Enhanced Mineralization of Polychlorinated Biphenyls in Soil

Transcription

1 APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Apr. 1993, p /93/ $2./ Copyright D 1993, American Society for Microbiology Vol. 59, No. 4 Enhanced Mineralization of Polychlorinated Biphenyls in Soil Inoculated with Chlorobenzoate-Degrading Bacteria W. J. HICKEY,t D. B. SEARLES, AND D. D. FOCHT* Department of Soil and Environmental Science, University of California, Riverside, Riverside, California Received 17 August 1992/Accepted 2 January 1993 An Altamont soil containing no measurable population of chlorobenzoate utilizers was examined for the potential to enhance polychlorinated biphenyl (PCB) mineralization by inoculation with chlorobenzoate utilizers, a biphenyl utilizer, combinations of the two physiological types, and chlorobiphenyl-mineralizing transconjugants. Biphenyl was added to all soils, and biodegradation of '4C-Aroclor 1242 was assessed by disappearance of that substance and by production of 14Co2. Mineralization of PCBs was consistently greatest (up to 25.5%) in soils inoculated with chlorobenzoate degraders alone. Mineralization was significantly lower in soils receiving all other treatments: PCB cometabolizer (1.7%); chlorobiphenyl mineralizers (8.7 and 14.9%o); and mixed inocula of PCB cometabolizers and chlorobenzoate utilizers (11.4 and 18.%). However, all inoculated soils had higher mineralization than did the uninoculated control (3.1%). PCB disappearance followed trends similar to that observed with the mineralization data, with the greatest degradation occurring in soils inoculated with the chlorobenzoate-degrading strains Pseudomonas aeruginosa JB2 and Pseudonwnas putida Pill alone. While the mechanism by which the introduction of chlorobenzoate degraders alone enhanced biodegradation of PCBs could not be elucidated, the possibility that chlorobenzoate inoculants acquired the ability to metabolize biphenyl and possibly PCBs was explored. When strain JB2, which does not utilize biphenyl, was inoculated into soil containing biphenyl and Aroclor 1242, the frequency of isolates able to utilize biphenyl and 2,5-dichlorobenzoate increased progressively with time from 3.3 to 44.4% between and 48 days, respectively. Since this soil contained no measurable level of chlorobenzoate utilizers yet did contain a population of biphenyl utilizers, the possibility of genetic transfer between the latter group and strain JB2 cannot be excluded. Polychlorinated biphenyls (PCBs) are frequently encountered as contaminants in soil environments, usually originating from electrical transformer leaks or improper disposal of PCB-containing wastes. Remediation of PCB-contaminated soils has typically involved excavation followed by impoundment in landfills or destruction by incineration. However, issues of long-term liability, high cost, and facility access stand to limit the continuing implementation of these strategies and have provided the impetus needed for the examination of alternative technologies, bioremediation being among these. The PCB-degrading capacity of numerous bacterial strains, many of which are common soil organisms, has been described previously (2-4, 9). However, the only PCB congeners typically utilized by bacteria as growth substrates are the monochlorobiphenyls, although recent exceptions have been reported (16). The more highly chlorinated PCB congeners, in contrast, are degraded by a cometabolic process in which biphenyl typically serves as the substrate inducer and cooxidant. The usual end products produced from the cometabolic degradation of PCBs are chlorinated benzoic acids and chlorinated aliphatic acids (1, 11), the former of which are known to serve as growth substrates (6, 12, 19, 22). Numerous studies have examined PCB degradation in soil systems (for reviews, see references 17 and 2); however, one of the first reports examining this process in the context * Corresponding author. t Present address: Department of Soil