War/d .~OLUG!/ of Microbio/ogy

& Biotechno/ogy

11, 271-279

Phenanthrene Pseudomonas M.A. Providenti,

mineralization sp. UGl4

C.W. Greet-,

by

H. Lee* and J.T. Trevors*

A phenanthrene-mineralizing Pseudmnontzs sp., designated UG14, was isolated from creosote-contaminated soil. It contained two plasmids, of approximately 77 kb and 76 kb, the smaller of which contained DNA sequences that hybridized with probes specific for ndoB and xylE, genes involved in catabolism of aromatic hydrocarbons. At initial phenanthrene concentrations of 10, 50, 200 and 1000 mg/l broth, 27%, 19%, 7.7% and 3.3%, respectively, after 36 days’ incubation at 3OT. Most ‘T-label was of the [9-Tlphenanthrene was recovered as ‘TO2 converted to a water-soluble metabolite tentatively identified as I-hydroxy-2-naphthoic acid. Rhamnolipid biosurfactants produced by P. uerugimsu UG2 enhanced mineralization of 50 mg phenanthrene/l by Pseudurnmus sp. UGl4. With the biosurfactant at 0, 25 and 250 mg rhamnose equivalents/l, 6.5%, 8.2% and 9.8%, respectively, of the phenanthrene was mineralized after 35 days. Key w&s:

Biodegradation,

biosurfactants,

PAH, phenanthrene,

Biological remediation of hydrocarbon-contaminated soil is an alternative to physical-chemical methods. Biodegradation of polycyclic aromatic hydrocarbons (PAH) has been researched and some work has indicated that bioremediation of PAH-contaminated soils is possible (Wang ef al. 1990; Mueller ef al. 1991, 1993; Lewis 1993). PAH are contaminants found in creosote and petroleum (Mueller et al. 1989; Weissenfels ef al. 1992) and characterized by their low aqueous solubility (Robotham & Gill l989), their tendency to sorb to soil and sediment (Means et al. 1980; Karickhoff 1981; Dzombak & Luthy 1984 and in some cases their mutagenic and carcinogenic properties (Cerniglia 1984, 1992). Some of the larger (more than three rings) and more toxic PAH persist in soil (Bossert 81 Bartha 1986) but bacterial strains capable of degrading some PAH (e.g. fluorene, fluoranthene, pyrene, benzo[a]pyrene, benzo[a]anthracene, chrysene) have been documented (Cerniglia 1992, 1993). Naphthalene and phenanthrene have been used as model compounds in PAH degradation studies. The bacterial oxidation pathways of these two compounds and other PAH M.A. Providenti, H. Lee and J.T. Trevors are with the Deoartment of Environmental Biology, University of Guelph, Guelph, Ontar&, NlG 2W1, Canada; fax: 519 637 0442. C.W. Greer is with the National Research Council Canada, Biotechnology Research Institute, 6100 Royalmount Awe, Montreal, Quebec, H4P 2FI2, Canada. ‘Corresponding authors. @ 1995 Rapid Communications

of Oxford

Ltd

Pse~&ncrras

sp.

have been reviewed (Cerniglia 1984, 1992). The genetics of naphthalene metabolism in Pseudomonas spp. have been studied and reviewed (Yen & Serdar 1988) and the genetics of phenanthrene metabolism have also been investigated. Possible involvement of plasmid in phenanthrene metabolism was suggested for a Beijerinkia sp. (Kiyohara et al. 1983), a A4ycobacferium sp. (Guerin & Jones 1988), and several Pseudomonas spp. with NAH7 or NAH7-like plasmids (Sanseverino ef al. 1993). The low bioavailability of many PAH may limit efficient biodegradation (Wodzinski & Coyle 1974; Stucki & Alexander 1987; Cemiglia 1993; Providenti er al. 1993). Research in our laboratory showed that rhamnolipid biosurfactants, produced by Pseudomonas aeruginosa UG2, enhanced removal of several structurally distinct hydrocarbons, including PAH such as phenanthrene, from soil (Van Dyke ef al. 1993a,b; Scheibenbogen ef al. 1994). The enhanced solubilization of PAH may increase their bioavailability and hence their biodegradation by specific PAH degraders. The present study was on the isolation and mineralization of phenanthrene by Pseudomonas sp. UGl4. In addition, the effect of rhamnohpid biosurfactants produced by P. aeruginosa UG2 on mineralization of phenanthrene was investigated. A paucity of information exists on the use of biosurfactants in biodegradation of compounds with low aqueous solubility.

M.A. Pronidenfi

Materials

et al.

