SGM Journals
CELL AND MOLECULAR BIOLOGY OF MICROBES

σ54-mediated control of the mannose phosphotransferase sytem in Lactobacillus plantarum impacts on carbohydrate metabolism

Marc J. A. Stevens, Douwe Molenaar, Anne de Jong, Willem M. De Vos, Michiel Kleerebezem

  • 1TI Food and Nutrition, PO Box 557, 6700 AN Wageningen, The Netherlands
  • 2NIZO food research, PO Box 20, 6710 BA Ede, The Netherlands
  • 3Laboratory of Microbiology, Wageningen University and Research Centre, Dreijenplein 10, 6703 HB Wageningen, The Netherlands
  • 4Department of Genetics, Groningen Biomolecular Sciences and Biotechnology Institute (GBB), Rijksuniversiteit Groningen, Kerklaan 30, 9751 NN Haren, The Netherlands
  • Correspondence
    Michiel Kleerebezem
    Michiel.Kleerebezem{at}nizo.nl

    • Microbiology 2010; 156(3):695–707 · https://doi.org/10.1099/mic.0.034165-0

    View at publisher PubMed

    Abstract

    Sigma factors direct specific binding of the bacterial RNA polymerase to the promoter. Here we present the elucidation of the σ54 regulon in Lactobacillus plantarum. A sequence-based regulon prediction of σ54-dependent promoters revealed an operon encoding a mannose phosphotransferase system (PTS) as the best candidate for σ54-mediated control. A σ54 (rpoN) mutant derivative did not grow on mannose, confirming this prediction. Additional mutational analyses established the presence of one functional mannose PTS in L. plantarum, the expression of which is controlled by σ54 in concert with the σ54-activator ManR. Genome-wide transcription comparison of the wild-type and the rpoN-deletion strain revealed nine upregulated genes in the wild-type, including the genes of the mannose PTS, and 21 upregulated genes in the rpoN mutant. The σ54-controlled mannose PTS was shown also to transport glucose in L. plantarum wild-type cells, and its presence causes a lag phase when cultures are transferred from glucose- to galactose-containing media. The mannose PTS appeared to drain phosphoenolpyruvate (PEP) pools in resting cells, since no PEP could be detected in resting wild-type cells, while mannose PTS mutant derivatives contained 1–3 μM PEP (mg protein)−1. Our data provide new insight into the role of σ54 in L. plantarum and possibly other Gram-positive bacteria in the control of expression of an important glucose transporter that contributes to glucose-mediated catabolite control via modulation of the PEP pool.

    • ↵†Present address: Laboratory for Food Biotechnology, Institute of Food Science and Nutrition, Swiss Federal Institute of Technology, Schmelzbergstrasse 7, CH-8092 Zurich, Switzerland.

    • ↵‡Present address: Centre for Integrative Bioinformatics, Vrije Universiteit, De Boelelaan 1081A, 1081 HV Amsterdam, The Netherlands.

    • The Gene Expression Omnibus (GEO, ) accession numbers for the array design and primary data for the microarray experiments in this study are GPL6368 and GSE11351, respectively.

    • Supplementary material is available with the online version of this paper.

    Edited by: P. W. O'Toole

    INTRODUCTION

    Bacterial transcription initiation requires a sigma (σ) factor to direct the RNA polymerase core enzyme to the promoter. Most bacteria possess more than one sigma factor, each with its own specific promoter sequence, generating an important mechanism for differential gene expression in bacteria. Based on structural and functional criteria, sigma factors can be categorized into two main classes (Merrick, 1993): a class containing all the Escherichia coli σ70-like factors and a class containing only a single member, the σ54 factor (Merrick & Gibbins, 1985). Although not immediately recognized as a sigma factor, σ54 was originally discovered in the Gram-negative bacterium Salmonella typhimurium, where it was found to regulate expression of the glutamine synthetase (Garcia et al., 1977). The σ54 transcription factor (also known as σN or σL) differs from all other sigma factors by binding to a promoter with a conserved −12/−24 motif (Buck et al., 1986; Morett & Buck, 1989) and by the absolute requirement for a dedicated activator protein to initiate transcription by the DNA-bound σ54–RNA polymerase (RNAP) complex. This activator protein catalyses ATP-dependent transition from a closed to an open complex of the RNAP, which enables initiation of transcription (Sasse-Dwight & Gralla, 1988). The activator protein usually binds upstream of the σ54 promoter site, and DNA looping is required for the activator to contact the RNAP, resembling a mechanism similar to transcriptional initiation in eukaryotes. For this reason the σ54 activators are also called bacterial enhancer binding proteins (EBPs) (Studholme & Dixon, 2003).

    Since transcription of σ54-dependent genes requires the activity of a dedicated activator protein, transcription of these genes is generally tightly controlled, with low levels of promoter leakage (Wang & Gralla, 1998). Another advantage of σ54-dependent regulation is the possibility to modulate gene expression from a promoter over a wide range without requirement for additional transcription regulators, thereby allowing rapid and strict regulation of gene expression under changing physiological conditions (Buck et al., 2000).

    Comparison of 186 σ54-dependent bacterial promoters led to a defined promoter sequence (−12/−24) and spacer-length consensus (Barrios et al., 1999). Deletion of one or more nucleotides in the spacer region of σ54-dependent promoters resulted in a dramatic decrease or complete loss of promoter activity (Buck, 1986), and although there are no data available about promoter activity after insertion of nucleotides, the spacer-length between the −12 and −24 regions is considered to be essential for promoter function (Barrios et al., 1999). Similarly, a range of σ54-dependent genes has been identified, and σ54 appears to be involved in a variety of cellular processes, including nitrogen assimilation and fixation, glutamine synthesis and transport, and sugar transport (for a review see Studholme & Buck, 2000). In silico analysis of bacterial genomes has led to the identification of σ54-encoding genes in various bacteria, representing all bacterial phyla. Bacteria that possess a σ54 gene generally also have one or more predicted σ54-activator genes, which are usually genetically linked to the corresponding regulated genes (Reitzer & Schneider, 2001).

    In low-GC Gram-positive bacteria, σ54-dependent genes have been described in several species. In Bacillus subtilis, σ54 (there designated σL) is involved in levanase and acetoin import (Ali et al., 2001; Debarbouille et al., 1991) and in arginine and valine degradation (Gardan et al., 1997; Debarbouille et al., 1999). In Listeria monocytogenes, Enterococcus faecalis and Lactobacillus casei the expression of the operon encoding the mannose phosphotransferase system (mannose PTS) appears to be strictly σ54-dependent (Dalet et al., 2001; Hechard et al., 2001; Yebra et al., 2004).

    Here we describe the in silico and experimental characterization of the σ54 regulon of Lactobacillus plantarum, a lactic acid bacterium that is found in vegetable, meat and dairy fermentations. L. plantarum is a natural inhabitant of mammalian gastrointestinal tracts and specific strains are marketed as probiotics (Ahrne et al., 1998; de Vries et al., 2006). The genome of L. plantarum WCFS1 contains an rpoN-like gene, encoding a 444-residue protein with high identity to σ54 proteins in other bacteria, including closely related species such as Pediococcus pentosaceus (53 % identity), Enterococcus faecalis (51 %) and Listeria monocytogenes (40 %). The in silico predictions and transcriptome analysis described here show that only the mannose PTS operon is directly controlled by σ54 in L. plantarum. Furthermore, this mannose PTS also functions as a glucose-uptake system in L. plantarum, as exemplified by the reduced growth rate of a mannose PTS mutant on glucose-containing media. σ54 is thus shown not only to control the expression of the mannose PTS but also to have a marked effect on transport of other carbon sources via an indirect regulatory mechanism, mediated by the intracellular concentration of the glycolytic intermediate PEP.