Science, Wisconsin-Madison, Madison, WI University of of potential bioremediation applications was that of Brunner et al. (5). These investigators examined the effect of inoculation with a PCB cometabolizer (Acinetobacter sp. strain P6) and biphenyl addition on mineralization of [14C]Aroclor While mineralization was more rapid in soils amended with biphenyl and inoculated with Acinetobacter sp. strain P6, the final levels of PCB mineralization (19%) were similar to that found in soils amended with biphenyl alone (17%). Similar findings were reported in a subsequent study except that levels were higher (3%) (7). Another result common to both studies was the occurrence of a several-day lag between the periods of maximal total CO2 and 14C2 evolution rates. Viable plate counts confirmed that the periods of maximal CO2 evolution coincided with that of the highest populations of biphenyl degraders. The corresponding peak for '4C2 was assumed to correspond to maximal activity for commensal populations that metabolized chlorobenzoates and other degradation products from PCBs. However, as the commensals did not metabolize PCBs, these organisms were believed to have no direct impact on the PCB transformation process itself. Several recent developments, however, have warranted a reexamination of the role of chlorobenzoate degraders in the PCB degradation process. First, chlorobenzoates have been reported to be inhibitory to PCB metabolism (16, 21). However, the inhibitory effect is indirect and is actually attributed to chlorocatechols, which are produced when PCB degraders cometabolically transform chlorobenzoates (18, 21). Second, bacteria that degrade di- and trichlorobenzoates, which are likely to be among the primary PCB degradation products (1, 11), have only recently been isolated (13, 14). Thus, the effect of inoculation of these Downloaded from on October 5, 218 by guest 1194

2 VOL. 59, 1993 PCB MINERALIZATION AND CHLOROBENZOATE DEGRADERS 1195 organisms on the PCB degradation process should be reevaluated since they are not likely to be common to most soils. For example, 3-chlorobenzoate, which is the most biodegradable and most studied compound of this class, has been reported in one soil to be metabolized only upon inoculation with a chlorobenzoate utilizer (8). Third, the recent construction and availability of chlorobiphenyl-mineralizing organisms (1, ) permit evaluation of the survival and efficacy of inoculants that theoretically should be exempt from the commensal relation and associated spatial-temporal problems of product diffusion between the two populations. The present study was undertaken to examine the role of chlorobenzoate degraders in conjunction with PCB cometabolizers in soils with respect to optimizing PCB mineralization. MATERUILS AND METHODS Soil preparation. An Altamont series (fine montmorillonitic, thermic, typic chromoxerert) surface soil (ph 6.6; P, 4.53%; total nitrogen, 1.9 g kg-'; organic matter, 41 g kg-1) was used in these studies. The PCB fortification solution (total of 6 ml of hexane) contained 2 mg of unlabelled Aroclor 1242 ml-' and 396,424 dpm of 14C-Aroclor 1,242 ml-'. An atomizer was used to apply this solution as an aerosol to 2 kg of soil. Residual hexane was then allowed to evaporate for 3 days by spreading the soil out in a thin layer in a foil-lined pan. Biphenyl and NH4NO3 were added to the bulk soil in crystalline form at rates of 4, and 3 mg kg-1, respectively. Samples (1 g) of air-dry soil were placed in 25-ml Erlenmeyer flasks, and a total of 27 ml of 5 mm potassium phosphate buffer (ph 7.5) and/or cell suspension was added to bring the soil to 5% moisture-holding capacity. The initial level of 14C-Aroclor 1242 in each flask was determined to be x 16 dpm 1 g of soil-'. Cultures. The chlorobenzoate-degrading organisms utilized in this study were Pseudomonas aenuginosa JB2 and Pseudomonas putida P1ll. Strain JB2 was isolated from a PCB-contaminated soil in Fontana, Calif. (14), while strain P1ll was isolated from sewage sludge