SEARCH

SEARCH BY CITATION

Keywords:

  • Escherichia coli ;
  • genomic SELEX;
  • purine nucleotide;
  • RbsR, ribose;
  • transcription factor

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

Escherichia coli is able to utilize d-ribose as its sole carbon source. The genes for the transport and initial-step metabolism of d-ribose form a single rbsDACBK operon. RbsABC forms the ABC-type high-affinity d-ribose transporter, while RbsD and RbsK are involved in the conversion of d-ribose into d-ribose 5-phosphate. In the absence of inducer d-ribose, the ribose operon is repressed by a LacI-type transcription factor RbsR, which is encoded by a gene located downstream of this ribose operon. At present, the rbs operon is believed to be the only target of regulation by RbsR. After Genomic SELEX screening, however, we have identified that RbsR binds not only to the rbs promoter but also to the promoters of a set of genes involved in purine nucleotide metabolism. Northern blotting analysis indicated that RbsR represses the purHD operon for de novo synthesis of purine nucleotide but activates the add and udk genes involved in the salvage pathway of purine nucleotide synthesis. Taken together, we propose that RbsR is a global regulator for switch control between the de novo synthesis of purine nucleotides and its salvage pathway.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

The commonest pentoses, d-ribose, d-xylose and l-arabinose, can support the growth of Escherichia coli on a mineral medium as the sole source of carbon and energy. Uptake of these pentoses is mediated by ABC (ATP-binding cassette)-type high-affinity transporters specific for the respective sugars (Horazdovsky & Hogg, 1987; Song & Park, 1998; Park et al., 1999). In the case of d-ribose, several low-affinity transporters also serve as alternative routes for its uptake, including the allose transporter (Kim et al., 1997), the xylose transporter (Song & Park, 1998) and the altered form of glucose phosphotransferase system (Oh et al., 1999). The system for transport and utilization of these sugars are regulated on the transcriptional level by transcription factors which belong to different structural families: the ribose regulator (RbsR) belongs to the LacI family (Mauzy & Hermodson, 1992a), whereas the arabinose (AraC) (Stoner & Schleif, 1983; Miyada et al., 1984) and xylose (XylR) (Song & Park, 1998) regulators are members of the AraC family. Members of these families have the opposite modes of action: RbsR is a classical repressor whose inducer is d-ribose, whereas AraC and XylR are activators of transcription.

The genes encoding the transporter and initial-step enzymes of ribose utilization form an inducible rbsDACBK operon (Lopilato et al., 1984; Mauzy & Hermodson, 1992a; Laikova et al., 2001). RbsACB proteins form the high-affinity ABC-type ribose transporter consisting of the extraplasmic binding protein (RbsB), membrane permease (RbsC) and cytoplasmic ATPase (RbsA) (Park et al., 1999). RbsD, a ribose mutarotase, catalyzes the conversion between the β-pyran and β-furan forms of d-ribose, whereas ribokinase RbsK converts d-ribose into d-ribose 5-phosphate. Since the majority of ribose exists as β-pyranose in solution, the intracellular supply of furanose is limiting when the ribokinase is actively consuming ribose (Sigrell et al., 1998). The regulator gene for this operon (rbsR) is considered to form an independent operon, located immediately downstream of the rbsDACBK operon (Laikova et al., 2001), but this organization has not been experimentally confirmed. RbsR, a member of LacI family transcription factor, forms the bundle of four helices characteristic of proteins implying the HTH DNA contact motif. The C-terminal 270 residues of RbsR are homologous to the ribose-binding protein RbsB (Mauzy & Hermodson, 1992b).

The rbsDACBK operon is the only target of regulation so far identified to be under the control of RbsR (Lopilato et al., 1984; Mauzy & Hermodson, 1992a; Laikova et al., 2001). RbsR binds to the operator of a palindromic sequence in the rbs operon promoter (Mauzy & Hermodson, 1992a). Using the newly developed ‘Genomic SELEX’ screening system (Shimada et al., 2005), we have so far determined the regulation targets for more than 200 transcription factors from E. coli (reviewed in Ishihama, 2010, 2012) and realized that the number of regulation targets of transcription factors in E. coli are more than those reported and listed in the databases such as EcoCyc (Keseler et al., 2012) and RegulonDB (Salgado et al., 2012). We then tried to list up the whole set of regulation targets for RbsR using the newly developed Genomic SELEX screening system. The results herein described indicate that RbsR is involved in regulation of not only the rbs operon for transport and utilization of d-ribose but also a set of genes for both de novo and salvage pathways of purine nucleotide synthesis.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

