INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Pseudomonads play an important role in nature because of their ability to metabolize natural and manufactured organic chemicals. Many of these compounds are environmental pollutants, such as benzene, toluene, xylene, ethylbenzene, styrene, and chlorobenzoates (18), and their removal has been named bioremediation. Although the enzymic pathways responsible for degradation of these pollutants may be effective when the target compound is the sole growth-supporting substrate, in nature these compounds are present as mixtures, and some substrates may be degraded preferentially. Catabolite repression control refers to the ability of an organism to preferentially metabolize one carbon source over another when both are present in the growth medium. Because of the importance of pseudomonads to bioremediation efforts, understanding the control of catabolite repression is important so that more efficient, genetically modified organisms can be utilized in the removal of these environmental pollutants.

The molecular mechanisms of catabolite repression control have been extensively characterized in enteric bacteria, where glucose is the preferred carbon source. In these organisms, enzymes of the phosphoenolpyruvate-dependent phosphotransferase system mediate catabolite repression control by regulation of cyclic AMP (cAMP) concentration via adenylate cyclase activity (22). The strongest repressing substrates in Pseudomonas spp. are acetate, tricarboxylic acid cycle intermediates, and glucose (4, 10, 26). Unlike Escherichia coli, in Pseudomonas species adenylate cyclase activity, cAMP phosphodiesterase activity, and cAMP pools do not fluctuate with carbon source, nor does the addition of cAMP relieve repression of catabolite responsive pathways (21, 25). In addition, only one phosphotransferase system (fructose) has been identified in Pseudomonas (5), suggesting that PTS components are not involved in catabolite repression control in pseudomonads. The only protein thus far shown to be involved in catabolite repression in Pseudomonas is Crc of P. aeruginosa, but a function has not been identified for this protein (11). Crc has some sequence similarity (25 to 32% identity) to DNA repair enzymes of both prokaryotes and eukaryotes. However, Crc does not appear to have endonuclease activity or to bind DNA, suggesting that it has some other function.

Expression of branched-chain keto acid dehydrogenase (BCKAD) of P. putida is regulated by carbon and nitrogen sources (29). The bkd operon, which encodes BCKAD, and its regulation by BkdR, a positive transcriptional regulator, has been well characterized in P. putida (13-16, 29); therefore, P. putida BCKAD is a good model system for studying catabolite repression control of catabolic pathways in pseudomonads.

The objective of the present research was to determine if Crc played a role in catabolite repression control in P. putida, as well as in P. aeruginosa, by using the bkd operon as a model system. The phenotypes of the P. aeruginosa and P. putida crc mutants were compared, and although the carbon sources responsible for catabolite repression in the two species differ, the pathways regulated by Crc are identical in the two species.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains, plasmids, and culture conditions. The bacterial strains and plasmids used in this study are listed in Table 1. P. putida was grown with aeration at 30°C, while P. aeruginosa and E. coli were grown with aeration at 37°C. Starter cultures were grown in 2 ml of 2xYT medium (23). For BCKAD assays, P. putida and P. aeruginosa were grown in nitrogen-free valine-isoleucine medium (17) which contained 0.3% valine and 0.1% isoleucine (wt/vol) as the sole carbon and nitrogen sources (17).

Enzyme assays. Cultures for enzyme assays were grown in 100 ml of medium to an A₆₆₀ of 0.6 to 0.8, harvested, and then treated as described previously for enzyme assays (17). The following enzyme assays were performed as described previously: BCKAD (27), amidase (26), urocanase (7), and G6PDH (10). One unit of BCKAD is 1 µmol of NADH formed/min/mg of protein, and one unit of G6PDH is 1 µmol of NADPH formed/min/mg of protein. Amidase assays were done on whole-cell suspensions, and the specific activity was measured as micromoles of lactamide per A₆₆₀. The data in Fig. 5 are expressed as the percent activity relative to the lactamide plus lactate culture.

