Rickettsial infections of dogs, horses and ticks in Juiz de Fora, southeastern Brazil, and isolation of Rickettsia rickettsii

Medical and Veterinary Entomology (2011) 25, 148 155 doi: 10.1111/j.1365-2915.2010.00915.x Rickettsial infections of dogs, horses and ticks in Juiz de Fora, southeastern Brazil, and isolation of Rickettsia rickettsii from Rhipicephalus sanguineus ticks R. C. P A C H E C O 1,J.MORAES-FILHO 2, E. GUEDES 3, I. SILVEIRA 2, L.J. RICHTZENHAIN 2, R.C. LEITE 3 and M. B. LABRUNA 2 1 Department of Basic Sciences and Animal Production, Faculty of Veterinary Medicine, Federal University of Mato Grosso, Cuiabá, MT, Brazil, 2 Department of Preventive Veterinary Medicine and Animal Health, Faculty of Veterinary Medicine, University of São Paulo, São Paulo, SP, Brazil and 3 Department of Preventive Veterinary Medicine, Veterinary School, Federal University of Minas Gerais, Belo Horizonte, MG, Brazil Abstract. The present study was performed in an area endemic for Brazilian spotted fever (BSF) in Juiz de Fora, state of Minas Gerais, Brazil, during the years 2007 and 2008, when fatal cases of BSF (caused by Rickettsia rickettsii) were reported. Adult ticks (Acari: Ixodidae) identified as Rhipicephalus sanguineus (Latreille) and Amblyomma cajennense (Fabricius) were collected from dogs and horses, respectively, and tested by polymerase chain reaction (PCR). Overall, 13.1% of the Rh. sanguineus ticks and none of the A. cajennense were found to be infected with R. rickettsii. Two isolates of R. rickettsii were successfully established in Vero cell culture from two Rh. sanguineus ticks. An indirect immunofluorescence assay (IFA) using R. rickettsii antigens detected blood serological reaction to R. rickettsii in 67.9% (53/78) of dogs and 41.0% (16/39) of horses living in the study area. Larval offspring from two Rh. sanguineus engorged females, naturally infected by R. rickettsii, were reared to adult stage in the laboratory. All active stages (larvae, nymphs, adults) remained 100% infected by R. rickettsii, which was efficiently transmitted to naïve rabbits. Overall, the results of the present study indicate a potential risk for transmission of R. rickettsii to humans by Rh. sanguineus, an occurrence yet to be documented in Brazil. Key words. Rhipicephalus sanguineus, Rickettsia rickettsii, dogs, horses, isolation in cell culture, serosurvey. Introduction Brazilian spotted fever (BSF) is an acute tick-borne disease caused by the bacterium Rickettsia rickettsii and is known as Rocky Mountain spotted fever (RMSF) in North America (Labruna, 2009). It has been classically reported to occur in the southeastern region of Brazil, which includes the states of Minas Gerais, São Paulo, Rio de Janeiro and Espírito Santo (Sexton et al., 1993; Rozental et al., 2002; Guedes et al., 2005; Nascimento et al., 2005). The main vector of BSF is the tick Amblyomma cajennense, which primarily parasitizes horses and capybaras in most of the BSF-endemic areas, as well as several other medium-sized animals such as opossums (Didelphis spp.) and domestic dogs (Sangioni et al., 2005; Perez et al., 2008). In these regions capybaras and opossums are believed to serve as amplifier hosts of R. rickettsii among tick populations (Horta et al., 2009; Souza et al., 2009). Amblyomma cajennense is also the most aggressive human-biting tick in southeastern Brazil (Guglielmone et al., 2006). The tick Amblyomma aureolatum (Pallas) is the main vector of BSF in the metropolitan area of São Paulo city, where the primary hosts of the tick adult stage are dogs (Guglielmone et al., 2003; Pinter & Labruna, 2006; Pinter et al., 2008). Correspondence: M. B. Labruna, Departamento de Medicina Veterinária Preventiva e Saúde Animal, Avenida Professor Orlando Marques de Paiva, 87, Cidade Universitária, São Paulo, SP 05508-270, Brazil. Tel.: +55 11 3091 1394; Fax: +55 11 3091 7928; E-mail: labruna@usp.br 148 Medical and Veterinary Entomology 2010 The Royal Entomological Society

Rickettsial infections in Brazil 149 The tick Rhipicephalus sanguineus is distributed throughout Brazil in most urban areas with large populations. It is a typical nidicolous tick that has adapted to live in human households and all active stages feed primarily on domestic dogs (Labruna & Pereira, 2001). Despite the proximity of Rh. sanguineus to humans, the tick has seldom been reported to bite humans in South America (Guglielmone et al., 2006); however, Rh. sanguineus is recognized as a vector of R. rickettsii to humans in Mexico and, more recently, the U.S.A. (Bustamante & Varela, 1947; Demma et al., 2005). Recently, Rh. sanguineus ticks were reported to contain R. rickettsii DNA in BSF-endemic areas of the states of São Paulo (Moraes-Filho et al., 2009) and Rio de Janeiro (Cunha et al., 2009; Gehrke et al., 2009). Piranda et al. (2010) showed that, under experimental conditions, domestic dogs can serve as amplifier hosts of R. rickettsii for Rh. sanguineus ticks. These findings indicate that dogs and Rh. sanguineus may play a role in the epidemiology of BSF in Brazil. The present study evaluated the rickettsial infection of dogs, horses and ticks collected in a BSF-endemic area in Brazil. The results showed high infection rates by R. rickettsii in Rh. sanguineus ticks, from which two isolates of R. rickettsii were obtained in cell culture. Materials and methods Study area and ticks The study was performed in an area of Juiz de Fora, Minas Gerais, Brazil (Fig. 1) that is endemic for BSF. From 1995 to 2008, 17 laboratory-confirmed cases of BSF were reported to the Juiz de Fora Health System, including five deaths (29% fatality ratio) (Juiz de Fora Municipal Heath Office, unpublished data, 2009). In April 2007, a total of 288 adult ticks were collected from 50 horses in two neighbourhoods (Nova Benfica and Barbosa Lage). These horses represented almost all of the entire horse population ( 60 horses) reared in these two localities. During July and August 2008, 122 adult ticks were collected from 250 free-ranging dogs captured by the Zoonosis Control Centre (CCZ), Juiz de Fora municipality (CCZ/Juiz de Fora), located in the Nova Benfica neighbourhood (Fig. 1). These dogs encompassed the entire canine population in the CCZ when ticks were collected. Ticks were identified by using taxonomic keys (Barros- Battesti et al., 2006) and selected for molecular testing for rickettsial DNA by haemolymph test or polymerase chain reaction (PCR) assays. Engorged female ticks collected from dogs were maintained at 27 C and 80% relative humidity (RH) until they had completed oviposition. Egg masses were subsequently tested by PCR. Haemolymph test Male ticks collected from dogs were individually processed using the haemolymph test as described (Burgdorfer, 1970). Briefly, a drop of haemolymph from each tick was dried on a glass slide and stained by the Giménez method (Giménez, 1964). Thereafter, ticks were frozen at 80 C. Isolation of rickettsiae Attempts to isolate Rickettsia in Vero cell cultures were performed using ticks containing Rickettsia-like structures according to the haemolymph test. For this purpose, frozen ticks were thawed and subjected to the shell vial technique as previously described (Kelly et al., 1991) with some modifications (Labruna et al., 2004a). Briefly, ticks were individually thawed in a water bath at 37 C and disinfected for 10 min in iodine alcohol, after which they were washed several times in sterile water. Washed ticks were triturated in 500 μl of brain heart infusion broth (BHI) and the homogenate was inoculated into shell vials containing a confluent monolayer of Vero cells. After inoculation, the shell vials were centrifuged for 1 h at 700 g at 22 C. The monolayer was washed once with minimal essential medium containing 5% bovine calf serum, and then incubated at 28 C with medium containing antibiotics (1% penicillin and 1% streptomycin). After 3 days, the medium was switched to antibiotic-free medium, and the aspirated medium was checked by Giménez staining for thepresenceofrickettsia-like organisms. If the result was positive, the monolayer of the shell vial was harvested and inoculated into a 25-cm 2 flask containing a monolayer of confluent uninfected Vero cells. Cells in the 25-cm 2 flask were checked by Giménez staining until >90% of them were infected, when they were harvested and inoculated into 150-cm 2 flasks of Vero cells. In all instances, inoculated Vero cells were incubated at 28 C. The level of infection of cells was monitored by Giménez staining of scrapped cells from the inoculated monolayer. A rickettsial isolate was considered to be established in the laboratory after at least three passages through 150-cm 2 Vero cell flasks, each achieving a proportion of infected cells >90% (Labruna et al., 2004a). Molecular tests DNA was extracted by the guanidine isothiocyanate phenol technique (Sangioni et al., 2005) from each frozen male tick (except for the ticks used for isolation attempts in cell culture) from pools of 20 eggs derived from each of the engorged females collected from dogs, and from individual ticks removed from horses. DNA of infected cell cultures was obtained by boiling (100 C for 10 min) as previously described (Eremeeva et al., 1993). DNA extracted from each tick and egg pools was tested by PCR using the primers CS-78 (forward) and CS-323 (reverse) (Table 1), which amplify a 401-bp fragment of the citrate synthase gene (glta) of possibly all Rickettsia species (Labruna et al., 2004a). For each set of reactions, negative (5 μl of water) and positive (5 μl of DNA extracted from Rickettsia parkeri-infected cells) controls were included. DNA from infected cell passages were tested by a battery of PCRs using all primer pairs listed in Table 1, targeting fragments of four

