emm and sof gene sequence variation in relation to serological typing of opacity-factor-positive group A streptococci

Approximately 40-60% of group A streptococcal (GAS) isolates are capable of opacifying sera, due to the expression of the sof (serum opacity factor) g ene. The emm (M protein gene) and sof 5' sequences were obtained from a div erse set of GAS reference strains and clinical isolates, and correlated wit h M serotyping and anti-opacity-factor testing results. Attempts to amplify sof from strains with M serotypes or emm types historically associated wit h the opacity-factor-negative phenotype were negative, except for emm12 str ains, which were found to contain a highly conserved sof sequence. There wa s a strong correlation of certain M serotypes with specific emm sequences r egardless of strain background, and likewise a strong association of specif ic anti-opacity-factor (AOF) types to sof gene sequence types. In several e xamples, M type identity, or partial identity shared between strains with d iffering emm types, was correlated with short, highly conserved 5' emm sequ ences likely to encode M-type-specific epitopes. Additionally, each of thre e pairs of historically distinct M type reference strains found to share th e same 5' emm sequence, were also found to share M serotype specificity. Ba sed upon sof sequence comparisons between strains of the same and of differ ing AOF types, an approximately 450 residue domain was determined likely to contain key epitopes required for AOF type specificity. Analysis of two So f sequences that were not highly homologous, yet shared a common AOF type, further implicated a 107 aa portion of this 450-residue domain in putativel y containing AOF-specific epitopes. Taken together, the serological data su ggest that AOF-specific epitopes for all Sof proteins may reside within a r egion corresponding to this 107-residue sequence. The presence of specific, hypervariable emm/sof pairs within multiple isolates appears likely to be a reliable indicator of their overall genetic relatedness, and to be very u seful for accurate subtyping of GAS isolates by an approach that has releva nce to decades of past M-type-based epidemiological data.