All tick collections were performed during a 2-year period in three different localities: (a) in the Center for Wildlife Rescue and Triage (CETAS-IBAMA), (b) in the Center of Ophiology and Venom Animals of the Federal University of Bahia (NOAP/UFBA), both located in the municipality of Salvador, and (c) in the Laboratory of Venom Animals and Herpetology of the State University of Feira de Santana (LAPH/UEFS), located in the municipality of Feira de Santana. These three locations receive animals that were rescued by the Environmental Police or civilians in periurban rainforest fragments. Right after the rescue process and in the moment that the animals were entering the conservation centers, the animals were examined for tick infestations, and ectoparasites were collected. The ticks collected from wild specimens were stored in 70% ethanol until identification and DNA extraction. The Chico Mendes Institute of Biodiversity (ICMBio), from the Brazilian Ministry of Environmental Issues approved this study (SISBIO 52141-2). The animals were screened, identified, and put into quarantine in accordance with Brazilian law.

Tick identification was performed as previously described by Onofrio et al. (7) for adults and Martins et al. (14) for nymphs. Identifications were performed in double-blind assays at the Laboratory of Veterinary Parasitology (ICS/UFBA) and the Laboratory of Zoological Collections of the Butantan Institute. A stereoscope coupled to a digital camera was used to register the main structures used for the tick identifications. After taxonomic identification, the ticks were hermetically sealed into flasks containing 70% ethanol.

For DNA extraction from isolated organs, specifically, salivary glands and midgut, adult male and female ticks of sufficient size were dissected as described by Edward et al. (15) in a modified protocol. Briefly, the ticks were fixed with entomological pins in paraffin-filled Petri dishes. After fixing, the ticks were covered with a sodium phosphate buffer solution (pH 7.4). Under a stereomicroscope, the dissection procedure started with a lateral incision of the body, followed by separation of the ventral and dorsal parts, exposing the internal organs. The salivary glands were withdrawn before the midgut to avoid contamination. Subsequently, both organs were washed in phosphate buffer saline (PBS) pH 9.6 and preserved in 70% ethanol until molecular evaluation.

PCR assays using genus-specific primers were performed to investigate the presence of pathogens of the Rickettsia, Ehrlichia, Babesia, and Hepatozoon genera and the Anaplasmataceae family. DNA samples were obtained from tick salivary glands or midguts, or the whole tick if the tick were too small for dissection.

DNA extraction was performed using the PureLink™ Genomic DNA Mini kit (Invitrogen®) as described by the manufacturer. When a host animal had more than one tick, a pool of the samples (salivary gland, midgut, or whole tick) was used for DNA extraction. The PCR assays used genus-specific primers and PCR experimental conditions previously described: for identification of the Rickettsia genus, primers CS78-F (5′-GCAAGTATCGGTGAGGATGTAAT-3′) and CS323-R (5′-GCTTCCTTAAAATTCAATAAATCAGGAT-3′) were used to amplify a gltA gene fragment of 401 bp (16); for the Ehrlichia genus, primers DSB720-F (5′-ATTTTTAGRGATTTTCCAATACTTGG-3′) and DSB720-R (5′-CTATTTTACTTCTTAAAGTTGATAWATC-3′) were used to amplify a dsb gene fragment of 349 bp (17); for the Anaplasmataceae family, the forward primer was GE2 (5′-GTTAGTGGCAGACGGGTGAGT-3′) and the reverse primer was HE3 (5′-TATAGGTACCGTCATTATCTTCCCTAT-3′) that amplify a ribosomal 16S gene fragment of 360 bp (18); for the Babesia genus, primers BAB143-167-F (5′-CCGTGCTAATTGTAGGGCTAATACA-5′) and BAB694-667-R (5′-GCTTGAAACACTCTARTTTTCTCAAAG-3′) were used to amplify a ribosomal 18S gene fragment of 551 bp (19); finally, for the Hepatozoon genus, primers HEP142-169-F (5′-GGTAATTCTAGAGCTAATACATGAGC-3′) and HEP743-718-R (5′-ACAATAAAGTAAAAAACAYTTCAAAG-3′) were used to amplify a ribosomal 18S gene fragment of 574 bp (17, 19). A negative control using ultra-pure water and a positive control. Genomic DNA purified from the blood of dogs infected with Hepatozoon canis, Ehrlichia canis, Babesia canis, Anaplasma platys, as confirmed by nucleotide sequencing, and from a Rickettsia parkeri reference strain, were used as positive controls. PCR products were marked with SYBR™ Green Master Mix (ThermoFisher Scientific^©) and submitted to agarose gel (1.5%) electrophoresis to verify the presence or absence of the specific amplicons and to confirm or dismiss the presence of the pathogens in tick sample DNA.

