Joanne V. Gilbert

Neisseria gonorrhoeae and neisseria meningitidis: extracellular enzyme cleaves human immunoglobulin A.

The gonococcus and meningococcus, which infect human mucosal surfaces, elaborate a highly specific proteolytic enzyme which cleaves the immunoglobulin A1 subclass of the principal mucosal antibody, immunoglobulin A (IgA). The susceptible Pro-Thr bond lies in a unique region of the IgA heavy chain; the IgA2 subclass, lacking this peptide bond, is enzyme resistant.

The effect of an IgA1 protease on immunoglobulins bound to the sperm surface and sperm cervical mucus penetrating ability.

A major site of impaired fertility in men with autoimmunity to sperm rests at the level of restricted sperm entry and motion within cervical mucus. We studied the effects of a protease derived from Neisseria gonorrhoeae, whose substrate specificity is limited to human IgA1, on the ability of antibody-bound sperm to penetrate human cervical mucus in vitro. IgA on the sperm surface, but not IgG, was degraded by IgA1 protease. A correlation was seen between the levels of IgA bound relative to IgG and the improvement in sperm cervical mucus penetrating ability after IgA1 protease exposure. These results provide evidence that antisperm autoantibodies of both IgA and IgG classes impair the ability of spermatozoa to populate the female reproductive tract. They implicate the Fc region of the immunoglobulin molecule in mediating this effect and offer the potential to restore male fertility by treating antibody-bound sperm in vitro with immunoglobulin-directed bacterial proteases, before insemination.

Inhibition of microbial IgA proteases by human secretory IgA and serum

Microbial IgA proteases cleave human serum IgA1 immunoglobulin, but human secretory IgA is resistant to hydrolysis. We have found this resistance to be due to an inhibition of protease activity that is mediated by the Fab region of secretory IgA. The IgA proteases of the genus Neisseria are more sensitive to inhibition than is the protease of Streptococcus sanguis. There is also a serum inhibitor of Neisseria proteases that co-chromatographs with IgG. Monoclonal (myeloma) human IgG proteins and plasma protease inhibitors such as alpha-1-antitrypsin and alpha-2-macroglobulin do not inhibit. Human sera do not contain inhibitor to S. sanguis protease activity. We conclude that microbial IgA proteases are subject to inhibition by IgA in secretions and IgG in serum, and this activity is most consistent with being an anti-enzyme antibody. The insensitivity of S. sanguis IgA protease to inhibition is unexplained but provides further evidence that the IgA proteases are structurally diverse.

Role of Hpn and NixA of Helicobacter pylori in Susceptibility and Resistance to Bismuth and Other Metal Ions

Background. Helicobacter pylori produces Hpn, a 60‐amino acid, histidine‐rich protein that avidly binds nickel and zinc ions, and NixA, a high‐affinity nickel transporter in the cytoplasmic membrane. We tested the hypothesis that Hpn and NixA govern susceptibility to metal ions in H. pylori.

In Vivo Complementation of ureB Restores the Ability of Helicobacter pylori To Colonize

ABSTRACT The objective of this study was to determine (i) if complementation of ureB-negative Helicobacter pylori restores colonization and (ii) if urease is a useful reporter for promoter activity in vivo. Strains used were M6, M6ΔureB, and 10 recombinant derivatives of M6 or M6ΔureB in which urease expression was under the control of different H. pylori promoters. Mice were orally inoculated with either the wild type or one of the mutant strains, and colonization, in vivo urease activity, and extent of gastritis were determined. Of eight M6ΔureB recombinants tested, four colonized mice. Of those, three had the highest in vitro urease activity of any of the recombinants, significantly different from that of the noncolonizing mutants. The fourth colonizing recombinant, with ureB under control of the cag-15 promoter, had in vitro urease activity which did not differ significantly from the noncolonizing strains. In vivo, urease activities of the four colonizing transformants and the wild-type control were indistinguishable. There were no differences in gastritis or epithelial lesions between mice infected with M6 and those infected with the transformants. These results demonstrate that recovery of urease activity can restore colonizing ability to urease-negative H. pylori. They also suggest that cag-15 is upregulated in vivo, as was previously suggested by demonstrating that it is upregulated upon contact with epithelial cells. Finally, our results suggest that total urease activity and colonization density do not contribute to gastritis due to H. pylori.

Differential Gene Expression from Two Transcriptional Units in the cag Pathogenicity Island of Helicobacter pylori

ABSTRACT Infection with Helicobacter pylori strains containing the cag Pathogenicity Island (cag PAI) is strongly correlated with the development of severe gastric disease, including gastric and duodenal ulceration, mucosa-associated lymphoid tissue lymphoma, and gastric carcinoma. Although in vitro studies have demonstrated that the expression of genes within the cag PAI leads to the activation of a strong host inflammatory response, the functions of mostcag gene products and how they work in concert to promote an immunological response are unknown. We developed a transcriptional reporter that utilizes urease activity and in which nine putative regulatory sequences from the cag PAI were fused to theH. pylori ureB gene. These fusions were introduced in single copies onto the H. pylori chromosome without disruption of the cag PAI. Our analysis indicated that while each regulatory region confers a reproducible amount of promoter activity under laboratory conditions, they differ widely in levels of expression. Transcription initiating upstream of cag15 and upstream of cag21 is induced when the respective fusion strains are cocultured with an epithelial cell monolayer. Results of mouse colonization experiments with an H. pylori strain carrying the cag15-ureB fusion suggested that this putative regulatory region appears to be induced in vivo, demonstrating the importance of the urease reporter as a significant development toward identifying in vivo-induced gene expression in H. pylori.

Assay and Properties of IgA Protease of Streptococcus Sanguis

An assay procedure for streptococcal IgA protease is described which uses isotopically labelled human serum IgA as substrate. Enzyme activity was monitored by the radioactive counts in the Fab alpha product, which was separated from other components in the digestion mixture by electrophoresis. Cleavage of IgA was linear with respect to time using catalytic amounts of the enzyme. Km was calculated to be 5.5 X 10(-6)M, pH optimum 6.0-7.0 at 37 degrees C, and the enzyme was fully inactivated at low concentrations of the metal chelator ethylenediaminetetraacetic acid.

Detection of IgA protease activity among multiple bacterial colonies

Abstract For genetic and screening studies of IgA proteases a technique is described for identifying the enzyme in unlimited numbers of bacterial colonies on agar plates. Human 125 I-labeled IgA (the only known substrate) bound to anti-IgA polyacrylamide beads is suspended in agar, and enzyme detected by autoradiography of released radioactive Fab α fragments.

IgA1 protease cleaves heavy chains independently in dimeric human IgA1.

Bacterial IgA1 proteases have substrate specificity for human IgA1 immunoglobulin, and cleave both the heavy (alpha) chains where they are paired by disulfide bonds in the hinge region. To determine if the close apposition of the alpha chains allows a single enzyme-substrate-binding event to cleave both hinge region peptides we quantitated the relative levels of intermediate products during the course of complete hydrolysis of an IgA1 paraprotein. The substrate had four Fab regions, analogous to a secretory IgA dimer. The experimental data were then compared to computer-generated models in which various levels of cooperativity among Fab regions were tested. The results most closely conformed to a model in which each individual alpha chain is proteolyzed independently, without regard to the total number of hinge region peptides available in the substrate IgA1. These results will be used to guide the design of IgA1 hinge region peptide analogues as IgA1 protease inhibitors.