Enzyme-linked immunosorbent assay for measurement of antibody to type III group B streptococci.

Neonates at risk for fulminant type III, group B streptococcal (III GBS) infection are those who lack antibody to the capsular polysaccharide. A newly developed enzyme-linked immunosorbent assay (ELISA) was compared with a standard radioactive antigen-binding assay (RABA) for quantitation of III GBS antibody in human sera. Although there was a significant correlation between the ELISA and RABA (r = 0.81; P less than 0.001) in general, the ELISA detected antibody both to core and native antigens of III GBS, whereas the RABA detected antibody to native polysaccharide exclusively. The results of the two assays were discordant when sera which had only high native or core antibody (not both) were assessed. Although the ELISA was reproducible and required less than 1 microliter of serum, interpretation of data obtained by the assay should be viewed with caution since only antibody to native III GBS has been correlated with human immunity.     

Immunoglobulin-binding structure on bovine group G streptococci different from type III Fc receptors on human group G streptococci.

The immunoglobulin G (IgG)-binding capacity of 54 group G streptococci of human and bovine origin was investigated. Of 20 human strains, 17 carried a surface component which could combine with human IgG and bovine IgG1 and IgG2. Inhibition experiments with unlabeled human IgG and with a panel of animal sera revealed that the same surface component was involved in the binding of human as well as bovine immunoglobulins. Of 16 beta-hemolytic bovine group G streptococci, 13 reacted with human IgG but not with bovine IgG1 or IgG2. This binding structure was different from the type III Fc reactivity found in human group G streptococci. All human strains, including the three IgG Fc-nonreactive strains, fermented trehalose, in contrast to all bovine beta-hemolytic strains, which were negative. Immunoglobulin Fc reactivity is thus a feature not only of human strains but also of some bovine strains.     

In vitro lymphocyte proliferation in response to type III group B streptococci.

The in vitro cell-mediated responses to group B streptococci (GBS) and the relationship of cell-mediated immunity to specific humoral immunity to type III GBS were investigated. Blood was obtained from 20 adult volunteers, and lymphocytes were isolated and cultured in microtiter plates. Each well contained 2 x 10(5) lymphocytes, 15% autologous serum, and either GBS (cell-to-organism ratio of 1:10, 1:1, or 1:0.1), phytohemagglutinin, streptokinase-streptodornase, or RPMI 1640. Cells were harvested at 5, 6, or 7 days, and DNA synthesis was quantitated. Serum antibody titers were determined with an enzyme-linked immunosorbent assay. Maximal lymphocyte responses occurred at 6 days of culture and at a cell-to-organism ratio of 1:1. Individuals with significant antibody titers to type III GBS, as well as those with undetectable antibody, responded to GBS (stimulation index greater than 10). There was a significant difference (P less than 0.001) between mean antibody concentrations in responders (stimulation index greater than 10) and nonresponders (stimulation index less than 10). Thus, the in vitro responses to GBS may be both to a specific antigen and to a nonspecific mitogen and may be important in host immunity to GBS.     

Factors influencing release of type III antigens by group B streptococci.

The release of serotype III group B streptococcal polysaccharides into the supernatant fluid was examined under a variety of physiological conditions. Release of both high- and low-molecular-weight type III antigens was fairly constant throughout exponential growth, but increased markedly upon entering the stationary phase of growth. Increased glucose and decreased phosphate concentrations both caused a large increase in release of antigens. Inhibition of protein synthesis in exponentially growing cells by chloramphenicol (10 micrograms/ml) caused a condition of unbalanced growth in which antigen release was increased greatly over control values. Strain variability in antigen release was also observed. Strains which are known to be high neuraminidase producers released elevated levels of both low- and high-molecular-weight type III antigens. Non-neuraminidase-producing strains released considerably less high-molecular-weight antigen, but similar levels of the low-molecular-weight antigen compared with the high neuraminidase producers. Strain D136C, a type III non-neuraminidase producer, released negligible quantities of the high-molecular-weight antigen in the supernatant fluid. These results indicate that both the physiological environment and the type III strain are important in determining the quantity of type-specific antigen released into the culture fluid.     

Protease production by clinical isolates of type III group B streptococci.

