Trypanosoma cruzi and Other Vector-Borne Infections in Shelter Dogs in Two Counties of Oklahoma,United States

Abstract

Trypanosoma cruzi is an emerging zoonotic vector-borne parasite infecting dogs and other mammals in the United States. In this study we evaluated shelter dogs in one northeastern and one southeastern county in Oklahoma for prevalence of exposure to T. cruzi. Dogs were tested for antibodies against T. cruzi using the Chagas STAT PAK^® assay and for T. cruzi in circulation by PCR. In addition, dogs were tested for evidence of infection with other vector-borne organisms using the SNAP^® 4Dx^® Plus Test and PCR. Overall, 26 of 197 (13.2%) shelter dogs had detectable antibodies against T. cruzi and 3 of 189 (1.6%) dogs were PCR positive. In addition, we found that 42 of 197 (21.3%) shelter dogs had evidence of exposure to or were infected with at least one vector-borne agent other than T. cruzi based on serology and/or PCR; 9 of 42 (21.4%) of these dogs were also positive for T. cruzi antibodies. Other infections identified in dogs included Anaplasma phagocytophilum, Anaplasma platys, Babesia sp. (Coco), Dirofilaria immitis, Ehrlichia canis, Ehrlichia chaffeensis, Ehrlichia ewingii, and Hepatozoon americanum. This study serves to boost state-wide veterinary and public health awareness of T. cruzi and other vector-borne pathogens infecting shelter dogs in Oklahoma. Results indicate the need for more comprehensive screening of shelter dogs in Oklahoma for exposure to vector-borne agents to enhance surveillance and to identify dogs in need of additional specific veterinary care.

Introduction

T rypanosoma cruzi is the vector-borne hemoflagellate causing Chagas disease in dogs and humans in the Americas (Curtis-Robles et al. 2017, Bern et al. 2019). The disease is often insidious, leading to fatal cardiac and/or gastrointestinal damage and dysfunction over time. Domestic dogs are regarded as reservoirs of T. cruzi infection to triatomine bugs (Hemiptera: Reduviidae, subfamily Triatominae), commonly known as kissing bugs, in some regions of Latin America. The role of infected dogs in the epidemiology of Chagas disease in the United States is not well understood (Curtis-Robles et al. 2017, Bern et al. 2019, Elmayan et al. 2019).

Cases of canine Chagas disease have been reported in southern states including Georgia, Louisiana, Oklahoma, South Carolina, Tennessee, and Texas (Bradley et al. 2000, Rowland et al. 2010). Limited surveillance studies assessing canine T. cruzi infection prevalence in states with documented cases collectively indicate that exposure rises with increased age, increased time spent outdoors, and housing within multidog and outdoor/open-air kennels (Rowland et al. 2010, Tenney et al. 2014, Curtis-Robles et al. 2017, 2018, Elmayan et al. 2019, Hodo et al. 2019, Busselman et al. 2021); these factors are thought to be associated with elevated risk of exposure to infected triatomine vectors.

In 2010, Rowland et al. (2010) found the seroprevalence of T. cruzi exposure among pet dogs across Tennessee to be 6.4%. More recently, Elmayan et al. (2019) found an overall seroprevalence of at least 6.9% in shelter dogs in southern Louisiana. In Texas, where the majority of Chagas surveillance studies have been conducted, shelter dog populations have a seroprevalence of 3.8–29.5% depending on county or ecoregion (Tenney et al. 2014, Garcia et al. 2016, Hodo et al. 2019).

In predominantly outdoor pet dogs in seven colonias along the Texas–Mexico border, an overall seroprevalence of at least 19.6% was found (Curtis-Robles et al. 2017). In kenneled working coon hounds in south-central Texas, an overall seroprevalence of 57.6% was reported (Curtis-Robles et al. 2017), and in government-owned working dogs along the Texas–Mexico border, a seroprevalence of at least 7.4%, possibly as high as 18.9%, was found (Meyers et al. 2017).

In Oklahoma, Bradley et al. (2000) detected T. cruzi reactive antibodies in 11 of 301 (3.6%) owned, shelter, or stray dogs that were associated with three canine Chagas cases in separate eastern counties; all but one of the antibody-positive dogs were owned. One antibody-positive dog was found to have circulating parasite by PCR. With the exception of a nationwide survey of U.S. Department of Homeland Security working dogs, which documented seronegative and PCR-negative dogs in Oklahoma (Meyers et al. 2020), the prevalence of T. cruzi exposure in dogs in Oklahoma has not been assessed since the investigation conducted by Bradley et al. (2000) over two decades ago.

