Detection of Rickettsia spp. in Animals and Ticks in Midwestern Brazil, Where Human Cases of Rickettsiosis Were Reported

Simple Summary Human cases of Brazilian spotted fever (BSF), diagnosed serologically by immunofluorescence assay, have recently been reported in the Goiás state, midwestern Brazil. Because serological cross-reactions among different rickettsial species that belong to the spotted fever group (SFG) are common, the agent responsible for the BSF cases in Goiás remains unknown. We evaluated the presence of anti-Rickettsia spp. antibodies in dogs, horses and capybaras (Hydrochoerus hydrochaeris), as well as rickettsial DNA in ticks collected from these animals and from the environment, in an area where BSF cases have been reported and two areas under surveillance in Goiás. The DNA of Rickettsia that did not belong to the SFG was detected in Amblyomma dubitatum, which was identified by DNA sequencing as Rickettsia bellii. Seroreactivity to SFG and R. bellii antigens was detected in dogs, horses and capybaras, with higher titers for R. bellii in dogs and capybaras. These data demonstrate the circulation of SFG rickettsiae in the region and the need for further research to definitively determine the agent responsible for rickettsiosis cases in this area. Abstract Brazilian spotted fever (BSF) is the most important tick-borne diseases affecting humans in Brazil. Cases of BSF have recently been reported in the Goiás state, midwestern Brazil. All cases have been confirmed by reference laboratories by seroconversion to Rickettsia rickettsii antigens. Because serological cross-reactions among different rickettsial species that belong to the spotted fever group (SFG) are common, the agent responsible for BSF cases in Goiás remains unknown. From March 2020 to April 2022, ticks and plasma were collected from dogs, horses and capybaras (Hydrochoerus hydrochaeris), and from the vegetation in an area where BSF cases have been reported and two areas under epidemiological surveillance in Goiás. Horses were infested by Amblyomma sculptum, Dermacentor nitens and Rhipicephalus microplus; dogs by Rhipicephalus sanguineus sensu lato (s.l.), Amblyomma ovale and A. sculptum, and capybaras by A. sculptum and Amblyomma dubitatum. Adults of A. sculptum, A. dubitatum, Amblyomma rotundatum and immature stages of A. sculptum and A. dubitatum, and Amblyomma spp. were collected from the vegetation. DNA of Rickettsia that did not belong to the SFG was detected in A. dubitatum, which was identified by DNA sequencing as Rickettsia bellii. Seroreactivity to SFG and Rickettsia bellii antigens was detected in 25.4% (42/165) of dogs, 22.7% (10/44) of horses and 41.2% (7/17) of capybaras, with higher titers for R. bellii in dogs and capybaras. The seropositivity of animals to SFG Rickettsia spp. antigens demonstrates the circulation of SFG rickettsiae in the region. Further research is needed to fully determine the agent responsible for rickettsiosis cases in this area.


Introduction
Among the rickettsiae circulating in Brazil, Rickettsia rickettsii and Rickettsia parkeri are the most important species from a public health perspective. R. rickettsii is the agent of Brazilian spotted fever (BSF), the most severe form of spotted fever and an important zoonosis with a high fatality rate in humans (≥50%), mainly in the southeast of the country [1][2][3]. R. parkeri is the causative agent of a milder spotted fever, which has not been related to fatal cases so far, but it is an emerging pathogen in different parts of the country [4][5][6][7][8]. Among other species circulating in Brazil, Rickettsia bellii has been detected in more than 25 tick species and is considered to be non-pathogenic for animals and humans [9,10]. However, it has been argued that R. bellii could play a role in the ecology and epidemiology of spotted fever group (SFG) rickettsiae by inhibiting their vertical transmission (transovarian transmission) to new generations of ticks [10,11].
Sampling domestic animals and capybaras has been useful for detecting the circulation of SFG rickettsiae in areas where Amblyomma sculptum, Amblyomma aureolatum and Amblyomma ovale are present [1,[12][13][14][15][16][17]. As an example, dogs and horses are hosts of tick vectors of SFG rickettsiae in Brazil and are also capable of producing antibodies against Rickettsia spp., which are detectable by indirect immunofluorescence assays (IFA). Thus, these animals are reliable sentinels to assess the epidemiological situation of BSF in a given area [12][13][14][15][16]18].
Capybaras also play a crucial role in the epidemiology of BSF. In addition to being sentinels, they act as amplifying hosts of R. rickettsii, infecting new populations of A. sculptum [19,20]. Although BSF cases have been confirmed in the state of Goiás (including in the Goiânia municipality), our current knowledge about the vectors, their respective hosts and the epidemiology of the disease in this region is still incipient [21,22]. Therefore, the present study aimed to investigate the presence of antibodies to Rickettsia spp. in dogs, horses and capybaras, in addition to detecting Rickettsia spp. DNA in ticks collected from these animals and from the environment in different areas of Goiânia, Goiás state, midwestern Brazil.

