Survey of Plasmodium in the golden-headed lion tamarin (Leontopithecus chrysomelas) living in urban Atlantic forest in Rio de Janeiro, Brazil

Communicating the presence of potential zoonotic pathogens such as Plasmodium spp. in wild animals is important for developing both animal and human health policies. The translocation of an exotic and invasive population of Leontopithecus chrysomelas (golden-headed lion tamarins) required the screening of these animals for specific pathogens. This studies objective was to investigate Plasmodium spp. infection in the L. chrysomelas, both to know its prevalence in these animals in the local area and to minimize the risk of pathogens being translocated to the destination site. To investigate Plasmodium spp. infection, blood samples from 268 animals were assessed for the presence of Plasmodium spp. by genus-specific PCR and stained thick and thin blood smears were examined by light microscopy. Data of human malaria infection in the studied region was also assembled from SINAN (Diseases Information System Notification—Ministry of Health of Brazil). Results from the PCR and microscopy were all negative and suggested that no L. chrysomelas was infected with Plasmodium spp. Analysis of SINAN data showed that malaria transmission is present among the human population in the studied region. This study is the first to provide information on Plasmodium spp. infection in L. chrysomelas.Plasmodium spp. infection of this species is rare or absent though malaria parasites circulate in the region. In addition, there is minimal risk of translocating Plasmodium spp. infected animals to the destination site.

non-human primates and humans are exposed. Plasmodium brasilianum and P. simium are genetically and morphologically very similar to human parasites Plasmodium malariae and Plasmodium vivax, respectively [6][7][8]. Additionally, the human parasite Plasmodium falciparum has also been documented in non-human primates in the Brazilian Amazon [9]. Recently, in the Venezuelan Amazon, 12 individuals were found to be infected with parasites that had identical 18S gene sequences to P. brasilianum that had been isolated from the spider monkey (Alouatta seniculus) [10]. Therefore, it is possible that non-human primates in Brazil are reservoirs for parasites that could infect humans.
The translocation of wild animals has been highlighted as an important factor in emerging infectious diseases of wild animals [11], which in turn can threaten the survival of endangered species [11]. There is a possibility in any translocation programme that the introduction of animals into a new geographical area will involve the risk of also introducing new pathogens [12,13], and therefore there is a responsibility to take steps to minimize this risk.
Due to the presence of non-human primate Plasmodium spp. infections [14] and the Plasmodium spp. vectors [15] in the Atlantic forest as well as locally acquired Plasmodium spp. infections in humans in Rio de Janeiro state [16,17] it was determined that non-human primates living in the Atlantic forest around Rio de Janeiro could be at risk of being infected with Plasmodium spp. parasites.
The translocation of an exotic invasive population of Leontopithecus chrysomelas (golden-headed lion tamarin) [18] living in the urban Atlantic forest in Niteroi, Rio de Janeiro state to their native area (Bahia State, Brazil) [19], required the screening of the L. chrysomelas animals for pathogens, including Plasmodium spp., according to a national ruling in Brazil (IN no 179/08) [20] and International recommendations [21]. The animals were being translocated due to the risk of competition and hybridization with the endemic and endangered [22] Leontopithecus rosalia (golden lion tamarin).
The primary outcome of the screening was to identify whether there was any risk of translocating Plasmodium spp. infected animals from the original site (Rio de Janeiro state) to the destination area (Bahia state) during the L. chrysomelas translocation program, where they may pose a risk to both translocated animals and the local animals (although there was no golden-lion tamarin in the release site, there are other primate species). Alternatively, in cases of zoonosis, potentially presenting a risk to the human population living in the area [11,13,23]. In addition, information on the presence of Plasmodium spp. parasites in these animals could provide important information on the likelihood of these animals being reservoirs of parasites that could cause human disease. A small survey of Plasmodium spp. has already been conducted for L. rosalia with negative results (reviewed in [2]), however, this is the first work investigation involving L. chrysomelas.

Animals and capture
Three hundred and thirty-five Leontopithecus chrysomelas, Family Callitrichidae, were captured in an urban Atlantic forest fragment in the municipality of Niterói (Serra da Tiririca State Park, 22°56′S, 43°00′W) in the state of Rio de Janeiro, Brazil ( Fig. 1) between June 2012 and November 2013. Their capture was part of a programme (the Tamarins Translocation Project) to remove L. chrysomelas from the range of L. rosalia and to translocate this invasive L. chrysomelas population to their natural area of occurrence in another state of Brazil (Bahia). No animals suffered, died naturally or were euthanized for the purpose of, or during the course of this research.