Current T. cruzi infection prevalence in dogs in Oklahoma may be similar to that recently documented in other states due to travel or displacement of dogs, or possibly from autochthonous transmission occurring within the state (Curtis-Robles et al. 2017, Elmayan et al. 2019). Our primary goal in this study was to reassess prevalence of canine T. cruzi exposure in Oklahoma. To this end, we tested shelter dogs in two regionally disparate counties in Oklahoma for antibodies against T. cruzi using the Chagas STAT PAK^® assay and for T. cruzi in circulation by PCR.

In addition, we tested shelter dogs for evidence of infection with other vector-borne pathogens using the SNAP^® 4Dx^® Plus Test and PCR. Results presented herein indicate that T. cruzi infection prevalence in shelter dogs in Oklahoma is similar to that reported in other states, particularly in some areas of the adjoining state of Texas. In addition, we document a variety of other vector-borne pathogens in shelter dogs in Oklahoma.

Materials and Methods

Shelter dogs

Shelter dogs resided in several small nonprofit animal shelters in LeFlore County and in a municipal animal shelter in Tulsa County during years 2018–2019. Sample collection was conducted multiple times over the course of the study depending on personnel availability. Blood was collected from nonaggressive dogs that were residing in the shelters at the time of each visit; other than amenable temperament, there were no exclusion or inclusion criteria of dogs. The length of time dogs spent in shelters before blood collection was unknown.

Whole blood (1–3 mL) in ethylenediaminetetraacetic acid was collected for testing. Demographic data including age (estimated or known), gender, and breed designation (mixed or specified) were available from the majority of dogs tested. Samples from dogs in LeFlore County were collected and posted by a local veterinarian to Oklahoma State University's College of Veterinary Medicine (OSU-CVM) through standard mail. Samples from dogs in Tulsa County were collected by OSU-CVM personnel. Blood samples were processed the same day of collection or were stored at 4°C for up to 72 h before processing once received in the mail or transported to OSU-CVM.

Serology

For serological evaluation, plasma was harvested from 1 mL of whole blood by centrifugation for 15 min at 560 g. Plasma and remaining whole blood samples were stored at 4°C or −20°C until testing. Plasma samples were tested for T. cruzi reactive antibodies using the Chagas STAT PAK^® assay (Chembio Diagnostic, Inc., Medford, NY) according to the manufacturer's instructions. The immunochromatographic rapid antibody assay exhibits a specificity of 100% and sensitivity of 99.8%, and although indicated for use in human medicine, it has been used in research studies for antibody testing of other animal species (Nieto et al. 2009, Curtis-Robles et al. 2017).

According to the manufacturer, any color change of the STAT PAK test line, no matter how faint, in conjunction with the positive control line indicates a positive test. Samples with obvious coloration of the test line were regarded as antibody positive and samples with no coloration of test line were regarded as antibody negative. In the interest of not artificially inflating canine seroprevalence in this study, weakly reactive samples were regarded as equivocal for T. cruzi antibodies. Equivocal results were those in which coloration of the test line was barely perceptible but not completely absent; equivocal results were considered antibody negative when making statistical comparisons of T. cruzi serostatus between dog groups.

The majority of shelter dogs that were screened for T. cruzi were also tested for serological evidence of infection with other vector-borne agents depending on sample availability using the SNAP^® 4Dx^® Plus Test (IDEXX, Westbrook, ME); the commercial point-of-care rapid assay detects antibodies against Anaplasma phagocytophilum, Anaplasma platys, Borrelia burgdorferi, Ehrlichia canis, Ehrlichia chaffeensis, and Ehrlichia ewingii, and antigen of Dirofilaria immitis (Little et al. 2021).

DNA extraction and PCR

All DNA extractions, PCRs, and PCR purifications were performed in separate dedicated laboratory areas to prevent DNA contamination of samples. Commercial kits were used for DNA extractions and purifications, which were performed according to manufacturer protocols. Water samples were included with each round of DNA extraction and PCR to verify that cross-contamination of samples had not occurred. Whole blood samples from shelter dogs were extracted for DNA with the QIAamp^® DNA Blood Mini Kit (Qiagen, Valencia, CA) or the Illustra™ blood genomicPrep Mini Spin Kit (GE Healthcare, Piscataway, NJ).

To detect T. cruzi DNA present in extracts from whole blood, a 188 bp segment of a repetitive nuclear sequence was amplified using primers TCZ1 and TCZ2 as previously described, but using Taq DNA polymerase recombinant (Invitrogen, Waltham, MA) (Kirchhoff et al. 1996, Virreira et al. 2003).

