Aerobic Biodegradation of Biphenyl and Polychlorinated Biphenyls by ...

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Sept. 1997, p. 3378–3384 0099-2240/97/$04.0010 Copyright © 1997, American Society for Microbiology

Vol. 63, No. 9

Aerobic Biodegradation of Biphenyl and Polychlorinated Biphenyls by Arctic Soil Microorganisms WILLIAM W. MOHN,1* KAROLINA WESTERBERG,1 WILLIAM R. CULLEN,2 2,3 AND KENNETH J. REIMER Department of Microbiology and Immunology1 and Department of Chemistry,2 University of British Columbia, Vancouver, British Columbia, and Environmental Sciences Group, Royal Military College, Kingston, Ontario,3 Canada Received 20 March 1997/Accepted 6 June 1997

We examined the degradation of biphenyl and the commercial polychlorinated biphenyl (PCB) mixture Aroclor 1221 by indigenous Arctic soil microorganisms to assess both the response of the soil microflora to PCB pollution and the potential of the microflora for bioremediation. In soil slurries, Arctic soil microflora and temperate-soil microflora had similar potentials to mineralize [14C]biphenyl. Mineralization began sooner and was more extensive in slurries of PCB-contaminated Arctic soils than in slurries of uncontaminated Arctic soils. The maximum mineralization rates at 30 and 7°C were typically 1.2 to 1.4 and 0.52 to 1.0 mg of biphenyl g of dry soil21 day21, respectively. Slurries of PCB-contaminated Arctic soils degraded Aroclor 1221 more extensively at 30°C (71 to 76% removal) than at 7°C (14 to 40% removal). We isolated from Arctic soils organisms that were capable of psychrotolerant (growing at 7 to 30°C) or psychrophilic (growing at 7 to 15°C) growth on biphenyl. Two psychrotolerant isolates extensively degraded Aroclor 1221 at 7°C (54 to 60% removal). The soil microflora and psychrotolerant isolates degraded all mono-, most di-, and some trichlorobiphenyl congeners. The results suggest that PCB pollution selected for biphenyl-mineralizing microorganisms in Arctic soils. While low temperatures severely limited Aroclor 1221 removal in slurries of Arctic soils, results with pure cultures suggest that more effective PCB biodegradation is possible under appropriate conditions. and degradative activities of such organisms. Furthermore, we know very little about their possible role in the response of Arctic soil microbial communities to pollution. Traits particular to psychrotolerant microorganisms may be necessary for bioremediation of Arctic soils and may be generally useful for bioremediation of other environments, such as soils and waters in temperate regions as well as in aquifers. Gounot noted in a review article (15) that psychrotolerant bacteria may be adapted to fluctuations in temperature and other challenging conditions that are also common in cold environments, such as low nutrient availability, low water activity, and high pressure. Therefore, microorganisms adapted to these environments may have important advantages in biotechnological applications. When mesophilic microorganisms are used in bioremediation, heating of the treatment system is often required. Using psychrotolerant microorganisms in bioremediation systems would mean a reduced need for heating and a possible corresponding reduction in costs (20, 22). In this study, we addressed some of the questions surrounding the potential for aerobic PCB bioremediation at Arctic sites. Studies of the anaerobic dechlorination of PCBs in Arctic soils are also in progress, with the ultimate objective being PCB bioremediation through sequential anaerobic-aerobic biological treatment. Here we show that aerobic microorganisms indigenous to soils from two sites in the Canadian Arctic can degrade biphenyl and PCBs with up to three chlorine substituents. We examined biphenyl, in addition to PCBs, because biphenyl degraders are known to metabolize PCBs and because biphenyl as a cosubstrate can stimulate PCB biodegradation. We found that the potentials of Arctic and temperateregion soil microflora to degrade biphenyl were similar when compared to 7 and 30°C. We show evidence suggesting that PCB pollution has selected for biphenyl-degrading microflora at the Arctic sites. We enriched and isolated microbial strains which grow on biphenyl and degrade PCBs at low temperature (7°C). We believe this is the first report of psychrotolerant

Numerous Arctic and sub-Arctic sites have been contaminated with polychlorinated biphenyls (PCBs). High concentrations of PCBs exist at some former military radar facilities and other sites where electronic equipment has been used. In some cases, such contamination is entering the marine (25) and terrestrial (10) food chains and spreading via atmospheric transport (9, 16). The Canadian Environmental Protection Act, as well as governmental policies and agreements, requires cleanup of many of the PCB-contaminated sites. Due to the remote location of these sites, conventional cleanup technologies are very expensive, and bioremediation could result in significant cost savings. However, the cold climate of these sites poses substantial technical challenges for bioremediation. Although PCB metabolism has been intensively studied (reviewed in reference 1), very little is known about PCB degradation by Arctic soil microflora or, more generally, by psychrotolerant (also referred to as psychrotrophic) organisms. The only such information we are aware of is a conference report by Williams (28) of aerobic biodegradation of di- and trichlorobiphenyl congeners in river sediment at 4°C. The biodegradation of pollutants other than PCBs in cold environments has been examined. The biodegradation of crude oil at low temperature has received the greatest attention (reviewed in reference 26). The degradation of phenol (22), chlorophenol (20), toluene (8, 27), naphthalene (27), dodecane (27), and hexadecane (27) also occurs at low temperatures (#10°C). Kolenc et al. (21) transferred genes encoding toluene degradation to a psychrophilic strain of Pseudomonas putida, which then used toluene as the sole carbon source at temperatures as low as 0°C. Although psychrotolerant microorganisms can degrade several pollutants, we know very little about the occurrence * Corresponding author. Mailing address: Department of Microbiology and Immunology, University of British Columbia, #300-6174 University Blvd., Vancouver, BC V6T 1Z3, Canada. Phone: (604) 822-4285. Fax: (604) 822-6041. E-mail: [email protected]. 3378

