Regulation of Glutarate Catabolism by GntR Family Regulator CsiR and LysR Family Regulator GcdR in Pseudomonas putida KT2440

Glutarate is an attractive dicarboxylate with various applications. Clarification of the regulatory mechanism of glutarate catabolism could help to block the glutarate catabolic pathways, thereby improving glutarate production through biotechnological routes. Glutarate is a toxic metabolite in humans, and its accumulation leads to a hereditary metabolic disorder, glutaric aciduria type I. The elucidation of the functions of CsiR and GcdR as regulators that respond to glutarate could help in the design of glutarate biosensors for the rapid detection of glutarate in patients with glutaric aciduria type I. In addition, CsiR was identified as a regulator that also regulates l-2-HG metabolism. The identification of CsiR as a regulator that responds to l-2-HG could help in the discovery and investigation of other regulatory proteins involved in l-2-HG catabolism.

Glutarate is a metabolic intermediate in the catabolism of several amino acids (such as L-lysine, L-hydroxylysine, and L-tryptophan) and aromatic compounds (such as nicotinate and benzoate) (11)(12)(13)(14). The classic glutarate catabolism pathway is the glutarylcoenzyme A (glutaryl-CoA) dehydrogenation pathway, where glutaryl-CoA dehydrogenase (GcdH) is the key enzyme (11). In this pathway, glutarate is first converted to glutaryl-CoA, followed by the dehydrogenation and decarboxylation of glutaryl-CoA by GcdH to produce crotonyl-CoA (15)(16)(17). Crotonyl-CoA can be converted to two molecules of acetyl-CoA, which are then channeled into the tricarboxylic acid (TCA) cycle (11,18). Recently, a glutarate hydroxylation pathway with glutarate hydroxylase (CsiD [carbon starvation-induced protein]) and L-2-hydroxyglutarate (L-2-HG) oxidase (LhgO) as its key enzymes was identified in both Pseudomonas putida (19) and Escherichia coli (20). CsiD is capable of converting glutarate and 2-ketoglutarate (2-KG) into L-2-HG and succinate. The L-2-HG produced is subsequently converted to 2-KG by LhgO. These two pathways have been found to cooperate in glutarate catabolism in P. putida KT2440 (19). However, the nature of the regulatory mechanism of glutarate catabolism in P. putida KT2440 and the issue of whether there is an interaction between the processes of regulation of these two pathways have not yet been studied.
In this study, the regulatory mechanism of glutarate catabolism was studied in P. putida KT2440, a model organism containing two glutarate catabolic pathways (19,34,35). A GntR family protein, CsiR, and a LysR family protein, GcdR, were identified as the regulators of the glutarate hydroxylation pathway and the glutaryl-CoA dehydrogenation pathway, respectively. There is no cross-regulation between these two pathways. In addition, it was confirmed that CsiR is also involved in L-2-HG catabolism and uses both L-2-HG and glutarate as its effectors. This report improves our understanding of the regulatory mechanisms of glutarate and L-2-HG catabolism.
P. putida KT2440 was cultured in minimal salt medium (MSM) supplemented with glutarate or pyruvate as the sole carbon source, and the transcription levels of the genes mentioned above were detected by reverse transcription-PCR (RT-PCR). Transcription of csiD and lhgO (involved in the glutarate hydroxylation pathway) was induced by glutarate (19), as was transcription of gcdH and gcoT (involved in the glutaryl-CoA dehydrogenation pathway) (see Fig. S1A in the supplemental material). The results of the carbon source feeding experiments indicate constitutive expression of genes csiR and gcdR (Fig. S1B). The relative expression levels of these genes were then analyzed by real-time quantitative PCR (qPCR). As shown in Fig. 1C, genes csiD, lhgO, gcdH, and gcoT were induced by glutarate. Additionally, the enzymatic activities of CsiD and GcdH in the glutarate medium were also found to be higher than those in pyruvate medium (Fig. 1D).
