Mutagenesis of the dengue virus envelope glycoprotein gene can significantly alter virus infectivity phenotypes in cultured cells and live mosquitoes
Date
2011
Authors
Erb, Steven Michael, author
Blair, Carol D., advisor
Roehrig, John T., committee member
Olson, Kenneth E., committee member
Chen, Chaoping, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
The dengue virus (DENV) envelope (E) glycoprotein is the primary determinant for initiation of host cell infection. To date, studies investigating the contribution of DENV genetics to mosquito infection are limited. A infectious clone cDNA of DENV type 2 strain 16681 (30P-NBX) provided the ability to introduce site-specific amino acid (AA) mutations into the E protein. The results of the studies herein analyze the effects that AA mutations in the E protein have on infectivity of cultured cells and live mosquitoes. The ability of 30P-NBX to infect Aedes aegypti RexD strain mosquitoes after oral infectious blood-meal was investigated and showed that both 30P-NBX and the parent virus 16681 have low, but equivalent midgut infection rates (MIRs). Mosquito midgut infection with 30P-NBX is not affected by the virus titer in the blood-meal as long as titers are above 6 log 10 pfu/ml or 7 log 10 TCID 50 /ml. Additionally, multiple experimental repetitions with at least 20 mosquitoes per infectious blood-feed were required to obtain an accurate average MIR for 30P-NBX. Serial passage of 30P-NBX in RexD mosquito midguts identified a single AA mutation at position 122 in domain II of the E protein from lysine to glutamic acid that correlated with increased MIRs. Introduction of this AA mutation into the infectious clone (mutant virus K122E) reproduced the results from the serial passage experiment. Compared to 30P-NBX, K122E was not only shown to infect a higher proportion of mosquitoes as early as day 2 post blood-feed, but also to produce a disseminated infection in a higher proportion of mosquitoes by day 6 post blood-feed. Also, K122E consistently produced a midgut infection that spread throughout the entire tissue while 30P-NBX stayed restricted by comparison. Virus attachment to midgut cells was compared and showed that 30P-NBX and K122E could attach with equal efficiencies via our midgut-virus attachment assay. Additionally, incorporation of a single AA mutation into the infectious clone at E protein AA 120 from arginine to threonine significantly enhanced mosquito midgut infection compared to 30P-NBX. This is the first time that mosquito infection determinants have been identified in the DENV E protein. Amino acid mutations were engineered into the E protein on the lateral ridge of domain III, the fusion peptide at the distal end of domain II, and the molecular hinge region between domains I and II. Mutant virus phenotypes were analyzed in cell culture and live mosquitoes. In contrast to previous suggestions, domain III mutant virus phenotypes showed that the FG loop structure (previously suggested as a mosquito-specific infection determinant) and not the specific AA sequence is important for infection of mammalian cells and live mosquitoes, while the structure and sequence of the FG loop is dispensable for infection of cultured C6/36 cells. Additionally, mutations that remove positively charged residues from the A strand in DIII significantly attenuate infection of mosquitoes after oral infectious blood-meal and completely abrogate infection in mammalian cells. The results of this study suggest that there may be multiple structures in the E protein that are contributing to virus-receptor interactions. Viruses with mutations in the fusion peptide and hinge region of the E protein were intrathoracically (IT) inoculated into mosquitoes and showed variable infectivity phenotypes. All of the mutants except for one virus from both the fusion peptide and hinge region viruses attenuated infection of mosquito tissues outside the midgut. Importantly, considering that almost all of these viruses were able to replicate as efficiently as wild type in C6/36 cells, the IT inoculation results provide evidence that C6/36 cells are not a complete surrogate for DENV replication in mosquitoes.
Description
Rights Access
Subject
envelope protein
Aedes aegypti
dengue virus
serial passage
C6/36 cells