Degradation of Dead Microbial Biomass in a Marine Sediment

Transcription

1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Sept. 1986, p /86/ $02.00/0 Copyright 1986, American Society for Microbiology Vol. 52, No. 3 Degradation of Dead Microbial Biomass in a Marine Sediment JAMES A. NOVITSKY Biology Department, Dalhousie University, Halifax, Nova Scotia B3H 4JI, Canada Received 3 March 1986/Accepted 23 May 1986 The availability of dead microbial biomass in a marine beach sand to degradation and mineraliation was examined. Microbial sand populations were labeled with [14C]glutamic acid, [3H]adenine, or [3H]thymidine and killed with chloroform. Live sand or seawater (or both) was added to the sterile labeled sand, and biochemical components of the populations were monitored for 10 days. Labeled RNA was degraded more quickly than labeled DNA, but both nucleic acids were degraded to approximately the same extent (60 to 70%). 3H20 was a major acid-soluble breakdown product. RNA (and possibly DNA) breakdown products were reincorporated into DNA (and possibly RNA) during the incubation period. In addition to metabolite salvage, 32% of the total macromolecular 14C was respired in the 10-day period regardless of whether sand or seawater was used as the inoculum. Respiration was essentially complete in 3 days, whereas nucleic acid degradation continued throughout the 10-day incubation. The results indicate that dead microbial biomass is a labile component of the sediment ecosystem. In seawater, and particularly in marine sediments, large numbers of bacterial cells can be observed microscopically, yet few of these bacteria can be cultured (5, 7), indicating that large portions of the population are active but not culturable, dormant, or dead. A number of techniques have been developed to demonstrate the metabolic activity, rather than the culturability, of aquatic bacteria. These methods include sensitivity to nalidixic acid to detect synthetically active bacteria (17), reduction of tetraolium salts to detect electron transport system activity (29), and the use of autoradiography (22) to detect cells actively taking up specific substrates. Using a combination of these methods, Tabor and Neihof (28) were able to demonstrate activity for >85% of the bacterial population of Chesapeake Bay. Unfortunately, a majority of the population in most other environments is inactive. Meyer-Reil (22), using autoradiography, found an average of only 31% of the microbial population of the Kiel Fjord and Bight active with respect to glucose uptake. Zimmermann et al. (29), using the electron transport system method, found no greater than 12% of coastal Baltic Sea microbial populations actively respiring. Using autoradiography I (25) found <10% of the population active in a coastal marine sediment. I have previously discussed several explanations for these observations (25) and would now like to consider one of these possibilities: that the nonactive cells are truly dead. If so, have the bodies that we observe under the microscope persisted for a long period of time, thus making the sediments a bacterial graveyard, or are dead microbes degraded quickly? In addition to whole cells, do cellular components likewise persist for long periods of time? Relatively large amounts of extracellular DNA or RNA or both have been measured in marine samples (26), and a few studies (20, 21, 23) have suggested that specific microbial components have short turnover times, but data on the turnover of most cellular macromolecules are scarce. Of particular interest to me was the stability of DNA and RNA, especially recently synthesied nucleic acids as measured with the [3H]adenine and [3H]thymidine incorporation methods for the determination of microbial carbon production and specific growth rates. The present study was initiated to provide data on the extent and rate of killed biomass degradation as well as the stability of RNA and DNA in marine sediments. If it can be 504 determined that dead microbial cells are rapidly broken down or mineralied or both, the graveyard hypothesis becomes less tenable. I hope that an estimate of the longevity of dead microbial biomass compared with the specific growth rate of the population will eventually lead to a good estimate of the magnitude of the viable, but nongrowing, population in marine sediments. MATERIALS AND METHODS Sand preparation. Sand and water were collected in clean plastic containers from the surf one (water depth, approximately 0.5 m) of the beach at Waimea Bay, Oahu, Hawaii. Approximately 10 cm3 of wet sand and 15 ml of filtersterilied seawater were amended with either 10,uCi of [U-'4C]glutamic acid (specific activity, 294 mci/mmol), 100,uCi of [2-3H]adenine (specific activity, 12.8 Ci/mmol), or 100,iCi of [methyl-3h]thymidine (specific activity, 77.2 Ci/mmol) and incubated with occasional gentle shaking at the in situ temperature of 25 C. After 6 h (glutamic acid) or 2 h (adenine and thymidine), the overlying water was carefully decanted and replaced with 40 ml of filtered seawater containing 400,ul of chloroform. After 2 h of gentle intermittent shaking, the supernatant was decanted, and the sand was gently washed four times with 10 ml of chloroformsaturated, filtered seawater. After the final wash the sand was dried in vacuo and stored in a desiccator until used. Gamma irradiation. Labeled or unlabeled sand was sterilied by exposure to gamma radiation (total dose, 2.5 megarads; approximately 10 h of exposure). Plastic centrifuge tubes containing 10 to 50 g (wet) of sand were capped and lowered into a Mark IV 'Co gamma irradiator. Upon retrieval the sand was aseptically dried in vacuo. Radioactive sand (prepared as described above) was irradiated after the replacement of the radioactive supernatant with 40 ml of filtered seawater (chloroform omitted). After irradiation the samples were aseptically washed and dried as described above. Experimental procedure. Labeled sand (0.25 g dry), freshly collected live sand (0.5 g wet), and 5 ml of seawater were added to enough 50-ml serum bottles so that two bottles could be sacrificed for each determination at each time of sampling. For samples inoculated with seawater only, the live sand was replaced with nonlabeled gamma-irradiated

2 VOL. 52, 1986 DEAD MICROBIAL BIOMASS DEGRADATION IN MARINE SEDIMENT ' o 120 cn V) % 100 L 80 Ft) o aiv I a I 0 ) TIME (days) FIG. 1. Fate of dead 14C-labeled biomass after inoculation of a chloroform-treated sediment sample. Sediment microbial populations were initially labeled with ['4C]glutamic acid and then killed with chloroform. Labeled sterile sand was then inoculated with fresh live sand. Symbols: 0, CO2; 0, acid-soluble metabolites; l, acid-insoluble biomass. I sterile sand. Controls consisted of labeled and gammairradiated nonlabeled sand suspended in filter-sterilied seawater or chloroform-saturated seawater. All bottles were capped with rubber stoppers and incubated in the dark at 25 C with intermittent shaking. "4C determinations. For the determination of 14CO2 production, the serum bottle stopper was replaced with a stopper fitted with a cup containing a filter paper wick saturated with 0.15 ml of,-phenethylamine. One-half ml of 2 N HCl was injected into the bottle; after 20 min the wick was removed, and radioactivity was determined by liquid scintillation counting. The bottles were then placed on ice, and 14.5 ml of cold 2 N HCl was added to dissolve the sediment and precipitate the macromolecules. After 1 h, two 10-ml samples were filtered onto Millipore Corp. type HA filters and washed twice with 5 ml of cold 1 N HCl. The filters were dried overnight at 60 C and then oxidied in an Oxymat model JAlOl sample oxidier in preparation for liquid scintillation counting. The radioactivity in two 1-ml portions of the acid-precipitated filtrate was also counted to determine the amount of acid-soluble metabolites present. 3H determinations. The reactions were terminated, the sediment was dissolved, and the macromolecules were precipitated by the addition of 10 ml of cold 2 N HCI followed by 5 ml of cold 1 N HCl. After 1 h on ice, two 5-ml portions of the mixtures were filtered onto Whatman GF/F glass fiber filters. Each filter was washed twice with 5 ml of cold 1 N HCI and then froen. A 500-pd sample of filtrate was assayed directly for radioactivity, and the 3H20 portion of the filtrate was determined by the method of Karl (15). The [3H]RNA and [3H]DNA collected on the filters were purified and separated by the method of Karl (15). Calculations. All calculated values are expressed per gram of dry labeled sand. Each point is the mean of two determinations from two individual incubation bottles. When sample sie permitted (see above), each determination consisted of the mean of duplicate subsamples. For the calculation of the percent carbon respired, the mean from 10 individual bottles was determined. The percent respired was calculated in two different ways: by dividing the 14CO2 collected by the sum of the '4CO2 collected and the macromolecular 14C and by dividing the loss of macromolecular "4C by the original 14C-labeled biomass. RESULTS At the onset of the present investigation, I assumed that gamma irradiation would be a nondestructive method for steriliation that would cause minimal chemical and physical changes in the