APPLIED AND ENVIRONMENTAL MICROBIOLOGY, June 1993, p. 1735-17410099-2240/93/061735-07$02.00/0Copyright © 1993, American Society for Microbiology

Development of Field Application Vectors for Bioremediationof Soils Contaminated with Polychlorinated Biphenylst

C. A. LAJOIE,l* G. J. ZYLSTRA,2 M. F. DEFLAUN,3 AND P. F. STROM4Center for Environmental Biotechnology, University of Tennessee, Knoxville, Tennessee 379321; Envirogen,Inc., Lawrenceville, New Jersey 08648 ; and Center for Agricultural Molecular Biology,2 Department ofEnvironmental Sciences, Cook College, and the New Jersey Agricultural Experiment Station,4 Rutgers

University, New Brunswick, New Jersey 08903

Received 20 November 1992/Accepted 15 March 1993

Field application vectors (FAVs), which are a combination of a selective substrate, a host, and a cloningvector, have been developed for the purpose of expressing foreign genes in nonsterile, competitive environmentsin which the gene products provide no advantage to the host. Such gene products are exemplified by theenzymes for the cometabolism of polychlorinated biphenyls (PCBs) through the biphenyl degradation pathway.Attempts to use highly competent PCB-cometabolizing strains in the environment in the absence of biphenylhave not been successful, while the addition of biphenyl is limited by its human toxicity and low watersolubility. Broad-substrate-specificity PCB-degradative genes (bphABC) were cloned from a naturally occur-ring isolate, Pseudomonas sp. strain ENV307, into broad-host-range plasmid pRK293. The resultingPCB-degrading plasmids were transferred to the FAV host Pseudomonas paucimobilis IIGP4, which utilizes thenontoxic, water-soluble, nonionic surfactant Igepal CO-720 as a selective growth substrate. Plasmid stability inthe recombinant strains was determined in the absence of antibiotic selection. PCB-degrading activity wasdetermined by resting cell assays. Treatment of contaminated soil (10, 100, or 1,000 ppm of Aroclor 1242) bysurfactant amendment (1.0% [wt/wtlIgepal CO-720 in wet soil) and inoculation with recombinant isolates ofstrain 1IGP4 (approximately 4 x 106 cells per g of soil) resulted in degradation of many of the individual PCBcongeners in the absence of biphenyl. Further improvements, including the use of non-antibiotic-resistancecloning vectors, addition of the bphD gene, and chromosomal integration of the PCB-degradative genes, mayultimately result in FAVs useful for both reactor-contained and in situ treatment of the partially dechlorinatedPCBs often found in contaminated soils and sediments.

Many potential applications of genetically engineered mi-croorganisms in environmental biotechnology involve intro-ducing genetic capabilities into nonsterile, competitive envi-ronments in which the genetic capabilities provide noadvantage to the host. This is exemplified by the cometabo-lism of polychlorinated biphenyls (PCBs) through the biphe-nyl degradation pathway. PCB contamination is widespreadas a result of industrial activities. The highly chlorinatedcongeners in commonly used Aroclors (commercial PCBmixtures) are persistent in aerobic environments.Workers have isolated many strains of microorganisms

which grow on biphenyl as a sole source of carbon andenergy and cometabolize many of the PCB congenerspresent in commercial Aroclors (3, 13). Degradation ismediated by four enzymes encoded by the bphABCD genes,which convert chlorobiphenyls to their corresponding chlo-robenzoic acids (12, 23). These organisms are unable todehalogenate the resulting ring fission products and cannotuse any of the highly chlorinated congeners as growthsubstrates (1, 13). However, other organisms are capable ofdehalogenating and mineralizing chlorobenzoic acids (24).The rate-limiting step in the metabolism of PCBs to CO2 inthe environment has been reported to be the initial cometa-bolic oxidation (11). Following stimulation of cometabolicactivity via biphenyl amendment, transient accumulations ofchlorobenzoic acids have been observed, and many of these

* Corresponding author.t Publication D-07531-1-93 of the New Jersey Agricultural Exper-

iment Station.

chlorobenzoic acids can be readily degraded by indigenousmicroorganisms (11, 15, 16, 18).Treatment of contaminated soils or sediments in situ has

the potential for considerable cost savings compared withexcavation and treatment in reactors. One difficulty withusing cometabolic approaches for either in situ or reactor-contained PCB degradation is that addition of biphenyl isproblematical because of its low water solubility and itshuman toxicity. In the absence of biphenyl, exogenousorganisms are unable to compete effectively with the highlyadapted indigenous species for alternative growth sub-strates, and also the biphenyl dioxygenase pathway is notfully induced (8, 20). Cloning of the PCB-degradative genesinto Eschenchia coli results in production of enzymes in theabsence of biphenyl induction (23). However, E. coli is notnative to soil and grows slowly at ambient temperatures.Additions of PCB-degrading microorganisms to competitivemicrobial environments without the development of a suit-able niche have not been shown to result in effectivebioremediation (15). Even with biphenyl addition, treatmentmight be limited because of the selective advantage ofnon-PCB-cometabolizing biphenyl utilizers.