Study Area
This study was performed from March 2020 to April 2022 in the municipality of Goiânia, the capital of the Goiás state, which is located in midwestern Brazil. From an ecological viewpoint, 70% of the state's territory is part of the Cerrado biome, a tropical savanna ecoregion with the following two distinct seasons: the rainy season (from October to April) and the dry season (from May to September) [23].
Field activities were carried out in an area with a confirmed case of BSF (site A) and two areas (sites B and C) considered by the Superintendência de Vigilância em Saúde de Goiás (SUVISA-GO) as areas under epidemiological surveillance ( Figure 1).
The study area at each chosen site was delimited by tracing a radius of 3 km from the place where ticks were collected from the environment (Figure 1). Site A corresponds to the school farm of the Veterinary and Animal Science School (EVZ) of the Federal University of Goiás (UFG) and to the neighborhoods close to the EVZ (16 •  W, altitude 717 m); and site C corresponds to the peri-urban and rural areas close to the dam of the Ribeirão João Leite reservoir, which is located within the Altamiro de Moura Pacheco State Park (PEAMP), an integral protection conservation unit (16°34′09.2″ S, 49°12′51.2″ W, altitude 742 m).
This study was previously approved by the Chico Mendes Institute for biodiversity  and by the Institutional Animal Care and Use Committee (CEUA/UFG) of the Federal University of Goiás (protocol 092/19).

Animal Sampling
The sample size for dogs was calculated using an expected frequency of 10% for the detection of antibodies against SFG Rickettsia spp. [14,24]. Due to the lack of knowledge of the size of the population of dogs in the study area, an infinite population was considered for the sample calculation, with a margin of error of 5% and a confidence level of 95%, using the EpiInfo ® program. The resulting calculated sample size was at least 138 dogs.
For horses and capybaras, we used a convenience sampling strategy because some horse owners were reluctant to participate in the research and due to the inherent difficulties associated with capturing and handling capybaras. Capybaras were drawn into a trap in site A of approximately 90 m 2 , using corn, corn silage and banana leaves as bait. The baits were placed in the afternoon between 4:30 pm and 5:30 pm, and capybaras were captured during the night, with the aid of a net catcher. Once trapped and physically restrained with the net, capybaras were anesthetized with an intramuscular injection of ketamine (10 mg/kg) plus xylazine (0.5 mg/kg). Capybaras were identified with a subcutaneous microchip (Allflex), clinically monitored during the procedure until recovery from anesthesia, and released at the same capture site, as suggested by Neves et al. [25].

Blood Sampling and Tick Collection
Blood samples were collected in tubes with EDTA by venipuncture of the cephalic vein for dogs, jugular vein for horses and saphenous vein for capybaras. The samples were This study was previously approved by the Chico Mendes Institute for biodiversity  and by the Institutional Animal Care and Use Committee (CEUA/UFG) of the Federal University of Goiás (protocol 092/19).

