INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Bacteria are endowed with systems that allow them to sense environmental and/or physiological signals, integrate them, and transmit an output response. Frequently, these responses are channeled through signal transduction pathways that ultimately allow the regulation of large sets of genes. One such global response is catabolic repression, which can be defined as the regulatory process allowing cells to preferentially use a certain carbon source over others when confronted by a mixture of them (25). To this end, cells should repress the expression of catabolic genes corresponding to the nonpreferred carbon sources present, which would otherwise be active. Catabolite repression is relatively well understood in Escherichia coli and Bacillus subtilis (20, 35, 36, 37) but not in Pseudomonas (7, 10). The mechanisms underlying this repression process in each of these bacterial species are quite different. For example, the levels of the CRP protein bound to cyclic AMP play an important role in catabolic repression in E. coli, but not in B. subtilis or Pseudomonas. In Pseudomonas aeruginosa and Pseudomonas putida, the levels of cyclic AMP do not vary appreciably with the carbon source, as they do in enteric bacteria (24), and the only CRP analog known, named Vfr, is involved not in catabolic repression but in quorum sensing (1). In addition, and as opposed to enteric bacteria, organic acids are usually preferred growth substrates for Pseudomonas species (24).

Very few proteins have been shown to participate in catabolic repression in Pseudomonas. The first to be described, Crc (catabolite repression control), is involved in the repression of a number of genes implicated in the metabolism of some sugars and nitrogenated compounds in both P. aeruginosa (10, 23, 45) and P. putida (18, 19). However, not all genes regulated by catabolic repression are influenced by Crc (18, 45). Despite several efforts, the mechanism by which Crc regulates gene expression is not clear. At the P. putida bkd operon, encoding the branched keto acid dehydrogenase, mutation of the crc gene led to elevated levels of the BkdR regulatory protein, although the levels of bkdR mRNA remained unchanged. This was interpreted as indicating that Crc has a direct or indirect posttranscriptional effect on bkdR expression (18, 19). Overall, available data suggest that Crc would be a component of a signal transduction pathway modulating carbon metabolism as well as other phenomena such as biofilm development (19, 24, 32).

We have investigated the role of Crc in catabolic repression of the alkane degradation pathway encoded in the OCT plasmid of P. putida GPo1, a strain previously known as Pseudomonas oleovorans GPo1 (43). The genes of this pathway are grouped in two clusters, alkBFGHJKL and alkST (Fig. 1) (43, 44). The alkBFGHJKL operon is transcribed from a promoter, named PalkB, whose expression requires the transcriptional activator AlkS and the presence of alkanes (22, 33). In the absence of alkanes, the alkST genes are expressed at low levels from promoter PalkS1, which is recognized by the ^S RNA polymerase and is autoregulated by AlkS (5, 6). When alkanes are present, the AlkS regulator represses PalkS1 more tightly and activates promoter PalkS2, which is located 38 nucleotides (nt) downstream from PalkS1 and which provides high expression of the alkST genes (5). Therefore, the pathway is controlled by a positive feedback mechanism governed by AlkS (Fig. 1). In addition, the levels of the enzymes in this pathway are modulated by catabolic repression, depending on the carbon source being used (16, 40). This superimposed control occurs by an unknown mechanism that regulates transcription from the promoters PalkB and PalkS2 (5, 47). Activation of these promoters by AlkS and the alkane inducer is very efficient when cells are grown in a minimal salts medium at the expense of citrate, but it shows a three- to fourfold reduction when organic acids such as lactate, pyruvate, or succinate are used as the carbon source. Repression is much stronger (about 30-fold repression) when cells grow exponentially in a rich medium, such as Luria-Bertani (LB) medium, or in minimal salts medium supplemented with Casamino Acids (47). Repression in rich medium abruptly disappears as cells enter the stationary phase of growth, which suggests the existence of elements that ensure a low expression of promoters PalkB and PalkS2 during exponential growth. In an attempt to identify factors involved in the modulation of this pathway, we have investigated the role of the Crc protein in the process. Our results indicate that Crc has essentially no role in the repression induced by organic acids, but it participates in the repression observed in rich medium. The mechanism through which Crc affects expression of this pathway was investigated.

