IDENTIFICATION OF PYRUVATE KINASE MODULATORS AS POTENTIAL THERAPEUTICS FOR SICKLE CELL DISEASE AND CANCER

Human pyruvate kinase enzyme (PK) plays a crucial role in the final step of glycolysis; catalyzing the transfer of phosphate group from phosphoenolpyruvate (PEP) to adenosine diphosphate (ADP) to form pyruvate and adenosine triphosphate (ATP) respectively. PK has four distinct isoenzymes in humans, including L, R, M1, and M2. The highly homologous L and R isoenzymes are expressed from the gene PKLR, while PKM1 and PKM2 are expressed from the gene PKM. PKLR is mainly found in red blood cell (RBC) and liver, involved in a major role of ATP synthesis and 2,3-bisphosphoglycertae (2,3-BPG) hemostasis. ATP is important for proper function of RBC, including deformability and stability, and a loss of ATP has a dire consequence on RBC health. PKLR is responsible for synthesizing 50% of ATP during glycolysis, and its deficiency in RBCs due to pathogenic mutation or complete loss of the enzyme leads to decreased ATP production, and in turn initiate RBC hemolysis and chronic nonspherocytic hemolytic anemia (CNSHA). In addition to producing ATP, PKLR is also involved in 2,3-BPG homeostasis. 2,3-BPG is a natural allosteric effector of the oxygen transport protein hemoglobin (Hb), allowing efficient offloading of oxygen to tissues. Despite PKLR importance, it is also linked to several disorders, most notably sickle cell disease (SCD) and liver cancer (HCC). SCD is an autosomal recessive genetic disorder that affects 100,000 individuals in the United States and millions worldwide. The pathophysiology of the disease is directly associated with a mutation in Hb resulting in sickle Hb (HbS). Under hypoxic condition, HbS tends to polymerize and aggregate, leading to sickling of red blood cells, which is the primary pathophysiology underlying the disease. The hypoxia-induced polymerization is made worse by high levels of 2,3-BPG in sickle RBC. In addition to increase production of ATP, activation of PKLR leads to depletion of 2,3-BPG, since it is downstream in the glycolytic cycle. From the foregoing, activation of PKLR, should be a viable treatment option for SCD. Hepatocellular carcinoma (HCC) is one of the most prevalent malignant tumors in the world, affecting about one in every 100 people. Cancer cells rewire their metabolism to increase production of ATP to promote growth, proliferation, and long-term maintenance. This phenomenon is known as the Warburg effect. Studies show that inhibiting PKLR to reduce ATP production has the ability to suppress cancer cell growth. For this reason, targeting PKLR should be a novel way to treat HCC. The objective of this project is to find PKLR modulators for the treatment of SCD and liver cancer. A virtual screening campaign was initiated to identify potential allosteric modulators of PKLR using AtomNet® technology and Mcule library. Concurrently, in-house virtual screening was also carried out using Mcule library. The study resulted in a total of 158 compounds from the ATOMWISE screen (both first and subsequent second-generation compounds), and six compounds from the in-house screen. The compounds were then studied for their effect on PKLR kinetic mechanism, mode of interaction with PKLR, and their ability to prevent cell proliferation. PKLR protein carrying an N-terminal His-tag was expressed in E. coli and purified using a Ni-NTA column to obtain yield of 5 mg/ml at 95% purity. The enzyme activity was determined at 37°C using the lactate dehydrogenase (LDH) coupled assay which was optimized to work in high throughput for studying the compounds modulatory effect on PKLR. The assay measures the change in UV/VIS absorbance at 340 nm due to oxidation of NADH by LDH while converting pyruvate to lactate. Several promising hits, including four inhibitors (DA1, DA2, DA3 and Mol-1), and two activators (DA4 and DA5) from initial virtual screening, and three inhibitors (DA6, DA7 and DA8) and two activators (DA9 and DA10) from subsequent virtual screening were discovered. The apparent of the substrate PEP (Khalf) and Vmax of the enzyme PKLR are 0.91±0.014 mM and 8.8±0.9 e-5 mM.S-1, respectively. In the presence of compounds DA1, DA2, DA3, DA4, DA5 and Mol-1, Khalf and Vmax ranged from 0.35-3.4 mM, and 1.3e-5-7.9e-5 mM.S-1, respectively. DA1, DA2 and Mol-1 were found to be non-competitive inhibitors, and DA3 a competitive inhibitor. As expected, DA1, DA2, DA3 and Mol-1 shifted the activity curve towards the low-affinity T-state. DA4 and DA5 showed an activation towards PKLR as both increased the activity of PKLR and shifted the activity curve towards the high-affinity R-state. The IC50 values for the inhibitors DA1, DA2, and DA3 were 4.0±0.33μM, 5.2±0.74μM, and 10.4±0.30μM respectively, while the EC50 for the activators DA4 and DA5 were 10.8±0.81 μM and 19.5±1.9μM, respectively. For the second-generation compounds, the IC50 for DA6 increased by 8-fold in comparison with the parent compound DA1, whereas the IC50 of DA7 and DA8 increased by 3-fold and 6-fold, respectively compared to the parent compound DA2. Unlike the first-generation compounds, the second-generation activators, DA9 and DA10 did not show significant increase in potency compared to the parent compound DA4. Likewise, the IC50 for Mol-1 was 16.0±1.1µM. The dissociation constant (KD) between PKLR and the compounds DA1, DA2, DA3, DA4 and DA5 were determined using Microscale Thermophoresis Technology (MST) and showed low to moderate affinity of 4-23µM, with DA5 showing the highest affinity (3.8±0.86 µM), while DA1 showed the weakest affinity (23±0.5 µM). Interestingly, the two parameters, KD and IC50/EC50 correlated well with the exception of DA1, which even though showed the most potent effect (IC50= 4.0±0.33 µM), exhibited the weakest binding affinity (23±0.5 µM). Two of the inhibitors, DA1 and DA2, were tested in cell-based assay using the wound-healing and (CCK-8) assay to determine their antiproliferative activities. Both compounds showed dose-dependent antiproliferative activity in Hep3B cell with IC50 of 10±0.5 µM, and 20±0.9µM respectively. Several of the second-generation potential inhibitor hits are undergoing similar studies. X-crystallographic studies were attempted to study the compounds’ mode of interactions with PKLR and to guide targeted modification. The study was unsuccessful due to poor crystal diffraction. Computational studies were used as alternate method to study the compounds putative interactions with PKLR. The docking was guided by the kinetic results, which led to four of the compounds DA1, DA2, DA4 and DA5 being docked at the AG-348 binding site, and one compound DA3 being docked to the PEP binding site. The putative interactions allowed for speculating on the differing mechanism of action of these compounds. It is expected that the totality of the results from this study and other ongoing studies will steer towards the identification of lead PKLR modulators that have potential to be developed or serve as leads for SCD or cancer therapeutics