Animal Sampling
The sample size for dogs was calculated using an expected frequency of 10% for the detection of antibodies against SFG Rickettsia spp. [14,24]. Due to the lack of knowledge of the size of the population of dogs in the study area, an infinite population was considered for the sample calculation, with a margin of error of 5% and a confidence level of 95%, using the EpiInfo ® program. The resulting calculated sample size was at least 138 dogs.
For horses and capybaras, we used a convenience sampling strategy because some horse owners were reluctant to participate in the research and due to the inherent difficulties associated with capturing and handling capybaras. Capybaras were drawn into a trap in site A of approximately 90 m 2 , using corn, corn silage and banana leaves as bait. The baits were placed in the afternoon between 4:30 p.m. and 5:30 p.m., and capybaras were captured during the night, with the aid of a net catcher. Once trapped and physically restrained with the net, capybaras were anesthetized with an intramuscular injection of ketamine (10 mg/kg) plus xylazine (0.5 mg/kg). Capybaras were identified with a subcutaneous microchip (Allflex), clinically monitored during the procedure until recovery from anesthesia, and released at the same capture site, as suggested by Neves et al. [25].

Blood Sampling and Tick Collection
Blood samples were collected in tubes with EDTA by venipuncture of the cephalic vein for dogs, jugular vein for horses and saphenous vein for capybaras. The samples were centrifuged at 5000× g for 10 min, and the separated plasma was kept at −20 • C until processing [14,26].
Each sampled animal was carefully inspected for the presence of ticks, for 3 min for dogs and capybaras, and 5 min for horses. During the morning, ticks were collected from the environment by flagging on the vegetation [27], covering an area of approximately 120 m 2 (30 × 40 m) on each site. For this, two white flannels were used, with a pair of collectors for each flannel. Flagging was carried out twice for 1:30 h in each study area and, during this period, flannels were examined every 5-10 m for the presence of ticks. Ticks were removed with the aid of toothless tweezers, placed in 15 mL conical tubes containing isopropyl alcohol and kept at room temperature until taxonomic identification in the laboratory.
Ticks were identified to the species level under a stereomicroscope using descriptions and taxonomic keys [28][29][30][31]. Because there is no taxonomic key for Brazilian Amblyomma larvae, they were identified to the genus level only [32].

Detection of Antibodies to Rickettsia spp.
Plasma from dogs, horses and capybaras were tested by indirect immunofluorescence assays (IFA) using crude antigens derived from the following four Rickettsia spp. isolates from Brazil: R. rickettsii (strain Pampulha) [33], R. parkeri (strain Atlantic rainforest) [6], R. bellii (strain Mogi) [34] and R. amblyommatis (strain Ac37) [35], as previously described by Labruna et al. [36]. Slides with crude antigens were produced as described by Horta et al. [15]. Plasma samples were tested individually using the methodology described by Horta et al. [15]. However, unlike Horta et al. [15], we used plasma instead of serum to detect IgG antibodies. Previous studies showed that there was no difference between serum and plasma samples for antibody detection [37,38]. Briefly, the plasma was diluted in two-fold increments with phosphate-buffered saline (PBS), starting from the 1:64 dilution. Slides were incubated with rabbit anti-dog IgG (Sigma, St Louis, MO, USA), rabbit antihorse IgG (Sigma, St Louis, MO, USA) and sheep anti-capybara IgG (CCZ, São Paulo, Brazil), coupled with fluorescein isothiocyanate at the 1:1000 dilution for dogs and horses, and at the 1:500 dilution for capybaras. Plasma samples reacting at the 1:64 dilution were titrated at two-fold increments to determine the endpoint titer to each of the four Rickettsia antigens. For plasma showing antibody titers to a given Rickettsia species at least four-fold higher than those observed for the other Rickettsia species, the reaction was considered to be possibly homologous to that Rickettsia species or to a closely related species [15,36]. In each slide, a serum previously shown to be non-reactive (negative control) and a known reactive serum (positive control), from the studies of Piranda et al. [39], Ueno et al. [40] and Ramírez-Hernández et al. [20], were tested at the 1:64 dilution.

