| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 CNRS, UMR7156, Génétique Moléculaire, Génomique Microbiologie, Département Microorganismes, Génomes, Environnement, 28 Rue Goethe, 67083 Strasbourg, France
2 Université Louis-Pasteur Strasbourg-I, Strasbourg, France
3 CNRS, FRC 1589, Plateforme Protéomique Esplanade, 15 Rue René Descartes, 67084 Strasbourg, France
Correspondence
Florence Arsène-Ploetze
florence.ploetze{at}gem.u-strasbg.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A supplementary figure showing the results of differential proteomic analysis in response to Ci availability is available with the online version of this paper.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
), Bifidobacterium species (Kawasaki et al., 2007) and Lactobacillus plantarum (Arsène-Ploetze & Bringel, 2004). Higher Ci concentrations, obtained by increased CO2 tension in the gas phase (around 4 %) or addition of KHCO3 in the medium, stimulate the growth of L. plantarum (Arsène-Ploetze & Bringel, 2004). L. plantarum is a versatile Gram-positive, lactic acid bacterium whose genome has been sequenced (Kleerebezem et al., 2003). This bacterium can be found in different biotopes such as plants and mammalian cavities, and it is involved in many food fermentation processes. In such biotopes, this homofermentative lactic acid bacterium often grows in the presence of other CO2- and acid-producing micro-organisms.
The modulation of bacterial gene expression in response to changes in Ci availability has been described for genes that encode enzymes involved in carbon assimilation in autotrophs (Price et al., 2008; Woodger et al., 2007). Several virulence genes are also regulated in response to Ci availability. CO2/bicarbonate is likely to be a physiologically significant signal encountered by pathogens in the host environment (Drysdale et al., 2005; Herbert et al., 2001; Koehler, 2002; Hyde et al., 2007). Ci-responding regulation is not restricted to carbon assimilation or virulence genes. Regulation of gene expression in response to Ci concentrations was proposed for genes involved in general stress response or purine metabolism in cyanobacteria (Wang et al., 2004), the cad operon in Escherichia coli (Takayama et al., 1994), uncharacterized genes in Pseudomonas sp. S91 (Stretton et al., 1996) and for pyr genes in two lactic acid bacteria species, L. plantarum and Lactococcus lactis (Arsène-Ploetze et al., 2006a; Maligoy et al., 2008). The L. plantarum de novo pyrimidine biosynthesis operon and pyrP gene that code for the PyrR1-B-C-Aa1-Ab1-D-E-F-P proteins and the uracil permease, involved in preformed pyrimidine rescue, are differently expressed in response to Ci availability. L. plantarum regulation of pyr gene expression is particularly unusual compared to other known Gram-positive bacteria, since two PyrR regulators, with 62 % amino acid identity, are present. PyrR1 is an RNA-binding protein responding to pyrimidine availability via a mechanism of transcriptional attenuation and regulating pyr gene expression in response to pyrimidine availability (Nicoloff et al., 2005). PyrR2 modulates the pyr gene expression in response to Ci concentration via an unknown mechanism (Arsène-Ploetze et al., 2006a). Deletion of the pyrR2 gene prevents growth in the absence of pyrimidines and arginine in limiting Ci concentrations, a phenotype called HCR (high-CO2-requiring) prototrophy (Arsène-Ploetze et al., 2006a). This HCR phenotype was clearly correlated with decreased pyr gene expression in several mutants (Nicoloff et al., 2005). However, other uncharacterized mutants sharing the HCR phenotypes were not impaired in the regulation of the pyr gene expression, suggesting that the molecular basis of HCR growth phenotype is not fully understood (Bringel et al., 2008). In this study, we questioned if other Ci-responding genes are controlled by PyrR2.
Ci has been proposed to be the growth-limiting factor in L. plantarum in normal air by limiting synthesis of carbamoyl phosphate (CP), a common intermediate of the pyrimidines and the arginine biosynthesis pathways (Fig. 1) (Nicoloff et al., 2005). Among the genes whose expression may confer Ci-dependent growth, the arginine-regulated genes, including the carbamoyl phosphate synthase (CPS) carAB operon, were good candidates. The carAB genes are close to, but oriented in the opposite direction from, the argCJBDF-ccl operon, which encodes the other enzymes required for citrulline biosynthesis (Nicoloff et al., 2000). Citrulline is subsequently metabolized into arginine by argininosuccinate synthase and argininosuccinate lyase, encoded by argG and argH, respectively. These two genes are located in a different genomic region from the other arg and carAB genes. The carAB and all biosynthetic arg genes are part of an arginine-repressed regulon. Expression of the arg and car genes is regulated in response to arginine by both ArgR1 and ArgR2 (Nicoloff et al., 2004). A genetic link between arginine regulation and another as yet uncharacterized CO2-dependent metabolism was proposed (Nicoloff et al., 2004). Another Ci-responding candidate gene was upp, which encodes a uracil phosphoribosyltransferase involved in a uracil salvage pathway as its mutation confers the HCR phenotype (Arsène-Ploetze et al., 2006b). The uncharacterized cah gene (accession number AL935263, EMBL database), encoding a putative carbonic anhydrase that catalyses the hydration of CO2 to 
, was also included in our study, because the cyanobacterial carbonic anhydrase is regulated at the transcriptional level in response to CO2 concentrations (Soltes-Rak et al., 1997). In order to identify genes that may be up- or downregulated in response to Ci availability, and to check if the proteins encoded by the car, arg, upp and cah genes accumulated differentially in response to Ci, a highly sensitive proteomic approach using differential 2D gel electrophoresis (DIGE) was combined with slot-blot experiments.