in Panama City, Panama (13). Both strain JB2 and strain Plll utilized the following as sole carbon and energy sources: 2-chloro-, 3-chloro-, 2,3-dichloro, 2,5-dichloro-, and 2,3,5-trichlorobenzoate. Neither organism grew on or metabolized biphenyl or chlorobiphenyls. Pseudomonas sp. strain PB133 was also isolated from sewage sludge in Panama City and utilized biphenyl as a sole carbon and energy source but did not grow on chlorobenzoates. In preliminary screening, strain PB133 had demonstrated a relatively broad spectrum of PCB congener degradation and was therefore selected as the test PCB cometabolizer. Pseudomonas sp. strains UCR1 and UCR2 were isolated from chemostat mating experiments (13a). Strain UCR1 mineralized 3-chlorobiphenyl, while strain UCR2 mineralized both 2-chloro- and 2,5-dichlorobiphenyl. Prior to use as inocula, the above-listed strains were grown in a mineral salts medium (14) with the following as selective carbon sources (relevant strains in parentheses): biphenyl (strain PB133), 2-chlorobiphenyl (strain UCR2), 3-chlorobiphenyl (strain UCR1), and 2,3,5-trichlorobenzoate (strains JB2 and Plll). Biphenyl, 2-chlorobiphenyl, and 2,3,5-trichlorobenzoate were added at 5 ppm, while 3-chlorobiphenyl was added at 25 ppm for batch culture growth. Cultures were harvested in late log phase by centrifugation (14) and resuspended in 5 mm phosphate buffer to a final density of approximately 19 cells ml-'. To enumerate the organisms, soil samples (1.27 g [wet weight]) were taken at specified times at and after inoculation from each of the duplicate flasks. These were then added to 99 ml of sterile 5 mm phosphate buffer and shaken for 1 min. Serial dilutions were made in sterile 5 mm phosphate buffer and inoculated (.1 ml) in duplicate onto the appropriate selective media. Following incubation for 1 days at 27 C, dilutions containing between 2 to 2 colonies were counted. CO2 evolution determinations. Flasks were connected to traps through which a vacuum continually withdrew air (flow rate, ca. 2 ml min-'). Incoming and outgoing air was scrubbed of CO2 by passage through a trap containing 25 ml of 1 N KOH. The solutions trapping outgoing CO2 were changed at 2-day intervals, at which time 1-ml samples were withdrawn for titration of total CO2. The samples were mixed with ml of.4 M BaCl2 and 1 drop of phenolphthalein and then titrated with 1 N HCl. The amount of CO2 evolved was then calculated on the basis of the quantity of alkali remaining. Material balance. Four parameters were examined for partitioning of the added 14C label: C2, soil alkali (e.g., HCO3), components extractable by hexane-acetone (residual PCBs and nonpolar metabolites), and nonextractable components ("'C immobilized in the soil organic matter or biomass). For 14CO2 determinations, 2-ml samples (collected as described in the preceding section) were added to 18 ml of AQUASOL-2. Chemiluminesence was allowed to subside for 12 h prior to counting on a Beckman model LS 5TD liquid scintillation counter. Each sample was counted twice, and counts were averaged and adjusted for the counting efficiency (typically 95%). To measure 14C-carbonates, soil samples (2 g) from each treatment were mixed with 2 ml of distilled H2O in a 25-ml Erlenmeyer flask. A vial containing.5 ml of 3 N H2SO4 was then placed in the flask with the soil-water mixture. When the flasks were reconnected to the trapping system and agitated, the introduction of the acid reduced the ph of the soil-water mixture to 4. After incubation for 2 days on the CO2 collection system, samples (2 ml) were withdrawn for 14C counting as described above. Hexane-acetone extractions were performed as described in the following section. Samples (1 ml) were taken from the extracts for liquid scintillation counting as described above. The residual radioactivity remaining in the extracted soil was determined by counting.1 g of extracted soil suspended in 2 ml of scintillation cocktail. Extraction of PCBs. The hexane-acetone extraction