Changing climatic conditions and the northward expansion of ticks and pathogens are of immense importance to human, animal, and environmental health assessment and risk management. From 2014 through 2015, a wildlife rehabilitation center in south-central Arizona, US noted soft ticks (Argasidae) infesting 23 birds, including Cooper's Hawks (Accipiter cooperii), Gray Hawks (Buteo plagiatus), Harris's Hawks (Parabuteo unicinctus), Red-tailed Hawks (Buteo jamaicensis), Great Horned Owls (Bubo virginianus), Common Ravens (Corvus corax), and a Greater Roadrunner (Geococcyx californianus), during the late summer seasonal rainy seasons. The parasites numbered in the hundreds on individual birds. Infested birds were moribund, obtunded, or paralyzed on presentation, with no prior histories of illness or evidence of trauma. Tick and avian blood samples were collected for vector-borne pathogen analysis focusing on the molecular detection of Rickettsia and Borrelia species. Ticks were identified as the neotropical species of soft tick, Argas (Persicargas) giganteus; their occurrence in the southern US on raptors represented an expansion of host range. Pathogen testing of ticks showed that 41% of 54 ticks were infected with Rickettsia hoogstraalii and 23% of 26 ticks were infected with a relapsing fever Borrelia; both agents are associated with uncertain health consequences. Among the blood samples, one was infected with the same Borrelia spp.; this Red-tailed Hawk also had Borrelia-positive ticks. With supportive therapy, the majority of birds, 74% (17/23), recovered and were released or permanent residents. The management of soft tick–infested birds and mitigation of future disease risk will require additional characterizations of these poorly studied soft ticks and their associated pathogens.
INTRODUCTION
Northward range expansions of arthropod vectors are increasingly detected, often associated with the concomitant expansion of associated pathogens and disease risks. For example, neotropical migratory birds have been implicated in the importation of exotic neotropical ticks to the US (Hasle 2013; Mukherjee 2014) with conservative estimates of >19 million neotropical ticks imported each spring, including ticks infected with Rickettsia spp. of human and animal health concern (Cohen et al. 2015). Climate change is attributed to shifting geographic ranges of several tick species (Sonenshine 2018); for example, Lyme disease risk in eastern and central Canada is predicted to increase with climate change (Ogden et al. 2008b), as Ixodes scapularis ticks deposited by northbound migratory birds (Ogden et al. 2008a) may encounter suitable climates for their survival and spread.
External parasites are not uncommon on avian patients presented to wildlife care facilities. Although the accurate identification of the parasites is essential to properly prevent infestation, treat patients, and assess risk of release back into the wild, birds arriving to rehabilitation centers can also serve as sentinels for local parasite occurrence.
Many neotropical avian species pass through the desert southwestern US during migration, and also nest in the region. Geographic range expansions of neotropical avian species are increasingly documented by both researchers and citizen science entities (Rappole et al. 2011; eBird 2018). Although limited data are available regarding the parasites of birds in the desert southwest, pathogens have been documented from diverse wildlife species in the adjacent country of Mexico (Aguirre et al. 1992) and from ticks found within the desert southwest (Armed Forces Pest Management Board 2012; Manzano-Román 2012; Hasle 2013; Justice-Allen et al. 2016).
From 2009 through 2015 during the late summer rainy season, soft ticks (Argasidae) were noted on several birds presented to the Tucson Wildlife Center, a wildlife rehabilitation facility in south-central Arizona that admits 3,000–4,000 animals per year, of which three quarters are birds. During our study in 2014–15, we noted tick infestations on approximately 1% of the birds taken in during these years, and all the infested birds were sampled. Each infested bird showed all or a combination of moribundity, obtundity, and paralysis on presentation without prior history of illness or evidence of trauma. The appearance of a previously undetected tick species in a situation where diverse bird species showed severe clinical manifestations was cause of great concern not only for bird health and conservation, but also for unknown zoonotic risk to humans. Our objectives were to identify the species of tick and to test ticks and avian blood samples for selected tick-borne pathogens to detail the host–vector–pathogen interactions that may be important for improving bird health.