Field application vectors (FAVs) have been developed forthe purpose of creating a temporary niche for the hostbacterial strain in such environments (21). This techniqueinvolves addition to the target environment of a selectivesubstrate that is readily utilizable by the host microorganismbut is unavailable to most indigenous species. The degrada-tive genes for a particular metabolic pathway are insertedinto this host, which functions to channel carbon and energyfrom the selective substrate into the production of the

1735

Vol. 59, No. 6

1736 LAJOIE ET AL.

TABLE 1. Bacterial strains and plasmids used in this study

Plasmid or strain Descriptiona Source or reference

PlasmidspUC19PCB E. coli, narrow host range, genes bphABCD, Apr This studypRK2013 Mating helper plasmid, Knr 10pRK293 Broad host range, Kr Tcr 9pGEM-7zf(-) Apr 24apCL1 pRK293 with two EcoRI sites, Kn' Tcr This studypCL2 Genes bphABC from pUC19PCB cloned into original EcoRI site This study

of pCL1pCL3 Genes bphABC from pUC19PCB cloned downstream of constitu- This study

tive kanamycin promoter in pCLlStrainsPseudomonas sp. strain Grows on biphenyl, broad substrate specificity, PCB degradation 26ENV307

P. paucimobilis 1IGP4 Grows on IGP, Kn' Tcs 21E. coli DH5a F- 480dlacZAM15 A(lacZYA-argF)U169 deoR recAl endAI 14

hsdRl7 (rk-, Mk+) supE44A- thi-1 gyrA96 relAlE. coli MC1061 F- araD139 A(ara-leu)7696 A(lac)X74 galUgalK hsdR2 (rk- Mk') 28

mcrBl rpsL (Smr)E. coli HB101 F- mcrB mrr hsdS20 (rB iB)recAm3supE44 ara-14galK 14

lacYl proA2 rpsL20 (Smr) xyl-5 A- leu mtl-la Abbreviations: Knr, kanamycin resistant; Kn', kanamycin sensitive; Tcr, tetracycline resistant; Tcs, tetracycline sensitive; Apr ampicillin resistant; Smr,

streptomycin resistant.

desired degradative enzymes. In this manner, inhibition ofgrowth of the exogenous strains due to competition with theindigenous microbial community is reduced.A previously developed FAV in which the nonionic sur-

factant Igepal CO-720 (IGP) was used as a nontoxic selectivesubstrate and Pseudomonaspaucimobilis 1IGP4 was used asthe host was able to grow in a competitive soil environment(the population increased from 2 x 106 to more than 109 cellsper g of soil), resulting in an increase in the presence ofnonadaptive tetracycline resistance marker genes (the pop-ulation of tetracycline-resistant organisms increased from 2x 106 to 4.3 x 108 cells per g of soil). Addition of the samestrain to unamended soil or soil amended with nonselectivesubstrates (tryptone, yeast extract, and glucose) did notresult in such growth (21). In this paper we describe thedevelopment of this FAV for the degradation of PCBsthrough addition of recombinant plasmids carrying genes forthe biphenyl metabolic pathway.

MATERIALS AND METHODS

Bacterial strains, plasmids, and culture conditions. Thebacterial strains and plasmids used in this study are de-scribed in Table 1. PCB-degrading strain ENV307, a Pseudo-monas sp. strain isolated from PCB-contaminated sediments(26), was grown on PAS medium (7), which consisted ofbasal mineral salts and yeast extract (0.05 g/liter) and wasamended with biphenyl. E. coli strains were grown onLuria-Bertani medium (10 g of tryptone per liter, 5 g of yeastextract per liter, 5 g of NaCl per liter) supplemented with0.1% glucose (LBG medium). LBG agar plates contained1.7% agar (Difco Laboratories, Detroit, Mich.). P. paucimo-bilis 1IGP4 was cultured in PAS medium amended with 0.2%IGP (Aldrich Chemical Co., Milwaukee, Wis.). IGP platescontained 1.7% Noble agar (Difco). When appropriate, theantibiotics tetracycline, kanamycin, and ampicillin wereused for selection at concentrations of 10, 50, and 50 ,ug/ml,respectively. Strain 1IGP4 and E. coli cultures were incu-bated at 25 to 28 and 37°C, respectively, in rotary shakerincubators or a laboratory rotator. Growth of strain 1IGP4

recombinants on biphenyl was tested by using inverted PASagar plates that had crystals of biphenyl placed in the petriplate lids.

