Search

The pathogenesis of infections caused by Neisseria gonorrhoeae, the causative agent of the sexually transmitted disease gonorrhea, can be studied using experimental infection of human male volunteers. The desire to avoid introducing new antibiotic resistance markers into strains to be used in human experimental infection has complicated the construction of genetically defined mutants in which expression of potential virulence factors is inactivated. To facilitate construction of such mutants, we have used a two-step mutagenesis strategy that allows for gene replacements without introducing new selectable markers into the final strain. The method uses a two-gene cassette containing both a selectable marker (ermC') and a counterselectable marker (rpsL). The cassette is cloned into the gene of interest and used to replace the wild-type gene on the chromosome by allelic exchange. A second transformation replaces the cassette-containing version of the gene with an engineered version with an unmarked deletion or other mutation. The rpsL gene of Escherichia coli functioned for the counterselection in the gonococcus, albeit with low efficiency. To improve the efficiency of the counterselection, we cloned the gonococcal rpsL gene and incorporated it into the cassette. This technique has been successful in creating defined mutants for human challenge, and also circumvents the limitation in the number of different selectable markers that are useful in Neisseria species.

Many mucosal pathogens, including Neisseria gonorrhoeae, produce proteases that cleave immunoglobulin A (IgA), the predominant immunoglobulin class produced at mucosal surfaces. While considerable circumstantial evidence suggests that IgA1 protease contributes to gonococcal virulence, there is no direct evidence that N. gonorrhoeae requires IgA1 protease activity to infect a human host. We constructed a N. gonorrhoeae iga mutant without introducing new antibiotic resistance markers into the final mutant strain and used human experimental infection to test the ability of the mutant to colonize the male urethra and to cause gonococcal urethritis. Four of the five male volunteers inoculated with the Iga- mutant became infected. In every respect-clinical signs and symptoms, incubation period between inoculation and infection, and the proportion of volunteers infected-the outcome of human experimental infection with FA1090iga was indistinguishable from that previously reported for a variant of parent strain FA1090 matching the mutant in expression of Opa proteins, lipooligosaccharide, and pilin. These results indicate that N. gonorrhoeae does not require IgA1 protease production to cause experimental urethritis in males.

Iron, an essential nutrient for most microorganisms, is sequestered by the host to decrease the concentration of iron available to bacterial pathogens. Neisseria gonorrhoeae, the causative agent of gonorrhoea, can acquire iron by direct interaction with human iron-binding proteins, including the serum glycoprotein, transferrin. Iron internalization from host transferrin requires the expression of a bacterial receptor, which specifically recognizes the human form of transferrin. Two gonococcal transferrin-binding proteins have been implicated in transferrin receptor function, TbpA and TbpB. We constructed a gonococcal transferrin receptor mutant without the introduction of additional antibiotic resistance markers and tested its ability to cause experimental urethritis in human male volunteers. The transferrin receptor mutant was incapable of initiating urethritis, although the same inoculum size of the wild-type parent strain, FA1090, causes urethritis in >90% of inoculated volunteers. To our knowledge, this is the first experimental demonstration that a bacterial iron acquisition system is an essential virulence factor for human infection.

A physical map of the chromosome of N. meningitidis Z2491 (serogroup A, subgroup IV-1) has been constructed. Z2491 DNA was digested with NheI, SpeI, SgfI, PacI, BglII, or PmeI, resulting in a limited number of fragments that were resolved by contour-clamped homogeneous electric field (CHEF) electrophoresis. The estimated genome size for this strain was 2,226 kb. To construct the map, probes corresponding to single-copy genes or sequences were used on Southern blots of chromosomal DNA digested with the different mapping enzymes and subjected to CHEF electrophoresis. By determining which fragments from different digests hybridized to each specific probe, it was possible to walk back and forth between digests to form a circular macrorestriction map. The intervals between mapped restriction sites range from 10 to 143 kb in size. A total of 117 markers have been placed on the map; 75 represent identified genes, with the remaining markers defined by anonymous cloned fragments of neisserial DNA. Comparison of the arrangement of genetic loci in Z2491 with that in gonococcal strain FA1090, for which a physical map was previously constructed, revealed complex genomic rearrangements between the two strains. Although gene order is generally conserved over much of the chromosome, a region of approximately 500 kb shows translocation and/or inversion of multiple blocks of markers between the two strains. Even within the relatively conserved portions of the maps, several genetic markers are in different positions in Z2491 and FA1090.