    METHODS

    Bacterial strains, media and growth conditions.

    The bacterial strains used in this study are listed in Table 1⇓. E. coli DH5α was used as an intermediate cloning host and was grown aerobically at 37 °C in TY medium (Sambrook et al., 1989). When appropriate, chloramphenicol was added to a final concentration of 8 μg ml−1. L. plantarum WCFS1 and its derivatives were anaerobically grown in MRS broth (De Man et al., 1960) supplemented with 2 % (w/v) of selected carbon source (mannose, glucose, galactose, etc.), or in a chemically defined medium (CDM) designed for L. plantarum (Teusink et al., 2005). Cells were grown at 30 °C. Carbon sources were obtained from Sigma-Aldrich. Growth was monitored by measurement of the OD600 in a spectrophotometer (Ultrospec 3000, Pharmacia Biotech).

    Table 1.

    Strains and plasmids used in this study

    DNA manipulations and gene disruption.

    Molecular cloning and DNA manipulations were performed essentially as described by Sambrook et al. (1989). Plasmids constructed in this study are listed in Table 1⇑. Large-scale plasmid DNA isolations from E. coli were performed using a Jet Star Maxiprep kit (Genomed). DNA was isolated from L. plantarum as described previously (Ferain et al., 1994). Restriction enzymes and Pwo polymerase were obtained from Promega. T4 ligase was obtained from Boehringer. Primers were purchased from Proligo.

    Construction of plasmids and strains.

    Gene disruption was performed using a double-crossover gene replacement strategy (except for strain NZ7308, manIIC : : pNZ7350; see below), which resulted in replacement of the target gene by a chloramphenicol-resistance gene cassette (P32cat; Bron, 2004). For disruption of the rpoN gene a 1.2 kb fragment of the upstream region of the target locus was amplified using a proofreading DNA polymerase (Pwo) and the primers rpoN-U-5′ and rpoN-U-3′ (Table 2⇓). Similarly, a 1.2 kb fragment of the downstream region of rpoN was amplified using primers rpoN-D-5′ and rpoN-D-3′ (Table 2⇓). A similar strategy was used for the manR gene, using primers manR-U-5′ and manR-U-3′ and primers manR-D-5′ and manR-D-3′ to amplify the upstream and downstream fragments, respectively (Table 2⇓). The up- and downstream fragments were sequentially cloned into the SwaI and Ecl136II sites, respectively, of the gene-replacement vector pNZ5318 (Lambert et al., 2007), resulting in the rpoN-replacement vector pNZ7373 and the manR-replacement vector pNZ7332 (Table 1⇑). All mutagenesis constructs aimed to replace the target gene by the chloramphenicol-resistance cassette, with the cat gene in the same orientation as the target gene. Correct plasmid construction was checked by restriction analyses.

    Table 2.

    Primers used in this study

    A similar double-crossover strategy was used to disrupt manIIABCD (pts9ABCD), manIIAB (pts9AB), manIIB (pts9B) and manIIA+B (pts10AB), with the modification that the pNZ5318 derivative pNZ5319 (Lambert et al., 2007) was used as cloning vector; this vector harbours lox sites up- and downstream of the P32cat cassette, which enables cat-gene deletion by a resolvase, resulting in a clean gene deletion as described by Lambert et al. (2007). Flanking regions of the mannose operon (manIIABCD) were amplified using the primer sets manIIAB-U-5′ and manIIAB-U-3′ for the upstream region and manPTS-D-5′ and manPTS-D-3′ for the downstream region. Flanking regions of manIIAB were amplified using primers manIIAB-U-5′ and manIIAB-U-3′ (upstream) and manIIAB-D-5′ and manIIAB-D-3′ (downstream), flanking regions of manIIB using manIIB-U-5′ and manIIB-U-3′ (upstream) and manIIAB-D-5′ and manIIAB-D-3′ (downstream), and flanking regions of manIIA+B using manIIA+B-U-5′ and manIIA+B-U-3′ (upstream) and manIIA+B-D-5′ and manIIA+B-D-3′ (downstream). The upstream regions obtained were cloned into the SwaI site and the downstream regions into the Ecl136II site of the gene replacement vector pNZ5319, resulting in the manIIABCD-replacement vector pNZ7366, the manIIAB-replacement vector pNZ7364, the manIIB-replacement vector pNZ7365 and the manIIA+B-replacement vector pNZ7372 (Table 1⇑).

    The resulting plasmids were transformed into L. plantarum as described previously (Josson et al., 1989), and primary plasmid integrants were selected on MRS plates with 8 μg chloramphenicol ml−1 and 1 % glycerol at 30 °C. To check for erythromycin sensitivity, colonies were picked and transferred to MRS plates with 20, 40, 60, and 80 μg erythromycin ml−1 and grown overnight at 30 °C. The anticipated genetic organization after correct gene replacement leads to chloramphenicol resistance and erythromycin sensitivity; integrants with this phenotype were checked by PCR using universal primers (Con-cam-for and Con-cam-rev) annealing in the cat gene and a site-specific primer (Con-primers in Table 2⇑) annealing outside of the chromosomal region used for homologous recombination. The 5′-primers were combined with primer Con-cam-rev and the 3′-primers were combined with Con-cam-for.

    For the disruption of manIIC (pts9C) a single-crossover strategy was used. A 400 bp internal fragment was amplified using primers manIIC-I-5′ and manIIC-I-3′ (Table 2⇑) and cloned into the SwaI site of pNZ5319, resulting in the manIIC-disruption vector pNZ7350. The resulting plasmid was transformed to L. plantarum, and primary plasmid integrants were selected on MRS plates with 8 μg chloramphenicol ml−1 and 1 % glycerol at 30 °C. Correct integration of the disruption plasmid was checked by PCR using primer Con-manIIC, annealing upstream of the integration site, and the Con-cam-rev primer.

    RNA extraction and quality control.

    RNA was isolated as described previously (Stevens et al., 2008), using a phenol/chloroform extraction followed by further purification with the High Pure RNA isolation kit (Roche). The yield and purity of the RNA were determined by measurement of A260 and A280 (Ultrospec 3000, Pharmacia Biotech). The RNA quality was assessed using an Agilent 2100 Bioanalyser (Agilent Technologies), following the manufacturer's instructions. Only RNA samples displaying 16S : 23S rRNA ratios of 1.6 or higher were labelled and used for microarray experiments.

    cDNA synthesis, labelling and hybridization.

    The Cyscribe Post-labelling kit (Amersham Biosciences) was used to synthesize and to label cDNA. Labelled cDNAs were hybridized on amplicon-based microarrays containing fragments of approximately 97 % of the genes of L. plantarum WCFS1 as described previously (Stevens et al., 2008). Array design was submitted to the Gene Expression Omnibus (GEO, ) under GEO accession no. GPL6368.

    Scanning, data extraction and analyses.

    The slides were scanned with a ScanArray Express 4000 scanner (Perkin Elmer) and the images were analysed with Imagene software 4.2 (BioDiscovery). Primary data were submitted to GEO under accession number GSE11351.