Six strains of serotype III group B streptococci isolated from confirmed cases of neonatal disease were examined for their ability to produce proteolytic enzymes. Three neuraminidase-producing strains and three non-neuraminidase-producing strains were employed in this study. Protease production was examined in 1,000-fold concentrated filtrates of stationary-phase cells with an insoluble substrate derived from horse hide powder labeled covalently with Remazol brilliant blue. Protease activity was not detected in any cultural supernatant fluids until they were fractionated on Sephadex G-100. After fractionation, the neuraminidase-producing strains were shown to elaborate approximately sixfold more protease than the non-neuraminidase-producing strains. The finding that clinical isolates of group B streptococci that elaborated high levels of neuraminidase also produced elevated levels of extracellular protease may indicate that the production of several different factors may determine the virulence of these organisms.     

Deposition and degradation of C3 on type III group B streptococci.

Antibody to the polysaccharide capsule of type III group B streptococci (GBS) and complement are essential to host defense against systemic infection in neonates. Interactions between C3 degradation products and specific neutrophil receptors mediate the attachment and ingestion of these organisms. To evaluate the influence of capsule on C3 disposition, we compared the C3 fragments released from a highly encapsulated clinical isolate (M861) with those from an unencapsulated mutant (COH 31-15) and an asialo mutant (COH 31-21) of type III GBS after opsonization with hypogammaglobulinemic serum. Upon sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot (immunoblot) analysis, the three strains displayed similar patterns of C3 degradation; both C3b and iC3b were detectable. However, as the duration of opsonization increased, C3 fragment bands became more prominent on the encapsulated strain. The capsule, and specifically sialylation of the capsular polysaccharide of type III GBS, promotes C3 fragment deposition. However, C3 was deposited and degraded to iC3b in the absence of capsule. Opsonization of strain M861 with serum containing antibody specific for the polysaccharide capsule facilitated C3 fragment deposition in the early phases of opsonization. Because iC3b is one of the C3 fragments on an encapsulated strain of type III GBS, the relative deficiency of neonatal neutrophil receptors for this ligand may contribute to the virulence of this organism. Sufficient concentrations of antibody may enhance opsonization by facilitating C3 deposition as well as by interacting with Fc receptors on neutrophils.     

Comparison of type III Fc receptors associated with group C and group G streptococci

The type III Fc receptors present on or secreted by a series of group C and G streptococcal strains were studied. All strains capable of binding radiolabeled human IgG were shown to do so via an antigenically related Fc receptor. Treatment of any of the bacterial strains with papain or trypsin resulted in solubilization of Fc receptor activity. The pattern of Fc receptor activity recovered following enzyme treatment was not uniform. Differences were observed both between group C and G strains as well as within group C and G strains. Analysis of secreted Fc receptors indicated the presence of five molecular forms of Fc receptor. Each form was present at some level in the supernatant of every group C and G strain studied. The relative concn of each form of receptor secreted varied from strain to strain. The Fc receptor activity secreted by each strain demonstrated a similar affinity for the Fc region of human IgG and all were antigenically related. These results suggest that there is a family of closely related Fc receptors associated with group C and G streptococcal strains.     

Neutrophil Fc receptor participation in phagocytosis of type III group B streptococci.

Human peripheral blood neutrophils bear receptors for immunoglobulin G, FcRII, and FcRIII that differ structurally and functionally. We investigated the role of FcRII and FcRIII in the phagocytosis of group B streptococci (GBS) by measuring neutrophil uptake of radiolabeled type III GBS. The mean uptake of GBS opsonized with human serum containing complement and specific antibody was 76%, but when this serum was heated, the mean uptake was only 22%. A monoclonal antibody to FcRIII, Leu-11b, inhibited in a dose-dependent manner uptake of GBS opsonized with heated or intact serum to maxima of 40 and 30%, respectively. Conversely, a monoclonal antibody to FcRII, IV.3, inhibited by 77% the uptake of GBS opsonized with heated serum but had no effect when GBS was opsonized with intact serum. Leu-11b and IV.3 had an additive inhibitory effect with heated but not with intact serum. Neither monoclonal antibody inhibited the uptake of GBS opsonized with hypogammaglobulinemic serum. Therefore, FcRII is the primary mediator of the phagocytosis of GBS opsonized by antibody alone, whereas FcRIII plays a lesser role. Surprisingly, FcRII is not necessary for phagocytosis when complement is also present. FcRIII participates, to a limited extent, in phagocytosis of GBS opsonized with antibody whether or not complement is present. These findings suggest that the function of FcRII in triggering phagocytosis may be particularly important in host defense against type III GBS in the setting of complement deficiency of young infants.     

Biological and immunochemical identity of M protein on group G streptococci with M protein on group A streptococci.

