Molecular detection of Ehrlichia canis and Rickettsia rickettsii in tick-infested dogs from the city of Chihuahua, Mexico: phylogenetic analysis of the gp36 and ompA genes

Abstract

Vector-borne diseases in dogs are an emerging problem worldwide due to their frequency, morbidity, and zoonotic relevance. Since there are no previous studies determining the frequency of Ehrlichia canis and Rickettsia rickettsii infection in three neighborhoods of the city of Chihuahua, Mexico, using molecular detection techniques, the objective of the present study was to detect, by nested polymerase chain reaction (PCR), a fragment of the gp36 gene of E. canis and the ompA gene of R. rickettsii in tick-infested dogs captured at the canine control center of the city of Chihuahua, and to phylogenetically analyze the genotypes present. Blood samples were collected from 123 tick-infested dogs. DNA was extracted from peripheral blood leukocytes and used in nested PCR protocols. The results showed a positivity frequency of 16.2% (20/123) for E. canis and 3.2% (4/123) for R. rickettsii. Six sequenced samples corresponded to E. canis and four to R. rickettsii. Phylogenetic analysis showed a close relationship with sequences previously described in the United States. This is the first molecular identification confirming E. canis and R. rickettsii infection in tick-infested dogs from some neighborhoods of the Chihuahua city, Mexico.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s11259-026-11442-7.

Introduction

Vector-borne infections in dogs are an emerging problem worldwide due to their frequency, morbidity, and zoonotic relevance (Sánchez Pérez et al. 2023). The brown dog tick, Rhipicephalus linnaei (R. sanguineus s.l. “tropical lineage”), is the main vector of pathogens such as Ehrlichia canis, Hepatozoon canis, and Rickettsia rickettsii, and is widely distributed in Latin America (Dantas-Torres et al. 2012; Šlapeta et al. 2022).

Ehrlichia canis (Anaplasmataceae) is an obligate intracellular Gram-negative α-proteobacterium with tropism for monocytes and macrophages, where it forms morulae (Mavromatis et al. 2006). It is the etiological agent of canine monocytic ehrlichiosis (CME), a severe and potentially fatal disease (Ebani 2026). Rickettsia rickettsii (Rickettsiaceae, spotted fever group) is the causative agent of Rocky Mountain spotted fever (RMSF), characterized by endothelial infection and multisystemic failure (Parola et al. 2013). Its vectors include R. linnaei on dogs and other ixodids such as Dermacentor andersoni, D. variabilis, and Amblyomma mixtum (Demma et al. 2005; Labruna et al. 2017).

In Mexico, a national sero-epidemiological study reported 33.1% seroprevalence of E. canis, with the highest values in northern states (Nuñez-Ochoa 2003). Molecular data on E. canis in Latin America remain scarce (Sosa-Gutierrez et al. 2016). For R. rickettsii, molecular identification in dogs from northern Mexico began in 2009 (Tinoco-Gracia et al. 2009; Castillo-Martínez et al. 2017). The objective of this study was to identify by PCR and sequencing the presence of E. canis and R. rickettsii in tick-infested dogs from Chihuahua and to phylogenetically analyze the gene fragments.

Materials and methods

This study was approved by the Internal Committee for the Care and Use of Experimental Animals (CICUAE) under protocol nC 26_11. All procedures followed the Mexican Official Standard NOM-087-ECOL-SSA1-2002.

The study was conducted on dogs captured at the canine control center of Chihuahua, Mexico, from three neighborhoods: 3 de Mayo (28°35’43.6"N 106°01’52.2"W), Villa Juárez (28°36’57.1"N 106°01’16.5"W), and Jardines de Oriente (28°39’13.5"N 105°57’47.8"W).

Blood samples were collected from 123 dogs. The sample size was determined by logistical constraints during the collection period and represents a convenience sample. All dogs were infested with ticks, showed no apparent clinical signs, and had not received antibiotic treatment. Four milliliters of whole blood were obtained by puncture of the cephalic or saphenous vein into heparinized tubes (BD Vacutainer^®, Mexico) using a 21-gauge needle (BD, Mexico) and transported to the laboratory under refrigeration. Samples were centrifuged at 3500 rpm for 15 min; plasma and buffy coat were separated, and peripheral blood leukocytes (PBL) were stored at − 70 °C until needed.

Some Ticks were morphologically identified as consistent with Rhipicephalus linnaei (R. sanguineus s.l. “tropical lineage”), the brown dog tick, however, no molecular confirmation of tick species was performed.

