TEXT

Top Abstract Text References

Streptococcus agalactiae (group B streptococci) (GBS) is an important cause of infections in humans, particularly neonatal infections. Serotyping is widely used in epidemiological studies of GBS infections and has traditionally been based on capsular polysaccharide antigens of which the antigens Ia, Ib, and II through VIII are presently known. In addition, the surface-anchored c proteins c and c (5) and the R proteins R1 through R4 (20) are important GBS serotype markers. Serotyping based on both the capsular polysaccharide and the protein antigens enable subdivision of GBS into a variety of serovariants (8). Both the capsular antigens and the surface-localized proteins function as virulence factors and as targets of protective antibodies (1, 10, 11, 17, 18).

The genes encoding the GBS protein serotype markers c, c, and the protein Rib, which may be identical to the R4 protein (3), have been sequenced (7, 15, 19). The beta gene product, the c protein, has a signal sequence, an N terminus, a cell wall- and cell membrane-spanning C-terminal domain, and two functional domains (A and B) which mediate binding of immunoglobulin A (7). The bca gene encoding the c protein is different and, in the GBS strain A909, encodes a signal sequence, an N terminus, nine identical tandem repeats (82 amino acids in each repeat), a partial repeat, and a C terminus (15). The bca gene may undergo mutational deletion of repeats, and this results in mutants which may escape immune clearance (9). PCRs for detection of both the bca and beta genes have been described (9, 13, 14).

Over the years, reference and prototype strains of GBS have been selected solely on the basis of phenotypic traits such as surface-exposed proteins defined on the basis of immunoprecipitation testing. However, as the technology makes it possible, data concerning phenotypes of these strains should be supplemented by data concerning genotypes. The importance of molecular analyses of such strains has recently been emphasized by the findings that clinical GBS strains which are negative in fluorescent antibody testing (FAT) of proteins c and c nevertheless may harbor corresponding gene elements (12, 13). In particular, bca/c protein discrepancy occurred quite frequently (12). Since properties of strains are crucial in the context of research and reagent developments, such as generation of antibodies for serotyping, we have tested our collection of reference and prototype GBS isolates for the presence of the bca and beta genes and expression of the products of these genes.

GBS was cultured on blood agar plates or in Todd-Hewitt broth. The beta gene PCR was performed as described previously using a primer set which resulted in the amplification of a 620-bp product (13). The PCR product included a part of each of the domains A and B of the beta gene (7). For the bca gene, primers were designed that supported amplification of a 202-bp region within the bca repeat unit (12). The PCR procedure, including detection of amplification products by agarose gel electrophoresis, was performed as described (12, 13). FAT for serotyping and Western blotting of sodium dodecyl sulfate-extracted whole cells of GBS were performed as described using polyclonal and/or monoclonal anti-c and -c antibodies (2, 4, 16). For Western blotting, the material which was sodium dodecyl sulfate extracted from 50 µg of lyophilized bacteria was applied per lane.

Complete agreement was found between the antibody-based detection of the c protein and the PCR for beta gene detection for the GBS strains listed in Table 1. Positive c FAT was shown by 3 of 14 GBS strains (category A of c FAT and bca PCR reactivity pattern) (Table 1), in accordance with the established GBS serotypes. However, 7 of the 11 c FAT-negative strains tested positive for bca by PCR, four of them demonstrating the category B pattern with two or more PCR products according to product size, and three of them demonstrated the category C pattern with a single amplification product of the minimum size of 202 bp. Examples of the various banding patterns and categories are shown in Fig. 1. The category B and C strains also failed to show c protein expression in the whole-cell-based immunoblotting when probed against polyclonal and monoclonal anti-c antibodies, respectively. These results were repeatedly obtained for all of the isolates.

View larger version (47K): [in this window] [in a new window]   FIG. 1.   (a) Gel electrophoresis of products of bca PCR of GBS strains ATCC 12401 (lane 1), NCTC 12907 (lane 2), ATCC 12400 (lane 3), and NCTC 11079 (lane 4). C FAT and bca categories are indicated. (b) Products of beta gene PCR of strains ATCC 12401 (lane 1) and 15626/81 (lane 2). DNA standards are shown in the left lanes.

We have previously confirmed by hybridization using internal probes that the bca and beta gene elements, respectively, were amplified with the primer sets used in this study (12, 13). For bca, this has also been confirmed by sequence analysis of PCR products of the C category (12). Previously, we found that the category C pattern was consistent with a bca gene with only one repeat unit and that category A and B patterns were compatible with two or more repeats, the multiplicity of PCR products probably resulting from amplification which extended over a variable number of repeats (12).

Other investigators have shown that mutational deletion of bca repeats may occur both in vitro and in vivo (9). Conceivably, the mutations have adaptive importance since the resulting c proteins, with a reduced number of repeats, show loss of protective epitopes and the mutants show reduced susceptibility to opsonophagocytic killing by c antibodies (6, 9). We assume that all or some of the seven category B and C strains described in this study have been subjected to similar mutational deletions of bca repeats. These strains, including that single clone, have been repeatedly subcultured for years. It is possible that the mutational changes have occurred during the subculturing, and that these changes have benefitted the bacteria when growing in vitro. Alternatively, the mutational changes could have been present at the time of the original isolation. Recently, we showed that of 40 clinical GBS isolates which were negative in the c FAT, 16 strains showed positive PCR for bca elements (12). None of the seven prototype and reference GBS strains with c FAT and bca PCR discrepancy showed c protein expression at levels which could be detected by the antibody-based methods used in this study, consistent with the results of earlier serotyping. This may be due to additional mutational changes, for instance in a regulatory region. Alternatively, one or more of these strains may express a c protein with conformational changes that excluded recognition by the anti-c antibodies used by us (6). These antibodies were raised against a c protein with many repeats (2) which is antigenically different from c with few repeats (6). In contrast to the genotype and phenotype divergency of c, PCR testing of the beta gene showed complete agreement with the testing of the c protein, although beta gene/c protein discrepancy may occur (13).

Our findings underscore the need for characterization of reference and prototypes strains of GBS at the genetic level and highlight problems involved when selecting strains for research purposes or for the generation of diagnostic reagents such as antibodies against serogroup or serotype markers. For instance, the possibility cannot be excluded that one or more of the strains which contained bca without c protein expression may express the protein when exposed to environmental changes in vitro or in vivo.