View larger version (20K): [in this window] [in a new window]   FIG. 1.   The P. putida GPo1 alkane degradation pathway. The genes are grouped in two clusters, alkBFGHJKL and alkST (43, 44). In the absence of alkanes, the AlkS regulator is expressed at low levels from promoter PalkS1, which is autoregulated by AlkS (6). In the presence of alkanes, AlkS activates expression of the PalkB and PalkS2 promoters, generating a positive amplification loop. To activate PalkS2, AlkS binds to a site overlapping PalkS1. This fact, together with the higher AlkS levels generated, leads to a strong shut-off of promoter PalkS1 in the presence of alkanes (5). Activation of the PalkB and PalkS2 promoters by AlkS is down-regulated by catabolic repression when cells are grown in a defined medium containing certain organic acids (lactate or succinate) as carbon source or in rich LB medium (5, 47). Modified from Molecular Microbiology (5) with permissoin of the publisher.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains and culture conditions. The bacterial strains used are described in Table 1. Cells were grown at 30°C in rich LB medium or in minimal salts M9 medium (38), the latter supplemented with trace elements (4) and a carbon source (the indicated organic acid at a 30 mM concentration). Antibiotics were used at the following concentrations (in µg/ml): ampicillin, 100; kanamycin, 50; rifampin, 200; streptomycin, 50; tetracycline, 12 in E. coli and 8 in P. putida.

Recombinant DNA techniques. General methods for DNA manipulation were performed as described previously (38). DNA was sequenced on both strands with an Applied Biosystems DNA sequencer. PCR amplifications were performed using standard protocols (annealing temperature, 55 to 65°C; elongation temperature, 70°C; 30 cycles). Plasmids were introduced into P. putida by conjugation in triparental matings by using plasmid pRK2013 as helper (11).

Cloning and mutagenesis of the P. putida crc gene. The P. putida strain KT2442 crc gene was PCR amplified from chromosomal DNA using the oligonucleotides 5'-CAGGATCCGCCCCAGCTCAGCCAGTG and 5'-AGGGATCCGGCCGATCAAATAACCT. Both oligonucleotides contain artificially introduced BamHI sites (underlined) followed by sequences which hybridize, respectively, 232 nt upstream from the ATG start site or 48 nt downstream from the stop codon. The PCR product obtained, 1,073 bp in size, was cut with BamHI (size reduced to 1,067 bp) and cloned into the BamHI site of plasmid pUC18. The plasmid obtained was named pCRC5. This plasmid was digested with NruI, which cuts approximately at the middle of the crc gene, and ligated to a ca. 2,000-bp DNA fragment specifying resistance to tetracycline. This Tc^r determinant was obtained from mini-Tn5Tc by using endonuclease SmaI. The plasmid generated, in which the crc gene is interrupted by the Tc^r determinant (crc::tet), was named pCRC8. The crc::tet gene was excised from pCRC8 as an EcoRI-DNA fragment, the ends were blunted with the Klenow fragment of DNA polymerase I in the presence of deoxynucleoside triphosphates, and the gene was cloned at the SmaI site of plasmid pKNG101. The plasmid obtained was named pCRC10. Plasmid pKNG101 is specifically designed for marker-exchange mutagenesis (21). It replicates in E. coli, but not in P. putida, and carries an Sm^r determinant and the sacB gene, which mediates sucrose sensitivity. Plasmid pCRC10 was transferred to P. putida PBS4 (a P. putida KT2442 derivative harboring a PalkB::lacZ transcriptional fusion and the alkS gene in the chromosome), and Tc^r Sm^r sucrose-sensitive cells were selected. These cells harbor the pCRC10 plasmid integrated into the chromosome by a single crossing-over event, presumably at the crc gene, generating a wild-type and a mutant crc allele. Cells were cultured for a few generations in LB medium, plated in solid LB medium supplemented with tetracycline, and Tc^r Sm^s sucrose-resistant cells were selected. In these isolates, a second crossing-over event should have occurred, excising pKNG101 sequences together with the wild-type crc gene and leaving a crc::tet allele. The absence of a wild-type crc gene and the presence of the crc::tet allele was verified by PCR (data not shown). An isolate named PBS4C1 was selected for further work. A similar procedure was followed to transfer the crc::tet allele to P. putida strains PS16 and PBS10. Strain PS16 is a KT2442 derivative which carries a PalkS::lacZ transcriptional fusion integrated into the chromosome (the term PalkS refers to the two adjacent promoters for alkS, PalkS1, and PalkS2, separated by 38 nt; Fig. 1). An isolate named PS16C1 was selected for further work. The alkS gene was delivered into the chromosome of PS16C1 by means of the suicide donor plasmid pSS1, a mini-Tn5Sm derivative harboring the alkS gene. The strain obtained was named PSCS1. In the case of strain PBS10, which derives from P. putida KT2440 by insertion of the PalkB::lacZ fusion and the alkS gene in the chromosome, the crc-deficient derivative selected was named PBS10C1. To obtain plasmid pCRC11, the crc gene was excised from plasmid pCRC5 with BamHI and cloned at the BamHI site of the broad-host-range vector pKT231.

