The pH-Dependent Expression of the Urease Operon in Streptococcus salivarius Is Mediated by CodY

Abstract

Urease gene expression in Streptococcus salivarius 57.I, a strain of one of the major alkali producers in the mouth, is induced by acidic pH and excess amounts of carbohydrate. Expression is controlled primarily at the transcriptional level from a promoter, p_ureI. Recent sequencing analysis revealed a CodY box located 2 bases 5′ to the −35 element of p_ureI. Using continuous chemostat culture, transcription from p_ureI was shown to be repressed by CodY, and at pH 7 the repression was more pronounced than that in cells grown at pH 5.5 under both 20 and 100 mM glucose. The direct binding of CodY to p_ureI was demonstrated by electrophoretic mobility shift assay and chromatin immunoprecipitation (ChIP)–quantitative real-time PCR (qPCR). The result of ChIP-qPCR also confirmed that the regulation of CodY is indeed modulated by pH and the binding of CodY at neutral pH is further enhanced by a limited supply of glucose (20 mM). In the absence of CodY, the C-terminal domain of the RNA polymerase (RNAP) α subunit interacted with the AT tracks within the CodY box, indicating that CodY and RNAP compete for the same binding region. Such regulation could ensure optimal urease expression when the enzyme is most required, i.e., at an acidic growth pH with an excess amount of carbon nutrients.

INTRODUCTION

Microbial ureases are multisubunit, Ni²⁺-dependent enzymes that catalyze the hydrolysis of urea to produce ammonia and CO₂ (1). The production of ammonia via urea hydrolysis, in which ammonia becomes protonated at physiological pH values, increases the local pH and provides nitrogen for the microbes. Bacterial ureases are generally composed of α, β, and γ subunits, encoded by ureC, ureB, and ureA, respectively. An exception is the urease from Helicobacter pylori, which is composed of UreA and UreB, corresponding to a fusion of β and γ subunits and an α subunit of other bacterial ureases, respectively (2). The assembly of a functional urease requires accessory proteins, encoded by ureE, ureF, ureG, and ureD, for incorporation of Ni²⁺ into the metallocenter of the apo-urease (1). Some bacterial urease operons also harbor genes encoding uptake systems for urea and Ni²⁺. For instance, ureI of H. pylori encodes an H⁺-gated urea transporter (3, 4) and ureMQO of Streptococcus salivarius encode a high-affinity, nickel-specific ATP binding cassette transporter (5).

As urea is the most abundant nitrogen source in saliva and crevicular fluids, hydrolysis of urea by bacterial ureases is the principal source of alkali in the oral cavity. Alkali generation in the mouth is crucial for dental caries prevention and plaque pH homeostasis (6–8). Among all ureolytic microbes in the oral cavity, S. salivarius is the most abundant and highly ureolytic species (9) and is able to use urea as a nitrogen source (10). The urease operon in S. salivarius 57.I consists of 11 genes (ureIABCEFGDMQO) (5, 11, 12). The expression of this operon is subject to regulation by multiple environmental cues, with elevated urease expression occurring in cells grown at acidic pH values. Expression is further enhanced by an excess amount of carbohydrates and a higher growth rate (13). Such regulation allows the maximal production of urease when it is most needed by the bacteria to survive inhibition of growth or killing by low pH (10). Differential expression of ure genes is predominantly via the activity of a σ⁷⁰ promoter located 5′ of ureI (p_ureI) (12), although overall urease activity can be modulated posttranslationally by the intracellular Ni²⁺ content (5). Previous studies using a p_ureI-reporter fusion revealed that the expression of p_ureI remains sensitive to pH in the nonureolytic bacterium Streptococcus gordonii CH1, suggesting that the regulation of p_ureI is part of a conserved regulatory circuit (14).

Recent sequence analysis of the 5′ flanking region of S. salivarius p_ureI revealed one putative CodY binding consensus sequence (5′-AATTTTCWGAAAATT) (15, 16), raising the possibility that the expression of p_ureI is regulated by CodY. CodY is a global regulator highly conserved in low-G+C-content Gram-positive bacteria. The CodY regulon generally covers hundreds of genes, most of which are repressed by CodY during exponential growth and derepressed when nutrients become scarce (17, 18). In addition to regulating the expression of genes involved in metabolism, virulence, and various cellular activities, CodY activates the transcription of a small RNA (19), BsrF, in Bacillus subtilis. The DNA-binding activity of B. subtilis CodY is enhanced by interaction with branched-chain amino acids (BCAAs; e.g., isoleucine, leucine, and valine) and GTP (20, 21), whereas BCAAs are the only known effectors for CodY in Lactococcus lactis (22), Streptococcus mutans (23), and Streptococcus pneumoniae (24).