Gonococcal pilin variation is thought to allow immune evasion and change the adherence properties of the pilus. We have examined the process of pilin antigenic variation in human volunteers inoculated with strain FA1090. Our data show that pilin variation occurred throughout the process of infection, that at each time sampled after inoculation multiple pilin variants were present, and that later pilin variants appear to be recombinants between previously expressed genes and the silent storage pilin copies. Thus, during infection a large repertoire of proteins are available to the population to help avoid immune responses, to provide pili with varying functions, and to transmit to a new host.

To increase the utility of the previously constructed physical map of the chromosome of Neisseria gonorrhoeae FA1090, 28 additional genetic markers were localized on the map. Cloned gonococcal genes were used to probe Southern blots of restriction enzyme-digested DNA separated on pulsed-field gels, thus identifying the fragment in each of several digests to which the probe hybridized and the map location of each gene. The addition of the new markers brings the total number of mapped loci for this strain to 68; the locations of all of those markers on the updated map are shown.

The opacity (Opa) proteins of Neisseria gonorrhoeae are a family of outer membrane proteins demonstrating phase and antigenic variation. N. gonorrhoeae strain FA0190 has 11 opa loci that encode at least 8 antigenically distinct Opa proteins. To determine if expression of one Opa protein or a subset of them is favored during gonococcal infection, we inoculated Opa-negative variants of strain FA1090 intraurethrally into male volunteers. The Opa phenotype of gonococci isolated from urine and urethral swab cultures from nine infected subjects was determined. Opa proteins were expressed in a large proportion of the reisolates from the infected subjects. Gonococci cultured from urine or urethral swab samples from six of the subjects were uniformly Opa positive, with the predominant Opa variants differing among subjects. Three different Opa proteins were represented as the predominant type in at least one subject each. In three subjects, there was more heterogeneity in Opa phenotype of the reisolates, including the presence of Opa-negative variants. An increase in the proportion of isolates expressing multiple Opa proteins occurred over time in most subjects. Passage of the inoculum in vitro did not result in similar changes in Opa expression. There was no detectable difference in infectivity of an Opa-negative variant and one expressing an Opa protein (OpaF) that was highly represented in reisolates from the original nine subjects. Reisolates from three infected volunteers inoculated with the OpaF variant showed continued expression of OpaF alone or in conjunction with other Opa proteins. These results demonstrate that there is strong selection for expression of one or more Opa proteins by strain FA1090 in vivo, but that no single protein is preferentially expressed during early infection in the male urethra.

PCR amplification and DNA sequencing of the expression locus from Neisseria gonorrhoeae contained in urine sediments collected from experimentally infected human subjects produced two observations. First, different pilin sequences were obtained when separate aliquots of the same sample were amplified and sequenced. In contrast, the same pilin sequence was obtained when repeated amplifications were performed on individual colonies grown from the clinical samples. Second, mixed sequences (i.e., more than one nucleotide at variable positions in the pilin gene sequence) were observed in both the direct clinical isolates and individual cultures grown from the isolates. These results suggest that when clinical samples are directly examined by PCR amplification and sequencing, multiple amplifications may be required to detect sequence variants in the sample and minority variant sequences will not always be detected.