procedure described by Brunner et al. (5) was followed for quantification of residual PCBs in soil. Lindane was included as an internal standard. Samples from all soils were analyzed at the beginning and end of the experiment. Gas chromatography. A Hewlett-Packard (Palo Alto, Calif.) model 589 gas chromatograph, fitted with a 3-m DB-5 megabore column (J & W Scientific, Folsom, Calif.) and electron capture detector, was utilized. The column was initially held at 25 C for 1 min then heated to 23 C (2 C min-') and held for 1 min. The injector and detector were held at 24 and 3 C, respectively. The helium carrier gas flow was maintained at 3.5 ml min-'. Chemicals. Both 2- and 3-chlorobenzoate were obtained from Aldrich Chemical (Milwaukee, Wis.), while 2,3,5- trichlorobenzoate was purchased from Transworld Chemical (Chevy Chase, Md.). Both 2- and 3-chlorobiphenyl were obtained from Karl Industries (Aurora, Ohio). Aroclor 1242 was purchased from Foxboro Analytical Laboratories (North Haven, Conn.), while [14C]Aroclor 1242 (24.4 mci Downloaded from on October 5, 218 by guest

3 1196 HICKEY ET AL. APPL. ENVIRON. MICROBIOL. TABLE 1. Recovery of 14C- and unlabelled Aroclor 1242 from soils % of 14C recovered' PCBse (% of total added) By solvent From From From C2 Total extraction carbonates residual soil JB ± b.72 ±.1 7. ± e Pill ± 2. b.73 ± ± ±.82 de PB ± 1.8 a ± ±.21 bc JB2 + PB ± a ± ± 2.56 bc Plll + PB ± a.56 ± ± ±.74 dc UCR ± a.48 ± ± ± 3.1 b ± 7.48 UCR2 7.7 ± ± 2.1 a.53 ±.4 7. ± bc ± 4.56 No inoculation 77.3 ± ± 6.3 a.55 ± ± ±.4 a ± 7.11 arecoveries from each replicate were determined by triplicate gas chromatographic analyses. Values are the averages from the treatment ± standard deviation and are corrected for an extraction efficiency of 72.8%. b Values are reported as the average (± standard deviation) from duplicate analyses of each of the replicates. The 1% value for applied 14C was x 16 dpm 1 g of soil-'. Values followed by a common letter (within a column) were not significantly different (95% level) by Duncan's Student t test. mmol-') and AQUASOL-2 were obtained from New England Nuclear (Boston, Mass.). RESULTS All treatments resulted in higher levels of PCB mineralization relative to that in the noninoculated control (Table 1). The highest PCB disappearance rates (solvent extraction) and mineralization levels (CO2), however, occurred in soils inoculated with the chlorobenzoate-degrading strain JB2 (25.5%) or strain P111 (23.%) alone. A second trend was lower PCB mineralization levels achieved by strains JB2 and P111 inoculated together with the PCB cometabolizer (strain PB133) than when inoculated separately. Mineralization was reduced from 25.5 to 11.3% with strain JB2 and from 23. to 17.9% with strain P1ll. The difference in mineralization with strain P1ll inoculated alone versus that with coinoculation with strain PB133 was marginally significant at the 95% level value (P =.575). Intermediate PCB mineralization levels were observed in soils inoculated with the PCB cometabolizer, strain PB133 (1.7%), or with the chlorobiphenylmineralizing strains UCR1 (8.7%) and UCR2 (14.9%). These levels were not significantly different from one another (Table 1). The PCB disappearance was similar to mineralization in that rates after inoculation with chlorobenzoate degraders alone were greater than those after coinoculation with a PCB cometabolizer (Table 1). Solvent extracts of [14C]PCBs from soils were significantly lower only when the soil was inoculated with strain Plll or JB2. Total amounts of CO2 evolved from all inoculated soils were similar (Fig. 1 to 4). In the absence of inocula, however, biphenyl amendment was less effective, as lower production of 14C2 paralleled that of total CO2 throughout the incubation (Fig. 1). CO2 evolution was not observed to increase in the noninoculated soil until after 5 days of incubation. In all inoculated soils, maximum CO2 evolution rates preceded the maximum rates of 14CO2 evolution by 12 to 14 days (Fig. 1 to 4). Populations