The strains and plasmids used in this study are described in Supplementary Table S1. Y. enterocolitica strains were cultured at 26°C or 37°C in LB medium (10 g/l tryptone, 5 g/l yeast extract, 5 g/l NaCl). E. coli strains were grown at 37°C in LB medium. Antibiotics were used at the following concentrations: nalidixic acid (Nal) – 30 μg/ml, chloramphenicol (Cm) – 25 μg/ml, kanamycin (Km) – 50 μg/ml, gentamicin (Gm) – 40 μg/ml, tetracycline (Tet) – 12.5 μg/ml, trimethoprim (Tp) – 50 μg/ml.

Triplicate overnight cultures of Y. enterocolitica strains Ye9N (the wild-type strain) and AR4 (OmpR-deficient mutant) were grown in LB, pH 7.0 at 26°C or 37°C with shaking (150 rpm) to an OD₆₀₀ of 3.0. Samples of 25 ml were centrifuged (8000 × g, 20 min, 4°C), and the cell pellets were flash frozen in liquid nitrogen and stored at −80°C prior to fractionation. Each of the bacterial pellets was resuspended in 12.5 ml of lysis buffer (200 mM Tris−HCl pH 8.0, 0.5 M sucrose, 250 μg/ml lysozyme, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride) and incubated at 4°C for 1 h. The cell suspensions were then sonicated on ice for 18 cycles of 30 s, separated by 30 s rest intervals, using a Sonics Vibra-Cell VCX 130 (Sonics & Materials, Newtown, CT, United States). The cell lysates were centrifuged for 10 min at 8000 × g at 4°C to remove unbroken cells and debris, then the supernatants were centrifuged at high speed (35,000 × g, 1.5 h, 4°C) to pellet total membranes. The protein concentrations in the supernatants (cytoplasmic protein extracts) were measured using the RC-DC protein assay (Bio-Rad, Hercules, United States) and equalized by dilution with lysis buffer. The protein extracts were precipitated by adding 4 volumes of cold acetone and holding at −20°C overnight. The samples were then centrifuged for 20 min at 14,000 × g at 4°C. The resultant protein pellets were rinsed with cold acetone twice, each time resuspending and centrifuging for 20 min at 14,000 × g at 4°C. After carefully removing the supernatants, the protein precipitates were briefly air-dried and resuspended in 40 μl of 5× Invitrosol LC/MS Protein Solubilizer (Invitrogen Life Science Technologies). To each sample, 4 μl of 20% (w/v) urea stock solution was added to facilitate dissolution of the pellet. The samples were then vortexed for 3 min, incubated at 60°C for 25 min and diluted with 160 μl of 25 mM ammonium bicarbonate to produce a final Invitrosol concentration of 1×.

Sample preparation for mass spectrometry analyses were performed as described previously (Nieckarz et al., 2016). The cytoplasmic proteomes of strains Ye9N and AR4 were compared to produce a differential proteome list. Proteins whose abundance differed significantly (q-value ≤ 0.05) between the wild-type strain Ye9N and OmpR-deficient mutant AR4, according to MS analysis, were defined by the ratios of ≤0.67 (protein more abundant in AR4) or ≥1.5 (protein less abundant in AR4).

The acid-sensitivity assay was performed essentially as described by Kakoschke et al. (2014). To measure survival in acid, overnight cultures grown in LB at 26°C or 37°C were diluted to 10⁷ CFU/ml in PBS (pH 7.0). Then, 0.5 ml of each bacterial cell suspension was mixed with an equal volume of PBS acidified with acetic acid to pH 4.0 (acid stress), or PBS at pH 7.0 (control). The cell suspensions were incubated at 26°C or 37°C for 90 min and then dilutions were plated on LB plates to determine the number of viable cells as CFU/ml. The percentage survival value was defined as the CFU/ml after acid treatment × 100/CFU/ml of the control.

DNA manipulations, polymerase chain reactions (PCRs), restriction digests, ligations and DNA electrophoresis, were performed according to standard protocols (Sambrook and Russell, 2001). Plasmid and genomic DNA were isolated using a Plasmid Miniprep DNA purification Kit and GeneMATRIX Bacterial & Yeast Genomic DNA Purification Kit (EurX), respectively. When amplified fragments were used for cloning, the PCR was performed using DreamTaq DNA polymerase or Phusion High-Fidelity DNA polymerase (Thermo Scientific). Oligonucleotide primers for PCR and sequencing were purchased from Sigma Aldrich and are listed in Supplementary Table S2. DNA fragments amplified by PCR were purified using a PCR/DNA Clean Up kit (EurX). The plasmids used in this study are described in Supplementary Table S1. DNA sequencing was performed by Genomed S.A. (Warsaw, Poland).

