TEXT

Top Abstract Text References

-Amylase, the predominant salivary enzyme in humans (1), catalyzes the hydrolysis of -1,4-glucosidic linkages in starch and binds to oral streptococci, a predominant group comprising the oral microflora (9, 10, 25, 28, 30). Amylase binds Streptococcus gordonii through high-affinity protein receptors that cluster around surface cell division sites (31). Bound amylase retains enzymatic activity (9, 29) and may hydrolyze dietary starch to provide fermentable carbohydrates to the bacteria. This hypothesis is supported by growth studies that found that only organisms expressing the amylase-binding phenotype are able to grow in starch-containing medium and only after preincubation of the cells with salivary amylase (9, 29; J. D. Rogers, R. J. Palmer, Jr., P. E. Kolanbrander, and F. A. Scannapieco, submitted for publication).

A 20-kDa amylase-binding protein (AbpA) mediates the binding of amylase to S. gordonii (9, 25, 31). An abpA mutant of S. gordonii did not grow in starch-containing medium despite preincubation with amylase. While AbpA is found in abundance in spent brain heart infusion (BHI) broth, which contains 0.2% glucose, it is absent in supernatants of cultures grown to mid-log phase in defined medium containing 1% glucose. These results suggested that carbohydrates may influence AbpA expression.

Carbon catabolite repression (CCR) is a regulatory mechanism that allows bacteria to use a set of proteins to metabolize a specific carbohydrate source while down-regulating proteins involved in the utilization of other carbohydrates. CCR in gram-negative bacteria is a positive regulatory mechanism that is mediated by cyclic AMP (cAMP)-dependent and -independent mechanisms (5). In cAMP-dependent regulation, the bacterial phosphoenolpyruvate:sugar phosphotransferase system (PTS) promotes a sequence of phosphoryl transfer events that activates transcription of CCR-sensitive operons. cAMP-independent regulation is mediated by a catabolite repressor and activator protein (Cra) that represses proteins involved in sugar metabolism and activates those involved in substrate oxidation. Gram-positive bacteria such as Bacillus and Streptococcus lack detectable levels of cAMP (5) and rely on cAMP-independent mechanisms of catabolite repression (23, 26, 33).

In Bacillus subtilis, catabolite control protein A (CcpA), a DNA-binding protein (14, 18), is thought to negatively regulate transcription by binding to a catabolite-responsive element (CRE) in the promoter region or within the transcribed gene (6, 16). CcpA binding is modulated by the phosphorylation state of HPr, the heat-stable phosphoryl carrier component of the PTS (24). Genes involved in utilization of starch (14), histidine (36), acetate and acetoin (12), gluconate (11), and xylose (19) are all repressed by CcpA. A second protein, CcpB, whose expression depends on growth conditions (7), is involved in catabolite repression of the gluconate and xylose utilization genes. Finally, CcpC regulates the expression of the citB (aconitase) and citZ (citrate synthase) genes (17).

Diauxic growth has been demonstrated in the oral streptococci, and the PTS has a regulatory role in Streptococcus mutans sugar metabolism (8, 20). Recently, RegM has been described as a CcpA homolog in S. mutans. Interestingly, the activity of RegM does not appear to conform to the model of CCR described for B. subtilis (35). For instance, disruption of regM does not affect diauxic growth of S. mutans in a number of sugars in the presence of glucose, and increased glucose repression was noted for -galactosidase, mannitol-1-P dehydrogenase, and P--galactosidase activities in the regM mutant (32).

To investigate the role of carbon catabolite repression on the expression of abpA, culture supernatant levels of AbpA were compared after growth with different carbohydrate sources. Also, a ccpA homolog designated regG was identified in S. gordonii, and the effects of insertional inactivation of this gene on the expression of abpA were determined.

Bacterial strains, plasmids, and culture conditions. The bacterial strains and plasmids used in this study are listed in Table 1. Streptococci were routinely cultured in a defined medium (FMC) (34) or in BHI for various time periods at 37°C without shaking in a candle jar. Broth media were supplemented to 1% with glucose, maltose, sucrose, lactose, or maltooligosaccharides. Escherichia coli strains were grown under aerobic conditions with shaking for 12 to 16 h at 37°C in Luria-Bertani (LB) broth and maintained on LB agar. Strains containing recombinant clones were plated on LB agar containing erythromycin (300 µg/ml).

