The biochemical mechanism and pharmacological inhibition of phosphatidylinositol phosphate kinases

Project Summary / Abstract The objective of the proposed research is to elucidate the biochemical mechanisms underlying the exquisite substrate binding and catalytic specificity of two phosphatidylinositol phosphate 5-kinases (PIP5K, PIKfyve). The PIPK family of lipid kinases include PIP5K (type 1), PIP4K (type 2) and PIKfyve (type 3), and is primarily responsible for converting phosphatidylinositol monophosphate lipids into PI(4,5)P2 and PI(3,5)P2. Despite sequence homology, these kinases are highly selective in substrate binding [PIP5K binds PI(4)P, PIP4K binds PI(5)P, and PIKfyve binds PI(3)P] and in catalytic activity [PIP5K and PIKfyve phosphorylate the C5 hydroxyl of the lipid's inositol head group, whereas PIP4K phosphorylates the C4 hydroxyl]. We and others have previously identified two structural elements within the kinase domain, the specificity loop and a conserved PIP-binding motif, that contribute to substrate selectivity, but how these two elements cooperate to confer kinase specificity remains undefined at the structural level. In aim 1, we plan crosslinking strategies to stabilize the specificity loop to facilitate co-crystallization with lipid substrates. We also plan to generate and crystallize a minimalistic catalytic core domain of PIKfyve. In aim 2, we propose genetic and chemical biological experiments to examine the role of PIKfyve in the life cycle of SARS-CoV-2. Several large-scale drug repurposing programs have identified apilimod, a PIKfyve inhibitor, as a top lead in suppressing SARS-CoV-2 replication in cell culture (a phase II clinical trial of apilimod in treating COVID-19 is ongoing at the Yale Center for Clinical Investigation). This discovery, together with earlier observations that apilimod also reduces infection by Ebola and Marburg viruses, has generated great interest in pharmacologically targeting PIKfyve. Drawing on structural and biochemical knowledge about the lipid kinase family, as well as chemical tools previously developed to target PIP4K, we have discovered a new class of potent PIKfyve inhibitors and plan to use them together with apilimod to interrogate how PIKfyve inhibition disrupts SARS-CoV-2 infection. All previously known PIKfyve inhibitors are structurally related to apilimod, and their binding mode to the lipid kinase is unknown. The new inhibitor class is significant because it not only adds confidence to the proposed involvement of PIKfyve in SARS-CoV-2 infection, but also has a known binding mode to PIPK, which should facilitate future optimization by medicinal chemistry.