To obtain ureABC:lacZ, ureEF:lacZ, ureGD:lacZ and ureR:lacZ transcriptional fusions, DNA fragments containing the promoters of the ureABC, ureEF and ureGD operons, and the ureR-like gene were amplified from Y. enterocolitica chromosomal DNA by PCR using the primer pairs NureABCKpnIL/NureABCKpnIP, LureEFEcoRI/PureEFKpnI, LureGDEcoRI/PureGDKpnI, and EcoRIReg1/KpnIReg2, respectively (Supplementary Table S2). The amplified fragments were digested with KpnI (in the case of the ureABC operon promoter) or EcoRI/KpnI and cloned into the corresponding sites of reporter vector pCM132Gm [derivative of plasmid pCM132 (Marx and Lidstrom, 2001) containing a gentamicin resistance cassette], upstream of a promoterless lacZ gene. The resulting constructs were verified by PCR using the primer pair pCM132GmSPR1/pCM132GmSPR2 (flanking the EcoRI and KpnI recognition sequences) followed by sequencing of the amplicons. The correct orientation of the cloned ureABC promoter was verified by PCR using the primer pair pCM132GmSPR1/lacZSprP and by sequencing. The constructs pCM132Gm-ureABC:lacZ, pCM132Gm-ureEF:lacZ, pCM132Gm-ureGD:lacZ and pCM132Gm-ureR:lacZ were introduced into E. coli CC118 λpir and transferred by conjugation into Y. enterocolitica Ye9N, the ompR mutant AR4, Ye8N and the ompB mutant KJ4, selecting transconjugants on LB plates containing Gm and Nal. The presence of these constructs in the Y. enterocolitica strains was confirmed by plasmid isolation and PCR with the primer pair pCM132GmSPR1/pCM132GmSPR2.

β-galactosidase assays were performed essentially as described by Thibodeau et al. (2004), using 96-well flat-bottomed plates (Nest Sc. Biotech.) and a Sunrise plate reader (Tecan). Briefly, cultures grown to stationary phase at 26°C or 37°C were diluted with LB medium to an OD₆₀₀ of 0.4–0.5 and 80 μl of each cell suspension were then mixed with 20 μl of POPCulture Reagent (EMD Millipore Corp) and incubated for 15 min to cause cell lysis. In the wells of a microtiter plate, 20 μl of each cell lysate were mixed with 130 μl of Z Buffer and 30 μl of ONPG (4 mg/ml). For kinetic assays, the absorbance at 415 nm (relative to a blank) was measured at time intervals of 10 s, with 2 s of shaking before each reading. The assays were performed at 26°C and monitored for up to 20 min. The β-galactosidase activity was expressed in Miller units calculated as described previously (Thibodeau et al., 2004). Each assay was performed at least in triplicate. To test the effect of acid pH, high osmolarity and urea on reporter gene expression, cultures grown to stationary phase at 26°C were diluted to an OD₆₀₀ of ∼ 0.4 in LB medium. Then 1 ml of each culture was centrifuged (4000 × g, 5 min, room temperature) and the cell pellets were resuspended in 1 ml of LB (86 mM NaCl, pH 7.0), LB adjusted to pH 4.5, LB supplemented with NaCl to a final concentration of 350 mM or LB containing 100 mM urea. These cultures were incubated for 2 h at 26°C with shaking (150 rpm). After two washes in LB medium by centrifugation (4000 × g, 5 min, room temperature) and gentle re-suspension, β-galactosidase activity in the final cell suspensions was measured.

The in vitro interaction between phosphorylated OmpR (OmpR-P) and the promoters of selected genes was examined using EMSAs, as described previously (Jaworska et al., 2018), with some modifications. The primers listed in Supplementary Table S2 were used in PCRs with Y. enterocolitica Ye9N genomic DNA to amplify DNA fragments comprising the regulatory regions of the operons ureABC (−305 to + 7 relative to the A of the ATG start codon of ureA), ureEF (−352 to + 33 relative to the A of the ATG start codon of ureE), ureGD (−272 to + 33 relative to the A of the ATG start codon of ureG), and gene ureR-like (−310 to + 52 relative to the A of the ATG start codon of ureR-like). Increasing amounts of recombinant OmpR-His₆ were phosphorylated in vitro by incubation for 30 min at room temperature in reaction buffer (50 mM Tris-HCl pH 8.0, 100 mM KCl, 1 mM EDTA, 1 mM DTT, 20 mM MgCl₂, 12% glycerol, 100 μg/ml BSA, 0.1% Triton X-100) containing 20 mM acetyl phosphate (lithium potassium acetyl phosphate; Sigma-Aldrich). After phosphorylation, 0.05 pmoles of the respective test fragments were added to form the separate EMSA binding reactions. As a negative control, a 211-bp fragment of the Y. enterocolitica 16S rRNA gene amplified by PCR (Supplementary Table S2) was also included in each reaction. The binding reactions were analyzed by electrophoresis (∼3 h at 110V) on 4.2% native polyacrylamide gels (19:1 acrylamide/bis-acrylamide, 0.2 × TBE, 2% glycerol). SYBRgreen 1× solution (Invitrogen) was used to stain the DNA bands in the gels, which were visualized with a GE Healthcare AI600 imager.

