Multiplex PCR Assay for Direct Identification of Group B Streptococcal Alpha-Protein-Like Protein Genes

Abstract

We developed a group B streptococcus multiplex PCR assay which allows, by direct analysis of the amplicon size, determination of the surface protein antigen genes of alpha-C protein, epsilon protein, Rib, Alp2, Alp3, and Alp4. The multiplex PCR assay offers a rapid and simple method of subtyping Streptococcus agalactiae based on surface protein genes.

Streptococcus agalactiae (group B streptococcus [GBS]) is the leading pathogen in newborn infants in both developing and industrialized countries throughout the world, causing sepsis and meningitis. It also causes disease in parturient women and immunocompromised adults (18).

Epidemiological studies on S. agalactiae infections are mainly based on capsule serotyping. Currently, nine different S. agalactiae serotypes have been described (serotypes Ia, Ib, and II to VIII). Serotype distribution varies with geographical region and ethnic origin, and the virulence of clinical isolates with similar capsular composition can vary widely, suggesting that other bacterial virulence factors, in addition to capsule, are involved in the pathogenesis of GBS (19).

The major surface-localized protein antigens of group B streptococci belong to a family of surface proteins named the alpha-C protein, Rib, Alp2, Alp3, Alp4, and the epsilon protein (alpha-protein-like proteins). They contain large internal tandem repeats and are potential virulence factors (2, 5, 10, 12, 15, 16, 20). All these proteins are encoded by stable mosaic genes, generated by a recombination of modules at the same chromosomal locus (4, 12). The proteins exhibit size variation between strains depending on the number of repeats in the corresponding gene. Moreover, it has been demonstrated that, in the course of infection, the number of repeats inside the alpha-C proteins can undergo internal deletions as a means for evading the host immune response (14).

Protein typing of GBS is determined by the use of C and R antisera. Indeed, antisera are not protein specific (10, 12). While the alpha-C and Rib proteins are immunologically distinct from each other, the alpha-C, Alp2 and Alp3 (Alp2/3), and the epsilon proteins cross-react during Western blotting (11).

C antiserum also recognizes beta antigen, a GBS immunoglobulin A binding protein that was not included in this study because its structure is unrelated to that of the family of alpha-protein-like proteins (7, 8).

Protocols to identify serotypes and surface protein antigen genes by molecular methods have been recently published (3, 9, 10). For the protein subtyping, parallel PCRs using different primer pairs for each gene were set up; detection of epsilon gene had not previously been considered. We developed a multiplex PCR that allows determination of the following GBS surface protein genes directly by the analysis of the amplicon size: the alpha-C protein, the epsilon protein, Rib, Alp2/3, and Alp4.

While the repeats present extensive homology between alpha-protein-like proteins, the N-terminal portion is distinctive. The analysis of a ClustalW amino acid multisequence alignment of the N-terminal portions of the alpha-C and epsilon proteins, Rib, Alp2, Alp3, and Alp4 identified distinctive strings for each protein (Fig. 1), except for the Alp2 and Alp3 proteins, which are identical over the first half of their length.

FIG. 1.

Amino acid multisequence alignment of the N-terminal portions of S. agalactiae alpha-C protein, Alp2, Alp3, Rib, and the epsilon protein. The nucleotide sequence corresponding to the string from positions 11 to 17, indicated in grey, was used as the forward universal primer. The reverse primers sequences, specific for each protein gene (except for Alp2/3), are also indicated in grey.

Primer nucleotide sequences corresponding to the distinctive strings were used in a multiplex assay as the reverse primers, while a nucleotide string, common to all the surface protein genes, was used as the forward primer (Table 1). Briefly, total DNA preparations were prepared from GBS cultures, according to the instructions provided with a DNeasy tissue kit (QIAGEN), and 50 ng was used as a template in a final volume of 25 μl of PCR mixture containing the following: 1× PCR buffer; 2 mM MgCl₂; 200 μM concentrations of dATP, dCTP, dGTP, and dTTP; 400 nM concentrations of each of the six primers; and 0.25 U of TaqDNA polymerase (Life Technologies).

TABLE 1.

Nucleotide primer sequences and amplicon size expected for each S. agalactiae surface protein gene considered in this study

Primer	Sequence (5′-3′)	GenBank accession no.	Position from start codon (nt)a	Amplicon size (bp)
Universal forward	TGATACTTCACAGACGAAACAACG		30
Alpha-C reverse	TACATGTGGTAGTCCATCTTCACC	M97256	428	398
Rib reverse	CATACTGAGCTTTTAAATCAGGTGA	U583333	325	295
Epsilon reverse	CCAGATACATTTTTTACTAAAGCGG	U33554	230	200
Alp2/3 reverse	CACTCGGATTACTATAATATTTAGCAC	AF208158	364	334
Alp4 reverse	TTAATTTGCACCGGATTAACACCAC	AJ488912	140	110

^ant, nucleotides.

The samples were amplified on a DNA thermal cycler (MJ Research, Inc.) by heating for 3 min at 96°C, followed by 30 cycles of 95°C for 60 s, 58°C for 45 s, and 72°C for 45 s and concluding with a cycle of 72°C for 10 min.

The PCR products were analyzed by electrophoresis in a 2% (wt/vol) agarose-1000 gel (Life Technologies). UV transillumination of the bands on the agarose gel showed different sizes of amplicons, which allowed direct identification of the surface protein gene possessed by each GBS strain tested (Fig. 2).