VOL. 63, 1997

PCB BIODEGRADATION BY ARCTIC SOIL MICROORGANISMS TABLE 1. Soils used in this study

Source

PCB concn (ppm)a

% Sand

% Silt

% Clay

% Organic C

Saglek Saglek Saglek Resolution Island Resolution Island Resolution Island

,1 5 .50 ,1 10 .50

78 82 82 90 92 95

18 14 14 7 3 2

4 3 4 4 5 3

0.15 0.38 0.33 0.22 1.2 0.22

a

Milligrams of PCBs per kilogram of dry soil.

microbial isolates with these capabilities. The results suggest that indigenous microorganisms may be useful for PCB bioremediation at Arctic sites. MATERIALS AND METHODS Soil samples. Samples from two locations in the Canadian Eastern sub-Arctic and Arctic were studied. The first site was Saglek, Labrador (58°N, 64°W), on the northern tip of Labrador. Although this site lies south of 60°N, its climate and ecology are typical of Arctic regions. The second site was Resolution Island, Northwest Territories (62°N, 65°W), north of Saglek and near the southern tip of Baffin Island. Areas at the sites became polluted with PCBs when they were used by both the U.S. and Canadian governments for military radar installations. From each site, we collected soil samples with different amounts of PCB contamination (mainly Aroclor 1260): uncontaminated, intermediately contaminated (5 to 10 ppm PCBs) and highly contaminated (.50 ppm PCBs). PCBs in the soils were analyzed by Axys Analytical Services, Sidney, British Columbia, Canada, using Environmental Protection Agency method 8081. The physical properties of these soils were characterized by Pacific Soil Analysis, Richmond, British Columbia, Canada (Table 1). The soils were very low in organic carbon. The soil from Saglek was a loamy sand, and that from Resolution Island was a sand (Canadian System of Soil Classification). Two uncontaminated soil samples from Vancouver (temperate climate) were also used. These were called garden soil and agricultural soil to indicate their origin. All the soil samples were sieved to remove any particles larger than 2 mm and were then stored at 7°C. Soil slurries. To prepare soil slurries, moist soil equivalent to 5 g (dry weight) was put in a clean 125-ml Wheaton bottle with a Teflon-lined screw cap. Distilled water was added to give a total water volume of 19.5 ml. Then, 0.5 ml of a stock solution containing NH4Cl, KH2PO4, and K2HPO4 (pH 7.0) was added to give final concentrations of 0.1 mM ammonium and 12.5 mM phosphate. All treatments were carried out in triplicate, and the slurries were incubated on shakers (150 rpm) at 30 or 7°C. Control slurries were autoclaved for 30 min on three consecutive days and incubated at their future incubation temperature between autoclavings. Biphenyl mineralization experiments. To soil slurries, [UL-14C]biphenyl (Sigma) and unlabeled biphenyl (Aldrich) were added from a stock solution in acetone to give 5 mg of biphenyl per liter and 8 3 105 dpm of 14C per slurry. 14 CO2 was trapped by placing a plastic tube (13 by 70 mm) containing 0.5 ml of 0.5 M NaOH in the slurry during incubation (because of its height, the free tube could not tip over in the bottle containing the slurry). The basic solution was replaced weekly, and the used solution was mixed with 10 ml of Ready Gel scintillation cocktail. These samples were held overnight in the dark and then counted in a Beckman LS 6000IC scintillation counter. PCB degradation experiments. Slurries with intermediately contaminated soils from both Saglek and Resolution Island were used for PCB degradation experiments. The soil slurries were prepared as described above, and biphenyl, Aroclor 1221 (Accu Standard), plus an internal standard were added from a stock solution in acetone. The final concentration was 50 mg/liter for both biphenyl and Aroclor 1221. Aroclor 1221 contains mostly monochlorobiphenyls (MCBs) and dichlorobiphenyls (DCBs) (see Table 2). The internal standard was 2,29,4,49,6,69-hexachlorobiphenyl (HCB) (Accu Standard) at a final concentration of 5 mg/liter. HCB was not present in the soils prior to its addition and was not degraded during incubations. The soil slurries were subsampled for PCB analysis by removing approximately 2 ml with a glass pipette with the tip cut off. The amount of solid material in each subsample varied, but the ratio of each compound to the internal standard was not affected by that variation. Culture methods. Microorganisms were extracted from soil by blending for 10 s in a solution of 0.1% (wt/vol) sodium pyrophosphate and 0.1% (vol/vol) Tween 80. For enumeration of total culturable heterotrophs, dilutions (in sterile saline solution) were spread plated in triplicate on Trypticase soy broth (BBL) supplemented with 1.5% agar. The plates were incubated at either 30 or 7°C. Colonies were counted at regular intervals for 3 weeks. The highest counts are reported here. Several methods were attempted to enumerate organisms able to grow on biphenyl. For these attempts PAS mineral medium (6) without yeast extract was