and Methods

Chemicals and Media [9-14C]Phenanthrene (13.1 mCi/mmol) was purchased from Sigma. Two mineral salts media were used: UG2 medium and UGl4 medium. IJG2 medium was used to prepare stock solutions of biosurfactants and contained (g/l) KZHPOa, 0.65; KHZPOa, 0.17; MgSOa.7HZ0, 0.10; NaNOJ, 1.0; and NaCl, 0.50; and (ml/l): 0.010 M FeSOa.7HZ0, 1.0; and 0.020 M Car& solution, 1.0 (Van Dyke et al 1993a). IJGI4 medium was used in phenanthrene degradation experiments and contained (g/l 10 rnM NaZHPOJ KHZPOd buffer, pH 7.2): (NH&SOa, 2.38; MgSOa.7HZ0, 0.25; and NaCl, 0.50; and (ml/l): 0.010 M FeSOa.7HZ0, 0.2; and 20 rnM CaCIZ, 5.0 (Kiyohara ef al. 1982). S erial dilutions of cultures were made in sterile 0.85% (w/v) NaCl. Unless otherwise stated, materials and media were sterilized by autoclaving. Isolation of Phenanthrene-degrading Bacferia A mixed culture, designated T-l, able to mineralize [o-Ylphenanthrene, was used as the source of microorganisms. The mixed culture was isolated from creosote-contaminated soil from a former wood-treatment facility. It was initially enriched by growth on 200 mg phenanthrene/l and was kindly provided by MI. Van Dyke (Department of Environmental Biology, University of Guelph, Guelph, Ontario, Canada). The T-l mixed culture (500 ~1) was inoculated into a 250 ml Edenmeyer flask containing 50 ml sterile UGl4 medium amended with 1 g yeast extract (Difco)/l, I g peptone (Difco)/l and 200 mg phenanthrene/l. This subculture, designated TlS-1, was incubated at 23’C at 200 rev/min until microbial growth was evident, as judged by culture turbidity. Phenanthrene-degrading isolates were identified by the spray-plate technique of Kiyohara ef al. (1982) with some modifications. Briefly, IOOfil were removed from the TlS-1 culture and serially diluted. Two sets of UGl4 agar plates containing I, 0.1 or 0.01 g each of yeast extract and peptone were prepared. Aliquots (200j& of serially diluted culture were spread on each set of plates. After 10 min, one set was sprayed with a 10% (w/v) solution of phenanthrene in ether to select for degraders. The other set did not receive phenanthrene and served as a control to enumerate viable cells. The plates were incubated at 30’C. Plates without phenanthrene were incubated until growth occurred (24 to 48 h). Plates sprayed with phenanthrene were incubated until phenanthrene-degrading cultures appeared (7 to 14 days). Degradation was monitored by loss of fluorescence when sprayed plates were illuminated with U.V. light. Phenanthrene-degrading cultures were streaked to purity on UGl4 agar plates containing 10 rnr.4 disodium succinate and incubated at 3O’C until distinct colonies grew (24 to 48 h). To test whether these colonies maintained their phenanthrene-degrading ability, single colonies were transferred aseptically to UGl4 agar plates with a needle, the agar surfaces sprayed with phenanthrene and plates incubated until phenanthrene degradation was observed. Colonies selected for transfer were chosen randomly. When more than one colony morphology was observed, two or three representatives of each were transferred. Producfion and Purification of Pseudomonas aeruginosa UG2 Biosurfactants Purified P. aeruginosa UG2 rhamnolipids were prepared as described by Van Dyke et aI. (1993a), with some modifications. Briefly, 25 ml sterile tryptic soy broth (Difco) in a 125 ml screwcap Erlenmeyer flask were inoculated with a single UG2 colony growing on tryptic soy agar (TSA; Difco). The flask was incubated

272

World]oumal

of Microbiology

6 Biotechnology, Vol 11, 1995

at 3O’C for 18 h with gyratory shaking at I75 rev/min A 500-ml screw-cap Erlenmeyer flask containing 200 ml sterile UG2 medium amended with 2% (w/v) glucose was inoculated with 1 ml of the 18-h culture and incubated at 30’C with shaking at 175 rev/min. After 9 days, the culture was centrifuged at 5000 x g for IO min at 4-C to pellet cells. The supematant was removed, acidified with 5 M HCl to pH 2 to precipitate the rhamnolipid biosurfactants and centrifuged at 12,000 X g for 15 min at 4°C. The precipitate was dissolved in 20 ml 0.05 M NaHCOJ, pH 7.2, acidified as before, and extracted five times with 2 vol. ethyl acetate. The extracts were pooled and dried under NZ. Each sample was dissolved in 25 ml 0.05 M NaHCOs and the pH adjusted to 7.2 with 5 M NaOH. The solution was filtered aseptically through a sterile 0.22.pm pore nylon filter into a sterile bottle and stored at 4’C in the dark until used. The L-rhamnose content of the biosurfactants was determined calorimetrically by digesting diluted l-ml samples with 4.5 ml 90% (v/v) HzS04 for 10 min at 95’C in 13 mm x 125 mm Pyrex screw-cap test tubes using the thioglycolic acid reagent, as described by Chandrasekaran & BeMiller (1980). L-Rhamnose was used as the standard. The concentrations of the biosurfactants were expressed as mg rhamnose equivalents (RE).

Phenanfhrene Minerakuxfion in Mineral Salfs Medium Phenanthrene mineralization was tested in defined MS medium by measuring release of Y02 from [o-Ylphenanthrene (Bartha & Pramer 1965). The biometer flasks were modified from 250-ml Edenmeyer flasks, with a 1.5 cm diam. x 7 cm length glass cylinder fused to the centre inside bottom. YOZ traps (sterile, 7 ml, high-density polyethylene, scintillation vials containing 5 ml 2 M NaOH) were placed inside the glass cylinder and the flasks sealed with sterile neoprene stoppers during mineralization was followed experiments. Mineralization of ‘Y-phenanthrene by periodically removing the whole trap and replacing it with a new trap containing 5 ml fresh NaOH. To quantify the amount of 14COZ captured, 1 ml trap solution was transferred to a 20-ml polyethylene scintillation vial (Fisher Scientific) foilowed by 1 ml acetic acid to reduce chemiluminescence, and IO ml Scintiverse II scintillation cocktail (Fisher Scientific). Vials were sealed, mixed by shaking, held in the dark overnight, and the amount of Y02 counted using a Beckman LS6000 Scintillation System. Crystalline phenanthrene was added to SO ml sterile UGl4 medium to give 200 mg/l in the biometer flask. Inocula from overnight cultures of bacterial colonies grown at 3O’C on TSA were washed twice with sterile UGl4 medium, resuspended in 1 ml sterile MS medium and added to flasks at a density of 1 X lo6 cells/ml. A control flask containing 200 mg phenanthrenejl received no inoculum. Flasks were sealed with foam plugs and incubated at 23’C with shaking at 200 rev/min. After 3 days, when culture turbidity had increased, 50,000 d.p.m. of [9-YIphenanthrene were added, the flasks sealed with neoprene stoppers and incubated as described previously. Periodically, the amount of phenanthrene mineralized was determined. Based on the results from this initial screen, one pure culture, designated UGl4, was selected for study. The second experiment investigated the ability of UGl4 to mineralize four different concentrations of phenanthrene: 10, 50, 200 and 1000 mg/l, each tested in duplicate. Twenty-four hours before inoculation, biometer flasks containing SO ml UGl4 medium received 50,000 d.p.m. Y-phenanthrene and different amounts of unlabelled phenanthrene from concentrated stock solutions of phenanthrene dissolved in acetone. UGl4 inocula were prepared by transferring a single colony