Obtention of a strain overexpressing the alkS gene from a heterologous promoter. A plasmid containing the alkS gene under the influence of the Ptrc promoter was constructed as follows. A NotI DNA fragment from plasmid pVTRC, containing the lacI^q gene, the Ptrc promoter, and a polylinker followed by transcription termination sites, was cloned between the NotI sites of vector pUC18Not; the resulting plasmid was named pUCVTR. A promoterless alkS gene was constructed by PCR using the alkS-containing plasmid pTS1 as substrate. Oligonucleotides 5'-TCCAGAAGCTTAAGAAGGAGATAGCATAATGAAAATAA, which includes a HindIII site, a ribosome binding site, and the start of the alkS gene, and 5'-CTCTCTCACACGGCTGA, which anneals between positions 333 and 317 of alkS relative to the translation start site, were used to amplify a ca. 300-bp DNA fragment containing the promoterless 5' end of alkS. The fragment was cut with HindIII and EcoRV and used to substitute the HindIII-EcoRV fragment of pTS1 that contains the start of the alkS gene. The plasmid obtained was named pTS21. A ca. 3-kbp DNA fragment containing the promoterless alkS gene was excised from plasmid pTS21 with EcoRI and HindIII (the latter cohesive end was blunted with mung bean nuclease prior to digestion with EcoRI), and the fragment was inserted between the EcoRI and NotI sites (the latter blunted with mung bean nuclease) of pUCVTR. In the resulting plasmid, named pUVS1, the alkS gene lies downstream from the strong Ptrc promoter. The alkS gene under the influence of the Ptrc promoter was excised from pUVS1 with NotI and inserted at the NotI sites of plasmids pUT-mini-Tn5Sm and pUT-mini-Tn5Km, obtaining plasmids pHLS1 and pHLS3, respectively. Plasmid pHLS1 was used to deliver Ptrc::alkS into the P. putida strain PBS4 by conjugation to obtain strain PBSH1. Similarly, pHLS3 was used to insert Ptrc::alkS into strain PS16S1 to obtain strain PSH2.

Assay for -galactosidase. An overnight culture of the appropriate strain was diluted in duplicate flasks to a final turbidity (A₆₀₀) of about 0.04 in fresh LB medium or in minimal salts M9 medium supplemented with the indicated carbon source. When cultures reached an A₆₀₀ of about 0.08, the nonmetabolizable inducer dicyclopropylketone (DCPK), which mimics the effect of alkanes (16), was added to one of the flasks (0.05% [vol/vol]), leaving the other one as a noninduced control. Growth was allowed to continue and, at different time points, aliquots were taken and -galactosidase activity was measured as described by Miller (28). Between three and five independent assays were performed in each case.