Bacterial strains and plasmids

Escherichia coli DH5α was used for plasmid amplification, E. coli BL21 for RbsR expression, and Ecoli BW25113 (W3110 lacIq rrnBT14 ΔlacZWJ16 hsdR514 ΔaraBADAH33 ΔrhaBADLD78) (Datsenko & Wanner, 2000) and JW3732 (a rbsR single-gene deletion mutant of BW25113) (Baba et al., 2006) were obtained from the E. coli Stock Center (National Bio-Resource Center, Mishima, Japan). Cells were grown in Luria–Bertani (LB) or M9-glucose medium at 37 °C under aeration with constant shaking at 150 r.p.m. Cell growth was monitored by measuring the turbidity at 600 nm.

Expression plasmid pRsbR of His-tagged RbsR protein was constructed essentially according to the standard procedure in this laboratory (Shimada et al., 2005; Yamamoto et al., 2005). The expression of RbsR was performed in E. coli BL21.

Expression and purification of RbsR protein

Escherichia coli BL21 (pRbsR) transformant was grown in LB broth in the presence of 50 μg mL−1 ampicillin, and expression of RbsR was induced in mid-log phase by adding 1 mM IPTG. After 3 h induction, cells were harvested and protein purification was carried out according to the standard procedure in this laboratory (Shimada et al., 2005; Yamamoto et al., 2005). In brief, lysozyme-treated cells were sonicated in the presence of 100 mM phenylmethylsulfonyl fluoride (PMSF). After centrifugation of cell lysate (30 mL) at 18 000 g. for 20 min at 4 °C, the resulting supernatant was mixed with 2 mL of 50% Ni-NTA agarose solution (Qiagen) and loaded onto a column. After washing with 10 mL of lysis buffer, the column was washed with 10 mL of washing buffer (50 mM Tris-HCl, pH 8.0 at 4 °C, 100 mM NaCl). Proteins were then eluted with 2 mL of an elution buffer (200 mM imidazole, 50 mM Tris-HCl, pH 8.0 at 4 °C, 100 mM NaCl), and dialyzed against a storage buffer (50 mM Tris-HCl, pH 7.6, 200 mM KCl, 10 mM MgCl2, 0.1 mM EDTA, 1 mM dithiothreitol and 50% glycerol). RbsR used throughout this study was more than 95% pure as analyzed by SDS-PAGE.

Genomic SELEX screening of RbsR-binding sequences

A mixture of DNA fragments of the E. coli K-12 W3110 genome was prepared after PCR amplification of the E. coli DNA library cloned in a multi-copy plasmid pBR322 (Shimada et al., 2005). For SELEX screening of RbsR targets, 5 pmol of the mixture of DNA fragments and 10 pmol His-tagged RbsR were mixed in a binding buffer (10 mM Tris-HCl, pH 7.8 at 4 °C, 3 mM magnesium acetate, 150 mM NaCl, and 1.25 mg mL−1 bovine serum albumin) and incubated for 30 min at 37 °C. The DNA-transcription factor mixture was applied to a Ni-NTA column, and after washing with the binding buffer containing 10 mM imidazole to remove unbound DNA, DNA-RbsR complexes were eluted with an elution buffer containing 200 mM imidazole. For SELEX-chip analysis, PCR-amplified products of the isolated DNA-RbsR complexes and original DNA library were labeled with Cy5 and Cy3, respectively. The fluorescent-labeled DNA mixtures were hybridized to a DNA microarray containing 43 450 species of 60-bp-long DNA probe, which are designed to cover the entire E. coli genome at 105-bp intervals (Oxford Gene Technology, Oxford, UK). The fluorescent intensity of RbsR sample at each probe was normalized with that of the corresponding peak of original DNA library. After normalization of each pattern, the Cy5/Cy3 ratio was measured and plotted along the E. coli genome.

