Identification of Rickettsia spp. and Babesia conradae in Dermacentor spp. Collected from Dogs and Cats Across the United States

In the United States, Dermacentor variabilis and Dermacentor andersoni are considered key vectors for Rickettsia rickettsii, the causative agent of Rocky Mountain spotted fever. Through regional surveillance, a wide diversity of Rickettsia spp. have been documented in D. variabilis, and Dermacentor spp. has been suggested as potential vectors for various other pathogens, including Babesia spp. and Ehrlichia canis. To better define the prevalence and diversity of pathogens in Dermacentor spp. across the United States, 848 ticks collected from dogs and cats in 44/50 states in 2018–2019 were tested by PCR for Rickettsia spp.-specific 17 kDa and ompA gene fragments; a subset of Dermacentor spp. was also tested with PCR, targeting fragments of the 18S and large subunit region rRNA genes of Babesia spp. and 16S rRNA genes of E. canis. Rickettsia spp. was identified in 12.5% (106/848) of ticks. Species detected include Rickettsia montanensis (n = 64 ticks), Rickettsia bellii (n = 15 ticks), Rickettsia rhipicephali (n = 13 ticks), Rickettsia peacockii (n = 8 ticks), Rickettsia amblyommatis (n = 3 ticks), Rickettsia cooleyi (n = 1 tick), and unclassified Rickettsia spp. (n = 2 ticks). Ticks with R. montanensis and R. bellii were submitted from every U.S. region; R. rhipicephali was predominantly detected in ticks from the southern half of the United States, and all R. peacockii-positive ticks were D. andersoni that originated from the Rocky Mountain states. Ehrlichia canis was not detected in any Dermacentor spp., and Babesia conradae was detected in two Dermacentor albipictus. Because most ticks had fed on dogs or cats before submission, these findings do not implicate a given Dermacentor sp. as a primary vector of these agents, but in regard to Rickettsia spp., the data do support other published work showing D. variabilis harbors a diversity of Rickettsia species with unknown implications for animal and human health.


Introduction
T icks transmit more vector-borne disease agents in the United States than any other arthropod (Parola et al. 2005). Tick-borne rickettsial pathogens are widely distributed across this region and pose a significant health risk to both humans and animals as nontreated, and some treated, cases may end in death (CDC 2004, Parola et al. 2013, Levin et al. 2014, Biggs et al. 2016, Jiang et al. 2021). In the United States, Rocky Mountain spotted fever (RMSF) caused by primary vectors for R. rickettsii in the region. However, Rhipicephalus sanguineus sensu lato (s.l.) has been identified as a competent vector in the southwestern United States and Amblyomma americanum has also been implicated as a secondary vector (Demma et al. 2005, Breitschwerdt et al. 2011, Biggs et al. 2016, Levin et al. 2017. Reports and concern for spotted fever group Rickettsia (SFGR) other than R. rickettsii as a significant cause of human disease have grown over the last two decades (Delisle et al. 2016, CDC 2018. For instance, R. philipii (Rickettsia 364D) transmitted by Dermacentor occidentalis in California may cause fever, headache, or a maculopapular rash, while Rickettsia parkeri, transmitted by Amblyomma maculatum across its southeastern range, can cause similar signs in humans (Parola et al. 2013, Paddock et al. 2018, Yoshimizu and Billeter 2018. In addition, a human case report describes rickettsiosis following a D. variabilis bite caused by R. montanensis, and recently, a novel Rickettsia sp. was identified as the cause of fever, lethargy, and thrombocytopenia in three dogs from south central U.S. states (McQuiston et al. 2012, Wilson et al. 2020. As many as 60% of dogs in disease-endemic areas may have evidence of exposure to SFGR (Levin et al. 2014). However, because of the cross-reactivity of various SFGR on routinely used Rickettsia indirect fluorescent antibody (IFA) assays, novel pathogens such as the one recently isolated from dogs, as well as other SFGR, likely go unrecognized (Parola et al. 2005, Wilson et al. 2020. Investigators in several U.S. regions have surveyed D. variabilis for Rickettsia spp. and demonstrated this tick carries a variety of SFGR, and, of particular importance, that R. rickettsii is rarely detected in D. variabilis, which is significant, given this tick is currently considered a primary vector for this pathogen (Table 1).