|
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
pyrR2 NC8 derivative strain (FB422) (Arsène-Ploetze et al., 2006a) were grown routinely on MRS agar plates (Difco) at 30 °C. Liquid cultures were grown in agitated DLA rich medium (Bringel et al., 1997) supplemented with 50 µg uracil l–1 (where appropriate). High Ci concentration in the liquid media was obtained by increasing the CO2 tension in the gas phase (4 %). This method has the same effect on growth stimulation as the addition of KHCO3 (2 g l–1 final concentration) to the liquid medium (Arsène-Ploetze & Bringel, 2004; Arsène-Ploetze et al., 2006a). For RNA and protein studies, cells were first cultivated in DLA in normal air-agitated precultures. The latter were used as inocula for 50 ml cultures in liquid DLA; the inoculum was added to obtain an OD600 of 0.05 and the culture was stopped when the OD600 reached 0.5 (approx. 4 h and two generations in the exponential growth phase).
Protein extractions and separations.
Cells were harvested by centrifugation (10 min at 5000 g). The cell lysis protocol used was as described previously (Arsène-Ploetze et al., 2006a). Protein concentration in the supernatant was quantified using the Bradford assay (Bradford, 1976) standardized with known concentrations of BSA.
For DIGE studies, the experimental design was implemented to analyse differences between samples prepared from cells grown with low or high Ci level, and labelled with CyDye DIGE fluors (Cy2, Cy3 or Cy5; GE Healthcare Biosciences), according to the manufacturer's instructions. Three hundred micrograms of extracted proteins was precipitated using the Clean-up kit (GE Healthcare Biosciences) according to the manufacturer's instructions. The pellet was resuspended in 30 µl sample buffer [8 M urea, 130 mM DTT, 4 % (w/v) CHAPS, 2 % (v/v) pharmalyte pH range 3–10 (GE Healthcare Biosciences)]. Protein concentration was estimated using the Proteins 2D quant kit (GE Healthcare Biosciences), according to the manufacturer's instructions. The pH of protein extracts was adjusted to between 8.5 and 8.8, by adding a few microlitres of 0.5 M NaOH. The protein extract was quantified once more before labelling was performed. Eight samples were labelled, corresponding to four replicates prepared from cells grown either with low or high Ci concentrations (four independent cultures in each condition). Minimal labelling was performed by mixing proteins (50 µg) from each of the four samples with either Cy3 (two samples) or Cy5 (two samples) DIGE fluors (400 pmol) (GE Healthcare Biosciences), to prevent any bias that may result from differential labelling efficiency. A pooled set of internal standards, comprising 25 µg aliquots from each of the eight samples (total 200 µg), was minimally labelled with Cy2 DIGE fluors. The labelling was performed for 30 min on ice, in the dark. Lysine solution (10 mM in aqueous solution) was added to stop the reaction (1 µl for Cy3- or Cy5-labelled extracts or 4 µl for Cy2-labelled extracts). Samples were incubated for 10 min on ice in the dark. Sample buffer was added to the labelled extracts in a 1 : 1 ratio and left on ice for 10 min. Finally, the protein samples that were separated on the same gel were pooled: one Cy3- and one Cy5-labelled sample of either low- or high-Ci level-grown cells and one-eighth volume of the pooled set of internal standards. Rehydration buffer was added to obtain 350 µl (final volume) before IEF separation.
Proteins were first separated according to their pI, on 18 cm IPG strips (linear gradient pH 4–7) in the IPGphor isoelectric focusing system equipped with a cup loading system (manifold) (GE Healthcare Biosciences), as previously described (Arsène-Ploetze et al., 2006a). IEF was carried out to obtain 53 000 V h. Strips were equilibrated in equilibration buffer (50 mM Tris/HCl pH 8.8, 6 M urea, 30 % glycerol, 2 % SDS, trace of bromophenol blue), twice for 15 min, first in the presence of DTT (10 mg ml–1) and then in the presence of iodoacetamide (25 mg ml–1). The second-dimension electrophoretic separation was performed by 12 % SDS-PAGE.
Differential protein expression analysis.
