Résumé : Group A Streptococcus (GAS) is a human specific bacterial pathogen which causes a variety of diseases and is responsible for significant mortality worldwide. GAS are known to interact with the immune system, notably by binding host serum proteins to the bacterial surface.Many of these binding functions are attributed to the GAS M protein, the archetypal GAS virulence factor, the substrate for GAS typing and the leading GAS vaccine candidate. The vast diversity of GAS strains has however made vaccine development challenging. We investigated the potential for cross-protective immunity within closely related strains in a clinical setting in Fiji. This study has shown that immunity to GAS skin infection was broader than previously believed and included some level of cross-opsonisation betweenGAS strains. The level of such cross-opsonisation was, however, variable among GAS lineages. We have also shown that the immunity to conserved M antigens was quite variable. These results inspired us to investigate other suitable vaccine antigens. We hypothesised that in cases where the M protein was less immunodominant, perhaps closely related M-like proteins played an important role in immunity. We therefore began global characterisation ofthe two major M-like proteins, called Mrp and Enn. A representative worldwide genomic collection including more than 2000 isolates originating from multiple continents and various clinical manifestations has been established collaboratively. We focused our analyses on theMga regulon which encodes M and M-like proteins. We found that mrp and enn genes were present in 85% of genomes suggestive of their importance to GAS survival and spread. We developed molecular definitions of the different genes families, clarifying nomenclature forthe worldwide reference laboratory at the American CDC. We established and validated an updated, more specific typing protocol for GAS which will reduce future misclassifications. We have also analysed the genetic linkage between M and M-like protein alleles and developed clusters of closely related protein sequences. By characterising this complex and variable genetic region, we provide a framework for future functional investigations. Finally, we began functional characterisation of Enn proteins by investigating the differential capacity of Enn proteins to bind to C4BP, an inhibitor of complement. Altogether our results suggest M-like proteins play an important role in GAS virulence and should not be neglected. These data support further functional analyses to better understand the contribution of M-like proteins to GAS infection.