In this study, 238 ticks were collected from 41 wild reptiles, amphibians, and mammalians rescued in rainforest fragments from Northeastern Brazil (Table 1). Regarding the life stage of all ticks identified, 85.7% were adults (204/238), represented by 87.7% (179/204) females and 12.3% (25/204) males, 13.0% (31/238) were nymphs and only 1.3% (3/238) were larvae (Table 2). Although for Amblyomma nodosum and Amblyomma varium both males and females were collected, there were no Amblyomma rotundatum males (Table 2). Regarding Ixodes loricatus only adults, male and female, were collected (Table 2).

The morphological identification of adults and nymphs (235/238) showed the presence of five species of ticks: A. rotundatum, A. nodosum, A. varium, I. loricatus, and Rhipicephalus sanguineus (Table 1). Three larvae (3/238) were found, and due to the absence of an identification key to classify larval species, the larvae were only classified by genus. One larva, found on Rhinella jimi, was identified as Amblyomma sp. and two larvae found on Didelphis albiventris as Ixodes sp. (Table 1). The morphological characteristics observed in each tick that allowed species identification are presented in Supplementary Figures 1–5.

The Amblyomma genus was the most abundant, representing 86.1% (205/238) of all ticks collected. A. rotundatum was the most prevalent tick species with a frequency of 72.3% (172/238) (Table 1). All ticks of this species were collected from either reptiles or the amphibian R. jimi. All reptiles rescued in this study were infested exclusively by A. rotundatum (Table 1).

Among snakes in our study, B. constrictor and Helicops carinicaudus had the highest infestation rate. The ticks obtained from these species contributed 83.7% of all A. rotundatum collected (144/172), with 41.9% (72/172) coming from each species of snake (Table 1). However, on B. constrictor, ticks were collected from 10 animals, while on H. carinicaudus all ticks were collected from one animal with a high degree of parasitism (Table 1). Additionally, A. rotundatum infested five additional snake species in this study: Eunectes murinus (11/172), Bothrops leucurus, (6/172), Python spp. (4/172), Waglerophis merremii (1/172), and Micrurus lemniscatus (1/172) (Table 1). Two more reptiles were infested by A. rotundatum, one Iguana iguana, and one Mesoclemmys tuberculata tortoise (Table 1).

The other tick species identified belonging to the Amblyomma genus were A. nodosum and A. varium with frequencies of 8.8% (21/238) and 4.6% (11/238), respectively (Table 1). Adults of A. nodosum and A. varium were found parasitizing mammals of the species Tamandua tetradactyla and Bradypus torquatus, respectively (Tables 1, 2).

Adults and nymphs of I. loricatus were found only in the mammalian hosts Didelphis aurita and Didelphis albiventris, accounting for 12.6% (30/238) of ticks collected. These two species of marsupials represent 50% of all infested mammals examined in this study (Table 1). The one specimen of R. sanguineus found in this study (0.4%; 1/238) was collected from Cavia aperea (Table 1).

A total of 36 DNA samples from A. rotundatum, A. nodosum, A. varium, and I. loricatus were obtained and evaluated for the presence of pathogens (Table 3). Our results showed no amplification for Babesia spp. or Ehrlichia spp. DNA in all samples tested. Rickettsia spp. were detected in three samples obtained from midgut or whole ticks. Of these three positives amplicons for Rickettsia spp., two originated from A. nodosum collected on T. tetradactyla, and one from A. varium collected on B. torquatus (Table 3). A. rotundatum and I. loricatus collected from reptile and amphibian hosts presented negative results for detection of Rickettsia spp. DNA.