Prior to mating, plasmid stability, and soil treatmentexperiments, individual strain 1IGP4 recombinants weregrown in PAS medium containing 0.2% IGP and tetracyclinein screw-cap tubes (16 by 125 mm) on a laboratory rotator toan optical density at 600 nm (OD600) of 0.5. The cultureswere centrifuged, washed twice, and resuspended in 0.05 Msodium phosphate buffer (pH 7.5) (22) or buffered dilutionwater (2) to an OD600 of 0.5.Recombinant DNA methods. Chromosomal DNA was iso-

lated and purified by a modification of the method of Beji etal. (5). Plasmids to be used in cloning procedures wereextracted from E. coli strains by using an alkaline-sodiumdodecyl sulfate method (19) and were purified by cesiumchloride-ethidium bromide density gradient centrifugation.Small-scale preparations of high-copy-number plasmids forrestriction mapping were obtained by performing a modifiedminilysis procedure (6). Low-copy-number plasmids forrestriction mapping were extracted by using a quick, large-scale plasmid preparation procedure (25). The techniquesdescribed by Maniatis et al. (22) were used for restrictionenzyme digestion and mapping of fragments. Restrictionendonucleases, T4 DNA ligase, and competent E. coli DH5acells were obtained from Bethesda Research Laboratories,Gaithersburg, Md. Digestions, ligations, and transforma-tions with plasmid DNA were performed according to man-ufacturer's instructions.

Plasmid construction. The PCB-degradative genes werecloned from strain ENV307. Large fragments (20 to 30 kb)from Sau3A partial digests of chromosomal DNA fromENV307 were ligated into the BamHI site of cosmid cloningvector pHC79 (Bethesda Research Laboratories, Gaithers-burg, Md.). The cosmids were then packaged into bacterio-phage X with an in vitro packaging system (Gigapak; Strat-agene, La Jolla, Calif.). The resulting A phage were used totransduce E. coli MC1061, and transductants were selectedon plates supplemented with ampicillin. Clones containingPCB-degradative genes were selected on the basis of their

APPL. ENvIRON. MICROBIOL.

FAVs FOR BIOREMEDIATION OF CONTAMINATED SOILS 1737

pUC19PCB bphA bphBC bphD Ap24.9 kbp Eg Bg x xx sI ~ ~~~~~~~~I, 11 Li,,. . I

E = E E E EE E E

pCLI /20.9 kbp E E/

Bg HOriV Kp _ Tc

OriV Kp OriV Kp,bphA EBgX

BgBEBg E Tc Bg BgEP;TC / / E bphAbphBC

pCL2 pCL3X 30.5 kbp 30.5 kbp bpBXxz;< bphBCX xXE ~~~~~~~~~~~H

TcFIG. 1. Construction of plasmids pCL2 and pCL3 by cloning

PCB-degradative genes bphABC from pUC19PCB into broad-host-range plasmid pCL1. Abbreviations for restriction sites: E, EcoRI;Bg, BglII; X, XhoI; H, HindIII; S, SstI. Abbreviations for genes:Tc, tetracycline resistance; Ap, ampicillin resistance; Kp, promoterof the kanamycin resistance gene; OriV, origin of replication. Theapproximate locations of bphABC in pUC19PCB are based oncomparisons with the restriction map of strain LB400 (23).

ability to convert 2,3-dihydroxybiphenyl (DHB) to the yel-low meta-cleavage product (23). The restriction fragmentscontaining the biphenyl-degradative genes were subclonedinto E. coli vector pUC19. Subclones with the highestactivity were selected on the basis of the results of anAroclor 1248 assay (3).An XhoI-HindIll fragment from the multiple cloning site

of pGEM-7zf(-) was cloned into the XhoI-HindIII region ofbroad-host-range plasmid pRK293, eliminating 0.5 kb fromthe kanamycin resistance gene. This plasmid was chosen onthe basis of its suitability for use in strain 1IGP4 (21). Theresulting plasmid (pCL1) (Fig. 1) contains two EcoRI sites,one of which is downstream of the kanamycin promoter.Plasmids pCL1 and pUC19PCB were partially digested withEcoRI, ligated (Fig. 1), and transformed into E. coli DH5a.Two strains harboring plasmids pCL2 and pCL3 (Fig. 1)were chosen for further study. The locations and orienta-tions of the PCB genes in these plasmid constructs weredetermined by restriction mapping.

Triparental matings. Strain 1IGP4, E. coli HB101 contain-ing the helper plasmid (pRK2013), and E. coli DH5ct isolatescarrying recombinant plasmids pCL1, pCL2, and pCL3 werespread on LBG agar plates at a cell concentration ratio ofapproximately 4:1:1, and the preparations were incubated atroom temperature. Control mating preparations containedonly strain 1IGP4 (no plasmids) and E. coli HB101 carryingplasmid pRK2013. After 24 to 48 h the cells were scrapedfrom the surface of the agar, suspended in dilution water,spread on plates containing PAS medium supplemented with1.0% IGP and tetracycline, and incubated at room tempera-ture until colonies appeared. Successful mating was indi-cated by colonies which turned yellow when they weresprayed with DHB.

