Tick-borne coinfections modulate CD8⁺ T cell response and progressive leishmaniosis

ABSTRACT

Leishmania infantum causes human visceral leishmaniasis and leishmaniosis (CanL) in reservoir host, dogs. As infection progresses to disease in both humans and dogs, there is a shift from controlling type 1 immunity to a regulatory, exhausted T cell phenotype. In endemic areas, the association between tick-borne coinfections (TBCs) and Leishmania diagnosis and/or clinical severity has been demonstrated. To identify immune factors correlating with disease progression, we prospectively evaluated a cohort of L. infantum-infected dogs from 2019 to 2022. The cohort was TBC-negative with asymptomatic leishmaniosis at the time of enrollment. We measured TBC serology, anti-Leishmania antigen T cell immunity, CanL serological response, parasitemia, and disease severity to probe how nascent TBC perturbs the immune state. At the conclusion, TBC+ dogs with CanL experienced greater increases in anti-Leishmania antibody reactivity and parasite burden compared to dogs that did not have incident TBC during the study. TBC+ dogs were twice as likely to experience moderate (LeishVet stage 2) or severe/terminal disease (LeishVet stage 3/4). Prolonged exposure to TBC was associated with a shift in Leishmania antigen-induced interferon gamma (IFN-γ)/interleukin-10 (IL-10) and enhanced CD8 T cell proliferation. Frequency of proliferating CD8 T cells significantly correlated with parasitemia and antibody reactivity. TBC exacerbated parasite burden and immune exhaustion. These findings highlight the need for combined vector control efforts as prevention programs for dogs in Leishmania endemic areas to reduce transmission to humans. Public health education efforts should aim to increase awareness of the connection between TBC and leishmaniosis.

INTRODUCTION

Disease caused by Leishmania infantum, visceral leishmaniasis (VL) in humans and leishmaniosis (CanL) in the reservoir host, dogs, is fatal in both species if not treated (1, 2). Only a fraction of infected human or canine hosts progress to disease for reasons still poorly understood (3, 4). Host cell microbicidal mechanisms have been shown to prevent the establishment of intracellular infection within phagocytic cells. During controlled infection, LeishVet stages 1 and 2, a type 1 immune response dominated by interferon gamma (IFN-γ)-secreting CD4 T cells contributes to maintaining low parasite burden (5–7). A shift toward a regulatory immune environment with concomitant immunosuppression diminishing the host’s ability to constrain parasite replication can lead to high IL-10 production, pathology-inducing levels of circulating antibodies (hypergammaglobulinemia), reduced T cell reactivity, parasite outgrowth, and pancytopenia (2, 8). This manifests clinically as anemia, malaise, and cachexia, characteristic of VL, with the addition of renal disease in CanL as well as HIV-coinfected people with VL, marked by elevated creatinine and proteinuria, a serious complication associated with high mortality rates (2). The temporal changes and factors that drive the shift from controlling type 1 immunity to a regulatory, exhausted T cell phenotype are poorly understood (7, 9). Increasing parasite burden and moderate clinical disease were associated with increased transmission from dogs to sand fly vectors (10). Therefore, understanding what immune factors correlate with disease progression using dogs, both as a model and a critical reservoir host, is crucial to interrupting the transmission cycle for overall public health.

VL is a neglected tropical disease found in regions where comorbid challenges predispose the host to progressive disease. Coinfection with a variety of pathogens from immunosuppressive viruses to immune-shifting helminth infections has been correlated with a higher likelihood of VL (11, 12). This phenomenon also occurs in the canine reservoir, where common coinfections are tick-borne bacterial pathogens. A causal association between tick-borne coinfections (TBCs) and the rate of Leishmania-positive diagnosis and/or clinical severity has been demonstrated in endemic areas (13). To better understand the immune alterations that occur in naturally infected hosts prior to disease progression, we followed a cohort of L. infantum-infected dogs from 2019 to 2022 to perform a prospective evaluation of anti-Leishmania antigen immunity and how inflammatory alterations like comorbid tick-borne infections perturb the immune state leading to progression with severe and/or fatal consequences.

We hypothesize that TBCs modulate the anti-Leishmania immune response to negatively impact parasitic control, pathogenic antibody progression, and disease. TBCs are acutely inflammatory (14), which may push a shift toward a regulatory environment in the setting of chronic Leishmania infection and promote the onset of immune exhaustion. During this study, we identified the kinetics and immune markers associated with this critical shift and resultant progressive leishmaniosis.

MATERIALS AND METHODS

Animals/cohort selection

We identified a cohort of naturally L. infantum-exposed dogs without clinical CanL and seronegative for the most common endemic canine tick-borne pathogens: Borrelia burgdorferi, Anaplasma phagocytophilum, Anaplasma platys, Ehrlichia chaffeensis, Ehrlichia canis, and Ehrlichia ewingii. This cohort was selected from a population of client-owned hunting hounds in which L. infantum is enzootic across three sites in the Midwestern USA (15). Due to the working nature of this cohort, there is relatively high exposure to ticks and tick-borne pathogens compared to companion dogs, similar to dogs in endemic areas of Brazil or Southern Europe (15). This cohort provided us a unique opportunity to monitor immune variables over time, as dogs are naturally exposed to environmentally occurring TBCs and concomitant CanL development.

Our research group assayed cohort Leishmania and TBC diagnostics, staged clinical CanL, and performed ex vivo T cell restimulation assays in response to Leishmania antigen at 3 month intervals over a course of 18 months, inclusive of two tick seasons. US hunting dogs from a naturally CanL enzootic cohort were screened for Borrelia burgdorferi, Anaplasma spp., and Ehrlichia spp. serology via IDEXX 4Dx SNAP test (IDEXX Reference Labs Inc.) and Leishmania serology via CVL DPP. DNA was isolated from whole blood and analyzed for the presence of Leishmania DNA via RT-qPCR described below. Physical exam was performed by licensed veterinarians for clinical signs of leishmaniosis (16).

Fifty dogs were enrolled based on the following inclusion criteria: a positive Leishmania diagnostic test (DPP or qPCR) or Leishmania diagnostic positive dam or full sibling (DPP or qPCR); a seronegative 4Dx SNAP test; two or fewer physical clinical signs of CanL. Dogs were excluded if they met any of the following criteria: negative Leishmania diagnostic test results and no Leishmania diagnostic positive dam or full sibling; a CVL DPP serology value >200; three or more physical clinical signs of leishmaniosis; a positive 4Dx SNAP test. Previous work in the same enzootic population identified that dogs born to dams ever positive for Leishmania had a 13.8-fold greater risk of becoming diagnostically positive themselves (P = 0.0002) (17).