Overnight cultures of Y. enterocolitica strains grown in LB, pH 7.0 at 26°C or 37°C were adjusted to an OD₆₀₀ of 1.0 (approximately 10⁹ bacterial cells per ml). Then 1 ml of each cell suspension was centrifuged (4000 × g, 3 min, 25°C). The cell pellets were resuspended in 0.1 ml of LB and 10 μl lots were spotted onto the surface of urease enzyme assay plates [Christensen’s urea agar – 1 g/L peptone, 1 g/L D(+)-glucose, 2 g/L potassium dihydrogen phosphate, 5 g/L sodium chloride, 0.012 g/L phenol red, 12 g/L agar, pH 6.8 with or without 2% urea]. Following incubation at 26°C or 37°C for 24 h, urease activity was observed as pink zones around the bacterial growth. Urease catalyzes the hydrolysis of urea to ammonia, which causes an increase in pH indicated by the pH indicator phenol red (Young et al., 1996).

The quantitative urease activity assay was based on the measurement of ammonia formed by enzymatic hydrolysis of urea, using Nessler’s reagent. Briefly, triplicate cultures of the Y. enterocolitica strains were grown overnight at 26°C in LB medium with shaking (150 rpm). Then, 1 ml of each culture was centrifuged (4000 × g, 3 min, 25°C) and the cell pellets washed once in Christensen’s urea medium [1 g/L peptone, 1 g/L D(+)-glucose, 2 g/L potassium dihydrogen phosphate, 5 g/L sodium chloride pH 6.8, with 2% urea] and resuspended in the same. The cell suspensions were diluted to an OD₆₀₀ of 0.1 in Christensen’s urea medium and incubated for 2 h at 26°C with shaking (150 rpm). The OD₆₀₀ of the cultures was measured to confirm no change, then 90 μl aliquots were mixed with 180 μl of 10% Nessler’s reagent (Sigma Aldrich) in the wells of a 96-well flat-bottomed plate (Nest Sc. Biotech.). Controls containing only Christensen’s urea medium were also included. The absorbance at 420 nm (relative to a blank) was measured 5 min after mixing the cells with Nessler’s reagent, using a TECAN Infinite M200PRO microplate reader. The levels of ammonia produced were determined from a standard curve constructed using defined concentrations of NH₄Cl prepared fresh in the assay buffer (Young et al., 1996).

The ompB:Tp deletion mutant of Y. enterocolitica Ye8N (replacement with Tp resistance gene cassette) was obtained by homologous recombination using suicide vector pDS132 (Philippe et al., 2004), carrying the mutagenic DNA fragment constructed by overlap extension PCR using primers listed in Supplementary Table S2. Briefly, three DNA fragments were PCR-amplified using strain Ye8N genomic DNA (for flanking regions) or plasmid p34E-Tp (for the Tp cassette) as the templates. A mixture of these three amplicons was then used as the template with flanking primers in a PCR. The mutagenic fragment was cloned into pDS132. The resulting construct pDSompB was transferred from E. coli CC118 λpir to Y. enterocolitica strain Ye8N by triparental mating, with E. coli strain DH5α carrying plasmid pRK2013 as a helper. To select for the second recombination, the transconjugant strains were plated on LB agar containing Tp and 10% (w/v) sucrose. Sucrose-resistant colonies were screened for the loss of the plasmid (Cm resistance). The correct allelic exchange was verified by PCR using primers listed in Supplementary Table S2 and by sequencing.

To complement the ompB mutation, the coding sequence of the ompB (ompRenvZ) operon with the native ribosome binding site was amplified by PCR using Ye8N chromosomal DNA as the template with primers OmpB1 and OmpB2 (Supplementary Table S2). The amplified fragment was digested with EcoRI/BamHI and cloned into the corresponding sites of reporter vector pBBR1MCS-2, containing a kanamycin resistance cassette (Kovach et al., 1995). The resulting construct pompB was verified by sequencing and used to transform E. coli S17 λpir. This plasmid was then introduced into the ompB mutant strain KJ4 by biparental conjugation. Exconjugants were selected on LB agar plates containing trimethoprim and kanamycin.