To investigate the possible involvement of CodY in the pH regulation of S. salivarius urease expression, here we examined the functions of CodY in the regulation of p_ureI in response to growth conditions. The results indicated that expression from p_ureI is repressed by CodY by direct binding to p_ureI. In the absence of CodY, the C-terminal domain of the RNA polymerase (RNAP) α subunit (α-CTD) interacts with the AT-rich sequence immediately 5′ to the promoter in p_ureI, indicating that CodY and RNAP compete for binding to the p_ureI promoter region.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

The bacterial strains and plasmids used in this study are listed in Table 1 and Table 2, respectively. S. salivarius 57.I and its derivatives were grown routinely in brain heart infusion (BHI; Difco) at 37°C in a 10% CO₂ atmosphere. When necessary, kanamycin (Km) at 1,000 μg ml⁻¹, spectinomycin (Sp) at 750 μg ml⁻¹, and erythromycin (Em) at 5 μg ml⁻¹ were added to the culture media for recombinant S. salivarius strains. Recombinant Escherichia coli strains were maintained in L broth supplemented, where indicated, with Sp at 50 μg ml⁻¹, Em at 300 μg ml⁻¹, or Km at 50 μg ml⁻¹. All chemical reagents and antibiotics were purchased from Sigma-Aldrich.

TABLE 1.

Bacterial strains used in this study

S. salivarius strain	Resistance	Description	Reference or source
57.I		Wild-type strain	9
MC308	Sp	57.I harboring spe-pureI-cat at lacZ	This study
MC308_ΔcodY	Sp, Em	MC308 codY::erm	This study
MC308_CΔcodY	Sp, Em, Km	MC308_ΔcodY harboring pDL276 codY(pSky10)	This study
MC308_ΔcodY_M1	Sp, Em	nt −52 to −44 of pureI in MC308_ΔcodY are mutated	This study
MC308_ΔcodY_M2	Sp, Em	nt −42 to −37 of pureI in MC308_ΔcodY are mutated	This study
MC308_ΔcodY_M3	Sp, Em	nt −51 to −46 and −42 to −37 of pureI in MC308_ΔcodY are mutated	This study

TABLE 2.

Plasmids used in this study^a

Plasmid	Relevant phenotype	Description	Reference or source
pDL276	Kmr	E. coli-Streptococcus shuttle vector	29
pGEM3zf(+)	Apr	General E. coli cloning vector	Promega
pMAL-c2X	Apr	Expression vector of MBP-tagged proteins	NEB
pMC195	Apr Kmr	Integration vector for S. salivarius	14
pMC286	Apr	pGEM3zf(+) harboring promoterless cat from pC194	This study
pMC300	Apr Spr	Spr integration vector for S. salivarius harboring a pureI-cat fusion	This study
pQE30	Apr	Expression vector of His-tagged proteins	Qiagen
pSF36	Apr	pGEM3zf(+) harboring codY and the flanking regions	This study
pSky2	Apr Emr	nt 73 to 234 of codY in pSF36 were replaced with a nonpolar erm	This study
pSky10	Kmr	pDL276 harboring codY	This study

^ar, resistance; MBP, maltose binding protein.

To obtain batch cultures grown under neutral or acidic pHs, cells were cultured to mid-exponential phase (optical density at 600 nm [OD₆₀₀] ≈ 0.6) in BHI containing 50 mM potassium phosphate buffer (pH 7.5) or in BHI that has been adjusted to pH 5.5 by the addition of HCl. Recombinant S. salivarius strains were grown in a Biostat Bplus bioreactor (Sartorius Stedim Biotech) at a dilution rate (D) of 0.3 h⁻¹ in medium containing 3% tryptone and 0.5% yeast extract (TY). The cultures were kept at pH 7 or pH 5.5 with 20 or 100 mM glucose as limiting or excess carbohydrate, respectively.

General genetic manipulations and qPCR.

Plasmid DNA was isolated from recombinant streptococcal strains by the method of Anderson and McKay (25). Plasmid DNA was introduced into S. salivarius 57.I and its derivatives by electroporation (12). PCR was carried out by using Phusion (Finnzymes) or Blend Taq DNA polymerase (Toyobo). Restriction endonucleases and DNA-modifying enzymes were purchased from New England BioLabs (NEB). The primers used in this study are listed in Table 3.

TABLE 3.