Study design

Cohort immune responses were followed over 18 months from April 2019 to November 2020 (Fig. S1). At 3 month time points (Fig. 1), whole blood and serum were collected, and a physical exam was performed to clinically stage disease. L. infantum serology, L. infantum blood parasite burden, and Borrelia burgdorferi, Anaplasma spp., and Ehrlichia spp. serology via IDEXX 4Dx SNAP test (IDEXX Reference Labs Inc.) were performed. After the conclusion of the main study period, remaining subjects were visited in the following year (2021) by the research team to evaluate progression and mortality outcomes. At the time of final analysis for this manuscript, dogs that never had positive PCR or serological Leishmania diagnostic results were censored from analyses (n = 9).

Fig 1.

Prospective study flow chart. A cohort of 50 dogs with asymptomatic Leishmania infantum exposure and seronegative for tick-borne pathogen exposure via 4Dx SNAP test was followed for 18 months. Evaluations were performed at each major time point (TP) occurring every 3 months, while checkpoints occurred monthly on non-major time point months. *Serum chemistry panel and complete blood count were not conducted at time point 2. After the main study conclusion, dogs were visited intermittently over the following 2 years for follow-up on long-term leishmaniosis outcomes.

To control for highly inflammatory bacterial TBC on immunity, dogs were block randomized by sex and age into two treatment groups receiving either oral tick prevention (Sarolaner, Zoetis Inc.) or placebo (Fig. S1). Treatment was administered monthly in a double-blinded manner; caretakers, veterinarians, and researchers were blinded to treatment status throughout the study.

Clinical staging

Clinical leishmaniasis was staged according to LeishVet staging system guidelines (2). Venous blood was drawn at 3 month intervals, except for time point 2. Complete blood count and serum chemistry panels were performed by IDEXX Reference Laboratories (Westbrook, ME). Physical exam findings, complete blood counts, and serum chemistry panel results were systematically used to assign LeishVet stage (1–4). Dogs without any physical signs of disease or clinicopathological abnormalities were assigned a stage of 0 or “healthy.” An overview of the clinical staging system applied is depicted in Table S1.

After the conclusion of the main study period, the research team was able to visit the remaining subjects at irregular time points in the following 2 years. At these post-study follow-up visits, physical examination, Leishmania serological and PCR diagnostic testing were performed, and caretaker-provided health history was obtained. Bloodwork and LeishVet staging were not performed at post-study time points. Study subjects were assigned a qualitative designation of mild or severe disease. Study dogs that underwent owner-elected euthanasia due to leishmaniosis during this time period were considered as a severe CanL outcome.

Parasites and standard curve

For preparations of total Leishmania antigen and standard curves, Leishmania infantum promastigotes (US/2016/MON1/FOXYMO4), originally isolated from a hunting dog with CanL, were cultured in complete hemoflagellate-modified minimal essential medium with 10% heat-inactivated fetal calf serum at 26°C (18). Stationary-phase parasites were enumerated on a hemacytometer and adjusted to 1 × 10⁸ parasites per milliliter in phosphate buffered saline (PBS). Ten-fold serial dilutions were made, and 10 µL of each dilution was spiked into 1 mL of Leishmania-negative control canine blood to create an exponential regression standard curve where circulating parasite load is calculated as parasite equivalents per milliliter canine blood (Fig. S2).

Quantitative PCR

DNA was isolated from 200 µL of canine peripheral blood at each major study time point with the Gentra Puregene Tissue Kit (Qiagen). For L. infantum detection by real time-qPCR, Leishmania small-subunit rRNA-specific probe (5´-[6-FAM]-CGGTTCGGTGTGTGGCGCC-MGBNFQ-3′) and primers (forward, 5′-AAGTGCTTTCCCATCGCAACT-3′; reverse, 5′-GACGCACTAAACCCCTCCAA-3′) were used as previously described (19).

Preparation of Leishmania antigen

Leishmania antigen was prepared after washing stationary-phase L. infantum promastigotes in cold PBS twice, followed by multiple freeze-thaws in 5 mM CaCl₂ and 10 mM Tris, pH 7.4. Lysed parasite solution was either used as is (total Leishmania antigen, TLA) or centrifuged at 5,000 relative centrifugal force (RCF) for 20 min at 4°C and supernatant was collected (soluble Leishmania antigen, SLA). Protein content of lysate or supernatant was quantified via Pierce BCA protein assay (Thermo Scientific) and aliquots frozen at −30°C.

Leishmania Enzyme-Linked Immunosorbent Assay (ELISA)

For indirect ELISA, 96-well flat bottom Enyzme Immunoassay (EIA)-treated plates were coated with 2 µg/mL SLA in bicarbonate buffer (pH 9.6) overnight, blocked with 1% wt/vol bovine serum albumin + 0.05% wt/vol Tween-20 in PBS, and probed with canine sera diluted 1:500 in blocking buffer. All samples and controls were run in duplicate. Peroxidase-conjugated AffiniPure anti-dog IgG (H + L) (Jackson ImmunoResearch) was diluted 1:20,000 in blocking buffer, and OptEIA TMB substrate (BD Biosciences) was used to detect peroxidase activity. The reaction was stopped with 2 M H₂SO₄. Absorbance was read at 450 nm and 650 nm on a spectrophotometer plate reader.

Serum samples from five CanL-negative dogs, from a non-endemic area, were included on each plate. A cut-off was calculated from the average absorbance of the negative controls plus three standard deviations. Test serum samples were considered positive if their average absorbance was greater than the cut-off. SLA ELISA results are reported as test sample absorbance ratio over the control cut-off. An SLA ELISA ratio >1 is considered positive.

Peripheral blood mononuclear cell (PBMC) stimulation assay

PBMCs were isolated from whole blood via density centrifugation over Ficoll-Paque PLUS (Cytiva, Fisherbrand). PBMCs were plated in 96-well round-bottom plates at 5 × 10⁵ cells per well in RP-10 media (Roswell Park Memorial Institute [RPMI] 1640 supplemented with 10% fetal bovine serum [FBS], 2 mM L-glutamine, penicillin-streptomycin, and non-essential amino acids). A subset of PBMCs was stained with carboxyfluorescein succinimidyl ester (CFSE) prior to stimulation. CFSE labeling was performed in the dark. CFSE was prepared at 5 µM in PBS. PBMCs at 5 × 10⁶ cells per milliliter were mixed with an equal volume of CFSE solution (final staining concentration 2.5 µM), incubated at room temperature for 5 min, and quenched with RP-10 for 5 min at room temperature.