MATERIALS AND METHODS
The hosts
Samples collected from 2014 to 2015 were used for this study. Each animal underwent physical examination and was routinely checked for external parasites. Birds included resident, active rehabilitation, pending release, and newly admitted from the wild but not yet housed in the facility. Affected birds were immediately placed in oxygen-supported intensive care units. Fluid therapy with 2.5% dextrose/0.45% sodium chloride (Baxter Healthcare, Deerfield, Illinois, USA) was given twice a day at 10% body weight subcutaneously. Antibiotic therapy initially used was ceftazidime (ceftazidime for injection, Sandoz Inc., Princeton, New Jersey, USA) at 75 mg/kg intramuscularly twice daily for 7 d. As pathogen identification became available, antibiotic treatment was changed to doxycycline (Vibramycin Monohydrate [doxycycline monohydrate] for Oral Suspension, Pfizer Laboratories, New York, New York, USA) at 20 mg/kg twice a day orally for 7 d. Necropsy examinations were performed on all deceased birds with ticks. Histopathologic and biopsy specimens were submitted to Zoo/Exotic Pathology Services (Carmichael, California, USA). Patients received ivermectin (IVOMEC 1% Injection, Merial Limited, Duluth, Georgia, USA) at 0.2 mg/kg orally and as topical application directly on the ticks and pyrethrin powder (Zodiac® Flea & Tick Powder for Dogs, Puppies, Cats & Kittens, Wellmark International, Schaumburg, Illinois USA) on individual birds for prevention of reinfestation upon release to outdoor pens. Permethrin powder for premise use (Prozap Garden & Poultry Dust, Neogen Corporation, Lexington, Kentucky, USA) was applied to access points, to soil, and to permanent perches of outdoor enclosures. Organic material and perches were removed and replaced with nonorganic, nonporous materials.
Ticks were removed from birds using fine-tipped forceps, stored in 70% ethanol, and saved for further analysis. Clinical pathology samples included blood collected via medial tarsal vein for determination of packed cell volume and blood smears examined in house. Additionally, blood samples were submitted for clinical pathologic analysis to Avian and Exotic Clin Path Labs (Wilmington, Ohio, USA) and for tick-borne pathogen analysis to Texas A&M University (College Station, Texas, USA).
The vectors
Ticks were identified morphologically at Texas A&M University and vouchered at the US National Tick Collection (Georgia Southern University, Statesboro, Georgia, USA) and the Texas A&M University Insect Collection. Morphologic identifications were made using a reference key (Estrada-Peña et al. 2006) and supporting information of the descriptions of Argas (Persicargas) giganteus (Kohls and Clifford 1968) and Argas keiransi (Estrada-Peña et al. 2003). In the laboratory, ticks were identified molecularly to genus using PCR DNA sequencing methods as follows. From a random subset of ticks from different birds, DNA was extracted from individual ticks using the E.Z.N.A. tissue DNA kit (Omega Bio-Tek, Norcross, Georgia, USA). Each tick was sliced into quarters with a sterile scalpel blade and incubated overnight at 55 C. We extracted DNA the next day as per the manufacturer's instructions. We subjected DNA extracts to PCR of the 12S mitochondrial ribosomal DNA (rDNA), resulting in a 360-base pair (bp) product using the T1B and T2A primers (Beati and Keirans 2001). A subset of samples was also subjected to two additional PCRs, one to amplify the internal transcribed spacer 2 (ITS2) region resulting in a 1.2-kb product using the ITS2-7923-F and ITS2-7923-R primers (Zahler et al. 1995), and a second to amplify the 16S mitochondrial rDNA resulting in a 460-bp product using the 16Sþ1 and 16S–1 primers (Mangold et al. 1998).The 12S and ITS2 PCR reactions were carried out as previously described (Cohen et al. 2015) and the 16S PCR reaction was carried out as previously described (Mangold et al. 1998). All positive samples were treated with ExoSAP-IT (Affymetrix, Santa Clara, California, USA) and submitted for Sanger sequencing (Eton Biosciences, San Diego, California, USA). Sequences were compared to those in the GenBank database of the National Center for Biotechnology Information to facilitate identification. Tick sequences generated in this study were deposited to GenBank (MH717088-9 and MK078102).