CsiR represses the transcription of csiD and lhgO. The transcriptional organization of csiR-csiD-lhgO was then assayed by RT-PCR. The intergenic region of csiD-lhgO could be amplified whereas the intergenic region of csiR-csiD could not be amplified (Fig. S2), indicating that genes csiD and lhgO were cotranscribed but that the transcript corresponding to csiD was different from that corresponding to csiR. The transcriptional start site (TSS) of the csiD gene was determined using the rapid amplification of cDNA ends (RACE) method. The TSS was identified as a guanine (G) residue found 62 bp upstream of the csiD start codon, with the putative Ϫ10 (TATTTT) and Ϫ35 (TAGACA) regions separated by 17 bp (Fig. 2A).
To characterize the csiD operon promoter, the 134-bp DNA fragment upstream of TSS (G) was fused to lacZ of promoter probe plasmid pME6522 to generate pME6522-P csiD (see Table S1 in the supplemental material). The resulting plasmid was transferred into P. putida KT2440 and P. putida KT2440 (ΔcsiR) (Table S1) by electroporation, and the promoter activity of P csiD was measured by ␤-galactosidase assays after culturing the resulting strains in MSM with glutarate or pyruvate as the sole carbon source. In the presence of glutarate, the promoter activity of P csiD was significantly higher than that of pyruvate (Fig. 2B), indicating that the P csiD fragment contains a functional glutarate-responsive promoter. Additionally, the promoter activity of P csiD in P. putida KT2440 (ΔcsiR) cultured in glutarate was about twice as high as that of P. putida KT2440, which indicated that CsiR represses the transcription of csiD.
To determine whether CsiR directly interacts with the csiD promoter region, the His 6 -tagged CsiR protein of P. putida KT2440 was expressed in E. coli BL21(DE3) and The corresponding steps of the glutarate degradation pathways are also shown. (C) qPCR analysis of the genes in csiR-csiD-lhgO and gcdR-gcdH-gcoT gene clusters. The relative expression levels of six genes, csiR, csiD, lhgO, gcdR, gcdH, and gcoT, were measured using RNA extracted from P. putida KT2440 grown in MSM with glutarate or pyruvate as the sole carbon source. The gene expression levels are represented as expression ratios of the indicated genes in glutarate medium versus pyruvate medium, normalized to 16S rRNA. (D) The activities of CsiD and GcdH in P. putida KT2440 grown in MSM with glutarate or pyruvate as the sole carbon source. Data shown are means Ϯ standard deviations (SD) (n ϭ 3 independent experiments). *, P Ͻ 0.05 in two-tailed t test; **, P Ͻ 0.01 in two-tailed t test; ns, no significant difference (P Ն 0.05 in two-tailed t test).
Regulation of Glutarate Catabolism in P. putida KT2440 ® purified by Ni-chelating chromatography. On the basis of the results of gel filtration and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), CsiR behaved primarily as a dimer ( Fig. 2C; see also Fig. S3). Electrophoretic mobility shift assays (EMSAs) were performed using the csiR-csiD intergenic region DNA fragment (F1) and purified CsiR. As shown in Fig. 2D, CsiR bound to F1 in a concentration-dependent (E) DNase I footprinting analysis of CsiR binding to the csiD promoter region. F1 was labeled with FAM dye and incubated with 1 g CsiR (blue line) or without CsiR (red line). The region protected by CsiR from DNase I cleavage is indicated with a dotted box. *, P Ͻ 0.05 in two-tailed t test; ***, P Ͻ 0.001 in two-tailed t test. manner and completely shifted the DNA fragment at a 6-fold molar excess. In addition, four DNA-CsiR complexes were detected at low concentrations of CsiR, while the bands shifted to form a more diffuse complex at higher concentrations (Ͼ60 nM) of CsiR. These results showed that CsiR can bind to the upstream region of the csiD operon.
A DNase I footprinting assay was also performed using purified CsiR. The csiR-csiD intergenic region DNA fragment (F1) end labeled with 6-carboxyfluorescein (FAM) on the noncoding strand was mixed with CsiR protein and was then digested with DNase I. Two clearly protected regions were observed (Fig. 2E). One of the protected regions contained the Ϫ10 and Ϫ35 regions relative to the TSS of csiD, which correlated with the fact that CsiR is a repressor.