PBMCs were stimulated with media, 10 µg/mL TLA, 1 µL/well canine distemper virus vaccine antigen (distemper, hepatitis, parainfluenza, parvovirus (DHPP), a relevant antigen-positive control), or 5 µg/mL Concanavalin A (ConA). Supernatants were collected from PBMCs receiving each stimulus after 72 h. Flow cytometry was performed on PBMCs stimulated with media, TLA, or DHPP for 7 days and with ConA for 3 days. For intracellular cytokine staining, 10 µg/mL brefeldin A (Sigma-Aldrich) was added 6 h prior to harvest.

Cytokine ELISAs

Supernatants from PBMCs restimulated with media, TLA, DHPP, or ConA as described above were collected after 72 h Sandwich ELISA was performed on supernatants using canine specific IFN-γ, interleukin-10 (IL-10), and tumor necrosis factor (TNF) DuoSet ELISA kits according to the manufacturer instructions (R&D Systems).

Flow cytometry

Cells were blocked with 50% FBS/PBS containing 0.9 µL/well dog gamma globulin for 15 min (Jackson ImmunoResearch). Surface-labeling antibodies were prepared in block solution: anti-canine CD4-AF647 (1:250, BioRad antibodies), anti-canine CD8-AF700 (1:250, BioRad antibodies), biotinylated anti-human programmed death 1 (PD-1) (1:40, R&D) and anti-human CD49d-BV605 (1:50, BioLegend Inc). Streptavidin-PE-Cy7 secondary label was used according to manufacturer instructions (BioLegend Inc). Cells were labeled on ice in the dark for 30 min. Samples were fixed for 15 min in fixation buffer (BioLegend Inc), washed with PBS, and stored at 4°C protected from light.

For intracellular cytokine staining (ICS), fixed cells were permeabilized in 1× perm/wash solution for 15 min at room temperature in the dark. Zenon conjugation kits were used to label anti-canine IFN-γ-R-PE (5 µg/mL, R&D Systems) and anti-canine IL-10-AF488 (5 µg/mL, R&D Systems) antibodies (ThermoFisher Z25255 and Z25002, respectively) according to manufacturer instructions. Cells were labeled on ice for 30 min in the dark. Cell events were acquired within 48 h on a Becton Dickinson LSR II flow cytometer, and data were analyzed using FlowJo software (Fig. S3 and S4).

Statistical analyses

Longitudinal modeling is described in the following section. All other statistical analyses were performed using GraphPad Prism software version 10.0.3. Statistical tests applied can be found in the figure legend for each comparison. Normality tests were applied to determine if data should be analyzed with parametric or nonparametric tests. If any normality test indicated non-normality, the test was performed as nonparametric. Unless indicated, outliers were not removed from analyses. Where indicated in the figure legend, outliers were identified and removed using the regression and outlier removal (ROUT) test.

Bayesian linear mixed-effects model

To better understand the temporal changes of the overall immune response, comorbid TBC, and clinical outcome by the different immune measurements, we used a Bayesian linear mixed-effects model. The primary goal of this model was to estimate within-subject changes relative to TBC seroconversion; thus, the model intentionally focused on time-centered effects rather than between-subject differences. Subject-specific random intercepts account for all stable individual-level characteristics (including unmeasured demographics), effectively serving as matched controls for pre-TBC baseline, while the fixed effect of TBC seroconversion isolates within-subject changes. Bayesian models including demographic information are considered in previous works by members of this research team (20, 21).

The time point of TBC seroconversion of each TBC+ study subject was denoted $t_S$. The analysis included six study time points for each TBC+ subject, including $t_s$, 3 and 6 months prior to $t_S$, and 3, 6, and 9 months post $t_S$. Study variables were normalized to mean 0 and variance 1 to facilitate interpretation of effect sizes across different measurement scales. LeishVet status (ordinal) was treated as a numeric variable to model its linear relationship with TBC seroconversion time. This parameterization allows the effect of TBC seroconversion to be interpreted consistently across all outcomes, regardless of their original units of measurement.

Given $Y_{ij}$ represents a study variable of interest for subject $i$ at time $j$, individual immunopathogenic variables were modeled using the following specification:

$$Y_{ij}^{\left(1\right)}\sim N\left({\beta }_0+b_{0i}+{\beta }_{1}I\left(j\ge t_s\right),{\sigma }_1^2\right)$$

where ${\beta }_0$ denotes the overall intercept (fixed effect), $b_{0i}$ the subject-specific random intercept, ${\beta }_1$ the effect of TBC+ time on $Y_{ij}^{\left(1\right)}$, and ${\displaystyle {\in }_{ij}}$ the error terms such that ${ϵ}_{ij}\sim N(0,{\sigma }_1^2)$ and $b_{0i}\sim N(0,{\sigma }_b^2)$. ${\sigma }_1^2$ and ${\sigma }_b^2$ denote population and individual-specific variances, using inverse-gamma prior distributions. ${\beta }_0$ and ${\beta }_1$ used normal prior distributions. The mean structure included the indicator function $I\left(j\ge t_s\right)$, which assigned a value of 1 at time points including or following $t_S$, and 0 at time points preceding $t_S$.

Missing data were handled using Bayesian imputation (22) via the JointAI package in R (23), which simultaneously fits the mixed model while imputing missing values under the missing-at-random assumption. This approach accounted for all types of missingness, whether before TBC seroconversion (e.g., absent baseline measurements), after seroconversion (e.g., dropouts), or intermittent gaps (e.g., a single missing LeishVet stage at timepoint 2). JointAI leverages the model structure itself to inform imputations, preserving longitudinal correlations by borrowing information from observed values from the same subject at other time points and patterns across the full cohort. Imputations were updated iteratively during Markov Chain Monte Carlo (MCMC) sampling, ensuring consistency with both the observed data and the hierarchical model structure.

Three MCMC simulations of 25,000 iterations and 5,000 iterations as a burn-in period were run, for a total of 20,000 iterations on each chain. To assess parameter convergence, we used the Gelman-Rubin diagnostic from the coda package (24) in R. All parameters in the model reached convergence.