S1 nuclease analyses of mRNAs. Total RNA was isolated from bacterial cultures as previously reported (29). S1 nuclease reactions were performed as described elsewhere (2), using 50 µg of total RNA and an excess of a 5'-end-labeled single-stranded DNA hybridizing to the 5' region of the mRNA. The single-stranded DNA probes were generated by linear PCR, as described previously (47), using as substrates either plasmid pPB7 linearized with PtsI (probe for the PalkB promoter) or pTS1 linearized with HindIII (probe for the PalkS2 promoter).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Crc has an important role in the down-regulation of the PalkB promoter in rich medium but not in that exerted by organic acids in a defined medium. Our previous analyses of the catabolic repression exerted on the PalkB promoter were performed using strain PBS4, a derivative of P. putida KT2442 harboring a PalkB::lacZ transcriptional fusion and with the alkS gene inserted into the chromosome (47). Strain KT2442 was selected because it is closely related to P. putida GPo1 (41, 43), is well characterized, and does not grow on alkanes. All regulatory features of the GPo1 alkane degradation pathway analyzed in both strains have been found to be conserved (5, 6, 33, 40, 47). To analyze the effect of the crc gene in the modulation of PalkB activity by carbon sources, an inactivated crc allele (crc::tet) was introduced by marker exchange in place of the wild-type allele of strain PBS4 (see Materials and Methods). As a first approach, the behavior of the PalkB promoter in the crc-deficient strain, named PBS4C1, was analyzed following expression of the lacZ reporter gene after addition of the nonmetabolizable inducer DCPK. It should be noted that previous analyses had indicated that -galactosidase levels in this strain faithfully reproduce the transcriptional activity of the promoter (47). Addition of the inducer DCPK to cells growing in a minimal salts medium containing citrate as the carbon source, a condition in which no catabolic repression is observed (40, 47), led to an immediate increase in the levels of -galactosidase, although these were somewhat lower in the crc-deficient PBS4C1 strain than in the wild-type PBS4 strain (Fig. 2). The use of lactate, succinate, or pyruvate as carbon sources is known to induce a three- to fourfold repression on PalkB induction (47). As shown in Fig. 2, when lactate was the carbon source used, a clear catabolic repression effect was observed on PalkB activity both in the wild-type and the crc-deficient strain, since -galactosidase levels were three- to fourfold lower at mid-exponential phase and two- to threefold lower at the late exponential phase than those observed when citrate was the carbon source (Table 2). Therefore, disruption of the crc gene had a very small effect (if any) on the catabolic repression exerted on PalkB activity by organic acids in a defined medium (1.3- to 1.5-fold reduction). However, when cells were grown in rich LB medium, a clear difference was noted between crc-deficient and wild-type strains. As previously reported (47), activity of the PalkB promoter in the wild-type strain remained very low (around 50 Miller units) after addition of the inducer throughout the exponential phase of growth and steadily increased when cells approached the stationary phase of growth, which occurred at an A₆₀₀ of about 1.2 to 1.5 (Fig. 2). The pattern was different for the crc-deficient strain. Expression was already evident during the mid-exponential phase, the levels of -galactosidase being about fivefold higher than those observed for the wild-type strain and reaching values close to 10-fold higher at an A₆₀₀ of 1.0 (Fig. 2 and Table 2). This corresponds to a relief in repression of almost 6-fold at mid-exponential phase and of 15-fold at the late exponential phase (Table 2). Nevertheless, repression was not totally relieved, since the -galactosidase levels observed in the LB-grown crc-deficient strain were still four- to sixfold lower than those in cells cultured in minimal salts medium with citrate as carbon source. Introduction into strain PBS4C1 of a broad-host-range plasmid harboring the crc gene (pCRC11) restored full catabolic repression on the PalkB promoter (data not shown). These results indicate that the Crc protein has an important role in the catabolic repression of the PalkB promoter in rich medium.

View larger version (20K): [in this window] [in a new window]   FIG. 2.   Effect of Crc on induction of the PalkB promoter in cells growing at the expense of different carbon sources. Strains PBS4 (wild type for crc; open circles) or PBS4C1 (lacking a functional crc gene; filled triangles), both of which contain a PalkB::lacZ transcriptional fusion and the alkS gene, were grown in duplicate flasks in either minimal salts media supplemented with citrate or lactate as carbon source or in rich LB medium. At an A600 of about 0.08, the nonmetabolizable inducer DCPK was added to one of the flasks (the other one was left as a control) and incubation continued. At various time points, -galactosidase activity was measured. Shown are the -galactosidase activities observed in the induced cultures, represented as a function of cell growth. -galactosidase activities in noninduced cultures were very low (30 to 100 Miller units, depending on cell density) and are not represented. The -galactosidase values shown correspond to five independent assays, all represented on the same plot.