GcdR activates the transcription of gcdH. The transcriptional organization of gcdR-gcdH-gcoT was also assayed by RT-PCR. The intergenic region of gcdH-gcoT was amplified, while the intergenic region of gcdR-gcdH could not be detected, indicating that genes gcdH and gcoT were cotranscribed but that the transcript corresponding to gcdH is different from that corresponding to gcdR (Fig. S4). The TSS of the gcdH operon was a G residue found 37 bp upstream of the gcdH start codon, with the putative Ϫ10 (TAGGCT) and Ϫ35 (TTGTCG) regions separated by 17 bp (Fig. 3A).
The 124-bp DNA fragment upstream of TSS (G) of gcdH was fused to lacZ of pME6522 to generate pME6522-P gcdH (Table S1). The resulting plasmid was transferred into P. putida KT2440 and P. putida KT2440 (ΔgcdR) by electroporation, and the promoter activity of P gcdH was measured by ␤-galactosidase assays. P. putida KT2440 harboring pME6522-P gcdH showed a higher ␤-galactosidase activity under conditions of culturing in the presence of glutarate (Fig. S5), which confirmed that the P gcdH fragment contains the glutarate-inducible promoter. However, very low activity (Ͻ40 Miller units) was detected in P. putida KT2440 (ΔgcdR) harboring pME6522-P gcdH in the presence of glutarate (Fig. S5). The gcdR gene in P. putida KT2440 (ΔcsiD) was then also disrupted to create another mutant strain. The mutant strain, P. putida KT2440 (ΔcsiD), harboring only the glutaryl-CoA dehydrogenation pathway, was still able to grow on glutarate, while the P. putida KT2440 (ΔcsiD ΔgcdR) strain lost the ability to utilize glutarate (Fig. 3B). These results indicated that GcdR is a transcriptional activator and is indispensable for the transcription of gcdH.
The His 6 -tagged GcdR of P. putida KT2440 was expressed in E. coli BL21(DE3) and purified. On the basis of the results of gel filtration and SDS-PAGE, GcdR behaved primarily as a tetramer ( Fig. 3C; see also Fig. S6). GcdR could not completely bind to the gcdR-gcdH intergenic region in EMSAs. The intergenic region was extended by adding 50 bp both upstream and downstream (the resulting region was named F2). GcdR bound to F2 in a concentration-dependent manner and completely shifted the DNA fragment at 10-fold molar excess (Fig. 3D). A DNase I footprinting assay was then performed using purified GcdR and F2. Two clearly protected regions were determined (Fig. 3E). One of the protected regions contained the T-N 11 -A consensus binding motif of LysR-type transcriptional regulators (36,37), 5=-GTGACAAAATATCAC-3=, and an interrupted inverted repeat sequence, 5=-AGTGA-N 7 -TCACT-3= (the start and end of the relevant sequences are indicated in bold).
CsiR and GcdR regulate their own target pathways. In order to determine whether there was cross-regulation between CsiR and GcdR, we used lacZ transcriptional fusions to measure the impact of csiR and gcdR mutants on the expression of the gcdH and csiD genes, respectively. As shown in Fig. 4A, the disruption of gcdR had no effect on the activity of the P csiD promoter, and the disruption of csiR had no effect on the activity of the P gcdH promoter too. It was shown that the regulator of one glutarate metabolic pathway could not activate or inhibit the transcription of genes in the other pathway, indicating that the two pathways are likely to be independent of each other and that there is no cross-regulation.
The gcdR gene in P. putida KT2440 (ΔgcdH) was then also disrupted. P. putida KT2440 (ΔgcdH) harbored only the glutarate hydroxylation pathway. The levels of growth and glutarate consumption of P. putida KT2440 (ΔgcdH ΔgcdR) were the same as those seen with P. putida KT2440 (ΔgcdH) (Fig. 4B), further suggesting that GcdR may not regulate the glutarate hydroxylation pathway. Results of EMSAs also indicated that CsiR could not directly interact with the gcdH promoter region (Fig. 4C) and that GcdR could not directly interact with the csiD promoter region (Fig. 4D). Taken together, these results suggested that CsiR and GcdR regulate their own target pathways.