With this model specification, we are interested in the posterior distribution of ${\beta }_1$, the impact of comorbid infection on clinical status, and the negative or positive direction of the posterior probability. We evaluate the strength of the evidence for an association between comorbid infection and clinical status through ${\displaystyle \mathrm{m}\mathrm{a}\mathrm{x}(P({\beta }_1>0\mid \cdot ),P({\beta }_1<0\mid \cdot ))}$, the higher one-tail probability. This probability is the criterion used to assess the strength of evidence for the corresponding hypothesis that TBC modulates each study variable being modeled. We considered the posterior probability (PP) of the TBC effect (positive or negative) on a given variable of interest as moderate if the PP lies in the interval $[0.65,0.85]$, while a PP in the interval $[0.85,1]$ was considered a strong effect. Table S1 shows all PP results.

RESULTS

Cohort inclusion and study design

For this prospective CanL cohort study, 159 subjects were screened and excluded based on listed exclusion criteria (Fig. 1). This excluded subjects with clinically apparent CanL and subjects exposed to current or recent TBC. Of 92 candidates, 50 dogs were enrolled in the study: 24 dogs with positive Leishmania diagnostic testing confirmed through serological or molecular testing, and 26 dogs that had either a dam or full sibling with a positive Leishmania diagnosis. Together, these inclusion and exclusion criteria resulted in a cohort of dogs with confirmed or highly suspected in utero Leishmania exposure but not presenting clinical CanL or comorbid TBC at the time of enrollment.

The enrolled cohort was followed over the course of 18 months between 2019 and 2020, with physical examination by Petersen Lab veterinarians monthly and blood collection for immunoassays and diagnostics every 3 months (Fig. 1, Fig. S1A). This cohort was comprised of 52% male and 48% female dogs, 42% aged 0–2 years, 48% aged 3–5 years, and 10% aged 6 years or greater (Fig. S1B). Throughout the course of the study, nine subjects had partial missing data for CanL-related (n = 3) or non-CanL-related circumstances such as ran away, not present on location during site visit, and behavioral issues (n = 6). This loss was determined to be at random.

Remaining subjects were assessed for long-term CanL outcomes by physical examination, Leishmania serological and PCR diagnostic testing, and caretaker-provided health history at intermittent follow-up visits over the 2 years following the study conclusion. Thirty-eight dogs had a positive Leishmania diagnosis at the end of the trial and follow-up monitoring period. Bloodwork and LeishVet staging were not performed at long-term outcome follow-up visits. Study subjects were assigned a CanL clinical outcome of mild or severe, which included dogs experiencing mortality due to leishmaniosis.

Twelve subjects were censored from data analysis due to never testing positive on a Leishmania diagnostic test (n = 9) or were lost to follow-up for non-Leishmania related reasons and did not have sufficient data (n = 3). Therefore, data from a total of 38 confirmed Leishmania-positive canine subjects were included in the analyses presented (Fig. 1).

Tick-borne coinfections exacerbate clinical presentation and CanL severity

At the end of the study, 19 dogs tested TBC-positive on the IDEXX 4Dx SNAP test at some point during the study period (Ever TBC+, Fig. S1D). We wanted to establish the extent to which TBC impacted CanL serological response and parasitemia, as measured by ELISA, qPCR, and disease severity. When comparing subject-matched first and final measurements for each group, we observed that both TBC-unexposed (P = 0.031) and TBC-exposed (P = 0.008) CanL dogs experienced a significant increase in blood parasite burden across the study period, demonstrating natural progression of CanL (Fig. 2A). Anti-Leishmania serological reactivity also increased significantly in both TBC-unexposed (P = 0.012) and TBC-exposed (P < 0.001) dogs across the study period (Fig. 2C). At the study conclusion, a larger proportion of TBC+ dogs had become parasitemic or seropositive for Leishmania compared to TBC− dogs (Fig. 2B and D), but this difference was not statistically significant.

Fig 2.

Significant progression of CanL in dogs with TBC exposure. L. infantum qPCR (A), SLA OD ratio (C), and LeishVet stage (E) of TBC− or TBC+ subjects. Wilcoxon matched-pairs signed-rank test (A, C, E). Subject matched initial time point value (white) compared to endpoint value (black) of blood parasite burden (B), serological status (D), LeishVet stage (F), or clinical outcome at post-study follow-up time point (G) between TBC− or ever TBC+ subjects. Chi-squared contingency analysis of parasitemia status by RT-qPCR. Chi-squared test P-value and relative risk (RR) with 95% confidence interval. TBC−, subjects that never experienced a TBC. Ever TBC+, subjects that were TBC+ at any point during the study. (H–K) Kinetics of Leishmania diagnostics among TBC-positive subjects at 3 month intervals preceding and following the time point of TBC seroconversion. PP of change in behavior after TBC seroconversion is indicated. *PP 0.65–0.85, moderate effect. **PP ≥0.85, strong effect. (H–J) Mean and SD shown.

Using the LeishVet clinical severity scale, dogs with subclinical CanL exposed to a TBC experienced a greater increase in LeishVet stages across the study period as compared to those unexposed (Fig. 2E). At their final study assessment, TBC+ dogs were at twice the risk of being assigned a LeishVet stage of 2 or higher compared to TBC− dogs (relative risk = 2.0, P = 0.051) (Fig. 2F). Furthermore, at follow-up visits, dogs that experienced a TBC during the study period (TBC+ dogs) had double the risk of developing severe disease or undergoing euthanasia due to leishmaniosis. Dogs that remained TBC− during the study period were more likely to remain subclinical or maintain mild disease (P = 0.022, relative risk = 2.0) (Fig. 2G). Together, these data support the conclusion that dogs with subclinical CanL experiencing a TBC during the study period significantly increased the likelihood of progression to clinical CanL by the completion of the study.

Using Bayesian linear mixed-effects modeling, we focused specifically on TBC+ dogs within our cohort and modeled how longitudinal outcomes varied between pre-TBC baseline and post-TBC trajectories, considering the first incidence of TBC+ test result as an inflection point and comparing the change in variable behavior prior to coinfection versus after the tick pathogen exposure event. The resulting PP of the model indicates the direction (increase, decrease, or no change) of the behavior kinetics across the coinfection time point and the magnitude of the probability of this behavior trend compared to pre-coinfection time points.

Applying this statistical model to TBC+ dog data, parasite burden and anti-Leishmania serological reactivity showed a moderate to strong increasing trend following the TBC event compared to the 6 months preceding TBC (Fig. 2H–K). This supports TBC as a triggering event for observed increased parasitemia and serological OD.