Primers used in this study

Primer	Sequencea	Purpose
codY_1900_AS	ATGCATGCGCCTTTGTCATATTACTTTCTCC	codY amplicon for MC308_CΔcodY construction
codY_800_S	TTGAGCTCCTGGACAAAAAGGCTTGTCC
codY_AS_PstI	TTCACTGCAGTTATTATTCGTATTCTTTCA	PCR for His-CodY construction
codY_S_BamHI	GAAGACAGGATCCATGGCAAATTTGCT
EMSA_pureI_AS	ACACATGTCAACGAATTTTCAGAAATTTTGCAACATTTAC	EMSA probe, for analyzing α-CTD and CodY binding
EMSA_pureI_S	GTAAATGTTGCAAAATTTCTGAAAATTCGTTGACATGTGT
pureI_4870_S	CGGACTATATTGTCAGAAACAGTC	ChIP-PCR
pureI_5090_AS	CACCTAACATAAGAACCTCCTAAG
pureI_AS_BamHI	TTTGGATCCCTCCTAAGTTTTTTATGTTAATATC	PCR of full-length pureI
pureI_M1_AS_SalI	ACGGTCGACCAGAAATTTTGCAACATTT	Inverse PCR to mutate the 3′ AT-rich sequence
pureI_M1_S_SalI	CTGGTCGACCGTTGACATGTGTTGTC
pureI_M2_AS_SmaI	CCGCCCGGGCAACATTTACATAT	Inverse PCR to mutate the 5′ AT-rich sequence
pureI_M2_S_SmaI	TTGCCCGGGCGGCGAAAATTCGTTGACATGT
pureI_M3_AS_SmaI	CAGCCCGGGTGCAACATTTACATATTAGTC	Inverse PCR to mutate both the AT-rich regions
pureI_M3_S_SmaI	GCACCCGGGCTGGTCGACCGTTGACA
pureI_S_SacI	TTTGAGCTCATTCCTGGGAGACTAGCTG	PCR of full-length pureI
rpoA_AS_PstI	ATCCTGCAGTTATTATTTATCGTTTTTGAGTCC	PCR for MBP–α-CTD construction
rpoA_S_BamHI	ACTGGATCCATGAAAGAAACTGAAAAAGTGAA
Ssal_00403_S	GGACAGCTTGTGTCAGTCTAG	codY amplicon for ΔcodY construction
Ssal_00405_AS	GCCCAATGGCAAAATCATGTGC
ureC_6301_S	GGACCAACTGTAGGTGATAGCGTAC	qPCR of ureC
ureC_6460_AS	TGGATTGTCACGTGTTTCCGTAGC
ureC_AS_PstI	AAACTGCAGGAACAAGAAATAACGTTGAGAC	PCR for His-UreC construction
ureC_S_BamHI	AAAGGATCCATGAGTTTTAAAATGGATCGTG

^aInserted restriction recognition sites are underlined.

Total cellular RNA was isolated from streptococcal strains as described previously (12) and further purified by using an RNeasy minikit (Qiagen). First-strand cDNA was generated from 2 μg total RNA with random hexamer primers. Quantitative real-time PCR (qPCR) was used to evaluate the level of ureC expression. The reactions were carried out using an iQ SYBR green Supermix (Bio-Rad) and a 7500 Fast real-time PCR system (Applied Biosystem). The data were analyzed by using 7500 software (v2.0.5). Each PCR mixture contained 250 nM primers ureC_6301_S and ureC_6460_AS. Thermal cycler conditions were as follows: 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. Each reaction was run in triplicate, and at least three samples were analyzed. Of note, a melting curve analysis was performed at the end of the amplification to ensure the amplification. The change in the quantification cycle (ΔC_q) of each sample was normalized with 16S RNA. As the expression of the ure operon is induced by acidic growth pH and excess amounts of carbohydrate, the ΔC_q derived from the wild-type strain grown at neutral pH with 20 mM glucose was used as the reference. The relative quantity of each sample was calculated as the ΔC_q of the sample compared to the C_q of the reference using the formula 2^−ΔΔCq.

Construction of p_ureI-cat fusion and codY-deficient and complemented strains.

The full-length p_ureI, from nucleotide (nt) −433 to the translation start site of ureI, was generated from S. salivarius 57.I by PCR with the primer pair pureI_S_SacI and pureI_AS_BamHI. Restriction sites were included in the primers to facilitate the following cloning. Briefly, the PCR product was digested with SacI and BamHI and ligated with a promoterless cat from plasmid pC194 of Staphylococcus aureus (26) on plasmid pMC286. The p_ureI-cat fusion was subsequently cloned into an integration vector modified from pMC195 (14) to generate pMC300. The modified integration vector contains an Sp resistance gene (spe) cassette (27) located within the lacZ locus of S. salivarius 57.I, which allows the integration of p_ureI-cat into the chromosome with the concomitant acquisition of a Sp resistance (Sp^r) phenotype. Plasmid pMC300 was introduced into S. salivarius 57.I, and the correct double-crossover recombination event was verified by colony PCR using lacZ-specific primers. The resulting recombinant S. salivarius strain harboring the full-length p_ureI-cat was designated MC308.

