Human antibody response to lipopolysaccharides from Neisseria gonorrhoeae

Biology of the Gonococcus: Disease and Pathogenesis

Neisseria gonorrhoeae infection is a major public health problem worldwide. The increasing incidence of gonorrhea coupled with global spread of multidrug-resistant isolates of gonococci has ushered in an era of potentially untreatable infection. Gonococcal disease elicits limited immunity, and individuals are susceptible to repeated infections. In this chapter, we describe gonococcal disease and epidemiology and the structure and function of major surface components involved in pathogenesis. We also discuss the mechanisms that gonococci use to evade host immune responses and the immune responses following immunization with selected bacterial components that may overcome evasion. Understanding the biology of the gonococcus may aid in preventing the spread of gonorrhea and also facilitate the development of gonococcal vaccines and treatments.

Gonococcal lipooligosaccharide sialylation: virulence factor and target for novel immunotherapeutics

Gonorrhea has become resistant to most conventional antimicrobials used in clinical practice. The global spread of multidrug-resistant isolates of Neisseria gonorrhoeae could lead to an era of untreatable gonorrhea. New therapeutic modalities with novel mechanisms of action that do not lend themselves to the development of resistance are urgently needed. Gonococcal lipooligosaccharide (LOS) sialylation is critical for complement resistance and for establishing infection in humans and experimental mouse models. Here we describe two immunotherapeutic approaches that target LOS sialic acid: (i) a fusion protein that comprises the region in the complement inhibitor factor H (FH) that binds to sialylated gonococci and IgG Fc (FH/Fc fusion protein) and (ii) analogs of sialic acid that are incorporated into LOS but fail to protect the bacterium against killing. Both molecules showed efficacy in the mouse vaginal colonization model of gonorrhea and may represent promising immunotherapeutic approaches to target multidrug-resistant isolates. Disabling key gonococcal virulence mechanisms is an effective therapeutic strategy because the reduction of virulence is likely to be accompanied by a loss of fitness, rapid elimination by host immunity and consequently, decreased transmission.

Experimental gonococcal urethritis and reinfection with homologous gonococci in male volunteers

BACKGROUND

Reinfection, a common occurrence with gonorrhea, may result from a lack of protective immune response, or from the tremendous gonococcal strain variation.

GOAL

A two-phase study in human volunteers tested whether experimental infection with Neisseria gonorrhoeae MS11mkC would protect against reinfection with the same organisms.

STUDY DESIGN

In phase 1, an intraurethral inoculum of 57,000 piliated, transparent (opacity protein-negative [Opa-]) MS11mkC N gonorrhoeae infected 14 of 15 (93%) volunteers. The volunteers were encouraged to delay treatment for at least 5 days. In phase 2, which began 2 weeks after treatment for the initial infection, volunteers were inoculated with 7,100 piliated, Opa- MS11mkC.

RESULTS

The phase 2 challenge infected 6 of 14 (43%) previously infected volunteers and 5 of 10 (50%) naïve control subjects. Phase 1 volunteers who resisted reinfection were significantly more likely to have had a fourfold or greater increase in lipooligosaccharide immunoglobulin G during phase 1 than those who did not resist reinfection (P = 0.026).

CONCLUSIONS

Although infection did not provide protection from reinfection under the conditions used, the results suggest that immunity to reinfection is more complex than anticipated by the experimental design.

Lipo-oligosaccharides (LOS) of mucosal pathogens: molecular mimicry and host-modification of LOS

Immunochemical studies of the lipo-oligosaccharides (LOS) of the Gram-negative bacteria Neisseria gonorrhoeae and Neisseria meningitidis have revealed some interesting structural characteristics of these LOS that might relate to their roles during pathogenesis. The carbohydrate moieties of the LOS of pathogenic Neisseria mimic carbohydrates present in glycosphingolipids of human cells. Firstly, an LOS component present among a number of Neisseria species is antigenically and/or chemically identical to lactoneoseries glycosphingolipids present in human cells. The lactoneoseries LOS becomes sialylated on Neisseria gonorrhoeae when they are grown in the presence of cytidine 5'-monophospho-N-acetyl-neuraminic acid (CMP-NANA), the nucleotide sugar for sialic acid. Examination of gonococci present in exudates from males with natural infection indicates that sialylation also occurs in vivo. The mechanism for this process apparently involves a bacterial sialyltransferase scavenging available host CMP-NANA ("host-modification" of LOS) and transferring the sialic acid to the lactoneoserieslike LOS. Strains of N. meningitidis and Haemophilus influenzae also express similarly sialylated LOS suggesting that this is a common mechanism of pathogenesis among these bacteria. Additional examples of LOS that mimic other glycosphingolipid series have been identified also and the fact that multiple series can be expressed in a single population of gonococci suggests that a diverse set of LOS can be presented to the host during infection. It is possible that this diverse set of LOS serve different functions for the bacteria in various hosts and/or environments during infection.