The pathogens
Tick and blood samples were screened for Rickettsia and Borrelia species. The DNA from a 50–100-µL sample of whole blood was extracted in the same manner as for the ticks described above, but with a 30-min incubation time. Initial Rickettsia screen was done by amplification of a 617-bp region of the gltA gene using the primers RrCS 372 and RrCS 989 (Williamson et al. 2010). Reactions were performed in 15-µL volumes using 1.5 µL of extracted DNA with 0.67 µM each primer and FailSafe PreMix E buffer and enzyme (Epicentre Technologies Corp., Chicago, Illinois, USA). Confirmation of positive results was accomplished by amplifying an 856-bp region of the ompB gene with primers rOmpB 120-M59 and rOmpB 120-807 (Raoult et al 2001). Additionally, all the gltA-negative blood samples and a small number of gltA-negative tick samples were subjected to the ompB reaction. The PCR purification and bidirectional DNA sequencing for both genes were performed as above.
Samples were tested for Borrelia in an iterative process using up to five independent PCRs. First, samples were subjected to a nested PCR to amplify the 16S-23S rRNA intergenic spacer (IGS) region of Borrelia species (Bunikis et al. 2004). Samples that produced a positive result in the IGS assay were then subjected to four additional assays: a qPCR (qualitative or real-time PCR) with Lyme group– and relapsing fever group–specific probes to amplify the 16S rRNA gene of Borrelia (Tsao et al. 2004), and three assays to detect the glpQ gene of relapsing fever Borrelia (Bacon et al. 2005; Ullmann et al. 2005; Fomenko et al. 2011). Sanger sequencing was performed on all amplicons of the target size as described earlier. Pathogen sequences generated in this study were deposited to GenBank (MH717090-5).
RESULTS
The hosts
In 2014–15, large numbers of soft ticks were found on 23 individuals of seven avian species (Table 1) at the Tucson Wildlife Center, most during the rainy season months of August, September, and October. Of the 23 infested birds, ticks were noted at the time of intake on seven birds (one nestling, two fledglings, three immatures, and one adult), whereas ticks were presumably acquired on site for the remaining 16 birds (Table 1). The ticks were primarily found on the feet, legs, and leg axillae and were distributed across all age groups and sexes (Fig. 1A, B). Species included Cooper's Hawk (Accipiter cooperii, n=4), Gray Hawk (Buteo plagiatus, n=2), Harris's Hawk (Parabuteo unicinctus, n=3), Red-tailed Hawk (Buteo jamaicensis, n=6), Great Horned Owl (Bubo virginianus, n=5), Common Raven (Corvus corax, n=2), and Greater Roadrunner (Geococcyx californianus, n=1). Birds newly admitted with ticks were typically reported to have fallen from the nest or were found recumbent on the ground.
Table 1
Twenty-three birds of seven species are shown with basic data identifying their species, age category, the presence of Argas (Persicargas) giganteus larval ticks immediately on admission (not acquired at the facility) or on examination after being in the facility for treatment of other conditions or preparation for release (assumed to have been acquired on the premises), differential blood cell counts, if ticks collected from them were positive for Rickettsia hoogstraalii, and final disposition of each bird.a
Figure1
On physical examination, numerous larvae of Argas (Persicargas) giganteus ticks were discovered on the tarsometatarsal area of a nestling Great Horned Owl (Bubo virginianus), from Tucson, Arizona, USA, May 2015. This photograph illustrates (A) the dense spatial arrangement of the ticks and (B) the subcutaneous reaction with hemorrhage (black arrows). Numerous small ticks were firmly attached and remained so following direct acaricide application.
Once in rehabilitation, physical examination of birds showed all or a combination of peracute sternal recumbence, distal-to-proximal ascending paralysis, pale mucous membranes, dehydration, air hunger, and inability to swallow with normal level of consciousness if they were not obtunded (Table 1 and Fig. 2A, B). Birds showed diminished distal reflexes, diminished response to toe pinch, and lack of facial expression or attempt to move. Hundreds of small, firmly attached ticks were present on the birds (Fig. 3), with edema, ecchymoses and petechiae at attachment sites. Presumptive identification was that of an argasid tick of unknown species.