E’henanfhrene wherahzfio~ by Pseudomonas from cerol 30’C

a TSA plate to UGl4 medium containing 1% (w/v) giyand 0.2% (w/v) yeast extract. The culture was incubated at with shaking at 200 rev/min until mid-log phase = 0.6). Two ml of the inoculum were added to each K)Dm biometer flask which was sealed and incubated at 3O’C and 175 rev/min shaking, Undegraded, particulate phenanthrene which gathered on the sides of the flasks was resuspended into the culture medium by vigorous shaking for several seconds twice daily, if necessary. Periodically, the amount of “COZ evolved was determined. In addition, the amount of ‘QZphenanthrene transfomed to a water-soluble form was monitored. Initially, this was done by extracting 2 to 5 ml of culture supematant with z vol. CHZCIZ and measuring the amount of radioactivity remaining in the aqueous phase, as described by Guerin & Jones (1988). Subsequently, it was found that centrifuging 1.5 ml of the culture at 12,100 X g for 5 min and determining the amount of radioactivity in I ml of supematant yielded similar results. Although this latter method gave slightly higher readings (by up to 5%), it avoided a lengthy extraction procedure and provided an indication of the amount of 14C-phenanthrene transformed to a water soluble form. The amount of radiolabel in supematant samples was measured using the same procedure as for the 14COL traps except that acetic acid was not added. The third mineralization experiment investigated phenanthrene mineralization in the UGl4 medium amended with various amounts of rhamnolipid biosurfactants produced by P. ueruginosu UG2. Biometer flasks containing 50 ml sterile UGl4 medium and 50 mg phenanthrene/l were prepared as in the second mineralization experiment. Twenty-four hours before inoculation, flasks received 100,000 d.p.m. [9-14C]phenanthrene followed by concentrated stock solutions of UG2 rhamnolipid biosurfactants to final concentrations of either 25 or 250 mg RE/l, Control flasks received 0.625 ml 0.05 M NaHCOJ, pH 7.2. Three flasks were prepared for each treatment. The UGl4 inoculum, pregrown in a phenanthrene-amended medium to induce phenanthrene-degrading activity, was prepared as follows: sterile UGl4 medium amended with 1% (w/v) glycerol and 0.2% (w/v) yeast extract was inoculated with a single UG14 colony from TSA, and incubated at 30’C with shaking at 175 rev/min until the ODeoO was 0.6. Several 500-m], screw-top, Erlenmeyer flasks, each containing 200 ml sterile UGl4 medium supplemented with 50 mg phenanthrene/l, were each inoculated with 16 ml of the growing culture and incubated for 48 h at 30’C with shaking at 175 rev/min. Undegraded ohenanthrene which gathered on the sides of the flasks “was resispended by vigorois shaking. This was done two or three times in the first 24 h but was unnecessary thereafter. The cultures were pooled, divided into 25ml aliquots and centrifuged at 5000 X g for 10 min. The supernatant was discarded and the cell pellet resuspended in 1 ml sterile HZ0 and added to biometer flasks. The initial cell density was about 1 X 109 c.fu./m~. Flasks were incubated at 22’C with shaking at 175 rev/min. Periodically, the amount of phenanthrene mineralized was determined as above.

Identificution of ihe Phenn&rene-mineralizing UGl4 Struin UGl4 was subjected to some standard taxonomic tests (Gerhardt 1981; Krieg &I Holt 1984). To confirm the identity of the culture, the microorganism was subjected to Biolog GN metabolic tests (Biolog, Hayward, CA) and analysed after 24 to 48 h incubation at 3O’C. The microorganism was also sent to Microbial ID, Inc. (Newark, Delaware, USA) and subjected to GC analysis of total cellular fatty acids as a means of identification.

Anuiysis of Wuter-soluble Metubolites Produced by UGl4 During Growth on Phenunthrene Water-soluble metabolites in supematants of UG14 cultures grown on phenanthrene were analysed by TLC and HPLC. Flasks containing 50, 200 or 1000 mg unlabelled phenanthrenejl were prepared, inoculated and incubated as described for the second mineralization experiment. At intervals, 2 ml samples were removed and centrifuged at 12,000 X g, The supematant was acidified with 30 ~1 1 M HCl and extracted three times with 2 vol. ethyl acetate. The extracts were dried under NZ and each dissolved in 0.25 ml ethyl acetate. For TLC, 10 to 20 ~1 of the extracts were applied to 0.1.mm silica gel sheets with fluorescent indicator (Eastman Kodak, Rochester, NY, USA) and developed in a solvent of hexane/CHCIJ/glacial acetic acid (10: 3 :2, by vol.). Authentic phenanthrene (Sigma), I-hydroxy-2.naphthoic acid (Aldrich), 1,2-dihydroxy naphthalene (Aldrich), salicylic acid (Sigma) and catechol (Fisher Scientific) served as standards. Ten ~1 of a solution containing 1 mg of each standard compound/ml ethyl acetate were applied in lanes adjacent to samples. Resolved compounds were visualized under u.v. light at 254 nm. For HPLC analysis, 50 ~1 extract dissolved in ethyl acetate were dried under NZ, redissolved in 1 ml methanol, and 10 to 20 ml injected into a Gilson Model 305 liquid chromatograph equipped with a IO-pm Partisil IO ODS 3 analytical column (1 mm X 250 mm), Samples were eluted at 24'C with methanol/ water (7 : 3, v/v), at a flow rate of 1 ml/min. Prior to use, the pH of the water was adjusted to 2.75 with HsPOa and the water and methanol were passed through a OL?-prn pore nylon filter and degassed for 18 h. Metabolites were detected at 254 nm using a Gilson Model 115 U.V. detector and chromatograms recorded with a Hewlett Packard 3394A integrator. Five to 20 ~1 of a methanol solution containing 1Opg of each standard (phenanthrene, I-hydroxy-2.naphthoic acid, 1,2-dihydroxy naphthalene, salicylic acid, and catechol)/ml were used as standards. Effect of Pseudomonas

