Endurance, refuge, and reemergence of dengue virus type 2, Puerto Rico, 1986-2007

Abstract

To study the evolution of dengue virus (DENV) serotype 2 in Puerto Rico, we examined the genetic composition and diversity of 160 DENV-2 genomes obtained through 22 consecutive years of sampling. A clade replacement took place in 1994-1997 during a period of high incidence of autochthonous DENV-2 and frequent, short-lived reintroductions of foreign DENV-2. This unique clade replacement was complete just before DENV-3 emerged. By temporally and geographically defining DENV-2 lineages, we describe a refuge of this virus through 4 years of low genome diversity. Our analyses may explain the long-term endurance of DENV-2 despite great epidemiologic changes in disease incidence and serotype distribution.

Figure 1

Historic overview of dengue, Puerto Rico, 1986–2007. A) Number of suspected, clinically defined cases of dengue fever/dengue hemorrhagic fever by year reported to the Centers for Disease Control and Prevention’s Dengue Branch. B) Percentage of identifications of each serotype relative to the total of positive serotype identifications by using tissue culture isolation or reverse transcription–PCR per year. Numbers in parenthesis indicate numbers of dengue virus (DENV) serotype 2 identifications each year. Black, DENV-1; blue, DENV-2; white, DENV-3; red, DENV-4. C) Number of partially sequenced (E gene) autochthonous Puerto Rican isolates reported by previous studies (14,20) and whole genome sequences obtained in the present study by year of their corresponding case presentation. D) Bayesian coalescent inference of population dynamics and genetic diversity by using the Bayesian Skyline plot. Markov Chain Monte Carlo from BEAST version 1.4.7 (www.biomedcentral.com/1741-7007/8/114). Sampling procedures were used to estimate posterior distribution of DENV-2 genetic diversity in an effective population through the study period on the basis of full genome sequence data. x axis, time in years through the study period; y axis, product of the effective population size (relative genetic diversity) and generation length in years; black line, median estimate; blue shadow, 95% highest probability density.

Figure 2

Evolution of dengue virus (DENV) serotype 2, Puerto Rico. Maximum likelihood phylogeny of the 140 Puerto Rico and 20 international isolates of DENV-2 (see number of isolates by year below). Names of clades (I, II, and III) and subclades (IA, IB) are shown at the base of their respective branches on the phylogeny tree. Clade II (dark blue) circulated during 1986–1996 and clade I (light blue) during 1994–2007. Subclade IA and clade III represent foreign, transient reintroductions throughout the 22-year study period. Black dots indicate 18 isolates from Puerto Rico with phylogenetic associations closer to foreign isolates than to other Puerto Rico viruses. Two-letter geographic codes indicate origin (PR, Puerto Rico; JM, Jamaica; SJ, Saint John; SH, Saint Thomas; VE, Venezuela; Cl, Colombia; NC, Nicaragua; MX, Mexico; CU, Cuba; SI, Saint Kits; DR, Dominican Republic; SX, Saint Croix; BR, Brazil). PR code is accompanied by a 2-digit number that indicates geographic location (municipality) on the island. The 2-digit numbers between the geographic codes and the GenBank accession numbers indicate the year of the corresponding case of each isolate. Bootstrap values are shown for all clades (I, II, and II), subclades (IA, IB), and most immediate lineages. Twelve amino acid changes associated with IB/II differences are shown on the Table; 6 other selected changes across subclade IB are shown on the left at the base of the branches exhibiting the respective changes. Amino acid numbers refer to the position in each DENV protein. Relevant epidemiologic events are highlighted to the right with pie charts showing the relative levels of each serotype isolated for that period Black, DENV-1; blue, DENV-2; white, DENV-3; red, DENV-4. Scale bar indicates nucleotide substitutions per site.

Figure 3

Geographic clustering of Puerto Rico dengue virus lineages. A) Maximum-likelihood phylogeny of clade II. All isolates indicate year of case presentation and GenBank accession numbers. B) Maximum-likelihood phylogeny of subclade IB shows isolates by year and GenBank accession numbers. Six regions had >5 isolates (San Juan, Caguas, Ponce, Mayaguez, Aguadilla, and Arecibo). C) Eight regions of Puerto Rico with colors corresponding to isolates in panel A and year for the 3 regions with more isolates of that clade: San Juan, Mayaguez, and Ponce. D) Eight regions of Puerto Rico, showing colors and years corresponding to isolates in panel B. Correlation between phylogeny and geographic location of isolation for the isolates in this study was estimated by using Bayesian Tip-association Significance testing. Association index 6.51 (95% confidence interval 6.03–7.22); parsimony score statistic 52.27 (95% confidence interval 51–54). Monophyletic clade size Bayesian Tip-association Significance estimates are shown for 6 regions of Puerto Rico with >5 isolates represented in at least 1 subclade and statistically representative geographic associations (p<0.05).

Figure 4

Epidemiology of dengue virus (DENV) serotype 2 in Puerto Rico, 1997–2006. A) Municipalities with persistent DENV-2 transmission (Caguas, Juncos, Las Piedras, Carolina) versus those with discontinuous transmission (Morovis, Toa Alta, Toa Baja, Cataño, Guaynabo, Cidra, San Lorenzo, Canóvanas, Humacao, Naguabo, Ceiba, Fajardo), 1998–2002. Inset shows satellite view; red dot indicates national capital (San Juan), and yellow box indicates region where DENV-2 took refuge during 2000–2002. B–D) Satellite view depicts virus transmission corridors. White pins point to specific geographic locations where DENV-2 isolates were collected during the specified time period. Yellow lines connect isolates by their phylogenetic affiliations suggesting migration of virus. B) DENV-2 traveled to the San Juan region from the west during 1997–1999; C) DENV-2 transmission retracted to the eastern, refuge region with restricted dispersion patterns during 2000–2002; D) DENV-2 reemerged focused on the San Juan region and later dispersed throughout the island during 2003–2006.

Figure 5

Incidence of dengue virus (DENV) serotypes 2 and 3 in Puerto Rico, 1996–2005. Solid blue line, incidence of DENV-2 within the refuge region; dashed blue line, incidence of DENV-2 in the rest of the island outside the refuge reason; solid black line, incidence of DENV-3 within the DENV-2 refuge region; dashed black line) incidence of DENV-3 in the rest of the island outside the refuge region. Incidence was calculated as number of confirmed, positive cases of each serotype per thousand residents.

Figure A1

Dengue virus (DENV) serotype 2 case distribution, Puerto Rico, 1987–2006. Maps of Puerto Rico are shown as examples of distribution of total of DENV-2 cases reported on the island during 4 different years representing clade II expansion (1987) (A), clades II/I mixing (1995) (B), low evolution of subclade IB (2001) (C), and resurgence of subclade IB (D). All maps represent the 78 municipalities of Puerto Rico. White, 0–1 DENV-2 identifications; gray, 2–5 DENV-2 identifications; black, >6 DENV-2 identifications. Maximum number of identifications in any 1 municipality was 17.