Molecular Detection of Rickettsia spp. in Ticks and Blood Samples
Of the 5804 ticks collected in this study, a sample of 524 individuals (i.e., 377 A. sculptum (72, 98 and 207 from horses, capybaras and vegetation, respectively), 27 A. dubitatum (26 and 1 from capybaras and vegetation, respectively), 67 D. nitens from horses, and 53 R. sanguineus s.l. from dogs) were randomly selected and individually processed for DNA extraction using the guanidine isothiocyanate and phenol/chloroform technique [14] for adult ticks and the boiling extraction method for nymphs [41]. In addition to ticks, 17 blood samples collected from capybaras were also subjected to DNA extraction using the DNAeasy Blood and Tissue Kit (Qiagen, Valencia, CA, USA), following the manufacturer's recommendations. PCR with the blood of dogs and horses was not performed because previous studies showed the very low sensitivity of this technique for the detection of rickettsial DNA in the blood of dogs and horses that have been experimentally infected with R. rickettsii [39,40].
DNA from ticks and capybara blood samples were tested by a TaqMan real-time qPCR assay that targeted a 147-bp fragment of the rickettsial citrate synthase (gltA) gene [42][43][44]. The qPCR-positive samples were tested by the following two conventional PCR assays: one using primers CS-78 and CS-323 for the gltA gene [42], and another using primers Rr190.70p and Rr190.602n targeting the 190 kDa outer membrane protein (ompA) gene of SFG rickettsiae [45]. Negative samples were further tested using PCR protocols targeting the 16S rDNA gene of ticks [46] or the cytochrome b (cytB) gene of mammals [47] in order to validate the DNA extraction protocol. If a sample did not produce any product in these PCR assays, the sample was discarded from the study.
PCR products for the gltA, ompA, 16S rDNA and cytB genes were stained with SYBR Safe (Invitrogen, Carlsbad, CA, EUA), according to the manufacturer's recommendations, and visualized by electrophoresis in a 1.5% agarose gel using an ultraviolet transilluminator. The gltA PCR products were sequenced and the obtained sequences were subjected to BLAST analyses (www.ncbi.nlm.nih.gov/blast; accesed on 20 October 2022) to infer the closest similarities available in GenBank.

Data Analyses
The positivity to antibodies against Rickettsia spp. was compared for each animal species and sample sites using a chi-square or Fisher's exact test with Bonferroni correction. Differences were considered to be significant when p < 0.05 and all the analyses were performed in the program R (version 4.2.0), with packages stats and rcompanion. Ninety-five (95%) confidence intervals (CI) were calculated for prevalence data.  (Table 1). Of the 42 positive canine plasma, 20 (47.6%), 5 (11.9%), 4 (9.5%) and 1 (2.4%) showed possible homologous reactions to R. bellii, R. rickettsii, R. parkeri and R. amblyommatis, respectively. Regarding the horses, 40% (4/10) and 10% (1/10) of the seroreactive animals showed possible homologous reactions to R. rickettsii and R. bellii, respectively. Among the capybaras, 57.1% (4/7) of the reactive plasma showed possible homologous reactions to R. bellii (Table 1). No. = number of animals; PAIHR: possible antigen involved in homologous reaction; PSI: probable site of infection with confirmed cases of Brazilian spotted fever (BSF); RA = R. amblyommatis; RB = R. bellii; RP = R. parkeri; RR = R. rickettsii. * Seroreactivity values followed by different superscript letters represent a significant difference in the seroreactivity of the animals between the sampled areas (p < 0.05).
Although the IFA positivity for all animals showed different absolute values at site A (31.5%; 28/89), as compared to other sites (22.9% (16/70) and 22.4% (15/67) in site B and C, respectively), there was no significant difference (X 2 = 2.186; df = 2; p = 0.335). The positivity in horses was higher at site A as compared to other sites (Table 1), but statistically significant only between sites A and C (Fisher's exact test, OR = infinite; p = 0.045). The two-tailed p value between the sites A and B was 0.164 and between B and C, it was 1.000. Regarding the dogs, the positivity was similar across all sites and there was no statistical difference between them (X 2 = 0.116; df = 2; p = 0.944).