Bioinformatic analyses of the mass spectrometric data were performed as described previously (Nieckarz et al., 2016). In silico analyses of the ure gene cluster and ureR-like gene in Y. enterocolitica strains were performed on a shotgun genome sequence of Y. enterocolitica subsp. palearctica Ye9N (bio-serotype 2/O:9; NCBI/GenBank: JAALCX000000000) and Y. enterocolitica subsp. enterocolitica 8081 (bio-serotype 1B/O:8; NCBI/GenBank: AM286415). The Needleman-Wunsch algorithm was used for sequence alignment (Altschul et al., 1997). Promoter prediction was performed using the web-based software BPROM in the Softberry package (Solovyev and Salamov, 2011). Manual homology searches of predicted urease sub-units were conducted by BLAST analysis¹ to describe a urease protein complex. The MOTIF bioinformatics tool provided by GenomeNet, Japan, was used to search for sequence motifs². The identified proteins were described according to the UniProt databases or GenBank, or homologous sequences obtained using BLAST. Statistical analyses were performed using Prism 5 software (v. 5.01, GraphPad). Student’s t-test was used to determine statistically significant differences.

To study the effect of temperature on the cytoplasmic proteome of Y. enterocolitica, cells of wild-type strain Ye9N, grown in standard lysogeny broth (LB) medium at 26°C or 37°C, were fractionated using ultracentrifugation and the cytoplasmic fraction was analyzed by shotgun label-free quantitative LC-MS/MS. This analysis revealed 23 differentially expressed proteins (q-value ≤ 0.05, at least two peptides per protein, minimal acceptable fold change 1.5, Table 1). The more abundant proteins at the higher growth temperature included molecular chaperones such as GroEL (Hsp60; ∼4-fold) together with its co-chaperone GroES (∼7-fold), DnaK (Hsp70, ∼1.7-fold), ClpB (∼6-fold) and HtpG, a prokaryotic Hsp90 homolog (∼4-fold). In contrast, the level of trigger factor (TF), a ribosome-associated peptidyl cis/trans isomerase, which represents the only ribosome-associated chaperone known in bacteria (Hoffmann et al., 2010), was decreased at 37°C (∼2.7-fold). The level of YopE, the secreted effector of the Yersinia Ysc-Yop T3SS, which disrupts the actin cytoskeleton of eukaryotic cells (Viboud and Bliska, 2005; Aepfelbacher et al., 2007; Trosky et al., 2008), was higher at 37°C compared to 26°C (∼9-fold), confirming previous reports (Lambert de Rouvroit et al., 1992; Akopyan et al., 2011). Upregulation of YerA (∼6.7-fold), a chaperone required for YopE secretion at 37°C, was also noted (Parsot et al., 2003). Proteins more abundant at the lower temperature (26°C) included the periplasmic maltose-binding protein MalE (∼14.39-fold), amino acid metabolic enzymes histidine ammonia-lyase HutH (∼7.57-fold) and succinylornithine transaminase AstC (∼7.99-fold), as well as bacterioferritin Bfr (∼7.43-fold). Interestingly, this proteomic analysis revealed downregulation at 37°C of a structural subunit of urease, UreC (∼2.3-fold), and the urease accessory proteins UreE (∼2.8-fold) and UreG (∼3.2-fold).

Shotgun Proteomic analysis was applied to identify cytoplasmic proteins subject to regulation by OmpR. We analyzed cytoplasmic fractions of the wild-type strain Ye9N and the isogenic ompR null mutant AR4, cultured at 26°C or 37°C in LB medium, by LC-MS/MS analysis. The mutant AR4 (ΔompR:Km) was constructed previously by a reverse genetics PCR-based strategy (Brzostek et al., 2003). The proteomic analysis revealed 28 differentially expressed proteins (q-value ≤ 0.05, at least two peptides per protein, minimal acceptable fold change 1.5, Table 2). The major difference in cytoplasmic protein abundance between the strains at 26°C, was a more than 130-fold decrease in the level of HU-alpha, a subunit of histone-like bacterial protein important in maintenance of the nucleoid structure (Boubrik et al., 1991; Zhou et al., 2005; Kamashev et al., 2017), in the ompR mutant. This indicated that OmpR positively affects hupA gene expression. It was previously demonstrated that the Y. pestis PhoP protein, which belongs to the OmpR-family of two-component system response regulators, positively influences the production of HU-alpha (Zhou et al., 2005). Differential quantitative LC-MS/MS analysis of cytoplasmic samples revealed that urease subunit gamma UreA was less abundant (∼3-fold) in the ompR-negative strain AR4 compared with the parental strain at 26°C, suggesting positive OmpR-dependent regulation.