In addition to SFGR, Dermacentor spp. has been implicated as vectors for other pathogens in the United States. Experimentally, D. variabilis can transmit Ehrlichia canis to dogs, causing ehrlichiosis, a potentially life-threatening illness with some dogs showing no sign of infection, while others display evidence of fever, lethargy, anorexia, myalgia, lymphadenopathy, bleeding tendencies, and neurologic abnormalities ( Johnson et al. 1998, Little 2010. Also, a rapidly emerging and medically relevant pathogen in the western United States and Canada, Babesia duncani, has caused serious disease in immunocompromised humans, and preliminary research suggests Dermacentor albipictus as one of the tick vectors (Swei et al. 2019, Yang et al. 2021. The winter tick, D. albipictus, is occasionally found on dogs and cats, although it is maintained in nature by wild ungulates (Duncan et al. 2020).
Various Dermacentor spp. is encountered questing in the environment across its range and is commonly collected from dogs and cats (Thomas et al. 2016, Little et al. 2018, Lehane et al. 2020. In a survey of ticks infesting dogs and cats across the United States, D. variabilis was the tick species most commonly infesting dogs (Saleh et al. 2019). Recently, D. variabilis was shown to be the predominant Dermacentor spp. infesting pets in the United States, including in the Rocky Mountain states, where this tick was historically considered rare or absent (Duncan et al. 2021). With the potential to spread pathogenic agents coupled with their changing geographic distribution, Dermacentor spp. surveillance and pathogen screening remain warranted (Minigan et al. 2018, Sonenshine 2018, Lehane et al. 2020, Duncan et al. 2021). In addition, pets may be considered sentinels for both ticks and tick-borne pathogens, and those with evidence of disease may indicate a risk to other members of the household (Paddock et al. 2002, Levin et al. 2014. Hence, the objective of this study is to describe the diversity and geographic distribution of potential pathogenic organisms in Dermacentor species ticks removed from dogs and cats across the United States.

Tick samples
Dermacentor spp. (n = 848; 544 females, 297 males, and 7 nymphs) from 471 dogs (n = 770 ticks) and 56 cats (n = 78 ticks) from 44/50 states used in this study was obtained and identified through an ongoing national survey of ticks on dogs and cats as previously described (Saleh et al. 2019, Duncan et al. 2021. Ticks were collected in private veterinary practices during the course of normal physical examinations; institutional review board approval was not required. Ticks were selected for Rickettsia spp. testing with a goal of achieving extensive geographic and Dermacentor species representation. Since D. variabilis was identified in all 44 states with Dermacentor spp. submitted, D. variabilis specimens (n = 827) were selected for use to reflect a broad geographic distribution across the four U.S. regions (Northeast, South, Midwest, and West) (Blagburn et al. 1996, Duncan et al. 2021. D. andersoni (n = 12) and D. albipictus (n = 9) were included to diversify the species tested, even though low numbers were submitted. To determine if less commonly identified pathogens were present, a subsample of ticks (n = 150) from the southern United States where canine ehrlichiosis is known to occur was tested for E. canis, and a subset of ticks (n = 189) primarily from the southern United States and from along the Atlantic and Pacific Coasts where babesiosis has been reported or suspected to occur was tested for Babesia species (Beall et al. 2012, Birkenheuer et al. 2020).

