Focal adhesion kinase (FAK) and its close paralogue, proline-rich tyrosine kinase 2 (PYK2), are key regulators of aggressive spreading and metastasis of cancer cells. While targeted small-molecule inhibitors of FAK and PYK2 have been found to have promising antitumor activity, their clinical long-term efficacy may be undermined by the strong capacity of cancer cells to evade anti-kinase drugs. In healthy cells, the expression and/or function of FAK and PYK2 is tightly controlled via modulation of gene expression, competing alternatively spliced forms, non-coding RNAs, and proteins that directly or indirectly affect kinase activation or protein stability. The molecular factors involved in this control are frequently deregulated in cancer cells. Here, we review the endogenous mechanisms controlling FAK and PYK2, and with particular focus on how these mechanisms could inspire or improve anticancer therapies.

Significance: Phosphoinositide 3-kinase activation typically occurs following stimulation by upstream receptor tyrosine kinases (RTKs), which alleviate p110α inhibition by p85α. p85α and p110α driver mutations have been reported to activate p110α by disrupting the inhibitory interface between p85α and p110α. This study revealed that driver mutations in the p85α N-terminal SH2 domain can enhance p110α activity by inducing the activation of multiple RTKs. Furthermore, combination treatment with RTK and AKT inhibitors provides synergistic therapeutic efficacy. This previously uncharacterized oncogenic mechanism presents the exploitable vulnerability of a class of p85α mutant tumors.

The Nef protein of human and simian immunodeficiency viruses boosts viral pathogenicity through its interactions with host cell proteins. By combining the polyvalency of its large unstructured regions with the binding selectivity and strength of its folded core domain, Nef can associate with many different host cell proteins, thereby disrupting their functions. For example, the combination of a linear proline-rich motif and hydrophobic core domain surface allows Nef to bind tightly and specifically to SH3 domains of Src family kinases. We investigated whether the interplay between Nef's flexible regions and its core domain could allosterically influence ligand selection. We found that the flexible regions can associate with the core domain in different ways, producing distinct conformational states that alter the way in which Nef selects for SH3 domains and exposes some of its binding motifs. The ensuing crosstalk between ligands might promote functionally coherent Nef-bound protein ensembles by synergizing certain subsets of ligands while excluding others. We also combined proteomic and bioinformatics analyses to identify human proteins that select SH3 domains in the same way as Nef. We found that only 3% of clones from a whole-human fetal library displayed Nef-like SH3 selectivity. However, in most cases, this selectivity appears to be achieved by a canonical linear interaction rather than by a Nef-like ‘tertiary' interaction. Our analysis supports the contention that Nef's mode of hijacking SH3 domains is a virus-specific adaptation with no or very few cellular counterparts. Thus, the Nef tertiary binding surface is a promising virus-specific drug target.

The Nef protein of human and simian immunodeficiency viruses (HIV and SIV, respectively) boosts viral pathogenicity through its interactions with host cell proteins. Nef has a folded core domain and large flexible regions, each carrying several protein interaction sites. By combining the polyvalency intrinsic to unstructured regions with the binding selectivity and strength of a 3D folded domain, Nef can bind to many different host cell proteins, perturbing their cellular functions. For example, the combination of a linear proline-rich motif and a hydrophobic core domain surface allows Nef to increase affinity and selectivity for particular Src family SH3 domains. Here we investigated whether the interplay between Nef’s flexible regions and its core domain can allosterically influence ligand selection. We found that the flexible regions can bind back to the core domain in different ways, producing distinct conformational states that alter the SH3 domain selectivity and availability of Nef’s functional motifs. The resulting cross-talk might help synergising certain subsets of ligands while excluding others, promoting functionally coherent Nef-bound protein ensembles. Further, we combined proteomic and bioinformatic analyses to identify human proteins that select SH3 domains in the same way as does Nef. We found that only 2–3% of clones from a whole human fetal library displayed a Nef-like SH3 selectivity. However, in most cases this selectivity appears to be achieved by a canonical linear interaction rather than a Nef-like ‘tertiary’ interaction. This analysis suggests that Nef’s SH3 recognition surface has no (or marginally few) cellular counterparts, validating the Nef tertiary binding surface as a promising unique drug target.