To construct a codY-deficient strain, a 2-kbp amplicon containing codY and the flanking regions was amplified from S. salivarius 57.I by PCR using primers Ssal_00403_S and Ssal_00405_AS. The PCR product was cloned into pGEM3zf(+) to generate plasmid pSF36. The region encoding the 25th to 78th amino acids (aa) of CodY on pSF36 was deleted by inverse PCR and subsequently replaced by an Em resistance gene (erm) without a terminator (28) to generate pSky2. Plasmid pSky2 was introduced into S. salivarius MC308 to generate MC308_ΔcodY. The correct insertion inactivation event was initially verified by colony PCR using codY-specific primers. The absence of the CodY protein in the ΔcodY strain was further confirmed by Western blotting using anti-CodY antiserum. To construct a CodY complementation strain, a 1.1-kbp amplicon containing codY, its 5′ flanking region of 250 bp, and 3′ flanking region of 90 bp was generated by PCR using primers codY_800_S and codY_1900_AS. The PCR product was cloned into the E. coli-Streptococcus shuttle vector pDL276 (29) to generate pSky10. The identity of pSky10 was confirmed by sequencing analysis, and the correct recombinant plasmid was introduced into the ΔcodY strain to generate the codY complementation strain MC308_CΔcodY.

Site-directed mutagenesis of the AT-rich sequences within the putative CodY box.

The AT-rich sequences 5′ to the −35 element of p_ureI were mutated by site-directed mutagenesis and established in S. salivarius MC308_ΔcodY. Briefly, two primers that are complementary at the 5′ region and contain mutated sequences within each target were used in an inverse PCR using pMC300 as the template. Restriction endonuclease recognition sequences were included in the primers. The PCR products were digested, self-ligated, and then established in E. coli. The constructs were confirmed by sequencing analysis and then introduced into S. salivarius MC308_ΔcodY. The allelic exchange event at lacZ was confirmed as described above. The resulting strains, MC308_ΔcodY_M1 and MC308_ΔcodY_M2, harbor mutations in the 3′ and 5′ AT-rich sequences, respectively. Strain MC308_ΔcodY_M3 lost both AT-rich regions.

Purification of recombinant CodY, UreC, and α-CTD and preparation of polyclonal antisera.

The coding regions of codY and ureC were amplified from S. salivarius 57.I by PCR using the primer pairs codY_S_BamHI plus codY_AS_PstI and ureC_S_BamHI plus ureC_AS_PstI and cloned into pQE30 (Qiagen). The identity of the recombinant plasmid was confirmed by sequencing analysis. The histidine-tagged CodY (His-CodY) and UreC (His-UreC) were induced and purified under denaturing conditions using the standard procedure. The identity of the recombinant protein was confirmed by matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS). Using a similar approach, a construct to produce maltose binding protein (MBP)-tagged α-CTD (MBP–α-CTD) (using primers rpoA_S_BamHI and rpoA_AS_PstI) was generated by using pMAL-c2X (NEB) as the cloning vector. The protein concentrations of the purified proteins were measured by a Bio-Rad protein assay based on the method of Bradford (30) with bovine serum albumin (BSA) as the standard. Purified recombinant CodY and UreC proteins were used to generate polyclonal antisera in rabbits (Genesis, Taiwan). The specificity and titer of the antisera were examined by Western blotting. The recombinant proteins used in the electrophoretic mobility shift assay (EMSA) were purified under native conditions by the standard method.

Western blot analysis.

Membranes containing protein samples were blocked with 10% skimmed milk in phosphate-buffered saline (PBS) with 0.1% Tween 20 (PBST) overnight prior to hybridizing with UreC antiserum diluted 1:12,000 in PBST. Horseradish peroxidase-conjugated antirabbit and luminol-based reagents (Millipore) were used to detect UreC.

EMSA.

Two commercially prepared (Genomics BioSci & Tech), biotin-labeled probes, EMSA_pureI_S and EMSA_pureI_AS, were annealed and used as the probes in EMSA with the purified recombinant His-CodY. These two oligonucleotides contain the complementary sequences of the 29 nt immediately 5′ to the −35 element of p_ureI, the −35 element, and its downstream 5 nt. Briefly, the annealed product was incubated with increasing amounts of His-CodY in 20 μl of the binding buffer [10 mM MgCl₂, 100 mM KCl, 10 mM Tris, pH 7.5, 1 mM EDTA, 50 μg ml⁻¹ BSA, 40 μg ml⁻¹ poly(dI-dC)]. BCAA (20 mM) was included in the binding reaction mixtures. Specific competition was carried out by including the same probe without labeling in a 500-fold excess in the reaction mixture. All reactions were carried out at 4°C for 30 min and then resolved on 6% nondenaturing polyacrylamide gels. The DNA and protein complex was electrotransferred onto nylon membranes and detected by using a chemiluminescent nucleic acid detection module kit (Pierce). In the competition EMSA, various amounts of His-CodY and MBP–α-CTD were mixed with the same probe used above.