of strain JB2 or Plll increased several orders of magnitude during the first 22-day interval and then either leveled out or decreased slightly during the balance of the study (Table 2). While coinoculation with strain PB133 did not appear to affect the growth rates of strains JB2 and Pill, slightly lower final population densities of strains JB2 and Pill were enumerated in soils receiving the dual inocula than in the soils inoculated with strain JB2 or Plll alone. Populations of 2-chlorobenzoate or 3-chlorobenzoate degraders were not detected in the noninoculated soil at dilutions as low as 1' at any time during the study. Plate counts of biphenyl degraders indicated that while the dynamics of these organisms varied by treatment, substantial populations were detected in all soils, including the noninoculated soil (Table 2). Population densities of the biphenyl 8c: Cu co c 18- K a, 2 D E - 4 1: : 9 le nn. -- u iu zu ;3U 4U ou 1% Qu FIG. 1. Cumulative 14C2 (top) and total CO2 (bottom) evolved over time from noninoculated soils () and soils inoculated with biphenyl-utilizing strain PB133 (O). Downloaded from on October 5, 218 by guest

4 VOL. 59, 1993 PCB MINERALIZATION AND CHLOROBENZOATE DEGRADERS 1197 ' 25 2 i Q, 1-_ 'O on N o 25-: E FIG. 2. Cumulative 14CO2 (top) and total CO2 (bottom) evolved over time from soils inoculated with transconjugant chlorobiphenylmineralizing strain UCR1 () or UCR2 (-). degraders were more uniform than those of the chlorobenzoate degraders, with the former usually not fluctuating more than 2 orders of magnitude. The exceptions were the soils that received the mixed inocula, in which the population densities of biphenyl degraders increased several orders of magnitude. Viable plate counts above were determined solely by the ability to use either biphenyl, 2-chlorobenzoate, or 3-chlorobenzoate as a sole carbon source. The ability of an isolate to utilize both biphenyl and a chlorobenzoate had not been considered and was not tested. Therefore, one of the experiments using JB2 as the inoculant was repeated by determining viable plate counts on separate media containing 2,5- dichlorobenzoate or biphenyl. Since no indigenous bacteria could be found from viable plate counts on any of the chlorobenzoates used in this study, colonies obtained on 2,5-dichlorobenzoate presumably owed their origin to P. aeruginosa JB2. Colonies obtained from plates containing biphenyl owed their origins, at the onset of the experiment, to indigenous biphenyl utilizers. To determine whether JB2 acquired the ability to utilize biphenyl, every colony obtained from growth on agar plates containing 2,5-dichlorobenzoate was transferred to agar plates containing biphenyl as the sole carbon source. Colonies that grew were then transferred several times from 2,5-dichlorobenzoate to bi E - J S Cumulative "4CO2 (top) and total CO2 (bottom) evolved FIG. 3. over time from soils inoculated with chlorobenzoate-utilizing strain JB2 () or strain Plll () alone. phenyl to verify their purity and ability to utilize both substrates for growth. The number of colonies found to utilize both substrates increased over time (Table 3). DISCUSSION The results from this study indicated that chlorobenzoate metabolism in soils was a limiting factor in the mineralization of Aroclor The greatest enhancement in mineralization levels was measured in soils inoculated with strains JB2 and Plll alone, even though these organisms could not metabolize PCBs but could degrade a wide range of mono-, di-, and trichlorobenzoates. The enhancement in PCB mineralization and transformation levels in response to the inoculation with either strain JB2 or Plll alone suggests that the indigenous population lacked broad-spectrum chlorobenzoate degradation capabilities. This point was verified by the inability to recover any colonies, by the viable plate count method, from soil on media containing either 2- or 3-chlorobenzoate as the sole carbon source. The question about the mechanism by which inoculation with chlorobenzoate degraders alone enhanced PCB degradation then arises. Specifically, did the inoculant strains simply metabolize substrates (chlorobenzoates) that otherwise accumulated in their absence, or were PCB minerali- Downloaded from on October 5, 218 by guest

