Establishment of novel enterovirus mouse models to assess the efficacy of new antivirals and the potential development of drug resistance.

Within the Enterovirus genus (EV, family Picornaviridae), rhinoviruses (RVs) are grouped in three species: RV-A, RV-B and RV-C. For long, these viruses were considered non-harmful pathogens of the human upper respiratory tract. However, RV infections of the lower respiratory tract are increasingly reported. Moreover, these viruses have been associated with severe exacerbations of asthma and chronic obstructive pulmonary disease (COPD). With more than 160 identified RV genotypes and very limited cross-neutralization, the development of a vaccine is extremely challenging. Different steps in the replication cycle have been exploited as antiviral targets. The best-studied class of antivirals are the so-called capsid binders. These compounds prevent entry and uncoating by binding to the viral capsid. Some antivirals, including the capsid binder pleconaril, have been clinically evaluated for the treatment of RV infections, but so far, none have reached market approval due to safety or efficacy issues. There is hence an urgent need for potent and safe antivirals for the prophylaxis and treatment of RV infections, particularly for high-risk patients with underlying chronic conditions (chapter 1). RVs are extremely species-specific human pathogens which hampers the development of a small animal model to study the in vivo efficacy and the effect of long-term treatment on the emergence of drug-resistant variants. Therefore, we employed a surrogate virus, namely coxsackievirus B4 (CV-B4), to develop a lethal mouse model (chapter 2). Like RV, CV-B4 is a member of the Enterovirus genus, but in contrast to RV, this virus is able to replicate efficiently in mice. We established a dose-dependent systemic CV-B4 infection in adult severe combined immunodeficient (SCID) mice, which induced a severe pancreatitis with diffuse necrosis of the exocrine pancreas. Next, we assessed the in vivo efficacy of the capsid binder pleconaril and a highly potent RV/EV replication inhibitor (compound A) in our mouse model. Interestingly, prophylactic treatment with compound A protected the SCID mice against CV-B4 infection for up to 60 days in contrast to pleconaril that lacked any antiviral effect in vivo. When treatment with compound A was initiated one or two days after infection, a significantly delayed disease progression was observed. Finally, a reduced daily dosage for a prolonged time was used to study the effect of long-term treatment on the genetic variation within the virus population, i.e. to explore whether resistance would develop. Altogether, we established an enterovirus infection mouse model that allows to characterize antiviral compounds during preclinical studies and to explore the emergence of potential drug-resistant variants, thereby aiding to the rationale for further clinical development. In chapter 3, we aimed to develop yet another relevant mouse model that would allow to assess the antiviral activity of RV inhibitors in the upper respiratory tract. We showed that following intranasal CV-B4 inoculation of adult SCID mice, virus replication was detected in the glandular epithelial cells of the nasal mucosa up to five days after infection. Next, we explored the effect of systemic (oral) or local (intranasal) administration and prophylactic or therapeutic treatment. Interestingly, both oral and intranasal administration of compound A fully protected the nasal mucosa from CV-B4 infection. Moreover, therapeutic treatment with compound A was able to cure the mice from an already established upper respiratory tract infection. In contrast, neither systemic nor local administration of pleconaril could protect the mice against this upper respiratory tract infection. Finally, we provided first evidence that antiviral treatment can also minimize the CV-B4-induced inflammatory cascade. In summary, this physiologically relevant enterovirus mouse model will constitute an invaluable tool to select the most optimal candidate for further clinical development in a more rational way. In close collaboration with medicinal chemists, a class of RV capsid binders was improved to increase the potency and broad-spectrum activity. In chapter 4, we report our efforts whereby we started from compound LPCRL_2034 (EC50 RV-B14 = 3.4 ± 1.0 µM, CC50 HeLa > 263 µM) and carried out pharmacomodulations with five- and six-membered heterocycles on the pore-side of the VP1 pocket. In addition, heterocyclic pharmacomodulations towards the toe-end side of the VP1 pocket were explored (chapter 5). Finally, the most active compounds 10e and 10h reached an EC50 value of 0.2 µM against RV-B14 (CC50 HeLa > 600 µM), which is a 17-fold improvement compared to the LPCRL_2034. By enantioseparation the activity of both compounds significantly increased (EC50 RV-B14 = 0.035 ± 0.006 µM and EC50 RV-B14 = 0.06 ± 0.03 µM resp.). In contrast to the initial hit, compounds 10e and 10h showed antiviral activity against a panel of both RV-A and RV-B types. The results of a time-of-addition study revealed that compounds 10 have a similar inhibition profile as the capsid binder pleconaril and thereby show that the pharmacomodulations did not alter the mechanism of action. Importantly, a resistance-development study provided first evidence that RV-B14 is likely less prone to develop resistance against compounds 10 than to pleconaril. Overall, we identified a new series of RV inhibitors with a promising pharmacological profile.status: publishe