Multiple conformational states of the HPK1 kinase domain in complex with sunitinib reveal the structural changes accompanying HPK1 trans-regulation

A perspective on HPK1 as a novel immuno-oncology drug target

“…The initial clue that the kinase activity of HPK1 plays a role in inhibiting TCR-induced activation signal came when it was reported that the ectopic expression of HPK1 in Jurkat T cells reduced TCR-induced Erk MAPK and AP-1 activation while the expression of a catalytically inactivated point mutant form of HPK1 could not (Liou et al, 2000). This early finding is consistent with our early observation that Jurkat cells, when treated with Sunitinib malate (Sutent), could produce more IL-2 in response to TCR engagement, resulting in enhanced phosphorylation of Erk MAPK and elevated IL-2 production [unpublished observation and (Johnson et al, 2019)]. Sutent, the multi-kinase inhibitor was capable of inhibiting HPK1 kinase activity in vitro at an IC 50 of 15 nM (Anastassiadis et al, 2011;Davis et al, 2011).…”

“…The caspase-cleaved N-terminal domain of HPK1 becomes catalytically active upon separation from its C-terminal fragment, which suggest that the C-terminal region of HPK1 is involved in the auto-or trans-regulation of HPK1 kinase activity. This speculation is strengthened by the recent X-ray crystallographic studies revealing that the three-dimensional structure of HPK1 kinase domain exists as a head-to-head domain-swapped dimer in both the crystal form, as well as in the liquid form Wu et al, 2019;Johnson et al, 2019. In the non-phosphorylated, inactive state, the activation loop of each HPK1 monomer partially occupies the ATP-and substrate-binding sites of its partner monomer, thus serving as a possible mechanism to quiesce the basal kinase activity of HPK1. This trans-inhibitory posture of the activation loop is disrupted when the threonine residue 165 and serine residue 171 are phosphorylated (Johnson et al, 2019).…”

“…This speculation is strengthened by the recent X-ray crystallographic studies revealing that the three-dimensional structure of HPK1 kinase domain exists as a head-to-head domain-swapped dimer in both the crystal form, as well as in the liquid form Wu et al, 2019;Johnson et al, 2019. In the non-phosphorylated, inactive state, the activation loop of each HPK1 monomer partially occupies the ATP-and substrate-binding sites of its partner monomer, thus serving as a possible mechanism to quiesce the basal kinase activity of HPK1. This trans-inhibitory posture of the activation loop is disrupted when the threonine residue 165 and serine residue 171 are phosphorylated (Johnson et al, 2019). This finding demonstrates that the dynamic structural positioning of the HPK1 activation loop is controlled by conformational changes induced by phosphorylation of HPK1.…”

“…The uncommon domain-swapped dimeric conformation and the structural changes accompanying HPK1 trans-regulation revealed by these studies may facilitate future insights regarding the strategy to enhance selectivity of the small molecule inhibitor of HPK1. (Wu et al, 2019;Johnson et al, 2019;Sawasdikosol and Burakoff, 2019). As an alternative approach, large pharmaceutical companies with sufficient resources might engage in an unbiased high-throughput screen of their small molecule libraries for allosteric inhibitors of HPK1.…”

“…In the non-phosphorylated, inactive state, the activation loop of each HPK1 monomer partially occupies the ATP- and substrate-binding sites of its partner monomer, thus serving as a possible mechanism to quiesce the basal kinase activity of HPK1. This trans-inhibitory posture of the activation loop is disrupted when the threonine residue 165 and serine residue 171 are phosphorylated ( Johnson et al, 2019 ). This finding demonstrates that the dynamic structural positioning of the HPK1 activation loop is controlled by conformational changes induced by phosphorylation of HPK1.…”