Plasmid stability. The stabilities of the PCB-degradingplasmids in strain 1IGP4 were determined in the absence oftetracycline selection. For each strain 200 ,ul of culture wasinoculated into 5 ml of PAS medium containing 0.2% IGP in

a screw-cap culture tube (16 by 125 mm). The culture wasthen grown at room temperature on a laboratory rotator toan OD6. of 0.5. The initial and final concentrations of strain1IGP4 were determined by plate counts on PAS agar con-taining 0.2% IGP. The initial and final percentages of tetra-cycline-resistant colonies were determined by picking 100colonies of each strain from the plates containing PAS agarsupplemented with 0.2% IGP and transferring them to platescontaining PAS medium supplemented with 0.2% IGP andtetracycline. The percentages of the tetracycline-resistantcolonies maintaining the bphC gene were determined byspraying the plates containing PAS agar supplemented with0.2% IGP and tetracycline with DHB.

Resting cell assays. For resting cell assays (3) of PCB-degrading ability, ENV307 was grown in PAS mediumcontaining biphenyl as the carbon source. The strain 1IGP4isolates were grown in PAS medium containing 0.2% IGPand tetracycline.

Soil treatment experiments. Soil (Nixon loam) for PCBdegradation studies was obtained from a Cook Collegeagricultural field in New Brunswick, N.J. (21). The soil wasprepared by adjusting the pH from 6.8 to 7.0 with Ca(OH)2and air drying it to 40% of field capacity. A 20-g soilsubsample for each treatment was amended with Aroclor1242 in 2 ml of acetone, and the acetone was allowed toevaporate. This 20-g subsample was mixed with 24 g ofuncontaminated soil, and then inorganic macronutrients (169ppm of NH4Cl, 42 ppm of K2HPO4), 1.0% (wt/wt) IGP in wetsoil, and sufficient distilled water to achieve a final wetweight of 50 g at a moisture content equivalent to 80% offield capacity were added. The final Aroclor concentrationswere 10, 100, and 1,000 ppm of PCBs (10, 100, and 1,000mg/kg of wet soil).For each strain, 0.5 ml of culture was added to 50 g of

PCB-contaminated soil to achieve a soil concentration ofapproximately 4 x 106 tetracycline-resistant cells per g. The50-g soil samples were placed in 150-ml Erlenmeyer flasks,and the weights of the flasks plus soil were determined.Incubation was at room temperature. Once per day distilledwater was added to the flasks to achieve the original weights,and the soils were mixed with dedicated stainless steelspatulas. Duplicate 1-g samples were removed periodicallyfor up to 48 days for extraction and gas chromatographicanalysis.PCB extraction methods. PCBs were extracted from rest-

ing cell assay preparations by adding 3 ml of ether(nanograde; Baxter, McGaw Park, Ill.) and shaking thepreparations for 1 h on a reciprocating platform shaker. A1-ml ether sample was transferred from the upper solventlayer to a Teflon-capped tube (16 by 125 mm) containing 4.0ml of hexane and 0.5 g of silica gel (35-60 mesh). The tubeswere vortexed, shaken for 15 min on a reciprocating shaker,and centrifuged for 5 min at approximately 190 x g with aBeckman model TJ-6 centrifuge.

Soils were extracted by adding 1-g portions of soil to 25-mlglass centrifuge tubes with Teflon-lined caps and then adding1 g of anhydrous Na2SO4, 5 ml of distilled water, and 5 ml ofether to each tube. The tubes were shaken for 1 h on arecipricating shaker and centrifuged for 15 min at 760 x g.An ether subsample was then treated as described above.Gas chromatography analysis. The relative concentrations

of PCB congeners in extracts of resting cell assay mixturesand soils were determined by using a model GC-14A gaschromatograph (Shimadzu, Kyoto, Japan) equipped with amodel AOC-14 auto injector, as well as an electron capturedetector and a split/splitless injector, both of which were

VOL. 59, 1993

APPL. ENVIRON. MICROBIOL.

TABLE 2. Plasmid stability in strain 1IGP4 recombinants containing pCL1, pCL2, and PCL3

Cell Concn (107) % of tetracycline-resistant cellsa % of cells expressing bphC genebStrain

In initial inoculumc At OD600 of 0.5d In initial inoculum At OD600 of 0.5 In initial inoculum At OD60 of 0.5

lIGP4(pCL1) 2.5 87.3 70 51 0 0lIGP4(pCL2) 2.6 79.8 53 15 100 100lIGP4(pCL3) 2.4 74.0 82 58 100 100

a The percentage of tetracycline-resistant cells was determined by picking 100 colonies from plates containing PAS medium supplemented with 0.2% IGP andtransferring them to plates containing PAS medium supplemented with 0.2% IGP and tetracycline.

b Percentage of colonies on tetracycline-containing plates.c Inoculum grown in PAS medium containing 0.2% IGP and tetracycline.d After growth in PAS medium containing 0.2% IGP.

kept at 300°C. An SPB-1 capillary column (30 m by 0.25 mm[inside diameter]; Supelco, Inc., Bellefonte, Pa.) was usedwith nitrogen as the carrier gas (0.88 ml/min) and as themake-up gas (35 ml/min). Injection of 1- or 2-,ul samples wasperformed by using the splitless mode. The column oventemperature was kept at 40°C for 2 min, raised to 80°C at arate of 10°C/min and then to 225°C at a rate of 6°C/min, andthen kept at 225°C for 35 min (4).For resting cell assays individual congeners were quanti-

fied and identified by comparison of congener profiles withthe profiles reported by Bedard et al. (4), using peak 41(congeners 2,3,4,3',4' and 2,3,4,2',3',6') as an internal stan-dard. For soil treatment experiments peak 31 (congeners2,3,3',4' and 2,3,4,4') was chosen as an internal standardbecause its retention time, area, and resolution are similar tothose of the biodegradeable congener peaks and it is notdegraded by the PCB-degrading recombinants of strain1IGP4. The percentage of degradation (to the nearest 5%)was calculated by comparison with the results obtained withkilled controls in resting cell assays and by comparison withthe initial concentrations in soil treatment studies.