Characterization of the effectors of CsiR and GcdR. The effects of glutarate and other compounds involved in the glutarate metabolism on the activity of the csiD promoter were evaluated. Plasmid pME6522-P csiD was transferred into P. putida KT2440 (ΔdavT Δalr) by electroporation, and the resulting strain was incapable of converting L-lysine and 5-aminovalerate into glutarate (Fig. 5A). The promoter activities of P csiD were measured using ␤-galactosidase assays, and the wild-type strain and P. putida KT2440 (ΔdavT Δalr) harboring pME6522-P csiD were cultured in MSMs with 2.5 g liter Ϫ1 pyruvate and different compounds as the carbon sources. Significant levels (324 to 1,015 Miller units) of promoter activities were detected in P. putida KT2440 harboring pME6522-P csiD in the presence of L-lysine, 5-aminovalerate, glutarate, and L-2-HG (Fig. 5B), whereas the ␤-galactosidase activities were observed only when P. putida KT2440 (ΔdavT Δalr) harboring pME6522-P csiD was grown in the presence of glutarate and L-2-HG (Fig. 5C). These results suggested that glutarate and L-2-HG can induce the P csiD promoter whereas L-lysine and 5-aminovalerate cannot induce the P csiD promoter.
The effects of L-lysine, 5-aminovalerate, glutarate, and L-2-HG on CsiR binding to the csiD promoter region were assessed by EMSAs. L-Lysine and 5-aminovalerate had no effect on the binding of CsiR to F1, whereas glutarate and L-2-HG prevented the binding (Fig. 5D). The effects of glutarate and L-2-HG (10 mM, 20 mM, 40 mM, 60 mM, 80 mM, 100 mM, and 120 mM) on the capacity of binding of CsiR to F1 were further analyzed by EMSAs. When the concentrations of glutarate and L-2-HG were increased, the amount of the CsiR-DNA complex decreased and the amount of the free DNA increased (Fig. S7). Therefore, glutarate and L-2-HG are the effectors of CsiR.
As for GcdR, L-lysine, 5-aminovalerate, and glutarate could induce the gcdH promoter in P. putida KT2440 harboring pME6522-P gcdH (Fig. 6A). However, only glutarate could induce the gcdH promoter in P. putida KT2440 (ΔdavT Δalr) harboring pME6522-P gcdH , while L-lysine and 5-aminovalerate could not (Fig. 6B). Thus, glutarate is the effector of GcdR. Regulation of Glutarate Catabolism in P. putida KT2440 ® CsiR regulates the catabolism of L-2-HG. As CsiR was able to use both glutarate and L-2-HG as the effectors, it was speculated that CsiR also regulates the metabolism of L-2-HG. Although P. putida KT2440 was able to use L-2-HG as the sole carbon source, it lost the ability to utilize L-2-HG after the lhgO gene was deleted (Fig. 7A), suggesting that lhgO was indispensable for L-2-HG utilization. The disruption of csiR significantly increased the growth rate and L-2-HG consumption rate of P. putida KT2440 in MSM with L-2-HG as the sole carbon source (Fig. 7B), implying that CsiR may repress the utilization of L-2-HG.
The activity of LhgO in P. putida KT2440 was detected using the wild-type strain and P. putida KT2440 (ΔcsiR) cultured in media with different compounds as the sole carbon sources. L-2-HG, L-lysine, 5-aminovalerate, and glutarate were found to induce the expression of LhgO, whereas 2-KG, glucose, succinate, and pyruvate did not (Fig. 7C). Moreover, the enzymatic activity of LhgO in P. putida KT2440 (ΔcsiR) was detected in the media with all of the compounds described above as the sole carbon sources (Fig. 7D).
The growth rate and L-2-HG consumption rate of P. putida KT2440 (ΔgcdR) were consistent with those of the wild-type strain when L-2-HG was used as the sole carbon source (Fig. S8A). When P. putida KT2440 (ΔgcdR) was cultured in MSM with different  Regulation of Glutarate Catabolism in P. putida KT2440 ® compounds as the sole carbon sources, the activity of LhgO in all the tested compounds was consistent with that of the wild-type strain (Fig. S8B). These results indicate that CsiR regulates the catabolism of L-2-HG, while GcdR does not.

