Seroepidemiological monitoring in sentinel animals and vectors as part of arbovirus surveillance in the State of Mato Grosso do Sul , Brazil

Introduction: From February-September 2010, seroepidemiological surveys were conducted on non-human primates and transmitter vector capture was used to investigate the possible circulation of arboviruses in the municipalities of Bonito, Campo Grande, and Jardim, State of Mato Grosso do Sul, Brazil. Methods: A total of 65 primates from the wild and captivity were used, and potential vectors were captured using Castro and dip nets. Serum samples were tested at the Instituto Evandro Chagas, State of Pará, using the hemagglutination inhibition test to detect total antibodies against 19 different arboviruses. Virus isolation was attempted from serum samples and arthropod suspensions using newborn mice and the C6/36 cell line clone. In addition, identification of the vector species was conducted. Results: From the 19 serum samples from Campo Grande, 1 sample had a 1:20 titer for Flavivirus. From the 35 samples collected in Bonito, 17 samples had antibodies to arboviruses, 4 (11.4%) were positive for Alphavirus, and 5 (14.2%) were positive for Flavivirus. Monotypic reactions were observed for the Mayaro (n = 10) and Oropouche (n = 5) viruses, and 6 (17.1%) samples had titers for >1 virus. We captured 120 Culicidae individuals that were potential arbovirus transmitters in Jardim; however, all the samples were negative for the viruses. Conclusions: Mato Grosso do Sul has a variety of vertebrate hosts and transmission vectors, thereby providing ideal conditions for the emergence or reemergence of arboviruses, including some pathogenic to human beings.

Arboviruses are almost completely maintained in the natural environment; consequently, people who have contact with enzootic foci are most at risk of acquiring infections that represent a public health problem 1 .Arboviruses are classified into 5 families according to their antigenic properties: Bunyaviridae, Flaviviridae, Reoviridae, Rhabdoviridae, and Togaviridae.In Brazil, some arboviruses have appeared regularly in urban areas (e.g., dengue virus [DENV] and Oropouche virus [OROV]) or in peri-urban areas (e.g., Mayaro virus [MAYV] and yellow fever virus [YFV]) as epidemics by infecting susceptible communities and causing febrile rash illness and/or hemorrhagic fevers or central nervous system diseases and meningitis, as is the case for Saint Louis encephalitis virus (SLEV), Rocio virus (ROCV), eastern equine encephalitis virus and western equine encephalitis virus¹ , ².
Studies on arboviruses involve virology, serology, entomology, ecology, and zoology³.Because they are usually diurnal arboreal species, non-human primates are more often infected by an arbovirus than other terrestrial animals; however, most of them have low hemagglutination inhibition antibody titers 4 .Interactions between humans and non-human primates that live in tropical forests are increasingly seen due to the development of ecotourism, subsequently increasing the risk of viral transmission.Non-human primates can act as important hosts in the cycle of several zoonoses, and as they belong in a habitat with high biodiversity, they serve as natural sentinels in the surveillance of several emerging viruses.It is common to use sentinel animals in arbovirus seroepidemiological studies as this represents a feasible method to obtain information about the circulation of these viruses in the natural environment 5,6 .
Serological studies on free-living or captive animals can determine antibody seropositivity rates, which often can lead to an understanding of virus circulation dynamics and host susceptibility;  however, susceptibility may be speculated upon even if the agent has not been isolated 4 .
It is extremely important to study this relationship, considering the direct or indirect participation of non-human primates in maintaining the epidemiological cycles of several arboviruses and that some of these viruses are responsible for diseases that range from asymptomatic cases to hemorrhagic fevers with a high mortality rate.
In these municipalities, the ecosystem is predominated by cerrado, which consists of bushy, small, bent, and twisted trunk trees.The cerrado has well-defined seasons, i.e., a rainy summer season, which occurs from October to April, and a dry winter season, which extends from June to August.
The City of Campo Grande, the capital of Mato Grosso do Sul, is located in the central region, with an elevation of ~532 m above sea level (20°26'34"S and 54°38'47"W).It has an estimated area of 8,096.05km 2 , with a population of ~765,000 inhabitants.CraS is located in an urban area of Park Estadual do Prosa.The park has an approximate area of 1,35km² in Serra de Maracaju Plateau, and it has enclosures for birds, mammals, and reptiles.The non-human primates are housed according to their biological characteristics in collective places and they are fed on fruits, seeds, and vegetables, while the carnivores are fed with live prey in order to encourage them to follow their intuitive hunting practices.