    Statistical analyses were performed with R () using the linear models for microarray database limma (Smyth, 2005). Background-corrected spot intensities in both channels (I1 and I2) were converted to M-A coordinates, where M=log2(I1/I2) and A=log2(I1/I2)/2 and subsequently normalized using a LOESS fit, assuming that, on average, M is independent of A and centred around 0 (Smyth, 2005). Normalized intensities were used for further analysis. Log odds for differential expression (B-value) higher than 1 were taken as a cut-off. As amplicons were spotted in duplicate, two measurements per gene were performed, and only genes of which both measurements matched the criteria mentioned above were taken into account.

    Physiological characterization.

    The rate of lactate production was estimated by measuring the acidification rate in acidification buffer (0.5 mM potassium phosphate, 70 mM KCl and 1 mM MgSO4), pH 6.4, containing 0.5 % (w/v) of a specified carbon source, as described previously (Poolman et al., 1987). Cells were grown to OD600 1.0, washed twice in acidification buffer to remove all residual carbon sources and suspended to a density of OD600 1.0. After calibration of the cell suspension, a carbon source was added (t0) and pH change was followed over time. Acidification rates were calculated for each suspension within the linear range using Microsoft Excel XP. The pH change was converted to H+ production by calibration of the cell suspension with 2 μl aliquots of 50 mM HCI.

    Transcription analysis of the mannose operon.

    RNA was isolated from exponentially growing L. plantarum WCFS1 cells cultured in CDM (Teusink et al., 2005) supplemented with specific carbon sources as described above. Total RNA (5 μg) was blotted on a Gene Screen filter (New England Nuclear) as recommended by the manufacturer. An internal fragment of the manIIC gene was synthesized using the primers manIIC-I-5′ and manIIC-I-3′ (Table 2⇑). The fragment was labelled with [α-32P]ATP (Amersham Biosciences) by nick translation (Sambrook et al., 1989) and used for hybridization at 65 °C in 6× SSC/0.2 % BSA for 2 h. Blots were washed 10 min with 6× SSC followed by 30 min washing with 1× SSC and finally with 0.1× SSC at 65 °C for 30 min prior to autoradiography.

    Bioinformatic methodology.

    To identify putative σ54 promoters, the algorithm for fitting a mixture model by expectation maximization, MEME (Bailey & Elkan, 1994), was used on 17 experimentally verified σ54-dependent promoters from E. coli (Reitzer & Schneider, 2001). MEME parameters were set as follows: one motif per sequence should be found (−mod oops) and only the given strand should be searched. The motif in the form of a position-specific scoring matrix (PSSM) was used to search the complete genome of L. plantarum WCFS1, using the search tool MAST (Bailey & Gribskov, 1998). The E-value cut-off used in MAST was 1.00×10−6. Only motifs lying between the start codon and 300 bp upstream of the start codon were selected, and overlap with other genes was allowed.

    Analyses of glycolytic intermediates.

    L. plantarum cells grown in CDM (Teusink et al., 2005) containing 1 % glucose to an OD600 of 1.0 were harvested by centrifugation (3360 g, 30 °C, 5 min), washed twice with CDM without carbon source and resuspended in CDM without carbon source at an OD600 of 5.0. Samples (10 ml) taken after 0, 15, 30, 60, 120 and 180 min were quenched in −40 °C 60 % methanol/HEPES buffer as described by Pieterse et al. (2005). Quenched cells were harvested by centrifugation at 3500 g at −20 °C using a Sorvall RC5B Plus centrifuge and, to remove medium components, then washed once with quenching buffer (−40 °C) and once with ice-cold water. Subsequently, cells were stored overnight at −80 °C before lyophilization. The lyophilizate was resuspended in 500 μl water, and cell debris was removed by centrifugation (20 800 g, 4 °C, 5 min) to obtain a cell-free extract. The extract was passed through a 0.22 μm filter before HPLC analysis. HPLC analyses were performed as described previously (Bhattacharya et al., 1995; Boogaard, 2002), using an anion-exchange DX-300 column and a conductivity detector in combination with an anion self-regenerating suppressor (Dionex) to enhance signal-to-noise ratio. The flow-rate of the elution buffer was adjusted to 1 ml min−1 (half-speed) leading to doubled runtime (from 45 to 90 min), improving separation of the individual peaks.

    RESULTS

    In silicoσ54 regulon prediction in L. plantarum WCFS1

    To elucidate the role of σ54 in L. plantarum, we initially set out to predict the σ54 regulon in this species. To obtain a reliable prediction of the σ54 regulon, experimentally verified σ54-dependent promoters from E. coli (Reitzer & Schneider, 2001) were used to construct a position-specific scoring matrix (PSSM; see Supplementary material, Table S1) using MEME (Bailey & Elkan, 1994). This PSSM was used to search the L. plantarum WCFS1 genome using MAST (Bailey & Gribskov, 1998). A limited set of significant hits was obtained (Table 3⇓) and the best hit (significance score 5.6×10−9) was a sequence encountered 123 bp upstream of an operon encoding genes of a putative mannose PTS (PTS9). The next-best hit had a drastically poorer score, but still remained within the significance cut-off employed here (Table 3⇓).

    Table 3.

    Predicted σ54-dependent promoters in L. plantarum WCFS1

    The position of the promoter (−12 region) in relation to the start codon is listed in the last column.

    RNAP containing σ54 requires an activator protein to initiate transcription (Sasse-Dwight & Gralla, 1988). These activators possess a unique σ54-interaction domain, whose Pfam signature (accession no. PF00158) (Finn et al., 2006; Studholme & Dixon, 2003) was used to search in bacterial genomes for σ54-activator genes. Such a search revealed a single candidate σ54-activator-encoding gene (probability E=9×10−64; Supplementary material, Fig. S1) in the L. plantarum genome (lp_0585), which was designated manR. Genes encoding σ54 activators are commonly genetically linked to the genes regulated by the activator. Notably, the σ54-activator gene in the L. plantarum genome is located immediately upstream of a manIIAB gene cluster (lp_0586-0587, PTS10) that is predicted to encode components of a mannose PTS, but appears to be incomplete since it lacks a permease-encoding gene. Importantly, the predicted L. plantarum σ54-activator is located downstream of the mannose PTS operon (PTS9), which is preceded by the best-scoring σ54-dependent promoter sequence (Table 3⇑, Fig. 1⇓). However, these genetic loci are separated in the L. plantarum WCFS1 genome by a 15 kb insertion encoding a non-ribosomal-peptide synthesis machinery (Fig. 1⇓). This insertion is absent in other L. plantarum strains, as was concluded on the basis of genome-wide, array-based comparative genome hybridization (CGH) analysis of this species (Molenaar et al., 2005).

    Figure image not available in archive
    Fig. 1.

    Schematic representation of the mannose gene cluster in L. plantarum WCFS1. The 15 kb insertion encoding the non-ribosomal-peptide machinery is printed at a smaller scale. Predicted σ54-dependent promoters are indicated.

    In conclusion, the in silico analyses suggested a role for σ54 in transcriptional control of the mannose-PTS-encoding operon(s) found up- and downstream of the σ54-activator-encoding gene. Nevertheless, these findings cannot exclude the involvement of σ54 in regulation of other target genes that are preceded by a promoter resembling the σ54-dependent consensus sequence.