DISCUSSION
On the basis of the results described above, we proposed a model of the regulation of glutarate and L-2-HG catabolism in P. putida KT2440 (Fig. 8). The catabolism of glutarate is regulated by CsiR and GcdR, which control the glutarate hydroxylation pathway and glutaryl-CoA dehydrogenation pathway, respectively. CsiR is a transcriptional regulator in the GntR family and can specifically bind to the csiD promoter region and repress the transcription of the csiD and lhgO genes. GcdR is a transcriptional regulator in the LysR family and can specifically bind to the gcdH promoter region and activate the transcription of gcdH.
The catabolism of glutarate in P. aeruginosa PAO1 depends on GcdH, whose expression is under the control of the GcdR transcriptional activator (38). The catabolism of glutarate in E. coli depends on CsiD and LhgO (20). The expression of csiD in E. coli is significantly upregulated during carbon starvation (39). However, the two pathways cooperate in glutarate catabolism in P. putida KT2440 and both GcdH and CsiD are induced during carbon starvation (19). In this study, it was found that two regulators, CsiR and GcdR, control the two pathways described above in P. putida KT2440, respectively. CsiR cannot interact with the gcdH promoter region (Fig. 4C) and has no effect on the transcription of gcdH (Fig. 4A). Similarly, GcdR cannot interact with the csiD promoter region (Fig. 4D) and has no effect on the transcription of csiD (Fig. 4A). In contrast to GcdH, which is present universally in Pseudomonas species, CsiD and LhgO may be acquired via horizontal gene transfer and are sporadically distributed in only 25 strains of Pseudomonas (19). The independence of the processes of evolution of the two pathways may be the cause for their independent regulation. However, the absence of cross-regulation between the two pathways does not eliminate the possibility that CsiR and GcdR can also regulate other genes in addition to their respective targets in glutarate catabolism. The potential alternative binding sites of CsiR and GcdR might be identifiable through transcriptome-based bioinformatic analysis and successive experimental validation.
GcdR uses only glutarate, the substrate of the glutaryl-CoA dehydrogenation pathway, as its effector. However, CsiR in P. putida KT2440 uses two effectors, the substrate glutarate and the intermediate L-2-HG of the glutarate hydroxylation pathway (Fig. 8). The effector of the transcriptional repressor protein is usually the substrate of the operon or a specific intermediate in the metabolism of the substrate. For instance, the effector of the L-lactate operon in E. coli is the substrate L-lactate (40). The effector of the 2,3-butanediol operon in P. aeruginosa PAO1 is the metabolic intermediate acetaldehyde (41). Although the effector promiscuity of CsiR may be due to the structural similarity of L-2-HG and glutarate, it is also possible that the response of CsiR to L-2-HG has a physiological significance.
Interestingly, P. putida KT2440 can use L-2-HG as the sole carbon source for growth and the utilization of L-2-HG depends on lhgO, a gene in the csiD-lhgO operon regulated by CsiR (Fig. 7A). L-2-HG also has some important physiological functions, including helping cells adapt to hypoxia, enhancing immunity, and metabolizing several compounds (19,(29)(30)(31)(32). In addition to a metabolic intermediate that can be produced during the glutarate metabolism, L-2-HG can also be produced from the reduction of 2-KG by the promiscuous catalytic activity of L-malate dehydrogenase and L-lactate dehydrogenase (25). The excessive accumulation of L-2-HG can be toxic to cells (22,42,43). Under conditions where L-2-HG but not glutarate is present, using L-2-HG as the effector of CsiR could help P. putida KT2440 to quickly sense L-2-HG present in habitats or produced by the intracellular metabolism to regulate the utilization of L-2-HG. CsiR is the first regulator that has been identified as being involved in the L-2-HG catabolism and that uses L-2-HG as its effector. This finding could be helpful in the identification of the regulatory proteins of L-2-HG catabolism in other species.
Glutarate is a five-carbon dicarboxylic acid with important industrial applications (1,2). In recent years, increased attention has been paid to the biotechnological production of glutarate. For example, glutarate can be produced by the four-step degradation of L-lysine (1,2,8,44). In addition, glutarate can be produced through the reverse adipate degradation pathway or ␣-keto acid carbon chain extension pathway (3,10). The elucidation of the regulatory mechanism of the glutarate metabolism could help to block the glutarate metabolic pathway, thereby improving glutarate production through biotechnological routes. Glutaric aciduria type I is an inherited metabolic disorder (45,46). In most cases, the diagnosis of glutaric aciduria type I is established biochemically by the detection of glutarate (47,48). The elucidation of the regulatory response to glutarate could also help to develop glutarate biosensors and a rapid detection method for glutarate. Recently, both CsiR and GcdR have been used in the development of glutarate biosensors (49). In this study, we confirmed that CsiR uses both glutarate and L-2-HG as effectors. Thus, GcdR may be a more suitable regulatory protein for use in construction of these biosensors.
