Abstract
The transcriptional regulation of three distinct alcohol oxidation systems, alcohol dehydrogenase (ADH)-I, ADH-IIB and ADH-IIG, in Pseudomonas putida HK5 was investigated under various induction conditions. The promoter activities of the genes involved in alcohol oxidation were determined using a transcriptional lacZ fusion promoter-probe vector. Ethanol was the best inducer for the divergent promoters of qedA and qedC, encoding ADH-I and a cytochrome c, respectively. Primary and secondary C3 and C4 alcohols and butyraldehyde specifically induced the divergent promoters of qbdBA and aldA, encoding ADH-IIB and an NAD-dependent aldehyde dehydrogenase, respectively. The qgdA promoter of ADH-IIG responded well to (S)-(+)-1,2-propanediol induction. In addition, the roles of genes encoding the response regulators exaE and agmR, located downstream of qedA, were inferred from the properties of exaE- or agmR-disrupted mutants and gene complementation tests. The gene products of both exaE and agmR were strictly necessary for qedA transcription. The mutation and complementation studies also suggested a role for AgmR, but not ExaE, in the transcriptional regulation of qbdBA (ADH-IIB) and qgdA (AGH-IIG). A hypothetical scheme describing a regulatory network, which directs expression of the three distinct alcohol oxidation systems in P. putida HK5, was derived.
- ADH, alcohol dehydrogenase
- PQQ, pyrroloquinoline quinone
- qADH, quinoprotein ADH
- qEDH, quinoprotein ethanol dehydrogenase
Edited by: M. A. Kertesz
INTRODUCTION
Alcohol dehydrogenases (ADHs) whose reaction is independent of NAD(P) have been found in many aerobic bacteria. Most of these enzymes contain pyrroloquinoline quinone (PQQ) as a prosthetic group and are termed quinoprotein ADHs (qADHs). Methanol dehydrogenase, found in methylotrophic bacteria, was the first quinoprotein shown to have a prosthetic PQQ (Anthony, 1982). Although considerable research has been carried out on the biochemistry and physiology associated with the qADHs (Anthony & Williams, 2003; Chen et al., 2002; Matsushita et al., 1999; Toyama et al., 2004), much less is known about the transcriptional regulation of these enzymes. Insights into the complexity of the transcriptional regulation of qADHs have been obtained from studies of Pseudomonas aeruginosa (Gliese et al., 2004), Pseudomonas butanovora (Vangnai et al., 2002), Pseudomonas putida (Promden et al., 2008) and Methylobacterium extorquens, in which at least 28 genes have been shown to be involved in the oxidation of methanol to formaldehyde (Lidstrom et al., 1994; Springer et al., 1995). In P. aeruginosa, the transcription of a quinoprotein ethanol dehydrogenase (qEDH) promoter, namely the exaA promoter, is regulated by a two-component system: a histidine sensor kinase (ExaD), which is presumably located in the cytoplasm, and a response regulator (ExaE). The AgmR response regulator has been shown to control transcription of a regulon consisting of the three operons exaBC, exaDE and pqqABCDE, the gene products of which are essential for ethanol oxidation (Gliese et al., 2004; Schobert & Görisch, 2001).
P. putida HK5 was originally isolated as a 1-octanol utilizer. When grown on different alcohols, P. putida HK5 expresses three distinct soluble ADHs (ADH-I, ADH-IIB and ADH-IIG), each of which contains a PQQ prosthetic group. ADH-I and ADH-IIB are formed in cells either grown on or induced with short-chain-length alcohols, while the induction of ADH-IIG is mainly restricted to 1,2-propanediol (Toyama et al., 1995). P. putida HK5 is one of only two organisms reported so far that expresses more than one type of PQQ-ADH in response to exposure to alcohols (Vangnai et al., 2002). Therefore, it is interesting to examine how the expression of the three enzymes is distinguished and regulated at a molecular level. The cloning and molecular analysis of the three PQQ-ADH genes, encoding ADH-I, ADH-IIB and ADH-IIG, was performed previously (Promden et al., 2008; Toyama et al., 2003, 2005). The aim of the present study was to analyse the promoter activities under various induction conditions in P. putida HK5, using a transcriptional lacZ fusion promoter-probe vector as a reporter, and to investigate the role of a two-component system that controls the three ADH clusters.