3379

used. Where indicated, the medium contained 10% soil extract or 50 mg of yeast extract per liter. Solid medium contained purified agar (BBL) or agarose (GibcoBRL). Biphenyl was added to the solid media in an overlay or by being spread on the surface; alternatively, biphenyl was provided to the plates as a vapor. Biphenyl was added to the liquid media in screw-cap tubes from an acetone stock solution to give final concentrations from 100 to 500 mg/liter. Soil microorganisms were extracted and diluted as above and either spread on plates or used to inoculate liquid media. Solid and liquid media were incubated at 7 or 30°C. These liquid cultures were incubated without shaking. Enrichment cultures were prepared using 20 ml of PAS medium with 500 mg of biphenyl per liter as the sole organic substrate. These cultures were inoculated with 0.2 g of soil and incubated in 50-ml flasks on shakers at 200 rpm. Enrichment cultures were incubated at 7 or 30°C. If growth occurred in these enrichment cultures, they were transferred to homologous medium. After three serial passages, the enrichment cultures were serially diluted in PAS, and the dilutions were plated on PAS with 1.5% purified agar. These plates were incubated in the presence of biphenyl vapors. Representatives of all colony morphotypes which formed were purified by streaking on homologous plates. Isolated colonies from each streaked plate were then picked and used to inoculate tubes containing 2.5 ml of PAS with 500 mg of biphenyl per liter. The tubes were incubated on a tube roller. After two serial passages on this liquid medium, isolates were considered capable of growing on biphenyl as a sole organic substrate. The temperature range of enrichment cultures and isolates was tested in 2.5 ml of PAS with 200 mg of biphenyl per liter in sealed screw-cap tubes (to prevent loss of biphenyl) incubated without shaking at the specified temperatures. The ability of enrichment cultures and isolates to degrade PCBs was tested in 2.5 ml of PAS with 100 mg of biphenyl per liter plus 100 mg of Aroclor 1221 or Aroclor 1242 per liter in sealed screw-cap tubes incubated at 7°C on a tube roller for at least 1 week beyond the time when growth (monitored by optical density measurement) stopped. Entire cultures were used as samples for PCB analysis. Analysis of PCBs. Analytical samples were stored at 220°C. Soil samples were extracted once with acetone and twice with a 3:1 mixture of hexane and acetone. The pooled extracts were evaporated under a stream of nitrogen until the organic phase reached a volume of about 2 ml. The extract was then extracted three times with hexane, and the organic phase was removed and passed over Na2SO4. After the extract was evaporated down to approximately 1 ml, 5 ml of hexane was added and the sample was again evaporated down to 1 ml. The resulting volume was passed through a Florisil (Fisher) column. The eluate was collected, and hexane was passed through the column until the sample volume was 5 ml. Cultures were extracted twice with hexane, and the extract was passed over Na2SO4. PCBs were analyzed with a Varian 3400 gas chromatograph (GC) connected to a Varian Saturn 4D ion trap mass spectrometer. A Varian 8200 autosampler was used. The GC was equipped with a DB-5 capillary column (J&W Scientific; 30 m by 0.25 mm [inner diameter]). The volume injected was 2.5 ml, the temperature of the injector was held at 260°C, the temperature of the transfer line was 280°C, and the carrier gas was helium. The column temperature program was as follows: 104°C for 3 min, increased at 20°C/min to 160°C, increased 2.5°C/min to 233°C, increased 20°C/min to 290°C, and held for 3 min. The mass spectrum of each GC peak was used to determine if the peak corresponded to a PCB and, if so, to determine the number of chlorine substituents. The masses that were monitored were m/z 152 to 154 (biphenyl), m/z 188 to 190 (MCB), m/z 222 to 226 (DCB), m/z 256 to 262 (trichlorobiphenyl [TCB]) and m/z 358 to 370 (HCB). Congener assignments (Table 2) were made by comparing retention times of peaks to published data (2, 11, 13, 24) and by comparing peak areas to data on relative amounts of different PCB congeners in Aroclor 1221. The peaks were quantified by integrating the area under each peak and then comparing it to the area of the peak corresponding to the internal standard, which allowed accurate measurement of the percent change of individual peaks in cultures. Biological removal of each congener was determined as the difference between the amount of that congener in live cultures and the amount in autoclaved controls. Total PCB removal was calculated using Aroclor 1221 as a calibration standard and assuming that it had the expected percent composition of PCB congeners (Table 2). Since the percent composition of PCB congeners varies among lots of Aroclors (12), total PCB removal was only approximated.

RESULTS AND DISCUSSION Biphenyl mineralization in soil. Biphenyl was readily mineralized in slurries of the Arctic soils (Fig. 1). This mineralization activity was presumably biological, because it was completely inhibited by autoclaving. The extent of biphenyl mineralization in Arctic soils typically ranged from 35 to 67%. Such values are consistent with complete degradation of biphenyl, with the remaining 14C probably being incorporated into biomass. The Arctic soils had intrinsic potential for biphenyl mineralization which was very similar to that of soils from a temperate region, since the rates and extents of biphenyl mineralization in the Arctic and temperate soils were comparable (Fig. 2). One Arctic soil and one temperate-region soil

3380

MOHN ET AL.

