A novel mechanism of urease regulation in Yersinia enterocolitica

Abstract

Yersinia enterocolitica produces the enzyme urease which hydrolyses urea, resulting in the production of carbonic acid and ammonia and a net increase in pH. In the presence of urea, urease enhances survival of Y. enterocolitica in the stomach and presumably in other acidic environments the bacteria encounter during the course of infection. In this study we show that Y. enterocolitica urease is a cytosolic enzyme which has a low K_m value (0.15±0.01 mM urea), suggesting that it functions at close to maximum velocity even at the low concentrations of urea available to Y. enterocolitica in gastric fluid and other tissues. Y. enterocolitica urease was active over a wide pH range, but unlike most other bacterial ureases, displayed an optimal activity at pH 3.5–4.5, suggesting a physiological role in protecting the bacteria from acid. Higher levels of urease activity were attained at 28°C than at 37°C, and investigation of the regulation of urease production revealed that the enzyme was not induced by urea, or by nitrogen limitation. Instead maximal activity was attained during the stationary phase of growth which coincides with the period of maximum acid tolerance of the bacteria. This type of regulation has not been described for any other ureolytic bacteria and seems to be unique to Y. enterocolitica.

Yersinia enterocolitica, Urease, Enzyme regulation

1 Introduction

Before the enteric pathogen, Yersinia enterocolitica, can colonize and penetrate the intestinal mucosa, the bacteria must overcome the gastric acid barrier to infection [1]. We have demonstrated that a functional urease enzyme is essential for Y. enterocolitica to tolerate acidic conditions in vitro and, furthermore, that urease contributes to the ability of the bacterium to survive passage through the stomach [2]. This enzyme hydrolyses urea to form carbonic acid and two molecules of ammonia, leading to a net increase in pH. Accordingly, urease may also play a role in the survival of Y. enterocolitica in other acidic environments the bacteria encounter during the course of infection, such as the phagosomes of polymorphonuclear leucocytes and macrophages (De Koning-Ward and Robins-Browne, unpublished). Apart from a specific role in virulence (reviewed in [3]), urease may also provide some ureolytic bacteria with access to nitrogen for growth in tissues.

Ureases are highly conserved amongst different bacterial species in terms of their primary structure, and all urease enzymes characterized so far require nickel ions for full activity [4]. Nevertheless, ureases from different bacteria differ in terms of their affinity (K_m) for urea, the conditions under which optimal activity is attained, and their mode of regulation. For example, in Proteus mirabilis and Providencia species, urease activity is induced by urea [5, 6], whereas in Klebsiella aerogenes, urease is regulated by the global nitrogen control system, such that under low nitrogen conditions, synthesis of urease is increased [7]. Other ureases are regulated by low pH, while some ureases are unaffected by environmental conditions and are synthesised constitutively (for a review, see [3]).

For this study, we characterized the urease of Y. enterocolitica and investigated the regulation of its biosynthesis. The overall aim of this work was to determine the conditions which favour maximal expression of urease activity in Y. enterocolitica and, thus, gain some understanding of where this enzyme is likely to contribute to bacterial survival in host tissues.

2 Materials and methods

2.1 Bacterial strains

Y. enterocolitica W22703 is a restriction mutant (Res⁻Mod⁺) derived from the wild-type serogroup O:9 strain W227 [8]. Y. enterocolitica 584 is a urease-negative derivative of Y. enterocolitica W22703 which harbours the transposon, TnphoA, within ureC[2]. Unless otherwise indicated, Y. enterocolitica strains were grown in brain heart infusion (BHI) broth at 28°C.

2.2 Urease assays

Y. enterocolitica strains were grown in 100 ml of broth for 18 h with rotation at 200 rpm. Soluble protein (1–100 g), derived from French press lysates of the culture [2], was assayed for urease using a coupled enzyme assay [9]. Urease activity was calculated as μmol of urea hydrolysed min⁻¹ mg of protein⁻¹.

To determine the cellular localization of Y. enterocolitica urease, cells (0.7 g wet weight) from a 300 ml overnight culture were separated into cytosolic, membrane, and osmotic shock fluid fractions [6], the last of which represents the contents of the periplasmic space. Each fraction was then assayed for urease activity as well as for NADH dehydrogenase, β-lactamase [10] and catalase [11], which are known to partition with the membrane, periplasmic and cytosolic fractions, respectively.