Figure2
A deceased Red-tailed Hawk (Buteo jamaicensis) from Tucson, Arizona, USA, was examined and photographed August 2014. (A) A small number of larval Argas (Persicargas) giganteus ticks are seen in typical locations on the ventrum and leg axillae. (B) The right lateral view of the same bird shows the pectoral area with head to the left, illustrating the large ecchymoses and hemorrhage at the attachment site of the ticks.
Figure3
The deceased Red-tailed Hawk (Buteo jamaicensis) from Tucson, Arizona, USA, August 2014, is shown in the same view as Figure 1, with the ventral skin reflected to demonstrate extensive subcutaneous with hemorrhage and edema associated with Argas (Persicargas) giganteus tick insertion. The penetration of the bite into the underlying musculature is also visible.
Complete blood counts and blood chemistry values across species in general demonstrated marked anemia, leukocytosis with heterophilia, and eosinophilia (Table 1). Leukocytosis was detected in 78% (18/23) of the birds, with 61% (14/23) greater than 21,000 cells per µL total white blood cell count (WBC); heterophilia in 30% (7/23; 17% (4/23) of the birds had absolute heterocyte counts ≥20,000 cells/µL), and eosinophilia in 57% (13/23). Thirteen percent (3/23) of birds had white blood cell within normal ranges and 7% (2/23) were not sampled. Thirteen birds were sampled for pathogen analysis, including 17% (4/23) of birds from which only ticks were sampled, 30% (7/23) from which only blood was sampled, and 7% (2/23) from which both ticks and blood were sampled (Table 2).
Table 2
The details of wild avian infestation with larval Argas (Persicargas) giganteus ticks and tick and avian blood infections with agents in the genus Rickettsia and Borrelia are presented. We assayed a total of 54 individual ticks for pathogen prevalence comprised of a random selection of 6–10 ticks per bird. Based on the gltA screening for Rickettsia, 22 of 54 (41%) ticks were positive, in which four birds were associated with positive ticks in which 30–100% of the individual ticks from these birds tested positive. Several avian tick and blood samples in our study were associated with PCR amplicons and DNA sequences with imperfect matches to known relapsing fever Borrelia species in the GenBank database.a
A total of 25% (6/24) of birds died or were euthanized. In these birds, gross findings included focal ecchymoses and petechiae on the skinned surface at the sites of tick attachments. No other significant lesions were found. Histopathologic examination in one Red-tailed Hawk of feathered skin with and without attached parasites, liver, intestines, ventriculus and proventriculus, pancreas, gall bladder, spleen, kidneys, heart, and lungs showed significant severe diffuse lymphoplasmacytic perivascular dermatitis and cellulitis with multifocal acute hemorrhagic necrotizing dermatitis at the site of tick attachments. Incidental findings were protozoan enteropathy and erosive ventriculitis, and the remaining tissues exhibited no significant lesions.
A total of 75% (18/24) of the admitted birds survived. Removal of the ticks was associated with recovery from paralysis; following other supportive care and treatments, the time ranged from 3 d to 7 d for recovery from complete recumbency to normal perching and eating. The packed cell volume rebounded to normal levels in 1–2 wk, and eosinophilia returned to normal levels in 2–4 wk. Leukocytosis generally tended to respond more rapidly following substitution of doxycycline treatment for ceftazidime. Of the 17 survivors, two birds were transferred to another facility, one bird was a resident, and, after obtaining normal blood parameters and being declared free of ticks for more than 2 wk, 57% (13/23) of birds were released (Table 1).
