INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Bioremediation is a technology in which biological systems are harnessed to clean up environmental pollutants. This technology has the potential to be less polluting, less expensive, and less invasive than others for destroying pollutants. Among the biological systems with potential for bioremediation, microorganisms are especially promising due to their catabolic diversity and their plasticity. One factor that can limit the efficacy of bioremediation strategies is the inability of existing microbial catabolic activities to degrade the target pollutant (34). Failure to degrade a target pollutant can arise from the inability of catabolic enzymes to transform the pollutant or its metabolites, as well as from the inhibition of these enzymes by metabolites. Such catabolic blocks are particularly problematic for structurally diverse pollutants, such as polychlorinated biphenyls (PCBs), whose occurrence in the biosphere has only recently become widespread.

PCBs were produced over a 50-year period for a wide variety of industrial uses. Although their production has been banned in the industrial world for over 20 years, the destruction of PCBs is of continued relevance due to their pervasiveness and persistence, their impact on fragile ecosystems, and human health concerns (24). A number of microorganisms are able to transform some of the congeners found in commercial formulations of PCBs. This transformation is effected aerobically by a pathway encoded by the bph gene cluster, the first four enzymatic activities of which transform biphenyl via a catecholic intermediate to benzoate and a pentanoate derivative (14). Among PCB-degrading strains, Rhodococcus globerulus P6 (variously identified as Acinetobacter sp. strain P6, Corynebacterium sp. strain MB1, and Arthrobacter sp. strain M5 [3]) and Burkholderia sp. strain LB400 (originally identified as Pseudomonas sp. strain LB400 [6]) transform a particularly large number of congeners. However, even these strains do not effectively mineralize lightly chlorinated congeners and transform but a few highly chlorinated congeners (5, 15, 31). The development of strains to effectively mineralize mixtures of PCBs requires the identification of recalcitrant or inhibitory catabolites and the development of strategies to degrade them.

2-Hydroxy-6-oxo-6-phenylhexa-2,4-dienoate (HOPDA) hydrolase BphD (EC 3.7.1.8) has been identified as a key determinant in the aerobic degradation of PCBs in Burkholderia sp. strain LB400 (29). This /-fold serine hydrolase catalyzes the fourth step of the bph pathway, hydrolyzing HOPDA at a carbon-carbon bond to yield 2-hydroxypenta-2,4-dienoate and benzoate (Fig. 1) (27). Recent studies (29) with monochlorinated HOPDAs have established that 3-Cl and 4-Cl HOPDAs competitively inhibit BphD of strain LB400 (BphD_LB400). Moreover, HOPDAs undergo a spontaneous transformation to acetophenones, although 3-Cl HOPDA is much more stable than 4-Cl HOPDA. Consistent with these observations, HOPDAs and chloroacetophenones accumulate in the culture medium when Burkholderia sp. strain LB400 is incubated with congeners that are predicted to give rise to 3-Cl and 4-Cl HOPDAs, respectively (6, 13, 27).

View larger version (6K): [in this window] [in a new window]   FIG. 1.   Hydrolysis of HOPDA catalyzed by BphD.

It is unclear whether the specificity of BphD_LB400 represents that of other BphDs. For example, R. globerulus P6 transforms 4,4'-dichlorobiphenyl (4,4'-diClB) to an unidentified monochlorinated benzoate (15). As the biotransformation of 4,4'-diClB is expected to proceed via the formation of 3,10-diCl HOPDA, this observation suggests that BphD from R. globerulus P6 (BphD_P6) hydrolyzes 3-Cl HOPDAs more specifically than BphD_LB400 does. Interestingly, BphD_P6 shares only 42% sequence identity with BphD_LB400 (22). This is significantly less than the sequence identity between the other respective homologous enzymes of the bph pathway in these strains, which is between 52 and 61% (2). The study of evolutionary divergent BphDs is critical to establish structure-function relationships in this family of hydrolases and may also identify enzymes useful for the effective microbial degradation of PCBs.

We report here the heterologous expression, purification, and characterization of BphD_P6. The specificity of the enzyme for monochlorinated HOPDAs was established by steady-state kinetic studies and was compared to that of BphD_LB400. The specificity of both enzymes for two dichlorinated (diCl) HOPDAs was also determined. The specificities of the respective hydrolases are discussed in terms of their implications for the microbial degradation of PCBs.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. 4-Chloro-2,3-dihydroxybiphenyl (4-Cl DHB) was synthesized as described previously (26). All other chlorinated DHBs were synthesized and isolated as described elsewhere (20). Restriction enzymes and PfuI polymerase were from Amersham Pharmacia Biotech (Baie d'Urfe, Quebec, Canada) and Stratagene, respectively. All other chemicals were of analytical grade.

