Dengue virus type 1 clade replacement in recurring homotypic outbreaks

Abstract

Background: Recurring dengue outbreaks occur in cyclical pattern in most endemic countries. The recurrences of dengue virus (DENV) infection predispose the population to increased risk of contracting the severe forms of dengue. Understanding the DENV evolutionary mechanism underlying the recurring dengue outbreaks has important implications for epidemic prediction and disease control.

Results: We used a set of viral envelope (E) gene to reconstruct the phylogeny of DENV-1 isolated between the periods of 1987-2011 in Malaysia. Phylogenetic analysis of DENV-1 E gene revealed that genotype I virus clade replacements were associated with the cyclical pattern of major DENV-1 outbreaks in Malaysia. A total of 9 non-conservative amino acid substitutions in the DENV-1 E gene consensus were identified; 4 in domain I, 3 in domain II and 2 in domain III. Selection pressure analyses did not reveal any positively selected codon site within the full length E gene sequences (1485 nt, 495 codons). A total of 183 (mean dN/dS = 0.0413) negatively selected sites were found within the Malaysian isolates; neither positive nor negative selection was noted for the remaining 312 codons. All the viruses were cross-neutralized by the respective patient sera suggesting no strong support for immunological advantage of any of the amino acid substitutions.

Conclusion: DENV-1 clade replacement is associated with recurrences of major DENV-1 outbreaks in Malaysia. Our findings are consistent with those of other studies that the DENV-1 clade replacement is a stochastic event independent of positive selection.

Figure 1

DENV-1 isolated in Malaysia for the period 1967–2010. The number of DENV-1 isolated for the period 1975–1980 is not available.

Figure 2

Distribution of DENV-1 genotypes in recurring DENV-1 outbreaks in Malaysia. The numbers below the pie charts indicate the total numbers of DENV-1 used from each outbreak. Total numbers of DENV-1 used for 1987, 1997 and 2004 outbreaks included viruses for the period 1987–1989, 1997–1999 and 2004–2006, respectively.

Figure 3

Maximum clade credibility tree of complete envelope genes of DENV-1. Horizontal branches are drawn to a scale of estimated year of divergence. Coalescent times with 95% highest posterior density values (ranges in parentheses) and posterior probability values (all 1.0) of key nodes are shown. DENV-1 outbreaks are indicated at the end of branches according to the outbreak-causing clades. The consensus E amino acid sequences of Malaysian DENV-1 isolates used were indicated as Ia, Ib, Ic, IIa, IIb, IIIa and IIIb. The DENV-1 isolates from Malaysia are highlighted in grey. Viruses marked with * are the rare isolates. A total of 39 new Malaysian DENV-1 E gene sequences were used [EMBL:FN825674, EMBL:FR666920-FR666928].

Figure 4

Observed amino acid substitutions in Malaysian DENV-1 E gene consensuses. The non-conservative amino acid substitutions which resulted in the polarity changes are highlighted in grey. TM = Transmembrane.