LIG_CSK_EPIYA_1

Several pathogenic bacteria use SH2 domains (Pfam:PF00017) in host proteins to mediate binding of tyrosine-phosphorylated effector proteins (Backert,2005). For example, Helicobacter pylori CagA (Q9RF15) binds to the SH2 domain of C-terminal Src kinase (Csk) (with a KD<1120 nM; Tsutsumi,2003) as well as to both SH2 domains in SHP-2 (Higashi,2002); Tarp protein from Chlamydia trachomatis serovar D (O84462) binds to the SH2 domain of SHC1 (with a KD<200 nM; Mehlitz,2010) while the homologous Tarp protein from C. trachomatis serovar L2 binds to the SH2 domain of Nck (Mehlitz,2010); as a last example, AnkA protein from Anaplasma phagocytophilum (A0S0L6) binds to the SH2 domain of SHP-1 (IJdo,2007).

In all the mentioned cases the interaction is dependent on a tyrosine phosphorylation that occurs inside the host cell after the translocation of the bacterial protein. As in the case of CagA, the phosphotyrosines have been described as promiscuous binders of up to five partner proteins, as shown by a SILAC experiment using peptides of 15 amino acids long with tyrosine in the central position of different bacterial effectors like Bartonella henselae BepD (Q5QT02), BepE (Q5QT01) and BepF (Q5QT00), or Enteropathogenic Escherichia coli Tir protein (Selbach,2009). Nevertheless, the adjacent sequence to the tyrosine can give the specificity to certain proteins, which can explain the presence of different tyrosines able to be phosphorylated along the sequence of these proteins.

The specificity for the binding of the mentioned Csk protein, a non-receptor tyrosine kinase involved in cell growth, differentiation, migration and immune response (Roskoski R,2015), is defined by the positions -3 to +1 of the tyrosine. This linear motif – consensus EPIYA - is found in predicted intrinsically disordered polypeptide (IDP) and recognizes the SH2 domain in an unknown way (Nesic,2010), though it is dependent on the presence of a serine at the position 109 in rat Csk (P32577; Tsutsumi,2003; Safari,2011). At time of creating this entry, there is no structure of a solved EPIYA-SH2 complex.

This five residue-long motif also mediates the binding to CSK of the mammalian protein Pragmin (D3ZMK9; Safari,2011; also named as Sgk223 in human) which is a negative regulator of Src family kinases (SFK). This highly conserved motif is in a region of predicted IDP. (EPIYA-like motifs are also reported in the human SRC kinase signaling inhibitor 1 (SNIP or p140Cap; Q9C0H9; Repetto,2013) but it is unclear if the motifs are in an accessible structural context.). Moreover, the motif is conserved in homologous proteins of Pragmin and p140Cap along mammals.

The tyrosine in the EPIYA motif has been shown to be phosphorylated by Csk in the case of Pragmin, so a feed-forward regulatory loop is created (Senda,2016); Csk was also found to be able to phosphorylate EPIYA of LspA1 protein from Haemophilus ducreyi (Q7VLE8) in vitro (Dodd,2014). Finally, CagA is reported to be phosphorylated by members of the Src Family Kinases (SFK) such as c-Src (Selbach,2002) and Lyn (Selbach,2002; Stein,2002).

The interaction of Csk with the phosphorylated LIG_SH2_Csk motif can lead to different and opposing scenarios. The binding of the cytoplasm-localized phosphorylated Pragmin sequesters Csk far from the membrane, preventing the inactivation of c-Src. On the other hand, the protein CagA is initially located next to the intracellular membrane and brings together Csk with c-Src (Safari,2011), a state that allows Csk to phosphorylate c-Src in a conserved tyrosine at position 530 in the human homolog. This leads c-Src to adopt a closed conformation where its kinase activity is inhibited (Roskoski R,2004). In a celullar system where SFKs are inactivated, substrates of this kinase like the Fcγ receptor in the immunoreceptor tyrosine-based activation motif (ITAM) domain are not phosphorylated and phagocytosis is deregulated, a cellular state abused by H. ducreyi using its EPIYA-containing effector protein LspA1 (Dodd,2014). There is surely much more to discover about the mechanisms whereby pathogenic EPIYA hijack perturbs these tyrosine kinase signalling systems.