Previous evidence for the presence of an M or M-like protein on group G streptococci has been based on the ability of these strains to survive in human blood. In addition, cross-reactions between group A and group G streptococci have been demonstrated, but they have relied either on whole bacterial cell vaccine-induced polyclonal sera or crude protein extracts of these cells. In this study two monoclonal antibodies prepared against the purified, native group A streptococcal M6 protein demonstrated a high degree of cross-reactivity with group G streptococcal clinical isolates (9 and 19 of 22 strains examined, respectively). Ten of these strains exhibited resistance to phagocytosis when rotated in human blood. In addition, immunoblot analysis of crude mutanolysin extracts of group G streptococci with one of the M6 monoclonal antibodies illustrated a remarkable similarity in the protein pattern of these extracts as compared with those of group A streptococcal M protein. The immunoblots further demonstrated a variation in the relative molecular weights of the extracted proteins from strain to strain over a range of 57,000 to 77,000. In addition, a purified, pepsin-derived fragment (Mr, 43,000) from a group G strain was capable of eliciting rabbit antibodies that were opsonic for group G cells in a bactericidal assay. These functional and immunochemical data, in concert with DNA hybridization between group G streptococcal DNA and a group A M6 gene probe (J. R. Scott, W. M. Pulliam, S. K. Hollingshead, and V. A. Fischetti, Proc. Natl. Acad. Sci. USA 82:1822-1826, 1985), provide strong evidence for the presence of an M protein on these organisms and indicate its probable role as a virulence molecule on the surface of group G streptococci.     

Complement and antibody participation in opsonophagocytosis of type IV and V group B streptococci.

Requirements for complement and antibody in neutrophil-mediated killing of serotype IV and V group B streptococci were investigated. Neutrophils from adults were tested in an opsonophagocytic assay with sera from healthy adults, healthy newborns, and hypogammaglobulinemic, agammaglobulinemic, and C4-deficient patients. For all serum sources, the bactericidal index for both serotypes exceeded 84% after 40 min of incubation. Heat inactivation of sera ablated killing. Blockade of neutrophil receptor FcIII effected a maximum of 16% inhibition of opsonophagocytosis, and FcII receptor blockade demonstrated negligible inhibition. When neutrophil complement receptor 1 or 3 blockade was employed, the maximum inhibition detected was 26%. Simultaneous blockade of complement receptors 1 and 3 effected maximum inhibition levels of 25 and 65% for serotypes IV and V, respectively. Blockade of complement receptor 3 and neutrophil receptor FcIII inhibited opsonophagocytosis by 56% for both serotypes. When serum complement concentrations were restricted, neutrophil-mediated killing diminished but was restored by the addition of hyperimmune rabbit antiserum. These findings suggest that complement and antibody are major participants in the opsonophagocytosis of serotypes IV and V group B streptococci. A low prevalence of carriage or mediation of efficient phagocytosis by interactions of neutrophil complement and Fc receptors may contribute to the rarity of human infections caused by these two serotypes.     

Extracellular antigens of serotype III group B streptococci.

Two sialic acid-containing type III group B streptococcal antigens were obtained from a supernatant growth medium, purified by anion exchange or gel filtration, and found to be free of group B reactivity. Quantitation of the high-molecular-weight extracellular type III antigen indicated that approximately 20-fold more antigen was recoverable from the growth medium than could be obtained by neutral buffer extraction of whole cells.     

Protective levels of human immunoglobulin G antibody to group B streptococcus type Ib.

We studied the concentration of circulating human immunoglobulin G (IgG) antibody to the native capsular polysaccharide of group B streptococcus (GBS) type Ib necessary to protect mice against lethal challenge by laboratory and clinical GBS Ib strains. Antibody was measured by an enzyme-linked immunosorbent assay in which native polysaccharide antigen coupled to human serum albumin was used. The assay was standardized by a quantitative precipitation test, using native antigen and specific human IgG antibody purified by affinity chromatography. IgG anti-GBS Ib antibody level-protection curves for 90% lethal dose challenge of mice were sigmoidal. The curves of whole serum and affinity-chromatographed IgG anti-GBS Ib were superimposable. The serum concentrations of human antibody required for complete protection of mice varied with the infecting strain and ranged from 0.038 to 0.175 microgram/ml. Protective levels of human IgG anti-GBS Ib were lower than those we found previously for homologous protection against GBS Ia challenge (range, 0.25 to 1.0 microgram/ml).     

