Suppression of Drug Resistance in Dengue Virus

Mateo, Roberto; Nagamine, Claude M.; Kirkegaard, Karla

To identify potential dominant drug targets for dengue virus, we sought published antiviral compounds that target oligomeric viral structures. Byrd et al. in 2013 (12) reported the inhibition of all four dengue virus serotypes by a compound termed ST-148, which was identified in a high-throughput screen (Fig. 1A) . Dengue virus core protein was determined to be the compound's target in direct binding experiments and by the mapping of a drugresistant mutation, S34L, to the coding sequences for core protein that were generated by bulk RT-PCR from RNA extracted from supernatants of singly or doubly infected BHK-21 cells. Wild-type virus had been engineered to contain a silent AflII restriction site in the core coding region. (G) Amplicons were cleaved with AflII and run in 1% agarose-TBE gels. The proportions of resistant and susceptible genomes were determined from the percentage of uncleaved (476-bp) and cleaved (344-bp) PCR products, respectively. Results shown are a representative example of coinfection with wild-type and NS5-A60T virus. All data shown are averages Ï® standard deviations of results from two to three biological replicates. Statistical analysis of the yield of drug-resistant virus in the coinfections, compared to single infection, was via an unpaired Student's t test. (12) (Fig. 1A and B) . Even for dominant drug targets, drugresistant viruses can be readily selected because the necessary mutations preexist in the inoculum. Then, when a stock is passaged in cultured cells at low multiplicities of infection (MOIs), these preexisting variants can infect cells and be selected in the absence of potentially dominant wild-type proteins. Our question is, what about the variants that newly arise?