Northern blot analysis

Total RNAs were extracted from exponentially growing E. coli cells (OD600 nm 0.5) by the hot phenol method. RNA purity was checked by electrophoresis on 0.3% agarose gel in the presence of formaldehyde followed by staining with Methylene Blue. Northern blot analysis was performed essentially as described previously (Shimada et al., 2007). Digoxigenin (DIG)-labeled probes were prepared by PCR amplification using E. coli W3110 genomic DNA (50 ng) as template, DIG-11-dUTP (Roche) and dNTP as substrates, gene-specific forward and reverse primers, and Ex Taq DNA polymerase (Takara). Total RNAs (1 μg) were incubated in formaldehyde-MOPS (morpholinepropanesulfonic acid) gel-loading buffer for 10 min at 65 °C for denaturation, subjected to electrophoresis on formaldehyde-containing 1.5% agarose gel, and then transferred to nylon membrane (Roche). Hybridization was performed with DIG easy Hyb system (Roche) at 50 °C overnight with a DIG-labeled probe. For detection of the DIG-labeled probe, the membrane was treated with anti-DIG-AP Fab fragments and CDP-Star (Roche), and the image was scanned with LAS-4000 IR multi-color (Fuji Film).

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

Search for the regulation targets of RbsR by Genomic SELEX screening

To identify the whole set of promoters, genes and operons under the direct control of RbsR, we performed Genomic SELEX screening (Shimada et al., 2005), in which purified His-tagged RbsR was mixed with a collection of E. coli genome fragments of 200–300 bp in length and RbsR-bound DNA fragments were affinity-isolated. The original mixture of genomic DNA fragments formed smear bands on PAGE, but after three cycles of Genomic SELEX, DNA fragments with high affinity to RbsR were enriched, forming sharper bands on PAGE gels (data not shown).

As a short-cut approach to identify the whole set of sequences recognized by RbsR, we subjected the isolated SELEX fragment mixture to the DNA chip analysis using an E. coli tilling array (Shimada et al., 2011a, b). In brief, the SELEX DNA fragments were labeled with Cy5 and the original DNA library with Cy3. The mixtures were then hybridized with the DNA tilling microarray (Oxford Gene Technology, UK) and the fluorescence intensities bound on each probe were measured. For identification of RbsR-binding sites, the Cy5/Cy3 ratio was plotted along a total of 43 450 probes aligned on the E. coli genome.

In the absence of inducer ribose, RbsR alone bound at least six regions (Fig. 1a), three within intergenic spacers and three within open reading frames (Table 1A). Since the prokaryotic DNA-binding transcription factors bind near promoters and control transcription from the promoters, the regulation target promoters were predicted based on the binding sequences of test transcription factors. As expected, the only known target rbsDACBK was included in the collection of three RbsR-binding spacers. The highest peak was identified for the promoter region of rbsDACBK operon (Fig. 1a), exhibiting the ratio of fluorescent intensity of 1925 between the RbsR-bound SELEX fragments (Cy5) and the original DNA library (Cy3) (Fig. 2e). In addition, RbsR-binding sites were identified within two spacer regions: one upstream of add (adenine deaminase) and downstream of malY (Fig. 2b), and another upstream of divergently transcribed purHD and rrsE-gltV-rrlE-rrfE rRNA operons (Fig. 2f). The add gene product (dCTP deaminase) is involved in the salvage pathway of purine synthesis. On the other hand, PurH (AICAR transformylase/IMP cyclohydrolase) and PurD (phosphoribosylamine-glycine ligase) are involved in the de novo synthesis of purine, indicating the involvement of RbsR in regulation of the genes for purine nucleotide synthesis.

Table 1. RbsR-binding sites identified by genomic SELEX
FunctionLeft geneRbsR siteRight geneFunction
  1. Genes shown in bold are the predicted regulation targets of RbsR. (A) Genes rbsD, purH and rrsE are the first gene of rbsDACBK, purHG and rrsE-gltV-rrlE-rrfE operon, respectively. (B) The udk gene forms a single operon with dcd encoding dCTP deaminase, while the add gene forms a single gene operon.

(A) Genomic SELEX in the absence of ribose
  proB > > proA > thrW tRNA-Thr
  malY >   > add Adenine deaminase
Uridine/cytidine kinase udk < > yegE < alkA  
  qseB > > qseC < ygiZ  
  kup >   > rbsD Ribose-binding protein
AICAR transformylase purH <   > rrsE 16S rRNA gene
(B) Genomic SELEX in the presence of ribose
  proB > > proA > thrW tRNA-Thr
  malY >   > add Adenine deaminase
Uridine/cytidine kinase udk < > yegE < alkA  
  qseB > > qseC < ygiZ  
  kup >   > rbsD Ribose-binding protein
AICAR transformylase purH <   > rrsE 16S rRNA gene
  kup >   > ytfI Predicted protein
image