sediment. A dose of 2.5 megarads was needed to completely sterilie the sediment used. In preliminary experiments this dosage not only sterilied the sediment but also almost completely destroyed the ATP, RNA, and DNA present. It can be assumed that other macromolecules were similarly affected. For these reasons chloroform fumigation was chosen for use in this study. Unlike gamma irradiation, chloroform treatment did not alter the amount of DNA, RNA, or total acid-precipitable macromolecules present. Because of these results I have assumed that such treatment did not cause major chemical changes in the sediment, but minor changes cannot be ruled out. The degradation of total dead microbial carbon is shown in Fig. 1. Macromolecular carbon was stoichiometrically converted to CO2 without the accumulation of acid-soluble, low-molecular-weight metabolites. The percentage of the biomass carbon that was respired during the 14-day incubation was calculated in two ways for both the sand-inoculated and the water-inoculated experiments. However, there is little difference among the various methods, with an overall average of 31% respired. The fate of DNA and RNA differed slightly, depending on the radioactive precursor used to label the nucleic acids. When [3H]thymidine was used, both DNA and RNA were labeled (Fig. 2) and subsequently degraded. Since the amount of radioactivity in the fractions containing both DNA and RNA accounted for greater than 99% of that measured for the total acid-insoluble fraction, the radioactivity contained in other macromolecules such as protein

3 506 NOVITSKY DAY 0 DNA(5.8%) DAY 14 DNA (1.4%) RNA (59.9%) (19.3%) kter (28.2%) FIG. 2. Inventory of 3H before and after inoculation of a chloroform-treated sediment sample. Sediment microbial populations were initially labeled with [3H]thymidine and then killed with chloroform. Labeled sterile sand was then inoculated with fresh live sand and examined after 14 days. The initial specific activity was 482,000 dpm/g of dry sand, and greater than 97% of the radioactivity was recovered on both sampling days. was assumed to be negligible. Although RNA was initially degraded more rapidly than DNA, both were significantly metabolied during the 14-day incubation (RNA, 62%; DNA, 72%). The products of this metabolism were acid APPL. ENVIRON. MICROBIOL. crsoluble low-molecular-weight components, of which 3H20 comprised 35.6% (Fig. 2). When [3H]adenine was used as a precursor, RNA was degraded, but DNA was observed to increase threefold (Fig. 3). The results were similar when unlabeled adenine (final concentration, 5 pm) was available, but when unlabeled RNA (bakers' yeast, type XI; final concentration, 75,ug/ml) was added to the incubation mixture, both RNA and DNA were degraded (Fig. 4). The results with the [3H]adeninelabeled sand plus unlabeled RNA were similar to the results obtained with the [3H]thymidine-labeled sand: initial rapid RNA degradation with extensive nucleic acid metabolism after 10 days (RNA reduced by 72%, DNA reduced by 53%). The [3H]adenine-labeled DNA and RNA were also degraded to acid-soluble low-molecular-weight metabolites; however, unlike the [3H]thymidine-labeled RNA and DNA experiment, over 93% of the breakdown products was 3H20 (data not shown). Over the 10-day incubation period sterile test samples (fumigated unlabeled sand plus filter-sterilied seawater) were tested for sterility by injecting [14C]glutamic acid and measuring the production of 14CO2. The results indicated that it was difficult to keep sterile controls sterile for periods in excess of several days, probably because the fumigation technique is not 100% effective. For this reason, chloroformsaturated seawater was used routinely to maintain sterility. In all control experiments, no abiotic degradation of macromolecules, DNA, or RNA was noted. Likewise there was no measurable abiotic production of 14CO2 or 3H20 (data not shown). However, since the effect of chloroform on enymatic activity is not known, abiotic degradation cannot be completely ruled out. DISCUSSION Various methods have been used to sterilie soils for the determination of microbial biomass. These methods include air drying, autoclaving, gamma irradiation, and chloroform or methyl bromide fumigation (8-10, 12, 27). Of these methods, chloroform fumigation has emerged as the pre- 170 ~0 U) c V c0 N co ) C TIME (days) FIG. 3. Fate of dead [3H]RNA and [3H]DNA after inoculation of a chloroform-treated sediment sample. Sediment microbial populations were initially labeled with [3H]adenine and then killed with chloroform. Labeled sterile sand was then inoculated with fresh live sand. Symbols: 0, DNA; *, RNA.

