Tracking the polyclonal neutralizing antibody response to a dengue virus serotype 1 type-specific epitope across two populations in Asia and the Americas

Abstract

The four dengue virus serotypes (DENV1-4) cause major public health problems worldwide. Highly neutralizing type-specific human monoclonal antibodies (hmAbs) target conformation-dependent epitopes on the DENV envelope protein, including 1F4, a DENV1 type-specific hmAb. Using a recombinant DENV2 virus displaying the DENV1 1F4 epitope (rDENV2/1), we measured the proportion and kinetics of DENV1 neutralizing antibodies targeting the 1F4 epitope in individuals living in Asia and the Americas where different DENV1 genotypes were circulating. Samples from 20 individuals were analyzed 3 and 18 months post-primary DENV1 infection, alongside samples from 4 individuals collected annually for four years post-primary DENV1 infection, from two studies in Nicaragua. We also analyzed convalescent post-primary DENV1 plasma samples from Sri Lankan individuals. We found that neutralizing antibodies recognizing the 1F4 epitope vary in prevalence across both populations and were detected from 20 days to four years post-infection. Additionally, both populations displayed substantial variability, with a range of high to low proportions of DENV1 type-specific neutralizing antibodies recognizing the 1F4 epitope seen across individuals. Thus, the 1F4 epitope is a major but not exclusive target of type-specific neutralizing antibodies post-primary infection with different DENV1 genotypes in Asia and Latin America, and additional epitopes likely contribute to type-specific neutralization of DENV1.

Figure 1

Amino acid residues comprising the 1F4 epitope in the EDI and EDI/EDII hinge region of a DENV1 E protein monomer were transplanted into a DENV2 backbone and analyzed by neutralization assay. (A) Diagram of the DENV1 E dimer (domains I, II and III in red, yellow and blue, respectively) with yellow and red spheres representing the 1F4 epitope footprint. (B) The amino acid residues within the 1F4 epitope, represented in red and yellow spheres, were transplanted into a DENV2 backbone, creating the rDENV2/1 virus. (C,D) Representative sigmoidal dose-response curves used to estimate the 50% neutralization titer (NT₅₀) to the rDENV2/1 and parental DENV1 and DENV2 viruses in plasma samples from patient 1449 at 3 (C) and 18 months (D).

Figure 2

Inclusion of the 1F4 epitope amino acid residues in a DENV2 backbone results in gain of neutralization against rDENV2/1 chimeric virus by polyclonal sera post-primary DENV1 infection. (A,B) Primary DENV1 plasma from 20 individuals in the Nicaraguan hospital-based study strongly neutralize the parental DENV1 virus and gain neutralization capacity against a DENV2 backbone containing the 1F4 amino acid residues (rDENV2/1) at 3 and 18 months post-infection. The NT₅₀ values to the rDENV2/1 virus and the DENV1 and DENV2 parental viruses were compared by one-way ANOVA (n = 20). (C–E) Paired analysis of the longitudinal samples at 3 and 18 months shows a significant decay of DENV1 NT₅₀ values, whereas DENV2 and rDENV2/1 titers are maintained constant over time. The t-test was used to compare the neutralizing antibody titers to each serotype at 3 and 18 months post-infection. Data are representative of two independent experiments, and samples were processed in duplicate for each plasma sample. **p < 0.01; ***p < 0.001; ****p < 0.0001.

Figure 3

The proportion of the DENV1 type-specific neutralizing antibody response targeted to the 1F4 epitope is highly variable across the Nicaraguan population in the hospital-based study at 3 and 18 months post-infection. (A) Analysis of the proportion of the DENV1 type-specific response attributable to the 1F4 epitope at 3 and 18 months reveals substantial variability, with individuals displaying a range from high to low proportions of the DENV1 response directed to the 1F4 epitope. (B) Paired analysis where each individual was assigned a color and a symbol to enable visualization of the trajectory of the DENV1 type-specific response to the 1F4 epitope at 3 and 18 months post-infection. (C) Distribution of individuals who gained recognition of the 1F4 epitope between 3 and 18 months (beige), lost recognition (brown) or remained constant (green). (D) The antigenic cartography map positions viruses (DENV1, DENV2 and rDENV2/1 in teal, purple and yellow, respectively) and plasma (open teal squares), with the distance between each virus and plasma derived from its respective neutralizing antibody titer. Each grid square corresponds to a 2-fold dilution in the NT₅₀. From 3 to 18 months post-primary DENV1 infection, the DENV1 and rDENV2/1 titers converge, as indicated by the rDENV2/1 arrow pointing towards DENV1.

Figure 4

The proportion of the DENV1 neutralizing antibody response directed to the 1F4 epitope is affected by the levels of homotypic and heterotypic titers. (A–C) To better understand the different patterns of the proportion observed in the individual trajectory analysis, we grouped the individuals who displayed increased (A), constant (B) or decreased (C) proportion over time (right Y-axis) and concomitantly analyzed their NT₅₀ values (left Y-axis) to the parental DENV1 (teal line) and DENV2 viruses (purple line) and the rDENV2/1 chimeric virus (yellow line). In group A, the increase in antibodies targeting the 1F4 epitope is associated with a decay in the DENV1 NT₅₀ values and concomitant maintenance of the rDENV2/1 titers. In group B, the constant proportion between 3 and 18 months is accompanied by changes in the magnitude of neutralizing antibody titers to the parental viruses, while in Group C, the loss of recognition of the 1F4 epitope is associated with an increase in the NT₅₀ values to DENV1 and DENV2 viruses.

Figure 5

Primary DENV1 plasma samples collected up to four years post-infection track to varying degrees with the 1F4 epitope in the Pediatric Dengue Cohort Study in Nicaragua. (A–C) The mean of the neutralizing antibody titers to the rDENV2/1 and parental DENV1 and DENV2 viruses did not change significantly over four years post-infection. The NT₅₀ values were compared by one-way ANOVA analysis (n = 4). (D) From year 1 to year 4 post-primary DENV1 infection, the 1F4 epitope accounts for 47, 48, 52 and 47% of the DENV1 type-specific response, respectively. The proportions were compared by one-way ANOVA analysis (n = 4).

Figure 6

The 1F4 epitope is recognized by polyclonal sera from Sri Lankan individuals in the convalescent phase post-primary DENV1 infection. (A) In the convalescent phase, the NT₅₀ values to the DENV1 infecting serotype and the rDENV2/1 are significantly higher than the NT₅₀ values to the DENV2 serotype in Sri Lankan DENV-infected individuals. (B) The proportion of the DENV1 type-specific response attributable to the 1F4 epitope is variable and averages 53% across 12 Sri Lankan individuals. (C) The antigenic cartography map positions viruses (DENV1, DENV2 and rDENV2/1 in teal, purple and yellow, respectively) and plasma (12 open teal squares) and indicates a close proximity between DENV1 and rDENV2/1 viruses. Each grid square corresponds to a 2-fold dilution in the neutralization titer. The NT₅₀ values were compared by one-way ANOVA analysis (n = 12). *p < 0.05, ***p < 0.001.