Construction and screening of P. putida genomic cosmid library. A complete BamHI digest of PpG2 chromosomal DNA was sized over a sucrose gradient, and fragments larger than 6 kb were ligated with BamHI-digested pLAFR3 (28) with T4 DNA ligase at a chromosome-to-vector ratio of 5:1. In vitro packaging and transformation of E. coli JM109 was performed according to the manufacturer's recommendations (Gigapack III Gold Packaging Extract; Stratagene). The library was screened with the 2.0 kb SstI fragment of pPZ352, which contains the PAO1 crc gene (11), and was labeled by using the RadPrime DNA labeling system (Life Technologies). The library was plated on 2xYT agar containing tetracycline, colonies were lifted by using Colony/Plaque Screen (Dupont/NEN Research Products), and hybridization and washing conditions were performed as suggested by the manufacturer.

Nucleic acid preparations and manipulations. Chromosomal and cosmid DNAs used for library construction were purified by using cesium chloride gradients. Plasmid DNA was purified by using QIAprep Spin Miniprep kit (Qiagen). Chromosomal DNA was purified by using the Qiagen Midi kit or the Gentra Systems Puregene kit. Basic DNA manipulations were performed as described earlier (23). Transfer of plasmids from E. coli to P. putida was done by triparental mating with pRK2013 (8).

Plasmid constructions. pUCP18/19 are pUC-derived plasmids which contain a fragment from RP1 (19) which permits stable maintenance of these plasmids in Pseudomonas species (24). These plasmids were further modified in this study so that they could be used to transform Pseudomonas species by triparental mating. In order to accomplish this, the XmaIII fragment from pLAFRI, (6), which contains the mobilization (mob) site, was blunted and cloned into the SmaI site of pUC18. The 750-bp mob fragment was isolated after digestion with KpnI and BamHI and blunted. Next, pUCP18/19 was digested with SspI and blunted, and the 750-bp blunted mob fragment was cloned into the blunted vector, providing pUCPM18 and pUCPM19.

Mutant construction. The coding region of crc was amplified from pJRS191 by using primers S163 (GAACAGGCCGGCATTGAAGAAATA) and S165 (GCGCTGGACATGAGCAAGCTGGGCG). The PCR product was digested with EcoRI and KpnI and ligated with pUCPM19 digested with the same restriction enzymes; this plasmid was designated pJRS194. The stabilization fragment was removed from plasmids which were to be used for homologous recombination. To remove the stabilization fragment of pUCPM19, pJRS194 was digested with NdeI and EcoRI, and the 4.6-kb fragment was religated. Transposon pUTKm (9) was digested with AlwNI, and the 1.25-kb fragment containing the Km^r gene was blunt ended and ligated with the construct described above digested with BsaAI. The resulting plasmid, pJRS197, contained the Km^r gene cloned in the opposite orientation to crc 447 bp downstream of the ATG of crc. Conjugal transfer of this suicide plasmid into PpG2 was accomplished by triparental mating with pRK2013 (8). Mutants were isolated by using Pseudomonas isolation agar containing kanamycin. Because all mutants isolated were also Cb^r and therefore were single crossovers, restriction digests and Southern blotting were used to identify recombination events that occurred upstream of the Km^r gene insertion site. Double crossovers were detected by the loss of carbenicillin resistance, and this recombination event was confirmed by restriction digests and Southern blotting with the NdeI/AlwNI fragment of pUC19 containing bla as a probe. The mutant with crc inactivated by insertion of a Km^r cassette was named P. putida JS394.