    Mannose utilization in L. plantarum WCFS1

    To study the σ54 regulon in L. plantarum, an rpoN deletion strain was constructed using a double crossover gene-replacement strategy, resulting in strain NZ7306 (rpoN : : P32cat, Table 1⇑). Since in our in silico analysis the best-scoring σ54 promoter sequence was predicted upstream of the mannose PTS operon, and a σ54-activator gene appeared to be located downstream of that same operon, we evaluated the ability of the rpoN deletion strain to grow on mannose. The wild-type strain had comparable growth rates in MRS medium supplemented with mannose or glucose as carbon source (Table 4⇓). The rpoN deletion strain NZ7306 displayed growth characteristics similar to its parental strain when grown on MRS containing glucose. In contrast, when cells were grown on mannose the growth rate was reduced dramatically (Table 4⇓), supporting a role for σ54 in the regulation of genes involved in mannose utilization in L. plantarum. Furthermore, biomass formation on MRS supplemented with mannose was at least two times lower by the mutant strain compared to the wild-type (Table 4⇓).

    Table 4.

    Growth rates of L. plantarum WCFS1 and its derivatives on different carbon sources

    The maximal growth rates (μmax) and optical densities after 24 h of culture are given for strains grown in MRS medium or in a chemically defined medium (CDM), both supplemented with 2 % (w/v) of the specific carbohydrate.

    There are two putative mannose PTS operons in L. plantarum WCFS1, of which one is complete (manIIABCD or pts9ABCD), while the second locus (manIIAB or pts10AB) appears to be truncated (Fig. 1⇑). Both loci are preceded by a σ54 promoter sequence (Table 3⇑) and genetically linked to the σ54-activator gene. To characterize the role of these genes in mannose uptake and their postulated σ54 dependency, gene-deletion mutant strains were constructed of the putative regulator (manR), both mannose operons, and of the separate manIIAB locus, the manIIB gene and the permease-encoding manIIC gene present in the manIIABCD (PTS9) operon (Table 2⇑, Fig. 1⇑). Glucose- and mannose-dependent growth characteristics of all strains were determined (Table 4⇑). All strains appeared to grow approximately equally well in MRS medium supplemented with glucose. In contrast, strains mutated in the manIIABCD (PTS9) operon or in the σ54-activator gene manR displayed impaired growth and lower biomass yields on MRS supplemented with mannose (Table 4⇑). The strain in which the manIIAB locus of PTS10 was deleted appeared to be unaffected in its capacity to grow on mannose-containing media. These experiments clearly establish that L. plantarum contains a single functional mannose PTS system, encoded by the genes lp_0575–0577 (manIIABCD or pts9ABCD), which is under control of σ54 in concert with the σ54-activator, or mannose-operon regulator, ManR (lp_0585).

    Comparative transcriptome analysis of wild-type L. plantarum and its rpoN derivative

    To evaluate the accuracy of the predictions of the σ54 target sites, transcriptome profiles of exponentially growing wild-type cells were compared to those of the rpoN-mutant cells, both grown on MRS medium supplemented with glucose, using genome-wide, amplicon-based DNA-microarrays.

    Microarray analysis identified nine genes that were expressed at a higher level in the wild-type compared to the rpoN mutant (Table 5⇓). The rpoN gene showed higher expression in the wild-type, which is in agreement with deletion of the gene in the mutant strain (Table 5⇓). The set of genes with highest relative expression in the wild-type was the manIIABCD genes of PTS9, confirming the involvement of σ54 in their transcriptional control. The other genes were expressed at a slightly higher level, indicating that they were less affected by rpoN deletion. Except for the mannose operon, the genes with higher expression levels in the wild-type compared to its rpoN derivative were not preceded by σ54-dependent promoters (Table 3⇑), suggesting that their regulation is not directly dependent on σ54.

    Table 5.

    Genes significantly up- and downregulated in the wild-type compared to the mutant NZ7306

    Putative operons are indicated in bold.

    In the rpoN mutant derivative, 23 genes showed higher expression compared to the wild-type. However, since sigma factors are not known to block transcription of genes, higher expression of these genes is most likely the result of secondary effects of the rpoN mutation.

    The global transcription analyses strongly suggest that the σ54 regulon of L. plantarum consists of a single locus, the mannose operon manIIABCD encoding PTS9. This is in agreement with the particular significance of the predicted σ54 promoter sequence upstream of this locus and its genetic linkage to the cognate, activator-encoding gene manR. As a consequence the consensus sequence for σ54-binding sites in L. plantarum is identical to the sequence upstream of the mannose operon encoded by lp_0575–0577 (see Table 3⇑).

    Characterization of the mannose PTS in L. plantarum WCFS1

    The mannose operon under control of σ54 belongs to a separate family of PTSs, the mannose family, characterized by a unique IID enzyme and a fused IIAB enzyme (Saier & Reizer, 1992; Zuniga et al., 2005). Although it is annotated as a mannose transporter, a similar PTS is also known to transport glucose in E. coli (Grenier et al., 1985), and additional homologues are presumably the major glucose-uptake systems in lactic acid bacteria (Chaillou et al., 2001). A similar role in L. plantarum would require expression of the mannose operon in cells grown on glucose, and therefore transcriptional analyses of the mannose PTS genes was performed in cells grown on different carbon sources. The mannose PTS appeared to be transcribed only in cells growing on glucose or mannose, and not in cells growing on lactose, maltose, fructose, cellobiose or sucrose (Fig. 2⇓), supporting a role for the mannose PTS in mannose and glucose transport.

    Figure image not available in archive
    Fig. 2.

    Transcription analysis of the mannose operon on different carbohydrates. Total RNA from exponentially growing cells was spotted and an internal 300 bp fragment of the manIIC gene was used as probe. The carbohydrates are indicated above the spots.

    Involvement of the mannose PTS in glucose transport could lead to decreased glucose uptake in strains lacking this PTS, as already described in Listeria monocytogenes and Lactobacillus pentosus (Chaillou et al., 2001; Vadyvaloo et al., 2004). When cells were grown on glucose-containing MRS, only a very minor, but consistent, reduction of the growth rate was observed in any of the mutants affected in mannose PTS expression (rpoN, manR, manIIABCD) compared to the wild-type (Table 4⇑). However, this difference in growth rates was more pronounced when cells were grown in chemically defined medium (CDM), again supporting a role for the mannose PTS in glucose import (Table 4⇑).

    Reduced glucose uptake rates in L. plantarum should be reflected in decreased rates of lactate formation rates, which can be evaluated by measuring the acidification rate in a weakly buffered cell suspension (Poolman et al., 1987). Wild-type cells were grown on either glucose or mannose, harvested and resuspended in the acidification assay buffer. Both suspensions acidified when glucose was added [acidification rate of 238±12.4 and 168±2.8 nmol protons (mg protein)−1 s−1, in glucose and mannose pre-grown cells, respectively]. This result primarily indicates that cells grown on mannose contain a complete glucose transport and utilization machinery, which also supports a role for the mannose PTS in glucose import in L. plantarum WCFS1.

    The acidification rate observed in glucose-containing assay buffer for glucose-grown NZ7306 (ΔrpoN) cells was significantly reduced compared to that of glucose-grown wild-type cells [acidification rate 160±8.5 nmol protons (mg protein)−1 s−1, versus 238±12.4 nmol protons (mg protein)−1 s−1 observed for the wild-type], indicating a reduced glycolysis rate in the rpoN mutant. In contrast to wild-type cells, the glucose-pre-grown NZ7306 cells were not able to convert mannose to lactate, confirming the crucial role of rpoN in regulation of mannose utilization in L. plantarum.