The proteomic analysis of cytoplasmic fractions from cells grown at 37°C suggested that OmpR exerts a positive influence on the expression of the structural subunit of urease, UreC, since levels of this protein were reduced in the ompR mutant at this temperature (∼6-fold). In addition, we observed a two-fold decrease in the abundance of the chaperone ClpB in the ompR mutant. Proteins displaying reduced levels in the ompR mutant at 37°C included the putative sigma(54) modulation protein (∼13.6-fold) and the housekeeping enzymes 2,5-diketo-D-gluconate reductase A (∼6.8-fold) and glycine dehydrogenase GcvP (∼11.8-fold). In agreement with our previous finding, the absence of the OmpR regulator also caused an increase in the proteins of the Ysc-Yop T3SS (Nieckarz et al., 2016). These subunits of the Yersinia injectisome, which translocates the effector Yop proteins into host cells, include the translocation pore protein YopB (∼4.7-fold), forming a pore in the host cell membrane, the needle tip protein LcrV (∼5.8-fold) and the protein YscF, which polymerizes into a needle-like structure (45-fold) (Lee et al., 2000; Nordfelth and Wolf-Watz, 2001; Dewoody et al., 2013). Our proteomic data also revealed an increased amount of ribosomal proteins in the ompR mutant strain AR4 compared to the parent strain Ye9N at 37°C (∼ 2- to 7-fold), as well as higher levels of the chaperones GroEL (∼12.6-fold) and HtpG (∼1.6-fold).

The results of the proteomic analysis indicated that OmpR affects (both positively and negatively) the production of a number of proteins that serve a variety of functions. Notably, two of the three structural subunits of urease (UreA and UreC) were identified among the positively regulated OmpR-dependent targets.

To investigate the influence of OmpR on the production of urease by Y. enterocolitica we examined the activity of this enzyme in the wild-type strains Ye9N (a low-level pathogenic strain of bio-serotype 2/O:9) and Ye8N (a high-level pathogenic strain of bio-serotype 1B/O:8), and their respective mutants AR4 (ΔompR:Km) and KJ4. The mutant KJ4, lacking the ompB operon (ΔompRenvZ:Tp), was constructed in the present study by targeted deletion via homologous recombination (see MATERIALS AND METHODS).

Ureolytic activity was estimated in a semi-quantitative manner by plating Y. enterocolitica strains on Christensen’s urea agar and monitoring the change in color after 24-h incubation at 26°C or 37°C (Figure 1A). Higher urease activity was observed in strain Ye9N than in Ye8N at both temperatures. Moreover, the ureolytic activity of Ye9N was higher at 26°C than 37°C. The ompR mutant (AR4) exhibited a decreased level of urease production compared to Ye9N at both temperatures. In contrast, no visible difference was observed between Ye8N and KJ4 (ompB), irrespective of the growth temperature. To confirm that OmpR promotes urease production in Ye9N, plasmid pompR carrying the wild-type ompR allele was used to complement the mutation in strain AR4. Complementation caused a significant increase in urease activity in this ompR mutant at 26°C but not at 37°C. However, expression of the ompB operon in trans (plasmid pompB carrying both the ompR and envZ genes) in the ompB mutant KJ4, did not affect urease activity at either of the tested temperatures.

To further examine the impact of OmpR on the production of urease we employed a quantitative assay using Nessler’s reagent. The results presented in Figure 1B confirmed the Christensen’s urea agar data. Strain Ye9N displayed the highest urease activity, and that detected in the ompR-deficient strain AR4 was significantly lower (2.6-fold). Moreover, complementation of the ompR mutation in strain AR4 caused a clearly visible increase in urease activity (1.8-fold). There was no difference in urease activity between strains Ye8N, mutant KJ4 and KJ4/pompB. These findings showed that the level of ureolytic activity is dependent on the Y. enterocolitica strain, the presence of OmpR protein, as well as the temperature.