Shelter dogs that tested positive on SNAP for Anaplasma spp. and Ehrlichia spp. antibodies were screened by PCR using previously described assays to potentially detect and identify the rickettsial organisms in circulation, indicating ongoing infection (Chen et al. 1994, Yabsley et al. 2008). Also, the majority of shelter dogs were tested by a PCR method amplifying tick-borne blood apicomplexa including Hepatozoon spp. and Babesia spp. (Gubbels et al. 1999, Allen et al. 2008, Yabsley et al. 2008) depending on sample availability.

PCR purification and sequencing

PCR products were electrophoresed in a 2% agarose matrix stained with GelRed^® (Biotium, Fremont, CA). According to kit manufacturer protocol, PCRs containing correctly sized amplicons were purified using the QIAquick^® PCR Purification Kit (Qiagen). Purified DNA extracts were assessed for quantity and quality using a NanoDrop™ spectrophotometer (ThermoFisher Scientific, Waltham, MA) before DNA sequence analysis (Sanger method). Sequencing was conducted by the Oklahoma State University Molecular Core Facility (Stillwater, OK). Sequences obtained were compared with those available in the National Center for Biotechnology Information database (GenBank™) to determine organism identity.

Statistical analyses

Excel, MedCalc, and QuickCalcs-GraphPad were used for statistical analyses. The level of significance for all analyses was set at α = 0.05. Mean ages of T. cruzi antibody-positive and antibody-negative dogs (included equivocal results) were compared using t-test. Prevalence of antibody and PCR-positive dogs was calculated with 95% confidence intervals (CIs).

Proportions of T. cruzi antibody-positive and antibody-negative dogs were compared using chi-square or Fisher's exact test with CIs to determine statistical differences in T. cruzi seroprevalence overall and between counties according to dog age bracket (<3 years [young], ≥3 to <6 years [middle age], and ≥6 years [senior]) (Meyers et al. 2017), gender, designated breed (specified or mixed), and positivity on SNAP. In addition, prevalence odds ratios (PORs) were calculated with CIs to determine whether there was an association between T. cruzi exposure (based on seropositive status) and exposure to other vector-borne infections as indicated by SNAP and PCR.

Results

Dogs

A total of 197 shelter dogs in Oklahoma were tested in this study; 69 (35%) dogs resided in LeFlore County shelters and 128 (65%) dogs resided in the Tulsa County shelter. Age, gender, and breed description was not known for 2, 11, and 10 dogs, respectively. Neither clinical history nor travel history was known for any dog. See Table 1 for demographic data of shelter dogs tested. Overall, shelter dogs ranged in age from 5 months to 13 years ( $\bar{x}$ = 2.3 years, 95% CI 2.0–2.6). When comparing demographic data from dogs between counties, the only significant difference was that more dogs in Tulsa County than in LeFlore County were designated at intake as mixed breed (p < 0.0001).

Table 1.

Chagas STAT PAK Results for Shelter Dogs in Oklahoma by County and Demographics

	LeFlore county (n = 69)				Tulsa county (n = 128)				Total (n = 197)
	Sample size, n (%)	Trypanosoma cruzi Ab positive, n (%)	T. cruzi Ab negative, n (%)	p	Sample size, n (%)	T. cruzi Ab positive, n (%)	T. cruzi Ab negative, n (%)	p	Sample size, n (%)	T. cruzi Ab positive, n (%)	T. cruzi Ab negative, n (%)	p
STAT PAK		13 (18.8, 95% CI 11.2–29.8)	56 (81.2, 95% CI 70.3–88.8)			13 (10.2, 95% CI 5.9–16.7)	115 (89.8, 95% CI 83.3–94.1)			26 (13.2, 95% CI 9.1–18.7)	171 (86.8, 95% CI 81.3–91.9)
Mean age	67	2.8 (95% CI 1.4–4.3)	2.2 (95% CI 1.6–2.9)	0.39	128	3.1 (95% CI 2.0–4.2)	2.2 (95% CI 1.8–2.5)	0.10	195	3.0 (95% CI 2.1–3.8)	2.2 (95% CI 1.9–2.5)	0.07
Age group				0.23				0.04				0.09
Young	51 (76.1)	9 (17.6)	42 (82.4)		87 (68.0)	5 (5.7)	82 (94.3)		138 (70.8)	14 (10.1)	124 (89.9)
Middle age	9 (13.4)	1 (11.1)	8 (88.9)		33 (25.8)	7 (21.2)	26 (78.8)		42 (21.5)	8 (19.0)	34 (81.0)
Senior	7 (10.5)	3 (42.9)	4 (57.1)		8 (6.2)	1 (12.5)	7 (87.5)		15 (7.7)	4 (26.7)	11 (73.3)
Gender	58			0.79	128			0.41	186			0.48
Female	31 (53.4)	6 (19.4)	25 (80.6)		63 (49.2)	5 (7.9)	58 (92.1)		94 (50.5)	11 (11.7)	83 (88.3)
Male	27 (46.6)	6 (22.2)	21 (77.8)		65 (50.8)	8 (12.3)	57 (87.7)		92 (49.5)	14 (15.2)	78 (84.8)
Breed	66			0.48	121			0.99	187			0.32
Mixed	26 (39.4)	4 (15.4)	22 (84.6)		84 (69.4)	9 (10.7)	75 (89.3)		110 (58.8)	13 (11.8)	97 (88.2)
Specific	40 (60.6)	9 (22.5)	31 (77.5)		37 (30.6)	4 (10.8)	33 (88.2)		77 (41.2)	13 (16.9)	64 (83.1)