150 R. C. Pacheco et al. BRAZIL State of Minas Gerais N 21 30 00 S W E 3 1 2 5 4 S 21 40 00 S JUIZ DE FORA MUNICIPALITY 21 50 00 S k 22 00 00 S 43 40 00 W 43 30 00 W 43 20 00 W 43 10 00 W Fig. 1. Geographic location of the five localities of the present study in Juiz de Fora municipality, state of Minas Gerais, Brazil. 1, Nova Benfica; 2, Barbosa Lage; 3, Araújo; 4, Granjas Betânia; 5, Parque Guarani. rickettsial genes: glta; htra (17-kDa protein); ompa (190-kDa outer membrane protein), and ompb (135-kDa outer membrane protein). All PCR products of the expected size obtained in the PCRs described above were purified using ExoSAP-IT (USB Corp., Cleveland, OH, U.S.A.) and sequenced in an automatic sequencer (ABI Prism 310 Genetic Analyzer; Applied Biosystems, Inc., Foster City, CA, U.S.A.) according to the manufacturer s protocol. Partial sequences obtained were submitted to blast analysis (Altschul et al., 1990) to determine similarities to other Rickettsia species. A suspension of 3000 Rickettsia-infected Vero cells was inoculated intraperitoneally into a tick-naïve male guinea pig. The body temperature of the guinea pig was measured rectally

Rickettsial infections in Brazil 151 Table 1. Primer pairs used for amplification of rickettsial genes. Primer pairs Target genes and primers Nucleotide sequences (5 3 ) Reference and a serum known to be reactive were tested on each slide. Serum samples reacting at the screening dilution were tested in serial two-fold dilutions to determine the endpoint titre. glta 1 CS-78 GCAAGTATCGGT GAGGATGTAAT CS-323 GCTTCCTTAAAA TTCAATAAATC AGGAT 2 CS-239 GCTCTTCTCATCC TATGGCTATTAT CS-1069 CAGGGTCTTCGT GCATTTCTT htra 3 17k-5 GCTTTACAAAAT TCTAAAAAC CATATA 17k-3 TGTCTATCAATTC ACAACTTGCC ompb 4 120-M59 CCGCAGGGTTG GTAACTGC 120-807 CCTTTTAGATTAC CGCCTAA ompa 5 Rr190.70p ATGGCGAATATT TCTCCAAAA Rr190.602n AGTGCAGCATTC GCTCCCCCT Labruna et al. (2004a) Labruna et al. (2004a) Labruna et al. (2004b) Labruna et al. (2004b) Labruna et al. (2004a) Labruna et al. (2004a) Roux & Raoult (2000) Roux & Raoult (2000) Regnery et al. (1991) Regnery et al. (1991) Maintenance of Rickettsia-infected ticks in the laboratory Egg masses from two Rh. sanguineus female ticks that yielded Rickettsia-infected eggs by PCR, and one Rh. sanguineus female that yielded uninfected eggs by PCR, were held in an incubator at 25 C and 85% RH. Larval offspring from each of the three females were reared separately in the laboratory until they reached adult stage. For these purposes, each active stage (larvae, nymphs or adults) derived from each of the three females was allowed to feed on a ticknaïve rabbit, as previously described (Pinter et al., 2002; Horta et al., 2009). Engorged larvae and nymphs were held for moulting inside the incubator cited above. Samples of 10 unfed larvae, 10 unfed nymphs and eight unfed adult (four males, four females) offspring from each of the original three female ticks were submitted individually for DNA extraction followed by PCR targeting the rickettsial glta gene, as described above for field-collected ticks. All infested rabbits were tested by IFA on days 0 and 30 after infestation, using R. rickettsii antigens as described above. The body temperature of each rabbit was measured rectally each day. The present study was approved by the Bioethical Committee in Animal Research of the Faculty of Veterinary Medicine of the University of São Paulo. each day until 21 days post-inoculation, when the guinea pig was bled and its serum tested by immunofluorescence assay (IFA) against R. rickettsii antigens, as described below. Sera collection and indirect immunofluorescence assay During May and June 2007, blood sera samples were collected from 39 horses in two neighbourhoods (Nova Benfica and Barbosa Lage). In April 2008, 78 canine serum samples were collected in the four neighbourhoods of Granjas Betânia, Parque Guarani, Araújo and CCZ/Juiz de Fora (Fig. 1). Animals were randomly selected within the canine and horse populations of the areas and represented 25 70% of the animals available for sampling in each neighbourhood. Immunofluorescence assay was performed using R. rickettsii (Taiaçu strain) antigens as previously described (Zavala-Velasquez et al., 1996; Labruna et al., 2007). Sera were diluted in phosphate-buffered saline (PBS) and screened at a dilution of 1 : 64. Briefly, 10 μl of diluted sera were added to each