RESULTS

Plasmid stability. The results of the plasmid stabilitydeterminations are presented in Table 2. For all strains theinoculum, grown in PAS medium containing 0.2% IGP and10 ,ug of tetracycline per ml, did not consist entirely oftetracycline-resistant cells, even though no growth of parentstrain 1IGP4 was observed in the same medium containingtetracycline concentrations at least as low as 2.5 ,ug/ml inrange-finding studies. The initial percentage of tetracycline-resistant cells in the strain lIGP4(pCL2) inoculum was only53%, whereas the percentage was higher for the other twostrains, indicating that plasmid pCL2 is more unstable thanplasmid pCL1 or pCL3.

After approximately five divisions in the absence of selec-tion the percentage of tetracycline-resistant cells decreasedto 15 to 58% for the three strains. Spraying of the platescontaining PAS medium supplemented with 0.2% IGP withDHB yielded some colonies that did not turn yellow for allthree strains, indicating that the bphC gene was lost. How-ever, all tetracycline-resistant colonies did turn yellow whenthey were sprayed with DHB.

Degradative activity of recombinant organisms. None of thestrain 1IGP4 recombinants was capable of growth on biphe-nyl. Resting cell assays demonstrated that strains 1IGP4(pCL2) and lIGP4(pCL3) were able to degrade many of theindividual PCB congeners present in Aroclor 1242 (Table 3).No degradation was observed in resting cell assays per-formed with control strain lIGP4(pCL1). The extents ofdegradation of many of the individual congeners were no-

ticeably higher in lIGP4(pCL3) cultures than in 1IGP4(pCL2) cultures. Resting cell assay degradation by ENV307was more extensive than resting cell assay degradation byany of the recombinant 1IGP4 cultures. Restriction mapping(Fig. 1) of the original subclone (pUC19PCB) indicated thatthe PCB-degradative genes are closely related to the genesobtained from strain LB400 (23).

Soil treatment experiments. Addition of 1.0% IGP to soilcontaminated with 100 ppm of Aroclor 1242 and inoculationwith lIGP4(pCL3) resulted in a pattern of congener degra-dation similar to that observed in resting cell assays (Table3). Limited disappearance of only one peak (congeners 2,2'and 2',6) was observed in soil inoculated with control strain1IGP4(pCL1), and limited disappearance of only four peakswas observed in soil inoculated with lIGP4(pCL2).The rate of PCB degradation in soils inoculated with

lIGP4(pCL3) is indicated by the time course concentrationsof congeners 2',3,4 and 2,5,2',6' (peak 15) in soils contami-nated with 10, 100, or 1,000 ppm of Aroclor 1242 andamended with 1.0% IGP (Fig. 2).

DISCUSSION

The results of the laboratory scale treatment studiesindicate that effective concentrations of plasmid-encodedPCB-degradative gene products could be achieved in soil byusing FAVs, even though these products provide no benefitto the host in this environment. The FAVs are not expectedto derive any carbon and energy return from the degradationof the congeners present in Aroclor 1242. Strain 1IGP4 doesnot grow on biphenyl, and the use of only the first threegenes of the biphenyl degradation pathway (bphABC) in theplasmid constructs ensures, for demonstration purposes,that no energy is derived from even the lower chlorinatedcongeners.

It is important to note that the tetracycline resistancegenes are incorporated only as an aid in cloning experimentsand are not intended to impart a selective advantage on thehost strain in soil environments. The selective agent in theseFAVs is the surfactant. The use of tetracycline as a selectiveagent in soil treatment systems is impractical because of thepresence of resistant indigenous strains, cost, and regulatoryconstraints due to the possibility of a resulting increase in thepopulations of organisms which carry resistance to thismedicinally useful antibiotic.