Ab, antibody; CI, confidence interval.

Chagas STAT PAK serology

In total, 26 (13.2%, 95% CI 9.1–18.7) shelter dogs tested positive for T. cruzi antibodies by the Chagas STAT PAK assay (Table 1). Equivocal results (weakly reactive) and negative results (no color development of the test line) were obtained for 14 (7.1%, 95% CI 4.2–11.7) and 157 (79.7%, 95% CI 73.5–84.8) dogs, respectively. Thus, 171 (86.8%, 95% CI 81.3–91.9) dogs were considered antibody negative for data analyses (Table 1).

All dogs testing positive for T. cruzi antibodies ranged in age from 1 to 8 years ( $\bar{x}$ = 3.0 years, 95% CI 2.1–3.8) (Table 1). Overall, higher proportions of senior dogs and middle-aged dogs were antibody positive than young dogs, but not significantly so. However, in Tulsa County specifically, there was a significantly higher prevalence of antibody-positive middle-aged dogs than young dogs (p = 0.02). There were no significant differences in proportions of T. cruzi seropositive and seronegative dogs by gender or breed designation overall, or between and within LeFlore County and Tulsa County shelters (Table 1).

SNAP 4Dx Plus serology

Overall, 41 of 188 (21.8%, 95% CI 16.5–28.3) dogs tested positive on SNAP for at least one analyte; Anaplasma spp. antibodies were detected in 5 dogs (2.7%, 95% CI 1.0–6.3), D. immitis antigen was detected in 15 dogs (8.0%, 95% CI 4.8–12.8), and Ehrlichia spp. antibodies were detected in 28 dogs (14.9%, 95% CI 10.5–20.7) (Table 2). There were 5, 1, and 1 dogs that were positive for both Anaplasma sp. and D. immitis, Anaplasma sp. and Ehrlichia sp., and D. immitis and Ehrlichia sp., respectively.

Table 2.

4Dx SNAP Plus Results for Shelter Dogs in Oklahoma by County and Trypanosoma cruzi Serostatus

	Leflore county				Tulsa county				Total
	Tc Ab+ (n = 13)	Tc Ab eq. (n = 1)	Tc Ab– (n = 55)	Total (n = 69)	Tc Ab+ (n = 13)	Tc Ab eq. (n = 13)	Tc Ab– (n = 93)	Total (n = 119)	Tc Ab+ (n = 26)	Tc Ab eq. (n = 14)	Tc Ab– (n = 148)	Total (n = 188)
SNAP Plus
Positive (%)	5 (38.5)	1 (100)	16 (29.1)	22 (31.9)	4 (30.8)	3 (23.1)	12 (12.9)	19 (16.0)	9 (34.6)	4 (28.6)	28 (18.9)	41 (21.8)
Anaplasma spp. (%)	1 (7.7)	0 (0)	4 (7.3)	5 (7.2)	0 (0)	0 (0)	0 (0)	0 (0)	1 (3.8)	0 (0)	4 (2.7)	5 (2.7)
Borrelia burgdorferi (%)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Dirofilaria immitis (%)	1 (7.7)	0 (0)	6 (10.9)	7 (10.1)	3 (23.1)	1 (7.7)	4 (4.3)	8 (6.7)	4 (15.4)	1 (7.1)	10 (6.8)	15 (8.0)
Ehrlichia spp. (%)	3 (23.1)	1 (100)	10 (18.2)	14 (20.3)	2 (15.4)	3 (23.1)	9 (9.7)	14 (11.8)	5 (19.2)	4 (28.6)	19 (12.8)	28 (14.9)
Negative (%)	8 (61.5)	0 (0)	39 (70.9)	47 (68.1)	9 (69.2)	10 (76.9)	81 (87.1)	100 (84.0)	17 (65.4)	10 (71.4)	120 (81.1)	147 (78.2)

Tc Ab+, T. cruzi antibody positive; Tc Ab−, T. cruzi antibody negative; Tc Ab eq., T. cruzi antibody equivocal.