The study group included 66 females and 57 males; 63.4% were mixed-breed and 36.6% purebred. Age distribution was: 56 dogs < 1 year (45.5%), 26 aged 1–3 years (21.1%), 29 aged 3–6 years (23.6%), and 12 > 6 years (9.8%). Samples per neighborhood were: 67 (54.5%) from 3 de Mayo, 37 (30%) from Villa Juárez, and 19 (15.5%) from Jardines de Oriente.

DNA was extracted from PBL using the FavorPrep™ Tissue DNA Extraction Mini Kit (FavorGen^® Biotech Corp., Pingtung, Taiwan) following the manufacturer’s instructions. DNA was quantified at 260–280 nm using a NanoDrop Lite (Thermo Scientific, USA) and stored at − 70 °C until use. In order to verify the integrity of the genetic material, the hypoxanthine-guanine phosphoribosyltransferase 1 (HPRT1) gene was used as a housekeeping site using the primers and amplification conditions described by Ramos Santes (2016).

Primers for nested PCR are shown in Table 1. For E. canis, first-round primers targeting gp36 were those of Almazán et al. (2016); second-round primers were designed in this study. For R. rickettsii, both rounds targeting ompA were designed in this study using Primer3 input^®, BLAST^®, and Oligoanalyzer^®. In silico specificity was verified by BLASTn against the NCBI nucleotide database, confirming no significant similarity with non-target sequences.

Table 1.

Primer sequences used in the nested PCR to amplify gene fragments of E. canis and R. rickettsii and the size of the generated amplicon

Bacterium	Primer sequence (5′–3′)	Amplicon size	Reference
E. canis	C36-F1c: GTATGTTTCTTTTATATCATGGC	748 bp	Almazán et al. (2016)
E. canis	EC36-R1: GGTTATATTTCAGTTATCAGAAG
E. canis	Fwgp36-2-MSP: ATGGGTTATTGTATGCTTCATTT	435 bp	This study
E. canis	RVgp36-2-MSP: AGCAGAAACAGAATCTTCAGTAA
R. rickettsii	FwompA-1-MSP: GGCTAAAGTAACCGCACCTG	685 bp	This study
R. rickettsii	RvompA-1-MSP: CAAGACGTTTGGTGGAGCTC
R. rickettsii	FwompA-2-MSP: CAACACCGCTAACACTACCG	495 bp	This study
R. rickettsii	RvompA-2-MSP: TCAAAGGAGCGGGAGATTGT

The positive control for nested PCR for E. canis were DNA extracted from the spleen tissue of a dog naturally infected previously confirmed by Sanger sequencing and BLASTn analysis, showing identity only with sequences of E. canis. The positive control used for R. rickettsii was obtained from a PCR-positive PBL sample and sequenced by the Sanger method, confirming the amplicon specificity using BLASTn. These sequenced DNA samples were used as positive control in all subsequent PCR runs. Molecular-grade water was used as a negative control in every PCR run. All positive controls consistently yielded the expected amplicon size, and all negative controls were negative.

PCR mix: 5 µL DNA (500 ng), 1X buffer with 1.5 mM MgCl₂, 225 µM dNTPs, 600 nM each primer, 5 U/µL DNA polymerase, final volume 30 µL. Amplification reactions were performed in a thermal cycler (Thermo Fisher Scientific™, USA). Cycling conditions for E. canis: first round 95 °C/5 min, 40 cycles (95 °C/40 s, 45 °C/40 s, 72 °C/50 s), final extension 72 °C/10 min; second round: 95 °C/5 min, 40 cycles (95 °C/40 s, 50 °C/30 s, 72 °C/35 s), 72 °C/10 min. For R. rickettsii: first round 95 °C/5 min, 40 cycles (95 °C/40 s, 58 °C/40 s, 72 °C/50 s); second round: 95 °C/5 min, 40 cycles (95 °C/40 s, 58 °C/30 s, 72 °C/35 s), 72 °C/10 min. Products were electrophoresed on 1.5% agarose gels, stained with ethidium bromide (BIO-RAD, USA), and visualized on a UV transilluminator (UVP, M-20E, USA). The size of amplified fragments was verified using a commercial 1 kb DNA ladder (Cleaver Scientific, UK).