APPL. ENVIRON. MICROBIOL.

TABLE 2. Relative retention times, congener assignments, and relative amounts of PCB congeners in Aroclor 1221 Peak no.

Relative retention timea

Congener assignment(s)

Amt (wt %)b

1 2 3 4 5 5 6 6 7 8 8 9 10 11 11 12 13 14 15 15 16 17 18 19 20 20 21 21 22 23

0.306 0.373 0.424 0.430 0.458 0.458 0.501 0.501 0.518 0.531 0.531 0.564 0.597 0.607 0.607 0.611 0.615 0.624 0.631 0.631 0.649 0.652 0.697 0.700 0.718 0.718 0.737 0.737 0.756 0.864

Biphenyl 2 3 4 2,29 2,6 2,4 2,5 2,39 2,3 2,49 2,29,6 3,39 3,4 3,49 2,29,5 2,29,4 4,49 2,3,6 2,39,6 2,29,3 2,49,6 2,39,5 2,39,4 2,4,49 2,49,5 2,3,39 29,3,4 2,3,49 3,4,49

19 27.25 2.85 15.07 5.51 0.49 1.45 1.38 2.91 0.72 9.81 0.11 0.09 0.6 1.07 0.76 0.38 3.87 0.04 0.15 0.34 0.17 0.15 0.12 0.61 0.55 0.07 0.50 0.25 0.18

a Retention times are relative to the internal standard, 2,29,4,49,6,69-HCB, which had a retention time of 22.49 min. b See reference 29.

had exceptionally low potentials for biphenyl mineralization at 30°C and apparently lacked biphenyl degraders which tolerate such a high temperature. Biphenyl degraders in Arctic soils do not appear better adapted to low temperature, or worse adapted to high temperature, than biphenyl degraders in temperate-region soils. Generally, biphenyl mineralization in Arctic soils was stimulated more at 30°C than at 7°C (Fig. 1). The onset of biphenyl mineralization in each of the Arctic soils was faster at 30°C than at 7°C. The maximum detected rate of mineralization was usually higher at 30°C (1.2 to 1.4 mg of biphenyl g of dry soil21 day21) than at 7°C (0.52 to 1.0 mg of biphenyl g of dry soil21 day21). An exception to this trend was the uncontaminated soil from Resolution Island, which at 30°C had a maximum rate of only 0.28 mg of biphenyl g of dry soil21 day21. This treatment was repeated to confirm the low extent and rate of mineralization, and the same result was obtained. Finally, the extent of biphenyl mineralization in each soil was usually higher at 30°C than at 7°C. Exceptions to this trend were the extents of biphenyl mineralization in uncontaminated and moderately contaminated soils from Resolution Island. Considering the usual doubling of enzymatic rates for each 10°C increase in temperature, the differences in biphenyl mineralization at 7 and 30°C appear relatively small. These small differences could be due to the coexistence in the soils of biphenyl-mineralizing populations with different temperature optima. Generally, biphenyl mineralization was stimulated more in PCB-contaminated Arctic soils than in uncontaminated ones

(Fig. 1). At both 7 and 30°C, the onset of biphenyl mineralization in contaminated Arctic soils was always faster than in the uncontaminated one from the same site (i.e., there was a longer lag period before biphenyl mineralization occurred in the uncontaminated soils). Contamination by PCBs did not noticeably affect the maximum detected rate of biphenyl mineralization in soils from either Arctic site. However, the biphenyl mineralization data do not have sufficient resolution for precise kinetic analysis of mineralization. Further, the maximum rates of mineralization may have been limited by the availability of biphenyl as a function of its interactions with the soil matrix. The extent of biphenyl mineralization was always higher in contaminated Arctic soil than in the uncontaminated one from the same site; it was usually significantly higher. Differences in the time before the onset of mineralization and in the extent of mineralization are most easily explained by higher populations of biphenyl degraders in the polluted soils, suggesting in situ selection for biphenyl mineralizing organisms. Selective pressure in the polluted soils may have been for organisms using PCBs or cocontaminants as growth substrates or against organisms inhibited by those compounds or their metabolites. Alternatively, biphenyl degraders in these soils could be in a genetic or physiological state more ready for biphenyl mineralization. It does not appear likely that differences in the physical properties of the soil are responsible for differences in biphenyl mineralization, because the soils from each location are physically quite similar (Table 1). PCB biodegradation in soil. In the presence of added biphenyl, added PCBs were removed during incubation of slurries of the intermediately contaminated Arctic soils (Fig. 3). Removal of PCBs from these soils was due to biodegradation (not to sorption, polymerization, etc.), because there was no loss of PCBs from autoclaved control cultures. Removal of PCBs by biodegradation is also consistent with the above biological mineralization of biphenyl. Biphenyl and PCB removal were monitored weekly in one of four replicate cultures of each treatment. After 6 weeks at 30°C or 8 weeks at 7°C, PCB removal had ceased in the monitored replicates and the remaining three replicates were analyzed for final PCB removal (Fig. 3). The Saglek soil slurries incubated at 30°C were respiked with biphenyl after 4 weeks, since the biphenyl had been completely depleted after 3 weeks. This respiking was done to ensure that PCB degradation was not limited by a potential requirement for biphenyl as a cosubstrate. After 6 weeks of incubation of the intermediately contaminated Arctic soils at 30°C, PCB degradation was extensive, and the patterns of PCB congeners degraded in the two soils were very similar (Fig. 3A and C). The three MCBs, as well as 2,4-, 2,5-, 3,4-, and 3,49-DCB were completely removed. Also, 2,39-, 4,49-, and at least one of 2,49- and 2,3-DCB were partly removed. There was little or no degradation of 2,29and 3,39-DCB. As for the TCBs, there was significant degradation of 2,3,49- and 2,39,4-TCB in the soil from Saglek. It appears that these TCBs were also degraded in the soil from Resolution Island, but the evidence is less conclusive. The specificities for PCB congener degradation observed in the Arctic soils were consistent with earlier studies. In general, the fewer the chlorine substituents, the more likely a PCB congener will be degraded, and congeners with chlorine atoms on only one of the biphenyl rings are more easily degraded than congeners with the same number of chlorines on both rings (4, 6, 14, 18). Figure 3 shows the percent degradation of each congener. However, the congeners in Aroclor 1221 vary in abundance (Table 1); for example, each of the TCBs constitutes less than 1% of Aroclor 1221. All of the major components (.1%) of Aroclor 1221, with the