METHODS
Bacterial strains and culture conditions.
Bacterial strains used in this study are listed in Table 1⇓. Escherichia coli was cultivated at 37 °C in Luria–Bertani (LB) medium. P. putida HK5 wild-type and mutant strains were cultivated at 30 °C, either in LB medium or in basal medium (Promden et al., 2008). Alcohols used as inducers were added to a final concentration of 0.5 % (v/v) (equivalent to 50–85 mM, depending on the alcohol used), while aldehydes were used at 0.1 % (v/v) (10–18 mM). Antibiotics were added to the following final concentrations: ampicillin, 50 μg ml−1; piperacillin, 200 μg ml−1; tetracycline, 25 μg ml−1; kanamycin, 50 μg ml−1.
Bacterial strains and plasmids used in this study
DNA manipulations and construction of plasmids.
Routine recombinant DNA work was performed according to the protocols described by Sambrook et al. (1989). DNA plasmids used in this study are listed in Table 1⇑, and their principal constructs are shown schematically in Fig. 1⇓. The promoter-probe vector pQF50 (Farinha & Kropinski, 1990), harbouring the reporter lacZ and the test gene, was constructed to study the transcriptional regulation with the upstream region of the gene of interest. The upstream region of each gene was obtained by PCR amplification based on the nucleotide sequences available in the GenBank database, i.e. accession numbers AB333783, AB091400 and AB204833 for the ADH-I, ADH-IIB and ADH-IIG gene clusters, respectively (Table 1⇑, Fig. 1⇓). The gene complementation study was carried out by cloning a target gene including its promoter into the pCM62 broad-host-range plasmid (Marx & Lidstrom, 2001). Transformation of the plasmids pQF50, pCM62 and their derivatives into P. putida HK5 was performed as described by Choi et al. (2006). DNA sequencing was done using an ABI PRISM 310 (PE Biosystems).
Organization of the qADH genes within the (a) ADH-I, (b) ADH-IIB and (c) ADH-IIG clusters of P. putida HK5, and plasmid construction. The arrows at the insertion sites indicate the transcriptional direction of the inserted kanamycin-resistance genes. For construction of the lacZ promoter-probe vectors (pQW plasmids), PCR fragments of promoter region (divergent arrows) were cloned into pQF50. Black arrows indicate the lacZ gene. pCM plasmids were constructed by cloning PCR fragments into pCM62. Stippled boxes indicate the promoter regions of qedA (PqedA) and exaE (PexaE); striped boxes indicate the lacZ promoter region (PlacZ) of pCM62. The consensus sequence-predicted positions of the ribosome-binding sites (rbs) are shown.
RT-PCR procedure.
Total RNA was isolated from P. putida HK5 cells grown to late-exponential phase on ethanol or 1-butanol (Chuang et al., 1993). The RT-PCR kit (mRNA Selective PCR kit, AMV) was obtained from TaKaRa, and used according to the manufacturer's instructions. The primers used for RT-PCR were designed based on the intergenic region of the two adjacent genes of interest to generate a PCR product of approximately 230–500 bp. A negative control, for the exclusion of contaminating genomic DNA amplification, was performed in each case using Taq polymerase without any reverse transcriptase.
β-Galactosidase assay.
P. putida HK5 containing pQF50-lacZ derivatives was grown overnight in 5 ml LB medium containing 200 μg piperacillin ml−1 at 30 °C on a rotary shaker at 200 r.p.m. Cell cultures (1 ml) were collected by centrifugation at 15 000 g for 5 min, washed with basal medium, and resuspended in an equal volume of basal medium. The cell suspension was then diluted fivefold in basal medium containing piperacillin and the desired alcohol (at 0.5 %, v/v, final concentration) and/or 20 mM glucose. After shaking the culture for 6 h at 30 °C, the β-galactosidase activity was determined with cells treated with chloroform, according to the procedure of Miller (1992).
