Abstract
Blood serum samples were collected from 451 bats captured within the São Paulo city from April 2007 to November 2008, and individually tested by indirect immunofluorescence assay against antigens derived from five Rickettsia species reported to occur in Brazil: the spotted fever group (SFG) species R. rickettsii, R. parkeri, R. amblyommii, R. rhipicephali, and the ancestral group species R. bellii. For this purpose, an anti-bat immunoglobulin G was produced and used in the present study. Overall, 8.6% (39/451), 9.5% (34/358), 7.8% (28/358), 1.1% (4/358), and 0% (0/358) serum samples were reactive to R. rickettsii, R. parkeri, R. amblyommii, R. rhipicephali, and R. bellii, respectively. Endpoint titers of reactive sera ranged from 64 to 256. From 20 bat species of 3 different families (Molossidae, Vespertilionidae, and Phyllostomidae), 46 animals were shown to be reactive to at least one rickettsial antigen. Seropositivity per bat species ranged from 0% to 33.3%. Most of the serologically positive sera reacted with two or more rickettsial antigens. Seropositivity for SFG rickettsial antigens in the absence of reactivity against R. bellii (ancestral group species) suggests that bats from São Paulo city can be infected by SFG rickettsiae. The possible role of soft ticks in serving as vectors of SFG rickettsiae to bats within the São Paulo city, associated to its public health risks, is discussed.
Introduction
The genus Rickettsia has recently been classified into four groups of bacteria: the spotted fever group (SFG), the transitional group, the typhus group, and the ancestral group (AG) (Gillespie et al. 2007). In Brazil, at least seven Rickettsia species have been reported, namely, the SFG R. rickettsii, R. parkeri, R. amblyommii, and R. rhipicephali, all associated with ticks; the transitional group R. felis, and the typhus group R. typhi, both associated with fleas; and the AG R. bellii, associated with ticks (Labruna 2009). R. rickettsii is a recognized human and canine pathogen causing Rocky Mountain spotted fever (RMSF) (Dumler and Walker 2005, Labruna et al. 2009). R. parkeri was also shown to be pathogenic for humans in the United States and possibly in Uruguay (Paddock 2005), but human infection in Brazil remains unreported (Labruna 2009). R. felis and R. typhi are also pathogenic for humans, causing disease less severe than RMSF (Parola et al. 2005a). The remaining Rickettsia species in Brazil have never been associated with human disease, although R. amblyommii was recently incriminated as the etiological agent of human febrile illness in the United States (Apperson et al. 2008).
Ticks that have been recognized as vectors of rickettsioses to humans or animals are within the Ixodidae family, composed of the hard ticks (Parola et al. 2005b). These ticks are common parasites of a number of vertebrate species, but seldom found feeding on bats (Guglielmone et al. 2003, Barros-Battesti et al. 2006). On the other hand, the Argasidae tick family, composed of the soft ticks, bears at least 10 species that feed primarily on bats in Brazil (Barros-Battesti et al. 2006). Some of these ticks (Carios spp.) are important human-biting species that feed on humans inside their homes. In this case, the tick population is maintained by bats living in the roof of human dwellings (Carvalho 1942). During the last few years, one of the authors (M.B.L.) received several complaints of human bites by bat ticks inside homes in suburban areas of the states of São Paulo and Minas Gerais. The ticks were identified as Carios sp. and were associated with bats living in the roof of the human dwellings.