well of the antigen slides. The slides were incubated, washed, then incubated with the species-corresponding fluorescein isothiocyanate-labelled conjugate anti-immunoglobulin G (IgG) (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, MD, U.S.A.) and washed again. The slides were mounted with Gel Mount (Biomeda Corp., Foster City, CA, U.S.A.) and read using an ultraviolet microscope (BX60; Olympus Corp., Tokyo, Japan) at 400 magnification. A serum previously identified as non-reactive Results All adult ticks collected from dogs were identified as Rh. sanguineus (22 males, 100 engorged females). All ticks collected from horses were identified as A. cajennense (84 males, 204 females). In total, 100 egg pools (each containing 20 eggs) derived from each of the 100 Rh. sanguineus engorged females, plus 20 Rh. sanguineus males (the other two Rh. sanguineus males were used for inoculating cell culture), and 288 A. cajennense adult ticks were tested by PCR targeting a fragment of the rickettsial glta gene. None of the A. cajennense ticks contained rickettsial DNA. By contrast, five(25%)of 20 Rh. sanguineus males and egg samples from nine (9%) of 100 Rh. sanguineus engorged females contained rickettsial DNA, which, after sequencing (350 nucleotides excluding the region corresponding to the primers), showed 100% identity with R. rickettsii in GenBank (CP000766). Two Rh. sanguineus males that contained Rickettsia-like organisms by the haemolymph test were further processed by the shell vial technique. Rickettsiae were successfully isolated and established in Vero cell culture from these two ticks. Rickettsial DNA amplified by PCR from the two isolates was sequenced. Identical segments of 1106, 708, 497 and 471 nucleotides of the glta, ompb, htra and ompa genes, respectively, were obtained from each isolate, and each showed 100% identity to the corresponding sequences of R. rickettsii in GenBank (accession nos. CP000766, CP000848, CP000766 and EU109179, respectively). These isolates, designated as strains Rs1 and Rs2, have

152 R. C. Pacheco et al. Table 2. Results of indirect immunofluorescence assay for antibodies to Rickettsia rickettsii in horse and canine sera from five localities in Juiz de Fora municipality, state of Minas Gerais, Brazil. Number of IFA reactive sera*/number of sampled animals (%) Localities Dogs Horses CCZ/Juiz de Fora 42/63 (66.7) (Nova Benfica) Nova Benfica 6/7 (85.7) Barbosa Lage 10/32 (31.2) Araújo 3/5 (60.0) Granjas Betânia 4/5 (80.0) Parque Guarani 4/5 (80.0) Total 53/78 (67.9) 16/39 (41.0) *IFA titres 64 for R. rickettsii antigen. IFA, immunofluorescence assay; CCZ, Zoonosis Control Centre. been deposited in the Rickettsial Collection of the Faculty of Veterinary Medicine of the University of São Paulo, where they are available upon request. A sample of infected Vero cells from one of the rickettsial isolates was inoculated intraperitoneally in a male guinea pig, which became febrile (>40 C) on day 4 post-inoculation, subsequently developed scrotal oedema and necrosis, and made a spontaneous recovery after 2 weeks. By day 21 post-inoculation, the guinea pig showed strong serological reaction against R. rickettsii antigens at the 1 : 1024 serum dilution (no further serum dilution was tested). Immunofluorescence assay revealed antibodies reactive with R. rickettsii (titres 64) in 16 (41.0%) horses and 53 (67.9%) dogs. Endpoint titres ranged from 64 to 16 384 in horses, and from 64 to 131 072 in dogs. At least five (12.8%) horses and 40 (51.3%) dogs showed endpoint titres >1024. Rickettsia rickettsii-reactive antibodies were detected in horses or dogs from all five neighbourhoods sampled in the present study, with seropositivity frequencies varying from 31.2% to 85.7% (Table 2). In the CCZ/Juiz de Fora, where R. rickettsii-infected ticks were detected, 66.7% of the dogs were seroreactive. From the nine Rh. sanguineus egg masses containing R. rickettsii DNA by PCR, the two largest were selected on the basis that they would produce nearly 100% viable larvae. The same procedure was performed to select one egg mass from the PCR-negative egg masses. Larval offspring from these three females were reared until the adult stage, using rabbits as hosts. Polymerase chain reaction performed on unfed larvae, nymphs and adults derived from each of the three females indicated that 100% of the ticks derived from infected egg masses remained infected by rickettsia through to adult stage. By contrast, no tick from the uninfected egg mass was found to be infected by rickettsia. In addition, all rabbits exposed to infected larvae, nymphs and adults became febrile and developed anti-r. rickettsii antibody titres ( 1 : 1024 serum dilution), whereas no rabbit exposed