Virus HI* Results
Alphavirus Flavivirus Mayaro Oropouche Total Jardim municipality is located in the southwest of the state (21º28'49"S and 56º08'17"W).It has a territorial extension of 2,207.6km²,with a humid subtropical climate and temperatures of 15-39°C (Figure 1).The City of Bonito is located in Serra da Bodoquena in the southwest of the state; it is ~315m above sea level (21°07'16"S and 56°28'55"W) and has a population of ~25,000 inhabitants.The city is known worldwide for its caves, rivers, resorts, and ponds of clear water.
The animals were trapped using terrestrial Tomahawk traps, armed in a predetermined area, and easily visualized 7 .The animals were anesthetized using tiletamine hydrochloride and zolazepam hydrochloride (4.4mg/kg), with the aid of dip nets, zest leather gloves, and 1mL syringes.The dose was adapted to the animal's weight and it was administered intramuscularly 8 .We collected 3-5mL of blood from sedated animals with a weight of up to 3kg and 6-10mL from animals with a weight of over 3kg, by puncturing the femoral vein or brachial artery.
Later, the collected blood was centrifuged at 1,000 rpm for 10 min to obtain serum.Serum aliquots and whole blood samples were placed in cryovials.The samples were frozen in liquid nitrogen and stored at -70ºC until processing 9 .Biometric data as well as data pertaining to body temperature and heart and respiratory rate were collected (data not reported).For future studies, microchips (transponders) for identification were implanted in the subcutaneous interscapular region, with later confirmation using a specific reader.Until they recovered completely from the anesthesia, the animals were kept in appropriate cages, covered with canvas, and kept away from sources of stress.After recovery from the anesthesia, the animals were released.
The transmitter vectors were captured over 3 days, from to 15:00.The captured Culicidae specimens were placed in 5-mL cryovials, frozen in liquid nitrogen, and stored at -70°C until vector identification and virus isolation.
For virus isolation, newborn mice (Mus musculus) were inoculated intracerebrally with 0.02µL serum and blood isolated from the non-human primates, which were diluted in 1.8mL penistrep (100UI/mL penicillin and 100µg/mL streptomycin).The animals were observed daily for 21 days and no changes were noted on their identification cards.Concomitantly, 40µL of blood and serum from the studied animals (diluted 1:10 in culture medium) were added to different culture flasks that contained an Aedes albopictus cell line (clone C6/36) and Leibnitz culture medium modified with l-glutamine (L-15) plus tryptose, nonessential amino acids, penicillin (100UI/mL), streptomycin (100µg/mL), and fetal bovine serum (5% for growth and 2% for maintenance).Flasks containing confluent cell monolayers were observed daily for 10 days by using an inverted optical microscope for cytopathic effect verification (CPe).Similarly, after filtering, the arthropod vector suspensions were inoculated at a ratio of 1:10 in culture medium.
When observing the CPe, the presence of the virus in the studied samples was confirmed using indirect immunofluorescence 10 with polyclonal Alphavirus and Flavivirus 1 antibodies.
The information obtained was stored in a database using Microsoft ® Excel (2007).

Ethical considerations
This study was approved by the Ethics Committee on Animal Use of the Federal University of Mato Grosso do Sul (no.251/2010), IBAMA (n 21,808-1), and the Institute of Environment of Mato Grosso (no.23/108,891/2009).In addition, this study had technical, logistical, and administrative support, and permanent and consumable materials provided by the State Secretariat of Health of Mato Grosso do Sul.The animal capture and identification techniques were designed to be less invasive in order to preserve the welfare of the animals and to relieve potential stress.
The 11 serum samples collected from non-human primates captured in the town of Jardim and tested using the HI test were negative for the virus.From the 19 serum samples collected from the CraS non-human primate colony, 1 Cebus primate had titers of 1:20 for Flavivirus.From the 35 serum samples collected in Bonito, 17 (48.5%)showed evidence of antibodies for arboviruses, with 4 (11.4%)positive for Alphavirus and 5 (14.2%) positive for Flavivirus.Monotypic reactions were observed for MAYV (n = 10) and OROV (n = 5) (Table 2).Six (17.1%) samples showed positive results for >1 arbovirus (Table 3).
All samples studied yielded negative results for the presence of the viruses in newborn mice and C6/36 cell culture.We captured 120 competent vector species that were associated with the transmission of Alphavirus, Flavivirus, and Orthobunyavirus (Table 4).Many serological studies on free-living animals are limited to transversal seroepidemiologic surveys 4 .With the HI test, it is possible to compare the serological results of 2 samples and detect cases of recent infection with arboviruses, when the second sample has antibody titers that are 4-fold higher than the first.This test is widely used in serological surveys since it can detect antibodies to arboviruses for a long period after natural infection.It is considered a high sensitivity and low specificity test when compared to other tests such as the capture ELISA immunoglobulin M (IgM) antibodies, which is considered the gold standard 1 .