When comparing results between counties, significantly more dogs in LeFlore County (22/69; 21.9%, 95% CI 22.1–43.6) than Tulsa County (19/119; 16%, 95% CI 10.4–23.7) were positive for at least one analyte on SNAP (p = 0.002) (Table 2). However, the only significant difference in proportions of dogs between counties testing positive for different analyte(s) on SNAP was that LeFlore County, being the location where all Anaplasma spp. positive dogs were found, had a significantly higher prevalence of Anaplasma spp. positive dogs than Tulsa County (p = 0.01).

Canine serostatus comparisons

Overall, 9 of 41 (22%, 95% CI 11.8–36.9) dogs tested positive for both T. cruzi antibodies and at least one analyte on SNAP (Table 2). Although the proportion of dogs testing both T. cruzi antibody positive and SNAP positive was higher than expected by random chance, the difference was not significant compared with T. cruzi antibody-negative dogs (p = 0.12). There were no significant differences in proportions of both T. cruzi antibody-positive and SNAP-positive shelter dogs between counties (p = 1.0).

Although 4 of 15 (26.7%, 95% CI 10.5–52.3) and 5 of 28 (17.9%, 95% CI 7.4–36.1) dogs positive for D. immitis antigen and Ehrlichia spp. antibodies, respectively, on SNAP also tested positive for T. cruzi antibodies, the proportions were not significantly higher than for T. cruzi antibody-negative dogs (p ≥ 0.13). The odds ratio (OR) calculation, although also not quite significant (p = 0.09), indicated that shelter dogs in Oklahoma that were exposed to at least one vector-borne agent detectable on SNAP were twice more likely to be seropositive for T. cruzi (OR = 2.15, 95% CI 0.87–5.27).

PCR

Three of 189 dogs (1.6%, 95% CI 0.3–4.8) were PCR positive for T. cruzi. See Table 3 for all positive PCR results of dogs in conjunction with Chagas STAT PAK and SNAP 4Dx Plus serology. One PCR-positive dog was antibody positive and a second dog was antibody equivocal on STAT PAK. T. cruzi amplicons from two dogs contained occasional base pairs that could not be reliably discerned in chromatograms due to double or mis-spaced peaks. Comparisons with sequences available in GenBank indicated organism that was 99–100% similar to LT220297 in two dogs and 96% similar to LT220289 in one dog; both GenBank sequences used for reference have been reported in kissing bugs (Triatoma gerstaeckeri) in Texas (Aleman et al. 2017).

Table 3.

Positive PCR Results, Chagas STAT PAK Results, and 4Dx SNAP Plus Results from Shelter Dogs in Oklahoma by County

County	Age (years)	Gender	Breed	Chagas STAT PAK	4Dx SNAP Plus	PCR results
LeFlore	5	M	Collie	−	A, Eh	A. phagocytophilum (99% KC4553636); E. canis (97% EU7816686)
	2	F	Hound mix	−	A, Di	A. platys (99% CP046391); T. cruzi (100% LT220297)
	1	F	Collie	+	A	A. phagocytophilum (99% KC455363)
	2	M	Walker Coonhound	Eq	Eh	E. ewingii (99% MN336353); Babesia sp. Coco (100% EU109716)
	1	F	Border Collie	−	Eh	E. ewingii (99% MN336353)
	2	M	Labrador mix	−	Eh	E. ewingii (99% MN336353)
	3	F	Beagle	−	Eh	E. ewingii (99% MN336353)
	1	F	Labrador/Chow mix	−	Negative	Babesia sp. Coco (100% EU109716)
	6	M	Labrador	+	Eh	H. americanum (99% EU146062)
Tulsa	2	M	Catahoula mix	−	Eh	E. chaffeensis (100% MN368552)
	2	F	Great Pyrenees	−	Eh	E. ewingii (99% MN336353)
	3	F	German Shepherd mix	Eq	Di, Eh	E. ewingii (99% MN336353)
	5	NR	Border collie	−	Di, Eh	E. ewingii (99% MN336353)
	1.5	M	Labrador mix	Eq	Negative	T. cruzi (99% LT220297)
	2	M	Beagle mix	+	Negative	T. cruzi (96% LT220289)

A, Anaplasma; Di, Dirofilaria immitis; Eh, Ehrlichia; Eq, equivalent; F, female; M, male; NR, not recorded.