Tick-borne coinfections alter Leishmania antigen-induced cytokine profile

Control of leishmaniasis is mediated by CD4⁺ T cells producing IFN-γ to augment microbicidal activity of parasitized phagocytes (5–7). To counter this, Leishmania has evolved to induce numerous regulatory pathways that hamper pro-inflammatory responses. IL-10 induction during leishmaniosis is well documented in vitro and in vivo (8, 25). IL-10 antagonizes the effects of IFN-γ and other inflammatory cytokines, dampening damage from IFN-γ, but chronic antigen exposure leads to high levels of IL-10, antibody production, and high parasite burden (26). To assay cytokine production in our cohort, PBMCs from TBC+ CanL dogs were restimulated with Leishmania antigen (TLA) ex vivo. IFN-γ and IL-10 concentrations were measured in supernatants by ELISA (Fig. 3A and B), and cytokine-producing CD4⁺CD49d^hi lymphocytes were detected by intracellular staining and flow cytometry (Fig. 3C and D).

Fig 3.

Leishmania antigen-stimulated IFN-γ and IL-10 production following TBC. (A–F) TBC+ dogs only. Kinetics of TLA-stimulated IFN-γ (A, C) and IL-10 (B, D) production among all TBC+ dogs at 3 month intervals preceding and following the TBC seroconversion time point. (A, B) Cytokine concentration from stimulated PBMC supernatant via ELISA. (C, D) Frequency of cytokine-expressing cells among TLA-stimulated CD4⁺CD49d^hi lymphocytes by ICS. Median with 95% CI shown. Asterisks correspond to Bayesian model posterior probability. (E) Posterior probability of change after TBC seroconversion indicated. *PP 0.65–0.85, moderate effect. **PP ≥0.85, strong effect. (F) Spearman correlation coefficient for each combination. Bolded values indicate P ≤ 0.05. Heatmap shows red for strongest positive correlation coefficients and blue for strongest negative correlation coefficients. (G–I) Endpoint measurement percent IFN-γ or IL-10-positive CD4⁺CD49d^hi lymphocytes or ratio from TLA- or ConA-stimulated intracellular staining in TBC− dogs, dogs TBC+ for ≤6 months or ≥12 months. Kruskal-Wallis with Dunn’s test or one-way analysis of variance with Holm-Sidak test. Mean and SD provided. Outliers removed from G–I.

Within TBC+ dogs, we observed a moderate increase in IFN-γ secreted by bulk PBMCs in response to TLA following the TBC conversion event. However, IFN-γ supernatant concentration became more variable with median amounts trending downward after TBC conversion (Fig. 3A and E). IFN-γ produced by TLA-stimulated PBMCs negatively correlated with parasite burden among TBC+ dogs (Fig. 3F). TLA-induced IL-10 levels in bulk PBMC supernatants remained relatively steady up to 9 months post-TBC (Fig. 3B).

CD49d integrin expression differentiates antigen-experienced T lymphocytes from naïve cells (27). The frequency of IFN-γ-producing CD4⁺CD49d^hi T cells did not significantly alter from the months prior to coinfection to post-TBC (Fig. 3C and E). The proportion of IL-10⁺ CD4⁺CD49d^hi T cells moderately declined following the co-exposure event (Fig. 3D and E) and significantly negatively correlated with circulating parasite burden (Fig. 3F).

Following seroconversion, TBC+ dogs remained seropositive to tick pathogen antigen for variable periods of time. Of the 19 dogs that experienced TBC, 11 dogs were seropositive for only 6 months or less, and 8 dogs were seropositive for 12 months or more. Extended seropositivity could indicate ongoing coinfection or re-exposure. We hypothesized that a prolonged coinfected state would have a more significant impact on the anti-Leishmania immune response. Thus, we examined cytokine expression in dogs experiencing shorter versus more chronic coinfection periods. In Fig. 3G through I, we assessed endpoint cytokine production from TBC−, ≤6 months TBC+, and ≥12 months TBC+ dog cells (Fig. 3G through I). To better visualize patterns between these groups, outliers were removed from comparisons of TBC duration.

At endpoint measurements, in response to TLA, ≤6 months TBC dogs showed a significant increase in % IFN-γ⁺ CD4⁺CD49d^hi T cells compared to TBC− dogs (Fig. 3G). However, among dogs TBC+ for long-term periods (≥12 months), CD4⁺CD49d^hi T cells showed a significant increase in %IL-10 production (Fig. 3G). In response to the mitogen ConA, we observed a non-significant trend of decreased IFN-γ-producing cells among long-term TBC+ dog CD4⁺CD49d^hi T cells (Fig. 3H).

The ratio of %IFN-γ⁺/%IL-10⁺ CD4⁺CD49d^hi lymphocytes among long-term TBC+ dog cells also decreased non-significantly compared to TBC− dog cells at endpoint measurements (~1.9-fold reduction in response to TLA and ~1.8-fold reduction in response to ConA) (Fig. 3I). Together, these data may indicate dogs co-exposed to L. infantum and TBC for extended time periods are undergoing a potential shift to a more regulatory profile.

Relationship between PD-1 expression and proliferation kinetics among coinfected dogs

In addition to active immune evasion by Leishmania parasites, T cell exhaustion can occur in settings of chronic antigen exposure and T cell activation such as visceral leishmaniasis, leading T cells to become less responsive to stimuli (28). Therefore, we measured surface PD-1 expression, a key marker of T cell activation but also exhaustion (29), on CD4⁺ and CD8⁺ T cells in relation to proliferation in response to Leishmania antigen. In dogs with CanL, TLA-stimulated CD4⁺CD49d^hi T cell proliferation (%CFSE^lo) showed a strong probability of decreasing following the tick pathogen seroconversion time point (Fig. 4A and E) and positively correlated with CD4⁺CD49d^hi T cell PD-1 expression across time points following TBC (r = 0.73, P < 0.001, Fig. 4F). Looking more specifically at endpoint measurements, this correlation remained strongly significant in all dog groups TBC−, ≤6 months TBC+, and long-term TBC+ (Fig. 4I through K). Our results show that following TBC in CanL dogs, CD4⁺CD49d^hi cell proliferation decreases in a subject-matched, longitudinal model; however, PD-1 expression by these cells was highly correlated and thus does not explain this decrease, indicating other regulatory mechanisms are contributing to this decreased proliferative phenotype.

Fig 4.