5 1198 HICKEY ET AL CL a 3 o 25 E E 1_ lo......,31..., FIG. 4. Cumulative 14C2 (top) and total CO2 (bottom) evolved over time from soils inoculated with strain JB2 in combination with strain PB133 () and soils inoculated with strain P1ll in combination with strain PB133 (). zation levels higher because inoculants stimulated PCB cometabolizers in some manner to enhance the initial transformation of PCBs? If the former occurred, the recoveries of PCBs might be expected to be similar even though mineralization rates would differ. However, this clearly was not the TABLE 2. Viable plate counts of biphenyl-utilizing (BP) and 2-chlorobenzoate-utilizing (2-CBa) bacteria in soil amended with biphenyl and Aroclor 1242 log CFU/g of dry soil ona: Treatment Day Day 22 Day 49 2-CBa BP 2-CBa BP 2-CBa BP No inoculation < < < UCR1 ND 7.11 ND 7.7 ND 7.25 UCR2 ND 7.61 ND 8.44 ND 7.81 JB ND 7.89 ND 8.4 ND P ND 6.2 ND 7.64 ND PB133 ND 7.46 ND 9.57 ND 9.4 JB2-PB P111-PB a Lowest dilution plated (on agar medium) = 1-3. ND, not determined. TABLE 3. Increase in putative recombinant strains (able to utilize 2,5-dichlorobenzoate and biphenyl) over time Time (days) nn Putative recombinants a Number of colonies examined. APPL. ENVIRON. MICROBIOL. case, as significantly lower levels of PCBs were recovered from the soils inoculated with chlorobenzoate degraders alone. The argument that enhanced PCB mineralization was at least in part a consequence of enhanced PCB transformation may be thus made. One possible mechanism to account for the enhancement in PCB degradation and apparent stimulation of PCB cometabolizers relates to the role of chlorobenzoates as inhibitor precursors. Chlorobenzoates are metabolized to the corresponding chlorocatechols (18, 21), which are potent inhibitors of 2,3-dihydroxybiphenyl dioxygenase (1, 21), a central enzyme involved in PCB cometabolism. Thus, there exists the possibility that the efficacy of PCB cometabolizers is attenuated by an end product of their degradative activity. It is conceivable that the chlorobenzoate degraders could assimilate metabolites produced by PCB degraders that would otherwise accumulate and inhibit the further transformation of the PCBs. However, this begs the question of why coinoculation of the biphenyl utilizer PB133 with either chlorobenzoate utilizer JB2 or Plll gave lower rates of PCB metabolism than did inoculation with either chlorobenzoate utilizer alone. To keep both sides of the argument in perspective, it should be noted that the addition of either chlorobenzoate degrader with the biphenyl degrader PB133 did not reduce the efficacy of PCB degradation compared with that attained by PB133 alone (Table 1). Clearly, PCB degradation rates in soil, as measured by disappearance of PCB and 14C2 liberation, were greater when the soil was inoculated with either a biphenyl degrader (PB133), a chlorobenzoate degrader (Plll or JB2), or combinations thereof than when the soil received no inoculum. Thus, addition of chlorobenzoatedegrading inoculants did not assist in removing PCB metabolites produced by the biphenyl utilizers. Moreover, an argument that the two populations were inhibitory to or competitive with each other cannot be made, since there were no discernible differences in population counts among the single- or dual-inoculum treatments (Table 2). In the design of the initial experiment, it had been presumed that colonies formed on biphenyl plates owed their origin solely to indigenous bacteria when soil was inoculated with either Plll or JB2. Thus, the possibility of genetic exchange between the chlorobenzoate degraders and the indigenous biphenyl utilizers was not considered a priori. Since strain JB2 gave the highest rates of mineralization, an additional experiment was designed to determine whether JB2 could have acquired the ability to utilize biphenyl, presumably as a result of genetic exchange with indigenous biphenyl utilizers in soil. The results (Table 3) clearly indicate that this hypothesis could not be rejected, since the frequency of putative recombinants (i.e., isolates capable of utilizing 2,5-dichlorobenzoate and biphenyl as sole carbon sources) continued to increase with time. Although a control Downloaded from on October 5, 218 by guest