Bands of expected size were excised and purified using the FavorPrep™ GEL/PCR Purification Mini Kit (Favorgen Biotech Corp., Pingtung, Taiwan) following the manufacturer’s instructions. Six gp36 and four ompA amplicons were sequenced. Genetic similarity was analyzed using BLASTn, which compares sequences with those available in GenBank.

Phylogenetic trees were constructed using MEGA12 (Kumar et al. 2024) with the Maximum Likelihood method and the general time-reversible model of nucleotide substitution. A discrete Gamma distribution with 5 categories was used to model evolutionary rate differences among sites (+ G, parameter = 0.0401), with 0.00% of sites considered evolutionarily invariable (+ I). The tree with the highest log likelihood (− 401.64) is shown. Node support was assessed with 1000 bootstrap replicates. The analytical procedure encompassed 24 nucleotide sequences with 292 positions in the final dataset. Evolutionary analyses were conducted using up to 8 parallel computing threads.

Sequences were deposited in GenBank under accession numbers PV297777–PV297780 (ompA) and PV297781–PV297786 (gp36).

All statistical analyses were performed using R software (version 4.6.1; R Core Team, Vienna, Austria). PCR positivity for Ehrlichia canis was taken as the response variable, and the explanatory variables were sex, breed type, age, and neighborhood of origin of the dogs.

Fisher’s exact test was used to assess the association between PCR positivity and each categorical variable. Generalized linear models (GLM) with a binomial distribution and logit link function were fitted to estimate the effect of each variable on the probability of E. canis infection. Results were expressed as odds ratios (OR) with their respective 95% confidence intervals (95% CI) and P-values. A significance level of α = 0.05 was considered in all analyses.

Results

Of 123 dogs, 20 (16.2%) were PCR-positive for E. canis (40% males, 60% females; 45% purebred, 55% mixed-breed). Amplification of the housekeeping gene HPRT1 was achieved in more than 90% of the samples, confirming the integrity of the genetic material evaluated. Although Rhipicephalus linnaei ticks were visually identified, it is not certain that they were the only species of ticks infesting the dogs in the study. Detailed individual data are provided as Online Resource 1.

No statistical significance was identified for the variables sex, age, or breed type. Infection prevalence differed among the sampled neighborhoods, with positivity rates of: 3 de Mayo, 20.90% (14/67); Villa Juárez, 2.70% (1/37); and Jardines de Oriente, 26.32% (5/19). Fisher’s exact test confirmed this association (P = 0.013).

Multiple comparisons using Fisher’s exact test showed that Villa Juárez had a significantly lower prevalence than 3 de Mayo (adjusted P = 0.025; OR = 9.37) and Jardines de Oriente (adjusted P = 0.025; OR = 12.23). In contrast, no significant differences were observed between 3 de Mayo and Jardines de Oriente (adjusted P = 0.755), suggesting that both neighborhoods present similar risk levels for E. canis infection. A figure showing the spatial location of the neighborhoods where the dogs in the study were located is shown in Online Resource 3.

The six gp36 sequences obtained (PV297781–PV297786) clustered in a single clade, related to a sequence from a dog from Cuba (Fig. 1).

Fig. 1.

Phylogenetic tree constructed with MEGA12 software from partial nucleotide sequences of the gp36 gene of Ehrlichia canis from different geographical origins. The tree was inferred using the Maximum Likelihood method and the general time-reversible model of nucleotide substitution. Node support was assessed with 1000 bootstrap replicates. The Mexican sequences obtained in this study are identified with a white triangle, while those obtained in other studies are identified with black triangles (America), circle (Asia), and square (Africa)

Four dogs (3.2%) were positive for R. rickettsii (3 males, 1 female; 3 mixed-breed, 1 purebred). Individual data are shown in Online Resource 2. The ompA sequences were distributed in two clades: sequence PV297780 clustered with three USA tick-derived sequences, while PV297777–PV297779 formed a second clade related to USA tick sequences and a R. rickettsii isolate (Fig. 2).

Fig. 2.

Phylogenetic tree constructed with MEGA12 software from partial nucleotide sequences of the ompA gene of Rickettsia rickettsii from different geographical origins. The tree was inferred using the Maximum Likelihood method and the general time-reversible model of nucleotide substitution. Node support was assessed with 1000 bootstrap replicates. The Mexican sequences obtained in this study are identified with a white triangle, while sequences from other countries are identified with a white diamond (Europe) and a black triangle (America)

One dog (identified as 61c, 24-month-old mixed-breed female) was coinfected with E. canis and R. rickettsii.