Immunization of guinea pigs with epitope C-rich lipopolysaccharide from Neisseria gonorrhoea; in vivo selection of gonococcal variants

The immunogenic potential of lipopolysaccharide (LPS) of a variant of Neisseria gonorrhoeae strain Gc 40 selected by growth in vivo (vivo variant) was investigated in guinea pigs. LPS extracts obtained from the water (WLPS) and the phenol (PLPS) phases of a hot phenol-water extraction were compared for their antigenic capacity and protective effect against infection in subcutaneous chambers. Immunization with PLPS induced significant levels of anti-LPS and anti-epitope C antibodies but WLPS was not antigenic. Two days after challenge, all guinea pigs immunized with WLPS had infections similar to those seen in unimmunized control animals while most animals immunized with PLPS and challenged with either 10(3) or 10(5) gonococci per ml showed low numbers of or no viable gonococci in their chambers. Five days after challenge, however, the same animals had chamber infections with high viable counts similar to controls. Gonococci reisolated from three such animals had physically and antigenically altered lipopolysaccharide and showed patterns of serum sensitivity to pre-challenge chamber fluid from immunized animals which were different from those of the parent vivo variant used for immunization and challenge. The results demonstrate that selection of LPS variants occurs in vivo. This could constitute an immune evasion mechanism.

Passive hemagglutination test for detection of antibody to gonococcal ribosomal antigen in sera from patients with asymptomatic gonorrhea

Ribosomal fractions were obtained from a culture of type 2 Neisseria gonorrhoeae strain P-17 which was isolated from a patient with an acute gonococcal infection; these fractions were purified to eliminate the components of the outer membrane complex by affinity chromatography (Sepharose-anti-outer membrane complex antibody conjugates were used as the solid immunosorbent), and the resulting preparation was designated the purified ribosomal fraction, The purified ribosomal fraction was used to detect antibody activity in sera obtained from culture-positive asymptomatic carriers and healthy controls by a passive hemagglutination test. This passive hemagglutination test had a specificity of 100% for both sexes and sensitivities of 99.4 and 88.2% for female and male carriers, respectively, when an antibody titer of more than 1:3 was defined as abnormal. Absorption of the sera with nongonococcal organisms did not affect the antibody activity, and no significant difference in antigenicity among various N. gonorrhoeae strains was observed in ribosomal fractions. An enzyme-linked immunosorbent assay was also used to measure the relative amounts of specific antibodies to the purified ribosomal fraction, and this assay revealed that the anti-purified ribosomal fraction antibodies were immunoglobulin G.

Serum immunoglobulin response in uncomplicated gonorrhoea

Sera from 225 men and 140 women were examined by an indirect immunofluorescent antibody technique for antibody reactive with Neisseria gonorrhoeae. Antigonococcal IgM was demonstrated at a titre of greater than or equal to 16 in about 45% of infected, but in only 3% of non-infected, patients. Most of this antibody occurred in sera of patients who had been infected for less than 14 days. Antibody of the IgA class was found at a titre of greater than or equal to 16 in over half the infected, but in none of the non-infected, patients. IgG antibody reactive with the gonococcus was found in each infected patient at a titre of greater than or equal to 16 but in only 8% of controls. The mean log titre of this antibody was significantly higher in patients who had been infected for more than seven days than in those whose infection was of shorter duration.

Serological diagnosis of gonorrhoea by an enzyme-linked immunosorbent assay (Elisa)

IgG antibody to an outer membrane protein extracted from Neisseria gonorrhoeae was measured in patients with gonorrhoea. The level in such patients was significantly higher than in normal controls or in patients with other conditions who were attending the clinic for sexually transmitted diseases. Significantly higher antibody levels were detectable in half the patients within a few days of infection and in a similar proportion of carriers--both male and female. Men with rectal gonorrhoea had particularly high antibody levels. Sixteen per cent of presumptively normal men and 11% of normal women gave positive results but the actual false positive rate could be lower.