In summary, we have demonstrated that the catabolism of glutarate in P. putida KT2440 is regulated by the glutarate hydroxylation pathway regulator, CsiR, and by the glutaryl-CoA dehydrogenation pathway regulator, GcdR. The two metabolic pathways are independent of each other in regulation. CsiR is a special transcriptional repressor with two effectors, i.e., glutarate and L-2-HG. It is also the first regulatory protein to be identified in the regulation of the catabolism of L-2-HG.

MATERIALS AND METHODS
Bacterial strains and culture conditions. The bacterial strains and plasmids used in this study are listed in Table S1 in the supplemental material. E. coli was grown in Luria-Bertani (LB) medium at 37°C. P. putida KT2440 and its derivatives were cultivated in minimal salt medium (MSM) supplemented with different compounds as the sole carbon source at 30°C. If necessary, antibiotics were used at the following concentrations: kanamycin, 50 g ml Ϫ1 ; ampicillin, 100 g ml Ϫ1 ; and tetracycline, 30 g ml Ϫ1 . Cell growth was monitored by measuring turbidity at 600 nm.
RT-PCR and qPCR. P. putida KT2440 and its derivatives were cultivated in MSM supplemented with the appropriate carbon sources. Total bacterial RNA was purified by the use of an RNAprep Pure cell/bacteria kit (Tiangen Biotech, China) according to the manufacturer's directions. Contaminating DNA in the RNA preparations was removed by the use of RNase-free DNase I (TransGen, China). Synthesis of cDNA was performed using Superscript II reverse transcriptase (TransGen, China).
For transcriptional and cotranscriptional analysis, RT-PCR analyses were performed using mRNAs of P. putida KT2440 cells cultured in MSM and the appropriate primers. RT-qPCR was performed by the use of TransStart Top Green qPCR SuperMix (TransGen, China) and a LightCycler 480 system (Roche). The mbio.asm.org 11 relative levels of expression of the genes were calculated using the threshold cycle (2 ϪΔΔCT ) method (50). The results were normalized to the 16S rRNA gene level. Enzymatic assays of CsiD, GcdH, and LhgO. P. putida KT2440 was grown to mid-log stage in MSMs supplemented with the appropriate compounds as the sole carbon sources at 200 rpm and 30°C. Cells were harvested, centrifuged, washed, and resuspended in phosphate-buffered saline (PBS). Cells were adjusted to a final optical density at 600 nm (OD 600 ) of 20 and sonicated with a Sonics sonicator (500 W, 20 KHz). The homogenate was centrifuged at 13,000 ϫ g for 5 min at 4°C, and the supernatants were used as the crude cell extracts for activity measurement.
The activity of CsiD was measured at 30°C in 500 l of a reaction solution containing 20 mM imidazole (pH 6.7), 1 mM glutarate, 1 mM 2-KG, 0.4 mM ascorbate, 50 M Fe 2ϩ , and 40 l crude cell extracts. The consumption of oxygen was measured using a Clark-type oxygen electrode (Oxytherm; Hansatech, United Kingdom) equipped with an automatically temperature-controlled electrode chamber. One unit (U) of CsiD activity was defined as the amount that catalyzed the reduction of 1 mol of oxygen per min.
The activity of GcdH was determined at 30°C in 200 l of a reaction solution containing 50 mM Tris-HCl (pH 7.4), 0.25 mM glutaryl-CoA, 0.01 M flavin adenine dinucleotide (FAD), 0.2 mM ferricenium hexafluorophosphate (51), and 30 l crude cell extracts. The absorbance at 300 nm was measured using a Spectramax Plus 384 spectrophotometer (Molecular Devices, USA). One unit (U) of GcdH activity was defined as the amount that catalyzed the reduction of 1 mol of ferricenium hexafluorophosphate per min.