ChIP-qPCR.

The chromatin immunoprecipitation (ChIP) assay was performed by the method of Grainger et al. (31) with minor modifications. Briefly, chemostat-grown S. salivarius 57.I was cross-linked with formaldehyde, washed, and then resuspended in 1/50 of the original culture volume of the lysis buffer (31). The cell suspensions were subjected to mechanical disruption in the presence of glass beads (diameter, 0.1 mm) in a Mini-Beadbeater apparatus (Biospec Products) for a total of 5 min at 4°C. The cellular DNA in the clear lysate was sheared by sonication to generate DNA fragments with an average size of 0.5 to 1 kbp. Prior to precipitation with the antisera, the DNA suspension was incubated first with protein A/G agarose (Millipore), salmon sperm DNA, and BSA at 4°C for 1 h. The insoluble complexes were removed by centrifugation, and an aliquot of the supernatant was used in immunoprecipitation reactions with anti-CodY antiserum. Control reactions were carried out with preimmunized rabbit serum. Immunoprecipitated samples were un-cross-linked at 65°C for 12 h. DNA was then purified from the samples by using a GeneClean III kit (MP Biomedicals) prior to qPCR with primers pureI_4870_S and pureI_5090_AS.

CAT and urease assays.

The total protein lysates of S. salivarius MC308 and its derivatives were prepared as described previously (14). The chloramphenicol (Cm) acetyltransferase (CAT) activity in each lysate was determined by the method of Shaw (32), and the specific activity was calculated as the number of nmol of Cm acetylated min⁻¹ mg⁻¹ total protein. Urease activity was measured as described previously (11). The specific activity of urease was expressed as the number of nmol of urea hydrolyzed min⁻¹ mg⁻¹ total protein. All reactions in both assays were performed in triplicate, and negative controls were reactions done in the absence of the substrates (Cm or urea).

Statistical analysis.

Statistical analysis was performed using an unpaired two-tailed Student's t test or one-way analysis of variance (ANOVA) with GraphPad Prism (v5) software. Differences were considered significant if P was <0.05.

RESULTS

CodY negatively regulates p_ureI expression.

Previous studies revealed that the 20 bp immediately 5′ to the −35 element of p_ureI participates in the repression of p_ureI (14). Sequence analysis revealed a putative CodY box that differed by 1 nt from the consensus sequence derived from L. lactis (16) and is located 2 nt 5′ to the −35 element of p_ureI, suggesting that CodY is responsible for the repression observed previously. To examine p_ureI expression, a recombinant strain, MC308, harboring a p_ureI-cat fusion integrated at lacZ was constructed as detailed in the Materials and Methods. The spe cassette located at the 5′ end of p_ureI contains a strong terminator which blocks readthrough from the 5′ flanking region. To verify the repression of CodY on p_ureI in S. salivarius MC308, p_ureI activity was examined in a wild-type host and its codY-deficient (ΔcodY) derivative. The codY gene is flanked by alaA and Ssal_00405 (33) and transcribed monocistronically in S. salivarius 57.I (Fig. 1A). Inactivation of codY (Fig. 1B and C) led to the enhanced expression of p_ureI at both neutral and acidic pHs (Fig. 2A). Similarly, urease activity in the ΔcodY strain was significantly higher than that in the wild-type strain under the conditions tested (Fig. 2B), confirming that CodY negatively regulates p_ureI. A wild-type level of CAT and urease activity was observed in the complemented strain (MC308_CΔcodY). Interestingly, the CAT activity in MC308_ΔcodY at acidic pH was only slightly higher (P < 0.01, Student's t test) than that at neutral pH, whereas a greater than 7-fold increase in urease activity was seen in cultures grown under acidic conditions. As p_ureI is the major promoter of the urease operon, if not the only one, the elevated urease activity in MC308_ΔcodY grown at pH 5.5 compared to the activity in cells grown at neutral pH may result, at least in part, from the fact that urease is a comparatively stable enzyme.

FIG 1.

Inactivation and complementation of codY in S. salivarius 57.I. (A) Schematic diagram of codY and the flanking loci. The region that is replaced by erm in strain MC308_ΔcodY is indicated by a horizontal black line above codY. The putative terminator of codY is indicated by a lollipop-shaped symbol. The limits of the region established in the complemented strain (MC308_CΔcodY) are shown by two vertical arrows below the line. 405, Ssal_00405. (B) Confirmation of the genotypes of wild-type MC308, MC308_ΔcodY, and MC308_CΔcodY by PCR. The same set of primers used to generate strain MC308_CΔcodY was used here. Lanes 1 to 3, PCR products generated from wild-type MC308, the ΔcodY strain, and MC308_CΔcodY, respectively; lane M, DNA marker. (C) Western blot analysis using anti-CodY antiserum. Lanes 1 to 3, signals of lysates from wild-type MC308, the ΔcodY strain, and MC308_CΔcodY, respectively.