Bacterial strains and plasmids. Strains used for protein expression or DNA propagation included Escherichia coli DH5 (17), BL21 (DE3) (33), and LE392 (7) and Pseudomonas putida KT2442 (18). Plasmids used in this work were pT7-7 (32, 33), pVLT31 (12), and pEMBL18 (13). E. coli strains were grown at 37°C, while P. putida was grown at 30°C. In experiments designed to test the expression of BphD_P6, bacterial strains were grown on Luria-Bertani broth (LB) supplemented with carbenicillin (15 µg/ml; ICN Biomedicals Inc.) or tetracycline (12.5 µg/ml; ICN Biomedicals Inc.). Bacterial strains used for other purposes were grown on LB supplemented with the appropriate antibiotics.

DNA manipulation and amplification. DNA was purified, digested, and ligated using standard protocols (28). The bphD gene of strain P6 was amplified by PCR using two synthetic oligonucleotides designed to be complementary to the 5' and 3' ends of the published bpdF sequence from Rhodococcus sp. strain M5, which encodes a HOPDA hydrolase (22). The respective sequences of these oligonucleotides were 5'-CGGGCATATGATCCAAAAAATTG-3' and 5'-GGGGCTGCAGGTCAACTTAGATCAA-3'. These primers introduced NdeI and PstI restriction sites at the 5' and 3' ends, respectively, of the amplified product, facilitating subsequent cloning of the gene. The PCR mixture contained approximately 0.5 µg of BamH1-digested genomic DNA of R. globerulus P6, 0.75 U of Pfu DNA polymerase (Stratagene), 20 nmol of each deoxynucleoside triphosphate, and 100 pmol of each primer in a final volume of 100 µl. Twenty amplification cycles were performed as follows: 95°C for 1 min, 48°C for 1 min, and 72°C for 2 min. The PCR product was purified using a QIAquick PCR purification kit (Qiagen). DNA was purified from agarose gels using a Qiagen QIAEX-II gel extraction kit. Plasmid DNA was sequenced using an ABI model 373 Stretch DNA sequencer at the Nucleic Acid Analysis Unit, Université Laval. Sequencing reactions were performed according to the ABI dye-deoxy terminator protocol.

Purification of BphD_P6. Buffers containing 20 mM sodium HEPES (pH 7.5) were used throughout the purification. Chromatography was performed using an ÄKTA Explorer with resins and columns from Amersham Pharmacia Biotech. A crude extract prepared from 27 g of cells was loaded onto a Source 15Q anion-exchange column (2 by 9 cm) and eluted with a linear gradient of 0.05 to 0.25 M NaCl in 15 column volumes. Activity-containing fractions (0.11 M NaCl) were pooled and concentrated to 6 ml by ultrafiltration using an Amicon stirred cell equipped with a YM10 regenerated cellulose membrane. This preparation was brought to 5% saturation of ammonium sulfate and loaded onto a phenyl-Sepharose column (1 by 9 cm). The enzyme eluted around 1% saturated ammonium sulfate in a decreasing concentration gradient (5 to 0% in 5 column volumes). Activity-containing fractions from two runs were pooled, concentrated to 2.5 ml, and loaded onto a HiLoad 26/60 Superdex 200 gel filtration column. Fractions containing pure enzyme were pooled, concentrated by ultrafiltration as described above, and rediluted with fresh buffer. This was repeated several times to remove NaCl from the solution. Aliquots of concentrated enzyme were stored at 80°C.

Purification of other enzymes. Recombinant BphD_LB400 was purified as described previously (30). 2,3-Dihydroxybiphenyl dioxygenase (DHBD) of Burkholderia sp. strain LB400 was purified anaerobically as previously described (35).

Determination of protein concentration, purity, and molecular mass. Protein concentrations were determined using a bicinchoninic acid protein assay reagent kit (Pierce Chemical Co.) and bovine serum albumin as the standard. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed using a Bio-Rad Miniprotean II apparatus, and gels were stained with Coomassie blue according to established procedures (21). Molecular weight markers of the range 14,000 to 200,000 were used (Bio-Rad). The molecular mass of BphD was determined using high-pressure liquid chromatography-electrospray mass spectrometry (HPLC-ESMS). The instrumentation consisted of a microbore HPLC (Michrom UMA) connected on-line to a Perkin Elmer-Sciex model API III ESMS. Samples contained 5 mg of purified protein/ml. The molecular mass of the protein was determined using the Sciex deconvolution software.

