Activation of Helicobacter pylori CagA by tyrosine phosphorylation is essential for dephosphorylation of host cell proteins in gastric epithelial cells.

Helicobacter pylori type I strains harbour the cag pathogenicity island (cag-PAI), a 37 kb sequence,which encodes the components of a type IV secretion system. CagA, the first identified effector protein of the cag-PAI, is translocated into eukaryotic cells and tyrosine phosphorylated (CagAP-tyr) by a host cell tyrosine kinase. Translocation of CagA induces the dephosphorylation of a set of phosphorylated host cell proteins of unknown identity. CagA proteins of independent H. pylori strains vary in sequence and thus in the number and composition of putative tyrosine phosphorylation motifs (TPMs). The CagA protein of H. pylori strain J99 (CagAJ99) does not carry any of three putative tyrosine phosphorylation motifs (TPM-A, TPM-B or TPM-C) predicted by the MOTIF algorithm in CagA proteins. CagA,n is not tyrosine phosphorylated and is inactive in the dephosphorylation of host cell proteins. By site-specific mutagenesis,we introduced a TPM-C into CagA,. by replacing a single lysine with a tyrosine. This slight modification resulted in tyrosine phosphorylation of CagAJ99 and host cell protein dephosphorylation. In contrast, the removal of the indigenous TPM-C from CagAP12 did not abolish its tyrosine phosphorylation, suggesting that further phosphorylated sites are present in CagAP12. By generation of hybrid CagA proteins, a phosphorylation of the most N-terminal TPM-A could be excluded. Our data suggest that tyrosine phosphorylation at TPM-C is sufficient, but not exclusive,to activate translocated CagA. Activated CagAPtr might either convert into a phosphatase itself or activate a cellular phosphatase to dephosphorylate cellular phosphoproteins and modulate cellular signalling cascades of the host.