4 VOL. 52, 1986 DEAD MICROBIAL BIOMASS DEGRADATION IN MARINE SEDIMENT 507 Z Z ' -o 50- I WT) (a ) 40Jc,_4 mo~~~~~~~~~~ <i0 0 4 TIME (days) FIG. 4. Fate of dead [3H]RNA and [3H]DNA after inoculation of a chloroform-treated sediment sample. Sediment microbial populations were initially labeled with [3H]adenine or [3H]thymidine and then killed with chloroform. Labeled sterile sand was then inoculated with fresh live sand. Unlabeled RNA was added to the [3H]adenine-labeled sand incubations to reduce the reincorporation of labeled metabolites into RNA or DNA. Symbols: 0 and 0, [3H]adenine-labeled sand; O and *, [3H]thymidine-labeled sand; 0 and O, RNA; 0 and *, DNA. ferred method, partly becausc of its simplicity and partly due to a report (2) showing that chloroform did not alter microbial cells or metabolites in any way that made them more or less susceptible to subsequent microbial degradation. In the present study chloroform fumigation proved to be an adequate method for steriliing sediment. The integrity of the microbial cells after chloroform treatment, however, is not known. Due to the acceptance of the chloroform method, other potential methods have not been rigorously examined. It can only be assumed that microbial biomass killed in a specific way will be representative of natural microbial death. The stability of DNA and RNA in microbial populations is of interest due to the use of these molecules as measures of microbial growth rates and biomass production in the marine environment (6, 13, 15, 24). It is interesting to note that [3H]thymidine extensively labeled RNA in these populations (Fig. 2). The thymidine incorporation method for the determination of biomass prodution in these sediments, therefore, would be invalid unless the [3H]RNA was chemically separated from the [3H]DNA. For this study the labeling of both RNA and DNA by both precursors was advantageous because each experiment confirmed the initial rapid degradation of RNA. DNA in both cases was degraded at a rate from 5 to 16% of the initial [3H]DNA per day. Extrapolating the slower rate ([3H]thymidine-labeled DNA), all of the DNA would be degraded in 19.6 days. This figure is greater than the estimate of Paul (DNA turnover on the order of 24 h or less for seawater; J. Paul, personal communication) but is in general agreement with the DNA turnover time of 10 to 20 days calculated by Baelyan and Ayatullin (4) for surface ocean waters. Also comparable are the data of Maeda and Taga (21) who observed that 75% of the DNA added to marine sediments (6 mg of DNA per g wet weight) degraded in 10 days. Initial experiments with [3H]adenine-labeled cells indicated that the amount of [3H]DNA actually increased (Fig. 3). Assuming that the nucleic acid base(s) produced by RNA degradation was being recycled into DNA, experiments were conducted with unlabeled adenine or RNA added to the incubation mixture. Unlabeled adenine had no effect but unlabeled RNA eliminated the [3H]DNA increase (Fig. 4), probably by reducing the specific activity of the RNA and subsequently the RNA breakdown product pool. These results indicate that the breakdown intermediates are recycled (salvaged) into DNA (and probably RNA), thus conserving energy the cells would otherwise expend with de novo synthesis. Why added unlabeled adenine had no effect remains unknown. This recycling of metabolites indicates that any degradation-mineraliation calculations made in this manner are minimum estimates unless the reuse of lowmolecular-weight metabolites is measured. The recycling of [3H]thymidine-labeled intermediates was not noted, either because the methods used were insensitive or because the [3H]thymidine-labeled products were metabolied in a different manner. Since the same methods were used for both precursors, the latter hypothesis appears more plausible. Although the results are similar for the breakdown of either [3H]adenine- or [3H]thymidine-labeled nucleic acids, one major difference was the production of a large amount of acid-soluble, nonwater metabolites (Fig. 2) from the [3H]thymidine-labeled macromolecules. The exact composition of these low-molecular-weight compounds is unknown. It is possible that in the short course of the experiment these compounds represent a precursor pool awaiting utiliation rather than an accumulation of end products. Obviously there are differences between the labeling patterns of the two precursors. Since the specific activity of the initial 3Hlabeled macromolecules was not determined, it is not known (but it is likely) that the molecules were labeled to different extents by the precursors used. This, in addition to the recycling observed with the [3H]adenine-labeled populations, could be a major reason for the observed results. Observing the fate of the 14C-labeled population, it is clear that at least a portion of microbial biomass is readily mineralied to CO2 after the death of the microbial cells. The rapid mineraliation appears to be complete within 3 to 7 days. Undoubtedly, carbon continues to be mineralied past 7 days, but at a much slower rate. The calculated 10-day respiration figure of 31%, therefore, must be viewed as a