DNA from the Anaplasmataceae family was identified in six (6/36) samples originating from ticks collected on mammalian, reptile, and amphibian hosts (Table 3). One of the positive samples originated from the midgut of A. varium collected from B. torquatus (1/36), four originated from midgut of A. rotundatum collected from various species [M. tuberculata (1/36), B. constrictor (1/36), H. carinicaudus (1/36), and R. jimi (1/36)], and one from the salivary glands of A. rotundatum collected on R. jimi (1/36). Hepatozoon spp. DNA was positively amplified in six of the 36 tick samples. Positive amplicons originated from both midgut and salivary glands DNA samples from A. rotundatum that had infested two B. constrictor snakes (4/36), from the midgut of one specimen collected on H. carinicaudus (1/36), and from salivary glands of another specimen collected on R. jimi (1/36) (Table 3). Co-infection of Hepatozoon spp. and the Anaplasmataceae family were found in midgut DNA samples from A. rotundatum collected on H. carinicaudus and from salivary glands of A. rotundatum infesting R. jimi.

DNA samples obtained from different organs, salivary glands and midgut, differed in amplification for pathogens in relationship to the species of tick. Pathogens were only identified in salivary glands from A. rotundatum. For all other ticks species, DNA samples only showed amplification of pathogen DNA in samples from the midgut or whole tick (Table 3).

Several studies have described ticks from the Amblyomma genus parasitizing reptiles, amphibians, avians, and mammalians (7, 20, 21). To date, 23 tick species have been identified in the Amazon region, with Amblyomma being the most prevalent genus (22). The state of Mato Grosso, located in the Midwestern region of Brazil, composed by the Amazon, Cerrado, and Pantanal biomes, showed a tick diversity of 27 species with Amblyomma once again being the most prevalent (23). Our results also found Amblyomma ticks to be the most prevalent, especially in reptile and amphibian hosts.

In the state of Pernambuco in the Northeast region of Brazil, A. rotundatum seems to be the most prevalent tick species infesting reptiles (24). Various reptile species and toads belonging to the Rhinella genus in the Amazon were infested by A. rotundatum (25). In the state of Mato Grosso, three species of Rhinella, R. bergi, R. marina, and R. schneideri were infested with adults and nymphs of A. rotundatum and Amblyomma dissimile, with A. dissimile being the most frequent tick species found (23). The amphibian R. jimi rescued in this study were parasitized with A. rotundatum and one larvae of the Amblyomma genus. It has been previously described that Rhinella are predominately infested by ticks from A. rotundatum and A. dissimile species (26). In Brazil, the infestation by A. rotundatum on R. jimi was previously described in the states of Pernambuco (13, 27) and Ceará (28).

Snakes are frequently found parasitized by ticks (13, 29–31). The snakes of our study were all infested with A. rotundatum ticks. A previous evaluation of tick parasitism in snakes from a fragment of the Atlantic Forest showed A. rotundatum was the most prevalent, representing 99.1% of all ticks collected (29). However, other tick species, most frequently A. dissimile, are found infesting different snake species (23, 25, 31). Species from the Boidae family represented 50.1% of reptile species rescued in Salvador, Bahia, between 2012 and 2014 with B. constrictor making up 40.2% of these rescued reptile species. Rescue of wildlife animals in urban and periurban areas could be associated with the expansion of cities and increased deforestation, as these actions reduce biodiversity and require surviving native species to adapt to other environments (32). This high incidence of B. constrictor in periurban environments may explain the high prevalence of this snake species in our study. Pontes et al. (29) evaluated nine specimens of Helicops carinicaudus and found no tick infestations. However, the one H. carinicaudus captured in our study had the highest infestation rate among all animals rescued, showing a different pattern of parasitism between H. carinicaudus and hard ticks than previously reported.