Effect of Crc on expression of the AlkS regulator. Expression of the alkS gene occurs from two adjacent promoters which are autoregulated by the AlkS protein (Fig. 1). Promoter PalkS1, which is repressed by AlkS, is active only in the absence of alkanes (5, 6), and its activity is not influenced by the carbon source being used (5, 47). On the contrary, promoter PalkS2, which is activated by AlkS in the presence of alkanes, is affected by catabolic repression very much in the same way as promoter PalkB is (5). To analyze whether Crc has a role in the catabolic repression of promoter PalkS2, P. putida strain PSCS1 was constructed. This strain contains an inactivated crc allele (crc::tet) as well as a PalkS::lacZ transcriptional fusion (including promoters PalkS1 and PalkS2) and the alkS gene inserted into the chromosome. It is worth noting that, as explained above, under the conditions used (exponential growth in the presence of alkanes) promoter PalkS1 is inactive (5), so that the -galactosidase levels observed from the PalkS::lacZ fusion reflect the activity of the PalkS2 promoter only. Figure 3 shows that disruption of the crc gene reduced catabolic repression on the PalkS2 promoter in cells growing exponentially in LB medium but not in cells growing in a minimal salts medium containing lactate as the carbon source. When grown in LB medium, the repression observed for the wild-type strain was about 33- to 45-fold, depending on the growth phase considered (Table 3). In the case of the crc-deficient strain, repression decreased to only five- to sixfold. This indicates that disruption of the crc gene reduces catabolic repression of the PalkS2 promoter in LB medium by about sevenfold.

View larger version (18K): [in this window] [in a new window]   FIG. 3.   Effect of Crc on induction of the PalkS2 promoter in cells grown at the expense of different carbon sources. Strains PS16S1 (wild type for crc; open circles) or PSCS1 (lacking a functional crc gene; filled triangles), both of which contain a PalkS::lacZ transcriptional fusion and the alkS gene, were grown as described for Fig. 2 and -galactosidase activity was measured at various time points. The -galactosidase activities observed for the induced cultures are represented as a function of cell growth. PalkS activity in noninduced cultures was low (20 to 200 Miller units, depending on cell density) and is not represented. The -galactosidase values shown correspond to two to four independent assays, all represented on the same plot.

Disruption of crc leads to increased levels of the transcripts arising from the promoters PalkB and PalkS2 The higher levels of -galactosidase in crc-deficient cells containing the PalkB::lacZ and PalkS::lacZ transcriptional fusions should reflect an increase in the amounts of mRNA originated at the PalkB and PalkS2 promoters. As shown in Fig. 4, the levels of these transcripts in the crc-deficient strain PBS4C1 growing exponentially in rich medium were six- to sevenfold higher than those in the wild-type strain. As expected, promoter PalkS1 was not expressed in either strain (data not shown). The higher expression of PalkB and PalkS2 promoters in the crc-deficient strain suggests that Crc could affect (directly or indirectly) either transcription initiation or the stability of the mRNAs originated at these two promoters (but see Discussion). We therefore analyzed the stability of these transcripts by monitoring mRNA decay after addition of rifampin to induced cultures. However, since strains PBS4 and PBS4C1 are derived from P. putida KT2442, which is in turn a rifampin-resistant derivative of P. putida KT2440, we constructed new reporter strains equivalent to the above ones but that were derived directly from KT2440. To this end, the PalkB::lacZ fusion was delivered to the KT2440 chromosome with the help of the suicide donor plasmid pPBK2 to obtain strain PB10. The alkS gene was subsequently inserted into the chromosome of PB10 by using the suicide donor plasmid pTLS1 to obtain strain PBS10. Finally, the crc gene of strain PBS10 was replaced by the crc::tet allele by marker-exchange mutagenesis to obtain strain PBS10C1. Control assays similar to those shown in Fig. 2 showed that induction of the PalkB promoter in strains PBS10 and PBS10C1 behaved in the same way as in strains PBS4 and PBS4C1, respectively (data not shown). Therefore, strains PBS10 and PBS10C1 were used for the mRNA decay assays. As shown in Fig. 5, the half-life of the mRNA that originated at the PalkB promoter was about 1 min both in the strain bearing a wild-type crc gene (strain PBS10) and in that lacking a functional crc gene (strain PBS10C1). The message arising from promoter PalkS2 was more stable, showing a half-life of almost 3 min. Disruption of the crc gene reduced somewhat its decay rate, increasing the half-life to about 3.5 min. However, this small increase in half-life cannot explain the six- to sevenfold increase in the steady-state PalkS2 mRNA levels observed upon disruption of the crc gene. These results allow us to conclude that Crc down-regulates the levels of transcripts arising from the PalkB and PalkS2 promoters in cells growing exponentially in rich medium, but that it has little effect on the stability of these transcripts.