Leishmania antigen-stimulated CD4⁺ T cell proliferation and PD-1 expression decreased after TBC, while markers of CD8⁺ T cell activation increased. (A–E) Kinetics of TLA-stimulated proliferation and PD-1 expression as percent of CD4⁺CD49d^hi (A, B) or CD8⁺CD49d^hi (C, D) lymphocytes among TBC+ dogs at 3 month intervals preceding and following the time point of TBC seroconversion. Median and 95% CI shown. PP of change in behavior after TBC seroconversion is indicated. *PP 0.65–0.85, moderate effect. **PP ≥0.85, strong effect. (E) PP overview table. (F) Spearman correlation matrix of TBC+ dogs including all post-TBC measurements. Bolded values indicated P < 0.05. Correlation coefficient shown. Heatmap shows red for strongest positive correlation coefficients and blue for strongest negative correlation coefficients. (G–M) Endpoint measurements, outliers removed. Endpoint TLA-stimulated proliferation and PD-1 expression as percent of CD4⁺CD49d^hi (G, H) or CD8⁺CD49d^hi (L, M) lymphocytes among TBC−, dogs TBC+ for ≤6 months or ≥12 months. Mean and SD shown. One-way analysis of variance with Hold-Sidak post-test. (I–K, N–P) Correlation between TLA-stimulated proliferation and PD-1 expression as percent of CD4⁺CD49d^hi (I–K) or CD8⁺CD49d^hi (N-P) lymphocytes among TBC− dogs or TBC+ dogs. Spearman correlation coefficient (R) and P-value are shown. Linear regression and 95% CI are depicted.

Previous evaluation of CD8⁺ T cell responses to chronic Leishmania infection and disease indicated that these cells appeared to have an exhausted phenotype earlier in clinical progression compared to CD4⁺ T cells (7). How TBC and the likely highly inflammatory environment produced by this coinfection exposure alters CD8⁺ T cell responses during leishmaniosis has not been explored. Unlike their CD4 counterparts, TBC+ dog CD8⁺CD49d^hi T cell proliferation and PD-1 expression in response to TLA stimulation both moderately increased after the TBC seroconversion event (Fig. 4C through E) and were not significantly correlated considering all TBC+ time points (Fig. 4F). At endpoint measurements, chronic TBC+ dog CD8⁺CD49d^hi T cell proliferation in response to TLA was significantly higher than in TBC− dogs, while acutely TBC+ dogs showed an intermediate level of proliferating cells (Fig. 4L). CD8⁺CD49d^hi T cell proliferation did not significantly correlate with the level of PD-1 expression at the endpoint measurements in any groups. Together, these data indicate differential effects of TBC on antigen-induced proliferation of CD4⁺ vs CD8⁺CD49d^hi T cells. CD4⁺ T cell proliferation transiently but strongly decreases after TBC seroconversion, while CD8+ T cell proliferation increases and is sustained over time post-seroconversion.

To examine clinical diagnostic and TLA-stimulated immune readout relationships post-coinfection, Fig. S6 shows a correlation matrix and corresponding P-values, including the TBC seroconversion and subsequent time points. We wanted to establish whether the effect of increasing proliferating CD8 T cells was specific to the TBC+ dog group or applied to the entire cohort. Therefore, in Fig. 5, we performed correlation analysis between proliferating CD4 or CD8 T cells and Leishmania parasitemia vs serological results at the endpoint measurements including all subjects (Fig. 5). Indeed, we observed that TLA-stimulated %CFSE^lo CD8⁺CD49d^hi T cells were positively correlated with Leishmania blood burden (Fig. 5A, P = 0.045) and Leishmania serological OD ratio (Fig. 5B, P = 0.015) but not LeishVet stage. Enhanced proliferation among CD8⁺CD49d^hi T cells in this CanL cohort was associated with higher blood parasite burden.

Fig 5.

Leishmania antigen-stimulated CD8⁺ T cell proliferation significantly correlates with increasing Leishmania burden and serology. Correlation between endpoint measurement TLA-stimulated %CFSElo CD4⁺CD49d^hi (triangles, dotted line) or CD8⁺CD49d^hi (circles, solid line) lymphocytes and Leishmania parasite burden (A), serology (B), or LeishVet stage (C) including all study subjects. Spearman correlation coefficient (R), statistical result, and linear regression shown.

Impact of coinfection on long-term clinical outcomes

Finally, we tested if our immune readouts could delineate dogs that went on to develop severe leishmaniasis or mortality due to leishmaniosis. After the conclusion of the immunological study period, remaining subjects were visited for follow-up CanL clinical evaluation. We stratified the cohort by mild or severe CanL at follow-up assessment and by TBC status during the main study period (Fig. 6). At the endpoint study measurement, dogs that developed severe CanL already displayed a signature of significantly increased blood parasite burden, anti-Leishmania serology, and LeishVet stage compared to dogs maintaining mild disease (Fig. 6A through C). Additionally, both antigen-stimulated IFN-γ and, to a lesser extent, IL-10 secreted by PBMCs were already significantly reduced at the end of the study in dogs going on to develop severe CanL (Fig. 6D and E).

Fig 6.

Decreased cytokine production and a trend of increased CD8⁺CD49d^hi T cell proliferation prior to the development of severe CanL. Comparison of endpoint study measurements among dogs that went on to develop severe CanL and/or were euthanized due to leishmaniasis at post-study follow-up time points (severe) vs dogs that displayed mild disease at follow-up time points (mild) separated by their TBC status during the study period. Leishmania burden (A), Serology (B), LeishVet stage Clinical Severity (C), IFN-γ ELISA (D), IL-10 ELISA (E), %IFN-γ+ CD4⁺CD49d^hi cells (F) ,%IL-10+ CD4⁺CD49d^hi cells (G), %CFSElo CD4⁺CD49d^hi cells (H), %PD1+ CD4⁺CD49d^hi cells (I), %CFSElo CD8⁺CD49d^hi cells (J), and %PD1+ CD8⁺CD49d^hi cells (K). Outliers removed. P-values denoted between individual groups: Kruskal-Wallis with Dunn’s post-test or one-way analysis of variance with Holm-Sidak post-test. P-values denoted between mild and severe groups: unpaired t-test or Mann-Whitney U-test. Mean and SD shown.

When splitting the clinical outcome groups into dogs which did or did not experience TBC during the main study period, TBC+ dogs that maintained mild CanL disease showed significantly higher proportions of both IFN-γ⁺ and IL-10⁺ CD4⁺CD49d^hi T cells at the study culmination compared to TBC− dogs maintaining mild disease (Fig. 6F and G). However, TBC+ dogs that went on to develop severe CanL did not show enhanced CD4⁺ T cell cytokine expression, perhaps indicating immunosuppression associated with the severe CanL may already be occurring during the study in this group.