Determination of ADH activity and protein assay.
Cells from 100 ml culture of OD600 0.8–1.0 were harvested, washed once with ice-cold saline, and resuspended in 3 ml 50 mM Tris/HCl buffer (pH 8.0). The suspension was passed through a French pressure cell at 16 000 p.s.i. (110 400 kPa) at 4 °C. The cell debris was removed by centrifugation at 15 000 g for 20 min, and the resulting supernatant was obtained as a crude fraction. Phenazine methosulfate (PMS) reductase activity was measured for ADH-I activity, and ferricyanide reductase activity was used for ADH-IIB and ADH-IIG activity assays, as described previously (Toyama et al., 1995). The rate of reduction of electron acceptor observed with substrate was subtracted from that obtained without substrate. Protein content was estimated by a modified method of Lowry (Dulley & Grieve, 1975) with serial dilutions of BSA (Sigma) as a standard.
RESULTS AND DISCUSSION
ADH-I gene cluster: promoter activities of structural genes, qedA(B) and qedC
Genes encoding ADH-I (qedA) and cytochrome c (qedC) are adjacent, but divergently transcribed. This cytochrome c homologue is reported to be an essential component of the ethanol oxidation system in P. aeruginosa (Schobert & Görisch, 1999). The upstream region of qedA of P. putida HK5 contains the putative E. coli σ70 promoter consensus sequence (Harley & Reynolds, 1987), i.e. −35[TTCCCG] and −10[TATCTG]. This promoter region of P. putida HK5 also exhibits high similarity to the putative promoter sequence of the qEDHs of P. putida KT2440 and Pseudomonas fluorescens Pf5.
Promoter-probe vectors, i.e. pQW-ADHI and pQW-CYT (Fig. 1a⇑), were constructed and used to monitor the promoter activities of qedA and qedC genes, respectively. In response to various alcohol inducers, both promoters exhibited similar transcriptional patterns, though to different magnitudes. Significant promoter activities were observed towards primary and secondary short-chain-length alcohols (C2–C4), diols and glycerol. The highest promoter activity was obtained with ethanol induction, while the activity diminished upon induction with longer-chain-length alcohols (C5–C8) (Fig. 2a⇓, b). Previous results, based upon the relative activities of ADH-I towards different substrates, indicated that C2–C8 primary alcohols were good substrates for the enzyme (Toyama et al., 1995). Nonetheless, our present work suggested that longer-chain-length alcohols act as repressors for the qedA and qedC promoters. In contrast, though the secondary alcohols, diols and glycerol have been reported to be poor substrates for ADH-I (Toyama et al., 1995), significant promoter activities of qedA and qedC with these inducers were observed (Fig. 2a⇓, b). These results suggested that short-chain-length alcohols (C2–C4) act as signal molecules for a specific regulator that controls transcription of qedA and qedC. Although the similar transcriptional patterns of the two promoters under alcohol induction suggested that qedA and qedC might be under the control of the same regulator, the expression level of qedC was higher than that of qedA under the same conditions. This directly contrasts with an earlier study of P. aeruginosa, in which the promoter activity of exaB (cytochrome c550) was reported to be significantly lower than that of exaA (qEDH) (Schobert & Görisch, 2001). The high expression level of qedC was anticipated, because P. putida HK5 has three types of qADHs, each of which requires cytochrome c for electron translocation during alcohol oxidation. In the present work, qedA and qedC promoters were found to likely be affected by a catabolite repression control by glucose, as the combination of glucose with ethanol led to a low induction of the qedA and qedC promoter activities (Fig. 2a⇓, b). Lactate, acetate and tricarboxylic acid cycle intermediates such as citrate and succinate were also found to act as catabolite repressors of qedA (data not shown).