With the exception of B. constrictor and H. carinicaudus, in our study, A. rotundatum infested snakes at lower rates than previously reported. In a study conducted in Venezuela, six specimens of E. murinus were found infested with A. dissimile, but not A. rotundatum (33). In the Northern (34), Midwest (23), and Southern region (35) of Brazil, specimens of E. murinus were found infested by A. dissimile. The parasite-host relationship between B. leucurus and A. rotundatum has already been described in the state of Bahia (36). A. rotundatum was also found infesting B. leucurus and Bothrops erythromelas in the state of Ceará (28). The only record in Brazil of A. rotundatum infestation in the Python genus occurred in a Python molurus bivittatus (30). The authors characterized this Python species as an artificial host of A. rotundatum due to the exotic origin of this snake. Snakes from the Python genus are exotic species on the American continent (37). However, A. rotundatum has also been described to infest P. bivittatus in the state of Florida in the United States, where both the tick and snake are exotic species (38). In experimental conditions, A. rotundatum ticks were able to establish an artificial infestation in W. merremii (39), but, in natural environments, this interaction was not observed. Until now, Micrurus lemniscatus has not been described as a host snake species for A. rotundatum. Thus, this study describes a novel parasite-host relationship, although another species from this genus, Micrurus ibiboboca, has been shown to be parasitized by A. rotundatum in the Northeast region of Brazil (28, 30).

Although the majority of the reptiles examined in this study were snakes, the lizard, I. iguana, and tortoise, M. tuberculata, were also infested with A. rotundatum. Infestation in I. iguana by this species of tick has already been registered in the states of Pernambuco and Paraíba (24, 30). In contrast, in the Midwestern region of Brazil, 44 I. iguana specimens from either free-living or captivity niches were only infested by A. dissimile (23). Although A. rotundatum has already been described to infest Mesoclemmys vanderhaegei in the Midwestern region of Brazil (23), this is the first record of parasitism of this species of tick on M. tuberculata.

Adult forms of A. nodosum have been described in mammalian hosts, almost exclusively on anteaters, as on T. tetradactyla and M. tridactyla, while immature forms are frequently found on birds hosts (23, 40, 41). Infestation by A. nodosum in avians was previously described for different species (23, 42, 43). A. nodosum occurrence was described in Brazilian's Southeastern (44), Midwestern (40), Northern (25), and Northeastern region (30, 43). An infestation by A. nodosum was found on T. tetradactyla in the state of Ceará (45) and on M. tridactyla and T. tetradactyla in the state of Pernambuco (30).

Adult A. varium display high host specificity, found almost exclusively in sloths from the Bradypodidae family, such as Bradypus tridactylus, Bradypus variegatus, B. torquatus (46). A. varium infestation has also been described in the Choloepus hoffmanni and Choloepus didactylus sloth species (25, 46). Interestingly, in Colombia, A. varium was described to parasitize a domestic dog (47). The occurrence of A. varium in Brazil has been registered in the Southeastern, Northeastern, Midwestern, and Northern regions (22, 23, 43, 46, 48). In the Northeastern region, specimens of A. varium were found infesting B. torquatus in the state of Bahia (46) and B. variegatus in the state of Paraíba (30).

Our study found I. loricatus on D. aurita and D. albiventris as previously described in diverse Brazilian states such as Paraná, Rio Grande do Norte, São Paulo, and Pernambuco (21, 49–51). In the state of Pernambuco, I. loricatus was found infesting five marsupial species, including D. aurita and D. albiventris, with high infestation frequencies in these two marsupial species (50), similar to that observed in our study.

Rhipicephalus sanguineus mainly parasitizes domestic dogs and is considered an important vector for pathogens that have negative impacts on human and veterinary health (1). The majority of studies involving R. sanguineus have been in domestic animals, but this tick species is also found on wild animals (30, 52, 53). In Brazil, adults and nymphs of R. sanguineus have previously been found to parasitize black-tailed marmosets [Mico melanurus; (23)], and the carnivorous species Nasua nasua, Cerdocyon thous, Chrysocyon brachyurus, Pseudalopex vetulus, and Leopardus tigrinus (23, 30, 52, 54). However, the parasitism of R. sanguineus in C. aperea is described for the first time in this study.