A physical map of the chromosome of Neisseria gonorrhoeae FA1090 has been constructed. Digestion of strain FA1090 DNA with NheI, SpeI, BglII, or PacI resulted in a limited number of fragments that were resolved by contour-clamped homogeneous electric field electrophoresis. The estimated genome size was 2,219 kb. To construct the map, probes corresponding to single-copy chromosomal sequences were used in Southern blots of digested DNA separated on pulsed-field gels, to determine how the fragments from different digests overlapped. Some of the probes represented identified gonococcal genes, whereas others were anonymous cloned fragments of strain FA1090 DNA. By using this approach, a macrorestriction map of the strain FA1090 chromosome was assembled, and the locations of various genetic markers on the map were determined. Once the map was completed, the repeated gene families encoding Opa and pilin proteins were mapped. The 11 opa loci of strain FA1090 were distributed over approximately 60% of the chromosome. The pil loci were more clustered and were located in two regions separated by approximately one-fourth of the chromosome.

P.II outer membrane proteins of Neisseria gonorrhoeae are encoded by a family of closely related genes. Although the genes are highly conserved, major differences in sequence among them occur in two short regions, designated hypervariable regions 1 (HV1) and 2 (HV2). In this study, we determined the number and DNA sequence of the hypervariable regions in the P.II genes of strains FA1090. The FA1090 chromosome contained at least eleven P.II loci, having six different versions each of HV1 and HV2 among them. Southern blotting with HV-specific oligonucleotides showed that each version was present in one to three copies, and that there were nine unique combinations of HV1 and HV2 in the P.II genes. Although each of the versions of HV1 or HV2 had a unique DNA sequence, there were some similarities among them, particularly when certain ones were compared. Restriction fragments containing only the HV regions were cloned into an expression vector to demonstrate that the epitopes recognized by a set of monoclonal antibodies specific for different FA1090 P.II proteins were completely encoded by either HV1 or HV2.

This study showed that a protein II (PII) of Neisseria gonorrhoeae FA1090 appeared to act as a mediator of attachment to HeLa cells. Two colony variants of FA1090 were selected. Both gonococcal variants were nonpiliated, but one contained a PII and the other did not. A monoclonal antibody (1090-10.1), which was directed against the PII, inhibited the apparent PII-mediated attachment to HeLa cells. Antibodies produced from clone 1035-4, which had no PII specificity, did not inhibit the attachment and were used as controls. Inhibition of gonococcal attachment by the 1090-10.1 monoclonal antibodies was demonstrated by fluorescent microscopy analysis. Monoclonal antibody 1090-10.1 appeared to cause agglutination of the PII-containing organism. To block the clumping caused by the PII-specific monoclonal antibodies, Fab fragments of goat anti-mouse IgG were incubated with gonococci and the 1090-10.1 monoclonal antibodies. The results showed that the goat anti-mouse IgG Fab fragments partially blocked the agglutination caused by the PII-specific monoclonal antibody. The effect of the 1090-10.1 antibodies on attachment was also determined by monitoring the HeLa cells with attached iodinated gonococci. The monoclonal antibody appeared to inhibit the PII-mediated attachment.

Gonococci isolated from localized urogenital infections usually possess one or more protein II (P.II) species in the outer membrane, but the structural and antigenic variation of these proteins among different gonococcal strains has made it difficult to determine if specific proteins of the P.II class are associated with particular sites or types of infection. A recent outbreak of gonorrhea in Durham, N.C., was unusual in that over 200 isolates were derived from a single strain, which provided an opportunity to address these questions. The P.II profile of 54 isolates was determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of outer membrane proteins. At least seven distinct P.II proteins were seen; no single protein or combination of proteins predominated in the different isolates, and there was no association of P.II profile with site of isolation. Gonococci recovered from the same patient at different times had different P.II profiles, confirming that P.II variation occurred in vivo.

An increase in the apparent molecular weight of the principal outer membrane protein (POMP) of Neisseria gonorrhoeae is associated with introduction of the penB2 genetic marker, which results in low-level, relatively nonspecific antibiotic resistance. Limited proteolysis of the two forms of POMP showed that they had few if any peptides in common. The nonspecific antibiotic resistance of penB2 was separated from the change in POMP by genetic transformation and by isolation of spontaneous penB mutants that showed no change in POMP. The genetic locus involved in the change from one POMP to another, which we have designated nmp-1, is closely linked to, but not identical with, penB2.

