A mechanism of capsular polysaccharide phase variation in Neisseria meningitidis is described. Meningococcal cells of an encapsulated serogroup B strain were used in invasion assays. Only unencapsulated variants were found to enter epithelial cells. Analysis of one group of capsule-deficient variants indicated that the capsular polysaccharide was re-expressed at a frequency of 10(-3). Measurement of enzymatic activities involved in the biosynthesis of the alpha-2,8 polysialic acid capsule showed that polysialyltransferase (PST) activity was absent in these capsule-negative variants. Nucleotide sequence analysis of siaD revealed an insertion or a deletion of one cytidine residue within a run of (dC)7 residues at position 89, resulting in a frameshift and premature termination of translation. We analysed unencapsulated isolates from carriers and encapsulated case isolates collected during an outbreak of meningococcal disease. Further paired blood-culture isolates and unencapsulated nasopharyngeal isolates from patients with meningococcal meningitis were examined. In all unencapsulated strains analysed we found an insertion or deletion within the oligo-(dC) stretch within siaD, resulting in a frameshift and loss of capsule formation. All encapsulated isolates, however, had seven dC residues at this position, indicating a correlation between capsule phase variation and bacterial invasion and the outbreak of meningococcal disease.

Strains of Neisseria meningitidis responsible for an epidemic of meningococcal disease occurring in Norway since the mid-1970s and for recent increases in the incidence of disease in several other parts of Europe have been identified by multilocus enzyme electrophoresis as members of a distinctive group of 22 closely related clones (the ET-5 complex). Clones of this complex have also colonized South Africa, Chile, Cuba, and Florida, where they have been identified as the causative agents of recent outbreaks of meningococcal disease. There is strong circumstantial evidence that outbreaks of disease occurring in Miami in 1981 and 1982 were caused in large part by bacteria that reached Florida via human immigrants from Cuba.

The genetic structure of populations of Neisseria meningitidis was examined by an analysis of electrophoretically demonstrable allelic variation at 15 genes encoding enzymes in 650 isolates of eight serogroups (A, B, C, W135, X, Y, Z, and 29E) and 38 nonserogroupable isolates. A total of 331 distinctive multilocus genotypes (electrophoretic types, ETs) was identified, among which mean genetic diversity per locus (H = 0.547) was greater than in Escherichia coli and other bacterial species thus far studied. The intercontinental distribution of some ETs and the recovery of organisms of identical genotype over periods of many years strongly suggest that the genetic structure of N. meningitdis is basically clonal as a consequence of low rates of recombination of chromosomal genes. Variation among strains in serogroup, serotype, and the electrophoretic pattern of the major outer membrane proteins has little relationship to the complex structure of populations revealed by enzyme electrophoresis, which involves 14 major lineages of clones diverging from one another at genetic distances greater than 0.50. Genetic diversity among ETs of isolates of the same serogroup was, on average, 84% of that in the total sample. Clones of serogroup A were unusual in being genotypically less heterogeneous than those of other serogroups and in forming a single phylogenetic group. Isolates of the same serotype or outer membrane protein pattern were also highly heterogeneous; on average, 87 and 97%, respectively, of the total species diversity was represented by ETs of the same serotype or outer membrane protein.

The neural cell adhesion molecules (N-CAM) occur chiefly in two molecular forms that are selectively expressed at various stages of development. Highly sialylated forms prevalent in embryonic and neonatal brain are gradually replaced by less sialylated forms as development proceeds. Here we describe a monoclonal antibody raised against the capsular polysaccharides of meningococcus group B (Men B) which specifically distinguishes embryonic N-CAM from adult N-CAM. This antibody recognizes alpha 2-8-linked N-acetylneuraminic acid units (NeuAc alpha 2-8). Immunoblot together with immunoprecipitation experiments with cell lines or tissue extracts showed that N-CAM are the major glycoproteins bearing such polysialosyl units. Moreover we could not detect any sialoglycolipid reactive with this antibody in mouse brain or in the neural cell lines examined. By indirect immunofluorescence staining this anti-Men B antibody decorated cells such as AtT20 (D16/16), which expressed the embryonic forms of N-CAM, but not cells that expressed the adult forms. In primary cultures this antibody allowed us to follow the embryonic-to-adult conversion in individual cells. In addition, the existence of cross-reactive polysialosyl structures on Men B and N-CAM in embryonic brain cells for caution in efforts to develop immunotherapy against neonatal meningitis.

A B S T R A C T A noncovalent complex of meningococcal group B polysaccharide and type 2 outer membrane protein has been characterized and its potential as a vaccine against group B meningococcal disease investigated. The polysaccharide component was found to have a partition coefficient, Kd, of 0.34 on Sepharose CL-4B in the presence of sodium deoxycholate. The protein consisted of four to five major proteins including the principal outer membrane protein. Hydrophobic binding between the protein and polysaccharide was demonstrated by gel filtration and isopycnic CsCl density gradient centrifugation and found to involve all of the proteins. After demonstrating safety and immunogenicity in animals, two lots of vaccine were tested in a total of eight volunteers. Two 120-,ug doses were given subcutaneously at 0 and 5 wk. Mild local reactions occurred in all eight volunteers, but no systemic reactions were observed. 2 wk after the first dose, six of the volunteers had increased levels of bactericidal antibodies against both the group B polysaccharide and the outer membrane proteins. Antibody rises to the group B polysaccharide (mean 6-fold) were confirmed by passive hemagglutination assays and rises to the proteins (mean 10-fold) by a solid phase radioimmunoassay. The second dose resulted in little or no increase in antibody titers. Antibody titers declined over a period of 14 wk but mostly remained above preimmunization levels. Bactericidal antibodies with specificity for the group B polysaccharide were mostly of the immunoglobulin (Ig)M class, and were directed against a determinant associated only with high molecular weight polysaccharides. We conclude that both the group B polysaccharide and the outer membrane protein are immunogenic in man when presented as a complex and that the complex warrants further testing and development as a vaccine against group B meningococcal disease.