To examine whether there are other phenotypic differences between the Y. enterocolitica strains Ye9N and Ye8N, their survival abilities at acidic pH was analyzed. It was previously demonstrated that the regulator OmpR is necessary for the survival of Y. enterocolitica strains of bio-serotypes 1B/O:8 and 2/O:9 in acidic conditions in vitro (Dorrell et al., 1998; Brzostek et al., 2003). However, in these separate studies, the acid stress conditions and temperatures applied to these strains were different. Thus, it was thought necessary to examine the effect of OmpR on the susceptibility to acid stress of strains Ye9N and Ye8N by studying them in parallel. The acid survival abilities of these strains and their OmpR-deficient mutants were analyzed by exposing cells grown at 26°C or 37°C to pH 4.0 for 90 min, and then determining the viable cell number. This was expressed as a percentage of the number of viable cells following exposure to pH 7.0 for the same length of time (Figure 2). Compared to wild-type strain Ye9N, its ompR mutant AR4 displayed decreases in% survival in acid conditions of 46% at 26°C and 26% at 37°C (Figure 2A). Similarly, the OmpR-deficient mutant KJ4, exhibited decreased survival in acid conditions compared to wild-type parent strain Ye8N: by 23% at 26°C and 61% at 37°C (Figure 2B). To confirm that OmpR promotes acid survival in Y. enterocolitica, plasmids pompR and pompB carrying the wild-type ompR alleles were used to complement the mutations in strain AR4 and KJ4, respectively. Complementation caused a significant increase in survival at pH 4.0, except for strain AR4 grown at 37°C. Together, these findings showed that both Y. enterocolitica strains are able to survive acid shock at the two tested temperatures, but strain Ye9N is more acid sensitive than Ye8N at 26°C. Moreover, OmpR promotes resistance to acidic stress. Strong OmpR-dependent acid survival was observed for strain Ye9N at 26°C and for strain Ye8N at 37°C. Thus, the impact of this regulator on acid survival is both strain specific and affected by temperature.

A shotgun genome sequence has recently been generated for Y. enterocolitica strain Ye9N (2/O:9). The availability of raw sequence data files of the Ye9N genome (NCBI/GenBank: JAALCX000000000) allowed us to examine the ure locus (contig 39) and compare it to that of strain 8081 (1B/O:8, NCBI/GenBank: AM286415,Thomson et al., 2006), whose derivative is strain Ye8N. Bioinformatic analysis revealed the presence of chromosomally located multigenic ure loci in both strains (Figure 3 and Supplementary Figure S1A). Within each locus, the ure genes are all in the same transcriptional orientation. In these strains, the ure locus is flanked by conserved genes: upstream is a gene encoding a hypothetical protein, while downstream there is the yut gene encoding a urea transporter (Sebbane et al., 2002). Using BPROM software, three putative promoters (−10 and −35 motifs) located upstream of ureA, ureE and ureG were identified within the urease cluster of both strains (Figure 3 and Supplementary Figure S1A), suggesting its organization in three operons, ureABC, ureEF and ureGD, encoding the structural (UreABC) and accessory/regulatory proteins (UreEF and UreGD). Alignment of the nucleotide sequences of ureABC, ureEF and ureGD of strain Ye9N with whose of strain 8081 showed 97%, 98% and 99% sequence identity, respectively. Nucleotide sequence alignments of the intergenic region upstream of ureABC of the analyzed strains revealed 98% sequence identity. Analogous alignments of the intergenic regions upstream of ureEF showed 99% identity, while sequences upstream of ureGD exhibited 99% identity.

The sequence conservation of polypeptides comprising the urease system of both strains was examined further. The urease structural proteins UreA, UreB and UreC, as well as the accessory/regulatory proteins of the Y. enterocolitica strain Ye9N exhibit a high degree of amino acid sequence identity to those encoded by the corresponding open reading frames (ORFs) of Y. enterocolitica 8081 (98−100%) (data not shown).

Finally, using BLASTp searches to analyze the genomes of both Y. enterocolitica strains we identified a putative transcriptional regulator of the AraC family (289 aa) with homology to the regulator UreR identified in other ureolytic bacteria. Amino acid sequence alignment of this protein (strain 8081, GenBank: CAL12570) revealed 71% similarity (56% identity) to the urease operon transcriptional activator of Serratia ficaria (GenBank: SNW02772.1) and 43% similarity (22% identity) to UreR of P. mirabilis (GenBank: Z18752.1). Therefore, for consistency, this AraC-like protein of Y. enterocolitica will be termed UreR-like from here on. In silico analysis of this protein sequence using the MOTIF online tool (GenomeNet, Japan), identified helix-turn-helix motifs and an AraC-like binding domain.

The UreR-like protein of both strain 8081 (1B/O:8) (NCBI/GenBank: AM286415.1) and strain Ye9N (contig 5) is encoded by a gene located within a cluster of flagellar structural and assembly genes (downstream of the fliT gene and upstream of the gene encoding metal-dependent phosphohydrolase), and not in the vicinity of the ure cluster (Figure 3 and Supplementary Figure S1B). The nucleotide sequence identity of ureR-like genes of strains Ye9N and 8081 is 99% (and Supplementary Figure S1B), while that of the intergenic regions upstream of the ureR-like genes is 97%.

