Ticks, rickettsial and erlichial infection in small mammals from Atlantic forest remnants in northeastern Brazil

We evaluated infection by Rickettsia spp. and Ehrlichia spp in small mammals and their ticks from two Atlantic forest conservation areas in the state of Rio Grande do Norte, northeastern Brazil. A total of 39 small mammals were captured during 2012–2013, encompassing 33 marsupials (29 Didelphis albiventris, four Monodelphis domestica), three Cricetidae rodents (two Necromys lasiurus, one Rattus rattus), one Caviomorpha rodent (Thrichomys apereoides) and two armadillos (Euphractus sexcinctus). The ticks Amblyomma auricularium, Ixodes loricatus, and Ornithodoros mimon were collected from D. albiventris, whereas only A. auricularium was collected from armadillos. Through immunofluorescence assay with Rickettsia spp. antigens, 6/28 (21%) D. albiventris and the single R. rattus specimen reacted to at least one rickettsial antigen, with highest seroprevalence and endpoint titers to Rickettsia amblyommatis. A total of 150 ticks (126 A. auricularium, nine I. loricatus, 15 O. mimon) was tested for rickettsial infection by PCR, which detected only R. amblyommatis in most of the A. auricularium ticks. Lung and spleen samples were collected from small mammals (two N. lasiurus, six D. albiventris, three M. domestica, one T. apereoides, one R. rattus) and were tested by PCR for Anaplasmataceae agents. The spleen from one D. albiventris contained a new ehrlichial agent, here named as Ehrlichia sp. strain Natal. Phylogenetic analysis inferred from the dsb gene of Ehrlichia spp. indicates that this novel agent is potentially a new species. Future studies should monitor the possible role of rickettsial and/or ehrlichial microorganisms as agents of emerging diseases in these degraded areas of Atlantic forest, just as has occurred with other agents in degraded areas of this biome in southeastern Brazil.

The Atlantic forest biome of Brazil is currently reduced to less than 6% of its original pre-Colombian extent, and currently exhibits high levels of forest fragmentation (da Silva and Casteleti, 2003). Several recent studies have provided evidence that degradation of the Atlantic forest has enhanced tick infestation on birds or small mammals, and in some cases, increased the exposure of domestic animals and humans to pathogenic or potentially pathogenic rickettsiae (Ogrzewalska et al., 2011(Ogrzewalska et al., , 2012Dantas-Torres et al., 2012;Scinachi et al., 2017). In the case of small mammals, because of their limited dispersion and short lifespan, they can serve as good sentinels for the circulation of rickettsial agents in a particular area by natural environmental dispersion (Milagres et al., 2013). Besides, disorders in natural ecosystems can eventually bring humans into contact with wildlife-associated pathogens, resulting in the occurrence of emerging or re-emerging vectorborne diseases (Bradley and Altizer, 2007).
Most of the studies on tick-borne rickettsial organisms in Brazil have been done in the southern half of the country. In the northern half, studies have been concentrated in the Amazon biome, and a few ones in the semi-arid Caatinga biome. The scarcity of reports on tick-borne agents in the Atlantic forest remnants of northeastern Brazil prompted the current study, which evaluated infection by Rickettsia spp. and Ehrlichia spp. in small mammals and their ticks in two Atlantic forest conservation areas located in the state of Rio Grande do Norte, northeastern Brazil.

Ethical statements
Procedures of this study have been previously approved by the "Instituto Chico Mendes" (ICMBio -SISBio permit 32104 -2), "Instituto de Desenvolvimento Sustentável e Meio Ambiente" of Rio Grande do Norte (IDEMA-RN), and by the Ethics Committee on Animal Use of the Institute of Biomedical Sciences, University of São Paulo, protocol number 204/2013.

Study area
The city of Natal, state of Rio Grande do Norte, northeastern Brazil, has 10 Environment Protection Zones (EPZ). This study was performed in two of these protected zones (EPZ-1: Parque da Cidade Dom Nivaldo Monte -05°50′39.1″S 35°13′54.2″W; and EPZ-2: Parque Estadual das Dunas de Natal -05°49′30.5″S 35°11′35.6″W), which have an Atlantic forest matrix as original biome; yet the ecosystems within consist of dune formations covered mostly with salt marsh vegetation peculiar to the Coastal Tablelands, Atlantic forest, and scattered patches of Caatinga vegetation. The climate is tropical humid with average annual temperature of 26°C and annual rainfall of 2500 mm, with most intense rainy season between February and July (Freire, 1990;Ramalho and Pimenta, 2010).