Transposon mutagenesis and screening. To isolate transposon mutants affected in bkdR expression, triparental matings of P. putida JS386, E. coli CC118(pUTKm), and E. coli HB101(pRK2013) were plated on Difco Pseudomonas Isolation Agar plus kanamycin plus 0.004% X-Gal (wt/vol; 5-bromo-4-chloro-3-indolyl--D-galactopyranoside). Approximately 10,000 colonies were screened on the basis of color, since JS386 is light blue on this medium. Both white and dark blue colonies were collected, and phenotypes were confirmed by -galactosidase assays. Southern blots were used to identify mutants with insertions in the bkdR-bkdA1 intergenic region or in lacZ, and these mutants were discarded. Four white colonies and three dark blue colonies were selected for further study. Genomic DNA from each mutant was digested with several restriction enzymes and Southern blotted with the 3.0-kb EcoRI-Km^r fragment to identify fragments of a desired size for cloning. The transposon insertion sites for three of the white colonies and all of the dark blue colonies were identified by cloning into pUC19 and sequencing the DNA insert by the Oklahoma University Health Sciences Center DNA Sequencing Facility.

DNA sequencing and analysis of nucleotide sequences. DNA Sequencing was performed by Bruce A. Roe's lab at the University of Oklahoma Advanced Center for Genomic Technology (Norman, Okla.). Cosmid DNA was purified by means of a cleared-lysate diatomaceous earth method (20). Sequencing was undertaken by using the double-stranded, shotgun-based approach (2). The resulting sequences were screened to eliminate vector, assembled into contiguous fragments, and proofread by using the Phred/Phrap/Consed system developed by P. Green (http://chimera.biotech.washington.edu/uwgc). Contigs larger than 2 kb were deposited before publication in the "unfinished" division of the high-throughput genome sequencing GenBank database and were given accession number AC004396.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Involvement of Crc in catabolite repression of the bkd operon of P. aeruginosa PAO1. Each strain of P. aeruginosa was grown on valine-isoleucine medium alone or supplemented with the additional carbon sources lactate, glucose, or succinate. Addition of glucose or succinate to the inducing (valine-isoleucine) medium caused a 65 to 70% repression of BCKAD activity in the wild-type strain PAO1, and even lactate caused some repression (Fig. 1). Repression was completely relieved in the crc mutant PAO8020, where similar BCKAD activities were seen in inducing medium with or without glucose or succinate. Repression was restored when PAO8020 was complemented with crc⁺ from P. aeruginosa on a plasmid (pPZ352). These results indicate that crc is responsible for a significant fraction of repression of the P. aeruginosa bkd operon by glucose and succinate synthetic media. The levels of BCKAD activity in the crc mutant were elevated compared to the wild type under all growth conditions, suggesting that constitutive levels of Crc in P. aeruginosa PAO1 cause some repression of bkd operon expression in the absence of repressing carbon source.

View larger version (55K): [in this window] [in a new window]   FIG. 1.   Effect of mutation in crc on catabolite repression of BCKAD in P. aeruginosa. P. aeruginosa PAO1, PAO8020, and PAO8020(pPZ352) were grown in valine-isoleucine medium either alone or supplemented with lactate, glucose, or succinate (columns in each group from left to right, respectively) as described in Materials and Methods. Data are the averages of three separate experiments.

Isolation of P. putida crc. P. putida crc was isolated in order to determine if Crc was involved in catabolite repression of the bkd operon of P. putida. A PpG2 chromosomal DNA library was constructed in pLAFR3 and screened with a 2.0-kb SstI fragment of pPZ352 containing the P. aeruginosa crc gene. The positive clone with a ~40-kb insert of P. putida, pJRS191, was partly sequenced, and the consensus sequence of a 13-kb contig containing the crc gene was obtained (Fig. 2).

View larger version (7K): [in this window] [in a new window]   FIG. 2.   Annotation of contig 270 from pJRS191. Open reading frames were identified by similarity searching by using the BLAST program (1). Similarities were found to the following proteins (identities at the protein level are given in parentheses): glnA, glutamine synthetase (61% to E. coli); gmk, GMP kinase (61% to E. coli); rph, RNase PH (84% to P. aeruginosa); crc, Crc (86% to P. aeruginosa); pyrE, orotate phosphoribosyltransferase (80% to P. aeruginosa); argB, acetylglutamate kinase (46% to Methanococcus jannaschii); and algC, phosphomannomutase (75% to P. aeruginosa). The only similarity found to the open reading frame (ORF) was to other reading frames of unknown function identified by genomic sequencing.