DISCUSSION
In the HI test, it is common to observe cross-reactivity among viruses belonging to different genera.Saint Louis encephalitis virus, West Nile Virus (WNV), and ROCV form an antigenic complex with Japanese encephalitis virus, which complicates the interpretation of HI test results, making it necessary to carry out more specific tests.Thus, HI test results should be cautiously interpreted, especially when the investigated samples come from endemic areas where different arboviruses commonly co-circulate 11,12 .
We found cross-reactivity among viruses belonging to the Flavivirus genus in 5 serum samples, which precluded the identification of the infecting virus and the infection time.Positive results for antibodies against Flavivirus suggest that the host has been previously exposed to some of the arboviruses studied and produced antibodies against them 4 .Due to the presence of cross-immune protection, it is possible that some arboviruses remain silent in Brazil.
The negative results found in the virus isolation and cell culture experiments of this study are consistent with those found in a study of 35 non-human primates in San Pedro, located in the central region of Paraguay, which sought to isolate YFV using the E6 Vero cell line 13 .
For the inoculation of newborn mice and for reverse transcription polymerase chain reaction (RT-PCR), personnel at the Instituto Evandro Chagas isolated YFV from the blood and viscera samples of an Alouatta primate, which died in Anastácio-MS, in 2008 (PM Baptist: unpublished data).
Yellow fever virus was also isolated from a sample of a non-human primate captured in an epidemic area in the State of Rio Grande do Sul (RS).Therefore, the Ministry of Health has standardized epidemic surveillance as part of its yellow fever (YF) surveillance strategy in order to identify feasible wild areas where YFV outbreaks might occur.
During the active surveillance for YF in RS, which occurred from 2002-2007, 181 non-human primates from several regions were caught and antibodies for OROV (n = 1) and SLEV (n = 16) 14 were detected using a neutralization test.
During an epizootic investigation on non-human primates in Bolivia, YFV was detected using RT-PCR 15 .
A study with 570 sera from Alouatta caraya from the Porto Primavera region, which is located in the Presidente Epitácio municipality in the State of São Paulo, reported negative results for immunoglobulin G (IgG) antibodies against YFV, a similar result to those observed in the City of Jardim, suggesting the absence of circulating YFV in this region.Moreover, no outbreak or human cases were detected 16 .
Some arboviruses, such as MAYV, SLEV, and NOV, are introduced in certain regions by migratory birds at certain times of the year, e.g., spring and summer, as they move to regions with favorable climatic conditions for reproduction.Consequently, people who have contact with enzootic foci are most at risk of acquiring infections 17 .
Aiming to detect YFV in non-human primates and transmitter vectors, sentinel animals were used in an epidemiological survey in Chaco and Corrientes provinces in northern Argentina.On that occasion, 4 SLEV serum samples were positive for the HI test.From these, 2 were confirmed by neutralization, suggesting the possible involvement of these primates in the natural life cycle of the virus.SLEV was also isolated in samples from humans, rodents, and arthropods in Argentina 18 .In another seroepidemiologic study with wild primates from the border of Paraná with the State of Mato Grosso do Sul, 123 primates were captured.From the serum samples obtained, 21 were seropositive for SLEV by the HI test 6 , suggesting the involvement of primates in the SLEV maintenance cycle in the Southern Cone region.SLEV was also detected in the State of Pará in Culex declarator and in migratory birds during epizootic cases observed in primates 19 , confirming the importance of these hosts in the disease cycle.
A serological survey in French Guiana on 150 sentinel primates detected high titers of antibodies for MAYV, suggesting the possible movement of this virus in the country.Serum samples from humans showed a high prevalence of antibodies to arboviruses 20 .The high levels of antibodies to MAYV found in the sera of animals trapped in the City of Bonito suggest the possible circulation of this virus in this location.MAYV antibodies were also found in Callithrix argentata primates by the HI test during an investigation of Mayaro and YF outbreaks in Belterra in the State of Pará 19 .

METHODS
Batista PM et al -Arbovirus monitoring in Mato Grosso do Sul, Brazil

FIGURE 1 -
FIGURE 1 -Map of the State of Mato Grosso do Sul.The municipalities where the seroepidemiological non-human primate surveys were conducted are highlighted.