Expression of Neisseria gonorrhoeae Protein II (P.II) is subject to phase variation and antigenic variation. The P.II proteins made by one strain possess both unique and conserved antigenic determinants. To study the mechanism of antigenic variation, we cloned several P.II genes, using as probes a panel of monoclonal antibodies (MAbs) specific for unique determinants. The DNA sequences of three P.II genes showed that they shared a conserved framework, with two short hypervariable (HV) regions being responsible for most of the differences among them. We demonstrated that unique epitopes recognized by the MAbs were at least partially encoded by one of the HV regions. Moreover, we found that reassortment of the two HV regions among P.II genes occurs, generating increased structural and antigenic variability in the P.II protein family.

Antigenic diversity among and within strains of Neisseria gonorrhoeae and Neisseria meningitidis has complicated studies of the pathogenesis of these strains and obstructed vaccine development. We previously described a distinct surface antigen (H.8) common to pathogenic Neisseria. We have now demonstrated in vivo expression of the H.8 antigen by detecting antibody responses to the antigen in 13 patients with disseminated neisserial infections. Each serum sample from a convalescent patient blocked the binding between the infecting meningococcal or gonococcal strain and a monoclonal antibody directed to the H.8 antigen, as demonstrated by binding-inhibition studies in enzyme-linked immunosorbent assays (P less than .005). Testing by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and western blotting demonstrated an IgG response in each convalescent serum to an antigen co-migrating with the H.8 antigen. Specificity of this antibody response was confirmed by probing recombinant bacteriophage that expressed the H.8 antigen. The commonality and the immunogenicity of the H.8 antigen indicate its possible role in the pathogenesis of, and its potential as a vaccine component for, gonococcal and meningococcal diseases.

Previous work indicates that strains of Neisseria gonorrhoeae isolated in Seattle, Wash., and Atlanta, Ga., show an association between serotypes 1 and 2 of protein I of the outer membrane and the ability to cause disseminated infection (T.M. Buchanan and J.F. Hildebrandt, Infect. Immun. 32:985-994, 1981). By using the same serotyping system, we confirmed the association between those serotypes and both disseminated infection and serum resistance in strains from North Carolina. Some strains of the same serotype had protein I species with different apparent molecular weights.

The protein II (P.II) outer membrane proteins of Neisseria gonorrhoeae, which have been implicated in gonococcal pathogenesis, have been previously shown to undergo a type of phase variation in which expression of any of several different forms of the proteins may be switched on or off. We identified six electrophoretically distinct forms of P.II proteins (designated P.IIa through P.IIf) within strain FA1090, and we isolated colonial variants of FA1090 that expressed only one of the six different P.II protein forms. Two monoclonal antibodies that bound specifically and differentially to P.II proteins were produced. One antibody bound to proteins P.IIb and P.IId and was bactericidal for all colonial variants expressing P.IIb. The second antibody bound to P.IIa and was bactericidal for colonial variants expressing P.IIa. P.II protein profiles of survivors of antibody killing indicated that multiple P.II protein species may be expressed on a single bacterium and that P.II protein switching in the gonococcus is nonrandom.

Despite the inherent limitations imposed by working with an organism that still has a very limited genetic map and relatively few systems for manipulation of the genome, much has been learned in the past decade. Recent applications of the technology of monoclonal antibodies and recombinant DNA, coupled with careful studies of the immunobiology of pili and various outer membrane structures have provided exciting insights into the molecular pathogenesis of gonococcal infections. One of the principal lessons is the sophistication of the gonococcal strategies for evading the host defenses, including high-frequency variations of pili and P.II and an extracellular protease for specific cleavage of IgA1. It is unclear how much antigenic variation will limit vaccine development. Regardless, the pursuit of a vaccine has led to important new fundamental knowledge of the genetics and structure-function relationships of several cell surface components, and the future promises to be both rational and interesting.