2.3 Optimal pH of urease activity

Activity of Y. enterocolitica urease in cell lysates was determined over a pH range of 3.5–9.0, using 20 mM glycine buffer for pH 3.5–4.5, 20 mM citrate-trisodium citrate buffer for pH 3.5–5.8, 10 mM sodium phosphate buffer for pH 5.8–7.8 and 20 mM Tris buffer for pH 7.8–9.0. Urea was included in the reaction mixture at a final concentration of 2 mM; a physiologically relevant concentration which provides V_max conditions (see Section 3.1).

2.4 Effect of availability of nitrogen on urease activity

Y. enterocolitica was grown to stationary phase in either 50 ml of Luria broth (LB) or in 50 ml of ammonia-free 4-morpholinepropanesulfonic acid (MOPS) minimal medium [12] containing as the sole source of nitrogen, 10 mM ammonium chloride, 10 mM glutamine or 0.5% (w/v) casamino acids with 10 mM ammonium chloride. All media were supplemented with 100 μM NiSO₄. Urease activity was then assayed in whole cell lysates of each culture.

2.5 SDS-PAGE and immunoblotting

Proteins from lysates of Y. enterocolitica were spotted directly onto a nitrocellulose membrane. The 19-kDa urease antigen (UreB) was then revealed by immunoblotting as follows: incubation for 1 h with a 1:200 dilution of rabbit antiserum raised against the purified antigen (antiserum kindly supplied by S. Batsford), washing, incubation for a further 1 h with a 1:10.000 dilution of conjugated anti-rabbit immunoglobulin-horseradish peroxidase, and visualisation by enhanced chemiluminescence (Amersham) as specified by the supplier.

3 Results

3.1 Characterization of Y. enterocolitica urease

Rates of urea hydrolysis in cell extracts were measured at 11 different concentrations of substrate, ranging from 0.01 to 2 mM urea (data not shown). Analysis of the results indicated that the V_max and K_m of Y. enterocolitica urease were 0.31±0.02 μmol urea hydrolysed min⁻¹ mg protein⁻¹ and 0.15±0.01 mM urea (mean±standard deviation of three separate determinations), respectively.

Y. enterocolitica urease was active over a wide pH range (pH 3.5–8.5) (Fig. 1). Maximal activity was observed at pH 3.5–4.5, with V_max decreasing under more alkaline conditions such that no activity was observed at a pH ≥9.0. While this is in contrast to most other bacterial ureases, which show optimal activity around neutral pH [3], similar results were obtained when the assays were repeated using 20 mM Tris at pH from 3.5 to 9.0 (data not shown). Although Tris buffered across this wide range of pH produced salt concentrations that were not physiological, these investigations confirmed that urease activity was greatest at an acidic pH.

1

Effect of pH on urease activity of Y. enterocolitica. Cell lysates were assayed for urease activity in buffers ranging from pH 3.5 to 9.0, using 20 mM glycine for pH 3.5–5.8 (●), 10 mM sodium phosphate buffer for pH 5.8–7.8 (○), and 20 mM Tris buffer for pH 7.8–9.0 (▲). The values are the mean of at least two independent assays.

Even though maximal urease activity was observed at low pH, subsequent assays were performed in Tris buffered at pH 8.0 because (i) the enzyme glutamate dehydrogenase used in the coupled enzyme assay for urease activity has a pH optimum of 8.0, and (ii) the intracellular pH of Y. enterocolitica is likely to approximate 8 owing to its decarboxylase [13]. Therefore, pH 8.0 is probably physiologically more relevant for this enzyme than acidic pH.

Urease was found to be localised in the cytosol, with no detectable activity in the supernatant (data not shown), membrane or osmotic fluid fractions (Table 1). Since these results suggested that urease is not exported extracellularly but rather that urea must be able to gain access to the cytoplasm, we investigated if intact cells offer a permeability barrier to urea by measuring urease activity in whole cells and an equivalent sample ruptured by two passes through a French pressure cell. Although urease activity in whole cells was approximately one-third that of the cell lysate (0.115 compared to 0.313 μmol urea hydrolysed min⁻¹ mg protein⁻¹), transport rates were respectable, indicating that at physiological concentrations of urea there is relatively free access of urea to the cytosol and of ammonia from the cytosol into the surrounding medium.