FIG 2.

CAT (A) and urease (B) specific activities (Sp. Act.) in batch-grown wild-type S. salivarius MC308, the codY mutant (MC308_ΔcodY), and the ΔcodY complemented strain (MC308_CΔcodY). Values are the means and standard deviations from three independent experiments. Significant differences between the wild-type and ΔcodY strains were analyzed using Student's t test. ***, P < 0.001; *, P < 0.05.

As growth pH and carbohydrate concentration are the two major growth parameters influencing p_ureI activity (34) and the chemostat culture system can control precisely each of the growth parameters, wild-type MC308 and MC308_ΔcodY were cultivated in a chemostat at pH 7 and 5.5 with 20 or 100 mM glucose. The ureC-specific mRNA and urease activities of both strains were examined by using qPCR and a biochemical assay, respectively. When 20 mM glucose was included in the growth medium, glucose was undetectable in the culture supernatant, whereas approximately 50 mM glucose was detected in cultures supplemented with 100 mM glucose (data not shown). A larger amount of the ureC message was observed in MC308_ΔcodY than in the wild-type strain under all conditions (Fig. 3A), confirming that CodY negatively regulates p_ureI. Eightfold and 10-fold increases in the ureC message were detected in MC308_ΔcodY at pH 7 under glucose-limiting (20 mM) and excess glucose (100 mM) conditions, respectively, whereas only a 4-fold increase was seen in cells grown at pH 5.5 with both glucose concentrations, suggesting that the impact of CodY on p_ureI expression was more pronounced at pH 7 than at pH 5.5. As the transcription of ureC in MC308_ΔcodY remained sensitive to growth pH and carbohydrate concentration, these results also suggested that p_ureI is regulated by additional transcription factors.

FIG 3.

Levels of the ureC message (A), urease-specific enzyme activity (B), and UreC protein (C) in chemostat-grown wild-type S. salivarius MC308 and the codY mutant (MC308_ΔcodY). I and II, TY medium containing 20 mM glucose; III and IV, TY medium containing 100 mM glucose; I and III, the system was controlled at pH 7; II and IV, the system was controlled at pH 5.5. The amount of ureC mRNA in each sample was measured by qPCR. The ΔC_q of MC308 grown at neutral pH with 20 mM glucose was used as the reference. The UreC protein was detected by Western blotting with polyclonal UreC antiserum and quantified by densitometry, and the amount of MC308 grown at neutral pH with 20 mM glucose was used as the reference. The values are the means and standard deviations from three independent experiments. Significant differences between the wild-type and ΔcodY strains were analyzed using Student's t test. ***, P < 0.001; **, P < 0.01; *, P < 0.05.

Urease activity was also elevated in MC308_ΔcodY (Fig. 3B). The differences between wild-type MC308 and MC308_ΔcodY were more striking in cultures grown under limiting glucose, especially at pH 5.5, than in those grown with 100 mM glucose. Interestingly, the pattern of urease activity in MC308_ΔcodY under the four growth conditions was quantitatively different from that of ureC mRNA, suggesting that posttranscriptional and/or posttranslational effects influence overall urease activity. To investigate the difference between the ureC mRNA levels and urease activity, the amounts of UreC protein produced in both stains under various conditions were examined by Western blotting (data not shown) and densitometry analyses (Fig. 3C). The expression pattern of the UreC protein was more similar to that of ureC mRNA, although the mean value for cells grown under 20 mM glucose, pH 5.5 was comparable to, but not lower than, that for cells grown under 100 mM glucose, pH 7. This result indicated that there is a posttranslational control point governing urease activity, presumably at the stage of assembly and/or activation of the enzyme.

CodY is able to bind to the cognate target near p_ureI.

EMSA was carried out to determine whether CodY modulates p_ureI expression by binding to the putative CodY box described above. A positive shift was apparent with 0.5 μM the recombinant His-CodY, and a dose-dependent enhancement was detected with increasing amounts of His-CodY (Fig. 4A). Addition of BCAA but not GTP enhanced the shift (data not shown). The shifting pattern was abolished when unlabeled probe was included in the reaction mixture (Fig. 4A, lane 7), confirming that CodY binds specifically to its target.

FIG 4.