Neisseria gonorrhoeae DNA that encodes the pathogenic Neisseria H.8 common antigen was cloned in the lambda phage sep6. The recombinant phage, designated s6H.8, was detected immunologically with a monoclonal antibody that binds to the H.8 antigen. The gonococcal and s6H.8 forms of the antigen yielded identical partial proteolysis epitope maps. Neisseria species that did not manifest the H.8 antigen showed little or no DNA homology with s6H.8. This clone should facilitate investigation into the clinical utility of this antigen and determination of its possible function in gonococcal pathogenesis.

To determine the ultrastructural distribution and immunological accessibility of proteins II (P.IIs) on the surfaces of whole gonococci, anti-P.II gold probes were developed and used in electron microscopic studies of viable P.II-expressing variants of Neisseria gonorrhoeae FA1090. Anti-P.II probes clearly marked the surfaces of organisms and their associated outer membrane blebs. The surface-exposed portion of P.II is antigenically variable. With the use of two different sizes of gold probes, it was demonstrated that individual gonococcal cells can express more than one antigenic type of P.II simultaneously.

Hybridomas secreting monoclonal antibodies against an apparent strain-specific cell surface antigen of Neisseria gonorrhoeae were produced. Spleen cells from BALB/c mice immunized with whole gonococci were fused with mouse myeloma cell line Sp2/0, and hybrid cells were selected in culture. One hybridoma that secreted antibodies reactive with the immunizing strain was cloned by limiting dilution to obtain cell lines secreting monoclonal antibodies. These antibodies reacted with purified outer membranes from the immunizing strain as well as with whole gonococci. Binding of antibodies to whole gonococci was highly strain specific, with most gonococcal strains showing less than 1% of the binding with the immunizing strain. Antibodies did not bind to the other Neisseria species tested. Binding of monoclonal antibodies to whole gonococci of the immunizing strain was not dependent on state of piliation. The extent of antibody binding did vary in different colonial variants of the immunizing strain. Antibody bound to cells from colonies that were transparent or of intermediate opacity, but did not bind to cells from deeply opaque colony variants.

We compared the structure of P.II proteins of gonococcal strain FA1090 by N-terminal sequence analysis of purified proteins and by DNA sequencing of cloned P.II genes. Regulation of P.II gene expression does not involve major DNA rearrangements, but may involve generation of frame-shifts in unexpressed P.II genes. There are probably 8 or 9 P.II genes, each possessing a common leader sequence, in the gonococcal chromosome.

Neisseria meningitidis class 5 proteins are major outer membrane proteins that share many properties with the proteins II (P.II) of Neisseria gonorrhoeae. We generated two bactericidal monoclonal antibodies, each of which bound specifically to one of the two identified class 5 proteins produced by N. meningitidis FAM18. The monoclonal antibodies also bound to class 5 proteins of a limited number of other meningococcal strains. Using the bactericidal activity of the monoclonal antibodies, we demonstrated that expression of both class 5 proteins was subject to reversible phase variation in vitro. The N-terminal amino acid sequence of a purified class 5 protein revealed striking similarity to the N-terminal amino acid sequence of gonococcal P.II proteins. Using a cloned class 5 gene, we identified three potential class 5 gene loci in N. meningitidis FAM18. These class 5 sequences also had homology with gonococcal P.II gene sequences and contained the CTCTT repeat sequence believed to be important in the regulation of gonococcal P.II expression.

Expression of outer membrane protein II (P.II) of Neisseria gonorrhoeae is subject to reversible phase variation at a rate of 10(-3)-10(-4)/cell/generation. The signal peptide coding regions of P.II genes contain variable numbers of tandem repeats of the sequence CTCTT. Changes in the number of CTCTT units, leading to frameshifting within the gene, are responsible for changes in P.II expression. Phase variation mediated by the CTCTT repeat also occurred in E. coli, as assayed with a P.II-alkaline phosphatase (phoA) gene fusion. Phase variation in both the gonococcus and E. coli was recA-independent, occurred at similar rates, and involved insertions or deletions of one or more repeat units. The characteristics of the phase variation process were consistent with a model in which expression of P.II genes is regulated by slipped-strand mispairing of the DNA in the CTCTT repeat region.