6 VOL. 59, 1993 PCB MINERALIZATION AND CHLOROBENZOATE DEGRADERS 1199 (no inoculum) was not run, the counterargument that the addition of biphenyls and PCBs to soil simply enhanced the growth of a lower than detectable level of chlorobenzoate degraders in the soil is not supported by the data (Table 2). Moreover, rates of PCB degradation in the uninoculated soil were much lower than those with all other treatments. If genetic exchange between indigenous soil bacteria and JB2 accounted for the observed increase in isolates capable of utilizing both substrates, then mineralization of PCB congeners might be brought about by a single organism. A single strain would effect a far greater rate of PCB mineralization than a consortium (PCB cometabolizer plus chlorobenzoate utilizer), the latter of which is dependent upon diffusion of products from one to the other and is not kinetically capable of carrying out synchronous growth (7). If genetic exchange between indigenous biphenyl utilizers and chlorobenzoate-utilizing inoculants is the basis for enhanced PCB degradation, it is necessary to consider why mixed inoculations with PB133, the biphenyl utilizer, were not likewise more efficacious. Since the only treatment tested for the ability of single isolates to utilize chlorobenzoates and biphenyl was with JB2, it is not possible to know whether similar results might have occurred with the other treatments. Genetic exchange between JB2 and PB133 would not produce a strain having greater capacity towards the initial attack on Aroclor 1242 congeners than PB133. The diversity of the indigenous biphenyl utilizers with respect to congener attack is unknown. Their slow mineralization of biphenyl (Fig. 1) is no indication of their capacity or spectrum in attacking Aroclor 1242 congeners: conceivably they could be more diverse than PB133. The decreased mineralization levels achieved with dual inocula relative to that achieved with chlorobenzoate degraders alone illustrated the problems of inoculating multiple organisms into soil, as opposed to the potential advantages of constructed organisms for delivering both PCB-cometabolizing and chlorobenzoate-degrading activities. In this study, mineralization levels obtained with the constructed, chlorobiphenyl-mineralizing organisms (strains UCR1 and UCR2) were at best 2% greater than that achieved by strain PB133 alone and less than that achieved by strain JB2 or P1ll alone. The results could be attributed to the lack of a broad PCB and/or chlorobenzoate degradation spectrum. With regard to the latter activity, strain UCR1 was able to utilize only 3-chlorobenzoate while strain UCR2 degraded 2-chloro- and 2,5-dichlorobenzoate. This spectrum of chlorobenzoate-degrading activity was far less than that of strain JB2 or P1ll. Moreover, cometabolism of Aroclor 1242 by both strains was poor and restricted primarily to mono- and dichlorocongeners (2a). Nevertheless, the problems with coinocula indicated herein provide further impetus for developments in strain construction. In summation, PCB mineralization and transformation were enhanced in the tested soil by the introduction of chlorobenzoate degraders, strains Plll and JB2, which were not previously known to metabolize PCBs. Thus, the important role that chlorobenzoate-utilizing commensals may play in enhancing PCB degradation cannot be overlooked. Whether this enhancement is effected by a consortium or by genetic exchange among inoculant and indigenous bacteria remains unclear. ACKNOWLEDGMENTS We thank Fain Sutherland (U.C. Riverside) for his capable technical assistance and Blanca Hernandez (Universidad de Panama, Panama City) for supplying strains P1ll and PB133. This work was supported by grants from the University of California Biotechnology Research and Education Program and the Occidental Chemical Corp. REFERENCES 1. Adams, R. H., C.