Abstract:
C10 was identified in a FACS-FRET-based screen designed to identify small compounds that interfere
with HCV capsid protein (CP) interaction. Previous experiments demonstrated that C10 exhibited
antiviral activity against all tested members of the Flaviviridae family, exhibiting a favorable
therapeutic index (TI) of 400 to 600 and a specific crosslinking effect on the CP. Subsequent testing of
different C10 derivatives, revealed C45, a compound that exhibited enhanced antiviral efficacy in
reinfection, while maintaining comparable toxicity (TI=2800). C45 was found to also induce higher
molecular weight species of CP.
The objective of this project was to elucidate the mode of action of C10 and C45, regarding their ability
in the viral life cycle and to analyze the compound-CP-interaction.
To study the impact of C10 and C45 on different steps of the flavivirus replication cycle, cell-based
infection assays were employed and quantified by Western Blot (WB), qPCR, and immunofluorescence
microscopy. To analyzing the compound binding site, a method to purify compound-bound viral
particles from unbound particles was established, and compound-resistant DENV was generated and
sequenced to identify acquired mutations.
In this thesis it was observed that C10 and C45 can bind to assembled DENV and HCV particles reducing
their infectivity. Next, it was demonstrated that treatment with C10 or C45 had no impact on the viral
particle release and that the high antiviral activity of both compounds cannot be explained by
diminished genome packaging. Infection with C10-treated DENV particles resulted in capsid
accumulation without co-localized E-protein in cells, suggesting that C10 interferes with the uncoating
of the viral life cycle, thereby inhibiting viral replication and translation. The impact of C45 on the
uncoating and viral fusion has not yet been studied, however as both compounds are closely related it
is hypothesized that they act similarly. The acquired data supports a model in which C10 and C45 target
the flavivirus CP during viral assembly or in assembled viral particles, inducing crosslinking of CPs,
which leads to an inhibition of the CP disassociation during viral uncoating.
Upon selection of C10-resistant viruses, different mutations were mapped to all three structural
proteins and the non-structural proteins NS4B and NS5. Within the target protein CP, mutations of the
amino acids K9 and N93 were identified. Utilizing a published an in-silico prediction model of different
N-terminal orientations picked up during distinct phases of the viral life cycle, the hypothesis that C10-
binding to the residues K9 and N93 stabilizes the N-terminal part in a conformation acquired for
assembly was made. This hypothesis is supported by experimental evidence from multiple studies,
which demonstrate that the binding of C10 to the nucleocapsid hinders the disassembly process,
thereby preventing genome release. However, further validation is required to substantiate this
hypothesis, particularly through the identification of compound interaction sites.