Molecular assays
Each evaluated tick was dissected dorsoventrally and all internal contents collected for nucleic acid extraction, which was performed using a commercial blood kit with no modification added (Illustra GenomicPrep Kit; GE Healthcare). Successful extraction was measured by PCR of the partial fragments of an ITS-2 gene and 16S rRNA gene in a previous study (Duncan et al. 2021); there were no failed DNA extraction or notable PCR inhibitor in this study. Rickettsia spp. was detected and identified using previously described and optimized 17-kDa and ompA gene targets (Sumner et al. 2007, Heise et al. 2010). An E. canis-specific PCR targeting a 16S rRNA gene fragment was performed to determine if E. canis was present in the subsample described in the previous paragraph (Chen et al. 1994, Dawson et al. 1996. Babesia spp. primers targeting fragments of 18S and the large subunit region rRNA genes were used for PCR on the subset of ticks described above (Kim et al. 2013, Qurollo et al. 2017). All amplified products were confirmed on a 2% agarose gel, column purified, and sequenced with an ABI 3730 capillary sequencer (Applied Biosystems, Foster City, CA, USA) at the Oklahoma State University Molecular Core Facility (Stillwater, Oklahoma, USA). High-quality electropherograms were verified by visual inspection and sequences were compared to those available in GenBank (https://blast.ncbi.nlm.nih .gov/Blast.cgi). prevalence against tick stage, tick host, and tick U.S. region origin. The level of significance was set at a = 0.05, and Bonferroni correction was applied to multiple comparisons (i.e., U.S. regions). Maps were developed with MapViewer 8.7.752 (Golden Software, LLC, Golden, CO, USA) with geographic regions (Northeast, South, Midwest, and West) as previously established (Blagburn et al. 1996).
Overall prevalence of Rickettsia spp. in ticks from the four different defined regions (Northeast, South, Midwest, and West) was not significantly different (chi-squared test, w 2 = 7.06, df = 3, p = 0.07) with Rickettsia spp. detected in 13.4% (38/284; 95% CI 9.9-17.9) of Midwest ticks, 13.0% (24/185; 95% CI 8.8-18.6) of ticks from the West, 11.8% (24/ 203; 95% CI 8.0-17.0) of ticks from the South, and 11.0% (20/181; 95% CI 7.2-16.5) of ticks from the Northeast (Table 2). Across every region, R. montanensis was most often detected in the evaluated ticks. In addition, R. bellii was detected in ticks originating from every region. Rickettsia peacockii was only detected in D. andersoni from the Rocky Mountain states in the West, and no other Rickettsia spp. was present in the evaluated D. andersoni (n = 12). R. rhipicephali was detected in ticks that originated solely in the southern portions of the United States, and the three ticks with R. amblyommatis were submitted from states in the Northeast (Fig. 2).

Discussion
Even though D. variabilis and D. andersoni are considered primary vectors of R. rickettsii, the causative agent of RMSF, we did not detect this pathogen in any Dermacentor tick (n = 848) collected from dogs and cats in 44 U.S. states.    Dermacentor andersoni (n = 12) was only submitted from the western United States; Rickettsia spp. was not detected in any Dermacentor albipictus (n = 9). a Dogs with Dermacentor variabilis in which Rickettsia amblyommatis was identified were co-infested with Ixodes scapularis (n = 1) or not co-infested with other tick species at the time of presentation (n = 2). b Dogs with D. variabilis in which Rickettsia rhipicephali was identified were co-infested with Amblyomma americanum (n = 3), Amblyomma maculatum (n = 2), I. scapularis (n = 1), and Rhipicephalus sanguineus sensu lato (n = 1), or were not co-infested with other tick species at the time of presentation (n = 8).

FIG. 2.
Geographic distribution of Rickettsia spp. detected in Dermacentor spp. from dogs and cats across the United States; each symbol represents a tick with a detectable Rickettsia spp. (n = 106). Color images are available online.
