The Complete Genome Phylogeny of Geographically Distinct Dengue Virus Serotype 2 Isolates (1944-2013) Supports Further Groupings within the Cosmopolitan Genotype

Abstract

Dengue virus serotype 2 (DENV-2) isolates have been implicated in deadly outbreaks of dengue fever (DF) and dengue hemorrhagic fever (DHF) in several regions of the world. Phylogenetic analysis of DENV-2 isolates collected from particular countries has been performed using partial or individual genes but only a few studies have examined complete whole-genome sequences collected worldwide. Herein, 50 complete genome sequences of DENV-2 isolates, reported over the past 70 years from 19 different countries, were downloaded from GenBank. Phylogenetic analysis was conducted and evolutionary distances of the 50 DENV-2 isolates were determined using maximum likelihood (ML) trees or Bayesian phylogenetic analysis created from complete genome nucleotide (nt) and amino acid (aa) sequences or individual gene sequences. The results showed that all DENV-2 isolates fell into seven main groups containing five previously defined genotypes. A Cosmopolitan genotype showed further division into three groups (C-I, C-II, and C-III) with the C-I group containing two subgroups (C-IA and C-IB). Comparison of the aa sequences showed specific mutations among the various groups of DENV-2 isolates. A maximum number of aa mutations was observed in the NS5 gene, followed by the NS2A, NS3 and NS1 genes, while the smallest number of aa substitutions was recorded in the capsid gene, followed by the PrM/M, NS4A, and NS4B genes. Maximum evolutionary distances were found in the NS2A gene, followed by the NS4A and NS4B genes. Based on these results, we propose that genotyping of DENV-2 isolates in future studies should be performed on entire genome sequences in order to gain a complete understanding of the evolution of various isolates reported from different geographical locations around the world.

Fig 1. Phylogenetic analysis of DENV-2 complete genome nucleotide sequences.

(A) Maximum-likelihood trees were constructed using MEGA V5.05 software with bootstrap support of 1000 replicates. All nucleotide sequences were downloaded from the GenBank database for analysis (Table 1). The phylogenetic tree was constructed using the General Time Reversible (GTR) model. (B) Bayesian Maximum Clade Credibility tree of the 50 DENV-2 isolates. Seven groups including five major genotypes were identified. In the Cosmopolitan genotype, there were three groups (C-I, C-II and C-III) while C-I was sub-grouped into C-IA and C-IB. The DENV-1 US/Hawaii isolate was used as an out-group.

Fig 2. Phylogenetic maximum-likelihood trees of DENV-2 complete genome amino acid sequences.

Trees were constructed using MEGA V5.05 software with bootstrap support of 1000 replicates. All amino acid sequences were downloaded from the GenBank database for analysis, and the respective DENV-2 isolates with their accession numbers are listed in Table 1.

Fig 3. Phylogenetic maximum-likelihood trees of DENV-2 ORF nucleotide sequences.

Trees were constructed using MEGAV5.05 software with bootstrap support of 1000 replicates. All sequences of the ORF were manually separated from the whole-genome sequences that were downloaded from the GenBank database for analysis. The phylogenetic tree was constructed using the GTR model.

Fig 4. Phylogenetic maximum-likelihood trees of DENV-2 individual gene nucleotide sequences.

All sequences of the individual genes were manually separated from the complete genome sequences that were downloaded from the GenBank database for analysis. The phylogenetic trees were constructed using the GTR model. (A) C gene; (B) PrM/M gene; (C) E gene; (D) NS1 gene; (E) NS2A gene; (F) NS2B gene; (G) NS3; (H) NS4A; (I) NS4B; (J) NS5; (K) 3´ UTR; and (L) 5´UTR.