to the uninfected offspring became febrile or developed R. rickettsii reactive antibodies at the 1 : 64 serum dilution (Table 3). These results indicate the vector competence of larvae, nymphs and adults of Rh. sanguineus to transmit R. rickettsii to susceptible hosts. Discussion This study evaluated rickettsial infection in ticks and domestic animals from five neighbourhoods in the municipality of Juiz de Fora, where laboratory-confirmed cases of BSF were reported recently. All cases were associated with infestations by the tick A. cajennense, which has been incriminated as a vector of BSF in the area (Rodrigues et al., 2008). Results of the serological analysis of horses or dogs in the five neighbourhoods are similar to those of previous studies in other BSF-endemic areas of Brazil, in which horse seroprevalence to R. rickettsii was found to vary from 57.1% to 90.0% (Lemos et al., 1996; Sangioni et al., 2005), and canine seroprevalence ranged from 64.0% to 69.6% (Pinter et al., 2008; Moraes-Filho et al., 2009). In another study in Coronel Pacheco, a municipality neighbouring Juiz de Fora, Guedes et al. (2005) found 1.28% of A. cajennense free-living ticks to be infected by R. rickettsii. In the present study, none of the A. cajennense ticks collected from horses were shown to be infected by rickettsiae; however, at least 41% of these horses were seropositive for R. rickettsii with some indication of recent infection, as evidenced by high endpoint titres. As horses are primary hosts for the tick A. cajennense, they are sentinels for BSF within the Table 3. Rickettsial infection and vector competence of larvae, nymphs and adults of Rhipicephalus sanguineus derived from egg masses oviposited by three engorged females ( 1, 2, 3) collected from dogs in Juiz de Fora, Minas Gerais, Brazil. Egg masses from 1 and 2 wereshowntobe infected by Rickettsia rickettsii, whereas the egg mass from 3 was not. Data for female offspring Rabbits infested by larvae Rabbits infested by nymphs Rabbits infested by adults Females Unfed larvae* Fever IFA titres Unfed nymphs* Fever IFA titres Unfed adults* Fever IFA titres 1 10/10 (100) Yes <64 and 1024 10/10 (100) Yes <64 and 1024 8/8 (100) Yes <64 and 1024 2 10/10 (100) Yes <64 and 1024 10/10 (100) Yes <64 and 1024 8/8 (100) Yes <64 and 1024 3 0/10 (0) No <64 and <64 0/10 (0) No <64 and <64 0/8 (0) No <64 and <64 *Values presented as: number of positive ticks by polymerase chain reaction/number of ticks tested (% infection). Rectal temperature 40 C at 5 10 days post-infestation. IFA titres in sera collected on days 0 and 30 post-infestation (sera were tested from the 1 : 64 to the 1 : 1024 dilution). IFA, immunofluorescence assay.

Rickettsial infections in Brazil 153 distribution area of this tick in Brazil (Sangioni et al., 2005). It is very uncommon for Rh. sanguineus to parasitize horses in Brazil, where this tick is primarily associated with dogs within households (Labruna & Pereira, 2001). Thus, serological results in horses are indirect evidence that the A. cajennense population of Juiz de Fora is infected by spotted fever group rickettsiae. Reported infection rates by R. rickettsii in fieldcollected ticks have usually been very low (Burgdorfer, 1988; Guedes et al., 2005). Sangioni et al. (2005) did not find any infected A. cajennense ticks among 810 adult ticks sampled in three BSF-endemic areas in the state of São Paulo, where 80% of the horses were seropositive for R. rickettsii. These authors estimated that infection rates in these areas would be at most 0.8 1.43%. Thus, it is possible that Rickettsia-infected A. cajennense ticks would have been found in Juiz de Fora if a larger sample of ticks had been tested. In the U.S.A. and Brazil, infection rates by R. rickettsii reported among Dermacentor spp. (Acari: Ixodidae) and Amblyomma spp. populations from RMSF or BSF-endemic areas have varied from 0.05% to 1.3% (Burgdorfer, 1988; Guedes et al., 2005; Pinter & Labruna, 2006). Rates of Rh. sanguineus infection have been reported to be 2.9% (two of 70 ticks) in the U.S.A. (Demma et al., 2005) and 1.3% (two of 157 ticks) in Brazil (Moraes-Filho et al., 2009). Surprisingly, the present study found much higher infection rates by R. rickettsii among Rh. sanguineus ticks collected from dogs (13.1% among 122 samples of male ticks plus egg pools derived from engorged females). Seven of 22 male ticks were found to be infected; however, it is possible that some of these ticks were collected from rickettsaemic dogs, contributing to the high infection rates. At least 9% of the engorged female ticks were infected because these oviposited R. rickettsii-infected eggs. This appears to be the first reported demonstration of transovarial