Two-dimensional gel analysis was performed using the DIGE algorithms of ImageMaster 2D platinum software (v. 6.01, GE Healthcare Biosciences). Twelve images obtained from four gels (three images each) were analysed. After spot detection (around 300 spots per gel), each protein spot in a sample was compared to its corresponding spot in the internal standard on the same gel (Cy2-labelled extracts), to generate a ratio of relative protein levels. The Student's t-value and the ‘ratio’ (mean of the relative volumes obtained at limiting Ci condition divided by the mean of the relative volumes obtained at high Ci concentrations) were calculated for each spot according to the 2D platinum software manual. Only spots having a Student's t-value greater than 3 (P-value less than 0.025) and ratio greater than 1.2 were analysed. One of the DIGE gels was silver stained in order to pick spots to be stored at –20 °C before mass spectrometry analysis.
Mass spectrometry
In-gel digestion.
Picked spots were washed with 100 µl of 25 mM NH4HCO3 buffer and dehydrated with 100 µl acetonitrile, and the whole process was repeated. The samples were vacuum-dried for 10 min, reduced [10 mM DTT/25 mM NH4HCO3 buffer (100 µl) at 56 °C for 1 h] and alkylated [25 mM iodoacetamide/25 mM NH4HCO3 buffer (100 µl) at room temperature in the dark for 1 h]. After three washes for 5 min in 25 mM NH4HCO3 and acetonitrile alternately, samples were vacuum-dried, rehydrated overnight at room temperature in the presence of trypsin [3 volumes of 12.5 ng trypsin (Promega, V5111) µl–1 in 25 mM NH4HCO3 buffer, freshly diluted]. Tryptic peptides were extracted from the gels by sonication for 30 min in 5 µl of 35 % H2O/60 % acetonitrile/5 % HCOOH.
MALDI-MS analysis.
Mass measurements were performed on a BIFLEX III MALDI-TOF (Bruker Daltonics) equipped with the SCOUT High Resolution Optics with X-Y multisample probe and griddle reflectors. The instrument was run in positive ion reflector mode at a maximum accelerating potential of 19 kV. A saturated solution of 
-cyano-4-hydroxycinnamic acid (Sigma) in acetone was used as a matrix. Spreading and fast evaporation of 0.5 µl matrix solution enabled the formation of a fine layer of crystals, on which a droplet of 0.5 µl aqueous HCOOH (5 %) solution was mixed first with 0.5 µl peptide-containing digest, and then with 0.3 µl saturated matrix solution (in 50 % H2O/50 % acetonitrile). The preparation was vacuum-dried and washed once with 0.7 µl aqueous HCOOH (5 %). Mass spectra were internally calibrated with trypsin autolysis peaks (m/z 842.510 and m/z 2211.105). Monoisotopic peptide masses were assigned by the XTof v5.1.5 software (Bruker) and the peak list transferred through the MS BioTools program (Bruker Daltonics) as input to search against the protein database (NCBInr or Swiss-Prot) using MASCOT v1.9 software (Matrix Science). To improve accuracy, the results were verified by searching against both the NCBInr (v 20070111) and the Swiss-Prot database (v 51.4). Almost all proteins were found in both databases; one protein, GuaB, was identified using the MSDB database (v 20060831) from two independent pickings. Tryptic mass searches retained only data with up to one missed tryptic cleavage (cuts on the C-terminal side of KR unless next residue is P) and optional methionine oxidation or cysteine carbamidomethylation, with mass accuracy limited to 100 p.p.m.
Transcription analysis.
The RNA extraction and semiquantitative RT-PCR protocols have been described previously (Nicoloff et al., 2004). RT-PCR was performed using the Invitrogen Superscript one-step RT-PCR with platinum Taq kit, according to the manufacturer's instructions, with primers listed in Table 1. Optimized amounts of total RNA were used to target different transcripts (200 ng for argH and argG and 20 ng for rrn). For Northern hybridizations, DNA probes were amplified by PCR (95 °C for 1 min, followed by 35 three-step cycles of 94 °C for 40 s, 50 °C for 40 s and 72 °C for 2 min, with a final extension phase at 72 °C for 10 min), using primers listed in Table 1. The PCR products were DIG-labelled using the DIG-labelling kit (Roche Diagnostics). For each probe, the optimal amount of total RNA to be used was previously tested [0.1 µg for rrn; 1 µg for upp and adk; 2 µg for carA, argG and argH; 5 µg for pyrAb1, cah, and argC; 10 µg for guaB (Arsène-Ploetze et al., 2006a) or data not shown]. Slot-blot hybridizations were performed and quantified as described by Nicoloff et al. (2005), with Quantity One software (Bio-Rad, v 4.5.2). The background level was calculated and subtracted (local background subtraction option). To calculate the relative signal for each gene, the measured signal for each probe was divided by the signal obtained with the rrn probe. Under the tested conditions, the constitutive expression levels of rrn genes were verified using the upp gene, whose transcription is not regulated by Ci levels (Fig. 2) (Arsène-Ploetze et al., 2006a).
