Heme-regulated inhibitor (HRI) is one of the four mammalian kinases that phosphorylate eIF2α, facilitating a cellular response to stress through the regulation of mRNA translation. Originally identified as a heme sensor in erythroid progenitor cells, HRI has since emerged as a potential therapeutic target in both cancer and neurodegeneration. Here, we characterise two modes of HRI inhibition using structural mass spectrometry, biochemistry, and biophysics. We examined several competitive ATP-mimetic inhibitors – dabrafenib, encorafenib, and GCN2iB – and compared them with the heme-mimetic allosteric inhibitor, hemin. By combining hydrogen–deuterium exchange mass spectrometry with protein models generated by AlphaFold 3, we investigated the structural basis of inhibition by dabrafenib and hemin. Our analysis revealed that hemin inhibition induces large-scale structural rearrangements in HRI, which are not observed with ATP-mimetic inhibitors. Our results suggest that HRI may be inhibited using two distinctly different modalities, which may guide future drug development.
Recently it has been found that General Control Non-derepressible 2 (GCN2) can be activated by an array of small molecule ATP-competitive inhibitors, including clinically relevant compounds such as Ponatinib, and compounds specifically designed to be GCN2 inhibitors, such as GCN2iB. Furthermore, we recently showed that GCN2 can be activated in cells by clinically approved small molecule RAF inhibitors. GCN2 is a drug target, specifically in cancers such as mesothelioma, and a better understanding of this paradoxical activation is required to develop drugs which truly inhibit the enzyme. Using biochemical assays and structural mass spectrometry, we present a model for how GCN2 is activated by these compounds by promoting an active conformation in the HisRS domain while competitively inhibiting the kinase domain. This conformation promotes activating phosphorylation of GCN2, potentially through phosphorylation of other activated GCN2 molecules which are not bound to compound. Together this model suggests that efforts to inhibit GCN2 would benefit from exploring allosteric routes rather than targeting the ATP-binding pocket of the kinase domain.
Heme-Regulated Inhibitor (HRI) is one of the four mammalian kinases which phosphorylates eIF2α to facilitate a cellular response to stress through the regulation of mRNA translation. Originally identified for its role as a heme sensor in erythroid progenitor cells, it has since materialised as a potential therapeutic target in both cancer and neurodegeneration. Here we characterise two modes of HRI inhibition of using structural mass spectrometry, biochemical and biophysical techniques. We demonstrate that several ATP-mimetic compounds, including BRAF inhibitors and a compound, GCN2iB, thought to be specific to GCN2, are capable of potently inhibiting HRI. We demonstrate that hemin, a haem-like molecule, inactivates HRI structurally using hydrogen-deuterium exchange mass spectrometry (HDX-MS), and this results in wide-spread structural rearrangement of the protein and how that impacts on the kinase domain through a series of allosteric interactions. This inhibition mainly impacts autophosphorylation, which includes tyrosine phosphorylation, not observed before in the eIF2α kinases.