Capture of small mammals and ticks
Two field campaigns were conducted to capture small mammals: one during the dry season (October 2012) and one at the beginning of the rainy season (February 2013). During two weeks (14 nights) per campaign, Sherman and Tomahawk-like traps, baited with a mixture of cornmeal, sardines and bananas were distributed alongside hiking passages within both parks, in sites where signs of animal activity was observed. A total of 40 traps per EPZ, distributed in four passages per park, were set at the first day. Traps were checked every morning, and baits were daily replaced.
Trapped animals were anesthetized with the association of xylazine (5 mg/kg) and ketamine (50 mg/kg), Afterwards, collection of blood was done by cardiac puncture, or from the tail or cephalic vein. Blood samples were allowed to clot at room temperature, and then centrifuged for separation of the serum, which was collected and kept frozen until serological analysis.
From 13 small mammals that were euthanized, lung and spleen fragments were collected for molecular analyses. The skins of the euthanized animals were deposited at the Museum of Natural History at the Pontifical Catholic University of Minas Gerais, Belo Horizonte City.
Every animal had the entire body examined for the presence of ticks, which were stored in absolute ethanol and brought to the laboratory. Morphological identification to species level of adult ticks of the genera Amblyomma and Ixodes followed Onofrio et al. (2006Onofrio et al. ( , 2009, whereas identification of Amblyomma nymphs followed Martins et al. (2010), and Ornithodoros larvae followed Kohls et al. (1969) and Barros-Battesti et al. (2013). Larvae of the genus Amblyomma were separated by morphotype and identified to species level by molecular analysis. For this purpose, Amblyomma larval DNA was tested by polymerase chain reaction (PCR) with primers 5′-CCG GTC TGA ACT CAG ATC AAG T-3′ and 5′-GCT CAA TGA TTT TTT AAA TTG CTG T-3′, which amplify a ≈460 bp of the tick mitochondrial 16S rRNA gene, as previously described (Mangold et al., 1998). PCR products were purified and sequenced in an automatic sequencer (model ABI 3500 Genetic Analyzer; Applied Biosystems/Thermo Fisher Scientific, Foster City, CA) according to the manufacturer's protocol. The generated sequences were submitted to BLAST analysis (www.ncbi.nlm.nih.gov/blast) to infer the closest similarities available in GenBank.

Serology for anti-Rickettsia spp. antibodies
The presence of anti-Rickettsia spp. IgG antibodies in the sera of the captured animals was assessed by immunofluorescence assay (IFA) using, simultaneously, crude antigens of six Rickettsia isolates from Brazil: R. bellii strain Mogi, R. amblyommatis strain Ac37, R. rhipicephali strain HJ5, R. rickettsii strain Taiaçu, R. parkeri strain At24, and R. felis strain Pedreira, as previously described (Labruna et al., 2007). Samples that reacted at the screening dilution (1:64) were then titrated using two-fold serial dilutions to determine the IgG endpoint titer. Slides were incubated with fluorescein isothiocyanate-labelled sheep anti-opossum IgG (CCZ, São Paulo, Brazil) for sera from marsupials, goat anti-rat IgG (Sigma, St Louis, MO, USA) for sera from Cricetidae rodents, and goat anti-guinea pig IgG (Sigma, St Louis, MO, USA) for sera from Caviomorpha rodents. In each slide, a serum previously shown to be nonreactive (negative control) and a known reactive serum (positive control) were tested at the 1:64 dilution. These sera derived from the studies of Horta et al. (2009) and Krawczak et al. (2016).