|
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2004). Thus, the transcriptional regulation of the first genes of these operons, carA, argC and argG, as well as of the cah and upp genes, was analysed in response to Ci concentration, and compared with the transcriptional pattern of reference genes. The rrn genes and the pyrAb1 gene were used as characterized Ci-non-responsive and -responsive genes, respectively (Arsène-Ploetze et al., 2006a). The slot-blot hybridizations with probes designed to reveal these genes were performed with RNA extracts from cells cultivated at high or low Ci pool (Fig. 2
). As previously shown, the pyrAb1 gene was regulated in response to Ci concentration whereas the rrn gene was not (Arsène-Ploetze et al., 2006a). No significant change in expression (more than 20 % difference between low and high Ci pool, corresponding to the standard deviation obtained) was observed for carA, argC, argG, upp (Fig. 2a) and cah genes (data not shown). This suggests that transcription of these genes is not significantly modulated in response to Ci availability. However, these observations did not exclude a post-transcriptional regulation of these genes in response to Ci. Therefore, a DIGE proteomic approach was used to check if the proteins encoded by the car and arg, upp and cah genes accumulated differentially in response to Ci and, more generally, to identify new genes that may be up- or downregulated in response to Ci availability.
Twenty-five proteins are up- or downregulated when CO2 levels are increased
The accumulation pattern of 37 spots changed significantly when Ci concentrations changed (ratio >1.2 or <–1.2, t-test >5). Positive ratio means that the protein was more abundant at high Ci (4 % CO2-enriched air) than at low Ci (ordinary air) concentrations. A negative ratio means that the protein was less abundant at high than at low Ci levels (see Methods). From the 37 selected spots, 21 proteins were identified (Supplementary Fig. S1, available with the online version of this paper, and Table 2). Inorganic carbon-regulated proteins were designated Icr proteins. The PyrB, C, D, E, R1 proteins were identified (corresponding to 15 spots) and shared the same pattern of differences in accumulation as previously shown using a proteomic approach based on Coomassie staining (Arsène-Ploetze et al., 2006a); their amounts decreased when Ci concentration increased (Supplementary Fig. S1 and Table 2). In addition to these pyr-encoded proteins, when the Ci pool was high, 13 other proteins shared a lower relative abundance (ratio <0) and three proteins shared a higher relative abundance (ratio >0; Lp_1058, Lp_1188 and Lp_2367) (Table 2). Thus, 16 new proteins were identified in this DIGE study with different abundance in response to Ci availability. Together with nine previously identified pyr-encoded proteins (Arsène-Ploetze et al., 2006a), at least 16 other proteins show Ci concentration-dependent accumulation. According to the amplitude of the differences observed, Icr proteins were classified into two groups: class I included eight proteins with expression level ratio 
2, and class II clustered 13 proteins with significant (according to the t-test obtained) but lower ratios of less than 2 (Table 2).
|
). Among the Icr proteins, one enzyme involved in arginine biosynthesis, the argininosuccinate lyase encoded by argH, and two proteins involved in nucleotide metabolism, Adk and GuaB, were identified (Fig. 1).
Genomic exploration of the genes encoding Icr proteins
The genetic organization of the identified Icr protein-encoding genes was examined, as well as their possible co-transcription with adjacent genes. The 21 Icr-protein-encoding genes are grouped in 18 loci (Fig. 3) and included the previously characterized eight pyr genes of the de novo pyrimidine nucleotide synthesis pathway (Nicoloff et al., 2005) grouped in the polycistronic pyrR1, B, C, Aa1, Ab1, D, F, E transcript, whereas pyrP was localized in another locus. Among other Icr-protein-encoding genes, asp2, atpH, rfbC and dak2 are located within predicted operons encoding proteins involved in alkaline shock response, ATP synthesis-coupled proton transport (Bron et al., 2006), cell wall synthesis and an uncharacterized glycerol metabolism cluster, respectively. Two putative operons involved in glutamine import contained one Icr-protein-encoding gene. glnQ1 and glnQ3 are located at the end of predicted operons with glnPH1 and glnPH2, coding respectively for the permease protein and the substrate-binding protein of glutamine ABC transport systems. The genetic organization of lp_2340 suggests that this gene may be co-transcribed with lp_2339. The genetic organization of the Icr-protein-encoding genes guaA, guaB, adk, lp_0244, argS, hsp3, gshR4 and npr2 did not suggest any polycistronic structure with adjacent genes.