Promoter activities of (a) qedA; pQW-ADHI, (b) qedC; pQW-CYT, (c) aldA; pQW-ALDA, (d) qbdBA; pQW-ADHIIB, and (e) qgdA; pQW-ADHIIG in P. putida HK5 after induction by various substrates (0.5 %, v/v, alcohols, 0.1 %, v/v, aldehyde or 20 mM glucose). Data are shown as the mean± sd and are derived from three independent experiments. Abbreviations: BS, basal medium only; M, methanol; E, ethanol; 1P, 1-propanol; 1B, 1-butanol; 1Pn, 1-pentanol; 1H, 1-heptanol; 1O, 1-octanol; 2P, 2-propanol; 2B, 2-butanol; iB, iso-butanol; Ba, benzyl alcohol; EDO, ethanediol; PDO, 1,2-propanediol; Gly, glycerol; Acd, acetaldehyde; Bud, butyraldehyde; G, glucose.
Downstream of qedA lies a pentapeptide repeated sequence, namely qedB (Promden et al., 2008). The homologue of qedB was also observed downstream of type I ADH genes in P. aeruginosa ATCC 17933 (gene PA1981). Although it has been presumed that this possible ORF is co-transcribed with the exaA gene of P. aeruginosa (Görisch, 2003), the RT-PCR results from the present work showed that in ethanol-grown cells, qedA and qedB were not stably transcribed as a single transcript. Moreover, neither a σ54- nor a σ70-promoter consensus sequence could be found in the qedAB intergenic region, and no promoter activity was attained with the qedAB intergenic region-containing construct (pQW-PEN) under the conditions tested (data not shown). Thus, according to these results, qedB may not be a true ORF.
ADH-IIB gene cluster: C3/C4 alcohols and butyraldehyde as the inducers of ADH and ADH-IIB expression
The quinohaemoprotein ADH-IIB gene cluster of P. putida HK5 consists of the ADH gene (aldA), an ORF of unknown function (qbdB), and the ADH-IIB structural gene (qbdA) (Fig. 1b⇑) (Toyama et al., 2003). The RT-PCR result suggested that the qbdBA genes were co-transcribed from a promoter upstream of qbdB (data not shown). A σ54-consensus sequence, with the conserved sequence (underlined) centred around −24 and −12, respectively (Barrios et al., 1999), was found upstream of aldA [TGGCACAA]GGG[TTGCT] and qbdBA [TGGCACGA]AGC[CTGCT]. The promoter-probe vectors pQW-ALDA and pQW-ADHIIB were constructed (Fig. 1b⇑), and the transcriptional expression levels in response to various alcohol substrates were evaluated. The results, summarized in Fig. 2⇑(c, d), suggest that glucose acted as a catabolite repressor of the aldA and qbdBA promoters. The promoter activities of aldA and qbdBA demonstrated a similar induction pattern in response to primary and secondary C3 and C4 alcohols as well as butyraldehyde induction (Fig. 2c⇑, d). Since the qedA and qbdBA promoters were differentially induced by alcohols that differed in their chain lengths, their activities may be either controlled by different regulators or influenced by different regulatory factors. Nonetheless, it cannot be excluded that these regulators have broad specificity to other short-chain alcohols, resulting in the concurrent expression of ADH-I and ADH-IIB (Promden et al., 2008; Toyama et al., 1995).
ADH-IIG gene cluster: (S)-(+)-1,2-propanediol is a specific inducer
The ADH-IIG gene cluster consists of qgdA, a structural gene of quinohaemoprotein ADH-IIG, followed by the aldB gene, encoding an NAD-aldehyde dehydrogenase in the same direction, and orf1, encoding a hypothetical protein oriented in the opposite direction and located upstream of qgdA (Toyama et al., 2005). The activities of the divergent qgdA and orf1 promoters were investigated when induced with alcohols and aldehydes. The lacZ promoter-probe vector of orf1, i.e. pQW-ORFG1 (Fig. 1c⇑), showed a low constitutive activity with all substrates tested, suggesting that orf1 might not be directly involved in alcohol utilization in P. putida HK5 (data not shown). Since the σ54-dependent promoter consensus sequence, [TGGCATGG]CGG[TTGCG], was found at the upstream region of the qgdA gene, the qgdA promoter-probe vector, pQW-ADHIIG (Fig. 1c⇑), was constructed. The highest promoter activity of pQW-ADHIIG was observed after growth on (S)-(+)-1,2-propanediol. Ethanediol, glycerol and (R)-(−)-1,2-propanediol were weaker inducers, whereas ethanol, 1-butanol and butyraldehyde were ineffective inducers (Fig. 2e⇑). It should be noted that (S)-(+)-1,2-propanediol not only is a strong inducer for the qgdA promoter, but is also the preferred substrate of ADH-IIG with a higher substrate specificity (Km=0.055 mM) than that of (R)-(−)-1,2-propanediol (Km=3.32 mM) (Toyama et al., 2005). Interestingly, even though 1-butanol was a substrate of ADH-IIG (Km=0.043 mM) (Toyama et al., 1995, 2005), it was not found to be an effective signal molecule for the qgdA promoter. These results suggest that the qgdA promoter is specifically regulated by (S)-(+)-1,2-propanediol and that different regulators control the activities of the qedA, qbdB and qgdA promoters.
