Malaria remains an important public health problem despite of many efforts to control the disease around the World. According to the latest estimates of the World Health Organization, there were about 229 million human malaria cases and 409 000 deaths in 2019 (World Health Organization, 2020). Five Plasmodium species are more frequently associated with human infection, Plasmodium falciparum, P. vivax, P. malariae, P. ovale and P. knowlesi. The last species was recently described in a generalized zoonotic transmission in Southeast Asia (Cox-Singh, 2012). In Brazil, more than 150 000 cases of malaria were registered in 2019 (World Health Organization, 2020). Malaria is endemic almost exclusively within the Brazilian Amazon region, caused mainly by P. vivax (89%), followed by P. falciparum (11%) and with few notified cases of P. malariae (< 1%) (Ministerio da Saude, 2020). Nevertheless, 732 malaria cases were registered in 2018 in the extra-Amazon Region, including the Atlantic Forest region. Most of the cases in this region were diagnosed as P. vivax (de Pina-Costa et al., 2014). Recently, Brasil et al. reported that cases from an outbreak in the Atlantic Forest of Rio de Janeiro state, initially diagnosed as P. vivax infection, were in fact caused by Plasmodium simium, a neotropical primate (NP) parasite (Brasil et al., 2017).

In the Brazilian forests, there are two malaria parasites primarily infecting primates: Plasmodium brasilianum and Plasmodium simium. P. simium infects primates of the Atlantic Forest from South and Southeastern Brazil of Atelidae, Cebidae and Pitheciidae families (da Fonseca, 1951; Deane, 1992; de Alvarenga et al., 2015). Alouatta guariba clamitans (southern brown howler monkeys) is the species most frequently infected by P. simium (Deane, 1992; de Alvarenga et al., 2015; Abreu et al., 2019). P. brasilianum infects primates of all neotropical primate families distributed from Central America to the South of Brazil (Seidelin, 1912; Deane, 1992; Lourenço-de-Oliveira and Deane, 1995; Lalremruata et al., 2015; Alvarenga et al., 2017). Of importance, P. simium and P. brasilianum are morphological, genetic, and immunologically similar to human Plasmodium species, P. vivax and P. malariae, respectively (Coatney, 1971; Seed, 1976; Cochrane et al., 1985; Barnwell, 1986; de Arruda et al., 1989; Deane, 1992; Goldman et al., 1993; Escalante et al., 1995; Escalante et al., 1998; Fandeur et al., 2000; Volney et al., 2002; Tazi and Ayala, 2011; de Alvarenga et al., 2015; Lalremruata et al., 2015; Brasil et al., 2017). Brasil et al. and Lalremruata et al. molecularly characterized P. simium and P. brasilianum infections in humans and NPs, indicating their zoonotic transmission and reinforcing that non-human primates may act as malaria reservoirs (Lalremruata et al., 2015; Brasil et al., 2017). Mourier et al. showed the close similarity of whole genomes of P. simium from human and monkeys, and comparing the genes involved in the erythrocyte invasion identified a possible adaptation of the parasite for human and monkey infections in the Brazilian Atlantic Forest and reinforce the hypothesis of zoonotic transmission of P. simium/P. vivax (Mourier et al., 2019). In this context, zoonotic malaria constitutes a major challenge for malaria elimination.

Considering the genetic proximity between P. simium and P. vivax, and between P. brasilianum and P. malariae, the study of these species contributes to understanding their evolution in the America region. However, there are only a few studies of monkeys naturally infected by P. simium and P. brasilianum and even a smaller number of studies of the immune response in these animals (Duarte et al., 2006; Yamasaki et al., 2011; Costa et al., 2014; Monteiro et al., 2020). Thus, the identification of species from different families of NPs will enable a better understanding of these species as potential malaria reservoirs and whether these Plasmodium species in their natural hosts showed a similar immune response compared to human malaria.

The present study comprises an unprecedented study of malaria antibodies, using both Plasmodium pre-erythrocytic and different erythrocytic stage antigens in NP from distinct areas of the Brazilian Atlantic Forest.