Figure 1. RbsR-binding sites on the Escherichia coli K-12 genome identified by SELEX-chip. Genomic SELEX screening of RbsR-binding sequences was performed in the absence (a) or presence (b) of d-ribose. After the genomic SELEX, a collection of DNA fragments was subjected to SELEX-chip analysis using the tilling array of E. coli K-12 genome (for details see 'Materials and methods'). The Y-axis represents the relative number of RbsR-bound DNA fragments, whereas the X-axis represents the position on the E. coli genome. The regulation targets were predicted based on the location of RbsR-binding sites. When RbsR-binding sites are located within spacer regions, only the flanking genes are indicated. The direction of transcription is shown by arrows. The genes shown in parentheses carry the RbsR-binding sites on the open reading frames. The whole set of predicted regulation targets of RbsR is described in Table 1.

Download figure to PowerPoint

image

Figure 2. Peaks of the RbsR-binding sites along the Escherichia coli genome. The peak regions with the binding activity of purified RbsR protein were expanded. Patterns a, c and d with symbols shown against the black background indicate the RbsR-binding sites within ORFs, whereas patterns b, e and f with symbols shown against the white background indicate the RbsR-binding sites within intergenic spacers. Closed circles represent the level of RbsR binding in the absence of d-ribose, whereas closed circles represent the level of RbsR binding in the presence of inducer d-ribose.

Download figure to PowerPoint

RbsR-binding sites were identified in three open reading frames (ORFs) (Table 1A): proA encoding glutamate-5-semialdehyde dehydrogenase (Fig. 2a); yegE encoding predicted diguanylate cyclase (Fig. 2c); and qseB encoding quorum sensing sensory histidine kinase of QseBC two-component system (Fig. 2d). The binding on open reading frames has been identified only for a set of transcription factors but its functional role remains unresolved (Shimada et al., 2008). Downstream of the RbsR-binding site within yegE ORF, the udk-dcd operon exists, each encoding uridine/cytidine kinase (Udk) and dCTP deaminase (Dcd), respectively. Udk is involved in the salvage pathway of pyrimidine nucleotide synthesis (see Fig. 5). Since promoters of the E. coli genome are sometimes located within upstream gene ORF, the udk-dcd operon was supposed to be under the control of RbsR. This possibility was examined by Northern blot analysis of udk mRNA in the presence and absence of RbsR (see below).

Search for the regulation targets of RbsR in the presence of ribose

The rbsDACBK operon is induced by d-ribose that is sensed by RsbR. To confirm this regulation model for the ribose operon and to analyze the effect of d-ribose on the newly identified targets, we next performed the Genomic SELEX screening in the presence of 10 mM d-ribose. The peak of rbs operon promoter markedly decreased by 43.8-fold, from the fluorescent level of 1925 to 44 (Figs 1b and 2e). Likewise, a 12-fold reduction was observed in the height of the add operon peak (Figs 1b and 2b). As a result, the peak of qseC peak became the highest in the presence of ribose (Fig. 1b), even though the peak height of fluorescence remained virtually unchanged in both the absence (431) and the presence (232) of d-ribose (Fig. 2d). In the presence of d-ribose, the levels of other minor peaks apparently increased, including the peaks on proA and yegE ORFs (Fig. 1b). In addition, a new peak appeared within the spacer upstream of uncharacterized ytfI gene and downstream of cysQ (Fig. 1b, Table 1B).

To search for the recognition sequence of RbsR, the sequences of all probes that showed the high-level binding affinity to RbsR were subjected to the Logo pattern analysis using weblogo (http://weblogo.berkeley.edu/logo.cgi). Among five RbsR-binding regions, a high level similarity was found for 20-nucleotide-long sequences (Fig. 3), of which the central part of 8-nucleotide-long AA(A/T)CGTT(T/G)(T/C) sequences including the completely conserved CGT was highly conserved. We then predicted that the palindromic octanucleotide sequence AAACGTTT represents the RbsR-box (Fig. 3).

image

Figure 3. Search for the consensus sequence of RbsR-binding sites. Sequences of the probes with high level of RbsR-biding activity (rbsD, add, purH, qseC and udk) were analyzed using bioprospector (http://ai.stanford.edu/~xsliu/BioProspector/) and weblogo (http://weblogo.berkeley.edu/logo.cgi) was used for matrix construction. Among the five target RbsR-binding sites, the sequences conserved in more than three targets are shown in bold. The RbsR box was predicted to be composed of the palindromic octanucleotide sequence including the completely conserved CGT sequence.