transmission of R. rickettsii by naturally infected Rh. sanguineus ticks. Previous studies of experimental infections of Rh. sanguineus with R. rickettsii (Parker et al., 1933; Piranda et al., 2010) demonstrated that transovarial transmission occurred in engorged females that had been infected as larvae or nymphs, but no transovarial transmission occurred when the engorged females were first exposed to R. rickettsii during adult feeding. Based on this statement, it can be inferred that the 9% of R. rickettsiiinfected egg masses found in this study were oviposited by females that had acquired the infection before they reached adult stage, possibly from other rickettsaemic dogs (primary hosts for all developmental stages of Rh. sanguineus in Brazil), which suggests the establishment of R. rickettsii in the studied Rh. sanguineus population. Piranda et al. (2010) demonstrated that dogs can serve as amplifier hosts of R. rickettsii for Rh. sanguineus ticks. However, only a minority of moulted ticks (<40%) remained infected upon feeding on rickettsaemic dogs and, because the transovarial passage of R. rickettsii was characterized by low filial infection rates (usually <50%), these authors suggested that Rh. sanguineus alone is not capable of maintaining R. rickettsii through successive tick generations. Conversely, in the present study a 100% filial infection rate of R. rickettsii was estimated for at least two engorged females collected from dogs (Table 3). By contrast with the experimental infection performed by Piranda et al. (2010), the present study evaluated naturally infected ticks; thus it provides reliable evidence that 100% filial infection rates can occur in Rh. sanguineus under natural conditions. In addition, differences observed between the present study and that by Piranda et al. (2010) may be related to the strain of R. rickettsii andalsotothe Rh. sanguineus population. The remarkably high infection rate (13.1%) by R. rickettsii in Rh. sanguineus ticks identified in the present study may be related to a particular situation found in the CCZ-Juiz de Fora. When collecting samples, this CCZ was shown to harbour dozens of dogs that sustained a well-established Rh. sanguineus population in the area, as evidenced by numerous free-living ticks of different developmental stages found in the canine facilities (data not shown). As the CCZ received new dogs captured in different parts of the city almost every month, it is possible that many of these dogs were susceptible to R. rickettsii. Therefore, once parasitized by R. rickettsiiinfected Rh. sanguineus ticks in the CCZ, these recently introduced dogs would have developed rickettsaemia and created new lineages of infected ticks. The continuous introduction of susceptible dogs results in a more abundant source of rickettsaemic dogs for creating new lineages of R. rickettsiiinfected ticks, thus maintaining the infected Rh. sanguineus population continuously at high infection rates. Recent studies showed that Brazilian isolates of R. rickettsii are also pathogenic for dogs (Piranda et al., 2008; Labruna et al., 2009), as are isolates from the U.S.A. (Comer, 1991). Thus, we would expect dogs to become ill or even to die from R. rickettsii infection in the CCZ because they receive no specific antibiotic treatment. Indeed, canine mortalities in the CCZ occurred during the course of the present study, but most were not properly documented and a few were clinically suspected to be caused by ehrlichiosis. Most of the dogs in the present study were shown to be serologically reactive to Ehrlichia canis antigens (data not shown), which suggests that these were infected by E. canis. However, as the canine clinical profile of R. rickettsii infection is undifferentiated from that of E. canis infection, it is possible that some of the illnesses observed in the CCZ dogs were caused by R. rickettsii. Finally, serological results from dogs and horses from different areas of Juiz de Fora indicate that a spotted fever group rickettsia is circulating in different neighbourhoods of the municipality, a condition corroborated by recent human cases of BSF from different areas of Juiz de Fora (Rodrigues et al., 2008). Although these human cases have been associated with infestations by Amblyomma ticks (Rodrigues et al., 2008), the results indicate a potential risk for transmission of R. rickettsii to humans by Rh. sanguineus, although this occurrence remains as yet unproven in Brazil. Acknowledgements We are indebted to Chris Paddock (Centers for Disease Control, Atlanta, GA, U.S.A.) for revising the manuscript. We thank the Centro de Controle de Zoonoses (CCZ) of Juiz de Fora, Minas Gerais, for help and logistical support in field work.