FIG. 2.

Gel electrophoresis of multiplex PCR amplification products. Direct analysis of amplicon size allowed the determination of the genes encoding the surface protein of each GBS strain as follows: Rib (lanes 1, 2, and 8), alpha-C protein (lanes 3 and 10), Alp2/3 (lanes 4, 6, 9, and 12), Alp4 (lane 7), and the epsilon protein (lanes 5 and 1l). Lane 13 is the negative control. M, 2-log DNA ladder (0.1 to 10.0 kb; New England Biolabs, Inc.).

A serologically typed collection of GBS reference strains (10 strains), selected clinical isolates (55 strains), and bovine strains (9 strains) was analyzed by multiplex PCR, and results derived from the direct analysis of amplicon size were correlated with the gene sequence of the corresponding amplified product for each individual strain (Table 2). As noted above, C and R antisera cross-reacted with different protein genes. Moreover, as previously reported (4, 10), GBS strains can possess an alpha-protein-like protein gene even if the protein is not expressed on the surface. In two cases (reference strain Prague 10/84 and a bovine strain), no surface protein gene was detected. The absence of a surface protein gene in GBS isolates, although rare, has previously been reported (10).

TABLE 2.

Characteristics of GBS strains used in this study

Strain or isolate	Serotype	Alpha-protein-like proteina		Strain or isolate	Serotype	Alpha-protein-like proteina
Reference strains
NCTC 11078 (A909)	Ia/C	Alpha
090	Ia	Alp2b
NCTC 8187 (H36B)	Ib/C	Alpha
M 781	III/R	Rib
Prague 1/82	IV	Epsilon
Prague 10/84	V	Negative
Prague 118754	VI	Epsilon
Prague 7271	VII	Epsilon
JM9-130013	VIII/R	Alp3b
NCTC 9828	NT/R	Alp4
Clinical isolates
1319	Ia	Epsilon
1325	Ia	Epsilon
2129	Ib	Alpha
5518	Ib	Alpha
2088	Ia/C	Epsilon
5551	Ia/C	Alpha
1599	Ib/C	Alpha
1977	Ib/C	Alpha
5401	II	Alpha
3050	II	Rib
2452	II/C	Epsilon
5405	II/C	Alpha
2448	II/R	Rib
2141	II/R	Rib
5368	III	Rib
5435	III	Rib
5400	III	Alp2b
1056	III/C	Epsilon
1496	III/C	Epsilon
2601	III/C	Alpha
4383	III/R	Rib
1998	III/R	Rib
1999	IV	Epsilon
2274	IV	Alpha
1613	IV	Epsilon
2273	IV	Alpha
2620	IV	Epsilon
2102	V	Epsilon
2361	VII/R	Alp3b
2362	VII/R	Alp3b
5408	VIII	Alp2b
2179	VIII/C	Epsilon
1715	VIII/R	Rib
2198	VIII/R	Rib
2442	VIII/R	Alp2b
2925	NT	Alpha
2933	NT	Alpha
2450	NT	Alpha
2927	NT	Alpha
3002	NT	Rib
2277	NT/C	Alpha
2914	NT/C	Epsilon
2116	NT/R	Rib
5366	NT/R	Rib
Bovine strains
3492	III	Epsilon
3491	IV	Epsilon
3476	IV	Epsilon
3477	IV	Epsilon
3479	NT	Epsilon
3485	NT	Epsilon
3487	NT/C	Alpha
3490	NT	Epsilon
3488	NT	Negative
2628	V	Rib
2068	V	Alp3b
2110	V/C	Epsilon
2929	V/C	Alpha
5403	V/R	Rib
1989	VI	Epsilon
1992	VI	Epsilon
1990	VI/C	Epsilon
1994	VI/C	Epsilon
1272	VII/C	Alpha
2928	VII/C

^aThe designation of the alpha-protein-like protein encoded by the gene is based both on PCR results and sequencing of the PCR product.

^bThe differentiation between the proteins encoded by the alp2 and alp3 genes was performed by sequencing entire genes as described in reference 10.

Relationships between serotypes and surface protein genes were noted but to a lesser degree than previously reported (10, 12). The association of serotypes Ia, Ib, and II with the alpha protein, of serotype III with Rib, and of serotypes V and VIII with Alp3 was found, but these associations were not absolute. In particular, the alpha protein was present in most serotypes, Rib was detected in serotypes II, V, and VIII as well as in nontypeable strains, and Alp3 was also present in serotype VII. The Alp2 protein, first detected in serotype V (12) and subsequently detected in serotypes Ia and III (10), was also associated with type VIII in our study. No epidemiological studies surveyed the distribution of the epsilon protein that, in our study, was evenly distributed among GBS serotypes and prevalent in bovine strains.

The possibility of looking at the protein gene profile increases the potential of GBS subtyping by limiting the use of C and R antisera to the most appropriate strains after PCR has been used to ascertain protein expression.

A glycoconjugate vaccine to prevent GBS disease is being tested in clinical trials (1, 17). Promising data on using GBS surface proteins as protein carriers instead of tetanus toxoid have been published (6, 13). In this context, it is extremely important to have a rapid and reliable molecular test capable of determining the GBS surface protein genes by direct evaluation of amplicon size, which would permit extensive epidemiological studies on the profiles of protein subtypes of circulating GBS strains without consuming additional resources.