1

Cellular localization of urease of Y. enterocolitica.

Enzyme assayed	Enzyme activitya in fraction comprising:
Cytosol	Membrane	Periplasm
Urease	0.215	<0.001b	<0.001b
Catalase	5150	191	338
NADH
dehydrogenase	<0.001b	0.559	<0.001b
β-Lactamase	0.053	0.114	7.34

^aEnzyme activity is given as μmol substrate hydrolysed min⁻¹ mg protein⁻¹.^bValues below detectable levels.

3.2 Regulation of urease activity

Given that the ability of Y. enterocolitica to tolerate acidic conditions is dependent on urease activity [2] and that acid resistance is maximal during the stationary phase of growth, we investigated whether urease activity is affected by the growth phase of the bacteria. Fig. 2A demonstrates that expression of urease is dependent on growth phase, with activity increasing during late exponential phase (>16 h) and becoming maximal during stationary phase.

2

Effect of urea and growth temperature on urease activity. Cells from an overnight culture of Y. enterocolitica W22703 were diluted 1 in 1000 and grown in BHI broth with shaking at (A) 28°C or (B) 37°C in the presence or absence of 16.7 mM urea. Aliquots were withdrawn at intervals and the viable count and urease activity (starting at 12 h) were determined. Values are the mean of two independent experiments.

Urease activity was also examined at two growth temperatures: 28°C and 37°C, the former representing the optimal growth temperature of Y. enterocolitica, the latter the temperature in host tissues. Maximal urease activity at 28°C was 0.36 μmol urea hydrolysed min⁻¹ mg protein⁻¹ (Fig. 2A), whereas at 37°C, maximal activity was only 0.14 μmol urea hydrolysed min⁻¹ mg protein⁻¹ (Fig. 2B). Similar results were obtained with Y. enterocolitica strain W22703c, the plasmid cured derivative of W22703 (data not shown), indicating that regulation of urease was not influenced by the virulence plasmid, pYV.

Although the amount of urease produced at 37°C was less than that at 28°C, expression of urease activity was growth-phase dependent at both temperatures (Fig. 2B). Further experiments demonstrated that urease activity in Y. enterocolitica was not regulated by urea at either temperature (Fig. 2A and B).

As ammonia can serve as a nitrogen source for many species of bacteria, including Y. enterocolitica (unpublished data), and urease is required for the production of ammonia from urea, we also measured urease activity in Y. enterocolitica cultured under various nitrogen-limiting conditions to determine if urease expression is regulated by the availability of nitrogen. The media used were Luria broth (nitrogen rich) MOPS+casamino acids+NH₄Cl>MOPS+NH₄Cl>MOPS+glutamine (nitrogen limiting). Because preliminary investigations showed that low nickel concentrations (or unavailability of nickel in nitrogen-rich media) were rate-limiting for urease activity, all media were supplemented with 100 μM nickel, a concentration which generated maximal urease activity in Y. enterocolitica grown under nitrogen-rich and nitrogen-limiting conditions (data not shown). Regardless of the medium in which Y. enterocolitica was cultured, similar levels of urease activity were attained (data not shown), indicating that urease activity in Y. enterocolitica is not regulated by nitrogen limitation. These findings were confirmed by immunoblotting of lysates of Y. enterocolitica grown in nitrogen-rich and nitrogen-depleted media, with a rabbit polyclonal antibody to the purified 19-kDa subunit (UreB) of Y. enterocolitica urease (Fig. 3).

3

Effect of nitrogen availability on urease expression by Y. enterocolitica. Y. enterocolitica W22703 (lanes A–D) or Y. enterocolitica 584 (lane E) were grown in: lanes A and E, Luria broth; B, MOPS minimal medium containing 0.5% (w/v) casamino acids and 10 mM NH₄Cl; C, MOPS medium containing 10 mM NH₄Cl; D, MOPS medium containing 10 mM glutamine. Whole cell proteins were serially diluted and spotted directly onto a nitrocellulose membrane. Urease was revealed by incubation with rabbit antiserum raised against purified UreB subunit, washing, and then a further incubation with anti-rabbit immunoglobulin conjugated to horseradish peroxidase. The immunoblot was visualized by enhanced chemiluminescence. The resultant autoradiograph was scanned with a digital scanner and visualized by using Paint Shop Pro version 3.0 (JASC, Inc., Minnetonka, MN, USA).