To examine the impact of regulator OmpR on the activity of urease operon promoters in different Y. enterocolitica strains, we constructed transcriptional fusions of the putative promoter sequences with the lacZ gene in plasmid pCM132Gm. The P_ureABC, P_ureEF, and P_ureGD promoters located upstream of the corresponding operons were identified using the bacterial promoter prediction program BPROM (Solovyev and Salamov, 2011). The three reporter gene constructs were introduced into the wild-type Y. enterocolitica strains Ye9N and Ye8N, and their respective OmpR-deficient derivatives AR4 (ompR mutant) and KJ4 (ompB mutant). The expression of the transcriptional fusions was quantified by measuring β-galactosidase activity in cells grown to stationary phase at 26°C or 37°C in LB medium (Figure 4). The activity of the ureABC promoter in the ompR mutant (AR4 strain) was lower than in the wild-type strain Ye9N at 26°C and 37°C (19 and 33% lower, respectively) (Figure 4A). To confirm positive OmpR-dependent regulation of this urease operon in strain Ye9N, a plasmid carrying a wild-type copy of the ompR gene (pompR in Figure 4) was used to complement the ompR mutation in strain AR4. This caused a statistically significant increase in the level of β-galactosidase activity from ureABC:lacZ at both tested temperatures, indicating that OmpR acts as a positive regulator of the operon encoding the three structural subunits of urease in this strain. The positive regulatory effect of OmpR was also observed for fusions carrying the promoter regions of the ureEF (Figure 4B) and ureGD (Figure 4C) operons. Reduced activity of ureEF:lacZ (18% and 24% decrease in activity) and ureGD:lacZ (38% and 17% decrease in activity) fusions was observed in the ompR mutant strain AR4 compared to the wild-type Ye9N at 26°C and 37°C, respectively. However, an effect of complementation with plasmid pompR was only detected for the reporter fusion ureGD:lacZ at 37°C. Interestingly, the OmpR-dependent regulation of urease operon transcription was not observed in Y. enterocolitica Ye8N, at either 26°C or 37°C. However, the β-galactosidase activity of ureABC:lacZ was increased in complemented strain KJ4/pompB at both temperatures. We concluded that OmpR positively and specifically influences ure expression in Y. enterocolitica strain Ye9N, but its role is less clear in strain Ye8N.

With regard to the influence of temperature on urease promoter activity, β-galactosidase production by each urease fusion was higher at 26°C compared to 37°C in Y. enterocolitica strain Ye9N: by 12% for ureABC:lacZ, 27% for ureEF:lacZ and 40% for ureGD:lacZ. In the case of strain Ye8N, temperature had no significant effect on expression of the ureABC reporter fusion. However, for ureEF:lacZ and ureGD:lacZ, lower β-galactosidase activity was observed at 37°C compared to 26°C. Furthermore, we found higher ure operon expression in strain Ye9N than in Ye8N at both tested temperatures. These data indicated OmpR- and temperature-dependent modulation of the Y. enterocolitica Ye9N ure operons that parallels urease activity (see Figure 1).

The expression of many OmpR regulon genes in E. coli and Salmonella is osmoregulated and/or modulated in response to pH conditions (Chakraborty and Kenney, 2018). Next, we tested whether the expression of urease transcriptional units changes in response to high osmolarity and/or acidic conditions. Moreover, we were curious to see if urea could influence ure cluster gene expression. The effect of 350 mM NaCl, pH 4.5 and 100 mM urea on the β-galactosidase activity produced by ure operon:lacZ transcriptional fusions was tested in stationary phase cells of the wild-type Y. enterocolitica strains Ye9N and Ye8N grown at 26°C (Figure 5). No effect of high osmolarity (350 mM NaCl) or low pH (pH 4.5) was observed for the ureABC:lacZ fusion in strains Ye9N and Ye8N (Figure 5A). Only the addition of 100 mM urea caused a slight induction of ureABC expression in both tested strains. Measurements of β-galactosidase activity demonstrated that ureEF:lacZ expression was lower in strain Ye9N grown under acidic and high osmolarity conditions (Figure 5B). For strain Ye8N, a decrease in the expression of ureEF:lacZ was only observed in the presence of 350 mM NaCl. The addition of 100 mM urea significantly upregulated ureEF:lacZ expression only in strain Ye8N. The ureEF pattern of expression in response to high osmolarity, low pH and urea was also observed for the ureGD:lacZ fusion in strain Ye9N (Figure 5C). In strain Ye8N, significant changes in ureGD:lacZ expression were recorded in cells grown at pH 4.5 (downregulation) and in the presence of urea (upregulation).

