SummaryThe bactericidal activity of 20 maternal-neonatal paired sera was assessed employing a clinical type Ia, group B streptococcal isolate, strain 515, known to be opsonized by the classical complement pathway in a non-antibody dependent fashion. Twelve neonatal sera had efticient bactericidal activity for this isolate (mean 96%, range 91-99%) whereas eight had significantly lower bactericidal indices (mean 29%, range 0-54%) ( p < 0.001). The mean bactericidal index for 20 maternal sera (88%) did not differ from that of 20 adult control sera (91%). Bactericidal activity was not influenced by the concentration of antibody to the capsular polysaccharide antigen of type Ia, group B Streptococcus present in these sera. When the classical pathway of selected neonatal sera with high or low bactericidal activity was inactivated by MgEGTA chelation of C1, bacterial growth was observed uniformly when concentrations of specific antibody were low. The bactericidal index of neonatal sera correlated significantly with the total hemolytic complement titer ( r = 0.611, P < 0.01). The mean levels of each complement component or control protein assessed by radial immunodiffusion (Clq, C4, C3, B, H, and I) were lower in neonatal sera with low than in those with efficient bactericidal activity, and levels of Clq and H were depressed significantly (P < 0.025 and <0.001, respectively). Hemolytic titers of C4 but not C1 or C2 were also significantly lower for low than for selected high-killing neonatal sera (P < 0.05). Bactericidal activity in neonatal serum with a bactericidal index of zero was restored in a dose-dependent fashion by the addition of fresh frozen plasma. Abbreviations BI, bactericidal index CHSO, whole complement activity EGTA, ethyleneglycoltetracetate FFP, fresh frozen plasma GVB, Veronal-buffered saline containing 0.1% gelatin GVB++, GVB containing 0.5 mM Mg++ and 0.15 mM Ca++ MgEGTA, GVB supplemented to 4 mM Mg++ and 16 mM EGTA adjusted to pH 7.5 OD, optical density THB, Todd-Hewitt brothThe levels of individual classical complement pathway components in neonatal sera have been shown to be significantly lower than those in adults (1,6-7, 11-12, 19). Davis et al. (6) have shown that by the age of 6 months the mean concentrations of C2 and C4 have increased to levels not significantly different from those of adults, but Clq and C3 concentrations remain low. Few studies have compared the functional capacity of the classical pathway for a pathogen common in the neonate with the degree of depression of individual classical pathway components. We have shown previously that bactericidal activity in adult sera for a variety of clinical isolates of type Ia, group B streptococci was mediated by the classical complement pathway in a non-antibody dependent fashion (2). Both adult sera deficient in specific antibody and agammaglobulinemic serum, which had no detectable antibody to the capsular polysaccharide antigen of type la, group B Streptococcus supported efficient bactericidal activity for each of 18 clinical type Ia...

Since diabetes mellitus predisposes adults to group B streptococcal (GBS) bacteremia, a murine model of streptozotocin-induced diabetes and type II GBS bacteremia was developed to assess certain immune factors which might influence susceptibility to infection. In diabetic mice, the 50% lethal dose for two strains of type II GBS was significantly lower (greater than 1 log10 decrease in CFU per milliliter) than in control animals. This enhanced virulence of GBS for diabetic animals was associated with prolonged bacteremia, persistent sequestration of organisms in the splanchnic reticuloendothelial system, and a shift from splenic to hepatic clearance. Although immunization of control and diabetic animals resulted in high concentrations of type-specific serum antibody, it had no effect on late reticuloendothelial system sequestration in diabetics. In contrast, depletion of complement by treatment of mice with cobra venom factor blocked reticuloendothelial system clearance and resulted in fatal infection in both diabetic and control mice. These results indicate that neither type-specific antibody nor an intact complement system is adequate for effective clearance of type II GBS bacteremia in mice with experimentally induced diabetes. This clearance deficit could be the result of a defect in hepatocyte membrane receptors necessary for removal of this encapsulated microorganism.

At our hospital (Jefferson Davis Hospital, Houston, Tex.) since 1979, non-group D alpha-hemolytic streptococci have been isolated with increasing frequency from neonatal blood cultures with clinical findings of sepsis. A total of 47 such isolates were identified to the species level by the scheme of Facklam and were compared with 57 genital isolates from 167 maternity patients. Among the genital isolates, S. sanguis II and S. MG-intermedius accounted for 53 and 28%, respectively, and both were significantly less common in neonatal cultures (23 and 11%, respectively; P less than 0.05). Among neonatal isolates, S. mitis was the single most frequent species (35%), in contrast to its rare occurrence in maternal cultures (3.4%; P less than 0.001). The disparity between the prevalence of S. mitis in neonatal compared with maternal cultures suggests that this species of non-group D alpha-hemolytic streptococci may have increased virulence in neonatal hosts.

A radioimmunoprecipitin test was developed to determine the immunoglobulin class distribution of naturally acquired and vaccine-induced antibody to the native capsular polysaccharide of type III group B streptococci (III-GBS). In sera from adults and pregnant women with naturally acquired antibody, the mean percentage of antigen bound by immunoglobulin G (IgG) was 74.9 and 78.6, respectively, whereas antigen bound by IgM comprised less than 10% of the total. In contrast, early-convalescent-phase sera (mean, 16.3 days) from neonates responding to III-GBS infection with an increase in specific antibody had significantly more IgM (mean, 36%; P < 0.001, unpaired t test). However, in late convalescence, the immunoglobulin class distribution in sera from these neonates was similar to that of naturally immune adults. Four weeks after immunization with III-GBS polysaccharide vaccine, sera from adults with low (<2 ,ug/ml) preimmunization antibody levels in their sera and from those with moderate (mean, 5.5 ,ug/ml) preimmunization levels contained specific antibody predominantly of the IgG class. Although the percentage of IgG-specific antibody was greater in sera from naturally immune adults than in that from vaccinees with a presumed primary immune response, the major portion of antigen bound by sera at 4 weeks postimmunization (62.5%) was associated with IgG. These observations support the opinion that immunization of pregnant women with III-GBS capsular polysaccharide could be efficacious for the prevention of invasive neonatal III-GBS disease.

No abstract