Neotropical primates infected with P. simium and P. brasilianum in the Brazilian Atlantic Forest have been largely reported (Deane, 1992; Duarte et al., 2006; Yamasaki et al., 2011; Costa et al., 2014; de Alvarenga et al., 2015; Alvarenga et al., 2017; Monteiro et al., 2020; Nunes et al., 2020). However, the humoral immune response of NPs naturally exposed to Plasmodium spp. transmission in this biome is still poorly understood. This is the first study that evaluates the humoral response of neotropical primates from different areas of the Atlantic forest against pre-erythrocytic and erythrocytic antigens of malaria parasites. The detection of antibodies against CSP peptides, the major pre-erythrocytic antigen covering the entire sporozoite surface, was evaluated to access the immune responses of NPs exposure to infected mosquito bites. Additionally, the presence of IgG against erythrocytic P. vivax antigens was evaluated to identify immune response in potentially well-established Plasmodium blood infection.

In the three studied areas, most neotropical primates had antibodies against P. vivax CSP variants, in slightly higher frequency against P. vivax VK210 and VK247. These CSP variants were the most frequent for P. vivax and the only ones described for P. simium (Goldman et al., 1993; Oliveira-Ferreira et al., 2004; Santos et al., 2019). Herein, NPs also showed antibodies against P. brasilianum/P. malariae CSP, as previously reported for NPs from Atlantic Forest in the São Paulo state (Duarte et al., 2006; Yamasaki et al., 2011). High seroprevalence against P. falciparum CSP was observed here, in concordance to previous reports of NP from other states (Duarte et al., 2006; Yamasaki et al., 2011). Although, all our NP samples were negative for P. falciparum infection, and only few descriptions of P. falciparum infected humans and mosquitoes have been described in the Atlantic forest area (Cerutti et al., 2007; Maselli et al., 2014; Laporta et al., 2015). The occurrence of a certain cross-reactivity degree among CSP repeats from different Plasmodium species, and also among P. falciparum CSP and asparagine-rich proteins from asexual parasite stage of other Plasmodium species in humans as well as non-human primates has been reported (Hope et al., 1984; Cochrane et al., 1990; Cochrane and Maracic, 1991; Volney et al., 2002; Hall et al., 2019). However, the association of seroprevalence against CSP and PCR positivity for specific Plasmodium species (P.vivax/P.simium and P.brasilianum/P.malariae) argued against the cross-reactivity. Therefore, further experiments are still needed to confirm the circulation of P. falciparum in the Atlantic Forest and immune response studies using erythrocytic antigens of P. falciparum and other Plasmodium species, such as P. malariae/P.brasilianum.

Based on the high similarity between P. simium and P. vivax, recombinant proteins of three leading erythrocytic stage surface proteins of P. vivax, PvAMA-1, PvDBPII, and PvEBP-2, were used here to evaluate the antibody response of NPs from Atlantic forest. In general, 76% of NP showed antibodies against P. vivax erythrocytic antigens. PvEBP-2 was the most frequently detected antigen and with the highest levels of antibodies. Recently, PvEBP-2 was suggested as a novel ligand for a potential alternative invasion pathway of Duffy positive reticulocytes (Hester et al., 2013; Ntumngia et al., 2016). Papua New Guinean children had IgG against PvEBP-2 correlated with protection against clinical malaria (França et al., 2017; He et al., 2019). The first evaluation of antibodies against P. vivax EBP-2 in non-human primates was undertaken here. High frequency of seroprevalence (63%) against MSP-1, another erythrocytic antigen, from different Plasmodium species was also demonstrated for animals from Atlantic forest in São Paulo and Santa Catarina states (Yamasaki et al., 2011; Monteiro et al., 2020). This seroprevalence was higher for free-living animals from the Atlantic forest than for animals from Amazon or Cerrado biome regions (Monteiro et al., 2020). These findings corroborate our suggestion of a high circulation of Plasmodium among NP in the Atlantic Forest area (Deane, 1992; de Pina-Costa et al., 2014; Brasil et al., 2017). Significant correlations of the humoral response against pre-erythrocytic and erythrocytic antigens found herein have been also described in humans (França et al., 2017; He et al., 2019).