Instead, we identified various other SFGR, which support data from other studies reporting a range of SFGR and a low prevalence of R. rickettsii in Dermacentor ticks (Moncayo et al. 2010, Fritzen et al. 2011, Trout Fryxell et al. 2015, Santanello et al. 2018, Francis et al. 2020, Luedtke et al. 2020, Occi et al. 2020. In fact, when identified in D. variabilis, R. rickettsii has a reported prevalence less than 1%; however, this is in contrast to previous reports of greater than 8% of ticks harboring R. rickettsii (Schriefer and Azad 1994, Stromdahl et al. 2011, Kakumanu et al. 2018, Hecht et al. 2019. This change to a lack, or rarity, of R. rickettsii in its primary tick vectors could be explained by evidence suggesting tick endosymbionts have the potential to alter tick physiology and transmission of pathogens as indicated by the inverse relationship between R. rickettsii and R. peacockii in D. andersoni and documentation of R. peacockii infection, excluding R. rickettsii infection in the same tick (Burgdorfer et al. 1981, Narasimhan andFikrig 2015). Approximately two-thirds (66.7%) of D. andersoni had R. peacockii identified in this study, although only 12 specimens were evaluated. Other studies have documented R. peacockii in 1.3-80% of D. andersoni tested and found no R. rickettsii in those specimens (Burgdorfer et al. 1981, Niebylski et al. 1997, Francis et al. 2020, Lefcort et al. 2020.
For D. variabilis, a similar mechanism may exist as data suggest this tick is unable to maintain two different Rickettsia spp. through transovarial transmission (Macaluso et al. 2002). In addition, other tick genera may be responsible for transmitting the causative agent of RMSF more frequently than originally recognized. In the southwestern United States, the brown dog tick, R. sanguineus s.l., is capable of transmitting R. rickettsii, while Haemaphysalis longicornis and A. americanum can successfully maintain and transmit R. rickettsii under laboratory conditions, although their transmission potential in nature is not fully known (Demma et al. 2005, Labruna et al. 2008, Levin et al. 2017, Stanley et al. 2020. Besides R. rickettsii, other SFGR can cause significant disease in humans and therefore may be contributing to the rise in reported cases of rickettsiosis since many SFGR crossreact on commonly used serologic assays (Moncayo et al. 2010). Two such examples include R. philipii (Rickettsia 364D), the causative agent of Pacific Coast tick fever, transmitted by D. occidentalis along its range in California, and R. parkeri transmitted by A. maculatum in the southeastern United States (Paddock et al. 2010, 2018, Stromdahl et al. 2011, Yoshimizu and Billeter 2018. In this study, R. montanensis was the most prevalent Rickettsia spp. identified in D. variabilis, with this agent detected in over half (60.4%; 64/106; 95% CI 50.4-69.8) of the Rickettsia-positive ticks. Geographically, R. montanensis was found in ticks collected from every U.S. region and in 26 of the 44 states in which ticks were evaluated. Regional surveys from 22 states across the United States similarly found R. montanensis in D. variabilis (Table 1). Although the pathogenicity of R. montanensis in dogs and cats is currently unknown, mild disease has been induced in laboratory-reared guinea pigs and the agent was identified in a D. variabilis collected from a human before the development of a rash illness (McQuiston et al. 2012, Snellgrove et al. 2021. When considering the case report of human disease attributed to this SFGR, the widespread distribution of R. montanensis reported in this study is noteworthy. Before the human case report, R. montanensis was considered solely a nonpathogenic symbiont of D. variabilis (McQuiston et al. 2012, Parola et al. 2013. In ticks, the transition from endosymbiont to pathogen is not novel as evidenced by the emergence of disease-causing strains of Coxiella following numerous mutations and acquisition of new virulence factors (Duron et al. 2015). A recent example comes from the southern United States where a novel Rickettsia sp. has been identified as a cause of rickettsiosis in dogs (Wilson et al. 2020). Further investigation of intracellular pathogens and their evolution suggests this transition to a medically relevant pathogen may be achieved through inaccurate replication of the nonpathogenic Rickettsia spp. widely prevalent in ticks (Darby et al. 2007, Bonnet et al. 2017).
