Developing new inhibitors against flavivirus methyltransferase using a 'Fragment-Based Drug Discovery' strategy – FragVir

The discovery of new flaviviral inhibitors, in particular against Dengue virus, is based on a «Fragment-based drug discovery« approach which allows a rapid identification of small inhibitors (or fragments) which bind to the target protein.
Indeed, an experimental fragment screening approach by biophysical methods is used to identify fragment hits together with their binding site and their binding mode.
Knowledge of exactly how the fragments bind to the protein target allows the hits to be extended (or linked with other fragments) to create higher affinity compounds by successive rounds of improvements using medicinal chemistry and structure-based drug design. Thus, structural data obtained coupled with molecular modeling, chemical synthesis and evaluation of antiviral activities, offer the prospect of rapid and efficient optimization of these fragments potent inhibitors.

A multistep fragments screening against methyltransferase of dengue virus (DENV MTase) allowed the identification of seven fragment hits. Based on these data, three series of molecules resulting from the optimization of the fragment hits identified were designed by molecular modelling and twenty analogues were synthesized for each series.
The optimization process by molecular modelling was performed on the basis of a virtual ‘docking’ of targeted libraries. A rapid in silico assessment of these virtual libraries helped classify the «scoring function« for each virtual analogue from the lowest to the highest affinity. For each series, analogues showing higher affinity were synthesized. The inhibitory activity of each analogue was evaluated on the N-7 and 2'-O-MTase activities. The analogue/Mtase DENV affinity interaction was also studied by Thermal-shift assay. The most active analogues were then soaked with pre-formed crystals of DENV MTase to obtain X-ray structures of crystallographic complex and validate their binding mode. Regarding series 1, the synthesized analogues have shown no inhibition of N-7 and 2'-O-MTase activities. No crystallographic complex with MTase DENV was obtained by either soaking or co-crystallization. Therefore, the optimization process of this series was abandoned. Regarding series 2 and 3, the synthesized analogues showed a gain in the inhibition of N-7 and 2'-O-MTase activities. We have also obtained the crystallographic structures of some of these analogues in complex with the DENV MTase.

This optimization process will be repeated and refined around analogues of series 2 and 3 until the discovery of a drug candidate.

B. Coutard, E. Decroly, C. Li, A. Sharff, J. Lescar, G. Bricogne and K. Barral. Assessment of Dengue virus helicase and methyltransferase as targets for fragment-based drug discovery. Antiviral Research 2014, 106, 61-70

Flavivirus infections, such as those caused by mosquito-borne Dengue, West Nile, Kunjin, Japanese encephalitis and Yellow fever viruses, can give rise to life-threatening diseases of epidemic proportions that have a devastating economic impact. The dengue virus (DENV) alone causes 50 to 100 million human infections annually, leading to about 30,000 deaths; its transmission has increased dramatically and has become a major international public health concern. While effective vaccines are available against some flaviviruses, the pursuit of a DENV vaccine has proved elusive (recent clinical trials of a tetravalent DENV vaccine has shown only a low level of global protection). As no antiviral drugs are yet available to treat or prevent DENV infections, world health remains highly vulnerable to Dengue fever epidemics.
In such a context, the development of potent anti-flaviviral drugs represents an urgent as well as economically effective complement to any vaccination campaign, and may also be of prophylactic relevance in case of an epidemic. It is the aim of the present project to deploy an innovative drug discovery approach towards the development of potent antiviral compounds to treat flavivirus (and more specifically DENV) infections.
A natural focus for such a development is the flaviviral replication mechanism, many steps of which have been elucidated, leading to the identification of several targets whose inhibition can block replication. In particular the flaviviral protein NS5 has been shown to carry methyltransferase (MTase) activities involved in the RNA cap formation that are essential for the replication of flaviviruses. The investigation of these MTases constitutes one of the most actively growing fields amongst viral RNA capping targets for the discovery and development of anti-flavivirals.
The Fragment-Based Drug Discovery (FBDD) approach is now well established as a powerful method for the rapid identification of starting “hit” compounds and their subsequent elaboration into quality lead compounds. In the light of strong preliminary results and tools available in the laboratory, we decided to apply the FBDD approach to the development of specific flaviviral methyltransferase inhibitors.
This project involves an integrated multidisciplinary approach with the close interaction of structural biology, virology, biochemistry, synthetic and computational chemistry, combined with a stepwise medicinal chemistry strategy for rapid hit-to-lead optimization. A multistep fragment-based screening of a library of 500 fragments by Thermal-shift assays and X-ray crystallography has allowed the identification of seven fragment hits that bind to four different DENV MTase sites, three of which are novel binding sites. These seven fragments also show inhibition effect on DENV MTase activity assays.
We plan to develop these seven already validated fragment hits into drug-lead candidates by an iterative process involving cycles of (1) library design and modeling-driven evaluation of virtual libraries for synthesis, (2) parallel medicinal chemistry synthesis, (3) in vitro screening of newly synthesized compounds on flavivirus MTases to evaluate their inhibition effect, (4) crystal determination/biophysical studies to explore hit-protein interactions, (5) evaluation of the most potent compounds in infected cell-based assay with various flaviviruses and (6) assessment of the lead compounds for ADMETox properties. At the end of the project, we expect that at least one series of fragments will have led to potent antiviral lead compound(s) with good drug-like properties and, therefore, amenable to animal model testing.
This strategy is innovative in the field of anti-flaviviral research. We expect to make the first proof of concept for the successful use of FBDD approaches on emerging and neglected viruses and, in particular, on proteins involved in the capping of viral genomes.