Asymptomatic females: detection of antibody activity to gonococcal pili antigen by radioimmunoassay

A gonococcal pili antigen preparation was used to detect antibody activity sera obtained from 322 culture-positive asymptomatic females and 150 negative controls. Pili were obtained from a culture of type 2 Neisseria gonorrhoeae (strain 2686) and labeled with 125I for use in a double-antibody radioimmunoassay test system. Of the 322 sera obtained from culture-positive, asymptomatic females, 276 (85.7%) showed antibody activity greater than or equal to 1.8 mug/ml. Negative controls were obtained from three different groups of individuals, and 130 (86.7%) had undetectable antibody activity. Sera from asymptomatic, culture-positive females were absorbed with three different strains of N. gonorrhoeae, one of these strains being the organism used for pili antigen preparations. The absorbed sera were tested for antibody activity, and in each case the activity in the absorbed sera dropped to an undetectable level. When the same sera were absorbed with N. meningitidis, N. catarrhalis, N. perflava, Escherichia coli, Herellea vaginicola, Mima polymorpha, Staphylococcus aureus, and Candida albicans, little, if any, decline in the level of anti-pili antibody activity was observed.

Specificity of human serum antibodies against endotoxin from N. gonorrhoeae

Sera from patients with gonorrhoea were used to study the specificity of human antibodies against determinant a of endotoxin from gonococcal strains 8551, V, and VII. The sera were tested by an indirect haemagglutination technique before and after absorption with endotoxin from the same strains. The sera were used untreated and treated with mercaptoethanol (ME). The untreated and ME-treated sera showed antibody activity against all the a determinants when examined unabsorbed. After absorption with any of the endotoxin preparations the sera usually demonstrated antibody activity against one or both of the other preparations. The pattern of reactivity thus observed differed from one serum to another. Treatment of sera with ME resulted in an altered pattern of reactivity of four of the five sera. The results indicate multispecificity of human antibodies against the a determinants.

Mercaptoethanol-resistant human serum antibodies reacting with endotoxin from Neisseria gonorrhoeae

Sera from fifty patients with gonorrhoea, thirty with non-specific urethritis, and eighty blood donors were treated with mercaptoethanol (ME) and examined by the indirect haemagglutination test for antibodies against endotoxin from gonococci. Erythrocytes sensitized with determinant a of endotoxin from Strains 8551, V, and VII, or determinant b from Strain V were used. The percentage of sera active in the haemagglutination test was much higher in the gonorrhoea group than in the controls. The geometric mean titre was also significantly higher in the gonorrhoea group. This applied for all four antigens used. Results obtained in an anti-globulin test indicated that the titre of ME-treated serum was determined by IgG antibodies against the endotoxin. Many sera had titres which varied according to the strain origin of the antigen used in the test. The sensitivity of tests for antibodies was increased by using endotoxin from several different strains of gonococci for the examination of each serum. A simplified procedure for determination of antibodies against endotoxin from different strains of gonococci was elaborated.

Quantitative determination of antibody to gonococcal pili. Changes in antibody levels with gonococcal infection

Gonococcal pili, pure by the criteria of electron microscopic examination and polyacrylamide gel electrophoresis in sodium dodecyl sulfate, have been prepared by repeated cycles of precipitation with 0.1 M MgCl(2), followed by dissolution in 0.01 M Tris pH 8, 0.01 M NaN(3). Using a fluorescein-conjugated antibody prepared to pili from a single strain, pilar antigen(s) was found to be present in each of 18 strains of gonococci tested, and absent from strains of pilated meningococci, nonpathogenic Neisseria sp., and Escherichia coli. Purified pili, labeled with (125)I were used in an antigen binding assay to quantitatively measure antibody to pili in rabbit sera and in 561 human sera. The range of antibody activity for 133 persons unlikely to have experienced gonorrhea was 0.1-1.6 mug/ml with a geometric mean of 0.5 mug/ml. This geometric mean antibody activity was significantly lower than the geometric mean for asymptomatically infected males (1.0 mug/ml, P < 0.002), males with symptomatic gonococcal anterior urethritis (1.6 mug/ml, P < 0.001), or asymptomatically infected females (4.2 mug/ml, P < 0.001). Antibody appeared in elevated levels (> 1.6 mug/ml) 2-3 wk after infection and returned toward control levels 1 or more months after treatment. Antibody levels higher than 1.6 mug/ml were found in 26 (50%) of 52 males with gonococcal anterior urethritis, in 10 (33%) of 30 males with asymptomatic urethral infection and in 50 (89%) of 56 asymptomatically infected females. In a high-risk group of 103 females for whom culture results and antibody to pili were compared, 58 (57%) had elevated antibody levels to pili and 86% of the infected females were within this seropositive group. The antigen binding assay may provide a means to detect asymptomatic gonococcal infection in women.