Substrate preparation and determination of extinction coefficients. Extinction coefficents for HOPDA and monochlorinated HOPDAs were reported previously (29). Extinction coefficients for diCl HOPDAs were determined by addition of excess quantities of DHBD to solutions containing 100 mM phosphate buffer (pH 7.5) and weighed amounts of diCl DHBs, followed by the measurement of their absorbance spectrophotometrically on the instrument described below.

Kinetic measurements. Enzymatic activity was measured by following the consumption of the yellow substrate on a Varian Cary 3 spectrophotometer equipped with a thermojacketed cuvette holder. The spectrophotometer was interfaced to a microcomputer and controlled by Cary OS/2 multitasking software. The amount of enzyme used in each assay was adjusted so that the progress curve was linear for at least 2 min. Initial velocities were determined from a least-squares analysis of the linear portion of the progress curves using the kinetics module of the Cary software.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Construction of the expression vector. Amplification of the bphD gene from BamHI-digested genomic DNA from R. globerulus P6 yielded a DNA fragment of about 900 bp, as estimated from agarose gels. This fragment was digested with PstI and NdeI, extracted from agarose gels after electrophoresis, purified using a gel extraction kit, and then ligated into appropriately digested pT7-7 plasmid to produce plasmid pSS76. The entire sequence of the gene was determined from both strands of two separate clones. The sequences were identical to that of the bpdF gene of Rhodococcus sp. strain M5 except that the GGC codons corresponding to glycines 31 and 115 in the M5 sequence were replaced by the alanine codon GCC in the P6 sequence. The bphD gene, together with an upstream ribosomal binding site of T7 phage gene 10 protein of the T7-7 vector, was isolated from pSS76 as an XbaI/PstI fragment and inserted into two plasmids, pEMBL18 (13) and pVLT31 (12), a broad-host-range expression vector. These constructs were designated pSS186 and pSS316, respectively.

Expression, purification, and evaluation of molecular mass. Among the expression systems tested, including various E. coli strains containing each of the three described plasmid constructions, BphD_P6 was overexpressed only in P. putida KT2442 containing pSS316. The enzyme constituted about 15% of the total cellular protein as estimated from SDS-PAGE analysis and was therefore purified from this system. A total of 14 mg of BphD_P6, representing a yield of approximately 8%, was purified from 27 g (wet weight) of cells by anion-exchange, hydrophobic interaction, and gel filtration chromatographies. The final preparation of BphD_P6 was judged to be greater than 99% pure as judged from a Coomassie blue-stained denaturing gel (Fig. 2), and had a specific activity of 13.2 U/mg. HPLC-ESMS analysis revealed the subunit molecular mass of BphD_P6 to be 32,583.0 Da. This agrees well with the predicted molecular mass of the protein with the N-terminal methionine removed (32,576.3 Da). The concentrated enzyme preparation (9 mg/ml) was stable for at least 12 months when stored frozen at 80°C. Diluted enzyme preparations (less than 0.1 mg/ml) were stabilized by adding bovine serum albumin to a final concentration of 1 mg/ml.

View larger version (88K): [in this window] [in a new window]   FIG. 2.   Coomassie-blue-stained SDS-polyacrylamide gel of purified BphDP6. The gel was loaded with samples of BphD from R. globerulus P6: crude extract (37 µg, lane 2); preparation after anion exchange (30 µg, lane 3); preparation after hydrophobic interaction (10 µg, lane 4); and preparation after gel filtration (11 µg, lane 5). The molecular masses of the proteins in the standard (lane 1) are indicated at the left.

Properties of HOPDAs. The reported extinction coefficients of HOPDA and monochlorinated HOPDAs (29) were used in this study. The coefficients for 9,10-diCl and 9,11-diCl HOPDAs were 27.5 mM¹ cm¹ (_max = 439 nm) and 27.0 mM¹ cm¹ (_max = 438nm), respectively. The half lives of these compounds were 38 and 51 h, respectively, which are comparable to those of 9-Cl and 10-Cl HOPDAs (50 and 48 h, respectively [29]).