Capsular polysaccharide regulates neutrophil complement receptor interactions with type III group B streptococci.

The capsular polysaccharide of type III group B streptococci contributes substantially to the virulence of this organism. We explored the extent to which capsular polysaccharide influences neutrophil complement receptor interactions by using a poorly encapsulated strain (COH 31r/s), two well-encapsulated strains (M732 and M912), and strains produced from COH 31r/s by transposon mutagenesis that lacked capsule (COH 31-15) or had capsular polysaccharide lacking terminal sialic acid residues (COH 31-21). When tested with normal human serum, each strain had initially high bactericidal indices (85 to 96%). Monoclonal antibody blockade of neutrophil complement receptor 3 (CD11b/CD18) inhibited opsonophagocytosis to a significantly greater extent for the well-encapsulated strain than for the poorly encapsulated, asialo, or unencapsulated mutant strain. The addition of antibody with specificity for capsular polysaccharide reduced the inhibitory effect significantly for the encapsulated but not for the mutant strains. Blockade of neutrophil complement receptor 1 (CD35) effected only low-level inhibition. However, simultaneous blockade of complement receptors 1 and 3 augmented the inhibitory effect. When hypogammaglobulinemic serum was used as an antibody-free complement source, the initial bactericidal index was low (30% +/- 15%) for an encapsulated strain and was not affected for the mutant strains. Blockade of either neutrophil complement receptor 1 or 3 or the combination fully inhibited killing of the encapsulated strain. These results demonstrate that the type III group B streptococcal capsular polysaccharide regulates interactions with neutrophil complement receptors. We conclude that efficient phagocytic killing of encapsulated group streptococci in nonimmune serum requires ligation of complement receptors 1 and 3.     

Density profile of group B streptococci, type III, and its possible relation to enhanced virulence.

The buoyant densities of virulent and colonizing group B streptococci, type III, were determined by centrifugation of bacteria on a linear, hypotonic density gradient. A total of 28 strains were investigated. Eleven strains were obtained from blood cultures of babies with early-onset disease, and eight strains were isolated from the cerebrospinal fluid of babies with late-onset septicemia and meningitis. Nine colonizing strains were genital isolates from pregnant women subsequently giving birth to healthy children. In each strain the buoyant density was determined before and after neuraminidase treatment. All strains showed an increase in the buoyant density after enzymatic removal of sialic acid, and the density differences before and after desialylation were calculated. The mean values of these differences for blood, cerebrospinal fluid, and colonizing isolates were 23.4, 25.3, and 10.6 mg/ml, respectively. The mean value for the colonizing strains differed significantly from the mean value for each group of virulent strains. All colonizing strains banded singly in the gradient, whereas five of the virulent strains divided into two density populations. Extracts of the low-density cells produced markedly more dense immunoprecipitates with type antiserum than did extracts of the high-density bacteria. One double-banding strain was positive for R protein. After separation of the two density populations, this antigen was detected only in the low-density population. The results indicate that bacterial buoyant density is inversely related to the amount of capsular polysaccharide enveloping the cell and that a determination of the density profile of the bacteria may be used for discriminating strains with an increased pathogenic potential.     

Streptococcal group B type antigen X in group L streptococci.

Extracts from 19 of 29 group L streptococcal cultures reacted distinctly with antiserum against group B streptococcal type antigen X in coagglutination and immunodiffusion tests. The X antigen corresponded to those obtained by streptococci of serological groups B and G.     

Identification of a genetic locus essential for capsule sialylation in type III group B streptococci.

The type III capsular polysaccharide of group B streptococci (GBS) consists of a linear backbone with short side chains ending in residues of N-acetylneuraminic acid, or sialic acid. The presence of sialic acid on the surface of the organism inhibits activation of the alternative pathway of complement and is thought to be an important element in the virulence function of the capsule. We showed previously that a mutant strain of GBS that expressed a sialic acid-deficient, or asialo, form of the type III polysaccharide was avirulent, supporting a virulence function for capsular sialic acid. We now report the derivation of an asialo capsule mutant from a highly encapsulated wild-type strain of type III GBS, strain COH1, by insertional mutagenesis with transposon Tn916 delta E. In contrast to the wild-type strain, the asialo mutant strain COH1-11 was sensitive to phagocytic killing by human leukocytes in vitro and was relatively avirulent in a neonatal rat model of GBS infection. The asialo mutant accumulated free intracellular sialic acid, suggesting a defect subsequent to sialic acid synthesis in the biosynthetic pathway leading to capsule sialylation. The specific biosynthetic defect in mutant strain COH1-11 was found to be in the activation of free sialic acid to CMP-sialic acid: CMP-sialic acid synthetase activity was present in the wild-type strain COH1 but was not detected in the asialo mutant strain COH1-11. One of the two transposon insertions in the asialo mutant COH1-11 mapped to the same chromosomal location as one of the two Tn916 insertions in the previously reported asialo mutant COH31-21, identifying this site as a genetic locus necessary for expression of CMP-sialic acid synthetase activity. These studies demonstrate that the enzymatic synthesis of CMP-sialic acid by GBS is an essential step in sialylation of the type III capsular polysaccharide.     