    Overall, these experiments support an important role for the mannose PTS (PTS9) of L. plantarum in glucose uptake.

    Pleiotropic effects of the mannose PTS on overall carbohydrate metabolism

    Apart from mannose and glucose, other carbohydrates have also been reported to be transported by homologues of the mannose PTS family (Vadeboncoeur & Pelletier, 1997). The availability of a mutant (manIIABCD : : P32cat) provided the opportunity to evaluate the substrate spectrum of the L. plantarum mannose PTS system extensively. To this end, growth characteristics of the mutant on a range of carbon sources were monitored and compared to those of the parental strain. The mutant and wild-type strains were pre-cultured on glucose, washed twice to remove all traces of glucose and inoculated in fresh media at an OD600 of approximately 0.05. To minimize the effects of undefined medium composition, cells were grown in CDM (Teusink et al., 2005). The maximal growth rate of the wild-type and NZ7306 were similar for a range of carbon sources, including cellobiose, maltose, N-acetylglucosamine and sucrose (data not shown). However, significant differences were observed between the wild-type and mutant strain when they were transferred to medium containing galactose as the sole carbon and energy source. The wild-type strain WCFS1 displayed a growth arrest that lasted for 20–30 h (Fig. 3⇓), irrespective of the pre-culture growth phase at the time of transfer [stationary phase of growth (overnight culture), or exponential growth phase (OD600 1.0), data not shown]. Notably, the growth initiation delay of the wild-type cells when transferred to galactose was not observed when a trace amount (0.001 %, w/v) of glucose was added to the galactose-containing medium (Fig. 3⇓). In contrast, all mutant strains that lack mannose PTS expression (NZ7306, NZ7307, NZ7308, NZ7309, NZ7310 and NZ7311; see Table 1⇑) were able to initiate growth immediately after inoculation (illustrated for strain NZ7311 in Fig. 3⇓), suggesting involvement of the mannose PTS in repression of galactose utilization. The direct involvement of mannose PTS expression in the delayed growth initiation in the wild-type was further supported by the observation that wild-type cells pre-cultured on CDM with maltose (a carbon source that does not lead to mannose PTS expression; see above) started to grow on galactose immediately after the medium transfer (Fig. 3⇓).

    Figure image not available in archive
    Fig. 3.

    Growth of L. plantarum WCFSI and its derivatives on galactose. The wild-type strain WCFS1 (○) shows a prolonged growth arrest compared to the mannose PTS deletion mutant NZ7311 (◊) after transfer from glucose-containing to galactose-containing CDM. Strain WCFS1 grown on maltose (▪) or activated with glucose (△) initiates growth immediately after inoculation. For details see text.

    The mannose PTS drains the PEP pool

    Since no differential expression of galactose utilization genes was found in the comparative transcriptome analysis of the wild-type and its rpoN derivative, the regulation of galactose catabolism could be due to post-transcriptional regulation. This could possibly involve phosphorylation of the L. plantarum LacS-type galactose transporter by PEP, as has been described in Streptococcus thermophilus. In this bacterium galactose is taken up by a lactose/galactose transport protein (LacS), which catalyses two modes of transport: solute-H+ symport and lactose/galactose antiport (Foucaud & Poolman, 1992). The streptococcal LacS possesses a PTS-IIA-like domain, and phosphorylation of this domain led to an increased antiporter transport rate, while it did not affect its symporter-mode transport rate (Poolman et al., 1995). L. plantarum has two putative galactose transporters (lasS1 and lacS2) that share significant similarity with each other (38 % identity) and with the galactose/lactose transporter of S. thermophilus (37 % and 56 % identity, respectively). Furthermore, both transporters have a phosphate-accepting IIA domain, suggesting that a similar regulatory mechanism as has been observed in S. thermophilus might affect transport modes of the L. plantarum homologues.

    To investigate a potential role of LacS phosphorylation, the intracellular concentration of PEP was monitored in L. plantarum wild-type and the ManPTS mutant strain when grown in glucose-containing CDM followed by carbon starvation. To exclude the possibility of mannose PTS expression due to any leakage of the rpoN-dependent promoter, these measurements were performed using the manIIABCD deletion strain NZ7311 and the permease (manIIC) deletion strain NZ7308 (Table 1⇑). In samples taken from wild-type and NZ7311 cultures growing on CDM containing glucose, no PEP could be detected, indicating that intracellular PEP concentrations during growth are low (data not shown). To monitor PEP pool development upon carbon starvation, cells were grown to mid-exponential growth phase, harvested, washed, and resuspended in medium without a carbon source. Samples were taken from these suspensions at different time points after the medium transfer, and intracellular PEP concentrations were determined. In the wild-type strain, no PEP could be detected in any of the samples, indicating that this strain fails to accumulate PEP upon carbon starvation (Fig. 4⇓). In contrast, in the mutant strains NZ7308 and NZ7311, PEP was detected immediately after transfer to the starvation medium, and appeared to be maintained at a concentration of 1–3 μM up to several hours after carbon starvation (Fig. 4⇓). These results suggest that the expression of the mannose PTS significantly affects the capacity of the wild-type cells to accumulate PEP during starvation, which may hamper initiation of growth on galactose due to a lack of phosphate-donor molecules required for activation of the galactose transporter by LacS-IIA domain phosphorylation. In contrast, the relatively high PEP pools in resting cells of the mutant strain would support LacS activation, allowing immediate growth initiation upon transfer to galactose-containing media. The observation that trace amounts of glucose facilitated growth initiation of the wild-type strain on galactose can be explained by the production of PEP from glucose and the corresponding LacS activation under these conditions.

    Figure image not available in archive
    Fig. 4.

    Effect of mannose PTS deficiency on intracellular PEP pools: concentration of intracellular PEP in carbon-starved resting cells of L. plantarum wild-type (□) and its mannose PTS-mutant derivatives NZ7311 (▪) and NZ7308 (△). Mutants NZ7311 and NZ7308 differ significantly from the wild-type (P=3.66×10−5 and 1.0×10−4 respectively), but not from each other (P=0.19).

    Overall, these results indicate that the mannose PTS system of L. plantarum plays a pivotal role in carbon metabolism control. This role clearly exceeds the ‘simple’ role that is predicted on the basis of its annotated function, e.g. transport of mannose (and glucose), and includes control of central carbon metabolism by affecting the relative levels of important glycolytic intermediates such as PEP, which in their turn control the capacity to initiate growth on certain other carbon sources.

    DISCUSSION

    Regulation of gene expression by alternative sigma factors is a well-established mechanism of adaptation to environmental conditions in bacteria. Here we describe the prediction and validation of the σ54 regulon in L. plantarum WCFS1. The conserved binding sequence of this sigma factor enabled a prediction of the specific regulon using the pattern recognition algorithms MEME and MAST (Bailey & Elkan, 1994; Bailey & Gribskov, 1998). The prediction was verified by comparative whole-genome transcriptome analysis of the L. plantarum wild-type and its σ54-mutant derivative. The regulon was concluded to be restricted to the mannose PTS operon, which is in agreement with the genetic linkage between the gene encoding the σ54-activator protein and the target locus of the mannose PTS.