Steady-state kinetic analysis. BphD_P6 was found to obey classic Michaelis-Menten kinetics for all of the substrates tested here. Interestingly, the K_m of BphD_P6 for HOPDA was over twice that of BphD_LB400, although the specificity constant (k_cat/K_m) of BphD_P6 for this compound was within 70% of that of BphD_LB400 (Table 1).

View larger version (17K): [in this window] [in a new window]   FIG. 3.   Woolf-Augustinsson-Hofstee plot of the inhibition of the BphDP6-catalyzed hydrolysis of HOPDA by 4-OH HOPDA. Experiments were performed using 0 µM (), 2.9 µM (), 6.1 µM (×), 13.0 µM (), and 26.0 µM () 4-substituted HOPDA (100 mM phosphate [pH 7.5], 25°C). The solution of 4-substituted HOPDA contained a mixture of approximately 5% 4-Cl HOPDA and 95% 4-OH HOPDA. The lines represent a best fit of the data to an equation describing competitive inhibition using the least-squares and dynamic weighting options of LEONORA. The fitted parameters are Kic = 4.2 ± 0.2 µM, Km = 0.52 ± 0.01 µM, and Vmax = 14.81 ± 0.01 U/mg.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

This work describes the efficient heterologous expression and purification of BphD of R. globerulus P6, a serine hydrolase involved in the catabolism of biphenyl and some PCB congeners. Mass spectrometry analysis of the purified recombinant protein indicates that the N-terminal methionine is removed. Interestingly, heterologous expression of BphD_P6 was observed only in a pseudomonad host and not in those E. coli-based systems that worked well for BphD_LB400 (30). The heterologous expression of biphenyl dioxygenase of R. globerulus P6 was also observed only in P. putida KT2442 (25). Inefficient expression of the dioxygenase in E. coli was attributed to posttranscriptional host factors. Inspection of the nucleotide sequence of bphD_P6 revealed the presence of eight arginine codons (AGG and CGA), and six proline codons (CCC) that occur at a frequency of less than 1% in E. coli (19). However, mutation of a cluster of four of these codons, corresponding to residues 28, 33, 35, and 36, to frequently used codons resulted in only a slight improvement of the expression of BphD_P6 in E. coli (results not shown). Interestingly, a histidine-tagged variant of the rhodococcal dioxygenase is well expressed in E. coli (10). However, the recombinant dioxygenase did not contain its complement of Rieske-type iron sulfur cluster.

The steady-state kinetic data indicate that the respective specificities of BphD_P6 and BphD_LB400 are in some respects quite similar (Table 1), consistent with the notion that both enzymes evolved to degrade biphenyl. Accordingly, the specificity of BphD_P6 for HOPDA is similar to that of BphD_LB400. Moreover, to a first approximation, the two hydrolases displayed the same specificity toward monochlorinated HOPDAs: those bearing chloro substituents on the phenyl moiety are good substrates, while those bearing chloro substituents on the dienoate moiety are poor substrates if not competitive inhibitors. Considering that BphD_P6 and BphD_LB400 share 42% sequence identity, it seems likely that their specificities are representative of the majority of BphDs identified to date. However, it would be of interest to verify the specificity of even more divergent BphDs, such as that of Rhodococcus sp. strain RHA1 (23), which shares less than 30% sequence identity with either BphD_LB400 or BphD_P6.

The reactivity of BphD_P6 did, however, differ from that of BphD_LB400 in at least three respects. First, BphD_P6 catalyzed the hydrolysis of 9-Cl and 10-Cl HOPDAs at least twofold more specifically than BphD_LB400 did. Second, BphD_LB400 was approximately fivefold more specific for 5- and 8-Cl HOPDAs than BphD_P6. Finally, 4-Cl HOPDA was a more potent inhibitor of BphD_P6 than 3-Cl HOPDA, whereas 3-Cl HOPDA was a more potent inhibitor of BphD_LB400 than 4-Cl HOPDA. Thus, the respective specificities of BphD_P6 and BphD_LB400 for chlorinated metabolites are complementary to a certain extent.

In a recently published study, the specificity of BphD_LB400 for chlorinated HOPDAs was found to be largely consistent with the results of in vivo studies in which Burkholderia sp. strain LB400 was incubated with specific congeners. Accordingly, the accumulation of yellow-colored HOPDAs was observed when Burkholderia sp. strain LB400 was incubated with congeners such as 4,4'-diClB, 2,4,4'-triClB, and 2,5,4'-triClB, which are predicted to yield 3-Cl HOPDAs (5). Similarly, chloroacetophenones were detected when this strain was incubated with congeners such as 2,3'-diClB and 2,3,3'-triClB, which are predicted to yield unstable 4-Cl HOPDAs.