Free-living animals showed significantly higher levels of antibodies against Pb/PmCSP, PvEBP-2 and PvDBPII as compared to captive animals from Indaial/SC. This finding of high levels of antibodies in free-living animals can be attributed to the forest ecosystem of the region and closer proximity of free-living animals and anopheline species, which are responsible for the parasite transmission (Marrelli et al., 2007). No significant differences were found in the percentage of responders neither in levels of antibodies when were compared males and females NP, suggesting similar exposure to malaria infection. In humans, men from some endemic areas showed higher responsivity against malaria antigens, probably due to their occupation which might increases exposure to the malaria transmission (Del Giudice et al., 1990; Kale et al., 2019). The humoral response against CSP peptides from different Plasmodium species was higher in adults than non-adult animals, suggesting that the humoral immune response is correlated with the cumulative exposure to malaria infection throughout their lives, as demonstrated in humans (Tapchaisri et al., 1985; Esposito et al., 1986; Griffin et al., 2015). No significant differences were observed in the frequency of responders against erythrocytic antigens of P. vivax comparing adult and non-adult animals. However, the levels of antibodies against PvEBP-2 and PvDBPII were significantly higher in adults, suggesting that the magnitude of humoral response depends on the repetitive cycles of erythrocytic stages. Since this study was not designed for comparison between adult and non-adult monkeys and because of the difficult to access non-adult animals, the number of adults is 5 times higher than non-adult animals. Therefore, these findings must be carefully analyzed. Although of this limitation, at least one comparison had a strong statistic power (DBP-II, 0.847), confirming that in fact adults might have stronger response for specific antigens. Moreover, in humans, the humoral immune response against erythrocytic antigens also increases with age due to repeated exposure to malaria parasites (Del Giudice et al., 1990; Kano et al., 2012; Hester et al., 2013; Griffin et al., 2015; Stanisic et al., 2015).

The level of antibodies and rate of seroprevalence against Pb/PmCSP were higher in NP infected by P. brasilianum/P. malariae. The seroprevalence against PvEBP-2 and PvDBPII and level of antibodies against DBPII were higher in P. simium infected NP. Similar results were observed in Papua New Guinean children with a concurrent P. vivax infection which had significantly higher IgG levels against PvDBPII and PvEBP-2 (He et al., 2019). The absence of correlation between PCR positivity and antibody responses against the other antigens PvCSP and PvAMA-1 observed here was also previously observed in humans (Burkot et al., 1989; Kilombero Malaria Project, 1992; Campo et al., 2011; Pereira et al., 2018). However, in PCR negatives monkeys, we cannot rule out the possibility of parasites being in hematopoietic niches, which would explain the absence of parasites in the peripheral circulation (Obaldia et al., 2018; Silva-Filho et al., 2020).