Quarantine, anesthesia and sample collection
After capture, animals were kept in quarantine for 30 days at the Rio de Janeiro Primatology Centre (CPRJ/ INEA) in Guapimirim (22°32′S, 42°59′W) before being translocated to Bahia state. During quarantine, the animals were kept in a quiet place, away from human contact, in large and modular cages. The size of the cages varied according to the number of animals per group and dividers for visual separation between the groups were used. In each cage, there were tree trunks for the animals to move around, and wooden housing for the animals to sleep or hide in (mimicking the behaviour of the species in the wild). Protective measures were taken to prevent excessive exposure to heat including water diffusers on the ceiling of the room and air conditioners inside the building. The animals received water ad libitum and food (fruits, vegetables, chow primates, eggs, mealworms) two times a day. All measures to avoid stress and to ensure the well-being were applied and this also helped with the translocation process. To collect blood, chemical restraint was performed using 8-10 mg/kg of ketamine (CETAMIN ® , Syntec) with 0.25 mg/kg midazolam hydrochloride (Dormium ® , União Química) in order to conduct a clinical evaluation which included venous blood sampling and noting the animals age and sex.

Processing and analysis of samples for Plasmodium spp
Thick and thin blood smears of venous blood were made and stained with Giemsa, using standard methods [24] and examined under 1000× light microscopy. One to two hundred µl of packed red blood cells were stored at −20 °C. Within 3 weeks of collection, DNA was extracted from the packed red blood cells using the Illustra ™ blood genomicPrep Mini Spin kit (GE Healthcare) following the manufacturer's instructions and genus specific nested PCR (first reaction rPLU1: 5′ tcaaagattaagccatgcaagtga 3′ and rPLU6: 5′ cgttttaactgcaacaattttaa 3′; second reaction PLU3: 5′ tttttataaggataactacggaaaagctgt 3′ and rPLU4: 5′ tacccgtcatagccatgttaggccaatacc 3′ was carried out according to previously described protocols [25,26], in order to detect the presence of Plasmodium spp. DNA. Further, 10 % of samples from adults were randomly selected and re-tested by using Plasmodium ssrRNA primers rPLU6 5′ ttaaaattgttgcagttaaaacg 3′ and rPLU5 5′ cctgttgttgccttaaacttc 3′, as defined by Snounou et al. [27]. A blood sample from an individual known to be infected with Plasmodium was run as positive control in each PCR reaction. The DNA was extracted in the same way for the control as for the samples.

Obtaining information of local infections
Data on Plasmodium spp. infections in humans in Rio de Janeiro state during the period 2001-2014 were obtained from the Ministry of Health of Brazil-Diseases Information System Notification (Sistema de Informação de Agravos de Notificação-SINAN)-website [28]. The area of interest was defined as Rio de Janeiro State and then within Rio de Janeiro State the municipalities around the Tiririca State Park (Rio de Janeiro, Duque de Caxias, Magé, Guapimirin, Itaboraí, São Gonçalo, Niterói, and Maricá) (Fig. 1).

Results
Three hundred and thirty-five animals from 56 groups were captured; most groups were between 4 and 8 animals in size. Of the 335 animals, 268 animals (127 male, 116 female, 25 of unknown sex, including 24 infants) were tested for Plasmodium spp. and had results for all three samples (thick and thin smears and PCR of peripheral blood); some animals were not sampled due to body size (see Fig. 2). No Plasmodium spp. were visible in any of the smears and none of the blood samples from the animals contained DNA that was amplified by Plasmodium spp. specific primers.
According to SINAN from the Brazilian Ministry of Health . Moreover, reported cases from the municipalities near to the Serra da Tiririca State Park, comprise of malaria infections of residents or non-residents that were acquired elsewhere and from residents or non-residents that were acquired in the region ( Fig. 1; Table 1).

Discussion
This study describes the screening of a large wild nonhuman primate population (L. chrysomelas) for Plasmodium spp. None of the animals were infected. Results agree with previous work, which found Plasmodium spp. infection absent in other tested animals of the genus (L.

rosalia), and only very low frequency infection of family
Callitrichidae, even in areas where infection in other species was common [2,9,25,29,30]. For example, infection of the family Callitrichidae with Plasmodium spp. has been found in only four animals out of 604 Callitrichidae animals tested in Brazil between 1937 and 2013 [2,9,25,29,30] with the parasite being identified as P. brasilianum. In addition two species Callithrix penicillata, and Callithrix jacchus that were also introduced in Rio de Janeiro, and can be observed in the same area as the introduced L. chrysomelas [31], have also been surveyed for Plasmodium spp. infection (at other geographical locations) by light microscopy and like the L. chrysomelas were negative for infection [2].
Parasites do not randomly infect hosts, it is unlikely that any Plasmodium spp. infects all available primates, although it is still unclear which Plasmodium spp. can infect which species of primates. Plasmodium brasilianum has been shown to be able to infect a wide variety of non-human primates (reviewed in [3]), but L. chrysomelas may not be susceptible to infection. This could be because their position in the canopy, sleeping in hollows of trees from dusk to sunrise [32], when Anopheles are more active, and their small body size may result in low vector exposure (reviewed in [33]). Alternatively, L. chrysomelas may not provide an environment suitable for parasite survival [3]. It is also possible that the negative results are due to a low frequency of Plasmodium spp. in the area.
It is unlikely that this work contains false negative results as the negative results were due to agreement between both the microscopy results of thin and thick smear slides as well the as genus specific PCR, furthermore, the results were further confirmed with a different set of PCR primers in 10 % of randomly selected samples.