    The relatively higher expression in the wild-type strain of genes involved in maltose utilization (Table 5⇑) indicates involvement of σ54 in control of utilization of other carbohydrates. However, the predicted regulation of the maltose genes by a LacI-transcription regulator (Francke et al., 2008), and the virtually identical growth characteristics of the σ54 mutant and the wild-type strain on CDM supplemented with maltose (data not shown), contradict such a role for σ54. The higher expression in the wild-type of a ribonucleotide reductase, an enzyme that catalyses the reduction of ribonucleosides, thereby providing the building blocks required for DNA replication, is possibly caused by the higher growth rate of the wild-type. Careful manual inspection of upstream regions of the genes showing higher expression failed to identify candidate σ54-dependent promoters (Table 3⇑), and their regulation is therefore likely to be due to unidentified secondary effects and not directly dependent on σ54.

    The gene located immediately downstream of rpoN, the central glycolytic gene regulator (lp_0788; cggR), is expressed at a higher level in the mutant, which is probably due to readthrough of the P32 promoter upstream of the cat gene that replaces rpoN in the mutant. CggR probably regulates the expression of four glycolytic genes in L. plantarum (gap, pgk, tpi, eno) that lie immediately downstream of cggR. However, no differential expression of these glycolytic genes was observed in the transcriptome analyses, indicating that the effect of promoter readthrough is minimal. Furthermore, preliminary results in our laboratory suggest that overexpression of cggR does not affect the expression of glycolytic genes in strain WCFS1 (I. Rud and others, unpublished results). However, to exclude that any polar effects of the rpoN mutation were confounding the PEP measurements, the experiments to elucidate the role of the mannose PTS in modulation of the intracellular PEP pools were performed with mannose-PTS-deficient mutants.

    The operon encoding the fatty acid biosynthesis machinery (lp_1670–1681) showed higher expression in the σ54 mutant, probably due to the decreased glycolytic rates observed in this strain, which may lead to an altered flux to malonyl-CoA, an essential intermediate in fatty acid biosynthesis which relieves repression of the fab genes in B. subtilis (Schujman et al., 2006). The apparently higher expression in the mutant of an oxidoreductase encoded by ORF lp_0291 is an artefact due to the presence of a 189 bp fragment of this gene on the plasmid used for cloning, which is located directly downstream of the cat gene (Lambert et al., 2007). Overall, the increased expression of these genes in the rpoN mutant is most likely due to secondary effects, which is in agreement with the assumption that sigma factors are generally not involved in the repression of genes.

    The inability of the σ54-mutant strain NZ7306 to grow on mannose as sole carbon source showed that transcription of the mannose operon is strictly σ54-dependent, which is in agreement with the observation that σ54-regulated genes are generally not influenced by other transcription factors (Buck et al., 2000). Growth analyses of strains mutated in the mannose PTS showed that ManIIABCD is the only functional mannose transporter in L. plantarum.

    Regulation of the mannose operon by σ54 has already been described in the Gram-positive bacteria Enterococcus faecalis, Listeria monocytogenes and Lactobacillus casei (Dalet et al., 2001; Hechard et al., 2001; Yebra et al., 2004), indicating that this mode of regulation is conserved among some Gram-positive bacteria. In Ent. faecalis and Lis. monocytogenes mutation of the σ54-encoding gene led to resistance to mesentericin Y105, a class II bacteriocin produced by Leuconostoc mesenteroides, which was shown to be due to impaired manIIC expression (Dalet et al., 2000; Ramnath et al., 2004; Robichon et al., 1997). Additional studies proposed a conserved mechanism for bacteriocin sensitivity in which ManIIC acts as a docking protein for antimicrobial peptides (Diep et al., 2007). However, L. plantarum appeared to be resistant to mesentericin Y105 (M. J. A. Stevens, unpublished data), suggesting that the L. plantarum mannose PTS does not act as a docking protein for this bacteriocin.

    Homologues of the mannose PTS studied in this paper are found throughout the bacterial kingdom. It has been shown that this PTS also transports glucose and it is additionally thought to be important as a glucose PTS in several lactic acid bacteria (Chaillou et al., 2001). The observation that mannose PTS deletion leads to a decreased growth rate on glucose has already been described for Lis. monocytogenes and Lactobacillus pentosus (Chaillou et al., 2001; Vadyvaloo et al., 2004). Our data provide further evidence for glucose transport by the mannose PTS in L. plantarum. Notably, the codon adaptation index of the mannose-PTS-encoding genes was high, suggesting that this transport system can be produced at a high level (Kleerebezem et al., 2003); this is in agreement with its importance in uptake of favourable sugars such as glucose.

    The role of the mannose PTS in glucose transport suggests a role of this PTS in canonical systems for control of carbon utilization such as carbon catabolite repression (Chaillou et al., 2001). Indeed, this has been shown in the close relative of L. plantarum, L. casei, in which the mannose PTS regulates lactose operon expression via terminator modulation; only strains lacking CcpA and mannose PTS activity are able to express the lactose operon (Chaillou et al., 2001; Gosalbes et al., 1997). However, the expression of both lacS1 and lacS2 in a strain lacking a functional CcpA but still harbouring a mannose PTS indicates a different mechanism in L. plantarum (Stevens, 2008). Since we did not observe an effect in the transcription analysis, the regulatory role of the mannose PTS seems to be exerted at another level, possibly involving metabolic control of transport.

    Phosphorylation of a IIA-like domain of the S. thermophilus antiporter/symporter LacS leads to an increased antiporter transport rate, whereas symporter transport rate is not affected (Poolman et al., 1995). The antiporter mode (galactose/lactose exchange) is the most relevant transport mode as it is much faster than the proton-motive force (PMF)-driven symporter (lactose/H+) mode (Knol et al., 1996). The wild-type cells in our experiment were depleted of metabolic energy, as no PEP could be detected in these cells. Consequently, the prolonged growth arrest could be due to inability to build up the PMF needed for galactose transport initiation. The addition of a trace amount of glucose to resting cells (as in our experiments) has been shown to lead to changes in the concentrations of glycolytic intermediates (Neves et al., 2002) and to the generation of a PMF (Poolman et al., 1995), enabling galactose/H+ transport and eventually growth initiation.

    Cells mutated in the mannose PTS maintain a high intracellular PEP pool. Phosphorylation of LacS is (His∼P)-HPr and PEP-dependent (Gunnewijk & Poolman, 2000); hence it is likely that the high PEP pool in the mutant strain leads to a permanent LacS-(IIA-P) state, allowing efficient galactose import immediate after inoculation.

    At first sight, it may seem that σ54 has only a minor role in L. plantarum since only the mannose operon is transcribed from a σ54-dependent promoter. However, the mannose PTS operon encodes a glucose-uptake system in L. plantarum and σ54-dependent transcription allows expression of the operon at high level without the involvement of additional transcriptional regulators in glucose-containing media. Thereby, σ54 exerts metabolic control via the strict control of mannose PTS expression, which affects the concentration of PEP in carbon-starved cells, thereby influencing the energy state of these cells and modulating their capacity to initiate growth on other carbon sources, such as galactose. This metabolic control is apparently directly linked to glucose transport, and the conservation of mannose PTS regulation by σ54 in other Gram-positive bacteria suggests similar control in other lactic acid bacteria and Listeria spp.

    Acknowledgments

    This work was supported by grant no. IGE1018 from the Dutch IOP-Genomics Program. We thank Guido Starring for performing the HPLC analyses, Yann Hechard (Université de Poitiers) for providing the mesentericin-producing strain Leuconostoc mesenteroides Y105, and Pascal Hols (Université de Louvain) for critically reading the manuscript.

