Serosurvey of antibodies against spotted fever group Rickettsia spp . in horse farms in Northern Paraná , Brazil Soroprevalência de anticorpos contra Rickettsia spp . do grupo febre maculosa em equinos de haras no Norte do Paraná , Brasil

Brazilian spotted fever (BSF) is an emerging disease most likely caused by Rickettsia rickettsii. The objective of the present study was to estimate the seroprevalence of BSF rickettsia infections in equines from six horse farms located in Londrina County, Paraná, Southern Brazil. Six owners of horse farms situated in Cambé, Santa Fé, Guaraci and Londrina municipalities participated in the study. All farms were located in areas where BSF has not been reported. A total of 273 horses were sampled and their sera were tested by indirect Immunofluorescence assay (IFA) using R. rickettsii and R. parkeri antigens. Titers equal to and greater than 64 were considered positive. Of 273 sera tested, 15 (5.5%) reacted to R. rickettsii and 5 (1.8%) to R. parkeri. Five out of the six farms studied revealed seropositive animals and seropositivity rate ranged from 0 to 13%. The titers ranged from 64 to 512, and four samples had a titer of 512. Nine animals reacted to R. rickettsii with titers four-fold higher than those for R. parkeri. These results suggest that horses in Northern Paraná may have been exposed to rickettsiae identical or closely related to R. rickettsii.

Brazilian spotted fever (BSF) is an acute tick-borne disease caused by R. rickettsii and its clinical signs include fever, headache, myalgia, nausea and rash.The first case of BSF was reported in 1929 in the State of São Paulo (DIAS; MARTINS, 1939) and in recent years BSF has been reported in the State of Minas Gerais, Rio de Janeiro, Santa Catarina, Rio Grande do Sul and Paraná (FREITAS et al., 2010).
Indirect immunofluorescence assay (IFA) is currently the test of choice for the diagnosis of rickettsioses in humans and animals.However, cross-reactivity between Rickettsia species is seen in IFA.IFA does not allow to determining the etiologic Rickettsia species, but by comparing titers among different antigens, serological similarities and presumptive etiologic agents can be inferred (LASCOLA; RAOULT, 1997;HORTA et al., 2004;PACHECO et al., 2007).
Rickettsia spp.are transmitted by ticks of Amblyomma genus, and the most important species responsible for transmitting R. rickettsii that causes BSF is A. cajennense, which is often identified in humans with BSF (MOREIRA; MAGALHÃES, 1935).Horses are primary hosts for all parasitic stages of A. cajennense, and they have been described as excellent sentinels for BSF in many studies (SANGIONI et al., 2005;HORTA et al., 2007;PINTER et al., 2008).
The objective of the present study was to obtain seroepidemiological information about spotted fever group (SFG) rickettsia infections among horses on farms located in Northern Paraná, southern Brazil, through serological testing by IFA using R. rickettsii and R. parkeri antigens.
The study was carried out on six horse farms located in Londrina County (23° 18' 39" S and 51° 09' 53" W), Northern Paraná, a region with no reported BSF cases.A total of 273 horse blood samples were collected and sera were separated and stored at -20 °C before use.The number of samples collected at each farm was as follows: 50 in Cambé, 44 in Santa Fé, 25 in Guaraci, 35 in Londrina 1, 34 in Londrina 2, and 85 in Londrina 3.All of them are in a rural area, except for Londrina 1 farm that is located in an urban area.Londrina 1 farm is an equestrian school where horses are kept in stables with no access to pastures.In all other farms the animals were kept in pastures and had access to a forested area.Tick control was carried out by spraying with pyrethroids and oral ivermectin when ticks were found on the animals.The collection of animal blood was approved by the Universidade Estadual de Londrina (UEL) Ethics Committee on Animal Experiments (number 82/2006).
All horse sera were tested by IFA following Horta et al. ( 2004) protocol using two Rickettsia species: R. rickettsii strain Taiaçu (PINTER; LABRUNA, 2006) and R. parkeri strain São Paulo (SILVEIRA et al., 2007).Briefly, Rickettsia species were cultivated on Vero cells until 100% of cells were infected.Slides were then prepared by harvesting cells and air drying onto multiwell Teflon-coated glass slides, fixed in acetone, and stored at -20 °C before use.The slides were examined using an epifluorescence microscope (Olympus, Japan) and titers ≥64 were considered positive.
Table 1 summarizes the serological results of IFA for the six horse farms studied.Seropositive titers ranged from 64 to 512.Of 15 positive sera, 11 had R. rickettsii titers at least four-fold higher than R. parkeri titers.These sera were considered reactive to R. rickettsii or very closely related species (Table 2).
Two different tick species identified as A. cajennense and Anocentor nitens were detected on the horses during sampling.The study variables ciliary forest (p = 0.029) and wild animals (capybaras / Hydrochoerus hydrochaeris and New World opossums / Didelphis spp.) (p = 0.029) showed significant statistical associations with R. rickettsii seropositivity.Only animals more than one year old were seropositive for R. rickettsii.There were no seropositivity differences between gender and breed of the animals.Ticks were found only on 2.9% of the horses studied.
The rate of seropositive horses found in this study is similar to that reported by Freitas et al. (2010)  discrepancy could be explained by environmental and animal tick infestations, which were higher in urban Londrina than in horse farms.
Our study found only few A. cajennense ticks on horses and low seropositivity rate in these animals.Thus, rickettsia infection in humans transmitted by A. cajennense in these areas is unlikely because of very limited exposure of humans to ticks.And, in fact, there are no human BSF cases reported in this area.Sangioni et al. (2005) recommended surveys of horse sera as useful methods for BSF surveillance in areas where humans are exposed to A. cajennense ticks.However, due to low seropositivity rates and under the conditions present in our study, horses cannot be used as sentinel animals because of low exposure to Amblyomma spp.ticks and consequent lower risk to humans and other animals living in these areas.
The titers against Rickettsia spp.found in this study ranged from 64 to 512 with three samples reaching 512.Nine animals reacted to R. rickettsii showing titers that were four-fold higher than those seen for R. parkeri suggesting the circulation of Rickettsia spp.homologous to R. rickettsii in the population studied.
IFA is the gold standard test for rickettsioses in humans and animals (LASCOLA; RAOULT, 1997).However, cross-reactivity between Rickettsia species is common and therefore identification of Rickettsia species causing infection in BSF is difficult (HORTA et al., 2004;LABRUNA et al., 2004;PACHECO et al., 2007).
The present study provides important evidence that horses in farms in Northern Paraná may have been exposed to rickettsiae identical or closely related to R. rickettsii.Monitoring rickettsial infections in horses is a valuable action for preventing BSF in humans exposed to ticks.Further epidemiological studies examining other potential domestic and wild reservoirs for BSF are needed in this area and other areas in the State of Paraná.

Table 1 .
Toledo et al. (2010)and, Southern Paraná (9.33%), and lower than those found byHORTA et al. (2007) and LEMOS et al. (1996) (19 and 27.3%, respectively)in Southeast Brazil in BSF endemic areas.In contrast with our results,Toledo et al. (2010), while studying work horses in an urban area of Londrina, found a seroprevalence of 38.5%.This Seropositivity rates by indirect immunofluorescence assay (IFA) using two different Rickettsia antigens in horses from six farms in Northern Paraná, Brazil.

Table 2 .
Antibody titers by indirect immunofluorescence assay (IFA) using two different Rickettsia spp.antigens in horses from six farms in Northern Paraná, Brazil.