|
). Two hypotheses could explain these results: first, argG and argH are not co-transcribed; second, argH expression is regulated at the level of translation or protein stability. To test these hypotheses, co-transcription of argG and argH was assessed at low or high Ci using RT-PCR (several sets of primers complementary to the argG or argH genes were compared with a negative control; see amplifications a, b, c and d in Fig. 4). The experiments clearly showed that argG and argH genes form an operon. However, only ArgH was characterized as an Icr protein. Therefore, to test whether the argH gene transcription was Ci-responsive, slot-dot hybridization were performed at low or high Ci levels. Moreover, since the proteins Adk and GuaB are metabolically related to nucleotide and arginine metabolism (Fig. 1), analysis of guaB and adk transcription was included. The transcription of the argH and guaB genes changed significantly in response to higher Ci levels (ratio <0, Fig. 2). These expression patterns were concurrent with the observation that the two corresponding proteins were less abundant in cells grown with high Ci. Moreover, the absolute values of the ratio obtained with this transcriptional analysis (–2 and –1.5, respectively; Fig. 2b) were similar to those calculated with the DIGE experiment (–2.3 and –1.6, respectively; Table 2). These results suggest that guaB and argH regulation in response to Ci pools occurs principally at the transcriptional level or through mRNA stability, as previously observed for the pyr genes (Arsène-Ploetze et al., 2006a). Concerning the adk gene, only a slight difference between low and high Ci was measured; this difference was not significant (Fig. 2b).
|
) is similar to that observed at the protein level (Table 2).
PyrR2 is not involved in Ci-mediated regulation of argH or guaB expression
Regulation of pyr gene expression in response to Ci availability involves the transcriptional regulator PyrR2 (Arsène-Ploetze et al., 2006a). Therefore, the role of PyrR2 in the regulation of argH and guaB was analysed. These two genes show a similar pattern of regulation to that of the pyr genes in the wild-type background of strains CCM 1904 and NC8 (compare Fig. 2a and c). A pyrR2 mutant was available in the genetic background of strain NC8 and not of strain CCM 1904 (Arsène-Ploetze et al., 2006a). As the same regulation pattern with respect to Ci and pyrimidine availability was observed between strains NC8 and CCM 1904, the role of PyrR2 was tested only in strain NC8. The Northern blot experiments were performed using mRNA extracted from wild-type (NC8) or the corresponding pyrR2 strain FB422, cultivated with low or high Ci concentration. The guaB expression was slightly reduced in the pyrR2 mutant, when the Ci level was low, as compared to the wild-type. However, for both genes, the Ci response was still observed (compare low and high Ci in Fig. 2c: ratio 3.4 for argH and 2.1 for guaB). These results suggest that PyrR2 is not the major regulator involved in the regulation of argH and guaB transcription in response to Ci concentration, and that pyr, argH and guaB are part of the same Ci modulon, but belong to different regulons.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
), suggests that expression of 25 proteins is regulated in response to Ci availability in L. plantarum and that the regulation of 11 of these proteins occurs at the transcriptional or mRNA stability level (summarized in Fig. 3). However, this number is not complete since some proteins might have been missed. For example, the PyrAb1 and PyrAa1 polypeptides were not detected in the DIGE experiments. PyrAa1, in contrast to PyrAb1, should be found in a region of the gel with good resolution (pI 5.7, mol. mass 40.1 kDa; and pI 4.9; mol. mass 116 kDa, respectively) (Arsène-Ploetze et al., 2006a). PyrAa1 contains lysine residues; therefore the absence of this protein is not due to a labelling defect. Among other known Icr proteins encoded by pyr genes, PyrP and PyrF (Arsène-Ploetze et al., 2006a) were not detected. PyrP is a membrane protein not found in the cytoplasmic protein extracts. PyrF, with a pI of 7.7, would not have been detected using gels that targeted proteins with pI between 4 and 7. Some of the genes encoding Icr proteins are probably expressed in operons with other genes. Unlike the pyr genes, none of the proteins encoded by atpH, dak2 or rfbC co-transcribed genes was found among the Icr proteins. These proteins might have been missed in our study or their abundance might not be Ci-regulated. On the other hand, the differential accumulation observed for AtpH, Dak2 and RfbC proteins may result from post-translational regulatory mechanism.
PyrR2-dependent and independent mechanisms regulate Ci-responsive genes
PyrR2 is involved in pyr gene expression, but it was shown in this study that PyrR2 is not required in Ci-mediated regulation of argH or guaB expression. In other bacteria, several regulatory mechanisms control their Ci response. For instance, among the cyanobacteria, some, but not all of the genes of the Ci stimulon (McGinn et al., 2003; Wang et al., 2004) are regulated by CcmR (previously called NdhR), a LysR family regulator. In Gram-positive bacteria, regulators of the LysR family (Koehler, 2002) and regulator families such as the two-component signal-transducing systems (Federle et al., 1999; McIver & Myles, 2002) have been described. Thus, the Ci response regulation in L. plantarum may also involve several families of regulators, as found with other bacteria.
In this work, argH was co-transcribed with the upstream argG gene (Fig. 4). However, these two genes are differentially regulated in response to Ci. Regulation of transcription (initiation or termination) or RNA stability may occur only for the downstream gene argH in response to Ci abundance. The downstream gene argH may be expressed with argG as a bicistronic mRNA, but also from its own promoter. This hypothesis would explain why these two genes were differentially regulated in response to Ci.