Eleven of 26 (42.3%, 95% CI 25.5–61.1) SNAP-positive dogs were shown to have ongoing infections with Anaplasma spp. and/or Ehrlichia spp. by PCR (Table 3). Infections identified included A. phagocytophilum, A. platys, E. canis, E. chaffeensis, and E. ewingii. Three of 185 (1.6%, 95% CI 0.3–4.9) dogs were PCR positive for blood apicomplexa; infections identified were Babesia sp. (Coco) and Hepatozoon americanum (Table 3).

OR calculations did not show that dogs infected with Anaplasma spp. and Ehrlichia spp. based on PCR were at higher risk of T. cruzi exposure. However, although not significant, dogs infected with a blood apicomplexan (here, H. americanum) were over three times as likely (OR = 3.64, 95% CI 0.3–41.8, p = 0.3) to have been exposed to T. cruzi.

Interestingly, although still not significant, if equivocal STAT PAK results were regarded as positive, dogs infected with a blood apicomplexan (Babesia sp. [Coco] or H. americanum) were nine times more likely to have been exposed to T. cruzi (OR = 9.03, 95% CI 0.8–102.6, p = 0.08). Although peripheral to the objectives of this study, it is noteworthy that dogs infected with a blood apicomplexan were >12 times more likely to have Ehrlichia spp. antibodies on SNAP (OR = 12.4, 95% CI 1.1–142.3, p = 0.04).

Discussion

In this study we found an overall prevalence of T. cruzi exposure in shelter dogs in Oklahoma of 13.2% using the Chagas STAT PAK assay. A second optimized in-house serological test was not available at the time of this study. It is possible that if a second serological assay was used, which is the case for most reported canine surveillance studies, and only agreeing results between assays were regarded as positive or negative, that the seroprevalence found in our study would be lower. To limit upward skewing of seroprevalence, however, dogs showing faint test lines on STAT PAK were regarded as seronegative for statistical analyses.

The seroprevalence found in shelter dogs in Oklahoma does not seem unreasonable compared with surveillance data from other states, particularly from the adjacent state of Texas (Hodo et al. 2019). A similar finding of our study to other investigations was that, although not statistically significant, T. cruzi antibody-positive shelter dogs in Oklahoma tended to be middle-aged to senior, suggesting that risk of kissing bug exposure increases with lifespan (Meyers et al. 2017, Elmayan et al. 2019).

We also found three PCR-positive dogs, one of which was antibody negative and the other very weakly antibody positive (equivocal), suggesting recent exposure and acute stage infections in which T. cruzi reactive antibodies were not yet readily detectable by the STAT PAK assay. PCR-positive but antibody-negative dogs have been documented in other studies (Tenney et al. 2014, Elmayan et al. 2019).

Travel histories were not known for any of the dogs. It is possible that several to all of the T. cruzi seropositive and PCR-positive shelter dogs were infected elsewhere (Brown et al. 2010, Aleman et al. 2017, Bern et al. 2019, Elmayan et al. 2019). However, it is a logical supposition that at least some, possibly the majority, of positive dogs were exposed to T. cruzi in Oklahoma, given that autochthonous T. cruzi transmission in dogs has been reported (Bradley et al. 2000).

Also, two competent kissing bug species (Triatoma lecticularia and Triatoma sanguisuga) are endemic to the state and T. cruzi-infected wildlife species have been documented (Bern et al. 2019). Furthermore, kissing bugs harboring T. cruzi DNA have been documented in counties spanning the state of Oklahoma; domestic dog DNA was detected in a kissing bug from the same study, indicating that vector–canine interactions occur within the state (Allen and Lineberry 2022).

In this study, SNAP data revealed that 21.8% of the shelter dogs tested had been exposed to at least one vector-borne agent other than T. cruzi. Associations between positive T. cruzi serology and exposure to other vector-borne pathogens as indicated by serology or PCR were not significant according to POR calculations. The finding of A. phagocytophilum infection in two shelter dogs likely signifies instances of dog relocation from other states, as the zoonotic tick-borne rickettsial organism is not thought to be endemic in the central and southcentral United States (McMahan et al. 2016).

Another surprising finding was the identification of Babesia sp. (Coco) in two dogs. Babesia sp. (Coco) was documented previously in Oklahoma in two dogs that were clinically unwell due to other health factors (J. Meinkoth, pers. comm., November 11, 2019; Fujita et al. 2015). Case reports of Babesia sp. (Coco) have also occurred in Texas and several eastern states in dogs with immune-compromising conditions (Baneth 2018, CAPC 2019). It is possible that the two Babesia sp. (Coco)-positive shelter dogs had underlying health issues, but clinical histories were not known for any dogs tested in this study; one dog, however, was coinfected with E. ewingii and was weakly T. cruzi antibody positive.