DNA and RNA manipulations. Standard procedures were used for plasmid extraction from E. coli (3). DNA was prepared from S. gordonii as previously described (25). Total RNA was isolated from S. gordonii cells grown to the mid-logarithmic phase resuspended in 300 µl of diethyl pyrocarbonate-treated distilled H₂O (followed by 900 µl of Trizol reagent (Gibco-BRL) using the FastRNA Blue system (Bio 101, Inc, Vista, Calif.). RNA concentration and purity was determined using standard methods (27).

Influence of carbohydrate source on abpA expression. S. gordonii Challis was cultured to mid-log phase (~8 to 10 h) in FMC supplemented with various sugars. AbpA was detected using a solid-phase amylase ligand-binding assay (9, 13). Relative concentrations of bands were quantitated using a GS 300 scanning densitometer (Hoefer). AbpA was nearly undetectable in the supernatants of bacteria grown to mid-log phase in FMC supplemented with glucose, sucrose, maltose (Fig. 1A, lanes 2, 4, and 5 respectively), or lactose. AbpA was detected when the cells were cultured with maltooligosaccharides (Fig. 1A, lane 6). Growth in unsupplemented BHI resulted in the recovery of ~50-fold-greater amounts of AbpA than growth in BHI supplemented with 1% glucose. Northern blots probed with biotinylated abpA also demonstrated a large decrease in the abpA transcript when cells were grown in glucose-supplemented BHI (Fig. 1B).

View larger version (60K): [in this window] [in a new window]   FIG. 1.   (A) Influence of carbohydrate source on AbpA expression. A solid-phase amylase ligand-binding assay was carried out with culture supernatants (100 µg per lane) of S. gordonii Challis grown in defined medium with 1% concentrations of the following sugars: glucose (lane 2), lactose (lane 4), maltose (lane 5), and maltooligosaccharides (lane 6). Lane 1 contains molecular mass standards, and lane 3 is a positive control (BHI with no added sugar). The blot was first probed with purified amylase followed by rabbit antiamylase. Bound antibodies were visualized using goat-anti rabbit immunoglobulin G conjugated to an alkaline phosphatase reporter. (B) Northern blot of cells grown in BHI supplemented with 1% glucose (lane 1) or BHI with no added sugar (positive control) (lane 2). The Northern blot was probed with biotinylated abpA.

Time course of abpA transcription. BHI (10 ml) was inoculated with 100 µl of an overnight BHI culture of S. gordonii Challis. Cells were incubated (to an optical density of 0.5 at 600 nm), and 500 µl was then removed. Glucose was then added to the cultures to a final concentration of 1%, and 500-µl aliquots were removed at 2, 5, 10, and 15 min. Total RNA was isolated from each sample and analyzed. Biotinylated abpA was used to probe Northern blots (27). Cultures that did not have glucose added were monitored in parallel to control for potential mRNA degradation during processing.

View larger version (112K): [in this window] [in a new window]   FIG. 2.   Time course study of abpA mRNA transcript levels. (A) Total RNA (5 µg) was isolated from S. gordonii at mid-logarithmic growth (lane 1) and at 2 min (lane 2), 5 min (lane 3), 10 min (lane 4), and 15 min (lane 5) after the addition of 1% glucose to the culture. (B) Total RNA from parallel cultures grown without additional glucose is included for comparison. Biotinylated abpA was used as the probe.

Analysis of the S. gordonii abpA gene. cis-acting CREs are nucleotide sequences that mediate CCR for several genes in a variety of bacteria (33). A consensus CRE sequence for the binding of catabolite control proteins was identified by comparing the binding sites for genes known to be regulated by CCR (16). A potential major CRE (CRE 1) 153 bp downstream of the translational start site of the abpA gene was identified (AGTAACCGAATACTT). This is not unusual, since CREs show variability of their positions relative to the transcriptional and translational start sites of regulated genes (15). The abpA CRE shows conservation of bases at 13 of 14 positions with the derived consensus sequence. An additional CRE (CRE 2) at position +109 relative to the translational start site was also identified (CGACGGCGCATACT). The nucleotide sequence of this cre showed conservation at 10 of 14 bases.