Molecular analyses of tick-borne bacteria
Ticks, and fragments of spleen and lung were submitted to DNA extraction by using the Wizard genomic DNA purification kit (Promega corporation, Madison, USA) following manufacturer's instructions. Adult ticks were tested individually; nymphs or larvae were processed in pools of three ticks from the same individual host. The concentration of extracted DNA was measured in a spectrophotometer UV (Bio Photometer plus, Eppendorf, Hamburg, Germany). Only samples with at least 20 ng/μl of DNA were subjected to PCR assays.
Tick DNA samples were tested by PCR using primers CS-78 (5′-GCA AGT ATC GGT GAG GAT GTA AT-3′) and CS-323 (5′-GCT TCC TTA AAA TTC AAT AAA TCA GGA T-3′), which amplify a 398-bp fragment of the citrate synthase gene (gltA) of all known Rickettsia species (Labruna et al., 2004). Samples yielding amplicon for this PCR assay were further tested by another PCR assay with primers Rr190.70F (5′-ATG GCG AAT ATT TCT CCA AAA-3′) and Rr190.701R (5′-GTT CCG TTA ATG GCA GCA TCT-3′), which amplify a 631-bp fragment of the 190-kDa outer membrane protein (ompA) of most of the spotted fever group Rickettsia species (Roux et al., 1996). In order to test the suitability of the DNA extraction protocol, tick samples with negative results for both rickettsial genes were further tested by the tick mitochondrial 16S rRNA gene PCR protocol described above.
The dsb partial sequence of an erlichial agent detected in this study was aligned with corresponding sequences of different Ehrlichia species available in GenBank, using the T-COFFEE 8.93 program (McWilliam et al., 2013). A phylogenetic tree was inferred by Bayesian method with Mrbayes_3.2.5 software with 1,000,000 generations; the tree being sampled every 1000 generations, running 4 times beginning with random starting trees. The Jukes-Cantor model was used combined with the models of gamma distribution (G) (Huelsenbeck and Ronquist, 2001). The first 20% of the trees represented burning, and the remaining trees were used to calculate Bayesian posterior probability (BPP).

Results
A total of 39 small mammals were captured during the study, encompassing 33 marsupials (29 Didelphis albiventris, four Monodelphis domestica), three Cricetidae rodents (two Necromys lasiurus, one Rattus rattus), one Caviomorpha rodent (Thrichomys apereoides) and two armadillos (Euphractus sexcinctus). A total of 221 ticks were collected from 16 animals of only two mammal species, namely D. albiventris and E. sexcinctus (Table 1); no ticks were found on the remaining four mammal species. Three tick species, Amblyomma auricularium, Ixodes loricatus, and Ornithodoros mimon were collected from D. albiventris, whereas only A. auricularium was collected from E. sexcinctus. Taxonomic identification of A. auricularium larvae relied on morphological comparisons with lab-reared larvae from our laboratory (data not shown) and by molecular comparisons of three 16S rRNA gene partial sequences from three larval pools, which were identical to each other and 100% identical to a GenBank sequence of A. auricularium from northeastern Brazil (KR869154).
A total of 150 ticks (126 A. auricularium, nine I. loricatus, 15 O. mimon) was tested for rickettsial infection by PCR; 42 ticks (32 adults and 10 nymphs) were tested individually, and 108 ticks (nymphs and larvae) were tested in pools of three ticks each. While no rickettsia was detected in I. loricatus or O. mimon, most of the A. auriculrium ticks contained rickettsia (Table 2). Partial sequences of the gltA gene generated from 14 of these PCR-positive ticks (seven adults, three nymphal pools and one individual nymph, and two larval pools) were 100% (350/350 bp) identical to R. amblyommatis strain AaPE (KJ534310). All gltA-PCR positive tick samples yielded amplicon by the ompA PCR; from these, DNA sequences were generated from six adults, which were also 100% identical to R. amblyommatis strain AaPE (KJ534312). All rickettsia-negative tick samples yielded amplicons by the tick mitochondrial 16S rRNA gene PCR, validating the DNA extraction protocol.
Lung and spleen samples were collected from 13 animals, namely two N. lasiurus, five D. albiventris, and three M. domestica from EPZ-1, and one T. apereoides, one R. rattus, and one D. albiventris from EPZ-2. Only the spleen from one D. albiventris from EPZ-1 yielded PCR Table 1 Ticks collected from opossums (Didelphis albiventris) and armadillos (Euphractus sexcinctus) in two Environmental Protection Zones (EPZ-1 and EPZ-2) of Natal City, state of Rio Grande do Norte, northeastern Brazil.   amplicons for the ehrlichial genes 16S rRNA and dsb. DNA sequencing of the 16S rRNA amplicon generated a sequence that by BLAST analysis was closest (99% identity; 324/327 bp) to several uncultured Ehrlichia spp. from Oceania and Asia (MF069159, KR063138, FJ966352). The dsb sequence was closest (81%; 268/331 bp) to E. chaffeensis from the United States (JQ085942). The ehrlichial agent detected in opossum in the present study was named Ehrlichia sp. strain Natal. Phylogenetic analysis inferred from dsb partial sequences indicated that Ehrlichia sp. strain Natal was distinct from all known ehrlichial agents, since it formed an isolate clade, sister to Ehrlichia ewingii (Fig. 1). With 100% posterior probability, Ehrlichia sp. strain Natal and E. ewingii grouped in a clade composed by different haplotypes of Ehrlichia ruminantium and multiple haplotypes of unnamed ehrlichial agents that have been reported in Brazil and Argentina. Novel DNA sequences generated in this study have been deposited in GenBank under accession numbers KY207547 and KY207546 for Ehrlichia sp. strain Natal 16S rRNA and dsb, respectively.