Tick Identification
The percentage of infested animals and the tick species found according to the study site are detailed in Table 2 A total of 3974 ticks were collected from the environment (one location for each of the sites A, B and C), of which 68.3% (2713) were Amblyomma spp. larvae, 31.6% were A. sculptum (214 adults and 1043 nymphs), 0.08% were A. dubitatum (two adults and one nymph) and one (0.02%) was a female of Amblyomma rotundatum ( Table 2).

Molecular Detection of Rickettsia spp. in Ticks
Out of 524 tested ticks (Table 3), only 0.4% (2/524) were positive for the rickettsial gltA gene upon qPCR testing. These two samples were from A. dubitatum, thus corresponding to a positivity of 7.7% in this tick species (2/26) (Table 3). However, none of the samples tested positive for the ompA gene. A PCR product was successfully sequenced from the one A. dubitatum sample and the corresponding gltA gene fragment (350 bp) was 100% identical to R. bellii. The rickettsial sequence generated in the present study was deposited in GenBank under the accession number OP718791. No. = number All 17 capybaras were negative for the presence of Rickettsia spp. All PCR negative results for Rickettsia spp. were confirmed by positive amplification of the 16S rDNA or cytB genes from these ticks or capybara blood samples, respectively.

Discussion
Our results confirmed the exposure of dogs, horses and capybaras to Rickettsia spp. with regard to dogs, the IFA positivity (range: 24-26.8%) was lower than that reported in BSF-endemic areas in the southeastern and southern regions of Brazil, which is around 66% [14,15,18]. In our study areas, we found dogs with antibodies to SFG rickettsiae and parasitized by ticks that can act as vectors of SFG rickettsiae to humans in Brazil [7, 48,49]. Nevertheless, in all three study sites, most of the dogs presented serological reactions with higher titers to R. bellii than to SFG rickettsiae. These data highlight the presence of R. bellii, although the circulation of SFG rickettsiae cannot be ruled out because the lower endpoint titers to SFG could be related to other rickettsial antigens not tested in this study [39]. Regardless, our serological results do not allow us to confirm the circulation of SFG rickettsiae in the study sites.
Our results agree with Neves et al. [50] who carried out a serological survey with dogs from the south and central regions of Goiás. Using the very same rickettsial antigens, they detected an overall positivity of 19%, with 73% of the seropositive dogs showing a homologous reaction to R. bellii. The highest frequency of homologous reactions to R. bellii among seropositive dogs in our study can be attributed to the fact that this rickettsia Animals 2023, 13, 1288 9 of 13 frequently infects different species of ticks in Brazil, including A. ovale [51,52], a tick species found on dogs in one of the study sites.
Regarding horses, we found that the IFA positivity to Rickettsia spp. was higher in the area with a BSF-confirmed case (site A) than in the area under surveillance for rickettsioses (site C). However, only three horses from site A showed possible homologous reactions to R. rickettsii and with low titers (128-256), which does not confirm the circulation of R. rickettsii in site A, as the titers were relatively low and there are no studies that detected this rickettsia in ticks in site A. Our results were lower than the rates of 77.3% and 44.5% reported by Horta et al. [15] and Souza et al. [53], respectively, in BSF-endemic areas in the southeast region of Brazil. However, in our study, A. sculptum was the second most common tick species (23.5%) in horses. These animals can sustain a high density of A. sculptum ticks [54] and, together with other factors such as the presence of capybaras and riparian forests, indicate that the sampled areas present factors that are favorable to an increased risk of R. rickettsii transmission and, consequently, the occurrence of BSF cases [13,55].