The activity of LhgO was measured at 30°C in 800 l of a reaction solution containing PBS, 0.1 mM L-2-HG, 0.05 mM dichlorophenol-indophenol (DCPIP), 0.2 mM phenazine methosulfate (PMS), and 40 l crude cell extracts. The absorbance at 600 nm was measured using a UV/visible light spectrophotometer (Ultrospec 2100 pro; Amersham Biosciences, USA). One unit (U) of LhgO activity was defined as the amount that catalyzed the reduction of 1 mol of DCPIP per min.
Construction of P. putida KT2440 mutants. The P. putida KT2440 (ΔcsiR) mutant was generated as follows: the homologous arms upstream and downstream of the csiR gene were PCR amplified using primer pair csiR-uf/csiR-ur and primer pair csiR-df/csiR-dr, respectively (Table S2). The upstream and downstream fragments were fused via recombinant PCR with primers csiR-uf and csiR-dr. The generated fusion was digested with BamHI and HindIII and cloned into pK18mobsacB (52) cut with the same enzymes. The resulting plasmid, pK18mobsacB-ΔcsiR, was transferred into P. putida KT2440 by electroporation, and the mutant with integration of the plasmid pK18mobsacB-ΔcsiR into the chromosome was obtained by selection on an LB plate containing 50 g ml Ϫ1 kanamycin. Kanamycin-resistant transformants were plated onto LB plates containing 10% (wt/vol) sucrose to screen the csiR deletion mutants. All the constructed strains were confirmed by PCR and sequence analysis. Other mutants of P. putida KT2440 were generated using the same procedure.
Expression and purification of recombinant CsiR and GcdR. The csiR and gcdR genes were amplified using genomic DNA of P. putida KT2440 and primer pair csiR-F/csiR-R and primer pair gcdR-F/gcdR-R, respectively (Table S2). The amplified csiR gene was cut with BamHI and HindIII and cloned into His tag expression vector pETDuet-1 to generate pETDuet-csiR. The amplified gcdR gene was digested with NcoI and HindIII and cloned into His tag expression vector pET28a to generate pET28a-gcdR. The resulting expression plasmids were transformed into E. coli BL21(DE3) for CsiR and GcdR expression.
The recombinant E. coli BL21(DE3) strains containing either plasmid pETDuet-csiR or plasmid pET28a-gcdR were grown at 37°C in LB medium to an OD 600 of 0.5 to 0.6 and induced at 16°C for 10 h in the presence of 1 mM isopropyl-D-1-thiogalactopyranoside (IPTG). The cells were collected by centrifugation and washed twice with buffer A (20 mM sodium phosphate and 500 mM sodium chloride, pH 7.4). Pellets were resuspended in buffer A containing 1 mM phenylmethylsulfonyl fluoride (PMSF) and 10% glycerol (vol/vol) and were then lysed by sonication. The cellular lysate was centrifuged at 16,000 ϫ g for 30 min at 4°C to remove bacterial debris. The supernatant was loaded onto a HisTrap HP column (5 ml) that was preequilibrated with buffer A. Proteins were eluted with buffer B (20 mM sodium phosphate, 500 mM imidazole, and 500 mM sodium chloride, pH 7.4), and the eluted fractions were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) with 12.5% polyacrylamide gels. Protein concentrations were determined by the use of Bradford assays.
Determination of the transcriptional start sites. RNA samples were isolated from P. putida KT2440 grown in MSM supplemented with glutarate as the sole carbon source. The transcriptional start sites of the csiD and gcdH operons were determined using a 5= rapid amplification of cDNA ends (RACE) system (Invitrogen, China). As for csiD, the first strand cDNA was synthesized from total RNA using primer csiD-GSP1 (Table S2). The resulting cDNA was tailed with terminal deoxynucleotidyl transferase and dCTP and was subsequently amplified via PCR with the abridged anchor primer (APP) and csiD-GSP2 (Table S2). A nested PCR was then performed using the PCR product as a template with AAP and csiD-GSP3 (Table S2). The resulting PCR product was cloned into pMD18-T vector (TaKaRa, China) for sequencing. The transcriptional start site of the gcdH operon was determined using the same procedure.