Effect of Crc on catabolite repression of the bkd operon of P. putida. P. putida JS394, which possesses a chromosomal crc inactivated by insertion of a kanamycin cassette, was created by homologous recombination. For complementation studies, P. putida crc was cloned into pUCPM19. This construct was numbered pJRS196 and was used to transform P. putida JS394. P. putida PpG2, JS394, and JS394(pJRS196) were grown in valine-isoleucine medium alone and with lactate, glucose, or succinate (Fig. 3). BCKAD activity obtained in the presence of lactate and glucose was about one-half to one-fourth the activity obtained in cells grown in valine-isoleucine medium. BCKAD activity obtained in the presence of succinate was about three-fourths of the control value. There was some relief of catabolite repression in P. putida JS394 when succinate was the repressor (Fig. 3). However, when P. putida JS394 was complemented with crc in trans, there was distinct evidence of catabolite repression by lactate, glucose, and succinate, probably due the effect of multiple copies of crc (Fig. 3).

View larger version (47K): [in this window] [in a new window]   FIG. 3.   Effect of mutation in crc on catabolite repression of BCKAD in P. putida. P. putida strains PpG2, JS394, and JS394(pJRS196) were grown in valine-isoleucine medium either alone or supplemented with lactate, glucose, or succinate (columns in each group from left to right, respectively). Kanamycin was always added to medium in which JS394 was grown, and carbenicillin was added to medium in which strains containing pJRS196 were grown. Data are the averages of two separate experiments.

Effect of P. putida Crc on catabolite repression of G6PDH and amidase. Crc also controls expression of G6PDH and amidase in P. aeruginosa (4), and the effect of Crc on expression of these enzymes was also studied in P. putida. G6PDH activities were studied in extracts of P. putida PpG2, JS394, and JS394(pJRS196) grown in BSM containing mannitol plus lactate or succinate. In this case, the control medium is mannitol plus lactate because growth in mannitol alone is very slow. As seen in Fig. 4, G6PDH activity in P. putida PpG2 is subject to strong catabolite repression by succinate; G6PDH activity in mannitol plus succinate was only about one-third that obtained in mannitol plus lactate medium. However, G6PDH activities in JS394 cultures grown in the presence of mannitol plus lactate and mannitol plus succinate were nearly identical, demonstrating relief of catabolite repression in the crc mutant. When JS394 was complemented with pJRS196, the repression by succinate was restored, demonstrating that P. putida Crc is involved in catabolite repression of G6PDH by succinate.

View larger version (69K): [in this window] [in a new window]   FIG. 4.   Effect of mutation in crc on catabolite repression of G6PDH of P. putida. P. putida PpG2, JS394, and JS394(pJRS196) were grown in BSM containing mannitol plus lactate or containing mannitol plus succinate (left and right columns in each group, respectively). Data are the averages of two separate experiments.

View larger version (68K): [in this window] [in a new window]   FIG. 5.   Effect of mutation in crc in catabolite repression of amidase in P. putida. P. putida strains PpG2, JS394, and JS394(pJRS196) were grown in BSM containing lactamide plus lactate, succinate, glutamate, glucose, or gluconate (columns in each group from left to right, respectively). Amidase activity is expressed as a percentage of the activity seen in P. putida PpG2 grown in lactamide plus lactate, which was included with each experiment.

