Qingyun Ren

Preclinical Characterization of GLS4, an Inhibitor of Hepatitis B Virus Core Particle Assembly

ABSTRACT Hepatitis B virus (HBV)-associated chronic liver diseases are treated with nucleoside analogs that target the virus polymerase. While these analogs are potent, drugs are needed to target other virus-encoded gene products to better block the virus replication cycle and chronic liver disease. This work further characterized GLS4 and compared it to the related BAY 41-4109, both of which trigger aberrant HBV core particle assembly, where the virus replication cycle occurs. This was done in HepAD38 cells, which replicate HBV to high levels. In vitro, GLS4 was significantly less toxic for primary human hepatocytes (P < 0.01 up to 100 μM), inhibited virus accumulation in the supernantant of HepAD38 cells (P < 0.02 up to 100 nM), inhibited HBV replicative forms in the liver with a significantly lower 50% effective concentration (EC50) (P < 0.02), and more strongly inhibited core gene expression (P < 0.001 at 100 to 200 nM) compared to BAY 41-4109. In vivo characterization was performed in nude mice inoculated with HepAD38 cells, which grew out as tumors, resulting in viremia. Treatment of mice with GLS4 and BAY 41-4109 showed strong and sustained suppression of virus DNA to about the same extents both during and after treatment. Both drugs reduced the levels of intracellular core antigen in the tumors. Alanine aminotransferase levels were normal. Tumor and total body weights were not affected by treatment. Thus, GLS4 was as potent as the prototype, BAY 41-4109, and was superior to lamivudine, in that there was little virus relapse after the end of treatment and no indication of toxicity.

Discovery of hepatitis B virus capsid assembly inhibitors leading to a heteroaryldihydropyrimidine based clinical candidate (GLS4).

Inhibition of hepatitis B virus (HBV) capsid assembly is a novel strategy for the development of chronic hepatitis B (CHB) therapeutics. Herein we described our lead optimization studies including the synthesis, molecular docking studies and structure-activity relationship (SAR) studies of a series of novel heteroaryldihydropyrimidine (HAP) inhibitors of HBV capsid assembly inhibitors, and the discovery of a potent inhibitor of HBV capsid assembly of GLS4 (ethyl 4-[2-bromo-4-fluorophenyl]-6-[morpholino-methyl]-2-[2-thiazolyl]-1,4-dihydro-pyrimidine-5-carboxylate) which is now in clinical phase 2. GLS4 demonstrated potent inhibitory activities in HBV HepG2.2.15 cell assay with an EC50 value of 1nM, and it also exhibited high potency against various drug-resistant HBV viral strains with EC50 values in the range of 10-20nM, more potent than the typical HBV polymerase inhibitors such as lamivudine, telbivudine, and entecavir. Pharmacokinetic profiles of GLS4 were favorable and safety evaluation including acute toxicity and repeated toxicity study indicated that GLS4 was safe enough to support clinical experiments in human.

Influences on viral replication and sensitivity to GLS4, a HAP compound, of naturally occurring T109/V124 mutations in hepatitis B virus core protein†

Heteroaryldihydropyrimidine (HAP) compounds inhibit HBV replication by binding to a hydrophobic pocket at the interface between hepatitis B virus core protein (HBcAg) dimer, which interrupts capsid assembly by changing the kinetics and thermodynamics during this process. Structure biological studies have identified several amino acids in HBcAg crucial for compound binding. Here, we investigated the polymorphisms of T109 and V124 amino acids in HBV sequences submitted to GenBank database. Naturally occurring T109 and V124 and/or possible compensatory mutations in neighbored amino acids were introduced into HBV‐expressing plasmids. Viral replication competence and sensitivity to GLS4, a HAP compound, were evaluated using transient transfection and in vitro infection cell models. All tested mutations in these amino acids led to decreasing viral DNA replication at different levels. Specially, T109N and all V124 mutants caused severe deficiencies in viral plus‐strand DNA synthesis. T109I single mutation and all T109S/M/C/N mutations impaired HBeAg secretion. T109I showed modestly decreased sensitivities with IC50 3.3‐ to 6.8‐folds higher than wild‐type virus. In vitro infection assay showed T109N and all V124 mutants failed to synthesize cccDNA and following viral proteins. The other mutants, however, produced functional cccDNA pools as wild‐type virus did. Taken together, we profiled the competences of viral replication and sensitivities to capsid inhibitor of naturally existing mutations in T109 and V124. This will help to understand the possible antiviral resistance issues in future clinical applications of capsid inhibitors.

3-((R)-4-(((R)-6-(2-Bromo-4-fluorophenyl)-5-(ethoxycarbonyl)-2-(thiazol-2-yl)-3,6-dihydropyrimidin-4-yl)methyl)morpholin-2-yl)propanoic Acid (HEC72702), a Novel Hepatitis B Virus Capsid Inhibitor Based on Clinical Candidate GLS4

The inhibition of hepatitis B virus (HBV) capsid assembly is a novel strategy for the development of chronic hepatitis B (CHB) therapeutics. On the basis of the preclinical properties and clinical results of GLS4, we carried out further investigation to seek a better candidate compound with appropriate anti-HBV potency, reduced hERG activity, decreased CYP enzyme induction, and improved pharmacokinetic (PK) properties. To this end, we have successfully found that morpholine carboxyl analogues with comparable anti-HBV activities to that of GLS4 showed decreased hERG activities, but they displayed strong CYP3A4 induction in a concentration-dependent manner, except for morpholine propionic acid analogues. After several rounds of modification, compound 58 (HEC72702), which had an (R)-morpholine-2-propionic acid at the C6 position of its dihydropyrimidine core ring, was found to display no induction of the CYP1A2, CYP3A4, or CYP2B6 enzyme at the high concentration of 10 μM. In particular, it demonstrated a good systemic exposure and high oral bioavailability and achieved a viral-load reduction greater than 2 log in a hydrodynamic-injected (HDI) HBV mouse model and has now been selected for further development.