VOL. 63, 1997

PCB BIODEGRADATION BY ARCTIC SOIL MICROORGANISMS

3381

FIG. 1. Biphenyl mineralization in slurries of soils from two Arctic sites at two temperatures. Symbols represent soils with different degrees of PCB contamination; ■, uncontaminated soil; å, intermediately contaminated soil (5 or 10 ppm PCBs); 3, highly contaminated soil ($50 ppm PCBs); F, autoclaved control with intermediately contaminated soil. Error bars show standard deviations (n 5 3). Different lowercase letters indicate data points which are significantly different according to analysis of variance (ANOVA) (P 5 0.05).

exception of 2,29-DCB, were biodegraded at 30°C in both of the intermediately contaminated Arctic soils. Overall, 71% (wt/wt) of Aroclor 1221 was degraded in the soil from Saglek at 30°C and 76% was degraded in the soil from Resolution Island. After 8 weeks of incubation of intermediately contaminated Arctic soils at 7°C, PCB degradation was much more limited than after 6 weeks at 30°C (Fig. 3). The pattern of PCB con-

FIG. 2. Biphenyl mineralization in slurries of soils from two temperate region sites at two temperatures. Squares, garden soil; circles, agricultural soil; solid symbols, incubated at 30°C; open symbols, incubated at 7°C. Error bars show standard deviations (n 5 3). Different lowercase letters indicate data points which are significantly different according to ANOVA (P 5 0.05).

geners degraded at 7°C in the soil from Saglek was similar to that obtained at 30°C. In the soil from Resolution Island, there was substantial degradation of only one congener, 4-MCB, at 7°C. Biphenyl degradation in each soil was also much more limited at 7°C than at 30°C. At 7°C, 40% of Aroclor 1221 was degraded in the soil from Saglek and only 14% was degraded in the soil from Resolution Island. There are a number of possible explanations for the limitation of PCB biodegradation in soil at 7°C. First, cometabolism of PCBs by biphenyl degradation enzymes may be more sensitive to low temperature than is biphenyl degradation. We are not aware of any studies examining the possible effect of temperature on the specificity of such enzymes. Second, distinct organisms may have been responsible for biphenyl degradation at low temperature (7°C) and may lack the capability to cometabolize PCBs. Third, the bioavailability of PCBs, controlled by their sorption to the soil, may have been affected by temperature. Finally, differences between the biphenyl mineralization and PCB degradation experiments may have been a factor. The latter experiments included more biphenyl (5 versus 50 ppm) and included Aroclor 1221. It is possible that the higher concentration of biphenyl was inhibitory to PCB degradation or that the higher concentration of biphenyl or the Aroclor 1221 was toxic to the biphenyl-degrading microorganisms active at 7°C. The soils were stored at 7°C for 4 months between the two experiments, and it is possible that the microorganisms that were active on biphenyl at 7°C did not survive the storage as well as those that were active at 30°C. Whatever the explanation, PCBs were persistent in soils at low temperatures in this

3382

MOHN ET AL.

APPL. ENVIRON. MICROBIOL.