Roles of regulatory genes, exaE and agmR, for alcohol oxidation in P. putida HK5
Downstream of qedA (ADH-I structural gene) lie exaE and agmR, two regulatory genes that encode likely DNA-binding response regulators, and orf9, which encodes a hypothetical protein similar to that in P. putida KT2440. The RT-PCR results suggested that orf9 and agmR did not form an operon (data not shown). In fact, no orf9 transcriptional product was detected under ethanol induction. Although the role of orf9 has not yet been conclusively determined, a previous report has shown that disruption of the orf9 sequence adversely affects the expression of ADH-I (Promden et al., 2008).
An earlier study of the genetic regulation of qADH in P. aeruginosa ATCC 17933 revealed a two-component regulatory system composed of the sensor kinase ExaD and the DNA binding response regulatory protein ExaE (Schobert & Görisch, 2001). However, in P. putida HK5, the homologous gene corresponding to exaD could not be observed within the 10 kb ADH-I cluster gene fragment (Promden et al., 2008). Nevertheless, the activity of the exaE promoter was investigated in this study using the promoter-probe vector pQW-EXAE, which was constructed with the putative σ70 promoter region −35[TCGACA] and −10[AGTAGT], upstream of the exaE gene (Fig. 1a⇑). The results revealed that the exaE promoter was transcribed to an approximately similar level with all alcohols and the aldehyde tested, but that this expression level was relatively low (Fig. 3⇓). In addition, it was found that the exaE promoter was likely to be affected by catabolite repression control by glucose, as the combination of glucose with ethanol led to a low activity of the exaE promoter (Fig. 3⇓). To investigate the promoter activity of agmR, the plasmid pQW-AGMR was used (Fig. 1a⇑). The results are consistent with the notion that the agmR promoter is constitutively transcribed (Fig. 3⇓).
Promoter activities of exaE; pQW-EXAE (white bars) and agmR; pQW-AGMR (grey bars) in P. putida HK5 cells after induction by various substrates (0.5 %, v/v, alcohols, 0.1 %, v/v, aldehyde or 20 mM glucose). Data are shown as the mean± sd and are derived from three independent experiments. Abbreviations: BS, basal medium only; E, ethanol; 1B, 1-butanol; PDO, 1,2-propanediol; Gly, glycerol; Acd, acetaldehyde; G, glucose.