aeruginosa

UG2

Rhumnolipids

on Growth

of

UG14

The effect of UG2 biosurfactants on growth of UGl4 was investigated on two solid media. The first medium was UGl4 medium amended with 1% (w/v) glycerol and 0.2% (w/v) yeast extract and solidified with 1.5% (w/v) agar. The second medium was the same as the first except that glycerol and yeast extract were at l/25 the concentration. UGl4 cells from a saline suspension of an overnight TSA streak were spread with a sterile cotton swab over an a!zar surface. Plates were left for 30 min to allow the agar to absorb the inoculum. Biosurfactants were presented to the microorganisms in two ways. The first method involved punching four, evenly spaced, 6.mm holes into the agar with a sterile cork borer and filling each well with 50 ~1 of biosurfactant solution. The second method involved placing four, sterile, 6-mm circular paper disks (Whatman) on the agar surface and applying 50 ~1 of biosurfactant solution to each disk. Plates were left undisturbed for 30 min to allow the biosurfactant solutions to absorb into the agar and then incubated at 30°C for 24 to 48 h, until growth was evident. Biosurfactant concentrations tested were 25, 100, 250 and 400 mg RE/l and 2, 20 and 42 mg RE/ml. Controls received 50 p/l 0.05 M NaHCOs, pH 7.2, containing no rhamnolipids. The ability of biosurfactant solutions to lyse sheep red blood cells was tested according to the method of Mulligan et al. (1984) and Jain et ul. (1991). Briefly, UGl4 medium agar plates containing 1% (w/v) glvcerol, 0.2% (w/v) veast extract and 5% (v/v) defibrinated sheeo n blood were ,oresented with biosurfactant solutions bv the two methods described above and treated in the same way as for the inoculated plates. Clear zones were measured after 24 h incubation.

World ~ownal of Mxmbiology

& Biotechnology, Vol 11, 1995

273

M.A. Prouidenfi et al. UC14 was probed for gene sequences that hybridized to .rylE and ndoB, genes involved in catabolism of aromatic compounds. LJG14 plasmid DNA was isolated by alkaline lysis, as described in Sambrook ef al, (1989). Genomic DNA was isolated by the miniprep protocol described in Ausubel er al. (1990). To further purify genomic DNA, samples were extracted after ethidium bromide treatment as described in Stemmer (1991). Restriction enzymes and digestion buffers were purchased from Boehringer Mannheim. Ail restriction digests of plasmid DNA were performed at 37’C for 4 to 12 h. Electrophoresis of DNA samples was performed in 0.8% (w/v) agarose in TAE buffer (40 rnM Trisjacetate, 1 rnM EDTA), pH 8.0, at 60 V wikh TAE as running buffer. DNA was visualized after staining gels for IS mm in a solution of OS pg ethidium bromide/ml. DNA was transferred from agarose gels to positively charged nylon membranes (Boehringer Mannheim) by vacuum blotting with a Hoefer TE-80 TransVac. Before vacuum blotting, agarose gels were depurinated for 20 min in 0.25 M HCl, rinsed in distilled water, denatured for 60 min in a 1.5 M NaCl and 0.5 M NaOH solution, and neutralized for 30 mm in a 1.5 M NaCt and I M Tris/ HCl solution, pH 7.4. DNA was vacuum blotted for 30 to 90 min in a transfer buffer consisting of 1.5 M NaCl and 0.15 M sodium citrate, pH 7. Nylon membranes were air dried for 15 min at 24’C and baked for I.5 min at 120°C to fix DNA. Template DNA used to prepare probes of the rqjE and rzdoB genes were prepared by PCR as described by Greer er al. (199.3). ndoB codes for one of the subunits of naphthalene dioxygenase (Kurkela ef al. 1988). The r&B gene probe was a 642-nucleotide (nt) fragment from within the coding sequence of the n&B gene from P. p&o NCIB9816 (Kurkela et a/. 1988). ry[E codes for metapyrocatechase. The xy/E gene probe was an 834~nt fragment from within the coding sequence of the +~yfEgene from I? p&&a mt-2 (ATCC 33015) carrying the TOL plasmid (Nakai et al. 1983). Probes labelled with 3zP-dATP were prepared by the rapid protocol described in the Multiprime Labelling Kit (Amersharn) used, except the mixture was incubated for 3 h and an additional 2 U Klenow enzyme were added halfway through the in~bation. unincorporated labels were removed by passing the reaction mixture through a Bio-Spin 6 column (Bio-Rad). Membrane probing was performed at high stringency (6S°C) according to the protocol provided with Zeta-Probe Membranes (Bio-Rad). Probed membranes were dried for IS min at YIZO’C and exposed to Kodak X-OMAT AR film for 8 h at - SO’C.

Results Screening for Phenanfhrerze-degrading Bacteria Cultures able to degrade phenanthrene were isolated using the spray-plate technique of Kiyohara ef al. (1982). Under visible light, phenanthrene degradation was noted by darkening of agar, which intensified as phenanthrene degradation proceeded. Under U.V. light, the darkened areas did not fluoresce and were distinguished from undegraded phenanthrene, which fluoresced brightly. Phenanthrene degradation on plates spread with diluted aliquots from liquid subcultures was evident within I week (data not shown). Three zones showing good phenanthrene-degrading activity were subcultured on MS medium supplemented with

10 rnM succinate for isolation of pure cultures. From each clear zones, 14 distinct colonies were tested for the ability to degrade phenanthrene on MS agar plates sprayed with phenanthrene. Four pure cultures which retained good phenanthrene-degrading ability were isolated after repeated subculturing in defined (MS agar amended with succinate or glucose) and undefined (TSA) media containing no phenanthrene. The four isolates were first screened qualitatively for the ability to mineralize 200 mg phenanthrene/l MS medium with phenanthrene as the sole carbon source. All four isolates mineralized phenanthrene to some extent. One culture, designated LJGl4, was observed to mineralize phenanthrene earlier and to a higher extent than the other three cultures, and was selected for further studies. Taxonomic Idenfificafion of Phenanfhrene-degrading llG14 Taxonomic tests showed UGl4 was a Gram-negative, catalase-positive, oxidase-positive, small rod. UG14 was motile, as determined by growth on motility agar, and produced a fluorescent pigment on King’s B agar after 24 h incubation at 22’C. It was unable to grow at or above 3?QC on TSA plates but did grow at 30, 24 and 4°C on TSA or in tryptic soy broth. UG14 was unable to hydrolyse starch, liquify gelatin or reduce nitrate. These tests indicated UG14 was ~se~dorno~s p~f~~a. Testing UGl4 with the Biolog GN system showed that it exhibited 70% to 80% similarity to P. pufida subgroup B. However, GC analysis of cellular fatty acids showed a 52% similarity between uG1.4 and P, anreofaciens. We have tentatively designated this strain as Pseudomonas sp. UGl4. Phen~nfhrene Minera~j~fion and Mef~~olife Producfion by pseudomonas sp, UGl4 The ability of pseudomonas sp. UGl4 to mineralize phenanthrene at four different initial concentrations was studied. The results indicated that, as the phenanthrene concentration decreased, the relative amount of radiolabel recovered as r*COr increased (Figure I). However, the total amount mineralized increased as phenanthrene concentration increased. After 36 days, the amount of phenanthrene mineralized with initial phenanthrene concentrations of 10, 50, 200 and 1000 mg/l was 2.7 (27%), 9.5 (19%), 15.4 (7.7%), and 33 mg/l (3.3%), respectively. The amount of radiolabelled phenanthrene transformed to water-soluble metabolite(s) was monitored. Under the experimental conditions used, a large amount of radiolabel remained in the culture medium (Figure I). At an initial phenanthrene concentration of 10 mg/l, about 65% of the added label was present in culture supematant after 1 day and the amount decreased to about 36% after 36 days. At 50 mg/l, about 65% was found in the supematant after 1 day and the amount decreased to about 50% after 36 days. At 200 mg/l, around 60% was found in the supematant after 1 day and the amount increased to 80% by day 36. At