A human IgG 3 is opsonicin vitro against type III group B streptococci

Preparations of IgG 3 isolated by absorption of IgG 1, IgG 2, and IgG 4 from a human iv immunoglobulin with protein A-Sepharose were evaluated for their opsonic activities against type III group B streptococcal (GBS) strains. The resulting preparations were free of IgG 1 and IgG 2 and contained only trace amounts of IgG 4 (<2% of total IgG). These IgG 3 preparations exhibited excellent opsonic activities against type III GBS strains, similar to those of the unfractionated iv immunoglobulin (based on total IgG concentrations in the opsonic assays). In contrast, preparations of IgG 1, 2, and 4 eluted from protein A-Sepharose with 2M acetic acid and 7M urea were significantly less effective in enhancing phagocytosis and killing of type III GBS than IgG 3 preparations or iv immunoglobulin. The reasons for excellent opsonic activity of IgG 3 preparations as well as for decreased opsonic activity of IgG 1, 2, and 4 preparations are not clear. Perhaps alteration of IgG by lower pH and high concentrations of urea may have impaired the functional activity of IgG 1, 2, and 4 preparations. The significant finding of this study is the first demonstration of the excellent opsonic activity of IgG 3, emphasizing the importance of having intact IgG 3 in commercial immunoglobulin preparations used in prophylaxis or treatment of GBS infections.     

M proteins of group G streptococci isolated from bacteremic human infections.

We studied seven strains of group G streptococci isolated from clinically severe bacteremic infections in six intravenous drug abusers. These group G strains multiplied luxuriantly in fresh human blood. On electron microscopy, they exhibited surface fibrillae similar to those observed in M-protein-rich group A streptococci, but they were not serologically M typable with a battery of 39 M antisera. Rabbit antisera raised against two of the group G strains (1618 and 1750) opsonized the homologous but not the heterologous isolates and exhibited type-specific Ouchterlony immunoprecipitin reactions. Moreover, antisera raised against peptic extracts of strain 1750 also promoted phagocytic killing of that strain. Anti-1750 reacted in high titer in an enzyme-linked immunosorbent assay against peptic extracts of the homologous strain; these antibodies were removed by absorption with 1750 cells but not by absorption with heterologous strains. These studies represent, to our knowledge, the first analysis of virulence factors of group G streptococci isolated from invasive human disease. The seven epidemiologically related blood isolates of group G streptococci possess distinct type-specific, antiphagocytic surface virulence factors analogous to the M proteins of group A streptococci.     

Association of elevated levels of extracellular neuraminidase with clinical isolates of type III group B streptococci.

The level of total extracellular neuraminidase produced by 74 clinical isolates of group B streptococci isolated from diseased or asymptomatically colonized infants was assayed. Extracellular neuraminidase was obtained from concentrated filtrates of exponentially growing cultures of group B streptococci grown in a chemically defined medium (FMC) containing supplemental protein. The total activity of extracellular enzyme produced by these clinical isolates ranged from less than 10 to 360 nmol of sialic acid released per min per mg of cell dry weight. Strains were arbitrarily classified as either nonproducers (less than 10 nmol/min per mg of cell dry weight), low producers (greater than 10 to less than or equal 140 nmol/min per mg of cell dry weight), or high producers (greater than 140 to 360 nmol/min per mg of cell dry weight). Type III isolates from diseased infants were significantly more often classified as high producers than strains of group B streptococci of other serotypes from diseased infants (P less than 0.001). Furthermore, the serotype III strains isolated from neonatal infections were more often high producers than those of the same serotype from asymptomatically colonized infants (P less than 0.025). These results suggest that the ability to produce elevated levels of neuraminidase may be related to the frequent association of type III strains with disease among neonates.