Gonococci express a family of related outer membrane proteins designated protein II (P.II), which undergo both phase and antigenic variation. Six P.II proteins have been identified in strain FA1090. We developed monoclonal antibodies specific for each P.II protein. Using these antibodies as probes, we purified the six different P.II proteins of this strain. Despite the relatedness of the proteins, we could not purify all of them by a single purification scheme. Four P.II proteins were purified by chromatofocusing, and the remaining two proteins were purified by hydrophobic interaction chromatography on phenyl-Sepharose. The N-terminal amino acid sequence of the proteins showed a high degree of sequence conservation. However, there was variability at specific amino acid residues, giving each P.II protein a unique N-terminal amino acid sequence. Thus P.II proteins of one strain differ among themselves not only in antigenic determinants and primary structure, but also in other characteristics affecting their properties in different chromatographic systems.

The pathogenic Neisseria, N. gonorrhoeae and N. meningitidis, possess an outer membrane protein (OMP), designated H.8, with a conserved monoclonal antibody (MAb)-binding epitope. We determined the DNA sequence of a gonococcal H.8 gene, and confirmed the relationship between the cloned gene and the H.8 OMP by constructing a gonococcal mutant lacking H.8. The predicted H.8 OMP is a lipoprotein 71 amino acids in length, composed of 13 repeats of a consensus sequence AAEAP with perfect 5-residue periodicity. The AAEAP units form a repeating epitope that comprises the entire predicted sequence of the protein.

A hybridoma derived from a mouse immunized with gonococcal outer membranes produced an antibody, designated H.8, that bound to all strains of Neisseria gonorrhoeae and N. meningitidis tested, and to N. lactamica and N. cinerea, but only rarely to other nonpathogenic Neisseria species. Studies with the gonococcal strain used in production of the antibody showed that the antibody bound to a surface-exposed, protease-sensitive, and heat-modifiable outer membrane antigen that we believe is distinct from previously described gonococcal outer membrane proteins.

A genetic locus affecting susceptibility to the bactericidal activity of normal human serum has been designated sac-1. This locus was shown to be closely linked to, but not identical with, a second locus (designated nmp-2) that affects protein 1 of the outer membrane. The sac-1 locus could be linked to known antibiotic resistance markers on the gonococcal chromosome by genetic transformation.

The protein II (PII) outer membrane proteins of Neisseria gonorrhoeae are a family of heat-modifiable proteins that are subject to phase variation, in which the synthesis of different PII species is turned on and off at a high frequency. Transformation of PII genes from a donor gonococcal strain into a recipient strain was detected with monoclonal antibodies specific for the PII proteins of the donor. Individual PII protein-expressing transformants generally bound only one donor-specific PII monoclonal antibody. Recovery of transformants expressing a donor-specific PII protein depended on the PII protein expression state of the donor: the transformed population bound only monoclonal antibodies specific for PII proteins that were expressed in the donor. Colony variants with an altered frequency of switching of PII protein expression were isolated, but the altered switch phenotype did not cotransform with the PII structural gene. These results provide genetic evidence that PII proteins are the products of different genes and that expressed and unexpressed forms of the PII gene are different from each other.

The pathogenic neisseriae, Neisseria gonorrhoeae and Neisseria meningitidis, possess an outer membrane protein, H.8, which contains a conserved monoclonal antibody (MAb)-binding epitope in all strains tested. We have cloned and sequenced a meningococcal H.8 gene, and determined the characteristics of the predicted protein. The predicted signal peptide has features characteristic of a prokaryotic lipoprotein. The region at the N-terminal end of the mature protein (39 amino acids) is primarily composed of alanine, glutamate and proline residues arranged in imperfect repeats with the consensus sequence AAEAP. The epitope for H.8 MAb-binding was localized to a 20-amino-acid sequence within this region. The remainder of the predicted amino acid sequence shows extensive homology to azurins, which are small blue copper-binding proteins found in a limited number of species of pathogenic bacteria.