The prevalence of exposure to T. cruzi and other vector-borne agents that we found in shelter dogs in this study is not likely reflective of the prevalence of exposure in pet dogs in Oklahoma. Shelter dogs generally receive less routine veterinary care than pet dogs, and are less likely to be treated with labeled products that kill or repel arthropods, thus increasing their risk of exposure to infected arthropod vectors. Also, shelter dogs often spend more of their lifetimes outdoors, and hence are more likely to encounter arthropods that may harbor pathogens (Dantas-Torres and Otranto 2016, Loza et al. 2017, Hodo et al. 2019).

Conclusion

This is the first reported investigation estimating prevalence of T. cruzi exposure in shelter dogs in Oklahoma in >20 years. The prevalence found appears to be similar to that documented in shelter dogs surveyed in other states, particularly in some areas of Texas. It is not known whether positive dogs in this study were relocated from other areas with known autochthonous canine Chagas disease, or whether they were infected while in Oklahoma.

However, given the spate of other vector-borne infections detected in this study that are known to occur in dogs in Oklahoma, it is possible that shelter dogs within the state are also exposed to kissing bugs harboring T. cruzi. Further research is warranted to better characterize the prevalence of T. cruzi and other vector-borne infections in shelter dogs in Oklahoma; our findings suggest the need for more comprehensive screening of shelter dogs for vector-borne diseases within the state to increase veterinary and public health awareness, and to better provide affected dogs with appropriate veterinary care.

Footnotes

Acknowledgments

The authors extend our sincerest gratitude to Melanie Foster, Doctor of Veterinary Medicine (DVM) for allowing us to collect blood from dogs residing in the Tulsa County municipal shelter, to personnel at the shelter assisting us, to other OSU-CVM personnel assisting with blood collections especially Kellee Sundstrom, and to Joe Dubois, DVM for sending blood samples from shelter dogs in LeFlore County. The authors also thank undergraduate, graduate, and veterinary students in the research laboratory for testing canine samples.

Authors' Contributions

K.E.A. and M.W.L. conceived and designed the surveillance study. M.W.L. obtained canine samples from the Tulsa County shelter, and coordinated and performed serological and molecular testing on canine samples from Tulsa and LeFlore counties. K.E.A. and M.W.L. both performed statistical analyses presented in the article. K.E.A. wrote the draft article and M.W.L. assisted with revisions of the article and created data tables.

Ethics Approval and Consent to Participate

The study received approval from the OSU Institutional Animal Care and Use Committee (IACUC).

Author Disclosure Statement

No conflicting financial interests exist.

Funding Information

No external funding was received for the research presented in this article.

References

Aleman

, Guerra

, Maikis

, Milholland

, et al. Prevalence of Trypanosoma cruzi, the causal agent of Chagas disease, in Texas Rodent Populations. EcoHealth, 2017; 14:130–143.

Allen

, Li

, Kaltenboeck

, Johnson

, et al. Diversity of Hepatozoon species in naturally infected dogs in the southern United States. Vet Parasitol, 2008; 154:220–225.

Allen

, Lineberry

. Detection of Trypansoma cruzi in Kissing Bugs (Hemiptera: Reduviidae: Triatominae) Collected Across Oklahoma. J Med Entomol, 2022; 59:675–680.

Baneth

. Babesia of domestic dogs. In: Florin-Christensen

, Schnittger

, eds. Parasitic Protozoa of Farm Animals and Pets. Cham: Springer, 2018:241–258.

Bern

, Messenger

, Whitman

, Maguire

, et al. Chagas Disease in the United States: A public health approach. Clin Microbiol Rev, 2019; 33:e00023-19.

Bradley

, Bergman

, Woods

, Crutcher

, et al. Prevalence of American trypanosomiasis (Chagas disease) among dogs in Oklahoma. JAVMA, 2000; 217:1853–1857.

Brown

, Roellig

, Gompper

, Monello

, et al. Seroprevalence of Trypanosoma cruzi among eleven potential reservoir species from six states across the Southern United States. Vector Borne Zoonotic Dis, 2010; 10:757–763.

Busselman

, Meyers

, Zecca

, Auckland

, et al. High incidence of Trypanosoma cruzi infections in dogs directly detected through longitudinal tracking at 10 multi-dog kennels, Texas, USA. PLoS Negl Trop Dis, 2021; 15:e0009935.

CAPC. 2019, March 26. Available at https://capcvet.org/

10.

Chen

, Dumler

, Bakken

, Walker

, et al. Identification of a granulocytotropic Ehrlichia species as the etiologic agent of human disease. J Clin Microbiol, 1994; 32:32589–32595.

11.

Curtis-Robles

, Hamer

, Lane

, Levy

, et al. Bionomics and spatial distribution of triatomine vectors of Trypanosoma cruzi in Texas and Other Southern States, USA. Am J Trop Med Hyg, 2018; 98:113–121.