4 Discussion

This study has shown that the urease of Y. enterocolitica has one of the lowest K_m values of all known bacterial ureases, being 0.15±0.01 mM urea. Only Helicobacter pylori urease has a comparable K_m value (0.17 mM) [14]. Interestingly, both of these pathogens have access to far lower concentrations of urea (approx. 1.4 mM) in the intestinal tract than urinary pathogens, such as Proteus mirabilis and Providencia stuartii, which are exposed to high concentrations of urea in the urinary tract and whose enzymes have far higher K_m values: 13 mM [15] and 9.3 mM urea [16], respectively. Our findings suggest that even at the relatively low concentrations of urea available to Y. enterocolitica in the stomach and other tissues, its urease enzyme can function at near maximal velocity.

The urease of Y. enterocolitica also displays optimal activity at a far lower pH than most other bacterial ureases. The only other known bacterial species which possess acid ureases are Lactobacillus fermentum, L. reuteri and Streptococcus mitior[17], which unlike Y. enterocolitica, are acidophiles. However, Y. enterocolitica may encounter extreme acid conditions, such as during passage through the stomach. Under these circumstances, an increase in urease production would act as a protective mechanism whereby the bacteria could quickly neutralise hydrogen ions which penetrate the bacterial cell wall to maintain its intracellular pH. Whether urease plays any role in more neutral environments, however, remains unclear.

The typical modes of urease regulation in other bacterial species, namely induction by urea or nitrogen limitation, were not observed in Y. enterocolitica. This is consistent with (i) the lack of ureR in the urease gene complex of Y. enterocolitica which is located upstream of ureA in gene complexes that are inducible by urea [18], and (ii) the absence of a detectable σ⁵⁴ promoter sequence upstream of ureA which governs the transcription of many nitrogen-regulated genes [19]. Instead, urease activity in Y. enterocolitica was regulated by the growth phase of the bacteria, with maximal activity during the stationary phase of growth. This coincides with the phase when Y. enterocolitica displays maximal resistance to acidic pH [2]. The finding that the acid resistance of Y. enterocolitica is also dependent on the concentration of urea adds further weight to the notion that urease activity and acid tolerance are linked.

Although growth phase dependency of urease activity has not been described for other ureolytic bacteria, several other virulence-associated proteins of Y. enterocolitica, including Yst enterotoxin, Inv, Ail and Myf, are also regulated by the stage of growth of the bacterium [20]. Some of these stationary-phase genes are regulated by an alternative sigma factor, RpoS, but an rpoS mutant of Y. enterocolitica produced similar levels of urease during the stationary phase of growth to its parent strain (data not shown), indicating this sigma factor is not involved in the regulation of Y. enterocolitica urease.

Urease activity was greater in Y. enterocolitica grown at 28°C than at 37°C. Other virulence-associated proteins that are maximally expressed when Y. enterocolitica is cultured at 28°C include invasin and Yst [21]. Given that urease activity is maximal at environmental temperatures and during stationary phase, it appears that the conditions for optimal enzyme activity would be met during growth outside the host, since nutrients are likely to be limiting and the bacteria would be in stationary phase. Passage of Y. enterocolitica through the stomach would generate even higher urease activity, enhancing the survival of Y. enterocolitica in gastric acid. If urease also contributes to the virulence of Y. enterocolitica at other stages of infection, it must also be expressed in sufficient amounts at 37°C. In this regard, Mikulskis et al. [22] have shown that Yst expression can be induced at 37°C by increasing the osmolarity and pH of the culture medium to values generally found in the ileal lumen where Yst is postulated to exert its effect. It would be interesting to determine if urease synthesis can also be increased at 37°C using culture media that more accurately reflect the conditions which Y. enterocolitica encounters in vivo.

Acknowledgements

We are indebted to Dr. M. Iriarte and Dr. G. Cornelis for the gift of the rpoS mutant of Y. enterocolitica, to Dr. S. Batsford for providing the polyclonal UreB antibody, to Dr. D. Adams for his assistance with the urease assays and to Dr A. Ward for rewarding discussions. This work was supported in part by grants from the Australian National Health and Medical Research Council and the Australian Research Council.