Electrophoretic mobility shift assays. The DNA fragments used in the EMSAs were obtained from P. putida KT2440 genomic DNA by PCR using primers F1-F/F1-R, primers F2-F/F2-R, and primers C-F/C-R (Table S2). EMSAs were carried out using 20-l reaction mixtures containing 10 nM DNA fragment and increasing concentrations (0 to 120 nM) of purified proteins in EMSA binding buffer (10 mM Tris-HCl [pH 7.4], 50 mM KCl, 0.5 mM EDTA, 10% glycerol, 1 mM dithiothreitol [DTT]). All binding reaction mixtures were incubated at 30°C for 30 min and then subjected to electrophoresis on 6% native polyacrylamide gels for approximately 50 min at 4°C and 170 V (constant voltage). The gels were stained with SYBR green I (TaKaRa, China) for 30 min and photographed under UV irradiation.
To analyze the possible effectors of CsiR, the protein was incubated with L-lysine, 5-aminovalerate, glutarate, or L-2-HG in EMSA buffer at 30°C for 15 min. The F1 DNA fragment (10 nM) was then added, and the reaction mixture was incubated for an additional 30 min before electrophoresis.
DNase I footprinting. For preparation of the probes, the fragments containing the intergenic csiR-csiD region or the gcdR-gcdH region were amplified by PCR using the appropriate primers (primers F1-F/F1-R and primers F2-F/F2-R) ( Table S2). The PCR products were cloned into pEASY-Blunt Simple Cloning Vector (TransGen, China), generating pEASY-Blunt-F1 and pEASY-Blunt-F2, respectively. The probes were obtained by PCR amplification using primers M13F-FAM and M13R and plasmid pEASY-Blunt-F1 or pEASY-Blunt-F2 as the template. Following gel purification, the FAM-labeled probes were quantified with a NanoDrop 2000 spectrophotometer (Thermo Scientific, USA).
For each assay, a 400-ng volume of probe was incubated with CsiR (for the probe containing F1) or GcdR (for the probe containing F2) in a total volume of 40 l in EMSA binding buffer. After incubation at 30°C for 30 min, all reaction mixtures were treated with 10 l of a solution containing about 0.015 U DNase I (Promega) and 100 nmol freshly prepared CaCl 2 and were further incubated at 25°C for 1 min. The reaction was terminated by addition of 140 l of a stop solution containing 30 mM EDTA, 200 mM sodium acetate, and 0.15% (wt/vol) SDS. Digested samples were extracted with phenol-chloroform, precipitated with ethanol, resuspended in 30 l MiniQ water, and analyzed as described before (53).
␤-Galactosidase assays. To construct the reporter plasmids of the csiD promoter, a 134-bp fragment upstream of the TSS (P csiD ) was PCR amplified from P. putida KT2440 genomic DNA using the appropriate oligonucleotide pairs (Table S2). The purified PCR products were digested with EcoRI and PstI and cloned into pME6522 to generate pME6522-P csiD . In the same way, a 124-bp P gcdH fragment was cloned into pME6522 to generate pME6522-P gcdH . The resulting plasmids were verified by DNA sequencing and subsequently transferred into P. putida KT2440 and its derivatives by electroporation.
P. putida KT2440 and its derivatives harboring pME6522-P csiD or pME6522-P gcdH were grown in MSMs with different compounds as sole carbon sources. Cells were obtained from cultures at the mid-log phase and permeabilized with chloroform and SDS. The ␤-galactosidase activity was determined using o-nitrophenyl-␤-D-galactopyranoside as the substrate, and the results were expressed in Miller units (54).
Quantification of L-2-HG. P. putida KT2440 and its derivatives were cultured at 200 rpm and 30°C in MSM supplemented with 5.0 g liter Ϫ1 L-2-HG as the sole carbon source. Samples (1.0 ml) were taken periodically, boiled at 100°C for 15 min, and then centrifuged at 12,000 ϫ g for 15 min to remove cell debris. The concentrations of L-2-HG were measured by the use of a high-performance liquid chromatography (HPLC) system (Agilent 1100 series) equipped with an Aminex HPX-87H column (Bio-Rad) and a refractive index detector (RID). The mobile phase was 0.1% formic acid at a flow rate of 0.4 ml min Ϫ1 .