The vectors
The vast majority of ticks seen in birds were six-legged larval soft ticks. Additionally, an unidentified eight-legged life stage was found on some birds. The soft ticks were examined morphologically (Fig. 4) and identified as larval A. (P.) giganteus (Kohls and Clifford 1968). Morphologic characteristics for the genus Argas included a flattened body with dorsal and ventral surfaces equal in area, and sutural line present with an overall leathery integument. Species-level identification was based upon setal patterns on tarsi I and on shape, dimensions, and dentition of the hypostome, as well as their relatively large size (“giganteus”). Seventeen specimens were selected at random for 12S rRNA tick identification PCR; all amplified and yielded sequences of sufficient quality for approximately 330 nucleotides. In this assessed region, sequences were identical to each other, yet not represented in GenBank, with the closest matches being Argas persicus (65% query coverage; 88% identity; KJ133581) and Argas miniatus (64% query coverage; 87% identity; KC769590). A total of seven specimens for which we had 12S rRNA sequences was randomly selected for ITS-2 and 16S tick identification PCR; all amplified, and three amplicons were selected at random for sequencing. Sequencing of ITS-2 resulted in one tick failing the sequencing reaction; the other two produced forward sequences that were of sufficient quality for approximately the first 450 nucleotides. In this assessed region, sequences were identical to each other, yet they were not represented in GenBank, with the closest matches being Argas walkerae (46% query coverage; 87% identity; KJ133611) and A. persicus (22% query coverage; 97% identity; KJ133607). 16S sequencing resulted in all three samples producing forward sequences of approximately 405 nucleotides that were of sufficient quality. In this assessed region, sequences were identical to each other, yet they were not represented in GenBank, with the closest matches being A. keiransi (94% query coverage; 96% identity; KJ465101) and A. persicus (99% query coverage; 90% identity; GU451248).
The pathogens
From each bird (n=6) for which ticks were collected (Table 2), a range of six to 150 ticks were collected per bird. We assayed a total of 54 individual ticks for pathogen prevalence, comprising a random selection of six to 10 ticks per bird. Based on the gltA screening for Rickettsia, 22 of 54 (41%) ticks were positive, in which four birds were associated with positive ticks in which 30–100% of the individual ticks from these birds tested positive (Table 2). In all cases, ticks testing positive in the gltA screening assay were confirmed as positive in the ompB confirmatory assay. Likewise, the random selection of gltA-negative ticks were also negative in the ompB assay. All gltA sequences were of sufficient quality for a length of approximately 596 nucleotides. In this assessed region, sequences were identical to each other, with three taxa that had an equivalent level of sequence homology in GenBank: uncultured Rickettsia sp. clone (94% query coverage; 99% identity; KF360024), Rickettsia endosymbiont of Haemaphysalis sulcata (94% query coverage; 99% identity; DQ081187), and Rickettsia hoogstraalii isolate Rg-731 (94% query coverage; 99% identity; MF379282). All ompB sequences were of sufficient quality for a length of approximately 671 nucleotides. In this assessed region, sequences were identical to each other, with the closest matches in GenBank being R. hoogstraalii strain RCCE3 (97% query coverage; 99% identity; EF629536) and Rickettsia sp. scc31 (93% query coverage; 99% identity; DQ105802).
Of the nine avian blood samples tested for Rickettsia using the gltA screening assay, three yielded an amplicon of the appropriate size but sequencing reactions failed. All nine avian blood samples were also tested for Rickettsia using the ompB assay, of which six yielded amplicons of the appropriate size but sequencing reactions failed (Table 2).
We assayed a total of 26 individual ticks using the Borrelia IGS assay. Of these, six ticks (23%) removed from two birds yielded amplicons of approximately 800 bp. Resulting sequences were identical to each other, but with poor matches to sequences in GenBank: Borrelia anserina Es. (100% query coverage; 85% identity; CP013704), Borrelia anserina BA2 (100% query coverage; 85% identity; CP005829), and Borrelia sp. IM/19 (100% query coverage; 76% identity; DQ000285). Using the 16S qPCR assay, all six of these ticks hybridized with the relapsing fever probe with cycle threshold values ranging from 27.5 to 30.1 in the 45-cycle assay; two amplicons were attempted to be sequenced but both reactions failed. Of the three assays for glpQ gene, five of the six ticks amplified using the Fomenko et al. (2011) assay but with failed sequences, all six ticks failed PCR using the Bacon et al. (2005) assay, and five of the six ticks amplified using the Ullmann et al. (2005) assay. Of these five ticks, three were successfully sequenced and identical to each other with the following related taxa in GenBank: Borrelia anserina Es. (99% query coverage; 96% identity; CP013704), Borrelia anserina BA2 (99% query coverage; 96% identity; CP005829), and uncultured Borrelia sp. isolate Carios spiro-2 (99% query coverage; 91% identity; EF688576). All six individual Borrelia-positive ticks were also positive for Rickettsia (Table 2).