12.

Curtis-Robles

, Snowden

, Dominguez

, Dinges

, et al. Epidemiology and molecular typing of Trypanosoma cruzi in naturally-infected hound dogs and associated triatomine vectors in Texas, USA. PLoS Neg Trop Dis, 2017; 11:e0005298.

13.

Dantas-Torres

, Otranto

. Best practices for preventing vector-borne diseases in dogs and humans. Trends Parasitol, 2016; 32:43–55.

14.

Elmayan

, Tu

, Duhon

, Marx

, et al. High prevalence of Trypanosoma cruzi infection in shelter dogs from southern Louisiana, USA. Parasites Vectors, 2019; 12:322.

15.

Fujita

, Starkey

, Allison

, Meinkoth

. Two dogs infected with Babesia sp. (Coco) in Oklahoma. Joint Annual Meeting of the American College of Veterinary Pathologists and Society of Toxicologic Pathology. Minneapolis, MN, October 17–21, 2015.

16.

Garcia

, O'Day

, Fisher-Hoch

, Gorchakov

, et al. One health interactions of Chagas disease vectors, canid hosts, and human residents along the Texas-Mexico Border. PLoS Negl Trop Dis, 2016; 10:e0005074.

17.

Gubbels

, de Vos

, van der Weide

, Viseras

, et al. Simultaneous detection of bovine Theileria and Babesia species by reverse line blot hybridization. J Clin Microbiol, 1999; 37:1782–1789.

18.

Hodo

, Rodriguez

, Curtis-Robles

, Zecca

, et al. Repeated cross-sectional study of Trypanosoma cruzi in shelter dogs in Texas, in the context of Dirofilaria immitis and tick-borne pathogen prevalence. J Vet Intern Med, 2019; 33:158–166.

19.

Kirchhoff

, Votava

, Ochs

, Moser

, et al. Comparison of PCR and microscopic methods for detecting Trypanosoma cruzi . J Clin Microbiol, 1996; 34:1171–1175.

20.

Little

, Braff

, Place

, Buch

, et al. Canine infection with Dirofilaria immitis, Borrelia burgdorferi, Anaplasma spp., and Ehrlichia spp. in the United States, 2013–2019. Parasites Vectors, 2021; 14:10.

21.

Loza

, Talaga

, Herbas

, Canaviri

, et al. Systemic insecticide treatment of the canine reservoir of Trypanosoma cruzi induces high levels of lethality in Triatoma infestans, a principal vector of Chagas disease. Parasites Vectors, 2017; 10:344.

22.

McMahan

, Wang

, Beall

, Bowman

, et al. Factors associated with Anaplasma spp. seroprevalence among dogs in the United States. Parasites Vectors, 2016; 9:169.

23.

Meyers

, Meinders

, Hamer

. Widespread Trypanosoma cruzi infection in government working dogs along the Texas-Mexico border: Discordant serology, parasite genotyping and associated vectors. PLoS Neg Trop Dis, 2017; 11:e0005819.

24.

Meyers

, Purnell

, Ellis

, Auckland

, et al. Nationwide Exposure of U.S. Working Dogs to the Chagas Disease Parasite, Trypanosoma cruzi . Am J Trop Med Hyg, 2020; 102:1078–1085.

25.

Nieto

, Boughton

, Dorn

, Steurer

, et al. Comparison of two immunochromatographic assays and the indirect immunofluorescence antibody test for diagnosis of Trypanosoma cruzi infection in dogs in south central Louisiana. Vet Parasitol, 2009; 165:241–247.

26.

Rowland

, Maloney

, Cohen

, Yabsley

, et al. Factors associated with Trypanosoma cruzi exposure among domestic canines in Tennessee. J Parasitol, 2010; 96:547–551.

27.

Tenney

, Curtis-Robles

, Snowden

, Hamer

. Shelter dogs as sentinels for Trypanosoma cruzi transmission across Texas, USA. Emerg Infect Dis, 2014; 20:8.

28.

Virreira

, Torrico

, Truyens

, Alonso-Vega

, et al. Comparison of polymerase chain reaction methods for reliable and easy detection of congenital Trypanosoma cruzi infection. Am J Trop Med Hyg, 2003; 68:574–582.

29.

Yabsley

, McKibben

, Macpherson

, Cattan

, et al. Prevalence of Ehrlichia canis, Anaplasma platys, Babesia canis vogeli, Hepatozoon canis, Bartonella vinsonii berkhoffii, and Rickettsia spp. in dogs from Grenada. Vet Parasitol, 2008; 151:279–285.