Since neither tetracycline resistance nor PCB-degradingability is expected to provide a significant adaptive advan-tage to strain 1IGP4 in soil, there is probably no selection fororganisms maintaining either the individual genes or theplasmid. It is more likely that carbon and energy channeledinto cometabolic pathways result in some selection againstplasmid maintenance. The results of plasmid stability exper-

1738 LAJOIE ET AL.

FAVs FOR BIOREMEDIATION OF CONTAMINATED SOILS 1739

TABLE 3. Degradation of individual PCB congeners in Aroclor 1242 by Pseudomonas sp. strain ENV307 and 11GP4 recombinantscarrying plasmids pCL1, pCL2, and pCL3 in resting cell assays and soil treatment experimentsa

% Degradationd

Peak Congener(s)" % of Resting cell assays Soil treatment exptAroclor 1242CENV307 1IGP4 1IGP4 1IGP4 1IGP4 1IGP4 1IGP4(pCL1) (pCL2) (pCL3) (pCL1) (pCL2) (pCL3)

2 2,2' and 2,6 2.7 >90 0 55 65 20 30 753 2,4 and 2,5 0.9 >95 0 >80 >80 0 20 >904 2,3' 1.2 >95 0 >80 >85 0 25 855 2,3 and 2,4' 5.8 95 0 >95 >95 0 35 956 2,6,2' 0.9 >90 0 0 0 0 0 07 2,5,2' 8.7 90 0 75 95 0 0 658 2,4,2' and 4,4' 4.3 90 0 >95 >95 0 0 509 2,3,6 and 2,6,3' 0.7 >95 0 0 45 0 0 1510 2,3,2' and 2,6,4' 4.8 90 0 40 50 0 0 4511 2,5,3' 1.1 >95 0 55 >90 0 0 4012 2,4,3' 7.0 >90 0 0 >85 0 0 2513 2,5,4' 7.0 90 0 >95 >95 0 0 10014 2,4,4' 6.5 85 0 0 0 0 0 015 2',3,4 and 2,5,2',6' 3.3 90 0 80 90 0 0 6016 2,3,4' and 2,4,2',6' 0.9 85 0 0 40 0 0 017 2,3,6,2' 0.5 75 0 0 15 0 0 018 2,3,2',6' 0.5 >85 0 0 0 0 0 019 2,5,2',5' 3.1 85 0 80 >95 0 0 3520 2,4,2',5' 2.6 80 0 0 90 0 0 2521 2,4,2',4' 1.1 80 0 0 0 0 0 022 2,4,5,2' 1.4 45 0 55 >90 0 0 3523 2,3,2',5' 3.4 85 0 30 95 0 0 2524 3,4,4' and 2,3,2',4' 3.7 60 0 0 0 0 0 025 2,3,4,2' and 2,3,6,4' and 2,6,3'4' 3.0 75 0 0 15 0 0 026 2,3,2',3' 0.9 >90 0 25 85 0 0 2527 2,4,5,4' 1.9 0 0 0 0 0 0 028 2,5,3',4' 3.9 70 0 20 >95 0 0 3029 2,4,3',4' and 2,3,6,2',5' 4.8 55 0 0 0 0 0 030 2,3,6,2',4' 1.3 20 0 0 20 0 0 031 2,3,3',4' and 2,3,4,4' 2.9 40 0 0 0 0 0 032 2,3,6,2',3' and 2,3,5,2',5' 1.0 65 0 0 25 0 0 033 2,3,5,2',4' and 2,4,5,2',5' 1.1 70 0 0 60 0 0 034 2,4,5,2',4' 1.0 60 0 0 0 0 0 035 2,4,5,2',3' and 2,3,5,6,2',6' 1.0 0 0 0 0 0 0 036 2,3,4,2',5' 1.0 65 0 0 15 0 0 037 2,3,4,2',4' 0.7 0 0 0 0 0 0 038 2,3,6,3',4' and 3,4,3',4' 1.1 65 0 0 0 0 0 039 2,3,4,2',3' 0.7 75 0 0 0 0 0 040 2,3,6,2',4',5' and 2,4,5,3',4' 0.7 0 0 0 0 0 0 041 2,3,4,3',4' and 2,3,4,2',3',6' 1.0 0 0 0 0 0 0 0

a In the resting cell assays the Aroclor 1242 concentration was 10 ppm, and preparations were incubated with shaking (100 rpm) at room temperature for 48h. In the soil treatment experiments the Aroclor 1242 concentration was 100 ppm, the IGP concentration was 1.0%, and the preparations were incubated at roomtemperature for 48 days.

b Data were based on a comparison with the results of Bedard et al. (4) (we used the same gas chromatography conditions). Boldface type indicates the majorcongeners in the peaks.

Percentage (by weight) of Aroclor 1242 represented by the congeners present in each peak (modified from the study of Bedard et al. [4]).d Determined by comparison with perchloric acid-killed controls (resting cells assays) or with initial soil concentrations (soil treatment experiments). Levels

of apparent degradation less than 15% are not reported. Reductions in the congener concentrations to values below the detection limits are calculated by usingthe gas chromatogram peak area unit lower limit of 1,000; i.e., percent degradation = > [(initial peak area - 1,000)/initial peak area] x 100.