Involvement of exaE and agmR in ADH-I and cytochrome c expression
To investigate the influence of exaE and agmR on ADH-I expression, the ADH-I-defective mutants were tested for complementation with plasmids harbouring individually the qedA, exaE or agmR genes. Two plasmids harbouring the qedA gene were constructed (Table 1⇑) for the complementation study, i.e. pCM-ADHI and pCM-ADHIZ, in which qedA was under the control of the qedA promoter (PqedA) and the lacZ promoter (PlacZ), respectively. Each plasmid was transformed into P. putida HK5 wild-type and mutant strains. The transformants were then grown on ethanol and ADH-I activity was determined (Table 2⇓). Introduction of pCM-ADHI into HK5 WT (wild-type) cells increased the observed ADH-I activity by 2.5-fold, while it fully restored the activity in the qedA : : Kmr mutant cells (Table 2⇓). The complementation results for pCM-ADHIZ in either wild-type or mutant cells indicated that qedA was expressed at a low level under the control of the lacZ promoter. Previous work has demonstrated that the insertion of a kanamycin-resistance cassette (Kmr) in qedA, exaE and agmR genes completely eliminated ADH-I activity (Promden et al., 2008). Complementation with either pCM-ADHI or pCM-ADHIZ, but not pCM-AGMRZ, in exaE : : Kmr could only partially restore the ADH-I activity, while complementation with pCM-EXAE (under PexaE control) and pCM-EXAEZ (under PlacZ control) resulted in ∼80–90 % ADH-I activity relative to that seen in wild-type cells (Table 2⇓). This result indicates a likely important role for ExaE as a regulatory protein involved in PqedA transcriptional activity. Thus, the potential role of AgmR as another regulatory protein was also examined. In the presence of ExaE alone, in agmR : : Kmr/pCM-EXAEZ, qedA expression was not detected. Complementation of agmR using the agmR : : Kmr/pCM-AGMRZ mutant resulted in a 95 % restoration of ADH-I activity relative to wild-type cells, indicating that the presence of both ExaE and AgmR is necessary for the transcriptional regulation of PqedA. Moreover, the lacZ promoter-probe vector harbouring the qedA promoter, pQW-ADH-I (Fig. 1a⇑), was examined in the wild-type, exaE : : Kmr and agmR : : Kmr mutant cells with ethanol induction (Fig. 4b⇓). The promoter activity of the qedA construct in wild-type cells was 835±136 Miller units compared with no detectable activity in exaE : : Kmr and agmR : : Kmr mutant strains, supporting the hypothesis that both ExaE and AgmR are necessary for the transcription of PqedA.
Promoter activities of (a) qedC; pQW-CYT, (b) qedA; pQW-ADHI, (c) exaE; pQW-EXAE, (d) qbdBA; pQW-ADHIIB, (e) aldA; pQW-ALDA, and (f) qgdA; pQW-ADH-IIG in HK5 wild-type and exaE : : Km and agmR : : Km mutant P. putida strains when induced with 0.5 % (v/v) ethanol (a–c), butanol (d–e) or 1,2-propanediol (f). Data are shown as the mean± sd and are derived from three independent experiments
Specific activities of three ADHs from P. putida HK5 wild-type, mutant and complemented mutant strains
Data shown are mean±sd, and are derived from at least three independent replicates. WT, wild-type.
To investigate the involvement of exaE and agmR in regulating the expression of genes under the cytochrome c promoter (PqedC), the lacZ promoter-probe vector harbouring the qedC promoter, pQW-CYT (Fig. 1a⇑), was constructed and its expression level was examined in wild-type, and exaE : : Kmr and agmR : : Kmr mutant strains, under induction with ethanol (Fig. 4a⇑). The promoter activity of qedC in wild-type cells (5704±740 Miller units) was significantly reduced in the exaE : : Kmr mutant (708±262 Miller units) and completely abolished (no detectable activity) in the agmR : : Kmr strain. The complementation of exaE (pCM-EXAE) and agmR (pCM-AGMRZ) in the presence of pQW-CYT could partially recover qedC promoter activity by ∼30 % (1725±80 Miller units) and ∼16 % (955±100 Miller units), respectively, relative to that seen in the wild-type cells (data not shown). These results are consistent with the notion that both ExaE and AgmR are necessary for the transcription of not only PqedA but also PqedC. The regulation of ethanol oxidation by ExaE and AgmR regulators in P. putida HK5 is partly distinct from that in P. aeruginosa. In P. aeruginosa, ExaDE regulators regulate only the expression of ethanol dehydrogenase, but not that of cytochrome c550 (Görisch, 2003; Schobert & Görisch, 2001). In the present work, it was shown that the AgmR regulator governed the activity of the exaE promoter. When the agmR gene was disrupted, the exaE promoter activity was almost completely eliminated (Fig. 4c⇑), suggesting that AgmR is a general regulator of quinoprotein ethanol oxidation in P. putida HK5, similar to the situation in P. aeruginosa ATCC 17933 (Gliese et al., 2004).