    References

    1. Ahrne, S., Nobaek, S., Jeppsson, B., Adlerberth, I., Wold, A. E. & Molin, G. (1998). The normal Lactobacillus flora of healthy human rectal and oral mucosa. J Appl Microbiol 85, 88–94.
    2. Ali, N. O., Bignon, J., Rapoport, G. & Debarbouille, M. (2001). Regulation of the acetoin catabolic pathway is controlled by sigma L in Bacillus subtilis. J Bacteriol 183, 2497–2504.
    3. Bailey, T. L. & Elkan, C. (1994). Fitting a mixture model by expectation maximization to discover motifs in biopolymers. Proc Int Conf Intell Syst Mol Biol 2, 28–36.
    4. Bailey, T. L. & Gribskov, M. (1998). Combining evidence using p-values: application to sequence homology searches. Bioinformatics 14, 48–54.
    5. Barrios, H., Valderrama, B. & Morett, E. (1999). Compilation and analysis of σ54-dependent promoter sequences. Nucleic Acids Res 27, 4305–4313.
    6. Bhattacharya, M., Fuhrman, L., Ingram, A., Nickerson, K. W. & Conway, T. (1995). Single-run separation and detection of multiple metabolic intermediates by anion-exchange high-performance liquid chromatography and application to cell pool extracts prepared from Escherichia coli. Anal Biochem 232, 98–106.
    7. Boogaard, V. D. (2002). Catabolite control of sugar metabolism in Streptococcus thermophilus. PhD thesis, Wageningen University, Wageningen, The Netherlands.
    8. Bron, P. A. (2004). The molecular response of Lactobacillus plantarum to intestinal passage and conditions. PhD thesis, Wageningen University, Wageningen, The Netherlands.
    9. Buck, M. (1986). Deletion analysis of the Klebsiella pneumoniae nitrogenase promoter: importance of spacing between conserved sequences around positions −12 and −24 for activation by the nifA and ntrC (glnG) products. J Bacteriol 166, 545–551.
    10. Buck, M., Miller, S., Drummond, M. & Dixon, R. (1986). Upstream activator sequences are present in the promoters of nitrogen fixation genes. Nature 320, 374–377.
    11. Buck, M., Gallegos, M. T., Studholme, D. J., Guo, Y. & Gralla, J. D. (2000). The bacterial enhancer-dependent σ54 (σN) transcription factor. J Bacteriol 182, 4129–4136.
    12. Chaillou, S., Postma, P. W. & Pouwels, P. H. (2001). Contribution of the phosphoenolpyruvate : mannose phosphotransferase system to carbon catabolite repression in Lactobacillus pentosus. Microbiology 147, 671–679.
    13. Dalet, K., Briand, C., Cenatiempo, Y. & Hechard, Y. (2000). The rpoN gene of Enterococcus faecalis directs sensitivity to subclass IIa bacteriocins. Curr Microbiol 41, 441–443.
    14. Dalet, K., Cenatiempo, Y., Cossart, P. & Hechard, Y. (2001). A σ54-dependent PTS permease of the mannose family is responsible for sensitivity of Listeria monocytogenes to mesentericin Y105. Microbiology 147, 3263–3269.
    15. De Man, J. D., Rogosa, M. & Sharpe, M. E. (1960). A medium for the cultivation of lactobacilli. J Appl Bacteriol 23, 130–135.
    16. de Vries, M. C., Vaughan, E. E., Kleerebezem, M. & De Vos, W. M. (2006). Lactobacillus plantarum – survival, functional and potential probiotic properties in the human intestinal tract. Int Dairy J 16, 1018–1028.
    17. Debarbouille, M., Martin-Verstraete, I., Kunst, F. & Rapoport, G. (1991). The Bacillus subtilis sigL gene encodes an equivalent of σ54 from Gram-negative bacteria. Proc Natl Acad Sci U S A 88, 9092–9096.
    18. Debarbouille, M., Gardan, R., Arnaud, M. & Rapoport, G. (1999). Role of BkdR, a transcriptional activator of the SigL-dependent isoleucine and valine degradation pathway in Bacillus subtilis. J Bacteriol 181, 2059–2066.
    19. Diep, D. B., Skaugen, M., Salehian, Z., Holo, H. & Nes, I. F. (2007). Common mechanisms of target cell recognition and immunity for class II bacteriocins. Proc Natl Acad Sci U S A 104, 2384–2389.
    20. Ferain, T., Garmyn, D., Bernard, N., Hols, P. & Delcour, J. (1994). Lactobacillus plantarum ldhL gene: overexpression and deletion. J Bacteriol 176, 596–601.
    21. Finn, R. D., Mistry, J., Schuster-Bockler, B., Griffiths-Jones, S., Hollich, V., Lassmann, T., Moxon, S., Marshall, M., Khanna, A. & other authors (2006). Pfam: clans, web tools and services. Nucleic Acids Res 34, D247–D251.
    22. Foucaud, C. & Poolman, B. (1992). Lactose transport system of Streptococcus thermophilus. Functional reconstitution of the protein and characterization of the kinetic mechanism of transport. J Biol Chem 267, 22087–22094.
    23. Francke, C., Kerkhoven, R., Wels, M. & Siezen, R. J. (2008). A generic approach to identify transcription factor-specific operator motifs; inferences for LacI-family mediated regulation in Lactobacillus plantarum WCFS1. BMC Genomics 9, 145
    24. Garcia, E., Bancroft, S., Rhee, S. G. & Kustu, S. (1977). The product of a newly identified gene, gInF, is required for synthesis of glutamine synthetase in Salmonella. Proc Natl Acad Sci U S A 74, 1662–1666.
    25. Gardan, R., Rapoport, G. & Debarbouille, M. (1997). Role of the transcriptional activator RocR in the arginine-degradation pathway of Bacillus subtilis. Mol Microbiol 24, 825–837.
    26. Gosalbes, M. J., Monedero, V., Alpert, C. A. & Perez-Martinez, G. (1997). Establishing a model to study the regulation of the lactose operon in Lactobacillus casei. FEMS Microbiol Lett 148, 83–89.
    27. Grenier, F. C., Waygood, E. B. & Saier, M. H., Jr (1985). Bacterial phosphotransferase system: regulation of the glucose and mannose enzymes II by sulfhydryl oxidation. Biochemistry 24, 4872–4876.
    28. Gunnewijk, M. G. & Poolman, B. (2000). HPr(His∼P)-mediated phosphorylation differently affects counterflow and proton motive force-driven uptake via the lactose transport protein of Streptococcus thermophilus. J Biol Chem 275, 34080–34085.
    29. Hechard, Y., Pelletier, C., Cenatiempo, Y. & Frere, J. (2001). Analysis of σ54-dependent genes in Enterococcus faecalis: a mannose PTS permease (EIIMan) is involved in sensitivity to a bacteriocin, mesentericin Y105. Microbiology 147, 1575–1580.
    30. Josson, K., Scheirlinck, T., Michiels, F., Platteeuw, C., Stanssens, P., Joos, H., Dhaese, P., Zabeau, M. & Mahillon, J. (1989). Characterization of a Gram-positive broad-host-range plasmid isolated from Lactobacillus hilgardii. Plasmid 21, 9–20.
    31. Kleerebezem, M., Boekhorst, J., van Kranenburg, R., Molenaar, D., Kuipers, O. P., Leer, R., Tarchini, R., Peters, S. A., Sandbrink, H. M. & other authors (2003). Complete genome sequence of Lactobacillus plantarum WCFS1. Proc Natl Acad Sci U S A 100, 1990–1995.