View larger version (30K): [in this window] [in a new window]   FIG. 4.   Effect of Crc on the amounts of transcripts originated at the PalkB and PalkS2 promoters in cells grown in LB medium. Strains PBS4 (wild type for crc) or PBS4C1 (lacking a functional crc gene), both of which contain a PalkB::lacZ transcriptional fusion and the alkS gene, were grown in LB medium in the presence of DCPK. At an A600 of about 0.8, cells were collected and processed to obtain total RNA. (A) Levels of mRNA originated at the PalkB and PalkS2 promoters in each strain, determined by S1 nuclease protection assays in duplicate RNA samples. Lane M, DNA size ladder. (B) The bands observed for promoters PalkB and PalkS2 in three independent assays equivalent to that shown in panel A were quantified on a phosphorimager; the plot shows the mRNA levels observed in the crc-deficient cells relative to those observed in wild-type cells.

View larger version (15K): [in this window] [in a new window]   FIG. 5.   Stability of the transcripts originated at the PalkB and PalkS2 promoters and the effect of Crc. Strains PBS10 (wild type for crc) and PBS10C1 (lacking a functional crc gene), both of which contain a PalkB::lacZ fusion and the alkS gene in the chromosome, were grown in LB medium in the presence of DCPK to an A600 of about 0.8. Rifampin (200 µg/ml) was added to inhibit transcription and, at 0, 1, 2, 4, 6, 8, and 10 min, aliquots were taken, frozen on dry ice, and processed to obtain total RNA. The amounts of mRNA originated at the PalkB and PalkS promoters in each sample were determined by S1 nuclease protection assays. The graphs show the percentage of mRNA remaining relative to that observed at the moment of addition of rifampin. The standard deviation ranged from 2 to 10%. The dashed arrows indicate the half-lives of the mRNAs in each case. Filled circles, strain PBS10; open squares, strain PBS10C1.

High levels of the AlkS protein eliminate the catabolic repression of the PalkB and PalkS2 promoters. The above results showing that Crc can modulate the expression of the alkS gene suggest that the catabolic repression strategy could be directed to decrease the levels of the AlkS regulator. It could be presumed that reduction of the AlkS concentrations below a certain threshold would lead to a decline in the activity of the PalkB and PalkS2 promoters. This hypothesis agrees with previous results showing that an increase in the alkS gene dosage, which leads to an increase in the concentration of the AlkS protein, partially relieves catabolic repression in rich medium both at the PalkB and at the PalkS2 promoters (5, 47). The decrease in repression was, however, moderate, and the experimental approach used did not allow us to distinguish whether the effect was due to the increase in the levels of AlkS or to the presence of a higher number of copies of the alkS gene which, as subject of the repression effect, may titrate any putative regulator involved in repression. To eliminate this drawback, we analyzed the effect of expressing the alkS gene from a strong heterologous promoter. To this end, a Ptrc::alkS transcriptional fusion was introduced into the chromosomes of the reporter strains PBS4 and PS16S1. As shown in Fig. 6, expression of alkS from the strong Ptrc promoter totally eliminated the catabolic repression effect observed in LB medium at both the PalkB and PalkS2 promoters. Therefore, high expression of the alkS gene in an AlkS-independent manner overcomes catabolic repression in rich medium, which supports the idea that the repression effect may be directed to limit the levels of the AlkS protein.

