A fragment-based approach applied to a highly flexible target: Insights and challenges towards the inhibition of HSP70 isoforms.

The heat shock protein 70s (HSP70s) are molecular chaperones implicated in many cancers and of significant interest as targets for novel cancer therapies. Several HSP70 inhibitors have been reported, but because the majority have poor physicochemical properties and for many the exact mode of action is poorly understood, more detailed mechanistic and structural insight into ligand-binding to HSP70s is urgently needed. Here we describe the first comprehensive fragment-based inhibitor exploration of an HSP70 enzyme, which yielded an amino-quinazoline fragment that was elaborated to a novel ATP binding site ligand with different physicochemical properties to known adenosine-based HSP70 inhibitors. Crystal structures of amino-quinazoline ligands bound to the different conformational states of the HSP70 nucleotide binding domain highlighted the challenges of a fragment-based approach when applied to this particular flexible enzyme class with an ATP-binding site that changes shape and size during its catalytic cycle. In these studies we showed that Ser275 is a key residue in the selective binding of ATP. Additionally, the structural data revealed a potential functional role for the ATP ribose moiety in priming the protein for the formation of the ATP-bound pre-hydrolysis complex by influencing the conformation of one of the phosphate binding loops.

The crystal structures of macrophage migration inhibitory factor from Plasmodium falciparum and Plasmodium berghei.

Malaria, caused by Plasmodium falciparum and related parasites, is responsible for millions of deaths each year, mainly from complications arising from the blood stages of its life cycle. Macrophage migration inhibitory factor (MIF), a protein expressed by the parasite during these stages, has been characterized in mammals as a cytokine involved in a broad spectrum of immune responses. It also possesses two catalytic activities, a tautomerase and an oxidoreductase, though the physiological significance of neither reaction is known. Here, we have determined the crystal structure of MIF from two malaria parasites, Plasmodium falciparum and Plasmodium berghei at 2.2 A and 1.8 A, respectively. The structures have an alpha/beta fold and each reveals a trimer, in agreement with the results of analytical ultracentrifugation. We observed open and closed active sites, these being distinguished by movements of proline-1, the catalytic base in the tautomerase reaction. These states correlate with the covalent modification of cysteine 2 to form a mercaptoethanol adduct, an observation confirmed by mass spectrometry. The Plasmodium MIFs have a different pattern of conserved cysteine residues to the mammalian MIFs and the side chain of Cys58, which is implicated in the oxidoreductase activity, is buried. This observation and the evident redox reactivity of Cys2 suggest quite different oxidoreductase characteristics. Finally, we show in pull-down assays that Plasmodium MIF binds to the cell surface receptor CD74, a known mammalian MIF receptor implying that parasite MIF has the ability to interfere with, or modulate, host MIF activity through a competitive binding mechanism.