EMSA and ChIP-qPCR analysis demonstrating the interaction of CodY with p_ureI. (A) EMSA of CodY binding to the CodY box in p_ureI. Lane 1, probe only; lanes 2 to 6, use of 0.5 to 8 μM His-CodY in the assay in 2-fold increments, respectively; lane 7, use of 8 μM His-CodY with a specific competitor. All reactions were carried out with 0.02 pmol biotin-labeled probe. (B) The relative quantity of p_ureI bound by CodY under different growth conditions was evaluated by ChIP-qPCR assay. Wild-type S. salivarius was grown in a chemostat in TY medium containing 20 mM glucose, pH 7 (I), 20 mM glucose, pH 5.5 (II), 100 mM glucose, pH 7 (III), and 100 mM glucose, pH 5.5 (IV). Primers pureI_4870_S and pureI_5090_AS were used to amplify p_ureI by PCR. The ΔC_q of the sample from IV was used as the reference. Significant differences between samples were determined using one-way ANOVA. ***, P < 0.001; **, P < 0.01; *, P < 0.05.

To assess the in vivo binding of CodY to p_ureI and the impact of pH and carbohydrate concentrations on the binding activity, a ChIP-qPCR assay with anti-CodY antiserum was employed with the chemostat-grown cells at pH 7 and pH 5.5 with 20 or 100 mM glucose. A direct interaction between CodY and p_ureI was clearly observed under all growth conditions, and a larger amount of CodY binding was observed with cells grown at pH 7 than with cells grown at pH 5.5, regardless of the glucose concentrations (Fig. 4B), confirming that the repression of CodY was modulated by pH. Different from the ureC message (Fig. 3A), the impact of the glucose concentration on CodY binding to p_ureI was seen only in cells grown at pH 7. At pH 5.5, a comparable amount of CodY binding was detected between cells grown in the presence of 20 mM glucose and cells grown in the presence of 100 mM glucose, indicating that additional factors participate in the transcription regulation of p_ureI. Of note, primers specific for manL, which encodes a component of the phosphoenolpyruvate sugar:phosphotransferase system (PTS) that is not regulated directly by either CodY, failed to generate a PCR product from the DNA samples obtained from anti-CodY antiserum, confirming the specificity of the assay.

The CodY box acts as an UP element of p_ureI.

As the CodY box is only 2 nt 5′ to the −35 element of p_ureI and, along with its 5′ flanking base, contains two 5′-AAAATT segments, the AT tracks may act as an upstream (UP) element to enhance p_ureI transcription. If the AT tracks are involved in the positive regulation of p_ureI, the positive effect would be more evident in a repressor-free host, i.e., the ΔcodY strain. Thus, the promoter activity of fusions with mutations in either AT track (the ΔcodY_M1 or ΔcodY_M2 strain) and both AT tracks (ΔcodY_M3 strain) was monitored in a CodY-deficient host. Mutations in any part of the CodY box resulted in reduced CAT activity at both neutral and acidic pHs (Fig. 5A and B), indicating that this region indeed enhances p_ureI activity in a codY-deficient background. EMSA was then carried out to verify the interaction between α-CTD and this region. A positive shift was evident with 15 μM recombinant α-CTD (data not shown), indicating that the AT tracks act as an UP element in the absence of CodY. To demonstrate that CodY and α-CTD compete for the same binding sequence, EMSA was also carried out again with a mixture of His-CodY and MBP–α-CTD at various ratios. MBP–α-CTD was used instead of His-CodY in this study for detection purposes. Both proteins formed stable complexes with the probe (Fig. 5C). The complex formed by His-CodY (complex a) disappeared upon the use of a 5-fold molar excess of MBP–α-CTD (complex c) and the complex formed by MBP–α-CTD disappeared upon the use of His-CodY, indicating that the two proteins compete for the same binding site. When the two proteins were used in an equal molar ratio (Fig. 5C, lane 5), a moderately migrated band (complex b) was seen, suggesting that the binding affinity of the both proteins to the target sequence was similar.

FIG 5.

Functional analysis of the CodY box as an UP element for p_ureI. (A) CAT activity in batch-grown cells at pH 7.5 and 5.5. One or both of the AT tracks in the p_ureI-cat fusion were mutated and established in a codY-deficient S. salivarius strain. Values are the means and standard deviations from three independent experiments. Significant differences between the strain with a wild-type p_ureI (MC308_ΔcodY) and strains with mutated sequences (MC308_ΔcodY_M1, MC308_ΔcodY_M2, and MC308_ΔcodY_M3) at each growth pH were analyzed by one-way ANOVA. A P value of less than 0.001 was detected in all pairs of comparisons. (B) The sequences of the CodY box region in the wild-type (WT) strain and strains MC308_ΔcodY_M1 (M1), MC308_ΔcodY_M2 (M2), and MC308_ΔcodY_M3 (M3) are listed. The wild-type sequence of the CodY box is boxed. Mutated bases in each strain are in black, and unchanged bases are in gray. (C) EMSA demonstrating the competition between His-CodY and MBP–α-CTD in binding to p_ureI. Lane 1, probe only; lanes 2 to 8, mixing of decreasing amounts of His-CodY (13, 10, 7, 3.5, 1.4, 1, and 0 μM, respectively) with increasing amounts of MBP–α-CTD (0, 1, 1.4, 3.5, 7, 10, and 13 μM, respectively). All reactions were carried out with 0.03 pmol biotin-labeled probe.