Cloning and sequencing of an S. gordonii ccpA homolog. S. gordonii genomic DNA was used as the template for PCRs using primers based on regions conserved between a hypothetical ccpA homolog from S. mutans (GenBank accession no. AF001316) and a potential ccpA homolog identified from the unfinished Streptococcus pneumoniae genome database (The Institute for Genomic Research website at http://www.tigr.org). The primers ccpA1 (5'-GGAAACAATATGAACACAGACG-3') and ccpA2 (5'-TTGACGTGTCGTACCACGC-3') were designed to yield a full-length amplicon of an S. gordonii ccpA homolog. The PCR consisted of a 3-min denaturation at 95°C followed by 35 cycles of 95°C for 1 min, 54°C for 1 min, and 72°C for 1 min. The resulting 1-kb product was cloned into the pGEM-T (Promega) vector, and the nucleotide sequences of both the sense and antisense strands were determined. A search of the incomplete S. gordonii nucleotide sequence database (The Institute for Genomic Research website at http://www.tigr.org) with this sequence revealed a contig containing the 5' portion of the cloned sequence along with upstream nucleotide sequence data.

View larger version (74K): [in this window] [in a new window]   FIG. 3.   Sequence analysis of the regG locus. The nucleotide and deduced amino acid sequences are given for regG. The ribosome binding site (RBS) and the 10 and 35 promoter elements for regG are indicated, as is the NdeI restriction site used for insertion of the ermAM gene and inactivation of regG. The nucleotides corresponding to the pepQ partial open reading frame are also indicated, with the putative translational start site underlined.

View larger version (105K): [in this window] [in a new window]   FIG. 4.   Transcriptional regulation of abpA by RegG. Total RNA was extracted from S. gordonii Challis (lanes 1 and 2) and Challis-CE1 (lanes 3 and 4) during exponential growth in BHI (lanes 1 and 3) or in BHI supplemented with 1% glucose (lanes 2 and 4). As a reference, sequencing reactions using the same primer (PE1) were performed (left).

Inactivation of regG. The ermAM gene from pTS19E (37) was restricted with PvuII, the 2-kb ermAM fragment gel purified on 1% agarose, and the blunt-ended fragment was cloned into the NdeI site of the regG gene in pCRegG. The resulting plasmid was designated pCRegG-E1. A BsaI/ScaI double digest of pCR2KB-E1 was used to cleave a 415-bp portion of the ampicillin gene from the plasmid. The 6.6-kb fragment was purified on 0.8% agarose gels, excised, religated, and designated pCRegG-E2. This vector was constructed to avoid introduction of ampicillin resistance into S. gordonii. pCRegG-E2 was transformed into E. coli DH5 and transformants were initially selected on erythromycin (300 µg/ml) containing LB agar plates. pCRegG-E2, which contained the ermAM gene inserted 200 bp downstream of the translational start site of regG, was linearized by restriction with AspI and gel purified prior to transformation into S. gordonii Challis for 2 h with shaking (120 rpm) at 37°C; aliquots were then plated on BHI agar supplemented with erythromycin (5 µg/ml) and incubated at 37°C overnight in a candle jar (2).

Diauxic growth of S. gordonii Challis-CE1. Growth of wild-type S. gordonii Challis on sugars in the presence of glucose was compared to that of the regG mutant (data not shown). Cells from overnight cultures in defined medium containing 0.5% glucose were harvested, washed twice in sugar-free medium, and added to fresh medium containing either glucose (6 mM) and/or a secondary carbohydrate (6 mM) to a standard optical density of 0.1 at 600 nm. Cultures were incubated at 37°C, and samples were taken every 30 min in order to monitor growth spectrophotometrically.

Transcriptional regulation of abpA. Total RNA from mid-exponential-phase cells was used throughout these studies. Primer extension products were obtained by using oligonucleotide PE1 (5'-GGTAGTATGCGCCGACG-3') corresponding to the complement of abpA nucleotide positions 102 to 118 or PE2 (5'-CAGCTACTACTGCTGG-3') corresponding to positions 205 to 220. The oligonucleotides were end labeled with [-³²P]ATP and T4 polynucleotide kinase, and the reverse transcriptase reaction was performed according to standard protocols (27). The extension products were resolved on polyacrylamide sequencing gels and visualized by autoradiography. As a reference, sequencing reactions were performed by the dideoxy-chain termination method using the same primers as in the primer extension experiments, with pCR2KB serving as the sequence template.

Discussion. CcpA, a LacI/GalR-type trans-acting regulatory protein, mediates CCR in gram-positive bacteria with a low GC content (33). A number of ccpA homologs have been described for Bacillus, Lactobacillus, and Streptococcus spp. (14, 33). RegG of S. gordonii appears to be another CcpA homolog that is closely related to RegM of S. mutans (32). In both cases, a prolidase gene is located upstream of the catabolite control protein gene, and inactivation of regG or regM does not appear to affect glucose-mediated diauxy.