View larger version (27K): [in this window] [in a new window]   FIG. 6.   Effect of overexpressing the alkS gene from a heterologous promoter on the catabolic repression observed in LB medium. Strains PBSH1 (contains PalkB::lacZ and Ptrc::alkS transcriptional fusions; gray squares on the upper plot) and PSH2 (contains PalkS::lacZ and Ptrc::alkS transcriptional fusions; gray squares on the lower plot), were grown in LB medium as described for Fig. 2 and in the presence of isopropyl--D-thiogalactopyranoside (1 mM) to induce the Ptrc promoter. After induction with DCPK, -galactosidase activity was measured at various time points. The values obtained in induced cultures are represented as a function of cell growth; the upper plot corresponds to promoter PalkB (strain PBSH1), while the lower plot corresponds to promoter PalkS (strain PSH2). -galactosidase activity in the absence of DCPK was low and is not represented. The -galactosidase values shown correspond to four independent assays, all represented on the same plot. Values for strains PBS4, PBS4C1, PS16S1, and PSCS1 correspond to those shown in Fig. 2 and 3 and are provided for comparison.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The results presented here show that the crc gene has an important role in the catabolic repression of the PalkB and PalkS2 promoters that occurs when cells grow exponentially in rich LB medium. The crc gene, however, has no significant role in the catabolic repression observed in defined media containing organic acids such as lactate or succinate as the carbon source. These findings provide the first experimental evidence that the catabolic repression effect observed in these two culture media can be genetically distinguished, which suggests that catabolic repression operates through at least partially independent mechanisms in each case. This conclusion agrees with previous results showing that an increase in the copy number of the alkS gene partially relieves the catabolic repression observed in rich LB medium but not that occurring on a defined medium containing organic acids as the carbon source (5, 47).

Disruption of the crc gene led to a sixfold increase at mid-exponential phase or to a 15-fold increase at late exponential phase in the amount of -galactosidase expressed from the PalkB promoter in cells growing in LB medium in the presence of inducer. Similarly, the -galactosidase expressed from the PalkS2 promoter in a crc-deficient background increased about sevenfold at both the mid-exponential and late exponential phases of growth. Although the decrease in repression is quite significant, the absence of a functional crc gene did not totally relieve catabolic repression in rich medium, since the amounts of -galactosidase expressed from the PalkB or PalkS2 promoters were still five- to sixfold lower than those found under conditions of no catabolic repression (defined medium containing citrate as carbon source). This suggests that catabolic repression of these two promoters in LB medium could be mediated by at least two different systems, one of them including Crc as an important component, the other one being independent of Crc.