DISCUSSION

A variety of regulators of gene expression in bacteria allow the microbes to respond to different environmental signals. In this study, we demonstrated that the expression of p_ureI is regulated directly by a global regulator involved in nitrogen metabolism, CodY. This is intriguing, as the expression of p_ureI is dominantly regulated by pH, yet it is not very sensitive at all to nitrogen sources. Specifically, only a slight induction of p_ureI activity was detected in cells grown in the defined medium (FMC) containing glutamate (limiting nitrogen) compared to the level of induction in cells grown in glutamine (excess nitrogen) (data not shown). We also demonstrated here that the AT tracks of the CodY box could act as an UP element (35) for enhancing promoter strength in the absence of CodY. Biosynthesis of a catalytic active urease is a costly process that requires the activity of a complicated assembly machinery composed of several proteins, yet ureolysis at elevated pH values can be detrimental to the cells, so multifactorial regulation could allow fine-tuning of the system and the optimal fitness of S. salivarius. As the expression of p_ureI remains sensitive to pH and carbohydrate concentrations in the codY-deficient background, it is suggested that additional trans-acting elements are involved in the regulation. Previous studies have demonstrated that carbohydrate concentrations in the growth medium affect urease expression (13) and the PTS participates in urease regulation (34, 36); thus, it is likely that the additional regulation is directed by carbohydrate availability.

The results of p_ureI expression in chemostat cultures (Fig. 3A) and ChIP-qPCR (Fig. 4B) indicate that growth pH modulates the activity of CodY. Such modulation is likely influenced by a shift of the amount of intracellular metabolites at different growth pH values. Studies in Bifidobacterium longum subsp. longum (37) and Lactobacillus sanfranciscensis (38) have shown that metabolic shifts to BCAA catabolism for ATP generation participate in acid tolerance. Thus, in the same culture medium, the amount of intracellular BCAAs present at low pH may be smaller than that found at neutral pH, leading to less activation of CodY at low pH. Similarly, it is possible that the enhanced p_ureI activity in cells grown under excess carbohydrate compared to the activity in cells grown under limiting carbohydrate is due to a reduced nitrogen supply. Thus, CodY may act as a control point for pH-dependent regulation, which may be critical in the oral cavity, where rapid and extreme changes in pH are common.

As a dominant repressor of p_ureI, it is expected that inactivation of codY would lead to higher levels of p_ureI activity. However, the difference between the amount of UreC protein and urease activity at neutral pH in cells growing with an excess amount of glucose suggests that only a portion of the apo-urease was assembled into holoenzymes under this condition. Activation of apo-urease depends on CO₂ availability and uptake for lysine carbamylation of the active center, the availability and transport of Ni²⁺, the delivery of Ni²⁺ to the active center, and the availability of GTP as an energy source (39). Any one of these factors alone or in combination could influence overall urease activity. The previous demonstration that supplementation of Ni²⁺ in the culture medium enhances urease activity in cells grown at neutral pH (5) indicates that the accumulation of intracellular Ni²⁺, but not apo-urease, is a limiting variable in urease activation. Thus, the difference between the ureC message and urease activity in a ΔcodY strain grown at neutral pH with 100 mM glucose is caused, at least in part, by differences in the amount of intracellular Ni²⁺. Whether other factors, such as intracellular CO₂ concentrations, also contribute to the observed results will require further investigation.

Taken together, a working model for the regulatory network governing urease production is proposed (Fig. 6). At neutral pH, CodY represses the expression of p_ureI by binding to the CodY box 5′ to the −35 element. The inhibitory effect of CodY is less active when the growth pH becomes acidic, presumably due to the reduced availability of BCAAs. The release of CodY from the CodY box allows the binding of α-CTD to the UP elements and enhanced expression from p_ureI. The complex regulation of urease likely evolved as a result of the need to tightly control the expression of the genes in response to multiple stimuli so as to optimize growth and survival in a constantly changing environment.

FIG 6.

Model for the urease regulatory circuit in S. salivarius. Gray square, CodY box on p_ureI; two open squares, the −10 and −35 elements of p_ureI; dashed arrows, negative regulation of CodY. The model suggests that BCAA is less available at an acidic growth pH and, thus, that repression of p_ureI by CodY is more active at neutral pH than that at acidic pH. In the absence of CodY, α-CTD (α) interacts with the sequence 5′ to the −35 element of p_ureI and further enhances expression.