The positivity value of 41.2% detected in capybaras was lower than that found by Pacheco et al. [56] (74%) in non-endemic areas in the southeastern region of Brazil, where a greater number of capybaras with possible homologous reactions to R. bellii and with A. dubitatum predominating over A. sculptum were observed. In our study, 57.1% (4/7) of the capybaras showed a possible homologous reaction to R. bellii, suggesting that these animals have been exposed to this bacterium [17,56]. However, A. sculptum was the predominant tick species (83.6%) in capybaras compared to A. dubitatum (16.3%). It is likely that the high population density of capybaras and the presence of horses at the study site (site A) directly influenced the population size of this tick in the environment [57] and, consequently, on capybaras.
In the three sampled areas, in addition to the presence of horses and capybaras, other factors also contributed to A. sculptum being the predominant tick species in the vegetation. These factors include the greater adaptability of A. sculptum to anthropized areas [17] and the characteristics of the study sites, which presented a dense vegetation cover with the predominance of degraded pastures, grasses and shrubs. These characteristics provide an adequate microclimate for tick survival [1,58]. Similar to what was observed in our study, de Paula et al. [27] studied the seasonal dynamics of A. sculptum at site A and confirmed the predominance of A. sculptum over A. dubitatum in the environment. The latter species is more abundant in areas that are seasonally flooded [58][59][60], which is not the case of our study sites. Only one specimen of A. rotundatum was sampled during the collections in the vegetation (site C). This species has amphibians and reptiles as the main hosts [61] and, although its collection on vegetation has been previously reported, this is unusual due to the fact that A. rotundatum displays hunter behavior, thus making its collection difficult by vegetation collection using the cloth dragging technique [62].
Although we detected rickettsial DNA (not belonging to the SFG) in only 0.4% (2/524) of the ticks, we cannot exclude the possible presence of SFG rickettsiae in the study sites. Indeed, less than 1% of A. sculptum [49,63] and about 10% of A. ovale [16,64,65] ticks tested in previous studies have been found to be positive to R. rickettsii and R. parkeri, respectively, in areas where human rickettsiosis is endemic in Brazil. Furthermore, recent works have demonstrated the presence of R. amblyommatis and R. felis in A. sculptum and R. parkeri in A. ovale in Goiás [66,67], thus confirming the circulation of SFG rickettsiae in this state.
In Brazil, R. bellii has been identified in different populations of A. dubitatum [11,42,68], a finding corroborated by the detection of R. bellii DNA in A. dubitatum collected from one of the capybaras sampled in the current study. In a similar fashion, the high frequency of possible homologous reactions to R. bellii in capybaras tested herein further suggests the circulation of this bacterium in the study areas.
Luz et al. [17] demonstrated that the detection of antibodies to SFG rickettsiae in sentinel animals, a high environmental infestation with A. sculptum and the presence of capybaras (amplifying hosts of R. rickettsii) are important factors for BSF endemicity in Brazil. We reported these same factors in our study carried out in Goiânia, where 2 cases of rickettsioses were confirmed between 2012 and 2016 [22]. In Goiás, 16 cases of non-fatal rickettsioses have been confirmed so far [21]. Considering that mild cases of rickettsioses may be underreported, the official data must be far from the reality. Continuous surveillance and strengthening the diagnostic capabilities of public health laboratories may help to clarify the epidemiological situation of rickettsioses in Goiás and other risk areas in Brazil.

Conclusions
Our results confirmed the presence of antibodies to Rickettsia spp. in dogs, horses and capybaras. Most importantly, we demonstrated the exposure of dogs and horses to SFG rickettsiae. Although we only detected R. bellii infection in ticks, a known non-pathogenic agent for humans, this does not exclude the possible risk of rickettsiosis transmission and emphasizes the need for continuous surveillance in this area.

Informed Consent Statement: Not applicable.
Data Availability Statement: The data generated or analyzed during this study are included in this published article. The consensus sequence generated in this study is available in GenBank (OP718791).