Download figure to PowerPoint

Transcription regulation in vivo of the set of RbsR target genes involved in nucleotide metabolism

To confirm the regulation in vivo of the set of RbsR target genes, we carried out Northern blot analysis of mRNA in cells of steady-state growth in M9-glucose medium. The level of rbsD transcription increased markedly in the rbsR knockout mutant (Fig. 4, rbsD lane). In the presence of d-ribose, the level of rbsD mRNA in wild-type significantly increased but it did not show any change in the rbsR mutant. These observations altogether support the repressive role of RbsR for the rbsDACBK operon. The level of purH mRNA also increased in the rbsR mutant (Fig. 4, purH lane), implying a repressive role of RbsR in transcription of the purHG operon. In agreement with this prediction, the purH mRNA level in wild-type increased in the presence of inducer D-ribosome. The purHD gene products are involved in the de novo synthesis of purine nucleotides in the pathway downstream of ribose 5-P (Fig. 5). Thus, we concluded that RbsR is involved in repression of the entire pathway of de novo synthesis of purine nucleotides starting from the uptake of d-ribose.

image

Figure 4. Northern blot analysis of mRNA for RbsR target genes. Wild-type Escherichia coli K-12 BW25113 and its rbsR mutant JW3732 were grown in M9-0.25% glucose medium at 37 °C in the presence of 0.2% ribose. Total RNA was prepared at exponential phase, and directly subjected to Northern blot analysis under the standard conditions (for details see 'Materials and methods'). DIG-labeled hybridization probes are shown on the left side of each panel. The amounts of total RNA analyzed were determined by measuring the intensity of 23S rRNA gene.

Download figure to PowerPoint

image

Figure 5. Model of the RbsR regulon. RbsR has been believed to be an ordinary LacI-type repressor for a single target rbsDACBK operon (shown in box under clear background). After Genomic SELEX screening, we identified novel regulation targets of RbsR, including the genes involved in the salvage pathways of purine nucleotide synthesis (shown against a gray background). Ado, adenosine; Cyt, cytidine; Ino, inosine; Urd, uridine.

Download figure to PowerPoint

On the other hand, the level of both add and udk mRNAs decreased in the rbsR mutant (Fig. 4, add and udk lanes), implying a positive role of RbsR in transcription of the add and udk-dcd operons, both participating in the salvage pathway of nucleotide synthesis. In concert with this prediction, neither add or udk mRNAs decreased in the presence of exogenous d-ribose, but rather slightly increased (Fig. 4, add and udk lanes). Taken together, we propose a model, shown in Fig. 5, that RbsR transcription factor is not only the repressor of rbsDACBK operon but also the regulator of some downstream-pathway genes for de novo synthesis of purine nucleotides. Moreover, RbsR is an activator for the salvage pathway genes of nucleotide synthesis.

Previously, we identified that RutR, originally identified as a regulator of the degradation of pyrimidine nucleotides (Loh et al., 2006), plays a switching role between the de novo synthesis of pyrimidine nucleotides and its salvage pathway after sensing the external level of uracil and thymine (Shimada et al., 2007). Likewise RbsR could be a regulator for switching between the de novo synthesis of purine nucleotides and its salvage pathway. Since RbsR influences transcription of the udk and dcd operons, RsbR may also be involved in the regulation of salvage pathway of pyrimidine nucleotides.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

The ribose operon repressor RbsR represses not only the uptake and initial step metabolism of d-ribose to generate d-ribose 5′-phosphate but also the de novo synthesis of purine nucleotides from d-ribose 5′-phosphate. In the presence of inducer d-ribose, however, RbsR activates the salvage pathway of purine nucleotide synthesis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References

This work was supported by a Grant-in-Aid for Grants-in-Aid for Scientific Research (A) (21241047), (B) (18310133) and (C) (25430173) to A.I. and Grant-in-Aid for Grants-in-Aid for Young Scientists (B) (24710214) to T.S. from MEXT (Ministry of Education, Culture, Sports, Science and Technology of Japan), and MEXT-Supported Program for the Strategic Research Foundation at Private Universities 208-2012 (S0801037).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Conclusion
  7. Acknowledgements
  8. References