Soft ticks have been poorly studied regarding tick-borne rickettsiosis, reflected by only a few reports of natural infection by rickettsiae in soft ticks in a few countries (Rehacek et al. 1977, Philip et al. 1983, Hoogstraal 1985, Loftis et al. 2005, Cutler et al. 2006). However, under laboratory conditions, soft ticks were excellent competent vectors and reservoirs of R. rickettsii (Davis 1943). Rickettsioses in bats have also been poorly investigated. To our knowledge, there has been only a single study that investigated natural rickettsial infection in bats (Reeves et al. 2006), in which sera from 56 Eptesicus fuscus (insectivorous bat) from Georgia were tested against R. conorii and R. rickettsii antigens. Only one (1.8%) bat was shown to be seropositive, with a serologic titer of 256 for both antigens. Earlier studies showed that under experimental conditions bats are susceptible to R. rickettsii (Steinhaus 1947, Tamsitt and Valdivieso 1970). In addition, the bat species Histiotus velatus (Vespertilionidae) and Carollia perspicillata (reported as Hemiderma perspicillatum; Phyllostomidae) were reported to be reservoirs of RMSF (Magalhães 1944); however, we are not aware of any specific data that support these statements.
Considering the real risks of human infestations by bat ticks, bats should be good sentinels for the presence of Rickettsia-infecting bat-associated ticks. Thus, the goal of our study was to perform a serological analysis for rickettsial infection in chiropterans captured between 2006 and 2008 in the city of São Paulo, Brazil.
Materials and Methods
Specimen collection
Bats were captured by the Zoonoses Control Center of São Paulo city from April 2007 to November 2008, within the São Paulo city. Bats were identified according to family, genus, and species using the identification key proposed by Vizotto and Taddei (1973). Bats received anesthesia (weight/volume). Subsequently, blood samples were collected by cardiac puncture and the animal was euthanized in a dioxide carbon chamber. This protocol has been previously approved by the Scientific Committee of Zoonoses Control Center of São Paulo city as part of official program of rabies surveillance in bats.
Indirect immunofluorescence assay
Bat sera were individually tested by indirect immunofluorescence assay (IFA) using crude antigens derived from five Rickettsia isolates from Brazilian ticks: R. rickettsii (strain Taiaçu), R. parkeri (strain At24), R. amblyommii (strain Ac37), R. rhipicephali (strain HJ5), and R. bellii (strain Mogi). Antigen preparation and IFA reactions were performed as previously described (Labruna et al. 2007), except that for the present study, we used an anti-bat conjugate produced as described below. Individual sera were initially screened at 1:64 dilution against each of the five rickettsial antigens. In case of a positive reaction, serial twofold serum dilutions were tested up to 1:1024.
Fluorescein-conjugated anti-bat antibody
Anti-bat conjugate was produced in the Immunology Section of the Zoonoses Control Center of São Paulo city according to previously described methods (Beutner et al. 1965, Camargo 1973, Hudson and Hay 1976). Briefly, a pool of serum from different bat species was precipitated with 18% sodium sulfate and dialyzed with a 0.9% sodium chloride solution. Immunoglobulins were obtained by ion-exchange chromatography in a diethylaminoethyl-celulose (DEAE)-cellulose solution (Sigma, St. Louis, MO), and protein fraction purity was checked by agar gel immunoelectrophoresis (Hudson and Hay 1976). Protein concentration was measured by biuret reaction (Gornall et al. 1949). A sheep was immunized with a series of 10 intramuscular inoculations of bat purified immunoglobulin (1 mg protein/mL) associated with Freund's complete adjuvant at 15-day intervals. The antiserum was analyzed quantitatively by radial immunodiffusion, and after identification of precipitating line at the 1:128 dilution, the animal was bled through jugular puncture. Then, serum was precipitated with an 18% sodium sulfate solution and purified using DEAE-cellulose resin, and the anti-bat immunoglobulin G serum was conjugated with fluorescein isothiocyanate in dialysis overnight. The excess of fluorochrome present in the serum was removed by filtration through Sephadex G-50 column. The conjugate was kept at −20°C until use.
The working conjugate was first tested in IFA with Toxoplasma gondii antigens, for which we had control sera. Once it was shown to work well, it was titrated and used for Rickettsia antigens. The first rickettsial slides were also used with canine positive and negative control sera, and the corresponding conjugate as previously described (Labruna et al. 2007), to certify the quality of the five rickettsial antigens. Once reactive bat sera were found, they were used as positive control in the remaining reactions.