Amblyomma rotundatum is described as a parthenogenetic tick species (55, 56) and the majority of the studies (13, 24, 30, 57) have only encountered female's ticks, as was observed in our study. However, four A. rotundatum males were reported in Brazil, in the Amazonian state of Rondônia (44, 58) and in the USA (59). Adults, both female and male, and nymph stages of A. varium (46) and A. nodosum (41) have been described to infest different mammalians. However, the life cycle and biology of these species is not well known in nature.

It is noteworthy to say that the stress level caused by the rescue, or even the particular situations that led the animals to be rescued, are important factors to be considered for the presence or not of these ectoparasites. This is a complex issue, and further studies made through an active collect of these animals in their own habitat would be valuable to verify in situ such host-parasite relationships.

The detection of pathogens in the tissues of ticks can provide relevant information about parasite-hosts relationships and their associated pathogens. In performing analysis of different organs (midgut and salivary glands), we aimed to provide information about the location of the parasite in the vector body. The midgut of ticks are the main organ involved in the acquisition of blood-borne pathogens, and the salivary glands have an important role in the transmission of parasites during blood uptake mediated by the tick bite (60). Diverse pathogens are described as being transmitted by saliva; for example, protobacteria belonging to the Anaplasmataceae family are mainly transmitted by hard ticks via saliva as ticks feed (61). Experiments have shown that tick saliva is capable of enhancing infection in vertebrate hosts for several pathogens (60), including Rickettsia conorii (62).

The Babesia genus comprises apicomplexa parasites that are naturally transmitted by ixodids (63). However, the literature associates Babesia transmission with ticks from at least four genera: Rhipicephalus, Ixodes, Haemaphysalis, and Hyalomma (64–66). In Brazil, ticks belonging to the Amblyomma and Ixodes genera collected from small mammals were negative for Babesia pathogens (67). Negative PCR amplification for Babesia DNA was observed in samples extracted from Amblyomma cajennense and Amblyomma ovale collected from dogs in Pantanal, Brazil (68). However, Babesia caballi was found infecting Amblyomma variegatum ticks isolated from cattle in the Republic of Guinea (69). Recently, one female tick of Amblyomma testudinarium collected from a dog in Taiwan was positive for Babesia gibsoni, but R. sanguineus and Heamaphysalis hystricis are the species most frequently infected by Babesia (70). Thus, ticks belonging to the Amblyomma genus may not be the main vectors of tick-borne pathogens from the Babesia genus.

Our samples showed positive PCR amplification for Rickettsia spp. only in DNA samples extracted from ticks collected from mammalian hosts. Rickettsia rickettsii and Rickettsia parkeri are causative species of tick-borne spotted fever in Brazil, and these pathogens were found in this country in ticks collected from capybaras in Mato Grosso state (71) and from dogs in Bahia state (72). Rickettsia bacteria have been reported in ticks infesting several wild mammalian species throughout Brazil (73–76), and in Brazilian wild birds (77, 78). Rickettsia parkeri and R. bellii have been found in A. nodosum collected from M. tridactyla and T. tetradactyla in the Southeast and Central-West regions of Brazil (76). One A. rotundatum collected in the Amazonian region of Brazil showed positive PCR results for R. belli (79). Although our PCR results showed no amplification of Rickettsia spp. DNA in samples collected from ticks infesting amphibians, a previous study of A. rotundatum collected from R. jimi from the Northeastern region in Brazil, detected Rickettsia in 100 % of tick DNA samples (27). For reptilian hosts, we again detected no PCR amplification for Rickettsia spp. DNA, but this hemopathogen has already was described in ticks collected from a tortoise, Chelonoidis denticulata, in the state of Espírito Santo in Southeastern Brazil (48). Rickettsia bellii was also found in A. rotundatum and A. dissimile ticks collected from snakes in several regions of Brazil (80).