Phenanthrene mineralization

0

10

20

by Pseudomonas

30

40

Time (days) Figure 2. Time course of mineralization of 50 mg phenanthrene/l by Pseudomonas sp. UG14 in MS medium amended with various amounts of P. aeruginosa UG2 rhamnolipid biosurfactants. No additional C-sources were added. Mineralization was measured by evolution of “CO* from radiolabelled phenanthrene. Fthamnolipid biosurfactant concentrations were 0 (a), 25 (A) or 250 (m) mg rhamnose equivalents/l. Values are means of three independent determinations. Statistically significant differences between treatments were calculated by the Student-Newman’s T-test (f = 0.05) on day 35. Differences are indicated by letters ta,b).

Time (days) Figure 1. Time course of phenanthrene mineralization and transformation by fseudomonas sp, UG14 at various initial phenanthrene concentrations in medium containing 0.04% (w/v) glycerol and 0.008% (w/v) yeast extract (A) Mineralization, as measured by evolution of “COz from radiolabelled phenanthrene. (B) Transformation, as measured by the amount of water-soluble radiolabelled metabolite(s) detected in culture supernatant. Initial phenanthrene concentrations were 10 (0) 50 (A), 200 (m) or lOOOmg/l (*). Values are means of two independent determinations.

1000 mg/l, around 30% was present after 1 day and the amount increased to about 70% by day 36. The identity of metabolite(s) produced was investigated by TLC and HPLC analysis of culture extracts. TLC analysis revealed one compound which co-migrated with Ihydroxy-2-naphthoic acid applied in the adjacent lane (data not shown). HPLC analysis revealed a major peak with a retention time matching that of I-hydroxy-2-naphthoic acid (data not shown). In addition, several minor peaks with retentions less than I-hydroxy-2naphthoic acid were detected in the chromatogram. The identity of these peaks was not investigated. Phenanthrene Minerakafion by Pseudomonas sp. UGl4 in Mineral Salts Medium amended wifh P. aeruginosa UG2 Biosurfactants The effect of P. aeruginosa UG2 rhamnolipid biosurfactants on phenanthrene mineralization by UGl4 was investigated. Increasing amounts of rhamnolipid biosurfactants enhanced

mineralization. After 35 days’ incubation in the containing SO mg phenanthrene/l and amended or 250 mg RE biosurfactants/l, 6.5 ? l.l%, and 9.8 & 0.7% (N = 3) was mineralized, (Figure 2). The rate and extent of mineralization phenanthrene/l in the absence of biosurfactants than those in the previous experiment.

MS medium with 0, 25

8.2 k 1.2% respectively of 50 mg were lower

Effecf of P. aeruginosa UG.2 Biosurfactanfs on fhe Growth of Pseudomonas sp. uG14 UG2 rhamnolipid biosurfactants were tested for their effect on the growth of Pseudomonas sp. UG14 cells using the diffusion method. No inhibitory zones were observed at any of the concentrations tested (25, 100, 250 and 400 mg RE/l and 2, 20 and 42 mg RE/ml) on either solid media tested. Haemolytic activity of the UG2 biosurfactant solutions was tested as a control. Haemolysis was observed only at the higher biosurfactant concentrations. At 20 and 42 mg RE/ml, 14 and 17 mm zones of lysis formed, respectively. Pseudomonas sp. UG14 DNA: Plasmid Sizing and Hybridimfion wifh xylE and ndoB Gene Probes Two large plasmids similar in size were detected in UG14. The larger plasmid was designated pEVB1 and the smaller pEVB2. pEVB1 was estimated to be 77 kb, as determined by co-migration of closed circular UGl4 plasmid DNA with the 77-kb plasmid of Escherichia coli strain Rl (Starodub & Trevors 1989) (Figure 3). Plasmid size was also estimated by comparing the logarithm of the electrophoretic mobility

World]oumuf

of Mimbiology

6 Bioteckmhgy,

Vol 1I, 1995

275

M.A. Providenti et al.

Figure 4. Southern blot of Pseudomonas sp. UG14 plasmid and genornic DNA probed with ndoB gene probe. (A) Ethidium bromide-stained agarose gel of UG14 plasmid and genomic DNA before vacuum blotting. 1--2 DNA digested with Hindlll (size standards); 2 to 5~plasmid DNA digested with BamHI, EcoRI, Hindlll and Pstl, respectively; 6~unrestricted plasmid DNA, which contained some contaminating genomic DNA near 23 kb; 7--unrestricted genomic DNA. Fragments which hybridized with the probe are indicated by arrows. (B) Southern blot of gel probed with 32P-labelled ndoB gene DNA. Lane designations are as in (A). Sizes of fragments which hybridized strongly are indicated. Undigested plasmid hybridized with the probe but was only visible upon extended exposure.

