Susceptibility to Plasmodium vivax malaria associated with DARC (Duffy antigen) polymorphisms is influenced by the time of exposure to malaria

Abstract

Malaria has provided a major selective pressure and has modulated the genetic diversity of the human genome. The variants of the Duffy Antigen/Receptor for Chemokines (DARC) gene have probably been selected by malaria parasites, particularly the FY*O allele, which is fixed in sub-Saharan Africa and confers resistance to Plasmodium vivax infection. Here, we showed the influence of genomic ancestry on the distribution of DARC genotypes in a highly admixed Brazilian population and confirmed the decreased susceptibility of the FY*A/FY*O genotype to clinical P. vivax malaria. FY*B/FY*O individuals were associated with a greater risk of developing clinical malaria. A remarkable difference among DARC variants concerning the susceptibility to clinical malaria was more evident for individuals who were less exposed to malaria, as measured by the time of residence in the endemic area. Additionally, we found that DARC-negative and FY*A/FY*O individuals had a greater chance of acquiring high levels of antibodies against the 19-kDa C-terminal region of the P. vivax merozoite surface protein-1. Altogether, our results provide evidence that DARC polymorphisms modulate the susceptibility to clinical P. vivax malaria and influence the naturally-acquired humoral immune response to malaria blood antigens, which may interfere with the efficacy of a future vaccine against malaria.

Figure 1

Genomic ancestry of the Rio Pardo population. (a) Genomic individual ancestry bar plot, including worldwide populations and the Rio Pardo population. Each individual is represented by a thin vertical line that is partitioned into three colored segments corresponding to the individual’s estimated membership proportion in the European (blue), African (green) and Native American (red) clusters. (b) PCA representation of individual ancestry. EURO, European; AMER, Native American; AFRI, African and BRAZ, Brazilian (Rio Pardo population).

Figure 2

Frequency distribution of DARC genotypes according to the estimated genomic ancestry for the Rio Pardo population. Fitted multinomial logistic regression models showing the association between the distribution of DARC genotypes and (a) European, (b) Native American and (c) African ancestry. The individual proportion of genomic ancestry is shown in the x-axis. The y-axis is labeled on the probability scale and shows the variation in the probability of a genotype along with the variation in ancestry proportions.

Figure 3

Analysis of the risk of clinical P. vivax malaria for the Rio Pardo population. The predicted number of clinical malaria episodes according to the DARC genotype, time of residence in the Amazon region and place of residence in the Rio Pardo community were derived from the Zero-inflated Poisson regression model. (a) Area along or close to the Rio Pardo stream with a higher risk of malaria infection (Riverine area) and (b) Non-riverine area with a lower risk of malaria infection.

Figure 4

IgG total antibody levels against PvMSP1₁₉ during the 12-month follow-up study in the Rio Pardo Settlement. (a) Individuals were clustered according to the anti-PvMSP1₁₉ antibody levels evaluated in three cross-sectional surveys (baseline, 6 months and 12 months): Group 1: low level of response; Group 2: high level of response. The two components explained 90% of the point variability. (b) Anti-PvMSP1₁₉ antibody levels in the follow up for the low-response (light blue) and high-response (dark blue) groups. The results were expressed as the reactivity index (RI).

Figure 5

Association between the DARC genotype and antibody levels against PvMSP1₁₉ during the 12-month follow-up study in the Rio Pardo Settlement. The logistic regression model was adjusted for the time of residence in the endemic area and recent P. vivax infection. Only variables associated with anti-PvMSP1₁₉ antibody levels at a significance level of 0.05 (P < 0.05) were maintained in the final model. The odds ratio (OR, 95% CI) was adjusted using FY*A/FY*B as the reference group. The text in bold indicates that the OR is statistically significant.

Figure 6

Individual anti-PvMSP1₁₉ response evaluated in three cross-sectional surveys (baseline, 6 months and 12 months) according to the DARC genotype. The results were expressed as the reactivity index (RI). The antibody response data for low- and high-response groups are colored by light blue and dark blue, respectively.