Purifying selection in the evolution of dengue virus type 1 in Malaysia

      Background: Dengue, a disease with zoonotic origin is endemic in Malaysia. Available data suggests that major homogenotypic and heterogenotypic dengue outbreaks occurred in cyclical patterns of approximately every 8 and 3 years, respectively. To date, there have been three major DENV-1 outbreaks in Malaysia; in 1987, 1997, and 2004. DENV-1 serially isolated since 1987 within a single locality provide an opportunity to investigate its temporal phylogenetic history in Malaysia.
      Methods: DENV-1 isolates used in this study were isolated from Klang Valley, Malaysia during the period spanning ∼20 years (1987-2005). Viral RNA was extracted and the envelope (E) gene was amplified and sequenced. The DENV-1 E gene dataset for phylogenetic analysis comprised of 32 Malaysian and 74 global sequences representing the 6 distinct DENV-1 subgenotypes. Phylogenetic tree was inferred by using the Bayesian Markov Chain Monte Carlo (MCMC) method implemented in BEAST. The selection pressure analysis was performed by using the SLAC, FEL and REL methods available in the Datamonkey web server (
      Results: Based on the Bayesian inference, the evolutionary rates of DENV-1 E gene were estimated to be ∼5.54 × 10-4 to 10.3 × 10-4 substitutions/site/year. These rates are relatively slower than those of non-vector-borne RNA viruses such as influenza viruses and HIV. Selection pressure analyses showed no positively selected codon site within the Malaysia isolates and within each of the subgenotypes, except for the REL analysis of subgenotype II viruses indicated evidence of positive selection at codon 88 and 297 of the E gene.Residue 88 was found to be located near the fusion loop in domain II, while residue 297 falls within the domain I region.
      Conclusion: In this study, the molecular clock and selection pressure analysis collectively showed that the DENV-1 evolution process is mainly constrained by purifying selection.This suggests the possible contributions of other ecological factors such as mosquito density and susceptible human population to the recurrence of dengue outbreaks.