iments indicated that cloning of the PCB-degradative genesinto the original EcoRI site of plasmid pRK293 (constructpCL2) resulted in a decrease in stability. Cloning the samegenes downstream of the promoter of the kanamycin genedid not result in a decrease in stability, indicating that thePCB genes themselves do not necessarily result in a de-crease in plasmid maintenance (despite the increase in size).The fact that all tetracycline-resistant isolates maintained anactive bphC gene suggests that losses of both activities aredue to segregative plasmid loss rather than to gene deletions.Although plasmid stability was sufficient to effect PCB

degradation in these experiments, the loss of the desired

nonadaptive genetic capabilities in FAVs has the potential tobe a significant problem in some applications. The selectivesubstrate is chosen to minimize competition with indigenousorganisms (21). If plasmid loss rates are high, competitionbetween functional FAV hosts and FAV hosts that have lostthe inserted genes may be a more serious difficulty thancompetition with indigenous strains. This could be especiallyimportant if the desired nonadaptive pathways exert sub-stantial demands on the organism. In these cases, loss of theinserted genes could provide a considerable selective advan-tage over plasmid maintenance. Chromosomal insertion ofthe foreign genes via transposons or homologous recombi-

VOL. 59, 1993

APPL. ENVIRON. MICROBIOL.

100

r 80

0 60

Z 40

X 20

00 4 8 12 16

TIME (days)20 24

FIG. 2. Percentages of degradation of congeners 2',3,4 and2,5,2',6' (peak 15) in soils contaminated with 10, 100, or 1,000 ppmof Aroclor 1242. Soils were amended with 1.0% IGP (surfactant) andinoculated with strain lIGP4(pCL3) (2.8 x 106 cells per g of soil).

nation may increase stability, resulting in improved FAVperformance.

In resting cell assays, degradation of PCB congeners bystrain ENV307 was more extensive than the degradationobserved with strain 1IGP4 (Table 3). Achieving higherlevels of gene expression in strain 1IGP4 by using strongerpromoters may result in more extensive PCB degradation. Instudies with strain LB400, growth on biphenyl resulted indegradation of individual congeners that were not degradedby the same strain grown under noninducing conditions (23).The importance of plasmid stability and gene expression infield applications is suggested by the results of a comparisonof the performance of lIGP4(pCL2) and the performance oflIGP4(pCL3) in soil treatment experiments (Table 3).A major advantage of FAVs over naturally occurring

PCB-degrading strains is the elimination of the requirementfor biphenyl, with all its associated limitations and environ-mental problems. The potential ability of the surfactant toenhance the bioavailability of contaminants may be anadditional advantage and may be more effectively exploitedby using soil-slurry reactors.

General safety issues concerning the use of FAVs havebeen discussed previously (21). The use of non-antibiotic-resistance cloning vectors (17) and chromosomal integrationof the PCB genes would probably be required for regulatoryapproval of field releases. PCB-degradative genes (bphABC)from a naturally occurring strain (Pseudomonas sp. strainENV307) have been inserted into a soil bacterium, P.paucimobilis 1IGP4, which neither possesses the ability togrow on biphenyl nor degrades PCBs. Addition of the bphDgene for actual field applications is warranted. The cloning ofPCB genes per se into strain 1IGP4 does not appear towarrant any serious safety concern. In fact, another strain ofP. paucimobilis (strain Q1) has been shown to cometabolizePCBs by the biphenyl degradation pathway, although itsbphC gene is not closely related to that of the strain 1IGP4recombinants (12, 27).

ACKNOWLEDGMENTSWe thank Laura Meagher of the Center for Agricultural Molecular

Biology, Rutgers University, Gary Sayler and Bruce Applegate of

the Center for Environmental Biotechnology, University of Tennes-see, and Randy Rothmel of Envirogen, Inc., for their technicalassistance. We also thank Michael Shannon of Envirogen, Inc., forperforming resting cell assays with strain ENV307.Funding for this work was provided by the New Jersey Depart-

ment of Environmental Protection and Energy, Envirogen, Inc., andthe New Jersey Agricultural Experiment Station (supported by statefunds). Additional support was provided by the Waste ManagementResearch and Education Institute (University of Tennessee), theElectric Power Research Institute, and the Center for AgriculturalMolecular Biology.

REFERENCES

1. Ahmed, M., and D. D. Focht. 1973. Degradation of polychlori-nated biphenyls by two species of Achromobacter. Can. J.Microbiol. 19:47-52.

2. American Public Health Association. 1985. Standard methods forthe examination of water and wastewater, 16th ed. AmericanPublic Health Association, Washington, D.C.

3. Bedard, D. L., R. Unterman, L. H. Bopp, M. J. Brennan, M. L.Haberl, and C. Johnson. 1986. Rapid assay for screening andcharacterizing microorganisms for the ability to degrade poly-chlorinated biphenyls. Appl. Environ. Microbiol. 51:761-768.

4. Bedard, D. L., R. E. Wagner, M. J. Brennan, M. L. Haberl, andJ. F. Brown. 1987. Extensive degradation of Aroclors andenvironmentally transformed polychlorinated biphenyls by Al-caligenes eutrophus H850. Appl. Environ. Microbiol. 53:1094-1102.

5. Beji, A., D. Izard, F. Gavini, H. Leclerc, M. Leseini-Delstanche,and J. Krembel. 1987. A rapid chemical procedure for isolationand purification of chromosomal DNA from gram-negativebacilli. Anal. Biochem. 162:18-23.

6. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extractionprocedure for screening recombinant plasmid DNA. NucleicAcids Res. 7:1513-1523.

7. Bopp, L. H. 1986. Degradation of highly chlorinated PCBs by aPseudomonas strain LB400. J. Ind. Microbiol. 1:23-29.

8. Brunner, W., F. H. Sutherland, and D. D. Focht. 1985. En-hanced biodegradation of polychlorinated biphenyls in soil byanalog enrichment and bacterial inoculation. J. Environ. Qual.14:324-328.

9. Ditta, G., T. Schmidhauser, E. Yakobson, P. Lu, X.-W. Liang,D. R. Finlay, D. Guiney, and D. R. Helinski. 1985. Plasmidsrelated to the broad host range vector, pRK290, useful for genecloning and for monitoring gene expression. Plasmid 13:149-153.

10. Figurski, D. H., and D. R. Helinski. 1979. Replication of anorigin-containing derivative of plasmid RK2 dependent on aplasmid function provided in trans. Proc. Natl. Acad. Sci. USA76:1648-1652.

11. Focht, D. D., and W. Brunner. 1985. Kinetics of biphenyl andpolychlorinated biphenyl metabolism in soil. Appl. Environ.Microbiol. 50:1058-1063.

12. Furukawa, K., N. Hayase, K. Taira, and N. Tomizuka. 1989.Molecular relationship of chromosomal genes encoding biphe-nyl/polychlorinated biphenyl catabolism: some soil bacteriapossess a highly conserved bph operon. J. Bacteriol. 171:5467-5472.

13. Furukawa, K., N. Tomizuka, and A. Kamibayashi. 1983. Meta-bolic breakdown of Kaneclors (polychlorobiphenyls) and theirproducts by Acinetobacter sp. Appl. Environ. Microbiol. 46:140-145.

14. Hanahan, D. 1983. Studies on transformation of Escherichia coliwith plasmids. J. Mol. Biol. 166:557-580.

15. 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.

1740 LAJOIE ET AL.

FAVs FOR BIOREMEDIATION OF CONTAMINATED SOILS 1741

Lin, C. B. Morgan, and W. L. Gately. 1993. In situ stimulationof aerobic PCB biodegradation in Hudson River sediments.Science 259:503-507.

16. Hernandez, B. S., F. K. Higson, R. Kondrat, and D. D. Focht.1991. Metabolism of and inhibition by chlorobenzoates inPseudomonasputida P1ll. Appl. Environ. Microbiol. 57:3361-3366.

17. Herrero, M., V. De Lorenzo, and K. N. Timmis. 1990. Transpo-son vectors containing non-antibiotic resistance selection mark-ers for cloning and stable chromosomal insertion of foreigngenes in gram-negative bacteria. J. Bacteriol. 172:6557-6567.

18. Hickey, W. J., and D. D. Focht. 1990. Degradation of mono-, di-,and trihalogenated benzoic acids by Pseudomonas aeruginosaJB2. Appl. Environ. Microbiol. 56:3842-3850.

19. Ish-Horowitz, D., and J. F. Burke. 1981. Rapid and efficientcosmid cloning. Nucleic Acids Res. 9:2989-2998.

20. Kohler, H. P. E., D. Kohler-Staub, and D. D. Focht. 1988.Cometabolism of chlorinated biphenyls: enhanced transforma-tion of Aroclor 1254 by growing bacterial cells. Appl. Environ.Microbiol. 54:1940-1945.

21. Lajoie, C. A., S.-Y. Chen, K.-C. Oh, and P. F. Strom. 1992.Development and use of field application vectors to expressnonadaptive foreign genes in competitive environments. Appl.Environ. Microbiol. 58:655-663.

22. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

23. Mondello, F. J. 1989. Cloning and expression in Escherichia coliof Pseudomonas strain LB400 genes encoding polychlorinatedbiphenyl degradation. J. Bacteriol. 171:1725-1732.

24. Pettigrew, C. A., A. Breen, C. Corcoran, and G. S. Sayler. 1990.Chlorinated biphenyl mineralization by individual populationsand consortia of freshwater bacteria. Appl. Environ. Microbiol.56:2036-2045.

25. Promega. 1992. Technical manual 16. Promega, Madison, Wis.26. Sharma, A., C. D. Chunn, R. K. Rothmel, and R. Unterman.

1991. Studies on bacterial degradation of polychlorinated biphe-nyls: optimization of parameters for in vivo enzyme activity,poster Q48. 91st Gen. Meet. Am. Soc. Microbiol. 1991.

27. Taira, K., N. Hayase, N. Arimura, S. Yamashita, T. Miyazaki,and K. Furukawa. 1988. Cloning and nucleotide sequence of the2,3-dihydroxybiphenyl dioxygenase gene from the PCB-degrad-ing strain of Pseudomonas paucimobilis Ql. Biochemistry 27:3990-3996.

28. Wertman, K. F., A. R. Wyman, and D. Botstein. 1986. Host/vector interactions which affect the viability of recombinantphage lambda clones. Gene 49:253-262.

VOL. 59, 1993