Discussion

This is the first molecular identification of E. canis and R. rickettsii in tick-infested dogs from some neighborhoods of the Chihuahua city, Mexico. The 16.2% E. canis prevalence is lower than that reported in other Mexican regions using 16 S rRNA PCR: 36% in Yucatán (Pat-Nah et al. 2015), 28.3% in central Mexico (Acevedo Monroy 2016), and 47.45% in Ciudad Juárez, Chihuahua (Rodríguez-Alarcón et al. 2020). However, studies targeting gp36 are scarce; Almazán et al. (2016) reported 4% in Torreón, Coahuila. The use of gp36 improves specificity compared to the highly conserved 16 S rRNA gene, which may explain the lower detection rate. The presence of E. canis in blood can also be intermittent due to tropism for spleen and bone marrow (Harrus and Waner 2011).

For R. rickettsii, the 3.2% detection rate aligns with the 3.13–6.9% reported in Coahuila and Durango using gltA PCR on ticks (Castillo-Martínez et al. 2017), although it is lower than the 17.53% found in free-roaming dogs in Sonora (Montaño Márquez 2016). The low bacteremia in blood may reduce PCR sensitivity (Kidd et al. 2008); skin biopsies during the acute phase are recommended (Paddock et al. 2004).

Phylogenetic analysis of gp36 sequences revealed that E. canis from Chihuahua clustered with a Cuban isolate and, in an adjacent clade, with sequences from the USA and Brazil (Doyle et al. 2005). For R. rickettsii, ompA sequences grouped with USA tick isolates, consistent with the geographical proximity of Chihuahua to the southern United States.

The detection of coinfection in dog 61c reinforces the role of R. linnaei and other ticks as an vectors for multiple pathogens (Dantas-Torres et al. 2012; Sosa-Gutierrez et al. 2016). Coinfections with E. canis and R. rickettsii have been previously reported in Mexico (Aragón-López et al. 2021; Ojeda-Chi et al. 2019) and may complicate clinical diagnosis, potentially exacerbate disease severity, and facilitate pathogen transmission to vectors. Moreover, the finding that all positive dogs were clinically healthy at the time of sampling highlights the role of asymptomatic carriers as silent reservoirs and sentinels for public health, as has been documented for both pathogens (Di Bella et al. 2024; Kordick et al. 1999).

This study has several limitations. First, the sample size (n = 123) was determined by logistical constraints and was not based on a formal sample size calculation. Second, the study population consisted of a convenience sample of tick-infested dogs captured at a single animal control center, with uneven distribution across three neighborhoods (54.5% from one site). Therefore, the reported frequencies cannot be extrapolated to the general canine population of Chihuahua. Third, formal analytical validation of the nested PCR assays was not performed, although specificity was confirmed by Sanger sequencing and BLASTn analysis of all sequenced amplicons. Fourth, tick identification was performed by morphological assessment only; molecular confirmation of the vector or vectors species was not conducted. Given the taxonomic complexity of the R. sanguineus sensu lato complex, this limits the precision of vector identification. Despite these limitations, this study confirms the circulation of both pathogens in the region and underscores the need for further surveillance including clinical cases and vector populations.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: H Ramírez Álvarez.Data curation: H Ramírez Álvarez.Formal analysis: MJ Sánchez Pérez, C. Rodríguez-Murillo.Funding acquisition: H Ramírez Álvarez.Investigation: MJ Sánchez Pérez.Methodology: MJ Sánchez Pérez, C. Rodríguez-Murillo.Project administration: H Ramírez Álvarez.Resources: H Ramírez Álvarez.Software: H Ramírez Álvarez.Supervision: H Ramírez Álvarez.Validation: H Ramírez Álvarez.Visualization: MJ Sánchez Pérez.Writing – original draft: MJ Sánchez Pérez.Writing – review and editing: G Gordillo Pérez, C. Rodríguez-Murillo, GE Acevedo-Jiménez, SE García-Vázquez, AI Hernández García.

Funding

This research was funded by Universidad Nacional Autónoma de México (www.unam.mx), Research Chair C12478 of Facultad de Estudios Superiores Cuautitlán, UNAM awarded to H Ramírez Álvarez. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval

This study was approved by the Internal Committee for the Care and Use of Experimental Animals (CICUAE) under protocol number nC 26_11. All procedures followed the Mexican Official Standard NOM-087-ECOL-SSA1-2002 for the management of hazardous biological-infectious waste.

Competing interests

The authors declare no competing interests.