Figure 3. Ethidium bromide-stained agarose gel of plasmid DNA from Pseudomonas sp. UG14. 1--Unrestricted 83- and 77-kb plasmid DNA from Escherichia coil R1 (size standards). The plasmid migrating below the chromosomal DNA has not been characterized; 2--unrestricted plasmid DNA from UG14 and chromosomal DNA. Note that fragments resulting from Pstl digestion are more distinct in Figures 4 and 5. The larger plasrnid was designated pEVB1 and the smaller plasmid pEVB2; 3~Pstldigested plasmid DNA from UG14; chr---chromosomal DNA.

of restricted plasmid DNA fragments to the logarithm of standards of known size. Using this method, pEVB2 was 76 kb, as determined by the sum of fragments resulting from Psfl digestion (calculations not shown). Pstl restricted only the smaller plasmid, pEVB2 (Figure 3). UGI4 plasmid and genomic DNA were probed for the presence of gene sequences that hybridized to the ndoB and xylE genes. These gene sequences were found to be present in UG14 following probing of whole colony blots (data not shown). Genomic DNA, unrestricted plasmid DNA, and restricted plasmid DNA were probed. Genomic DNA did not hybridize with either the ndoB or xyleE gene probes (Figures 4 and 5). Plasmid DNA hybridized with both probes, restricted DNA samples more strongly than unrestricted DNA samples (Figures 4 and 5). In Figures 4 and 5, the undigested plasmid sample did not show hybridization with probes. However, upon extended exposure, a positive signal was present (data not shown). Sequences showing homology with the probes are present on pEVB2; neither

2 76

World]ournalof Microbiology & Biotechnology, Vo111, 1995

Figure 5. Southern blot of Pseudomonas sp. UG14 plasmid and genomic DNA probed with xylE gene probe. (A) Ethidium bromide-stained agarose gel of UG14 plasmid and genomic DNA prior to vacuum blotting. 1--2 DNA digested with Hindlll (size standards); 2 to 5~plasmid DNA digested with BamHI, EcoRI, Hindlll and Pstl, respectively; 6~unrestricted plasmid DNA, which contained some contaminating genomic DNA at 23 kb; 7--unrestricted genomic DNA. Fragments which hybridized with the probe are indicated by arrows. (B) Southern blot of gel probed with 'zP-labelled xylE gene DNA. Lane designations are as in (A). Sizes of fragments which hybridized strongly are indicated.

probe hybridized with pEVB1, which remained unrestricted and clearly separated from restricted pEVB2 following PstI digestion (Figure 3). Probing of BamHI, EcoRI, HindlII, and Psfl digests of plasmid DNA with the ndoB gene probe showed a strong signal on a 11.4-, 15.5-, 6.8- and 3.7-kb fragment, respectively, indicating that the hybridizing se-

Phenanfhrene mineralizafion quence maps to these fragments. Weaker signals were also visible on a I&6-kb and 9.4-kb Hi&III fragment and a 1.5kb PsfI fragment (Figure 4). xylE gene probing of the same digests showed a strong signal on a I7-kb, 17-kb, 7.2-kb and x.4-kb, and 2.9-kb fragment, respectively. A weaker signal was also visible on a 2.6-kb PsfI fragment (Figure 5). Neither gene mapped to the same fragment, for the restriction digests studied.

Discussion Bacterial cultures able to mineralize phenanthrene were isolated for use in biodegradation studies. Pseudomonas sp. UGl4 mineralized phenanthrene in a liquid medium. The majority of the phenanthrene was transformed to a watersoluble metabolite that retained the 14C-label. The major compound present was tentatively identified as l-hydroxy2.naphthoic acid, based on TLC and HPLC analyses. Initial transformation of phenanthrene to more water-soluble metabolite(s) was also observed by Guerin & Jones (1988) and Keuth & Rehm (1991). A factor which affected phenanthrene mineralization by UG14 is availability of alternate carbon sources. Although UGl4 can utilize phenanthrene as the sole carbon source, growth was extremely slow. Attempts to measure growth rate with phenanthrene as the sole carbon source have been unsuccessful. We included additional carbon sources (glycerol, yeast extract) during preparation of inocula for mineralization experiments. This allowed UGI4 to reach a high cell density and improved the rate and extent of phenanthrene mineralization. In the second mineralization experiment (Figure 1), additional carbon sources were included, whereas in the third mineralization experiment (Figure 2), additional carbon sources were only used during inoculum preparation. No additional C-sources were present during mineralization of phenanthrene. This may explain the differences mineralization of SO mg observed during phenanthrene/ml. The availability of phenanthrene to UG14 cells also appears to affect its mineralization, since rhamnolipid biosurfactants enhanced phenanthrene degradation by stationaryphase UGI4 cells. Previous work indicated that phenanthrene is only utilized in its dissolved state by a Pseudomonas sp. (Wodzinski & Coyle 1974). Furthermore, phenanthrene dissolution rates determined its utilization by a BeQerinkia sp. and a Flaoobacferium sp. (Stucki & Alexander 1987). UG2 rhamnolipid biosurfactants may enhance phenanthrene solubilization and therefore availability. Previous studies have shown that rhamnolipid biosurfactants enhanced octadecane dispersion and mineralization by P. aeruginosa strain ATCC 9027 (Zhang & Miller 1992) and hexadecane uptake by P. aeruginosu strain P201 (Koch ef a/. 1991). The critical micelle concentration (CMC) of rhamnolipids is approximately 30 to 50 mg/l (Zhang & Miller 1992; Van Dyke ef