In total, eight different Rickettsia spp. were identified in this study. Diversity of SFGR appears common in surveys of D. variabilis with as many as 10 species having been found within a population of this tick (Stromdahl et al. 2011, Trout Fryxell et al. 2015, Kakumanu et al. 2018, Hecht et al. 2019). In general, Rickettsia spp. makes up the largest proportion of microbial biomass of D. variabilis (Sanchez-Vicente et al. 2019). Although none of the ticks examined in this study had more than one Rickettsia spp. detected, it has been reported that multiple Rickettsia spp. can infect a single D. variabilis or D. andersoni; however, the transovarial transmission of more than one Rickettsia species may not occur (Macaluso et al. 2002, Carmichael and Fuerst 2010, Kakumanu et al. 2018, Francis et al. 2020. Regardless, detecting the presence of Rickettsia spp. or any other tick-borne pathogen in a given tick does not confirm that species plays a role in transmission; vector competence can only be assessed through experimental transmission studies. Dermacentor spp. is most well known for their ability to transmit Rickettsia spp., but this genus may also serve as a vector for other medically relevant tick-borne pathogens. For instance, two D. variabilis ticks from Mexico, an endemic region for canine ehrlichiosis, were found to have E. canis, and experimentally, D. variabilis is capable of transmitting E. canis ( Johnson et al. 1998, Sosa-Gutierrez et al. 2016. However, none of the Dermacentor tested (n = 150) in this study had detectable E. canis. The ticks in this study were submitted primarily from regions of the United States where canine infection with E. canis is relatively rare, which may have precluded detection of the organism (Beall et al. 2012). Moreover, as with detection alone, proving vector competence in an experimental study does not confirm relevance to transmission in nature. Ixodes scapularis ticks from Louisiana, a state in the southern United States, are capable of transmitting Borrelia burgdorferi in the laboratory, but, due to differences in tick phenology, host preferences, and questing behavior, do not sustain a maintenance system for this agent in the field ( Jacobs et al. 2003, Goddard et al. 2015, Ginsberg et al. 2021).
In addition, some agents of babesiosis are suspected to be transmitted by Dermacentor spp. (Kjemtrup and Conrad 2006, Shock et al. 2014, Swei et al. 2019. In a multistate survey, 2.7% of the D. variabilis tested had detectable Babesia spp., and D. albipictus has been documented in the transmission cycle of B. duncani to humans (Shock et al. 2014, Swei et al. 2019. In this study, B. conradae, a potentially fatal agent infecting dogs in the southern United States, was identified in two D. albipictus collected from cats. The majority of reported cases of B. conradae is associated with kennels of coyote-hunting dogs, and although ticks are suspected to play a role in transmission, to date, no tick has been named as the primary vector (Kjemtrup andConrad 2006, Dear et al. 2018). Our findings suggest further work is needed to elucidate the relationship, if any, between D. albipictus and B. conradae.
This study had several limitations. The ticks examined in this study were collected from pets and may have ingested a blood meal before submission; therefore, the results of this study do not implicate a particular Dermacentor sp. as the primary vector for the pathogens identified. For instance, R. rhipicephali is more often identified in R. sanguineus (s.l.), R. amblyommatis is associated frequently with A. americanum, and R. cooleyi with Ixodes spp., and yet these agents have been found in various tick species (Hayes and Burgdorfer 1979, Moncayo et al. 2010, Fritzen et al. 2011, Barrett et al. 2014). In addition, if multiple Rickettsia spp. were present in a tick, reaction conditions may have favored amplification of the target in greater abundance or failed to amplify some Rickettsia spp. at all, precluding their detection. Similarly, although ticks were available from 44/50 states for analysis, our reliance on passive surveillance may have resulted in inadequate numbers of tick specimens to test from a given state, leading to failure of detection of less common pathogens.

Conclusion
This study highlights the diversity of SFGR in D. variabilis collected from dogs and cats across the United States. The majority of the detected species currently has an unknown pathogenicity in pets, but may induce cross-reacting antibodies on IFA, confounding diagnostic tests. Continued surveillance of tick-borne agents for the sake of veterinary and human health is warranted.