154 R. C. Pacheco et al. This work was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo [FAPESP (State of São Paulo Research Foundation)], Fundação de Amparo à Pesquisa do Estado de Minas Gerais [FAPEMIG (State of Minas Gerais Research Foundation)], Coordenação de Aperfeiçoamento de Pessoal de Nível Superior [CAPES (Agency for Support and Evaluation of Postgraduate Education)] and Conselho Nacional de Desenvolvimento Científico e Tecnológico [CNPq (National Council of Technological and Scientific Development)]. References Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. (1990) Basic local alignment search tool. Journal of Molecular Biology, 215, 403 410. Barros-Battesti, D.M., Arzua, M. & Bechara, G.H. (2006) Carrapatos de Importância Medico-Veterinária da Região Neotropical: um Guia Ilustrado para Identificação de Espécies. Butantan, São Paulo. Burgdorfer, W. (1970) The haemolymph test. American Journal of Tropical Medicine and Hygiene, 19, 1010 1014. Burgdorfer, W. (1988) Ecological and epidemiological considerations of Rocky Mountain spotted fever and scrub typhus. Biology of Rickettsial Diseases (ed. by D. H. Walker), pp. 33 50. CRC Inc., Boca Raton, FL. Bustamante, M.E. & Varela, G. (1947) Distribuicion de las rickettsiasis em Mexico. Revista del Instituto de Salubridad y Enfermedades Tropicales, 8, 3 14. Comer, K.M. (1991) Rocky Mountain spotted fever. Veterinary Clinics of North America: Small Animal Practice, 21, 27 48. Cunha, N.C., Fonseca, A.H., Rezende, J. et al. (2009) First identification of natural infection of Rickettsia rickettsii in the Rhipicephalus sanguineus tick, in the State of Rio de Janeiro. Pesquisa Veterinaria Brasileira, 29, 105 108. Demma, L.J., Traeger, M.S., Nicholson, W.L. et al. (2005) Rocky Mountain spotted fever from an unexpected tick vector in Arizona. New England Journal of Medicine, 353, 587 594. Eremeeva, M.E., Balayeva, N.M., Ignatovich, V.F. & Raoult, D. (1993) Proteinic and genomic identification of spotted fever group rickettsiae isolated in the former USSR. Journal of Clinical Microbiology, 10, 2625 2633. Gehrke, F.S., Gazeta, G.S., Souza, E.R., Ribeiro, A., Marrelli, M.T. & Schumaker, T.T. (2009) Rickettsia rickettsii, Rickettsia felis and Rickettsia sp. TwKM03 infecting Rhipicephalus sanguineus and Ctenocephalides felis collected from dogs in a Brazilian spotted fever focus in the State of Rio de Janeiro/Brazil. Clinical Microbiology and Infection, 15, 267 268. Giménez, D.F. (1964) Staining rickettsiae in yolk-sac cultures. Stain Technology, 39, 135 140. Guedes, E., Leite, R.C., Prata, M.C.A., Pacheco, R.C., Walker, D.H. & Labruna, M.B. (2005) Detection of Rickettsia rickettsii in the tick Amblyomma cajennense in a new Brazilian spotted fever-endemic area in the state of Minas Gerais. Memórias do Instituto Oswaldo Cruz, 100, 841 845. Guglielmone, A.A., Estrada-Peña, A., Mangold, A.J. et al. (2003) Amblyomma aureolatum (Pallas, 1772) and Amblyomma ovale Koch, 1844 (Acari: Ixodidae): hosts, distribution and 16S rdna sequences. Veterinary Parasitology, 113, 273 288. Guglielmone, A.A., Beati, L., Barros-Battesti, D.M. et al. (2006) Ticks (Ixodidae) on humans in South America. Experimental and Applied Acarology, 40, 83 100. Horta, M.C., Moraes-Filho, J., Casagrande, R.A. et al. (2009) Experimental infection of opossums Didelphis aurita by Rickettsia rickettsii and evaluation of the transmission of the infection to ticks Amblyomma cajennense. Vector-Borne and Zoonotic Diseases, 9, 109 118. Kelly, J.P., Raoult, D. & Mason, P.R. (1991) Isolation of spotted fever group rickettsiae from triturated ticks using a modification of the centrifugation vial technique. Transactions of the Royal Society of Tropical Medicine and Hygiene, 85, 397 398. Labruna, M.B. (2009) Ecology of Rickettsia in South America. Annals of the New York Academy of Sciences, 1166, 156 166. Labruna, M.B. & Pereira, M.C. (2001) Carrapatos em cães no Brasil. Revista Clínica Veterinária, 30, 24 32. Labruna, M.B., Whitworth, T., Bouyer, D.H. et al. (2004b) Rickettsia bellii and Rickettsia amblyommii in Amblyomma ticks from the state of Rondônia, Western Amazon, Brazil. Journal of Medical Entomology, 41, 1073 1081. Labruna, M.B., Whitworth, T., Horta, M.C. et al. (2004a) Rickettsia species infecting Amblyomma cooperi ticks from an area in the state of São Paulo, Brazil, where Brazilian spotted fever is endemic. Journal of Clinical Microbiology, 42, 90 98. Labruna, M.B., Horta, M.C., Aguiar, D.M., Cavalcante, G.T., Pinter, A., Gennari, S.M. & Camargo, L.M. (2007) Prevalence of Rickettsia infection in dogs from the urban and rural