Fig. 2 Flow chart of animals captured and tested for infection with
Plasmodium spp. between 2012 and 2013. *Animals weighing less than 300 g were not sampled. **Nine animals not tested for Plasmodium as they were not to be translocated to Bahia. ***Three PCRs, 5 thin smears and 10 thick smears from 17 animals did not give results due to bad quality or lost samples

Table 1 Plasmodium infections in humans a between 2001 and 2014 reported in the area of interest of Rio de Janeiro State b , according to the-Diseases Information System Notification (SINAN) from Brazilian Ministry of Health
a Includes infections that were reported (imported and autochthonous cases) within the municipalities nearby the park, infections could occur in residents (see in Fig. 1) or non-residents b The area of interest are the municipalities Rio de Janeiro, Duque de Caxias, Magé, Guapimirin, Itaboraí, São Gonçalo, Niterói, and Maricá* near the Serra da Tiririca State Park (see in Fig. 1 Though this study agrees with findings in the literature, which shows animals from the same genus or even from the same family are often negative for infection [2,9,25,29,30], it is still possible that other non-human primates in the area are infected with Plasmodium spp. Data from the website SINAN showed that endemic infections in areas around the park do occur (Table 1) and interestingly Anopheles aqualasis, an efficient transmitter of Plasmodium spp., has been identified specifically in the same area where the L. chrysomelas were captured [34]. In addition, other studies in the Atlantic forest have reported Plasmodium spp. infections in vectors [15] and non-human primates [14], including in 30 % of tested non-human primates living in and around the Rio de Janeiro Primatology Center (CPRJ/INEA) in Guapimirim [35] which is located the same area of this study.
Leontopithecus chrysomelas in Rio de Janeiro often live in high-risk interfaces, in close proximity with humans occasionally accessing backyards or housing, which could increase the risk of transmission, especially when human residents are infected with Plasmodium. In addition, if they were infected with Plasmodium, this proximity would identify them as possible reservoirs for Plasmodium spp. which may in turn infect the local population; however these results suggest that this is unlikely to occur.
With the translocation of the animals, there was the responsibility to ensure that this action would not also result in the transfer of pathogens to potentially susceptible wild animals and/or the local human population. The results from this work show that there is a minimum risk of transfer of Plasmodium spp. to Bahia state. However, establishing the risk factors for emerging infectious diseases is challenging and even the introduction of disease-free animals into an area could pose a threat as it has the potential to alter variables like population density [11], therefore, any translocation programme needs to be thoroughly thought through.
Much research still needs to be done in the field of Plasmodium in non-human primates. The evolutionary history of P. brasilianum and P. simium is unclear (primarily due to a lack of primate samples [7]). However, data suggests that the transfer from humans to nonhuman primates (or vice versa) was recent and in the case of P. brasilianum has occurred more than once [7]. Also, though P. brasilianum can infect humans in an experimental setting [5], it is still unclear if it is likely to pose a risk to humans. As many countries strive to eliminate malaria, whether species such as P. brasilianum and P. simium can be readily transmitted between non-human primates to human populations needs to be clarified. In addition, any animal population likely to be a reservoir needs to be identified.

Conclusions
This study is the first to provide information about Plasmodium spp. infection in a population of L. chrysomelas of the urban Atlantic forest of Rio de Janeiro state.
The results indicate that Plasmodium spp. infection in L. chrysomelas population, from Serra da Tiririca Park, is rare or absent despite malaria parasites circulating in the region among vectors, humans and other non-human primates. The absence of infection provides no evidence to support that this species is a reservoir for parasites infecting humans and also suggests that the translocation of the tested animals exhibits a minimal risk of translocating Plasmodium spp. infected animals to Bahia state. In addition, this study promotes knowledge about Plasmodium parasites in the same areas that endangered species inhabit, such as Leontopithecus rosalia, which could represent a further threat for these tamarins and other primates.

Authors' contributions
EHA carried out the genus-specific PCR, examined thick and thin smears by light microscopy and drafted the paper; MGB and AP collected the samples and clinical information of the animals, helped analyse the results and draft the manuscript; LSO helped carry out the genus specific PCR; MCMK helped collect the samples and draft the manuscript. JMA helped draft the manuscript. JLCD conceived the study and helped draft the manuscript. SE conceived the study, carried out the epidemiological analysis from the SINAN website and drafted the manuscript. All authors read and approved the final manuscript.