Biophysical, biochemical and inhibition studies of hexokinases

Abstract

Hexokinase is the first enzyme in glycolysis, a major pathway for the generation of energy in all eukaryotes. Mammalian cells have four isoforms (I, II, III, IV) that have different tissue distribution and kinetic properties. Among all isoforms, human hexokinase II (hHKII) has been found to be implicated in many cancers with an increased expression which serves a dual role. First, it maintains the high glycolytic rate of malignant cells (Warburg effect) and second it prevents apoptosis when is bound to mitochondria. Trypanosoma brucei is a parasite that causes Human African Trypanosomiasis (HAT) and has two isoforms with extensive sequence similarity (98%), TbHKI (active form) and TbHK2 (inactive form). The bloodstream-form parasites (BSF) depend exclusively on glycolysis for their survival. The enzyme from both organisms is a validated target for drug-discovery against both cancer and HAT. The aim of the present study is the discovery of novel and specific inhibitors of the enzymes based on their structure. Structure-based drug discovery is commonly used in pharmaceutical companies to aid in the discovery of potent lead compounds.

In silico studies were performed in this project using the known crystal structure of human hexokinase I and a model of TbHKI generated by the protein modelling tool Phyre2. The docking programs, AutoDock (AD) and AutoDock Vina (Vina), were chosen to perform the docking of ~3 million compounds to the target molecules and scoring functions calculated the predicted binding affinities of each compound. In total, 28 compounds were purchased to test on the target molecules.

In the experimental part of the project, the two enzymes were cloned, expressed and purified. hHKII was successfully purified giving a high yield of active and pure protein. The protein was characterised using many biophysical methods to establish the oligomeric state, the homogeneity and the secondary structure.

Crystallisation trials failed and for this reason, N and C domains of the hHKII were purified separately. Unfortunately, the domains also failed to crystallise thus SAXS data were collected and analysed to gain information of their shape at low resolution. A novel inhibition assay was developed and used to identify four weak inhibitors against full length hHKII.

TbHKI was difficult to express in a soluble form as most of the protein was expressed in inclusion bodies. The purification resulted in a small amount of active protein that was used entirely for biochemical assays. Four compounds were purchased from the docking of the TbHKI model and one was found to inhibit the enzyme over 65% at 100 μΜ. Because the active site of both enzymes (hHKII, TbHKI) is well conserved the compounds from hHKII docking were also screened against the TbHKI. Four compounds were found to inhibit this enzyme while one of them was also an inhibitor for human isoform. The remaining three were specific for inhibition of TbHKI.