&y Pseudomonas

al. I993a) and it is noteworthy that initial phenanthrene mineralization rates by Pseudomonas sp. UGl4 were similar in media containing rhamnohpids above and below the CMC (Figure 2). However, the extent of phenanthrene mineralized after 5 weeks was higher with rhamnolipids at 250 mg RE/l than in unamended medium (Figure 2). Pseudomonas sp. UGI~ was probed for the presence of gene sequences that hybridized to probes for ndoB and rylE. Genomic DNA did not hybridize with either probe, indicating that neither gene was present on the UG14 chromosome. These results corroborate existing knowledge that no convincing evidence for chromosomal location of NAH genes exists (Yen & Serdar 1988). Both genes were located on pEVB2, the smaller of the two plasmids in UG14. The ndoB gene, from a naphthalene-degrading plasmid, and the xy0! gene, from a toluene-degrading plasmid, may often be found together. In naphthalene-degrading bacteria, catechol degradation is encoded by nahH, which is homologous to xylE (Lehrbach ef al. 1983; Harayama ef al. 1987; Assinder & Williams 1988). Since DNA sequences from NAH, SAL and TOL plasmids coding for the mefacleavage pathway are homologous (Lehrbach ef al. 1983), probes against one of these genes may also hybridize with the other related isofunctional genes. The presence of DNA sequences in Pseudomonas sp. UGI~ sharing homology to ndoB and xylE is noteworthy. First, UG14 can grow on naphthalene in MS medium (data not shown). Second, the phenanthrene degradation pathway may converge with the naphthalene pathway at 1,2-dihydroxynaphthalene (Evans ef al. 1965). Naphthalene-degradative genes may be present in phenanthrene-degrading microorganisms. However, Foght & Westlake (1991) have shown that not all PAH-degrading microorganisms necessarily possess DNA capable of hybridizing with probes for TOL or NAH DNA. This may be due to the protocatechuate pathway employed by some microorganisms during degradation of phenanthrene (Kiyohara ef al. 1976; Kiyohara & Nagao 1978) and possibly other PAH. pEVB2 from Pseudomonas sp. UGI4 may be similar to the NAH7 plasmid and may be involved in phenanthrene metabolism. Recent work by Sanseverino ef al. (1%~) indicated that enzymes from the nah system, encoded on NAH7 (present in P. pfida G7) or NAH7-like plasmids such as pKA2 and pKA3 (present in naphthalene-degrading isolates Pseudomonas spp. DFC49 and DFC50, respectively), are involved in the metabolism of higher-molecular-weight PAH like phenanthrene. pEVB2 from Pseudomonas sp. UG14 contains DNA sequences which share homology with NAH7, pKA2, and pKA3, indicating that it has a similar role.

Acknowlegements This research was supported by the Ontario Ministry of Environment and Energy, the Institute for Chemical Science

M.A.

Providenfi

et al.

and Technology, and the Natural Sciences and Engineering Research Council of Canada (NSERC). The views and ideas expressed here are those of the authors and do not necessarily reflect the views and policies of the Ontario Ministry of Environment and Energy, nor does mention of trade names or commercial products constitute endorsement or recommendation for use. We thank M. Van Dyke for providing the phenanthrene-mineralizing mixed culture, J. Chan, R. Gold and C. Sopher for invaluable assistance and advice, and D. Hamilton for photographic services. MAP was the recipient of a NSERC postgraduate scholarship.

References Assinder, S.J. & Williams, P.A. 1988 Comparison of the rnefu pathway operons on NAH plasmid pWW60-22 and TOL plasmid pWW53-4 and its evolutionary significance. journal of General Microbiology 134, 276F2778. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D., Siedman, J.G., Smith, J.A. & Struhl, K. 1990 Currenf Profocols in Molecular Biology, Vol. I. New York: John Wiley. Bartha, R. & Pramer, D. 1965 Features of a flask and method for measuring the persistence and biological effects of pesticides in soil. Soil Science 100, 68-70. Bossert, I.D. & Bartha, R. 1986 Structure-biodegradability relationships of polycyclic aromatic hydrocarbons in soil. Bzdlefin of Enuironmen~a~ und Conkxminufion Toxicology 37, 49&495. Cemiglia, C.E. 1984 Microbial metabolism of polycyclic aromatic hydrocarbons. Advances in Applied Microbiology XI, 31-71. Cemiglia, C.E. 1992 Biodegradation of polycyclic aromatic hydrocarbons. Biodegradafion 3, 351-368. Cemiglia, C.E. 1993 Biodegradation of polycyclic aromatic hydrocarbons. Currenf Opinions in Biofechno!ogy 4, 331-338. Chandrasekaran, E.V. & BeMiller, J.N. 1980 Constituent analysis of glycosaminoglycans. Methods in Carbohydrafe Chemisty 8, 8+96. Dzombak, D.A. & Luthy, R.G. 1984 Estimating adsorption of PAHs on soils. Soil Science 137, 292-308. Evans, W.C., Fernley, H.N. & Griffiths, E. 1965 Oxidative metabolism of phenanthrene and anthracene by soil pseudomonads. Biochemisfy ]ournul95, 819-831. Foght, J.M. & Westlake, D.W.S. 1991 Cross hybridization of plasmid and genomic DNA from aromatic and polycyclic aromatic hydrocarbon degrading bacteria. Canadiun ]ournal of Microbiology 37, 924-932. Gerhardt, P. (ed) 1981 Mum& of Methods for General Backriology. Washington DC: American Society of Microbiology. Greer, C.W., Masson, L., Comeau, Y., Brousseau, R. & Samson, R. 1993 Application of molecular biology techniques for isolating and monitoring pollutant-degrading bacteria. Wafer Reseurch ]oumaj of Canuda 28, 275-287. Guerin, W.F. & Jones, G.E. 1988 Mineralization of phenanthrene by a Mycobucterium sp. Applied und EnDironmentul Microbiology 54,937-944. Harayama, S., Rekik, M., Wasserfallen, A. & Bairoch, A. 1987 Evolutionary relationships between catabolic pathways for aromatics: conservation of gene. order and nucleotide sequences of catechol oxidation genes of pWW0 and NAH7 plasmids. Molecular and General Genetics 210, 241-247. Keuth, S. & Rehm, H.J. 1991 Biodegradation of phenanthrene by Arfhrobacfer po1ychromogenes isolated from a contaminated soil. Applied Microbiology and Biofechnology 34, 804-808.

Kiyohara, H. & Nagao, K. 1978 The catabolism of phenanthrene and

naphthalene

by

bacteria.