What could be the transcription regulatory mechanism in response to Ci availability? The pyr genes are regulated via long non-coding mRNA (5'-UTR), which form a secondary structure (Nicoloff et al., 2005). guaA and guaB are regulated in response to guanine availability in Bacillus subtilis via long 5'-UTR and potentially forming secondary structures (Barrick et al., 2004; Mandal et al., 2003; Wels et al., 2006). Thus, besides the pyr genes, the involvement of 5'UTRs of mRNA of Icr-protein-encoding genes is possible; however, this needs to be experimentally shown.
Changing Ci availability affects several metabolic pathways and adaptation processes
The predicted functions of the Icr proteins fell into ten major categories (Fig. 3), suggesting that the response largely affects L. plantarum's general metabolism. It has previously been demonstrated that optimal growth is observed when Ci level is increased (4 %) (Arsène-Ploetze & Bringel, 2004). Therefore, the potential role of the Icr proteins was analysed to ascertain whether their accumulation pattern correlated with the growth rate of L. plantarum. When growth rate was optimal (high Ci levels), proteins involved in cell wall biosynthesis (RfbC) in energy metabolism or ATP homeostasis (Lp_2367 atpH and Lp_1058 adk) were more abundant. In contrast, the abundance of the proteins involved in UMP and GMP synthesis (pyr genes, guaA and guaB), in glycerol metabolism (Dak2), in glutamine import (GlnQ1 and GlnQ3) and arginine synthesis or incorporation (ArgS or ArgH) decreased when growth rate was optimal. A recent proteomic study of L. plantarum has identified regulated proteins in stationary phase as compared to the exponential phase (Cohen et al., 2006). Among genes encoding Icr proteins, ten were found to be upregulated in stationary phase. These proteins included a putative oxidoreductase Lp_0244; a universal stress protein Lp_2340; enzymes involved in pyrimidine and purine biosynthesis Lp_2697 (pyrE), Lp_2702 (pyrC), Lp_2703 (pyrB) and Lp_0914 (guaA); an alkaline shock protein (asp2); Lp_3352 (hsp3); Lp_1391 (argS); and Lp_3267 (gshR4). The latter protein, involved in glutathione homeostasis, was also more abundant when Ci was low (GshR4). Glutathione is involved in different cellular processes including adaptation to pH variation stress.
Most Icr proteins were characterized by a ratio of less than 2 (13 class II proteins, Table 2). The change of accumulation pattern of proteins may not be strong enough to postulate that Ci availability has a direct effect. Subtle switches in the metabolic pathways should occur when the Ci levels changed, and may be correlated with a general growth-rate-dependent effect or stress responses such as intracellular pH (pHi) variations. For example, a drop in the internal pH (pHi) at increased Ci levels could account for the higher relative abundance of Icr20, encoded by atpH. Previous studies have shown that the atp operon is induced at low pH ranges in Lactobacillus acidophilus (Kullen & Klaenhammer, 1999). During CO2-enriched cultures, CO2 accumulation may lead to unbalanced decarboxylation reactions. In some lactic acid bacteria, decarboxylation reactions are essential for pH homeostasis since they help to increase the pHi, a crucial process for these acidotolerant bacteria (Konings et al., 1997). In the present work, the external pH did not change significantly at the end of growth at increased Ci concentration (less than 0.05 pH units, data not shown), but the internal pH may change. We hypothesize that at the tested high Ci concentrations, the efficiency of decarboxylation reactions may be reduced, leading to decreased pHi and AtpH slight accumulation (as observed, Table 2). Several Icr proteins would be involved in adaptation responses to stimuli such as pHi variations or growth phase. These observations suggest that when growth rate was not optimal (low Ci), L. plantarum is subject to stress, and confirm the hypothesis that this bacterium is capnophilic.
Involvement of Icr proteins in the CO2 requirements and adaptation to CO2-enriched environments of L. plantarum
The molecular basis that sustains the CO2-dependent growth phenotype has been studied in isogenic mutants of strain L. plantarum CCM 1904 with impaired CP synthesis, increased CP consumption or increased CP requirements. In these mutants, low CP pools may drive the HCR phenotype by limiting the arginine and nucleotide supplies, as recently reviewed (Bringel et al., 2008). However, our understanding of the Ci-dependent growth phenotype is incomplete, as the reason why some mutants have lost the ability to grow in normal air without CO2 enrichment and with an arginine-deregulated phenotype remains unknown (Nicoloff et al., 2004). The ArgH enzyme that catalyses the production of arginine from argininosuccinate is a Ci-responsive gene. Why only the downstream gene of the argGH operon would be required in L. plantarum adaptation to low Ci levels is not clear. The HCR mutants that were previously isolated may harbour an impaired Icr-protein-encoding gene, possibly argH or a regulator involved in argH transcription or mRNA stability control. These mutants may be impaired in genes that would directly or indirectly modulate arginine synthesis in response to Ci availability. However, such mutants may be impaired in other genes, with no direct link to arginine biosynthesis.