-M. Huang, F. K. Higson, V. Brenner, and D. D. Focht Construction of a 3-chlorobiphenyl-utilizing recombinant from an intergeneric mating. Appl. Environ. Microbiol. 58: Ahmed, M., and D. D. Focht Degradation of polychlorinated biphenyls by two species of Achromobacter. Can. J. Microbiol. 19: a.Arensdorf, J. J. Unpublished data. 3. Baxter, R. A., R. E. Gibert, R. A. Lidgett, J. H. Mainprize, and H. A. Vodden The degradation of polychlorinated biphenyls by microorganisms. Sci. Total Environ. 4: Bedard, D. L., R. Unterman, L. H. Bopp, M. J. Brennan, M. L. Haberl, and C. Johnson Rapid assay for screening and characterizing microorganisms for the ability to degrade polychlorinated biphenyls. Appl. Environ. Microbiol. 51: Brunner, W., F. H. Sutherland, and D. D. Focht Enhanced biodegradation in soil by analog enrichment and bacterial inoculation. J. Environ. Qual. 14: Dorn, E., M. Hellwig, W. Reineke, and H.-J. Knackmuss Isolation and characterization of a 3-chlorobenzoate degrading pseudomonad. Arch. Microbiol. 99: Focht, D. D., and W. Brunner Kinetics of biphenyl and polychlorinated biphenyl metabolism in soil. Appl. Environ. Microbiol. 5: Focht, D. D., and D. Shelton Growth kinetics of Pseudomonas alcaligenes C-O relative to inoculation and 3-chlorobenzoate metabolism in soil. Appl. Environ. Microbiol. 53: Furukawa, K., F. Matsumura, and K. Tonomura Alcaligenes and Acinetobacter strains capable of degrading polychlorinated biphenyls. Agric. Biol. Chem. 42: Furukawa, K., K. Tonomura, and A. Kamibayashi Effect of chlorine substitution on the biodegradability of polychlorinated biphenyls. Appl. Environ. Microbiol. 35: Furukawa, K., N. Tomizuka, and A. Kamibayashi Metabolic breakdown of kaneclors (polychlorobiphenyls) and their products by Acinetobacter sp. Appl. Environ. Microbiol. 46: Hartmann, J., W. Reineke, and H.-J. Knackmuss Metabolism of 3-chloro-, 4-chloro-, and 3,5-dichlorobenzoate by a pseudomonad. Appl. Environ. Microbiol. 37: Hernandez, B. S., F. K. Higson, R. Kondrat, and D. D. Focht Metabolism of and inhibition by chlorobenzoates in Pseudomonasputida Plll. Appl. Environ. Microbiol. 57: a.Hickey, W. J Ph.D. thesis. University of California, Riverside. 14. Hickey, W. J., and D. D. Focht Degradation of mono-, di-, and trihalogenated benzoic acids by Pseudomonas aenrginosa JB2. Appl. Environ. Microbiol. 56: Mokross, H., E. Schmidt, and W. Reineke Degradation of 3-chlorobiphenyl by in vivo constructed hybrid pseudomonads. FEMS Microbiol. Lett. 71: Nawaz, M. S., and K. D. Chapatawala Simultaneous degradation of acetonitrile and biphenyl by Pseudomonas aenrginosa. Can. J. Microbiol. 37: Pal, D., J. B. Weber, and M. R. Overcash Fate of polychlorinated biphenyls in soil-plant systems. Residue Rev. 74: Parsons, J. R., D. T. H. M. Sum, A. van Laar, and. Hutzinger Biodegradation of chlorinated biphenyls and benzoic acids by a Pseudomonas species. Appl. Microbiol. Biotechnol. 29: Rusinger, S., U. Kiages, and F. Lingens Degradation of 4-chlorobenzoic acid by an Arthrobacter species. Arch. Microbiol. 11: Downloaded from on October 5, 218 by guest

7 12 HICKEY ET AL. APPL. ENVIRON. MICROBIOL. 2. Sklarew, D. S., and D. C. Girvin Attenuation of polychlorinated biphenyls in soils. Rev. Environ. Contam. Toxicol. 98: Sondossi, M., M. Sylvestre, and D. Ahmad Effects of chlorobenzoate transformation on the Pseudomonas testosteroni biphenyl and chlorobiphenyl degradation pathway. Appl. Environ. Microbiol. 58: Sylvestre, M., K. Mailhiot, and D. Ahmad Isolation and preliminary characterization of a 2-chlorobenzoate degrading Pseudomonas. Can. J. Microbiol. 35: Downloaded from on October 5, 218 by guest