IgM antibody responses against Plasmodium antigens in neotropical primates in the Brazilian Atlantic Forest

Abstract

Introduction: Zoonotic transmission is a challenge for the control and elimination of malaria. It has been recorded in the Atlantic Forest, outside the Amazon which is the endemic region in Brazil. However, only very few studies have assessed the antibody response, especially of IgM antibodies, in Neotropical primates (NP). Therefore, in order to contribute to a better understanding of the immune response in different hosts and facilitate the identification of potential reservoirs, in this study, naturally acquired IgM antibody responses against Plasmodium antigens were evaluated, for the first time, in NP from the Atlantic Forest.

Methods: The study was carried out using 154 NP samples from three different areas of the Atlantic Forest. IgM antibodies against peptides of the circumsporozoite protein (CSP) from different Plasmodium species and different erythrocytic stage antigens were detected by ELISA.

Results: Fifty-nine percent of NP had IgM antibodies against at least one CSP peptide and 87% against at least one Plasmodium vivax erythrocytic stage antigen. Levels of antibodies against PvAMA-1 were the highest compared to the other antigens. All families of NP showed IgM antibodies against CSP peptides, and, most strikingly, against erythrocytic stage antigens. Generalized linear models demonstrated that IgM positivity against PvCSP and PvAMA-1 was associated with PCR-detectable blood-stage malaria infection and the host being free-living. Interestingly, animals with IgM against both PvCSP and PvAMA-1 were 4.7 times more likely to be PCR positive than animals that did not have IgM for these two antigens simultaneously.

Discussion: IgM antibodies against different Plasmodium spp. antigens are present in NP from the Atlantic Forest. High seroprevalence and antibody levels against blood-stage antigens were observed, which had a significant association with molecular evidence of infection. IgM antibodies against CSP and AMA-1 may be used as a potential marker for the identification of NP infected with Plasmodium, which are reservoirs of malaria in the Brazilian Atlantic Forest.

Keywords: Atlantic forest; IgM antibodies; Plasmodium; erythrocytic stage antigens; malaria; neotropical primates; pre-erythrocytic stage antigen.

Figure 1

Map of Brazil with the location of the three studied areas. Malaria endemic areas (red) and the Atlantic Forest areas (green) are indicated. The states of the studied areas are highlighted (Rio de Janeiro and Santa Catarina). Icons show three studied municipalities: Indaial (blue), Joinville (yellow), and Guapimirim (grey). The map also highlighted Belo Horizonte (pink) in Minas Gerais state because the negative controls were collected there. The map was made with ArcGIS software.

Figure 2

Seroprevalence and levels of IgM antibodies against Plasmodium antigens in Neotropical primates. Seroprevalence and reactivity index (RI) of IgM antibodies against (A) circumsporozoite protein (CSP) peptides and (B) Plasmodium vivax erythrocytic stage antigens of NP from Indaial and Joinville in the state of Santa Catarina and Guapimirim in the state of Rio de Janeiro. RI>1 was considered positive (dotted line in violin plots). The radar chart indicates the percentage of NP with RI>1. In the violin plots, the data shown are the individual RI values for each sample (dots) with their median and the interquartile range. Pvc, variant classic VK210 of the P. vivax CSP (lilac); Pvk, variant VK247 of the P. vivax CSP (pink); Pvl, variant P. vivax-like of the P. vivax CSP (purple); Pb/Pm, CSP of P. brasilianum/P. malariae (dark blue); Pf, CSP of P. falciparum (dark green); PvAMA-1, P. vivax Apical Membrane Antigen-1 (yellow); PvEBP-2, P. vivax Erythrocyte Binding Protein 2 (Orange); PvDBPII, P. vivax Duffy Binding Protein region II (blue). At the bottom of each graph is shown the numbers (n) of NP included from each area. Statistically significant differences are indicated by lowercase letters for radar plots (a – compared to Pvc (P< 0.05); b – compared to Pvk (P< 0.05); c – compared to Pvl (P< 0.01); d – compare to Pb/Pm (P< 0.05); e – compared to Pf (P< 0.01); f – compared to PvEBP-2 (P< 0.05); g – compared to PvDBPII (P< 0.05), and asterisks for violin plots (*P> 0.05, **P> 0.01, ***P> 0.001, and ****P<0.0001).

Figure 3

Correlation between IgM responses against pre-erythrocytic and erythrocytic stage antigens of P. vivax among Neotropical primates. (A) UpSetR plot of the number of NP with IgM against PvCSP and/or P. vivax erythrocytic stage antigens. (B) Significant correlations between the reactivity index of IgM against PvCSP and P. vivax erythrocytic stage antigens and between the different erythrocytic stage antigens from this Plasmodium species. Data shown are the value of ρ (Spearman’s rank coefficient), its significance as indicated by the P-value, and the dotted lines indicate the 95% confidence intervals. PvCSP, P. vivax CSP repeats (VK210, VK247, and/or P.vivax-like variants); PvAMA-1, P. vivax Apical Membrane Antigen 1; PvEBP-2, P. vivax Erythrocyte Binding Protein 2; PvDBPII, P. vivax Duffy Binding Protein region II.