Of the nine avian blood samples tested for Borrelia using the IGS assay, 11% (1/9) yielded an amplicon of approximately 800 bp that produced a sequence that was identical to the sequences recovered from the ticks on the same birds reported above (Table 2). Using the 16S qPCR assay, this blood was positive for relapsing fever Borrelia with a cycle threshold value of 22.9 in the 45-cycle assay; the amplicon failed the sequence reaction. Of the three assays for the glpQ gene, this blood sample was negative on the Fomenko et al. (2011) assay yet positive on both of the others; yielding a failed sequence in the Bacon et al. (2005) assay and a sequence with the same related taxa in GenBank as reported above for the ticks in the Ullmann et al. (2005) assay. The single bird with a positive blood sample was a Red-tailed Hawk and this bird also had Borrelia-positive ticks attached with the same Borrelia sequence as in the blood.
DISCUSSION
Wildlife rehabilitators can play important roles in the front line of disease surveillance and sentinel pathogen detection. We document wild bird infestations with large numbers of the soft tick A. giganteus in which approximately one third of infested birds showed distal-to-proximal ascending paralysis and other infested birds were obtunded or showed air hunger and inability to swallow. Numbers of infested birds were small, totaling 23 raptors and related species over the 2-yr period of investigation at a rehabilitation center in Arizona, that admits 2,250–3,000 birds per year. Among the infested birds included individuals with attached ticks at the time of intake from the wild as well as resident birds that acquired ticks on site at the center. Ticks in general are rarely found on birds presented to this facility (well under 10 cases per year), and no infested bird showed paralysis prior to this investigation. In years 2009–13 records indicated similar ticks were noted on other birds, but data were not collected and no specimens of ticks were collected.
Globally, at least 10 species of argasid ticks are known to be involved with paralysis in birds, and even more species of argasid and ixodid ticks have been associated with paralysis in humans and other mammals (Mans et al. 2008). In this study, morbidity via direct injury, blood loss, and immunological impact to the patients was severe, in some cases mitigating their eventual release and potential survival in the wild. Argasid infestations in tropical climates are associated with extensive mortality and morbidity (Hoogstraal 1985). Direct forms of damage include tissue destruction caused by the tick mouth parts and by the local inflammatory reaction of host to tick saliva; loss of blood, which in massive infestations can cause acute anemia; and paralysis caused by tick salivary toxins (Luttrell et al. 1996). All these manifestations were documented here.
Treatment for tick paralysis includes immediate removal of ticks to stop further exposure to salivary toxins. We found A. giganteus difficult to remove manually and slow to die immediately from topical application of acaricides. Without accurate knowledge of the natural history of A. giganteus, the diagnosis, prevention, treatment, and release criteria for afflicted avian species are difficult to assess and define. Wildlife managers need to be advised how to treat suspect infested nests with appropriate control methods. Critical supportive care for the birds, standard doxycycline antibiotic treatment, and application of acaricides all contributed to the recovery and eventual release of most of the birds. Removal of organic perching, debris, and nesting material was likely essential to prevent expansion of the infestation. Follow-up clinical pathologic analysis and examinations need to meet criteria for release back into the wild.
Little is known about the natural history of A. giganteus. Its preferred habitat is presumed to be porous organic materials or crevices (Furman and Loomis 1984). It has been recorded from Arizona, California, New Mexico and Utah in the US, and from Sonora, Mexico, on small passerines (Kohls and Clifford 1968; Furman and Lewis 1984). Argas giganteus larvae attach firmly to the host in their hundreds and feed for prolonged periods through a number of days without retreat, similar to those of Argas brevipes (Kohls et al. 1961). It is postulated that A. giganteus could be brought northward from the neotropics by migratory hawks, and infest other species by nest sharing, nest building, and nesting behavior (Philips et al. 2002; Belozerov 2003; Butler et al. 2003). In the Sonoran Desert, nesting sites and materials for large raptorial species of birds are at a premium, and nest competition is well known (Stensrude 1965; La Porte 2017). Coopers' Hawks, Harris's Hawks, and Gray Hawks often build substantial nests; however, they will also pirate existing nests and steal nesting materials. Great Horned Owls use preexisting nests built by other species. These behaviors could contribute to dispersal of nidicolous and roost-inhabiting tick species, perhaps including A. giganteus. Ticks were found on newly admitted birds, birds being held in hospital cages who had not arrived with ticks, birds in flight aviaries ready for release that were uninfested on prior examination, and resident birds. Although it is possible that a few of the ticks were missed on initial examination, the ticks typically appeared in large numbers on the host and it is unlikely that a massive infestation would be missed at admission or on routine examination. It also appeared that the ticks were highly mobile and migrated to adjacent cages rapidly. Thus, it is difficult to determine if ticks emerged in situ, ex situ, or, most likely, both on newly admitted patients. Isolation and biosecurity measures of infested birds are warranted. Because A. giganteus was not represented with any DNA sequences in GenBank, the deposition of sequences from our study linked to the definitive morphologic identification of the species will facilitate future molecular-based identification of this tick species.