    32. Knol, J., Veenhoff, L., Liang, W. J., Henderson, P. J., Leblanc, G. & Poolman, B. (1996). Unidirectional reconstitution into detergent-destabilized liposomes of the purified lactose transport system of Streptococcus thermophilus. J Biol Chem 271, 15358–15366.
    33. Lambert, J. M., Bongers, R. S. & Kleerebezem, M. (2007). Cre-lox-based system for multiple gene deletions and selectable-marker removal in Lactobacillus plantarum. Appl Environ Microbiol 73, 1126–1135.
    34. Merrick, M. J. (1993). In a class of its own – the RNA polymerase sigma factor σ54 (σN). Mol Microbiol 10, 903–909.
    35. Merrick, M. J. & Gibbins, J. R. (1985). The nucleotide sequence of the nitrogen-regulation gene ntrA of Klebsiella pneumoniae and comparison with conserved features in bacterial RNA polymerase sigma factors. Nucleic Acids Res 13, 7607–7620.
    36. Molenaar, D., Bringel, F., Schuren, F. H., de Vos, W. M., Siezen, R. J. & Kleerebezem, M. (2005). Exploring Lactobacillus plantarum genome diversity by using microarrays. J Bacteriol 187, 6119–6127.
    37. Morett, E. & Buck, M. (1989). In vivo studies on the interaction of RNA polymerase-σ54 with the Klebsiella pneumoniae and Rhizobium meliloti nifH promoters. The role of NifA in the formation of an open promoter complex. J Mol Biol 210, 65–77.
    38. Neves, A. R., Ramos, A., Costa, H., van Swam, I. I., Hugenholtz, J., Kleerebezem, M., de Vos, W. & Santos, H. (2002). Effect of different NADH oxidase levels on glucose metabolism by Lactococcus lactis: kinetics of intracellular metabolite pools determined by in vivo nuclear magnetic resonance. Appl Environ Microbiol 68, 6332–6342.
    39. Pieterse, B., Jellema, R. H. & van der Werf, M. J. (2005). Quenching of microbial samples for increased reliability of microarray data. J Microbiol Methods 64, 207–216.
    40. Poolman, B., Smid, E. J., Veldkamp, H. & Konings, W. N. (1987). Bioenergetic consequences of lactose starvation for continuously cultured Streptococcus cremoris. J Bacteriol 169, 1460–1468.
    41. Poolman, B., Knol, J., Mollet, B., Nieuwenhuis, B. & Sulter, G. (1995). Regulation of bacterial sugar-H+ symport by phosphoenolpyruvate-dependent enzyme I/HPr-mediated phosphorylation. Proc Natl Acad Sci U S A 92, 778–782.
    42. Ramnath, M., Arous, S., Gravesen, A., Hastings, J. W. & Hechard, Y. (2004). Expression of mptC of Listeria monocytogenes induces sensitivity to class IIa bacteriocins in Lactococcus lactis. Microbiology 150, 2663–2668.
    43. Reitzer, L. & Schneider, B. L. (2001). Metabolic context and possible physiological themes of σ54-dependent genes in Escherichia coli. Microbiol Mol Biol Rev 65, 422–444.
    44. Robichon, D., Gouin, E., Debarbouille, M., Cossart, P., Cenatiempo, Y. & Hechard, Y. (1997). The rpoN (σ54) gene from Listeria monocytogenes is involved in resistance to mesentericin Y105, an antibacterial peptide from Leuconostoc mesenteroides. J Bacteriol 179, 7591–7594.
    45. Saier, M. H., Jr & Reizer, J. (1992). Proposed uniform nomenclature for the proteins and protein domains of the bacterial phosphoenolpyruvate : sugar phosphotransferase system. J Bacteriol 174, 1433–1438.
    46. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
    47. Sasse-Dwight, S. & Gralla, J. D. (1988). Probing the Escherichia coli glnALG upstream activation mechanism in vivo. Proc Natl Acad Sci U S A 85, 8934–8938.
    48. Schujman, G. E., Guerin, M., Buschiazzo, A., Schaeffer, F., Llarrull, L. I., Reh, G., Vila, A. J., Alzari, P. M. & de Mendoza, D. (2006). Structural basis of lipid biosynthesis regulation in Gram-positive bacteria. EMBO J 25, 4074–4083.
    49. Smyth, G. (2005). Statistical issues in microarray data analysis. In Functional Genomics: Methods and Protocols, pp. 111–136. Edited by M. J. Braunstein & A. B. Khodursky. Totowa: Humana Press.
    50. Stevens, M. (2008). Transcriptome response of Lactobacillus plantarum to global regulator deficiency, stress and other enviromental conditions. PhD thesis, Wageningen University, Wageningen, The Netherlands.
    51. Stevens, M. J., Wiersma, A., de Vos, W. M., Kuipers, O. P., Smid, E. J., Molenaar, D. & Kleerebezem, M. (2008). Improvement of Lactobacillus plantarum aerobic growth as directed by comprehensive transcriptome analysis. Appl Environ Microbiol 74, 4776–4778.
    52. Studholme, D. J. & Buck, M. (2000). The biology of enhancer-dependent transcriptional regulation in bacteria: insights from genome sequences. FEMS Microbiol Lett 186, 1–9.
    53. Studholme, D. J. & Dixon, R. (2003). Domain architectures of σ54-dependent transcriptional activators. J Bacteriol 185, 1757–1767.
    54. Teusink, B., van Enckevort, F. H., Francke, C., Wiersma, A., Wegkamp, A., Smid, E. J. & Siezen, R. J. (2005). In silico reconstruction of the metabolic pathways of Lactobacillus plantarum: comparing predictions of nutrient requirements with those from growth experiments. Appl Environ Microbiol 71, 7253–7262.
    55. Vadeboncoeur, C. & Pelletier, M. (1997). The phosphoenolpyruvate : sugar phosphotransferase system of oral streptococci and its role in the control of sugar metabolism. FEMS Microbiol Rev 19, 187–207.
    56. Vadyvaloo, V., Snoep, J. L., Hastings, J. W. & Rautenbach, M. (2004). Physiological implications of class IIa bacteriocin resistance in Listeria monocytogenes strains. Microbiology 150, 335–340.
    57. Wang, L. & Gralla, J. D. (1998). Multiple in vivo roles for the −12-region elements of σ54 promoters. J Bacteriol 180, 5626–5631.
    58. Woodcock, D. M., Crowther, P. J., Doherty, J., Jefferson, S., DeCruz, E., Noyer-Weidner, M., Smith, S. S., Michael, M. Z. & Graham, M. W. (1989). Quantitative evaluation of Escherichia coli host strains for tolerance to cytosine methylation in plasmid and phage recombinants. Nucleic Acids Res 17, 3469–3478.
    59. Yebra, M. J., Viana, R., Monedero, V., Deutscher, J. & Perez-Martinez, G. (2004). An esterase gene from Lactobacillus casei cotranscribed with genes encoding a phosphoenolpyruvate : sugar phosphotransferase system and regulated by a LevR-like activator and σ54 factor. J Mol Microbiol Biotechnol 8, 117–128.
    60. Zuniga, M., Comas, I., Linaje, R., Monedero, V., Yebra, M. J., Esteban, C. D., Deutscher, J., Perez-Martinez, G. & Gonzalez-Candelas, F. (2005). Horizontal gene transfer in the molecular evolution of mannose PTS transporters. Mol Biol Evol 22, 1673–1685.