Involvement of agmR, but not exaE, in ADH-IIB and ADH-IIG expression
The disruption of agmR completely abolished the ADH-I activity, while the activities of ADH-IIB and ADH-IIG were reduced by 38 and 33 %, respectively (Table 2⇑). Complementation of the agmR gene using pCM-AGMRZ fully restored the ADH-IIB and ADH-IIG activities, and neither enzyme activity was affected when the exaE gene was disrupted.
In addition, the lacZ promoter-probe constructs harbouring the qbdBA, aldA and qgdA promoters (Fig. 1b⇑, c) were examined in the wild-type, and exaE : : Kmr and agmR : : Kmr mutant strains, under induction with butanol (for qbdBA and aldA promoter-probe vectors) or 1,2-propanediol (for the qgdA promoter-probe vector) (Fig. 4d–f⇑). The promoter activities of qbdBA (pQW-ADHIIB) and aldA (pQW-ALDA) in the wild-type and exaE : : Kmr mutant strains were not significantly different (Fig. 4d⇑, e). For qgdA (pQW-ADHIIG), the promoter activities were higher. Nevertheless, for all the three promoters, the activities were significantly reduced in agmR : : Kmr mutant strains. These results suggest that the AgmR, but not the ExaE, regulator plays a major role in the regulation of ADH-IIB and ADH-IIG expression. The findings agree well with results from a previous report on total ADH activities demonstrated by native PAGE with in situ enzyme activity staining (Promden et al., 2008).
Hypothetical scheme of the regulatory network controlling the alcohol oxidation system of P. putida HK5
P. putida HK5 expresses three distinct qADHs, depending on the type of alcohol presented as the signal molecule. The regulatory network scheme controlling the transcriptional expression of these three ADH genes and alcohol oxidation systems in P. putida HK5 was derived from the properties of exaE- and agmR-disrupted mutants and the promoter activities of each ADH gene cluster examined under various growth and induction conditions (Fig. 5⇓). The results indicated that AgmR is a primary, although not sole, regulator involved in all three alcohol oxidation systems. The product of agmR also regulates the transcription of exaE. ExaE subsequently influences the transcription of qedA (for ADH-I expression), but ExaE cannot direct the transcription of qedA in the absence of AgmR. Both AgmR and ExaE are therefore annotated as response regulators, for which corresponding sensor kinases are generally required. Since a sensor kinase homologue of the AgmR regulator is not currently known (Gliese et al., 2004), and the exaD gene encoding the histidine sensor kinase of ExaE (Schobert & Görisch, 2001) could not be detected in P. putida HK5 cloned fragments, there may be other sensing regulatory factor(s) which respond to ethanol, and the primary and secondary C3 and C4 alcohols involved. As for the transcription of qbdBA (ADH-IIB) and qgdA (ADH-IIG), AgmR played a partial role in governing their transcription. In this case, AgmR may be differentially controlled by other promoter(s), if any. In addition, we cannot rule out that besides AgmR and ExaE, other regulator(s) or effector(s) could be involved. For instance, the involvement of a protein X (Fig. 5⇓), which responds to the primary and secondary C3- and C4-chain-length alcohols, butyraldehyde and 1,2-propanediol, in directing qbdBA transcription, and a protein Y (Fig. 5⇓), which senses the stereospecific (S)-(+)-1,2-propanediol for controlling qgdA transcription, could be proposed.
Hypothetical scheme of the regulatory network controlling transcription of the quinoprotein alcohol oxidation systems in P. putida HK5. The scheme is derived from results obtained with exaE and agmR regulatory mutants, promoter activities and complementation studies. Genes are indicated by letters in arrows, whilst the hypothetical regulatory genes proposed for ADH-IIB and ADH-IIG are indicated by X and Y, respectively.
Acknowledgments
Financial support from the Thailand Research Fund through the Royal Golden Jubilee PhD Program (grant no. PHD/0030/2548) to W. P. is acknowledged.