Chapter 25:  Regulation of Urease for Acid Habitation

Acid Resistance

Many microorganisms can survive gastric acidity during their transit to other sites of infection. A straightforward means of surviving the high acidity of the stomach would be to restrict the proton permeability of the inner membrane, so that passage of H⁺ across the F₁F₀ ATPase or other H⁺-coupled reactions would occur only at very acidic pH. Alternatively, Na⁺ gradients could be used to drive ATP synthesis and solute import/export as is found in Propionum modestum (5). A means employed by acidophiles is the presence of a proton conductance reversing the potential across the inner membrane in the face of a large inward pH gradient (8). Either relative proton impermeability or high proton conductance would make it difficult for the organism to survive at neutral pH, unless a series of neutral pH responses were present. These would require large changes in the amino acid composition of membrane proteins, for example, by inserting sufficient charged amino acids to elevate or reduce the isoelectric point of the membrane domain of transport proteins; this is an unlikely scenario. However, some modification of the isoelectric point of proton-conducting transporters could modify acid resistance. An example is the Fc subunit of H. pylori that has a significantly higher isoelectric point than the corresponding protein of Escherichia coli or Bacillus subtilis.

Yersinia enterocolitica has a cytoplasmic urease with a pH optimum of 5.0, and the enzyme is inactive at neutral external pH. At acidic pH, with entry of protons into the organism, the cytoplasm acidifies to a pH of 5.0, the urease becomes active, and its activity buffers the cytoplasm by the generation of NH₃ (24). Since protein synthesis is absent in Y. enterocolitica at a cytoplasmic pH of 5.0, this mechanism allows gastric survival but not gastric habitation.

H. pylori synthesizes large quantities of a dimeric urease with optimum activity at neutral pH, which accounts for 10 to 15% of its total protein. This protein synthesis is constitutive (11). In vitro and in vivo, urease, detected by either immunostaining or activity, is found both inside and outside the organism (14). The presence of external urease is likely due to lysis of the bacterium with adhesion of the urease to the outside surface (13). In principle, the activity of either urease or both could account for gastric survival and colonization. Their relative contributions are determined by their respective amounts and the pH of the medium. At acidic pH of 4.5 or less, the outer urease is inactive; therefore, it does not contribute to acid survival.

Urease activity is essential for infection of animal models by H. pylori (1, 6, 20). The experiments carried out to date with urease-negative organisms either show no infection or weak infection when acid secretion is inhibited with omeprazole, a PPI. No experiments have been described wherein infection by this class of mutant was induced in the presence of a PPI, and then disappeared when the PPI treatment was stopped after infection. This type of experiment would show that urease activity was essential not only for infection but also for colonization.

H. pylori urease has a neutral pH optimum between 7.5 and 8.5 but essentially no activity below a pH of 4.5, and activity is lost irreversibly at this pH. Hence, the urease component exposed to the medium on the surface of the organism cannot be responsible for acid resistance at a pH of 4.5 or below, but this urease component could contribute to elevating the pH of the gastric environment of the organism if the pH remained above 4.5. The result of pH elevation due to outer urease activity would be that pH would rise to greater than 8.2 in the absence of acid or buffer, and then this neutralophile would not survive. Thus, urease activity on the surface could be lethal at neutral pH.

Intrabacterial urease shows a marked increase of activity with medium acidification, whether urease activity is measured directly by CO₂ release or by pH elevation using a microphysiometer in contrast to the decrease in activity of surface or soluble urease (Fig. 2A and B) (15, 17). The increase of activity is between 10- and 20-fold when external urease activity is present but can rise much more if surface urease is first washed off or selectively inactivated.

Activation begins at a medium pH of 6.5 and reaches its maximum value at pH 5.5. Urease activity remains constant down to a pH between 2.5 and 3.0 and is detectable even at a pH of 2.0. This behavior corresponds to that of an acid resistance mechanism, wherein the neutralizing capacity is maintained over a broad range of acidic pH and is lost at pH levels where ammonia generation would generate a toxic environment for the organism. Since what is vital for the organism is the PMF (pH/potential gradient) across the inner membrane, uncontrolled internal alkalinization would be unsafe, and the objective of creating a safe environment would be achieved by neutralization of the periplasm. This goal is also economical because the volume of the periplasm, a fraction of a femtoliter, is small compared to the volume of the "microenvironment'' surrounding the organism. A membrane potential of about 100 mV is measured in the presence of urea at medium pH from 3.0 to 6.0, and the calculated or measured periplasmic pH is 6.2 (15, 17). These values of the components of the PMF are compatible with protein synthesis and growth.

The mechanism underlying this pH activation of urease is of vital interest in considering possible selective targets for monotherapy. A clue to understanding this mechanism was gleaned from measurements of the K_m,app for urea in solubilized enzyme and suspensions of intact bacteria at neutral and at acidic pH. At acidic pH the K_m,app for intact bacteria was ~1 mM, similar to the K_m,app for soluble enzyme at neutral pH. However, at neutral pH the K_m,app for urease in intact bacteria is ~200 mM. The apparent absence of saturation at neutral pH in intact organisms suggested that at this pH, the rate-limiting step was no longer urease activity but access of urea to intrabacterial urease. Hence, the observed activation step of urease appears to involve urea transport (17).

Passive urea permeability across phospholipid bi-layers is 4 × 10⁻⁶ cm s⁻¹. Owing to the small volume (~8 fl) and large surface-to-volume ratio of H. pylori, it is difficult to measure urea uptake by bacteria because equilibration is very rapid. To overcome this problem the Xenopus oocyte expression system was employed, since these oocytes have a volume of ~0.5 μl and a much smaller surface-to-volume ratio. Direct evidence for acid-activated urea transport was obtained by oocyte expression of one of the genes of the urease gene cluster, UreI.