In natural conditions, heterotrophic micro-organisms are exposed to fluctuating Ci levels, for example in the gastro-intestinal tract or fermenting matter (Maligoy et al., 2008), to which they respond by adjusting their metabolism to outcompete neighbouring microflora. One-third of L. plantarum-related strains isolated from natural environments are conditional Ci-dependent prototrophs (Bringel & Hubert, 2003), but the molecular basis of this CO2 requirement has only been partially explained (Bringel et al., 2008). Further insights into the Ci-responsive pathways are needed to evaluate the ecological physiology of heterotrophs in biotopes with fluctuating Ci levels.
|
ACKNOWLEDGEMENTS |
|---|
Edited by: M. Kleerebezem
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Arsène-Ploetze, F., Kugler, V., Martinussen, J. & Bringel, F. (2006a). The expression of the pyr operon of Lactobacillus plantarum is regulated by inorganic carbon availability through a second regulator PyrR2 homologous to the pyrimidine dependent regulator PyrR1. J Bacteriol 188, 8607–8616.
Arsène-Ploetze, F., Nicoloff, H., Kammerer, B., Martinussen, J. & Bringel, F. (2006b). Uracil salvage pathway in Lactobacillus plantarum: transcription and genetic studies. J Bacteriol 188, 4777–4786.
Barrick, J. E., Corbino, K. A., Winkler, W. C., Nahvi, A., Mandal, M., Collins, J., Lee, M., Roth, A., Sudarsan, N. & other authors (2004). New RNA motifs suggest an expanded scope for riboswitches in bacterial genetic control. Proc Natl Acad Sci U S A 101, 6421–6426.
Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72, 248–254.[CrossRef][Medline]
Bringel, F. & Hubert, J. C. (2003). Extent of genetic lesions of the arginine and pyrimidine biosynthetic pathways in Lactobacillus plantarum, L. paraplantarum, L. pentosus and L. casei: prevalence of CO2 dependent auxotrophs and characterization of deficient arg genes in L. plantarum. Appl Environ Microbiol 69, 2674–2683.
Bringel, F., Frey, L., Boivin, S. & Hubert, J. C. (1997). Arginine biosynthesis and regulation in Lactobacillus plantarum: the carA gene and the argCJBDF cluster are divergently transcribed. J Bacteriol 179, 2697–2706.
Bringel, F., Vuilleumier, S. & Arsène-Ploetze, F. (2008). Low carbamoyl phosphate pools may drive Lactobacillus plantarum CO2-dependent growth phenotype. J Mol Microbiol Biotechnol 14, 22–30.[CrossRef][Medline]
Bron, P. A., Molenaar, D., de Vos, W. M. & Kleerebezem, M. (2006). DNA micro-array-based identification of bile-responsive genes in Lactobacillus plantarum. J Appl Microbiol 100, 728–738.[CrossRef][Medline]
Cohen, D. P., Renes, J., Bouwman, F. G., Zoetendal, E. G., Mariman, E., de Vos, W. M. & Vaughan, E. E. (2006). Proteomic analysis of log to stationary growth phase Lactobacillus plantarum cells and a 2-DE database. Proteomics 6, 6485–6493.[CrossRef][Medline]
Drysdale, M., Bourgogne, A. & Koehler, T. M. (2005). Transcriptional analysis of the Bacillus anthracis capsule regulators. J Bacteriol 187, 5108–5114.
Federle, M. J., McIver, K. S. & Scott, J. R. (1999). A response regulator that represses transcription of several virulence operons in the group A Streptococcus. J Bacteriol 181, 3649–3657.
Herbert, S., Newell, S. W., Lee, C., Wieland, K. P., Dassy, B., Fournier, J. M., Wolz, C. & Doring, G. (2001). Regulation of Staphylococcus aureus type 5 and type 8 capsular polysaccharides by CO2. J Bacteriol 183, 4609–4613.
Hyde, J. A., Trzeciakowski, J. P. & Skare, J. T. (2007). Borrelia burgdorferi alters its gene expression and antigenic profile in response to CO2 levels. J Bacteriol 189, 437–445.
Kawasaki, S., Nagasaku, M., Mimura, T., Katashima, H., Ijyuin, S., Satoh, T. & Niimura, Y. (2007). Effect of CO2 on colony development by Bifidobacterium species. Appl Environ Microbiol 73, 7796–7798.
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.