5 508 NOVITSKY measure of the labile components of the microbial population. It is interesting to note the absence of any buildup of acid-soluble intermediates (dissolved organic carbon). Apparently the microbes are efficient in either immediately utiliing or respiring low-molecular-weight metabolites that result from the degradation of macromolecules. I was concerned that some of the CO2 generated during the 10-day incubation might leak out of the incubation bottles, so the percent respired was calculated in two ways. First, the 1'4CO2 collected was divided by the total initial '4C-labeled biomass. Second, the loss of "4C-labeled acid-insoluble material was divided by the total biomass. Both produced similar results for 10 samples: with the first method, and % respired when water and sand were used as the inoculum, respectively; with the second method, ± 6.12 and % respired, respectively. However, there was greater variation in the measurements of acid-insoluble material. This loss of precision in the second method offset any accuracy gained by not having to account for the escaped CO2. A chloroform fumigation method similar to the one described here has been used by soil microbiologists to determine the amount of microbial biomass in soils (3, 8-12, 19, 27). The soil is sterilied with chloroform and then reinoculated with a small amount of fresh soil. The CO2 subsequently evolved is then measured and converted to microbial biomass by using a conversion factor that represents the microbial mineraliation efficiency. This constant has been determined by adding 14C-labeled cells to soil and measuring their breakdown; it can vary from 30 to 50% (2, 8), depending on the source of the biomass. Since approximately 33% of bacterial carbon is mineralied compared with 44% of fungal carbon, a "biomass-weighted" figure of 41% is generally accepted (2). The value calculated in this study (31%) is on the low side of the range observed for soils and may represent the scarcity of fungi in marine sediments. Since my value represents the mixed, in situ population, no correction is necessary for the different groups of microbes present. Can this method be used to measure the microbial biomass in marine samples as suggested by Karl (D. M. Karl, in E. R. Leadbetter and J. S. Poindexter, ed., Bacteria in Nature: A Treatise on the Interactions of Bacteria and their Habitats, in press)? In principle there seems to be no reason against its use; however, methods (14; S. Y. Newell, R. D. Fallon, and P. S. Tabor, in Leadbetter and Poindexter, ed., in press) currently exist that are easier, faster, and probably no less accurate than the fumigation method. Similar results were obtained when sterile labeled sand was reinoculated with either sand or water, indicating that the sand and water populations are similar or that each has the equivalent capability of biomass degradation. Although both possibilities are plausible, the latter is supported by the results of Maeda and Taga (21), who found DNAhydrolying bacteria common and widely distributed in both seawater and sediment. Although no abiotic biomass degradation was observed in the present study, a number of researchers (1, 4, 18, 20) have demonstrated the activity of DNA-hydrolying enymes in seawater and sediment, thus raising the possibility of abiotic biomass degradation. The environmental importance of the results presented here is that approximately 30% of the biomass carbon and significant portions of specific cellular components are degraded quickly after cell death. King and White (16) and Moriarty (23) both conclude that the bacterial cell wall component, muramic acid, does not persist in the environment for long. King and White (16) determined the 50% APPL. ENVIRON. MICROBIOL. turnover of muramic acid to be as short as 3.2 h. These findings lend support to the hypothesis that the large number of microbial cells observed in marine sediments represents active or dormant, but not dead, cells. ACKNOWLEDGMENTS I am grateful to Jean Novitsky for excellent field and technical assistance and to D. Karl for critical discussions and ideas concerning this work. I thank D. Karl and the Department of Oceanography, University of Hawaii, for their facilities and cooperation that made this research possible during my sabbatical leave there. This research was funded by grants A-6548 and T-1925 from the National Research Council of Canada. LITERATURE CITED 1. Aardema, B. W., M. G. Loren, and W. E. Krumbein Protection of sediment-adsorbed transforming DNA against enymatic inactivation. AppI. Environ. Microbiol. 