To examine whether Y. enterocolitica ureR-like expression responds to the presence of 100 mM urea, acidic conditions (pH 4.5) or high osmolarity (350 mM NaCl), and to test the influence of the regulator OmpR, we constructed a transcriptional fusion of the putative ureR-like promoter sequence identified by BPROM (Solovyev and Salamov, 2011) with the lacZ gene in vector pCM132Gm. This construct pCM132Gm- ureR:lacZ was introduced into the strains used to test the urease operon fusions, i.e., Ye9N and Ye8N and their OmpR-deficient derivatives, AR4 (ompR mutant) and KJ4 (ompB mutant).

As shown in Figure 6A, high osmolarity had no effect on ureR:lacZ expression in either of the strains Ye9N or Ye8N. A 2-h exposure to an acidic environment resulted in decreased ureR-like expression in the strain Ye9N. A slight stimulatory effect of urea (1.2-fold) was observed, but only in the strain Ye8N, which parallels the induction of ure gene expression by urea seen in this strain.

The ureR:lacZβ-galactosidase activity detected in the ompR mutant (strain AR4) was lower than that measured in the wild-type strain Ye9N at 26°C or 37°C (3.3- and 2.2-fold lower, respectively), and this effect was complemented by the wild-type ompR allele provided in trans (Figure 6B). These data suggested that OmpR is a positive regulator of the ureR-like gene in strain Ye9N. Interestingly, we found that the absence of the ompB in strain KJ4 resulted in the upregulation of ureR-like gene expression at both tested temperatures (by 1.3-fold at 26°C and 1.4-fold at 37°C). However, the addition of plasmid-encoded ompB to strain KJ4 did not complement the ompB deletion and even upregulated the ureR-like gene expression. These findings indicated that OmpR has the opposite regulatory effect on ureR-like gene expression in the two tested Y. enterocolitica strains, i.e., positive in Ye9N and negative in Ye8N.

We next tested whether OmpR may influence ure operon expression directly. In silico analysis of the regulatory sequence of ureABC (Figure 7A left panel) identified two 20-bp putative OmpR-binding sites (OBS), with 45% and 40% homology to the E. coli OmpR consensus sequence [TTTTACTTTTTG(A/T)AACATAT] (Maeda et al., 1991; Harlocker et al., 1995). No potential OBS were recognized in either the ureEF or the ureGD regulatory regions.

To determine whether OmpR could modulate ureABC, ureEF and ureGD expression by directly binding to the regulatory regions of these operons, we employed electrophoretic mobility shift assays (EMSAs) (Figure 7A right panel and 7B). DNA fragments containing the regulatory regions of the three urease operons were amplified by PCR and used in EMSAs with increasing concentrations of phosphorylated OmpR-His₆ (OmpR-P). A specific shifted DNA-protein complex was formed by the interaction between the ureABC regulatory region fragment and phosphorylated OmpR (Figure 7A right panel). We observed a stepwise increase in shifted bands of decreasing mobility as the OmpR concentration was increased from 2.47 μM up to 4.95 μM, suggesting that more than one binding site is present. Although no potential OBS were identified in the ureEF promoter sequence, phosphorylated OmpR also interacted with this regulatory region and a slower migrating DNA-protein band appeared at an OmpR-P concentration of 2.47 μM (Figure 7B left panel). As shown in Figure 7B right panel, OmpR-P did not interact with the regulatory region of ureGD, which is consistent with the apparent absence of OBS. OmpR was unable to bind the negative control 16S rDNA fragment included in each binding reaction.

Following the demonstration that OmpR regulates ureR-like gene expression in Y. enterocolitica, we hypothesized that this regulator may do so directly. One potential OBS was identified in the ureR-like gene regulatory region by in silico analysis (Figure 7C left panel). The predicted OBS exhibits high sequence identity (65%) to the consensus OmpR-binding sequence of E. coli. An EMSA was then used to examine the ability of the phosphorylated OmpR regulator to bind to the ureR-like regulatory region in vitro (Figure 7C right panel). This analysis revealed the appearance of specific slower migrating OmpR-DNA complexes concomitantly with the disappearance of the ureR-like regulatory region fragment band. In comparison with the EMSAs performed using the ureABC and ureEF regulatory region fragments (Figures 7A,B), the OmpR-P/ureR-like complexes were formed at a lower OmpR concentration (i.e., at 0.77 μM rather than 2.47 μM) suggesting higher binding affinity of OmpR for the ureR-like target.