There are seven genes in the urease gene cluster; ureA and ureB encode the structural subunits of the enzyme, and ureE, F, G, and H code for proteins involved in urease assembly, although only the gene products of the last three are absolutely essential for generation of activity. Protein-protein interactions between UreE and UreG and UreF and UreH were observed by using a yeast two-hybrid system, and interactions of UreB with UreE and UreH were demonstrated by the same methodology. It appears that two pairs of proteins are formed that bind to UreB, and then assembly and nickel insertion into the UreA/B complex occurs (21).

The open reading frame ureI is located between ureB and ureE and preceded by a promoter region. This gene encodes a product with a sequence predicted to be an inner membrane protein (7). Nonpolar deletion of ureI results in loss of acid activation of intrabacterial urease (16), and acid activation is restored when the mutant is complemented by a vector containing ureI. It is important to note that in the ureI-negative mutant, urease is fully activated when a low concentration of the nonionic detergent C₁₂E₈ is added to permeabilize the bacterial inner membrane (14). The loss of activation in the ureI-negative mutant measured by CO₂ production and the restoration of urease activity with the addition of detergent are shown in Fig. 3. The data obtained with pH measurements are in accordance with the radioactivity assays. In general, nongastric helicobacters, although they have varying amounts of urease activity, do not show acid activation or express UreI (16). Thus, UreI expression appears to be a specialized adaptation within the urease gene cluster for gastric habitation.

The elevation of pH with the addition of urea was visualized with confocal microscopy of bacteria cocultured with AGS cells and BCECF-free acid as a fluorescent pH probe of either the medium or the periplasm. This dye does not penetrate cell or inner bacterial membranes. Addition of urea at a pH of 5.5 to weakly buffered medium results in immediate elevation of pH in the bacterial periplasm, as shown in Fig. 4, followed by elevation of the medium pH in wild-type cells but not in ureI deletion mutants. In contrast, addition of detergent results in initial alkalinization of the cytoplasm in both wild-type and mutant organisms (2).

The Mechanism of UreI

Interesting properties are revealed when the cRNA for UreI is injected into oocytes. First and foremost, the oocytes acquire an acid-activated urea uptake, which is similar to that found for the urease activity in intact bacteria shown in Fig. 2B and C. Urea uptake by oocytes is essentially nonsaturable and highly selective; for example, thiourea is not taken up. Acid-stimulated urea uptake is temperature independent (23). The protein contains six predicted membrane-inserted segments with N and C termini in the periplasm. There are two periplasmic loops containing five or six histidine residues. Removal of any of the last three histidyls, two found in the second periplasmic loop and one in the C terminus, results in inactivation of the protein. These data suggest that UreI is a proton-gated urea channel and that protonation of periplasmic histidines, at least in the second periplasmic loop, is involved in the gating reaction. Interestingly, a homologous gene in Streptococcus salivarius also encodes a protein that transports urea in Xenopus oocytes but does so at neutral and acidic pH. This gene lacks most of the region encoding the periplasmic loops. Cross-linking and sodium dodecyl sulfate gel profiles suggest that the transport protein is an oligomer of at least two monomers, but it is not known whether function requires oligomerization (Weeks, unpublished observations).

These data suggest that urease-dependent adaptation to acidity in vitro depends on the transport function of UreI. UreI deletion mutants are unable to colonize mice, supporting the role of UreI in gastric acid adaptation (18). In a gerbil model, infection and colonization with UreI deletion mutants were achieved when infection was carried out in the presence of inhibitors of the gastric H,K-ATPase, and colonization disappeared when the inhibitor was no longer administered. These results suggest that urease activity is required for both infection and colonization (G. Hanauer and K. Melchers, unpublished observations). A model for the function of UreI is shown in Fig. 5.

Conclusions

H. pylori shows a remarkable acid-resistance response to the gastric environment by a specialized means of utilizing an intracellular, neutral pH optimum urease. Within the urease gene cluster it has a gene encoding an acid-activated urea channel, UreI. Instead of an acidic pH optimum urease suitable only for passage through the stomach, it has a neutral pH optimum urease that allows the cytoplasm to remain relatively neutral while the organism protects itself against acid by enhancing urea access to intrabacterial urease via Urel and achieving buffering of the periplasm. This maintains an adequate PMF for survival and for growth in an otherwise damaging environment.

There are undoubtedly other subtleties involved in the regulation of urease activity inside the organism as well as the NH₃ efflux across the inner membrane to buffer the periplasm. For example, at low rates of intrabacterial urease activity, the NH₃ produced will not result in significant cytoplasmic alkalinization. This is the situation in the absence of acid activation of Urel. However, at a pH where urea transport is fully activated, NH₃ production intracellularly probably increases by more than two orders of magnitude. This rapid production of NH₃ could alkalinize the cytoplasm even in the presence of NH₃ efflux through a bilayer with permeability of 10⁻¹ cm s⁻¹. However, if NH₃ efflux were enabled through the urea channel, the combination of NH₃ efflux across the bilayer and through UreI could maintain intrabacterial pH near neutrality in the face of high intrabacterial urease activity. Further, as soon as the periplasmic pH reaches 6.5, urea entry is switched off, thus inactivating intra-cellular urease.

It would seem possible that this mechanism of H. pylori adaptation to life in the human stomach may form the basis for methods designed to eradicate the organism. Presumably, drugs could be developed targeting the periplasmic domain of UreI and preventing acid activation of urea transport. If the bacterium is exposed, even for a short time, to a pH of less than 4.0, interference with urea uptake at these acidic pH values could prove lethal to the organism.

References