FIG. 3. Degradation of Aroclor 1221 in slurries of intermediately contaminated soils from two Arctic sites at two temperatures. The peaks are identified in Table 2. Error bars show the standard deviation (n 5 3). Slurries at 30°C were incubated for 6 weeks; those at 7°C were incubated for 8 weeks.

study, and their more effective biodegradation in soil at low temperatures remains to be demonstrated. Culturable microorganisms. The temperate-region soils had 10 to 104 times as many culturable heterotrophs (able to form colonies on Trypticase soy agar) as did the Arctic soils (Table 3). The low populations of culturable heterotrophic microorganisms in the Arctic soils seem reasonable, given the low organic content of these soils (Table 1). Soils from Saglek generally had more culturable heterotrophs than did soils from Resolution Island. The Arctic soils typically had at least as many culturable heterotrophs which grew at 7°C as at 30°C, while the temperate-region soils typically had fewer culturable heterotrophs which grew at 7°C than at 30°C. For the Arctic soils, the low populations of culturable heterotrophs seem inconsistent with the relatively high potential for biphenyl mineralization. This discrepancy suggests that the fraction of the microbial population with the capability to degrade biphenyl was larger in the Arctic soils than in the temperate-region soils. This situation seems plausible for the PCB-polluted Arctic soils but is surprising for uncontaminated Arctic soils. One possible explanation for this situation would be the presence of

similar levels, in the soils of both regions, of natural substrates for biphenyl degraders, possibly plant-derived compounds. All attempts to enumerate microorganisms in the Arctic soils which can grow on biphenyl failed. When microorganisms

TABLE 3. Total viable counts at different incubation temperatures Soil

Total viable counts (CFU/g of dry soil) 6 SD at: 30°C

Saglek, 0 ppm PCBs Saglek, 5 ppm PCBs Saglek, 50 ppm PCBs Resolution Island, 0 ppm PCBs Resolution Island, 10 ppm PCBs Resolution Island, 50 ppm PCBs Garden Agricultural

7°C 3

9.7 3 10 6 9.7 3 10 1.1 3 105 6 7.7 3 103 2.8 3 105 6 5.0 3 104 1.8 3 104 6 1.3 3 103

2.4 3 10 6 6.7 3 104 1.5 3 105 6 1.7 3 104 3.1 3 105 6 2.5 3 104 1.8 3 104 6 5.8 3 103

9.5 3 104 6 6.9 3 104

7.7 3 104 6 1.9 3 104

1.4 3 104 6 4.2 3 103

2.5 3 105 6 3.5 3 104

1.1 3 108 6 3.5 3 107 1.3 3 107 6 4.9 3 106

1.4 3 107 6 2.5 3 106 4.2 3 106 6 8.4 3 105

4

5

VOL. 63, 1997

PCB BIODEGRADATION BY ARCTIC SOIL MICROORGANISMS

FIG. 4. Growth of enrichment cultures on biphenyl at various temperatures. Open symbols, growth rate; solid symbols, final optical density; squares, Sag-0; circles, Sag-50.

were extracted from the Arctic soils and grown on solid mineral medium with biphenyl, the number of colonies on control plates with no biphenyl added was the same as that on plates with biphenyl. When microorganisms were extracted from the Arctic soils, serially diluted, and used as inocula for liquid mineral medium with biphenyl as the sole organic substrate, no growth occurred. When microorganisms were extracted from the intermediately contaminated Arctic soil samples, serially diluted, and used as inocula for the same liquid mineral medium with biphenyl as well as 10% soil extract or 50 mg of yeast extract per liter, there was no conclusive evidence of growth on biphenyl in either of these media. Of the Arctic soils, only two yielded transferable enrichment cultures which grew at 7°C on biphenyl as the sole organic substrate. The two enrichment cultures were from uncontaminated Saglek soil (Sag-0) and highly contaminated Saglek soil (Sag-50). The two successful enrichment cultures were incubated at 7°C; surprisingly, none of the enrichment cultures incubated at 30°C was successful. After growth at 7°C through five serial transfers, both Sag-0 and Sag-50 contained organisms capable of growing on biphenyl at 7°C as well as at higher temperatures (Fig. 4). In both enrichment cultures, dominant cell morphologies varied markedly with temperature. It is noteworthy that these different morphotypes persisted in the enrichment cultures through the previous five serial transfers at 7°C. The different morphotypes are consistent with the above suggestion that biphenyl-mineralizing populations having different temperature optima coexist in the Arctic soils. The maximum yields (based on optical density) were similar for both enrichment cultures and occurred at 15°C. The maximum growth rates of both enrichment cultures occurred at 24°C, but organisms in Sag-50 grew much faster at higher temperatures than did those in Sag-0. Both enrichment cul-