The specificity of BphD_P6 for chlorinated HOPDAs extends the interpretation of in vivo studies and raises some important questions. Consistent with the low specificity of this enzyme for 3-Cl HOPDAs, yellow-colored HOPDAs accumulate when R. globerulus P6 is incubated with 2,4,4'-triClB and 2,5,4'-triClB (15, 16). However, chlorobenzoates were reported to be the principal metabolites when the strain was incubated with 3,3'-diClB and 4,4'-diClB (15), which would be predicted to give rise to 4,9-diCl and 3,10-diCl HOPDA, respectively (Fig. 4). As benzoates are not formed spontaneously from HOPDAs (29), the in vivo data suggest that R. globerulus P6 harbors an enzyme which hydrolyzes 3,10-diCl, and 4,9-diCl HOPDA. It is possible that this enzyme is not the BphD_P6 characterized in this study. Significantly, R. globerulus P6 contains three DHBD isoenzymes, the genes of which map to different locations in the genome (4). Moreover, the dibenzofuran-degrading strain Sphingomonas sp. strain RW1 contains at least three BphD-type hydrolases (1, 8).

View larger version (15K): [in this window] [in a new window]   FIG. 4.   Proposed transformations of 3,3'-diClB to 4,9-diCl HOPDA (a) and 4,4'-diClB to 3,10-diCl HOPDA (b) in R. globerulus P6. Enzyme-catalyzed reactions are identified as A (biphenyl dioxygenase), B (2,3-dihydro-DHB dehydrogenase), and C (DHBD). The absolute configuration of the HOPDAs is not intended.

The transformation of 3,3'-diClB and 4,4'-diClB to chlorobenzoates by R. globerulus P6 might alternatively be explained by the influence of chloro substituents on the phenyl ring. Thus, BphD_P6 may be able to hydrolyze 3,10-diCl and 4,9-diCl HOPDA more efficiently than 3-Cl and 4-Cl HOPDA. Significantly, 9-Cl and 10-Cl HOPDAs are better substrates for the enzyme than unchlorinated HOPDA (Table 1). The reactivity of HOPDAs that are chlorinated on both the dienoate and phenyl moieties is particularly intriguing in light of the low K_m of BphD_LB400 for 9,11-diCl HOPDA. Should meta- and para-chloro HOPDA substituents increase the affinity of the enzyme for these compounds, then compounds such as 3,9,11-triCl HOPDA may be a exceptionally potent inhibitors of BphD_LB400. Further studies on the specificity of polychlorinated HOPDAs are required to resolve these issues.

The specificity and inhibition data point to significant electronic or structural differences in the respective active sites of BphD_P6 and BphD_LB400. For example, the higher K_ic of BphD_LB400 for 4-Cl HOPDA than for 4-OH HOPDA was suggested to arise from the configurations of these compounds (29). Thus, the proposed configuration of 4-OH HOPDA is more similar to that of the trans transoid enol tautomer of the unchlorinated substrate. However, the K_ic of BphD_P6 for 4-Cl HOPDA is an order of magnitude lower than that for 4-OH HOPDA, suggesting that there is a fundamental difference in how BphD_P6 and BphD_LB400 recognize the 4-substituted HOPDAs. Differences in these enzymes' active sites are also suggested by their different specificities for 5-Cl and 8-Cl HOPDA. Chloro substituents at these positions are relatively close to the site of nucleophilic attack at C-6 of HOPDA and thus could sterically hinder the binding of the substrate or its intermediates to the enzyme. Regardless of the origin of these deleterious effects, they are more significant in BphD_P6 than in BphD_LB400. The relevant differences in the active sites of these proteins may be revealed by the structure of BphD_LB400 (29).

Commercial mixtures of PCBs typically contain up to 60 different congeners. The effective microbial degradation of these environmental pollutants thus requires enzymes of broad or complementary specificities. The present study of BphD_P6, and its comparison to BphD_LB400, identifies this hydrolase as a key determinant of PCB degradation in evolutionarily divergent biphenyl-degrading microorganisms. In particular, the predicted competitive inhibition of BphD by 3-Cl and 4-Cl HOPDAs will slow the degradation of all PCB congeners at this step. Judicious screening of naturally occurring microorganisms may identify enzymes that are able to efficiently hydrolyze these compounds. Alternatively, in vitro evolution strategies may yield enzymes of the requisite specificity.