Discussion
The tick species A. auricularium, I. loricatus and O. mimon, have been previously reported in northeastern Brazil, where the same tick-host species associations found in the present study were recorded (Horta et al., 2011;Dantas-Torres et al., 2012;Saraiva et al., 2013;Labruna et al., 2014). In this study, most of the trapped small mammals were the white-eared opossum D. albiventris; very few rodents were trapped, in addition to the presence of an exotic species, the black rat R. rattus. These findings are compatible with Atlantic forest degradation, as previously reported (Bonvicino et al., 2002).
A high proportion of larvae, nymphs and adults of A. auricularium ticks collected from small mammals were found to harbor the spotted fever group agent R. amblyommatis. This finding is corroborated by a recent study that demonstrated 100% transovarial transmission and transstadial perpetuation of R. amblyommatis in A. auricularium ticks (Saraiva et al., 2013). Our serological analyses employing antigens of six Rickettsia species indicated highest seroprevalence and endpoint titers to R. amblyommatis, suggesting that at least part of the sampled small mammals have been infected by this agent, possibly via the parasitism by A. auricularium ticks. This statement is also corroborated by Saraiva et al. (2013), who demonstrated that this tick species was a competent vector of R. amblyommatis. Currently, R. amblyommatis is considered to be a potential human pathogen, based on serological evidence of human infection in the United States (Apperson et al., 2008;Vaughn et al., 2014). Noteworthy, one A. auricularium nymph was found attached to one of us (M.G.L.) during fieldwork (data not shown). Altogether, these results highlight the potential risk of human exposure to R. amblyommatis in the degraded Atlantic forest fragments of the present study.
A novel ehrlichial agent, Ehrlichia sp. strain Natal, was detected in an opossum D. albiventris. Phylogenetic analysis inferred from the highly polymorphic dsb gene of Ehrlichia spp. (Doyle et al., 2005) indicates that this novel agent is potentially a new species, yet to be formally described after in vitro isolation. All Ehrlichia species are known to be biologically transmitted by ixodid ticks (Dumler et al., 2001). Our acarological results incriminate two potential vectors of Ehrlichia sp. strain Natal, the ticks I. loricatus or/and A. auricularium. Further studies are needed to confirm this statement and the pathogenicity of Ehrlichia sp. strain Natal to animals and humans.
This work was performed in two environment protection zones of the Atlantic forest biome in northeastern Brazil, where human activities have almost completely destroyed or degraded the ecosystems during the last hundred years (da Silva and Casteleti, 2003). Even under such circumstances, the diversity of small mammals, their ticks, and potential tick-borne pathogens found in this study is noteworthy. Indeed, future studies should monitor the possible role of rickettsial and/or ehrlichial bacteria as agents of emerging diseases in these degraded areas of Atlantic forest, just as has occurred with other agents in degraded areas of this biome in southeastern Brazil (Ogrzewalska et al., 2012;Scinachi et al., 2017).

Conflicts of interest
The authors declare no conflicts of interest.