The proportion of proliferating and PD-1 expressing T cells trended to be increased among dogs developing severe disease post-study (Fig. 6H through K) compared to those maintaining mild CanL but did not show a significant difference driven by TBC status.

DISCUSSION

Cross-sectional studies showed dogs with CanL have a higher rate of coinfections with other vector-borne diseases than non-CanL dogs in Leishmania endemic areas (30–33). TBCs are highly prevalent canine coinfections in Leishmania-endemic countries due to the extensive distribution of tick vectors and range of associated tick-borne pathogens (15). Furthermore, TBCs have been associated with more severe clinicopathological abnormalities (Rickettsia conorii, E. canis, A. phagocytophilum and Bartonella henselae) and higher LeishVet clinical stage scores (A. phagocytophilum) (32). Whether dogs with CanL are more susceptible to TBCs due to Leishmania-induced immunosuppression or if TBCs trigger the evolution of underlying subclinical Leishmania infections is not clear. This study is unique as we were able to follow the natural development of CanL and associated immune parameters, alongside concurrent TBC under real-world conditions.

Over the course of this 18-month study, subclinical Leishmania-exposed individuals had significant increases in parasitemia, anti-Leishmania serum antibodies, and associated progressive clinical disease. Within the study period, 50% of subjects developed serum antibodies reactive against TBC antigens. Secondary infection with TBC led to more severe clinical presentation compared to TBC− CanL dogs. In addition, dogs positive for TBC during the study showed a significantly increased risk of mortality or progression to severe CanL following the study conclusion compared to TBC− dogs, indicating TBC had long-lasting impacts on the host (Fig. 2G). These results are consistent with findings from a previous prospective study associating TBCs with more severe CanL disease and mortality (13). In the TBC+ cohort, LeishVet stage (based off physical exam, complete blood count, and serum chemistry abnormalities), Leishmania blood parasite burden, and Leishmania serological reactivity values all strongly significantly correlated with one another, supporting that these readouts together act as a practical gauge of disease severity in this cohort.

In CanL enzootic areas of Europe and Brazil, E. canis often predominates, with up to 80% of dogs seropositive for E. canis in previous coinfection studies in Brazil (13). In the current study, Ehrlichia spp. accounted for 63% of TBCs and the majority of long-term TBC. Borrelia burgdorferi accounted for 37% of TBCs, and six out of seven Borrelia seropositive dogs were positive for ≤6 months (Fig. S1B). TBCs likely contribute to CanL progression via multifactorial processes.

Anaplasma and Ehrlichia species are both intracellular bacteria that infect and replicate within phagocytes systemically, the same cell types parasitized by Leishmania (34, 35). This significant overlap in affected tissues such as spleen, liver, and bone marrow may synergistically disrupt the normal function of these organs, accelerating clinical manifestations of disease. In a Brazilian study, dogs co-exposed to L. infantum and E. canis presented significantly more pronounced hematological and biochemical abnormalities including hypoalbuminemia, increased alkaline phosphatase, and lower hematocrit than dogs positive for Leishmania only (36). Intracellularly, tick-borne pathogens employ multiple immune evasion mechanisms. For example, Anaplasma and Ehrlichia species have been shown to inhibit lysosomal fusion, downregulate major histocompatibility complex (MHC) class II expression, decrease toll-like receptor (TLR) and mitogen activation protein kinase (MAPK) signaling, and prevent NADPH assembly (11). Pessôa-Pereira et al. found in vitro coinfection with L. infantum and Borrelia burgdorferi in a canine macrophage cell line resulted in significantly greater intracellular Leishmania infection rate and burden. This was associated with increased SOD2 mRNA expression and decreased mitochondrial reactive oxygen species activity (37). We hypothesize that TBCs facilitate Leishmania survival in part by hindering mechanisms controlling intracellular Leishmania replication.

Innate and adaptive cytokines induced by TBC may alter the effectiveness of the type 1 immune response controlling Leishmania replication in dogs maintaining asymptomatic leishmaniosis. Borrelia burgdorferi infection induces a mix of type 1 and type 17 inflammatory as well as regulatory IL-10 responses from canine and human cells (37–41). E. canis induces IFN-γ-secreting cells in dogs (42). Both IFN-γ- and IL-17-producing T cells occurred during canine infection with Ehrlichia chaffeensis (43). In mice, CD4⁺ T cell-derived IFN-γ was required for resistance against lethal Ehrlichia infection, while inflammatory CD8⁺ T cells contributed to pathogenesis (44, 45). IL-10 is upregulated transiently during acute time points (46, 47).

Together, these studies show TBCs can induce a mix of type 1, type 17, and IL-10 responses, and other innate inflammatory cytokines may occur. We speculate the inflammatory milieu is altered by TBC systemically and at focal sites of bacterial infection, in dogs concurrently infected with Leishmania parasites. Skewing away from type 1 mediator production, especially in an environment with ongoing IL-10 production such as CanL (8), could reduce the ability of IFN-γ to activate macrophages to control Leishmania replication. Meanwhile, excess co-stimulation by innate inflammatory cytokines in settings of chronic antigen presentation, which occurs during CanL, is known to promote the onset of T cell exhaustion (28). In addition, another prospective study found dogs with clinical CanL are at a significantly increased risk of contracting an Ehrlichia coinfection. Thus, a chicken and egg feedback scenario between TBC and CanL likely alters the host immune environment to promote the pathogenesis of and susceptibility to both organisms (30).

To uncover signs of potential immune exhaustion in coinfected dogs, we assayed T cell IFN-γ and IL-10 production, proliferation, and PD-1 expression. We saw a transient increase in the amount of TLA-stimulated IFN-γ produced by bulk PBMCs 3 months after TBC seroconversion, which declined progressively at 6 and 9 months following TBC (Fig. 3A). Similarly, CD4⁺CD49d^hi T cells showed an increased frequency of TLA-stimulated IFN-γ production in dogs acutely TBC, but the increase subsided in dogs with chronic TBC exposure (Fig. 3G). Transiently increased IFN-γ production post-TBC may be driven in response to the bacterial infection directly in the form of a bacteria antigen-specific cellular response, or in a bystander fashion due to Toll-like receptor activation or cytokines acting on existing effector T cells nonspecifically (48).