P. putida Crc is not involved in catabolite repression of urocanase activity. Repression of urocanase and histidase activities by succinate was not relieved in P. aeruginosa crc mutants (P. Phibbs and C. MacGregor, unpublished data); therefore, Crc does not play a role in catabolite repression control of the hut operon. For comparison, P. putida PpG2 and JS394 were grown on BSM containing histidine plus lactate, succinate, or glucose and then assayed for urocanase activity. The specific activities of urocanase in PpG2 and JS394 grown on histidine plus glucose were the same for both strains and were about two-thirds that of the activity obtained when they were grown in histidine plus lactate. Thus, the results also suggest that Crc is not involved in catabolite repression control of the hut operon in P. putida, although repression was not very strong.

Crc and RNase R affect expression of bkdR. Crc could act directly on transcription of the bkd operon or indirectly on expression of bkdR. To test the latter possibility, transposon mutagenesis of the bkdR-lacZ fusion of strain JS386 was undertaken to isolate and identify mutants affected in bkdR expression (see Materials and Methods). The strategy was to isolate dark blue colonies, which should be derepressed for expression of bkdR. Three dark blue colonies were isolated, two of which had insertions in crc. The other dark blue colony had an insertion in vacB, which encodes a protein involved in posttranscriptional regulation of virulence genes in Shigella flexneri (30). VacB has recently been identified as RNase R in E. coli (3), and this terminology will be used here. The structural gene for RNase R is rnr and was found on the same contig which contains crc (Table 1). The amino acid sequence of RNase R from P. putida is 50% identical to RNase R from E. coli and is 48% identical to RNase R from S. flexneri.

View larger version (39K): [in this window] [in a new window]   FIG. 6.   Effect of mutations in crc and rnr on expression of bkdR in P. putida strains with a bkdR-lacZ translational fusion. P. putida strains JS386, JS391, and JS393 were grown in synthetic medium with valine-isoleucine plus 2xYT, lactate, succinate, or glucose (columns in each group from left to right, respectively). Data are the averages of at least three separate experiments.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Crc is involved in the control of catabolite repression of BCKAD (Fig. 1), G6PDH, and amidase in P. aeruginosa (4), but not of urocanase (Phibbs and MacGregor, unpublished). The same can be concluded for Crc of P. putida (Fig. 3 to 5), although the effect on catabolite repression of BCKAD is not as strong, and in P. aeruginosa it is weak (Fig. 1 and 3). While the same catabolic enzymes are affected in both species, the catabolite repressors have different degrees of effectiveness. Succinate is a good repressor in P. aeruginosa (Fig. 1) (4) but is a weak repressor in P. putida (Fig. 3 and 5). The high degree of sequence similarity between Crc from P. aeruginosa and P. putida and the results of the catabolite repression control studies (Fig. 3 to 5) support the conclusion that Crc carries out the same functions in both species. There are two differences which bear examination, namely, the small intergenic region between crc and rph in P. putida (69 bp) compared to that in P. aeruginosa (479 bp) and the increased expression of BCKAD relative to the wild type seen in P. aeruginosa (Fig. 1) but not P. putida (Fig. 3).

The fact that catabolite repression in pseudomonads does not involve cAMP makes the role of Crc in catabolite repression all the more interesting. Crc does not appear to be a DNA-binding protein and is therefore not likely to interfere with transcription. Other possibilities are that Crc acts posttranscriptionally by interfering with the expression of the mRNA or posttranslationally by modifying BkdR. However, the sequence similarity of Crc to endonucleases (11) makes the latter possibility remote. It has been pointed out that it was not possible to demonstrate the binding of Crc from P. aeruginosa DNA (11) and that has been our experience with P. putida DNA as well (K. L. Hester and J. R. Sokatch, unpublished data). The transposon mutants (Fig. 6) were isolated to test the possibility that Crc had an effect on expression of bkdR and, indeed, of the three colonies isolated with increased expression of LacZ from the crc-lacZ fusion, two were crc mutants. The other mutant with a similar phenotype was RNase R, which is interesting because Crc has some similarity to exonucleases (11). The accompanying study (9a) presents evidence that Crc acts posttranscriptionally.