3383

tures removed significant amounts of PCBs in Aroclor 1221 (not shown). The extent of removal and the pattern of congener removal were very similar for the enrichment cultures and for the isolates described below. Isolates capable of growth on biphenyl were obtained from both Sag-0 and Sag-50 enrichment cultures. There were a total of five distinct isolates based on colony and cell morphology (Table 4). The gram-negative isolates were psychrotolerant and grew at 30°C, while the gram-positive isolates were mildly psychrophilic and grew only below 22°C. The psychrotolerant isolates were both from the highly contaminated Saglek soil. If these psychrotolerant organisms are zymogenous (i.e., colonized the soil to exploit new nutrients), that would be consistent with the suggestion by Baross and Morita (3) that zymogenous populations in cold environments will tend to be psychrotolerant rather than psychrophilic. However, one might expect most microorganisms in Arctic surface soils to be psychrotolerant, since these soils may fluctuate greatly in temperature (7). Psychrophilic microorganisms are believed to be best adapted to permanently cold habitats (3, 19), where they can theoretically outcompete psychrotolerant microorganisms (17). Unfortunately, our inability to culture organisms from the other soil samples at 7°C or from any Arctic soil samples at 30°C precludes generalizations based on comparisons of culturable organisms. Of the isolates, the two gram-negative psychrotolerant strains from Sag-50 were capable of substantially degrading PCBs at 7°C (Fig. 5). These two isolates, Sag-50A and Sag50G, removed a total of 54 and 60% of Aroclor 1221, respectively. In tests with Aroclor 1242, these two isolates did not substantially remove any PCB congeners in addition to those removed from Aroclor 1221 (not shown). In the presence of PCBs, these two strains turned their medium yellow. The two isolates had similar patterns of congener removal (Fig. 5), and these patterns included removal of the same congeners removed in Saglek soil slurries (Fig. 3), with the exceptions that the isolates substantially removed 3,39-DCB and failed to remove 2,39,4-TCB. None of the gram-positive psychrophilic isolates substantially degraded PCBs at 7°C (data not shown). If such psychrophilic organisms were relatively more abundant when the Arctic soils were incubated at 7°C than when they were incubated at 30°C, this may help explain the reduced PCB biodegradation in the soils at the lower temperature. One of the psychrophilic isolates, Sag-50H, failed to grow on biphenyl in the presence of Aroclor 1221 or 1242, which is consistent with the above suggestion that PCBs may be inhibitory to some biphenyl degraders. Conclusion. Psychrotolerant PCB degrading-organisms may be useful for in situ PCB bioremediation in cold or temperate climates. The fact that the extent of PCB degradation at 7°C was greater in pure cultures (Fig. 5) than in soil slurries (Fig. 3) suggests that with an improved understanding of such organisms, more extensive biodegradation of PCBs in soil may be possible at low temperature. The psychrotolerant PCB-degrading organisms exhibited reasonably high potentials for degra-

TABLE 4. Characteristics of isolates grown on biphenyl Strain

Sag-0D Sag-0H Sag-50A Sag-50G Sag-50I

Cell type

Irregular rod Irregular rod Rod Rod Rod

Cell size (mm)

0.8–1.2 3 2–5 0.8–1.2 3 2–5 0.7 3 1–3 0.7 3 1–3 0.8 3 2–4

Motility

2 2 1 1 2

Growth at temp (°C):

Gram reaction

7

15

22

30

37

1 1 2 2 1

1 (1) 1 1 1

1 1 1 1 1

2 2 1 1 2

2 2 1 1 2

2 2 2 1 2

3384

MOHN ET AL.

APPL. ENVIRON. MICROBIOL.

FIG. 5. Degradation of Aroclors 1221 at 7°C by two isolates (n 5 2). Peaks are identified in Table 2. The cultures were incubated for 2 weeks.

dation of MCBs and DCBs. While such organisms alone would not be effective in bioremediation of most PCB mixtures, they could be useful in a sequential process in which PCBs are first reductively dechlorinated by anaerobic organisms (1, 5, 23). Also, the existence of these psychrotolerant PCB-degrading organisms suggests that similar organisms which are capable of degrading more highly chlorinated PCBs may exist. ACKNOWLEDGMENTS We thank Bianca Kuipers and Andrew Mosi for advice on PCB analyses, Gordon Stewart and Roshie Banisadr for technical assistance, and John Poland and Bianca Kuipers for soil samples. This work was supported by contributions from the Canadian Departments of Northern and Indian Affairs and of National Defense. REFERENCES 1. Abramowicz, D. A. 1990. Aerobic and anaerobic biodegradation of PCBs: a review. Crit. Rev. Biotechnol. 10:241–251. 2. Albro, P. W., J. K. Haseman, T. A. Clemmer, and B. J. Corbett. 1977. Identification of the individual polychlorinated biphenyls in a mixture by gas-liquid chromatography. J. Chromatogr. 136:147–153. 3. Baross, J. A., and R. Y. Morita. 1978. Microbial life at low temperatures: ecological aspects, p. 9–71. In D. J. Kushner (ed.), Microbial life in extreme environments. Academic Press, Ltd., London, United Kingdom. 4. Bedard, D. L., M. L. Haberl, R. J. May, and M. J. Brennan. 1987. Evidence for novel mechanisms of polychlorinated biphenyl metabolism in Alcaligenes eutrophus H850. Appl. Environ. Microbiol. 53:1103–1112. 5. Bedard, D. L., and J. F. Quensen III. 1995. Microbial reductive dechlorination of polychlorinated biphenyls, p. 217–241. In L. Y. Young and C. E. Cerniglia (ed.), Microbial transformation and degradation of toxic organic chemicals. Wiley-Liss, New York, N.Y. 6. Bedard, D. L., R. Unterman, L. H. Bopp, M. J. Brennan, M. L. Haberl, and C. Johnson. 1986. Rapid assay for screening and characterizing microorganisms for the ability to degrade polychlorinated biphenyls. Appl. Environ. Microbiol. 51:761–768. 7. Boyd, W. L., and J. W. Boyd. 1971. Studies of soil microorganisms. Inuvik, Northwest Territories. Arctic 24:162–176. 8. Bradley, P. M., and F. H. Chapelle. 1995. Rapid toluene mineralization by aquifer microorganisms at Adak, Alaska: implications for intrinsic bioremediation in cold environments. Environ. Sci. Technol. 29:2778–2781. 9. Bright, D. A., W. T. Dushenko, S. L. Grundy, and K. J. Reimer. 1996. The weathering and dispersal of polychlorinated biphenyls from a known source in the Canadian Arctic. Environ. Sci. Technol. 30:2661–2666. 10. Dushenko, W. T., S. L. Grundy, and K. J. Reimer. 1996. Vascular plants as sensitive indicators of lead and PCB transport from local sources in the Canadian Arctic. Sci. Total Environ. 188:29–38. 11. Erickson, M. D., J. S. Stanly, K. Turman, G. Radolovich, K. Bauer, J. Onstut, D. Rose, and M. Wickham. 1982. Analytical methods for by-product PCBs—preliminary validation and interim methods. Report EPA-560/5-82006, NTIS no. PB 83 127 696. Office of Toxic Substances, U.S. Environmental Protection Agency, Washington, D.C.