areas of Monte Negro municipality, western Amazon, Brazil. Vector-Borne and Zoonotic Diseases, 7, 249 255. Labruna, M.B., Kamakura, O., Moraes-Filho, J., Horta, M.C. & Pacheco, R.C. (2009) Rocky Mountain spotted fever in dogs, Brazil. Emerging Infectious Diseases, 15, 458 460. Lemos, E.R.S., Machado, J.R., Coura, J.R., Guimarães, M.A.A.M. & Chagas, N. (1996) Epidemiological aspects of Brazilian spotted fever: serological survey of dogs and horses in an endemic area in the state of São Paulo, Brazil. Revista do Instituto de Medicina Tropical de São Paulo, 38, 427 430. Moraes-Filho, J., Pinter, P., Pacheco, R.C. et al. (2009) New epidemiological data on Brazilian spotted fever in an endemic area of the state of São Paulo, Brazil. Vector-Borne and Zoonotic Diseases, 9, 73 78. Nascimento, E.M., Gehrke, F.S., Maldonado, R.A., Colombo, S., Silva, L.J. & Schumaker, T.T. (2005) Detection of Brazilian spotted fever infection by polymerase chain reaction in a patient from the state of São Paulo. Memórias do Instituto Oswaldo Cruz, 100, 277 279. Parker, R.R., Philip, C.B. & Jellison, W.L. (1933) Rocky Mountain spotted fever: potentialities of tick transmission in relation to geographical occurrence in the United States. American Journal of Tropical Medicine, 13, 341 378. Perez, C.A., Almeida, A.F., Almeida, A. et al. (2008) Carrapatos do gênero Amblyomma (Acari: Ixodidae) e suas relações com os hospedeiros em área endêmica para Febre Maculosa no estado de São Paulo. Revista Brasileira de Parasitologia Veterinária, 17, 210 217. Pinter, A. & Labruna, M.B. (2006) Isolation of Rickettsia rickettsii and Rickettsia bellii in cell culture from the tick Amblyomma aureolatum in Brazil. Annals of the New York Academy of Sciences, 1078, 523 529. Pinter, A., Labruna, M.B. & Faccini, J.L.H. (2002) The sex ratio of Amblyomma cajennense (Acari: Ixodidae) with notes on the male feeding period in the laboratory. Veterinary Parasitology, 105, 79 88. Pinter, A., Horta, M.C., Pacheco, R.C., Moraes-Filho, J. & Labruna, M.B. (2008) Serosurvey of Rickettsia spp. in dogs and humans from

Rickettsial infections in Brazil 155 an endemic area for Brazilian spotted fever in the state of São Paulo, Brazil. Cadernos de Saúde Pública, 24, 247 252. Piranda, E.M., Faccini, J.L., Pinter, A., Saito, T.B., Pacheco, R.C., Hagiwara, M.K. & Labruna, M.B. (2008) Experimental infection of dogs with a Brazilian strain of Rickettsia rickettsii: clinical and laboratory findings. Memórias do Instituto Oswaldo Cruz, 103, 696 701. Piranda, E.M., Faccini, J.L.H., Pinter, A., Pacheco, R.C., Cançado, P.H.D. & Labruna, M.B. (2010) Experimental infection of Rhipicephalus sanguineus ticks with the bacterium Rickettsia rickettsii, using experimentally infected dogs. Vector-Borne and Zoonotic Diseases, doi:10.1089/vbz.2009.0250. Regnery, R.L., Spruill, C.L. & Plikaytis, B.D. (1991) Genotypic identification of rickettsiae and estimation of intraspecies sequence divergence for portions of two rickettsial genes. Journal of Bacteriology, 173, 1576 1589. Rodrigues, A.F.S.F., Monteiro, C.M.O. & Clemente, M.A. (2008) Estudo epidemiológico da febre maculosa em Juiz de Fora, Minas Gerais. Revista Estação Científica, Juiz de Fora, 5, 1 9. Roux, V. & Raoult, D. (2000) Phylogenetic analysis of members of the genus Rickettsia using the gene encoding the outer membrane protein rompb (ompb). International Journal of Systematic and Evolutionary Microbiology, 50, 1449 1455. Rozental, T., Bustamante, M.C., Amorim, M., Serra-Freire, N.M. & Lemos, E.R. (2002) Evidence of spotted fever group rickettsiae in state of Rio de Janeiro, Brazil. Revista do Instituto de Medicina Tropical de São Paulo, 44, 155 158. Sangioni, L.A., Horta, M.C., Vianna, M.C.B. et al. (2005) Rickettsial infection in animals and Brazilian spotted fever endemicity. Emerging Infectious Diseases, 11, 265 270. Sexton, D.J., Muniz, M., Corey, G.R. et al. (1993) Brazilian spotted fever in Espírito Santo, Brazil: description of a focus of infection in a new endemic region. American Journal of Tropical Medicine and Hygiene, 49, 222 226. Souza, C.E., Moraes-Filho, J., Ogrzewalska, M. et al. (2009) Experimental infection of capybaras Hydrochoerus hydrochaeris by Rickettsia rickettsii and evaluation of the transmission of the infection to ticks Amblyomma cajennense. Veterinary Parasitology, 161, 116 121. Zavala-Velazquez, J.E., Yu, X.J. & Walker, D.H. (1996) Unrecognized spotted fever group rickettsiosis masquerading as dengue fever in Mexico. American Journal of Tropical Medicine and Hygiene, 55, 157 159. Accepted 19 August 2010 First published online 18 October 2010