By 12 months post-TBC, CD4⁺CD49d^hi T cells showed enhanced IL-10 production in response to TLA (Fig. 3G). Overall, we observed a trend of lower IFN-γ/IL-10 production by CD4⁺ T cells in response to both Leishmania antigen and mitogen after a prolonged period of TBC compared to TBC− dogs. Other studies have observed lower or unchanged IFN-γ output in response to Leishmania antigen during active VL or CanL using various antigen stimulation and molecular readout techniques (7, 49–52). During active human VL, Leishmania antigen-stimulated PBMCs produce IL-10 but low IFN-γ, which recovers after treatment (53).

Statistical modeling of subject-matched longitudinal data showed TBC is followed by moderate to strong probability of decreased CD4⁺ T cell proliferation in response to TLA (Fig. 4). This was highly positively correlated with CD4⁺ T cell PD-1 expression in both TBC− and TBC+ subjects, indicating that at this stage of CanL, PD-1 expression is serving as a marker of CD4⁺ T cell activation. The decrease in CD4⁺ T cell proliferation in our TBC cohort is not explained by a concomitant increase in PD-1 expression. Other studies have reported high cytotoxic T lymphocyte associated protein 4 (CTLA-4) and PD-1 expression on CD4⁺ T cells during VL. While cells still produce IFN-γ, blockade of inhibitory receptor-ligand interactions in infected macrophage co-cultures enhanced their ability to induce Leishmania killing (7, 54), indicating these receptors are regulating functional capability. CD4⁺ T cell proliferation recovered after ~9 months post-TBC, and a difference was not observed at the endpoint compared to TBC− dogs. We did not assay the presence of CD4⁺ T cells responsive to tick pathogen antigens in this study.

In contrast to CD4⁺ T cells, TBC was associated with a moderate to strong probability that CD8⁺ T cells display increased proliferation and PD-1 expression in response to TLA following seroconversion, although these two factors were not significantly correlated (Fig. 4). CD8⁺ T cell proliferation in response to TLA in coinfected dogs was significantly higher than seen in non-coinfected dogs at the endpoint measurements and was highest in dogs coinfected for longer periods of time (Fig. 4L). Thus, TBC was associated with progressively increasing Leishmania-specific CD8⁺ T cell proliferation. We also saw that increased CD8⁺ T cell proliferation in response to Leishmania antigen correlated with increased parasite burdens and higher amounts of circulating anti-Leishmania antibodies (Fig. 5), both associated with worsening disease. Cortese et al. previously found a significant increase in CD8⁺ lymphocyte frequency in dogs with CanL (55). Interestingly, E. canis infection in dogs has also been associated with a significant increase in CD8⁺ T cells (56). CD8⁺ T cell proliferation in response to Leishmania antigen is exacerbated in dogs with tick-borne coinfections during CanL.

The roles of CD8⁺ T cells in human VL and CanL are not well studied compared to cutaneous leishmaniasis, where they contribute to both parasitic control and immunopathology through IFN-γ expression and cytotoxicity, respectively (57). It is thought that CD8⁺ T cells are more susceptible to immune exhaustion than their CD4⁺ counterparts, and CD8⁺ T cell dysfunction during VL has been previously reported (50, 57–59). In humans with VL, circulating CD8⁺ T cells expressed significantly higher levels of IL-10 mRNA and CTLA-4 and PD-1 protein than endemic control cells, and these markers were even higher in splenic CD8⁺ T cells than PBMCs, where parasite load is concentrated (50). Blocking PD-1 ligand in CanL dog PBMC cultures leads to increased CD8⁺ T cell proliferation (7). VL patient CD8⁺ T cells appear activated with a CD45RO-negative effector phenotype, expressing significantly higher levels of perforin and granulysin than endemic control cells, but did not contribute to Leishmania antigen-stimulated IFN-γ (50, 60). Additional inhibitory receptors lymphocyte activation gene-3 (LAG-3) and T cell immunoglobulin and mucin domain-containing-3 (TIM-3) have also been found enriched on VL patient CD8⁺ T cells (61). Together, in VL, CD8⁺ T cells exist in an activated state with robust inhibitory receptor expression, potentially making them more susceptible to immune exhaustion. This is supported by our observation of TBC-driven exacerbated CD8⁺ T cell proliferation during CanL.

After the completion of this 18 month study, we assessed members of the cohort for mild or severe CanL and/or dogs that underwent owner-elected euthanasia due to leishmaniosis. Endpoint study measurements of parasite burden, serological reactivity, and LeishVet stage were significantly higher in dogs that went on to develop severe disease outcomes, while TLA-induced IFN-γ and IL-10 production by bulk PBMCs were significantly reduced (Fig. 6). However, the level of IL-10 remained quite high in both groups. Among cellular responses, both CD4⁺ T cell PD-1 surface expression and CD8⁺ T cell proliferation trended to be higher in dogs that later developed severe disease.

Dogs in Leishmania endemic and enzootic areas experience high rates of exposure to additional vector-borne pathogens. Studying canine subjects in naturally infected states provides a more accurate but more complex representation of CanL host immune settings. Herein, we analyzed the clinical immune state of the subjects from the perspective of how TBC modulates CanL. We recognize that CanL also exerts modulatory effects on coinfecting species and immune responses. In addition, host-specific predispositions, external environmental pressures, and effects of sex and age all contribute to a web of interactions that is difficult, or unrealistic, to disentangle.

Whether the tick-borne pathogen or Leishmania parasite comes first, in this study, we observed tick-borne coinfections promote clinical progression in dogs with asymptomatic CanL and have complex effects on the anti-Leishmania immune response, particularly leading to alterations in the IFN-γ:IL-10 ratio and progressively increasing CD8⁺ T cell proliferation. More work needs to be done to explore the role of hyperactivation of CD8⁺ T cells and their dysfunction during VL. Preventing tick bites in subclinical L. infantum-infected individuals removes compounding effects of chronic exposure to TBC immune evasion mechanisms and inflammation that may contribute to the onset of immune exhaustion (62). Limiting the course of disease in dogs is important for reducing the risk of transmission to humans in endemic communities (10). Regular application of oral or topical acaricides, combined with environmental modifications and manual tick checks, is currently the most effective way to prevent TBCs at present (63). Clinically, tick preventatives should be incorporated into the management of subclinical L. infantum-infected dogs to prevent TBCs and reduce the risk of progression to clinical leishmaniosis.

ETHICS APPROVAL

All procedures involving animals during this study were approved by the University of Iowa Institutional Animal Care and Use Committee and performed under the supervision of licensed and, where appropriate, board-certified veterinarians according to International AAALAC accreditation standards.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/iai.00182-25.

Supplemental material. iai.00182-25-s0001.pdf.

Fig. S1 to S6; Tables S1 and S2.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.