Antiviral compound screening

Based on assessment of this data, a comprehensive in silico screening of three small compound libraries from the National Cancer Institute against the BOG binding site in the pocket was performed (Zhou & Post, manuscript in preparation). A total of 200,000 ligands were screened by docking them into the BOG binding site. After a rigorous hierarchical and iterative process the lowest-energy conform-ers were selected. Ultimately, based on visual inspection and consideration of drug-like properties, 23 compounds were selected for biological screening against virus infectivity.

To establish a reliable and robust screen, a luciferase reporter-based assay was developed for investigating the effects of the compounds on virus growth. The YFV genome was modified by inserting a fire-fly luciferase gene at the 3' end of the genome (in the full-length YFV cDNA plasmid pACNR-FLYF). The expression of the luciferase gene was controlled using an internal ribosomal entry site (IRES) obtained from the encephalomyocarditis virus (EMCV) genome (Fig. 4).

Back Chamfer Tool

FIG. 4. Schematic representations of the YFV, YF-Luc and YF replicon genomes. Straight lines depict non-coding regions (NTR); boxes depict coding sequences—white (structural proteins), grey (non-structural proteins) and black (signal sequences). (A) Schematic of the YFV genome; genes coding for the structural and non-structural proteins are shown. A fire-fly luciferase gene was inserted at the 3'end of the coding sequence and expressed using an Internal Ribosome Entry Site (IRES) to obtain the genome for the YF-Luc virus. (B) The replicon (YF-RP) was constructed by deletion of the structural protein coding sequence (shown enlarged). The cyclization sequence was retained to allow genome replication. A Renilla luciferase gene was inserted before the signal sequence for NS1. The sequence coding for the autoproteolytic FMDV-2A site was fused to the 3' end of the luciferase gene and thus, the Renilla luciferase could be cleaved off from the growing polyprotein during genome translation.

FIG. 4. Schematic representations of the YFV, YF-Luc and YF replicon genomes. Straight lines depict non-coding regions (NTR); boxes depict coding sequences—white (structural proteins), grey (non-structural proteins) and black (signal sequences). (A) Schematic of the YFV genome; genes coding for the structural and non-structural proteins are shown. A fire-fly luciferase gene was inserted at the 3'end of the coding sequence and expressed using an Internal Ribosome Entry Site (IRES) to obtain the genome for the YF-Luc virus. (B) The replicon (YF-RP) was constructed by deletion of the structural protein coding sequence (shown enlarged). The cyclization sequence was retained to allow genome replication. A Renilla luciferase gene was inserted before the signal sequence for NS1. The sequence coding for the autoproteolytic FMDV-2A site was fused to the 3' end of the luciferase gene and thus, the Renilla luciferase could be cleaved off from the growing polyprotein during genome translation.

The virus produced after transfection of in vitro transcribed RNA from this cDNA containing the luciferase gene was termed YF-Luc. Growth of YF-Luc could be determined as a measure of the luciferase activity in infected cells. The inhibitory activity of the compounds could be ascertained by the reduction in luciferase activity of YF-Luc infected cells in the presence of active compound. A YFV replicon (YF-RP) was also developed by deleting most of the coding region for the structural proteins from the YFV cDNA and retaining the genes coding for the replicase proteins. After in vitro transcription and transfection of RNA, this replicon could independently replicate in cells since it possessed all the signals and could produce the replicase proteins required for replication. However, it could not produce progeny viruses since it lacked genes coding for the structural proteins required for formation of virus particles. A Renilla luciferase gene was inserted at the 5' end of the replicon to yield YF-R.Luc2A-RP, and replication of this replicon could now be monitored by measuring luciferase activity in cells transfected with YF-R.Luc2A-RP (Figure 4). Thus, we could now test the activity of compounds specifically against YFV genome replication and not virus production using YF-R.Luc2A-RP.

Initially, standard cytotoxicity tests were performed using an XTT-based cytotoxicity assay kit (BioVision Inc.) and non-cytotoxic concentrations of the compounds were used for inhibition assays. The assay for screening compounds for activity against YFV growth consisted of infecting baby hamster kidney (BHK) cells with YF-Luc virus at a low multiplicity of infection (MOI = 0.1) followed by addition of the compounds at non-cytotoxic concentrations. A low MOI was employed so as to allow for re-infection of virus and thus gauge the effect of compounds on virus spread. At 36 hours post-infection, cell extracts were taken and luciferase assays performed. From the 23 compounds tested, five showed inhibitory activity against YF-Luc. These compounds that exhibited inhibitory activity against YF-Luc virus were tested for activity against YF-R.Luc2A-RP to determine whether the compounds were affecting genomic RNA replication as opposed to virus entry, assembly or exit. This assay required the transfection of BHK cells with in vitro transcribed YF-R.Luc2A-RP RNA and treatment of trans-fected cells with the compounds at concentrations that showed inhibitory activity against YF-Luc virus. At 36 hours post-infection, cell extracts were taken and luciferase assays performed. Out of the five compounds tested, four showed inhibition of YF-R.Luc2A-RP, indicating that the compounds are inhibiting steps during viral genome replication. Interestingly, three out of these four compounds were found to be active against YF-Luc virus at concentrations lower than those required for inhibition of YF-R.Luc2A-RP replication, suggesting that these compounds are affecting some steps other than genome replication.

Assays to determine whether the compounds affect steps during entry of the virus into cells, or during assembly and exit from the infected cells are in progress.

Based on our hypothesis, one would predict that the compounds are mimicking BOG and bind in the hydrophobic pocket on the E protein. This binding might prevent conformational changes that are required of the E protein during fusion. In order to substantiate the idea, cell-cell fusion assays of infected cells in presence of the compounds at low pH will have to be performed. The assay involves treating infected cells with low pH buffer in the presence of the compounds and then assaying for syncytia formation and comparing the efficiency to syncytia formation in untreated cells. However, flaviviruses bud from internal membranes and E protein is not found in substantial amounts on the plasma membrane of infected cells. A more direct fusion assay will be to purify pyrene labelled virus and test fusion efficiency in presence of the compounds in an in vitro fusion assay with liposomes. NMR and crystallographic methods can also be pursued to provide direct evidence of compound binding to the E protein. Similar assays will be employed to check whether the compounds have inhibitory activity against other flaviviruses such as DENV and WNV. Furthermore, using similar in silico molecule docking techniques using other target sites on the E protein, additional compounds have been identified and will be screened for inhibitors using the assays outlined above.

In summary, we have undertaken a structure-based approach for the rational design of small molecule inhibitors of flaviviruses, specifically directed against the outer envelope glycoprotein. Preliminary assays screening compounds for activity against YFV infection have been performed and the results are encouraging. Further assays to determine the exact site of action of the compounds have been developed and are in progress. Thus, using a powerful combination of structure-based drug design and informative biochemical assays, the goal is to identify novel compounds that selectively inhibit the target site and possess a desirable pharmacological profile that can be developed into potent inhibitors of flavivirus infection.

A ck nowledgements

This work was supported by a Public Health Service Program Project Grant (AI55672) from the National Institute of Allergy and Infectious Disease. The work was also funded as part of the NIH NIAID Great Lakes RCE (AI57153).

Was this article helpful?

0 0
Swine Influenza

Swine Influenza

SWINE INFLUENZA frightening you? CONCERNED about the health implications? Coughs and Sneezes Spread Diseases! Stop The Swine Flu from Spreading. Follow the advice to keep your family and friends safe from this virus and not become another victim. These simple cost free guidelines will help you to protect yourself from the swine flu.

Get My Free Ebook


Post a comment