For the first time, this study detected DNA from members of the Anaplasmataceae family in A. varium. Members of the Anaplasma and Ehrlichia genera are the causative agent of anaplasmosis or ehrlichiosis diseases in both domestic and wild animals and in humans (81). The Anaplasmataceae family was detected by PCR of Ornithodoros spheniscus ticks collected in Chile (82). However, the sequencing approach was unable to distinguish between Candidatus Neoehrlichia or organisms of the Anaplasma and Ehrlichia genera (82). In Northern Brazil, amplicons for the Anaplasma genus were detected in two A. dissimile collected from Bothrops atrox snakes (80). The genus Ehrlichia, belonging to the Anaplasmataceae family, are the most investigated pathogen of this family. However, in our results, although there was positive PCR amplification for the Anaplasmataceae family, there was no positive PCR amplification specifically for the Ehrlichia genus. The most studied species of this genus are Ehrlichia canis and Ehrlichia chaffeensis (83). In studies that, focused on wildlife, Ehrlichia spp. were found affecting terrestrial mammalian carnivores in a number of countries worldwide (84). Ehrlichia ruminantium was detected in Amblyomma ticks collected from tortoises in Kenya (85) and in the United States from animals imported from Zambia (86). In addition, DNA from E. canis was found in Amblyomma latum collected from monitor lizards in Kenya (85).

Six midgut and salivary glands from A. rotundatum samples garnered positive PCR result for Hepatozoon in our study. Hepatozoon spp. are the most common blood parasite found in reptiles (87). These hemoparasites affect reptiles on almost all continents of the world. However, description of Hepatozoon in ticks collected from reptiles are rare. Ticks belonging to the Amblyomma genus collected from snakes in Thailand, presented with a high prevalence of Hepatozoon infection (96%) (88). Amblyomma sp. ticks collected from a python in Australia were positive for Hepatozoon PCR amplicons (89). Hepatozoon sp. have only recently been found in Brazilian reptiles, specifically in two A. dissimile ticks collected from Bothrops atrox snakes in the Northern region of Brazil (80). In our work, we showed the presence of Hepatozoon sp. in ticks from two different snakes, B. constrictor and H. carinicaudus. The role of these ticks in transmission of Hepatozoon genus in snakes (80), and other reptiles was not determined. Experimental ingestion of rodent tissues infected with Hepatozoon ayorgbor by snakes successfully transmitted this parasite (90), suggesting that the feed of infected animals can be an important pathway in horizontal transmission of Hepatozoon sp.

Finally, we observed PCR amplification for Hepatozoon spp. and Anaplasmataceae family DNA in salivary glands in A. rotundatum, characterizing the first description of the co-infection of these hemopathogens in this tick species. The co-infection of different hemopathogens are expected once a host is infected by multiple pathogens (91). Descriptions of multiple tick-borne pathogens in the same tick specimen are increasing. Recently, in one Ornithodoros atacamensis tick from Chile, PCR detected three different pathogens from the Anaplasmataceae family and the Borrelia and Hepatozoon genera (82). In several Ixodes ricinus ticks, a predominant European tick species, co-infection of Anaplasma phagocytophilum and Rickettsia helvetica were found in both the salivary glands and midgut of males and females (92). In I. ricinus a co-infection with Borrelia burgdorferi, a causative agent of Lyme disease, and Babesia microtti has been described (93). In an investigation of pathogens in Amblyomma americanum, a tick species common in the United States, a co-infection of Rickettsia amblyommii and Ehrlichia chaffeensis in one tick and of R. amblyommii and Ehrlichia ewingii in two additional ticks was detected (94). In Brazil, an R. sanguineus tick collected from a dog was positive for E. canis and Leishmania infantum flagellate protozoa PCR amplicons (95). Our description of a co-infection with Hepatozoon spp. and Anaplasmataceae family pathogens in the midgut and salivary glands of A. rotundatum ticks collected from H. carinicaudus and R. jimi represents the first description of this double pathogen infection in the Brazilian territory.

In conclusion, our findings described known and new information about tick species and their respective hosts in Atlantic Forest fragments in Northeastern Brazil. The presence of several pathogen species in ectoparasite tissues provides new information that is important in human and veterinary medicine due to their zoonotic potential.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.