In Indaial and Joinville we sampled only Alouatta guariba clamitans, but in Guapimirim we included primates belonging to all Platyrrhini families. Thus, the evaluation of the immune responses among animals from Guapimirim provide an excellent opportunity to compare how each monkey family responds against Plasmodium sp. antigens under the same epidemiological conditions. Overall, most monkey members of the five families had IgG antibodies against at least one CSP repeat, 100% of Atelidae and Pithecidae, 92% of Cebidae, and 50% of Callitrichidae. The reactivity was homogeneous across different peptides, i.e., most animals responded to all three CSP peptides. However, reactivity against P. vivax erythrocytic antigens was observed for some antigens but not others. IgG antibodies against at least one of these antigens was detected in all specimens of Pitheciidae, in half of Atelidae specimens and half of Callitrichidae specimens, but in only 17% of Cebidae specimens. Alouatta guariba clamitans of Atelidae family, from Atlantic Forest of São Paulo and Santa Catarina states, previously showed antibodies against CSP, MSP1₁₉ and crude blood stage antigens (by IFA) of P. vivax, P. falciparum and P. malariae/P. brasilianum (Duarte et al., 2006; Yamasaki et al., 2011; Monteiro et al., 2020). Animals of Cebidae family previously showed antibodies against CSP peptides but did not against erythrocytic antigens, and animals of Callitrichidae family did not have antibodies against either pre-erythrocytic or erythrocytic antigens (MSP1₁₉ and crude blood stage antigens – IFA) for different Plasmodium species (Duarte et al., 2006; Yamasaki et al., 2011; Monteiro et al., 2020). Therefore, all families seem to be exposed to malaria transmission, but the establishment of blood malaria infection might only occur in Atelidae family, which could be the potential malaria reservoirs. Although, some species of Pitheciidae family also showed antibody response against erythrocytic antigens, they might not be important reservoirs in the Atlantic Forest because of the two studied species only Callicebus personatus is native of this region. Moreover, this conclusion must be taken with caution since some families were very poorly studied, in number of specimens and species. Interestingly, Alouatta guariba clamintas and Brachyteles arachnoides, species initially described infected by P. simium (Deane et al., 1968; Deane, 1992), showed antibodies against CSP and also erythrocytic antigens of Plasmodium sp. Alouatta is the genus with highest frequency of Plasmodium infection in the Atlantic forest (Deane, 1992; Duarte et al., 2006; Yamasaki et al., 2011; de Alvarenga et al., 2015; Alvarenga et al., 2017; Abreu et al., 2019). In conjunction with our findings of humoral immune response, this genus, particularly A. g. clamitans was suggested as the main potential reservoir of malaria in the Atlantic Forest (Abreu et al., 2019). However, the detection of gametocytes (sexual stage of parasite) and transmission experiments are still needed for evaluation of the potential for transmission and confirmation of their role as malaria reservoirs. Therefore, the growth of human intrusion in forest areas reinforces the need of the inclusion of the Atlantic Forest as malaria transmission area. In this region, malaria prevention and control depend on environmental education strategies for the local population and tourists, and epidemiological surveillance of zoonotic malaria.

Our results show that both male and females of neotropical primates from Brazilian Atlantic Forest have had contact with Plasmodium sporozoites. However, it seems that mainly adults of some species of Atelidae family constantly exposed to malaria transmission might develop an erythrocytic infection, potentially serving as reservoirs of malaria. Taken together, these data suggest that immune response recorded in naturally infected monkeys is similar to that reported in human exposed to malaria in endemic areas (Kano et al., 2012; França et al., 2017; Pereira et al., 2018; Kale et al., 2019).

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by Fiocruz Research Ethical Committee (CEUA license L037/2016) and by the Brazilian Ministry of Environment (SISBIO numbers 43375-4, 43375-6, 54707-137362-2, 52472-1, 28953-1).

CA, LC, JO, and GA designed and conception the study. GA, DA, APC, JS, AN, AP, SM, HP, VP, and ZH were responsible for sample collection. LM, HL, FK, IS, FN, JA, and JO were responsible for the production of Plasmodium proteins and peptides. GA, DA, MC, JS-A, and JO performed the experiments. GA, DA, TS, FK, JO, LC, CD-R, and CA analyzed and interpreted data. GA, DA, and CA wrote the first draft. All authors contributed to the article and approved the submitted version.

This study was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológia (CNPq - Grant nos. 457274/2014-0, 310477/2017-4), Fundação de Amparo a Pesquisa do estado de Minas Gerais (FAPEMIG – Grant no. CBB-APQ-02620-15), Fiocruz Inova Grant for innovative products (VPPIS-004-FIO-18-16), and the Secretary for Health Surveillance of the Brazilian Ministry of Health (IOC-017-FIO-17 and IOC-028-FIO-18). This study was partially supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer TO declared a shared affiliation, with no collaboration, with one of the authors, IS, to the handling editor at the time of review.