Although we cannot conclude that the Rickettsia and Borrelia detected in ticks or blood from the birds in this study contributed to the clinical status of the birds, the tick–host–pathogen associations and novel pathogen genetic data that we documented may be useful in future etiologic investigations. Based on DNA sequences from two different genes, we conclude that R. hoogstraalii or a closely related species infected several of the ticks removed from birds. Rickettsia hoogstraalii is a spotted fever group rickettsia with unknown pathogenicity originally isolated in 2006 from H. sulcata ticks in Croatia (Duh et al. 2010) and Carios capensis ticks in the US (Mattila et al. 2007). Rickettsia hoogstraalii infections occur in humans and wildlife (Vatansever 2018), with increasing detections in ticks removed from wild animals worldwide. Since its description, R. hoogstraalii has been detected in ticks from Italy, Spain, Japan, Ethiopia, Turkey, and the western Indian Ocean Islands (Kawabata et al. 2006; Marquez et al. 2008; Dietrich et al. 2014; Orkun et al. 2014; Chisu et al. 2017).
Several avian tick and blood samples in our study were associated with PCR amplicons and DNA sequences with imperfect matches to known relapsing fever Borrelia species in the GenBank database (<93% sequence homology). Tick-borne relapsing fever Borrelia spp. infect humans and animals worldwide; in the western US human infection is often linked to sleeping in rustic cabins or limestone caves in mountainous areas where Ornithodoros ticks are present, and the disease is characterized by recurring episodes of fever, headache, muscle and joint aches, and nausea (Dworkin et al. 2008). Relapsing fever remains a neglected disease, with little known about the ecology of disease as it relates to wild vertebrate reservoirs, tick species distributions, and zoonotic potential (Talagrand-Reboul et al. 2018). The presence of these agents in the birds at a wildlife care center should be considered as a zoonotic risk to humans and other patients, and adequate precautions and protocols should be taken. Introduction of new pathogens from exotic ticks of migrating birds is well documented and concern regarding zoonotic potential is well justified (Hamer et al. 2012; Maday 2018; Rainey et al. 2018).
The Sonoran Desert is supported by a decidedly tropical rainy season in the late summer and early autumn. With alterations in climate, including increased rainfall from El Nino Southern Oscillation events and stronger hurricanes over the Gulf of California, there may be increased northward expansion of neotropical flora and fauna, including tick vectors and vector-borne diseases (Estrada-Peña 2001; Gatewood et al. 2009; Randolph 2010). The Gray Hawks in this study are an example of such species that have expanded their breeding range northward in the last 10 yr (eBird 2018). Further investigations to characterize the natural history of A. giganteus, associated envenomation and pathogens, and options for tick control and bird treatment will aid in protecting avian health. Additionally, ecological tracking of changing patterns of avian migration and associated importation of ticks and pathogens is necessary to predict and mitigate future disease risk.
ACKNOWLEDGMENTS
Our study was made possible by the staff and volunteers at Tucson Wildlife Center and generous assistance and advice from Drury Reavill from Zoo/Exotic Pathology Services, Carmichael, California, US. The authors thank Lorenza Beati at the US National Tick Collection and Karen Wright at the Texas A&M University Insect Collection for archiving the specimens, and Mark Pilgard of the Centers for Disease Control and Prevention for assistance with Borrelia primers. We thank Gabriel Hamer for the photo of a larval tick as seen in Figure 4.