The mentioned increase in the levels of -galactosidase in the absence of a functional crc gene was paralleled by a sevenfold increase in the levels of the transcripts originated at the PalkB and PalkS2 promoters. Crc had little influence on the stability of these transcripts, which indicates that the higher mRNA levels observed in the crc-deficient background are due to an increase in transcription initiation. However, this increase in transcription initiation could be an indirect effect of the positive feedback mechanism that regulates the alkane degradation pathway. Since the promoters PalkB and PalkS2 are both activated by AlkS and this protein is present in the cell in limiting amounts (47; also, see below), an increase in the translation efficiency of the alkS mRNA would lead to higher levels of AlkS, which in turn could allow a more efficient activation of the PalkB and PalkS2 promoters. Unfortunately, it is difficult to distinguish between transcriptional and translational effects in the system under study since, as mentioned above, an increase in the levels of the AlkS protein due to enhanced translation would immediately lead to an increase in transcription initiation from the two promoters. Although this limitation does not allow us to identify at the present time which is the initial target of Crc, the results discussed here point to the AlkS protein as the central player in catabolic repression of the pathway. In spite of the positive amplification system that regulates the expression of the alkS gene, the levels of the AlkS protein present in induced cells appear to be low. We have previously shown that introduction of a high-copy-number plasmid bearing a PalkB::lacZ transcriptional fusion into strain PBS4, a strain which already bears a copy of the PalkB::lacZ fusion, and a copy of the alkS gene in the chromosome, does not lead to an increase in PalkB transcription compared to the situation where the fusion is in monocopy. However, when the alkS gene is included into the multicopy plasmid together with the PalkB::lacZ fusion, the activity of the PalkB promoter increases significantly (47). This result suggests that the AlkS levels generated from a single copy of alkS are sufficient to activate a single copy of the PalkB promoter, but not more. In addition, we have observed that the AlkS protein is highly unstable (G. Morales and F. Rojo, unpublished results). Finally, it has been observed that introduction of a multicopy plasmid containing the binding site for AlkS into a bacterial strain containing a PalkB::xylE transcriptional fusion and a copy of alkS strongly reduced the activity of the PalkB promoter, which indicates that the presence of multiple copies of the AlkS binding site reduces the amount of free (not bound to DNA) AlkS protein present (43). Altogether, the available data indicate that the levels of the AlkS regulator are probably limiting in induced cells. This provides an opportunity to modulate the activity of the pathway by controlling the levels of the AlkS protein. We show that overexpression of AlkS from a strong heterologous promoter leads to a total disappearance of catabolic repression in rich medium. Since catabolic repression interferes with the expression of the alkS gene, it is highly plausible that catabolic repression in rich medium is directed to down-regulate the levels of the AlkS protein.

Many degradation pathways for hydrocarbons and aromatic compounds are subject to catabolic repression in Pseudomonas species (8, 9, 12, 13, 26, 27, 30, 31, 39, 42). Our results provide the first evidence for a role of Crc in modulating the expression of a catabolic pathway for hydrocarbons. Crc is known to modulate the expression of branched-chain keto acid dehydrogenase (bkd operon), glucose-6-phosphate dehydrogenase, and amidase in both P. aeruginosa (10, 23, 45) and P. putida (18, 19). In the case of the P. putida branched-chain keto acid dehydrogenase, disruption of crc affected the catabolic repression observed in a defined medium containing organic acids as carbon source, as well as that observed in rich medium. This is in contrast to our results for the P. putida GPo1 alkane degradation pathway, which show that Crc has an important role in the catabolic repression in rich medium but hardly affects that exerted by organic acids in a defined medium. This may reflect differences in the pools of key metabolic intermediates in the different P. putida strains analyzed. Crc-mediated repression of the bkd operon in rich medium was shown to involve a decrease in the levels of BkdR, the positive regulator of the operon. The levels of bkdR mRNA did not increase in a crc-deficient background, which led to the proposal that Crc acts posttranscriptionally (19). It should be noted that, contrary to AlkS, BkdR does not seem to activate its own synthesis. It is possible that, as it has been proposed for bkdR, Crc regulates expression of alkS posttranscriptionally, the final effect being an increase in transcription initiation from the AlkS-activated PalkS2 promoter. Crc does not appear to be a DNA binding protein (18, 23). In agreement with this idea, we have observed that elimination of the DNA sequences upstream from the AlkS binding site at promoter PalkS2 (which is located at position 42.5 relative to the transcription start site) does not interfere with catabolic repression (our unpublished results). Therefore, catabolic repression does not seem to be mediated by a coregulator (Crc or any other protein) binding upstream from AlkS. Crc shows sequence similarity with a group of DNA repair enzymes, although no endo- or exonuclease activities have been identified (23). It has been discussed that Crc's nuclease or RNA binding activity could be directed to a specific secondary RNA structure, resulting either in degradation of the mRNA or in decreased translation (19). We have shown here that Crc has little influence on the stability of the transcripts arising from the PalkB and PalkS2 promoters, indicating that its effect is not exerted through RNA degradation. This leaves us with the alternative of Crc acting to decrease translation of the mRNA, a possibility that is compatible with all our results and with those reported for the bkd operon.