The research team, led by scientists at the Federal Rural University of Rio de Janeiro in collaboration with colleagues at the Federal University of the Jequitinhonha and Mucuri Valleys, analyzed whole blood samples from 386 dogs of both sexes, all breeds and all ages, collected from veterinary clinics and rural settlements across nine municipalities including Unaí, Arinos, Buritis and Formoso. The sample size was not arbitrary: the investigators calculated a minimum requirement of 384 individuals using a standard proportion-based formula assuming a 95 percent confidence level and a 5 percent margin of error, ensuring that their prevalence estimates would carry statistical weight. Blood was drawn by cephalic venipuncture with the owners’ consent, anticoagulated with EDTA, and processed initially in Minas Gerais before being shipped to Rio de Janeiro for the full suite of molecular analyses.

The screening strategy followed a cascading logic designed to distinguish true E. canis infection from other members of the Anaplasmataceae family that can infect dogs. After confirming the integrity of every extracted DNA sample through amplification of the canine housekeeping gene gapdh, which yields an approximately 400 base pair fragment, the researchers ran a conventional PCR targeting a 691 base pair region of the 16S rRNA gene shared across Anaplasmataceae bacteria. That screen returned 38 positives, or 9.8 percent of the sampled population. When those 38 samples were then interrogated with a more specific PCR targeting the trp19 gene, which is diagnostic for E. canis itself, 21 samples were confirmed, establishing an overall molecular positivity rate of 5.4 percent. The authors emphasize that this figure reflects active circulation of the agent in a region where molecular data were essentially absent, and they caution that infection frequencies reported across studies vary widely depending on sampling design, target populations and diagnostic approaches, since many surveys focus on clinically suspect animals and may overestimate prevalence relative to apparently healthy dogs.

What elevates the study beyond a simple prevalence survey is its two-gene characterization strategy. The team amplified an 843 base pair fragment of the p28 gene, which encodes an outer membrane protein produced by a multigene family known for generating antigenic variation, and separately targeted the trp36 gene, which encodes an immunoreactive glycoprotein distinguished by a variable region of tandem amino acid repeats. Among the 21 trp19-positive dogs, 12 were positive for p28 and eight for trp36, and the successfully sequenced products formed the basis of the phylogenetic and population-genetic analyses.

The p28 results were revelatory. Sequences showed between 99.27 and 99.88 percent nucleotide identity with reference strains Jake (CP000107) and Jaboticabal (EF014897), yet translation of the nucleotide data into amino acid sequences exposed meaningful protein-level variation. Compared against the Jake reference, the Brazilian sequences displayed 11 amino acid mutation sites at positions 64, 68, 79, 80, 82, 142, 144, 150, 152, 203 and 253, while comparisons against the Jaboticabal reference revealed point substitutions at positions 81, 153, 154, 156 and 208. More striking still was the haplotype analysis. Using DnaSP v6.12.03 to compute polymorphism indices and PopART v1.7 to construct a median-joining network incorporating 11 local sequences and 19 reference sequences from Brazil, the United States, India, the Philippines and Vietnam, the researchers identified nine haplotypes exclusive to the Unaí microregion. Haplotype diversity reached 0.9632, nucleotide diversity 0.0731, with 224 polymorphic sites across the alignment. No haplotype dominated; the most frequent one appeared in only three samples, a pattern the authors interpret as evidence of genetic dispersion rather than clonal expansion of a single strain.

The phylogenetic reconstruction reinforced this picture of a mixed population. Maximum likelihood analysis based on a 792 base pair p28 alignment of 30 E. canis sequences, using the Tamura 3-parameter model with gamma-distributed rate variation selected by the Akaike Information Criterion and bootstrap support estimated from 1,000 replicates, split the Unaí isolates between two clusters. Eight sequences grouped with strains from Rio de Janeiro, India and the United States, while three clustered with the Brazilian Jaboticabal strain, demonstrating that even within a single microregion, dogs carry bacteria belonging to divergent intraspecific lineages.

The trp36 gene told an equally compelling story, one with implications for how E. canis strains are classified globally. Nucleotide identity for the five sequenced trp36 samples ranged from 97 to 99 percent against GenBank references from multiple countries. Four of the sequences encoded a tandem repeat motif of nine amino acids, TEDSVSAPA, with copy numbers varying from 12 to 16 repeats, while a single sequence carried a different eight-amino-acid repeat, ASVVPEAE, in 13 copies. This copy-number plasticity matters because the TRP36 protein is a major target of the host immune response, and the tandem repeat region is thought to evolve under strong immunological selection, driving antigenic diversification and immune evasion. In the phylogenetic tree built from a 942 base pair trp36 alignment of 28 sequences, four of the Unaí isolates fell within the American (USA) genogroup, while one clustered squarely with Brazilian isolates of the BR genogroup. The co-circulation of these two genogroups in the same geographic area, the authors argue, points to a non-homogeneous bacterial population shaped by repeated introduction events and the long-term maintenance of divergent strains in an endemic environment.

The epidemiological and clinical stakes of this diversity are considerable. Previous work has suggested that different E. canis genogroups may be associated with distinct clinical profiles; the BR genogroup in particular has been linked to reduced hematocrit and hemoglobin levels, implying a greater impact on anemia in infected animals. Genotype may also influence bone marrow regeneration, the intensity of inflammatory responses, and the overall clinical course of infection. Yet the relationship between circulating genotypes and disease severity remains poorly resolved, a gap the authors highlight as a priority for future research. From a practical standpoint, the genetic variability documented in surface antigen genes such as p28 and trp36 could compromise the sensitivity of diagnostic assays that rely on these loci, undermine the performance of candidate vaccines, and favor the emergence of variants with different virulence profiles.

The study also underscores a methodological lesson: no single gene tells the whole story. Conserved loci such as 16S rRNA and dsb show limited variability in E. canis, which is precisely why the field has migrated toward the more labile p28 and trp36 markers. By combining the two, the researchers captured complementary facets of the population structure, the multigene-family-driven variation of p28 and the immune-pressured tandem repeat dynamics of trp36. All newly generated sequences have been deposited in NCBI GenBank under accession numbers PZ138103 through PZ138113 for p28 and PZ138114 through PZ138118 for trp36, making the dataset available for global comparative analyses.

For a region of Brazil where molecular data on E. canis were previously scarce, the study delivers a clear message: the Unaí microregion harbors a genetically rich and dynamic bacterial population, with multiple lineages, genogroups and repeat architectures circulating simultaneously in the dog population served by the brown dog tick Rhipicephalus linnaei, the principal vector. As canine monocytic ehrlichiosis continues to burden veterinary medicine across tropical and subtropical regions, and as Ehrlichia species increasingly attract attention in human medicine, including detections in human blood bank donors elsewhere in the Americas, fine-grained molecular surveillance of the kind performed in Minas Gerais will be essential for tracking how these pathogens diversify, spread and adapt.