12. Frame, G. M., J. W. Cochran, and Bøwadt. 1996. Complete PCB congener distributions for 17 Aroclor mixtures determined by 3 HRGC systems optimized for comprehensive, quantitative, congener-specific analysis. J. High Resolut. Chromatogr. 19:657–668. 13. Frame, G. M., R. E. Wagner, J. C. Carnahan, J. F. Brown, Jr., R. J. May, L. A. Smullen, and D. L. Bedard. 1996. Comprehensive, quantitative, congener-specific analyses of eight Aroclors and complete PCB congener assignments on DB-1 capillary GC columns. Chemosphere 33:603–623. 14. Furukawa, K., K. Tonomura, and A. Kamibayashi. 1978. Effect of chlorine substitution on the biodegradability of polychlorinated biphenyls. Appl. Environ. Microbiol. 35:223–227. 15. Gounot, A. M. 1991. Bacterial life at low temperature: physiological aspects and biotechnological applications. J. Appl. Bacteriol. 71:386–397. 16. Grundy, S. L., D. A. Bright, W. T. Dushenko, and K. J. Reimer. 1995. Evidence for short-range transport of polychlorinated biphenyls in the Canadian Arctic using congener signatures of PCBs in soils. Sci. Total Environ. 160/161:251–263. 17. Harder, W., and H. Veldkamp. 1971. Competition of marine psychrophilic bacteria at low temperatures. Antonie Leeuwenhoek 37:51–63. 18. Harkness, M. R., J. B. McDermott, D. A. Abramowicz, J. J. Salvo, W. P. Flanagan, M. L. Stephens, F. J. Mondello, R. J. May, J. H. Lobos, K. M. Carroll, M. J. Brennan, A. A. Bracco, K. M. Fish, G. L. Warner, P. R. Wilson, D. K. Dietrich, D. T. Lin, C. B. Morgan, and W. L. Gately. 1993. In situ stimulation of aerobic PCB biodegradation in Hudson River sediments. Science 259:503–507. 19. Herbert, R. A. 1986. The ecology and physiology of psychrophilic microorganisms, p. 1–23. In R. A. Herbert and G. A. Codd (ed.), Microbes in extreme environments. Academic Press, Ltd., London, United Kingdom. 20. Ja ¨rvinen, K. T., E. S. Melin, and J. A. Puhakka. 1994. High-rate bioremediation of chlorophenol-contaminated groundwater at low temperatures. Environ. Sci. Technol. 28:2387–2392. 21. Kolenc, R. J., W. E. Inniss, B. R. Glick, C. W. Robinson, and C. I. Mayfield. 1988. Transfer and expression of a mesophilic plasmid-mediated degradative capacity in a psychrotrophic bacterium. Appl. Environ. Microbiol. 54:638–641. 22. Kotturi, G., C. G. Robinson, and W. E. Inniss. 1991. Phenol degradation by a psychrotrophic strain of Pseudomonas putida. Appl. Microbiol. Biotechnol. 34:539–543. 23. Mohn, W. W., and J. M. Tiedje. 1992. Microbial reductive dehalogenation. Microbiol. Rev. 56:482–507. 24. Mullin, M. D., C. M. Pochini, S. McCrindle, M. Romkes, S. H. Safe, and L. M. Safe. 1984. High-resolution PCB analysis: synthesis and chromatographic properties of all 209 PCB congeners. Environ. Sci. Technol. 18:468–476. 25. Reimer, K. J., D. A. Bright, and W. T. Dushenko. 1995. Effects of local and distant contaminant sources: polychlorinated biphenyls and other organochlorines in bottom-dwelling animals from an Arctic estuary. Sci. Total Environ. 160/161:265–283. 26. Swannell, R. P. J., K. Lee, and M. McDonagh. 1996. Field evaluations of marine oil spill bioremediation. Microbiol. Rev. 60:342–365. 27. Whyte, L. G., C. W. Greer, and W. E. Inniss. 1995. Assessment of the biodegradation potential of psychrotrophic microorganisms. Can. J. Microbiol. 42:99–106. 28. Williams, W. A. 1996. Low temperature microbial aerobic degradation of polychlorinated biphenyls in sediment, abstr. Q-191, p. 419. In Abstracts of the 96th General Meeting of the American Society for Microbiology 1996. American Society for Microbiology, Washington, D.C. 29. Williams, W. A. Personal communication.