]ournul

of Generu/

Microbiology

105,69-75. Kiyohara, H., Nagao, K. & Nomi, R. 1976 Degradation of phenanthrene through o-phthalate by an Aeromonas sp, AgricuIfuru/ and Biological Chemisfy 40, 1075-1082. Kiyohara, H., Nagao, K. & Yano, K. 1982 Rapid screen for bacteria degrading water-insoluble, solid hydrocarbons on agar plates. Applied and Environmental Microbiology 45, 454-457. Kiyohara, H., Sugiyama, M., Mondello, F.J., Gibson, D.T. & Yano, K. 1983 Plasmid involvement in the degradation of polycyclic aromatic hydrocarbons by a Beijerinkiu species, Biochemical and Biophysics! Research Communicufions 111, 939-945, Koch, A.K., Kappeli, O., Fiechter, A. & Reiser, J. 1991 Hydrocarbon assimilation and biosurfactant production in Pseudomonas aeruginosa mutants. journal of Bacteriology 173, 4212-4219. Krieg, N.R. & Holt, J.G. (eds) 1984 Bergey’s Munua/ of Sysfemufic Bucferiology, Vol. 1. Baltimore: William & Wilkins. Kurkela, S., Lehvaslaiho, H., Palva, E.T. & Teeri, T.H. 1988 Cloning nucleotide sequence and characterization of genes encoding naphthalene dioxygenase of Pseudomonas pufidu strain NCIB9816. Gene 73,355-362. Lehrbach, P.R., McGregor, I., Ward, J.M. & Broda, P. 1983 Molecular relationships between Pseudomonas INC P-9 degradative plasmids TOL, NAH, and SAL. Plasmid 10, 164-

174. Lewis, R.F. 1993 SITE demonstration of slurry-phase biodegradation of PAH contaminated soil. Journal of fhe Air, Wasfe, Wufer Management Association 43, 503-508. Means, J.C., Wood, S.G., Hassett, J.J. & Banwxt, W.L. 1980 Sorption of polynuclear aromatic hydrocarbons by sediments and soils. Environmenfal Science and Technology 14, 1524-

1528. Mueller, J.G., Chapman, P.J. & Pritchard, P.H. 1989 Creosotecontaminated sites: their potential for bioremediation. Environmenfal Science und Technology 23, 1197-1201. Mueller, J.G., Lantz, S.E., Blattmann, B.O. & Chapman, P.J. 1991 Bench-scale evaluation of alternative biological treatment processes for the remediation of pentachlorophenoland creosote-contaminated materials: slurry phase bioremediation. Environmenfal Science and Technology 25, 1055-1061. Mueller, J.G., Lantz, S.E., Ross, D., Calvin, R.J., Middaugh, D.P. & Pritchard, P.H. 1993 Strategy using bioreactors and specially selected microorganisms for bioremediation of groundwater contaminated with creosote and pentachlorophenol. Environmenfal Science and Technology 2 7, 691498. Mulligan, C.N., Cooper, D.G. & Neufeld, R.J. 1984 Selection of microbes producing biosurfactants in media without hydrocarbons. Journal of Fermenfafion Technology 62, 311-369. Nakai, C., Kagamiyama, H., Nozaki, M., Nakazawa, T., Inouye, S., Ebina, Y. & Nakazawa, A. 1983 Complete nucleotide sequence of the metapyrocatechase gene on the TOL plasmid of Pseudomonas puti& mt-2. Journul of Biological Chemistry 258, 29232928. Providenti, M.A., Lee, H. & Trevors, J.T. 1993 Selected factors limiting the microbial degradation of recalcitrant compounds. ]ournul of Induskial Microbiology 12, 37‘+395. Robotham, P.W.J. & Gill, R.A. 1989 Input and behaviour of petroleum hydrocarbons. In The Fufe and Effects of CXl in Freshwafer, eds Green, J. & Trett, MW. London: Elsevier Applied Science. Sambrook J., Fritsch, E.F. & Maniatis, T. 1989 Molecular Cloning: a Iaborafory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Phtmanfhrene Sanseverino, J., Applegate, B.M., King, J.M.H. & Sayler, GS. 1993 Plasmid-mediated mineralization of naphthalene, phenanthrene, and anthracene. Applied and Environmental Microbiology 59,1931-1937. Scheibenbogen, K., Zytner, R.G., Lee, H. & Trevors, J.T. 1994 Enhanced removal of selected hydrocarbons from soil by Pseudomonas aeruginosa UG2 biosurfactants and some chemical surfactants. Jotirna\ of &mica/ Technology and Biotechnology 59, 53-59. Starodub, M.E. & Trevors, J.T. 1989 Silver resistance in Escherichia cob RI. journal of Medical Microbiology 29, 101-110. Stemmer, W.P.C. 1991 A 20.minute ethidium bromide/high-salt extraction protocol for plasmid DNA, BioTechniques 10, 726. Stucki, G. & Alexander, M. 1987 Role of dissolution and solubility in biodegradation of aromatic compounds. Applied and Environmental Microbiology 53, 292-297. Van Dyke, MI., Couture, P., Brauer, M., Lee, H. & Trevors, J.T. 1993a Pseudomonas aeruginosa IJG2 rhamnolipid biosurfactants: structural characterization and their use in removing hydrophobic compounds from soil. Canadian Journal of Microbiology 39,1071-1078. Van Dyke, M.I., Gulley, S.L., Lee, H. & Trevors, J.T. 1993b Evaluation of microbial surfactants for recovery of hydrophobic

mineralization

by Pseudomonas

pollutants from soil. Journal of InduskiaI Microbiology 11, 163170. Wang, X., Yu, Y, & Bartha, R. 1990 Effect of bioremediation on polycyclic aromatic hydrocarbon metabolism residues in soil. Environmental Science and Technology 24, 1086-1089. Weissenfels, W.D., Klewer, H.J. & Langhoff, J. 1992 Adsorption of polycyclic aromatic hydrocarbons (PAHs) by soil particles: influence on biodegradability and biotoxicity. Applied Microbiology and Biotecbnotogy 36, 689496. Wodzinski, R.S. & Coyle, J.E. 1974 Physical state of phenanthrene for utilization by bacteria. Applied and Environmental Microbiology 27, 1081-1084. Yen, K.M. & Serdar, C.M. 1988 Genetics of naphthalene catabolism in Pseudomonads. CRC Critical Reviews in Microbiology 15, 247-268. Zhang, Y. & Miller, R.M. 1992 Enhanced octadecane dispersion and biodegradation by a Pseudomonas rhamnolipid surfactant (biosurfactant). Appked and Environmenta Microbiology 58, 32763282.

(Received

in

November

19%)

form

revised

World]oumal

of

17

Microb&8y

November

19%;

accepted

& Biofechmlogy, Vo/ 11, 1995

,?I

279