Figure 4

Statistically significant differences in the seroprevalence and levels of IgM antibodies against Plasmodium antigens according to different sampling parameters. (A) Comparison of IgM seroprevalence and levels against CSP peptides of Plasmodium sp. between (i) free-living (purple) and captive (pink) individuals, and (ii) malaria parasite PCR-positive (PCR+, red for P. brasilianum/P. malariae and salmon for P. simium) and PCR-negative samples (PCR-, blue for P. brasilianum/P. malariae and light green for P. simium). (B) Comparison of IgM seroprevalence and levels against P. vivax erythrocytic stage antigens between (i) non-adult (brown) and adult (orange) individuals and (ii) malaria parasite PCR-positive (PCR+, salmon for P. simium) or PCR-negative (PCR-, light green for P. simium). Samples with an RI>1 were considered seropositive (dotted line). The filled bars indicate the percentage of NP with an RI> 1. The data shown on the graphs of IgM levels are the individual RI values for each sample (dots) with their median and the interquartile range (boxes). The ages of NP were estimated according to Carpenter (1965) and accordingly categorized into non-adult and adult. PvCSP, P. vivax CSP variants (VK210, Vk247, and P.vivax-like); Pb/PmCS, CSP repeats of P. brasilianum/P. malariae; PfCSP, CSP repeats of P. falciparum; PvEBP-2, P. vivax Erythrocyte Binding Protein 2; PvAMA-1, P. vivax Apical Membrane Antigen 1; PvDBPII, domain II of the P. vivax Duffy Binding Protein. Statistically significant differences are indicated by asterisks (*P> 0.05, **P> 0.01, and ****P> 0.0001). The free-living versus captive comparison was performed only for the group of NP from Indaial/SC.

Figure 5

Odds ratios for the IgM antibody responses of Neotropical primates against different Plasmodium antigens according to different sampling parameters. Comparison of the odds ratios for seroprevalence of IgM against various CSP peptides from different Plasmodium species and Plasmodium vivax erythrocytic stage antigens according to malaria parasite PCR positivity and whether sampled individuals were either free-living or captive. PvCSP, P. vivax CSP repeat variants (VK210, VK247, and P. vivax-like); Pb/PmCSP, CSP repeat of P. brasilianum/P. malariae; PfCSP, CSP repeat of P. falciparum; PvAMA-1, P. vivax Apical Membrane Antigen 1; PvEBP-2, P. vivax Erythrocyte Binding Protein 2; PvDBPII, domain II of the P. vivax Duffy Binding Protein. The data shown are only the significant values for the odds ratios (logistic Regression, P<0.05). The free-living versus captive comparison was performed only for the group of NP from Indaial/SC.

Figure 6

Association between active malaria infection and IgM antibody responses against different Plasmodium antigens in Neotropical primates. IgM seroprevalence and reactivity indices (RI>1) of NP against individual malaria parasite antigens, and different combinations of those antigens: P. vivax antigens (A), P. brasilianum/P. malariae antigens (B), and P. falciparum antigens (C). In the leftmost graphs, the absolute number of NP is indicated above the bars. In the graphs showing the IgM reactivity index, the data shown are the individual RI values for each sample (dots), and their median and the interquartile range (violin plot). In the graphs showing the correlations, the data shown are the value of ρ (Spearman’s rank correlation coefficient), with the P-value, and the dotted lines indicate the 95% confidence intervals. PvCSP, P. vivax CSP repeats (VK210, VK247, and/or P.vivax-like); Pb/PmCSP, CSP repeats of P. brasilianum/P. malariae; PfCSP, CSP repeats of P. falciparum; PvMSP1₁₉, fragment of 19kDa of the P. vivax Merozoite surface protein 1; PmMSP1_19: fragment of 19kDa of the P. malariae Merozoite surface protein 1; PfMSP1_19: fragment of 19kDa of the P. falciparum Merozoite surface protein 1; PvEBP-2, P. vivax Erythrocyte Binding Protein 2; PvDBPII, domain II of the P. vivax Duffy Binding Protein; PvAMA-1, P. vivax Apical Membrane Antigen 1. Statistically significant differences are indicated by asterisks (****P< 0.0001).

Figure 7

Heatmap of IgM antibody levels against Plasmodium antigens among different families of Neotropical primates. IgM antibody levels against CSP repeats from different Plasmodium species and P. vivax erythrocytic stage antigens among NP families from Guapimirim/RJ are indicated in different intensities of red. The IgM antibody response was expressed as Log10 (Reactivity Index +1). PvCSP, P. vivax CSP repeats (VK210, Vk247, and/or P.vivax-like); Pb/PmCS, CSP repeats of P. brasilianum/P. malariae; PfCSP, CSP repeats of P. falciparum; PvAMA-1, P. vivax Apical Membrane Antigen 1; PvEBP-2, P. vivax Erythrocyte Binding Protein 2; PvDBPII, domain II of the P. vivax Duffy Binding Protein. Different NP families are indicated by the colors in the first column, and PCR results for Plasmodium are indicated in the second, light green: PCR-positive for P. simium (de Alvarenga et al., 2018); dark green: PCR-positive for P. brasilianum/P. malariae (Snounou et al., 1993); and grey: PCR-negative for Plasmodium. The PCR results shown in this figure were previously published by our group (Costa et al., 2014; de Alvarenga et al., 2015; Alvarenga et al., 2017; de Alvarenga et al., 2018; Nunes et al., 2020).