soc. Am. J. 43:1140-1144. Focht, D. D., and H. Joseph. 1973. An improved method for the enumeration of denitrifyirig bacteria. Soil Sci. Soc. Am. Proc. 37:698-699. Hanson, R. S., and J. A. Philips. 1981. Chemical composition, p. 350-356. In P. Gerhart, R. G. E. Murray, R. N. Costilow, E. W. Nester, W. A. Wood, N. R. Krieg, and G. B. Phillips (ed.), Manual of methods for general microbiology. American Society for Microbiology, Washington, D.C. Mclnerney, M. J., and M. P. Bryant. 1981. Anaerobic degradation of lactate by syntrophic associations of Methanosarcina barkeri and Desulfovibrio species and effect of H2 on acetate degradation. Appl. Environ. Microbiol. 41:346-354. Mclnerney, M. J., M. P. Bryant, and N. Pfennig. 1979. Anaerobic bacterium that degrades fatty acids in syntrophic association with methanogens. Arch. Microbiol. 122:129-135. Payne, W. J. 1981. The status of nitric oxide and nitrous oxide as intermediates in denitrification, p. 85-103. In C. C. Delwiche (ed.), Denitrification, nitrification, and atmospheric nitrous oxide. John Wiley & Sons, Inc., New York. Smith, M. S. 1983. Nitrous oxide production by Escherichia coli is correlated with nitrate reductase activity. Appl. Environ. Microbiol. 45:1545-1547. Smith, M. S., and K. Zimmerman. 1981. Nitrous oxide production by nondenitrifying soil nitrate reducers. Soil Sci. Soc. Am. J. 45:865-871. Sorensen, J., J. M. Tiedje, and R. E. Firestone. 1980. Inhibition by sulfide of nitric and nitrous oxide reduction by denitrifying Pseudomonas fluorescens. Appl. Environ. Microbiol. 39: 105-108. Thauer, R. K., .K. Jungermann, and K. Decker. 1977. Energy conservation in chemnotrophic anaerobic bacteria. Bacteriol. Rev. 41:100-18Q. Tiedje, J. M. 1983. Denitrification. Agronomy 9:1011-1026. Valera, C. L., and M. Alexander. 1961. Nutrition and physiology of denitrifying bacteria. Plant Soil 154268-280. Weathers, P. J,I 1984: N20 evolution by green algae. Appl. Environ. Microbiol. 48:1251-1253. Wilhelm, E., R. Battino, and R. J. Wilcock. 1977. Low-pressure solubility of gases in liquid water. Chem. Rev. 77:225-231. Wolin, M. J., E. A. WoWib, and N. J. Jacobs. 1961. Cytochromeproducing anaerobic vibrio, Vibrio succinogenes, sp. n. J. Bacteriol. 81:911-917. Yoshida, T., and M. Alexander. 1970. Nitrous oxide formation by Nitrosomonas europaea and heterotrophic microorganisms. Soil Sci. Soc. Am. Proc. 34:880-882. Yoshinari, T. 1980. N20 reduction by Vibrio succinogenes. Appl. Environ. Microbiol. 39:81-84. Yoshinari, T., and R. Knowles. 1976. Acetylene inhibition of nitrous oxide reduction by denitrifying bacteria. Biochem. Biophys. Res. Commun. 69:705-710.