Koehler, T. M. (2002). Bacillus anthracis genetics and virulence gene regulation. Curr Top Microbiol Immunol 271, 143–164.[Medline]
Konings, W. N., Lolkema, J. S., Bolhuis, H., van Veen, H. W., Poolman, B. & Driessen, A. J. (1997). The role of transport processes in survival of lactic acid bacteria. Energy transduction and multidrug resistance. Antonie Van Leeuwenhoek 71, 117–128.[CrossRef][Medline]
Kullen, M. J. & Klaenhammer, T. R. (1999). Identification of the pH-inducible, proton-translocating F1F0-ATPase (atpBEFHAGDC) operon of Lactobacillus acidophilus by differential display: gene structure, cloning and characterization. Mol Microbiol 33, 1152–1161.[CrossRef][Medline]
Lee, J. W., Lee, S. Y., Song, H. & Yoo, J. S. (2006). The proteome of Mannheimia succiniciproducens, a capnophilic rumen bacterium. Proteomics 6, 3550–3566.[CrossRef][Medline]
Maligoy, M., Mercade, M., Cocaign-Bousquet, M. & Loubiere, P. (2008). Transcriptome analysis of Lactococcus lactis in coculture with Saccharomyces cerevisiae. Appl Environ Microbiol 74, 485–494.
Mandal, M., Boese, B., Barrick, J. E., Winkler, W. C. & Breaker, R. R. (2003). Riboswitches control fundamental biochemical pathways in Bacillus subtilis and other bacteria. Cell 113, 577–586.[CrossRef][Medline]
McGinn, P. J., Price, G. D., Maleszka, R. & Badger, M. R. (2003). Inorganic carbon limitation and light control the expression of transcripts related to the CO2-concentrating mechanism in the cyanobacterium Synechocystis sp. strain PCC6803. Plant Physiol 132, 218–229.
McIver, K. S. & Myles, R. L. (2002). Two DNA-binding domains of Mga are required for virulence gene activation in the group A Streptococcus. Mol Microbiol 43, 1591–1601.[CrossRef][Medline]
Nicoloff, H., Hubert, J. C. & Bringel, F. (2000). In Lactobacillus plantarum, carbamoyl phosphate is synthesized by two carbamoyl-phosphate synthetases (CPS): carbon dioxide differentiates the arginine-repressed from the pyrimidine-regulated CPS. J Bacteriol 182, 3416–3422.
Nicoloff, H., Arsène-Ploetze, F., Malandain, C., Kleerebezem, M. & Bringel, F. (2004). Two arginine repressors regulate arginine biosynthesis in Lactobacillus plantarum. J Bacteriol 186, 6059–6069.
Nicoloff, H., Elagöz, A., Arsène-Ploetze, F., Kammerer, B., Martinussen, J. & Bringel, F. (2005). Repression of the pyr operon in Lactobacillus plantarum prevents its ability to grow at low carbon dioxide levels. J Bacteriol 187, 2093–2104.
Price, G. D., Badger, M. R., Woodger, F. J. & Long, B. M. (2008). Advances in understanding the cyanobacterial CO2-concentrating-mechanism (CCM): functional components, Ci transporters, diversity, genetic regulation and prospects for engineering into plants. J Exp Bot 59, 1441–1461.
Soltes-Rak, E., Mulligan, M. E. & Coleman, J. R. (1997). Identification and characterization of a gene encoding a vertebrate-type carbonic anhydrase in cyanobacteria. J Bacteriol 179, 769–774.
Stretton, S., Marshall, K. C., Dawes, I. W. & Goodman, A. E. (1996). Characterisation of carbon dioxide-inducible genes of the marine bacterium, Pseudomonas sp. S91. FEMS Microbiol Lett 140, 37–42.[CrossRef][Medline]
Takayama, M., Ohyama, T., Igarashi, K. & Kobayashi, H. (1994). Escherichia coli cad operon functions as a supplier of carbon dioxide. Mol Microbiol 11, 913–918.[Medline]
Wang, H. L., Postier, B. L. & Burnap, R. L. (2004). Alterations in global patterns of gene expression in Synechocystis sp. PCC 6803 in response to inorganic carbon limitation and the inactivation of ndhR, a LysR family regulator. J Biol Chem 279, 5739–5751.
Wels, M., Francke, C., Kerkhoven, R., Kleerebezem, M. & Siezen, R. J. (2006). Predicting cis-acting elements of Lactobacillus plantarum by comparative genomics with different taxonomic subgroups. Nucleic Acids Res 34, 1947–1958.
Woodger, F. J., Bryant, D. A. & Price, G. D. (2007). Transcriptional regulation of the CO2-concentrating mechanism in a euryhaline, coastal marine cyanobacterium, Synechococcus sp. Strain PCC 7002: role of NdhR/CcmR. J Bacteriol 189, 3335–3347.
Received 6 March 2008;
revised 28 May 2008;
accepted 10 June 2008.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
Function of the Icr proteins (genes in black). 
Class I and II cluster Icr proteins with ratio higher or lower than 2, respectively; Ci concentration leading to maximal expression. 
Mechanism of regulation demonstrated or suggested in previous studies, with an indication of the level of regulation (transcription or translation).