Results
From 451 bats, serum samples of 358 bats were tested against the five rickettsial antigens, resulting in 9.5%, 9.5%, 7.8%, 1.1%, and 0% seropositivity for R. rickettsii, R. parkeri, R. amblyommii, R. rhipicephali, and R. bellii, respectively. Due to availability of antigens, another 93 serum samples were tested against only R. rickettsii, resulting in 5.4% positivity. Overall, 8.6% (39/451), 9.5% (34/358), 7.8% (28/358), 1.1% (4/358), and 0% (0/358) serum samples were reactive to R. rickettsii, R. parkeri, R. amblyommii, R. rhipicephali, and R. bellii, respectively. Endpoint titers for rickettsial antigens according to bat species are presented in Table 1. Frequency of seropositive bats according to bat species and families are shown in Table 2. From 20 bat species of 3 different families, 46 animals were shown to be reactive to at least one rickettsial antigen. Seropositivity per bat species ranged from 0% to 33.3%.
IFA, immunofluorescence assay; Neg., serum was nonreactive at the screening dilution 1:64; –, serum was not tested.
Discussion
IFA is currently the gold-standard test for serologic diagnosis of rickettsial infection in humans and animals (Brown et al. 1983, Raoult et al. 1986, La Scola and Raoult 1997). In the present study, we applied this serologic test to perform an indirect diagnosis of rickettsial infection in bats of São Paulo city. For this purpose, we used antigens of four SFG species and one AG species that have been isolated from Brazilian ticks. Seropositivity for SFG rickettsial antigens in the absence of reactivity against R. bellii (AG species) suggests that bats from São Paulo city can be infected by SFG rickettsiae. Since SFG species share numerous outer membrane and related proteins, serum cross reaction for two or more SFG species is expected (La Scola and Raoult 1997), as observed in the present study, in which most of the serologically positive sera reacted to two or more rickettsial antigens. For this reason, it was not possible to determine the SFG species that infected the bats.
The bat family Molossidae, the most numerous in the present study, and the family Vespertilionidae include insectivorous bats (Nowak 2003). These two families comprised 9 out of the 11 bat species with reagent sera. In São Paulo city, these bats live in groups of hundreds, or even larger groups, roosting body contact in closed spaces, mainly in the roof of buildings. Possibly, these shelters could be favorable for establishment of Argasidae ticks. Bats of the Phyllostomidae family have a remarkable variety of food habits, but the ones found to be seropositive in the present study are frugivorous (Nowak 2003). These bats roost in small groups in trees, where Argasidae ticks could also be established. Thus, it is possible that Argasidae ticks serve as vectors of SFG rickettsia to bat species in São Paulo city, although this statement deserves further investigation. Alternatively, other ectoparasites commonly found on bats, such as flies (Nycteribiidae, Streblidae), bugs (Polyctenidae), fleas (Ischnopsyllidae), or mites (Spinturnicidae, Macronyssidae, Trombiculidae) (Seneviratne et al. 2009), could be vectors of rickettsiae to bats.
Continued degradation and fragmentation of the natural habitat of bats has forced increased overlap of bats, domestic animals, and human coexistence, and this has created opportunities for bat-borne zoonotic diseases to emerge. The role of bats as carriers of Rickettsia is unknown. Regarding bat ticks (Argasidae), nothing is known about their association with Rickettsia under natural conditions in Brazil, and scarce information is available in other parts of the world. In contrast to a previous rickettsial serosurvey in bats in one site of the United States (Reeves et al. 2006), the present study showed that a significant number of bats living in the São Paulo city metropolitan area have been in contact with SFG rickettsiae. Further studies are needed to investigate the possible role of bat ticks as vectors of rickettsioses, especially in the urban areas where higher bat and human densities exist.
Footnotes
Acknowledgment
This work received financial support from Fundação de Amparo à Pesquisa do Estado de São Paulo (Grants 06/58210-7 and 06/60575-3).
Disclosure Statement
No competing financial interests exist.