46: Anderson, J. P. E., and K. H. Domsch Mineraliation of bacteria and fungi in chloroform-fumigated soils. Soil Biol. Biochem. 10: Anderson, J. P. E., and K. H. Domsch A physiological method for the quantitative mesurement of microbial biomass in soils. Soil Biol. Biochem. 10: Baelyan, V. L., and T. A. Ayatullin Kinetics of enymatic hydrolysis of DNA in seawater. Oceanology 19: Colwell, R. R., P. R. Brayton, D. J. Grimes, D. B. Rosak, S. A. Huq, and L. M. Palmer Viable but non-culturable Vibrio cholerae and related pathogens in the environment: implications for release of genetically engineered microorganisms. Biotechnology 3: Fuhrman, J. A., and F. Aam Thymidine incorporation as a measure of heterotrophic bacterioplankton production in marine surface waters: evaluation and field results. Mar. Biol. 66: Jannasch, H. W., and G. E. Jones Bacterial populations in seawater as determined by different methods of enumeration. Limnol. Oceanogr. 4: Jenkinson, D. S The effects of biocidal treatments on metabolism in soil. IV. The decomposition of fumigated organisms in soil. Soil Biol. Biochem. 8: Jenkinson, D. S., and D. S. Powlson The effects of biocidal treatments on metabolism in soil. 1. Fumigation with chloroform. Soil Biol. Biochem. 8: Jenkinson, D. S., and D. S. Powlson The effects of biocidal treatments on metabolism in soil. V. A method for measuring soil biomass. Soil Biol. Biochem. 8: Jenkinson, D. S., and D. S. Powlson Measurement of microbial biomass in intact soil cores and in sieved soil. Soil Biol. Biochem. 12: Jenkinson, D. S., D. S. Powlson, and R. W. M. Wedderburn The effects of biocidal treatments on metabolism in soil. III. The relationship between soil biovolume, measured by optical microscopy, and the flush of decomposition caused by fumigation. Soil Biol. Biochem. 8: Karl, D. M Measurement of microbial activity and growth in the ocean by rates of stable ribonucleic acid synthesis. Appl. Environ. Microbiol. 38: Karl, D. M Cellular nucleotide measurements and applications in microbial ecology. Microbiol. Rev. 44: Karl, D. M Simultaneous rates of ribonucleic acid and deoxyribonucleic acid synthesis for estimating growth and cell division of aquatic microbial communities. Appl. Environ. Microbiol. 42: King, J. D., and D. C. White Muramic acid as a measure of microbial biomass in estuarine and marine samples. Appl. Environ. Microbiol. 33: Kogure, K., U. Simidu, and N. Taga A tentative direct microscopic method for counting living marine bacteria. Can. J. Microbiol. 25: Loren, M. G., B. W. Aardema, and W. E. Krumbein Interaction of marine sediment with DNA and DNA availability

6 VOL. 52, 1986 DEAD MICROBIAL BIOMASS DEGRADATION IN MARINE SEDIMENT 509 to nucleases. Mar. Biol. 64: Lynch, J. M., and L. M. Panting Measurement of the microbial biomass in intact cores of soil. Microb. Ecol. 7: Maeda, M., and N. Taga Deoxyribonuclease activity in seawater and sediment. Mar. Biol. 20: Maeda, M., and N. Taga Occurrence and distribution of deoxyribonucleic acid-hydrolying bacteria in sea water. J. Exp. Mar. Biol. Ecol. 14: Meyer-Reil, L.-A Autoradiography and epifluorescence microscopy combined for the determination of number and spectrum of actively metaboliing bacteria in natural waters. Appl. Environ. Microbiol. 36: Moriarty, D. J. W Improved method using muramic acid to estimate biomass of bacteria in sediments. Oecologia 26: Moriarty, D. J. W., and P. C. Pollard DNA synthesis as a measure of bacterial growth rates in seagrass sediments. Mar. Ecol. Prog. Ser. 5: Novitsky, J. A Heterotrophic activity throughout a vertical profile of seawater and sediment in Halifax Harbor, Canada. Appl. Environ. Microbiol. 45: Novitsky, J. A., and D. M. Karl Influence of deep ocean sewage outfalls on the microbial activity of the surrounding sediment. Appl. Environ. Microbiol. 50: Powlson, D. S., and D. S. Jenkinson The effects of biocidal treatments on metabolism in soil. II. Gamma irradiation, autoclaving, air-drying and fumigation. Soil Biol. Biochem. 8: Tabor, P. S., and R. A. Neihof Direct determination of activities for microorganisms of Chesapeake Bay populations. Appl. Environ. Microbiol. 48: Zimmermann, R., R. Iturriaga, and J. Becker-Birck Simultaneous determination of the total number of aquatic bacteria and the number thereof involved in respiration. Appl. Environ. Microbiol. 36: