The study protocol was reviewed and approved by the Central Scientific Committee, College of Veterinary Medicine, University of Wasit (Approval No. CSC-CVM-71; January 25, 2025), and the Institutional Animal Care and Use Committee, College of Veterinary Medicine, University of Wasit (Approval No. WU-CVM.CVM-IPM-1085; February 14, 2025). All procedures involving the study horses, including clinical examination and the collection of blood and genital swab samples, were conducted in accordance with the approved study protocol and institutional animal care requirements.

The study was conducted from March to September 2025 in selected areas of four Iraqi provinces: Baghdad, Babylon, Nineveh, and Wasit. Baghdad, the capital of Iraq, is located in central Iraq at approximately 33°20′ N, 44°26′ E. Babylon is located in central Iraq, approximately 85 km south of Baghdad, at 32°32′ N, 44°25′ E. Nineveh is located in northern Iraq, approximately 410 km north of Baghdad, at 36°21′ N, 43°09′ E. Wasit is located in eastern Iraq, southeast of Baghdad, at approximately 32°30′ N, 45°49′ E. The study areas were selected principally according to the availability of horses and the cooperation and logistical support provided by horse owners and veterinary clinics during sample collection. The geographic distribution of the sampled horses across the four provinces is presented in Figure 1.

This study used a cross-sectional design to investigate the serological and molecular occurrence of T. equigenitalis and its associations with reproductive and selected epidemiological characteristics in horses from four Iraqi provinces. A total of 157 adult horses aged ≥4 years, with or without a history of reproductive disorders, were included. Clinical information was recorded for each horse before blood and genital swab samples were collected. Serum samples were examined for antibodies against T. equigenitalis using ELISA, whereas genital swab samples were examined for T. equigenitalis DNA using PCR. PCR-positive products were subsequently sequenced and subjected to phylogenetic analysis. The resulting serological and molecular findings were evaluated in relation to clinical and epidemiological variables. These procedures are consistent with the study population and diagnostic workflow reported in the manuscript.

A total of 157 adult horses aged ≥4 years, with or without a history of reproductive disorders, were included in the study. Following clinical examination, approximately 5 mL of jugular venous blood was collected aseptically from each horse into a labeled anticoagulant-free gel tube. In addition, genital swab samples were obtained from the vagina of mares and from the urethral fossa, urethra, and prepuce of stallions. Blood and swab samples were transported to the laboratory under cooled conditions. Blood samples were centrifuged at 3000 × g for 5 min, after which the separated serum was transferred into labeled 1.5-mL microcentrifuge tubes. Serum and genital swab samples were stored at −40°C until serological and molecular examination, respectively.

Serum samples were examined using a qualitative Horse Taylorella equigenitalis ELISA Kit (Cat. No. SL0011Ho; SunLong Biotech, China) according to the manufacturer’s instructions. Kit reagents and serum samples were brought to room temperature (22–27°C) before analysis. After completion of the assay procedures, absorbance was measured at 450 nm and expressed as optical density (OD). Following assay validation using the positive and negative controls, the critical cutoff value was calculated according to the manufacturer’s instruct-tions, and samples were classified as positive or negative based on their respective OD values.

DNA was extracted from genital swab samples using the Type A protocol of the G-Spin™ Total DNA Extraction Kit (iNtRON Biotechnology, Korea) according to the manufacturer’s instructions. DNA purity and concentration were evaluated using a NanoDrop spectrophotometer (Thermo Fisher Scientific, USA), with reported purity values of 1.8–2.0 and concentrations of 50–150 ng/µL. PCR targeted the 16S rRNA gene using the primers TEF (5′-AATACCGCATACGCCCTGAG-3′) and TER (5′-GTATTATCCCAGGGGGCTGC-3′), which were designed based on the T. equigenitalis reference sequence deposited in NCBI GenBank (accession No. NR_029234.1).

DNA amplification was performed using AccuPower® PCR PreMix (Bioneer, Korea). Each reaction had a final volume of 25 µL comprising 7 µL of DNA template, 2 µL of forward primer, 2 µL of reverse primer, and 14 µL of nuclease-free water. Amplification was performed in a thermal cycler (Bio-Rad, USA) under the following conditions: initial denaturation at 95°C for 5 min; 35 cycles of denaturation at 95°C for 40 s, annealing at 56°C for 40 s, and extension at 72°C for 40 s; followed by final extension at 72°C for 5 min.

PCR products were separated by electrophoresis on 1.5% agarose gels stained with ethidium bromide at 100 V and 80 mA for 60 min. Standard T. equigenitalis DNA purchased from Macrogen (Korea) was used as the positive control, whereas nuclease-free water was used as the negative control. Amplified products were visualized using an ultraviolet transilluminator, and samples producing the expected 583-bp amplicon were considered PCR-positive.

PCR-positive products for T. equigenitalis were sequenced by Macrogen (Korea) using a modified Sanger sequencing method. The resulting sequences were aligned using ClustalW implemented in MEGA version 11 and subsequently submitted to the NCBI GenBank database to obtain accession numbers. Multiple sequence alignment, phylogenetic tree construction, sequence identity, nucleotide diversity (π), mean genetic distance, and evolutionary divergence between groups were evaluated using MEGA version 11 and DnaSP version 5.1. Haplotype analysis was also performed using the sequence data. The manuscript subsequently reports that 23 PCR-positive sequences received GenBank accession numbers PZ146151.1–PZ146173.1.

Data were entered into Microsoft Excel 2016 and statistically analyzed using GraphPad Prism version 8.0.2. Associations between categorical variables were evaluated using Pearson’s chi-square (χ²) test or Fisher’s exact test, as appropriate. The Wilson method was used to calculate 95% confidence intervals (95% CI). Odds ratios (ORs), relative risks (RRs), and their corresponding 95% CIs were calculated using MedCalc Statistical Software. Data were expressed as number and percentage [n (%)] or mean ± standard error (SE), as appropriate. Differences were considered statistically significant at p < 0.05.

ELISA detected antibodies against T. equigenitalis in 52 of 157 horses (33.12%), whereas 105 (66.88%) were seronegative (Figure 2). The mean OD value of ELISA-positive samples (0.386 ± 0.009) was significantly higher than that of ELISA-negative samples (0.184 ± 0.003) (Figure 3).

PCR analysis of genital swab samples detected T. equigenitalis DNA in 23 of 157 horses (14.65%), whereas 134 (85.35%) were PCR-negative (Figure 4).

Among the 157 examined horses, 21 (13.38%) were positive by both ELISA and PCR, 31 (19.74%) were ELISA-positive/PCR-negative, two (1.27%) were ELISA-negative/PCR-positive, and 103 (65.61%) were negative by both methods (Figure 5). These distributions are consistent with the overall ELISA and PCR positivity reported for the study population.

Among mares, ELISA seropositivity differed significantly according to reproductive and clinical status (p = 0.0013). The highest seropositivity was observed among mares with a history of abnormal vaginal discharge (68.42%), followed by estrous cycle disturbances (52.94%), dystocia (40.00%), abortion (37.50%), vaginal lesions (33.33%), apparently healthy status (33.33%), and vaginitis (25.00%). No ELISA-positive cases were detected among the two mares classified as infertile (Table 1). These values correspond to the clinical distributions reported in the manuscript.

Among stallions, ELISA seropositivity also differed significantly among the recorded clinical categories (p = 0.0308). Seropositivity was detected in seven apparently healthy stallions (15.56%) and one stallion with genital inflammation (9.09%), whereas no ELISA-positive cases were detected among stallions with abnormal genital discharge or genital lesions (Table 2).

Among mares, PCR positivity differed significantly according to reproductive and clinical status (p = 0.0012). The highest proportion of PCR-positive animals occurred among mares classified as infertile (50.00%), followed by apparently healthy mares (25.00%), mares with a history of abortion (25.00%), abnormal vaginal discharge (21.05%), vaginal lesions (16.67%), estrous cycle disturbances (11.76%), and dystocia (10.00%) (Table 1).

Among stallions, PCR-positive results were detected only in the apparently healthy group. No significant association was detected between PCR positivity and clinical category (p = 0.0719), and no PCR-positive cases were identified among stallions with genital inflammation, abnormal genital discharge, or genital lesions (Table 2).

ELISA seropositivity differed significantly among age groups (p = 0.0041). Horses aged >10 years had the highest seropositivity at 62.86% (22/35), followed by those aged 7–10 years (33.33%, 17/51), >5–6 years (23.81%, 10/42), and the youngest study group (10.35%, 3/29). The corresponding OR and RR analyses also differed significantly among age groups (both p = 0.0001) (Table 3).

Seropositivity was significantly associated with sex (p = 0.0248), with a higher proportion observed among females (52.38%, 44/84) than males (10.96%, 8/73). The OR and RR analyses also differed significantly according to sex (both p = 0.0001) (Table 3).

Seropositivity also differed significantly among provinces (p = 0.0237). The highest proportion was recorded in Baghdad (39.71%, 27/68), followed by Wasit (36.84%, 14/38), Babylon (30.00%, 9/30), and Nineveh (9.52%, 2/21). The corresponding OR and RR analyses were reported as significantly different among geographic categories (both p = 0.0001) (Table 3). The manuscript reports these geographic distributions consistently.

PCR positivity differed significantly among age groups (p = 0.0345). The highest PCR positivity was detected among horses aged >5–6 years (38.10%, 16/42), followed by those aged 7–10 years (9.80%, 5/51) and >10 years (5.71%, 2/35); no PCR-positive animals were detected in the youngest study group (0/29). The OR and RR analyses also differed significantly among age categories (both p = 0.0001) (Table 4). The value of 38.10% is supported by 16/42 and by Table 4.

PCR positivity was significantly associated with sex (p = 0.0131), with a higher proportion detected among females (25.00%, 21/84) than males (2.74%, 2/73). The corresponding OR and RR analyses also differed significantly according to sex (both p = 0.0001) (Table 4).

Geographically, PCR positivity was 16.18% (11/68) in Baghdad, 15.79% (6/38) in Wasit, 13.33% (4/30) in Babylon, and 9.52% (2/21) in Nineveh. The manuscript reports an overall difference among provinces at p = 0.029, with the OR and RR analyses reported at p = 0.0001 (Table 4).

Sequences obtained from the 23 PCR-positive samples, designated GhIQf-Equine 1–GhIQf-Equine 23, were deposited in the NCBI GenBank database under accession numbers PZ146151.1–PZ146173.1. Phylogenetic relationships among the Iraqi T. equigenitalis sequences and reference sequences retrieved from GenBank were inferred from partial 16S rRNA gene sequences using the maximum-likelihood method. Branch lengths represented estimated evolutionary divergence, and the scale bar corresponded to 0.01 substitutions per site. Bootstrap support values were displayed at the corresponding internal nodes, with several major nodes showing support of 99%–100%.

Pairwise sequence comparison showed that the Iraqi sequences had 97.18%–99.12% nucleotide identity with the Iranian T. equigenitalis isolate Tay.Zahraei1 (GenBank accession No. DQ840456.1) (Table 5). GhIQf-Equine 10 showed the highest identity (99.12%), whereas GhIQf-Equine 2 and GhIQf-Equine 20 showed the lowest identity (97.18%). The phylogenetic tree did not contain a taxonomically distinct outgroup; sequences positioned toward the basal portion of the topology, including IR-CEM/Iran, KITE-1/Korea, HH139/Japan, 96-178/USA, and Kentucky 188/Japan, were also identified as T. equigenitalis (Figure 6). The manuscript reports the overall identity range and lack of a distinct outgroup accordingly.

Genetic diversity analysis identified 521 conserved sites, 32 variable sites, 32 polymorphic sites, 12 parsimony-informative sites, and 20 singleton variable sites. Thirteen haplotypes were detected, with a haplotype diversity of 0.909 and nucleotide diversity (π) of 0.0105. The average number of nucleotide differences was 5.81, and the overall mean genetic distance was 0.0105. Mean genetic distances were 0.0046 within the Iraqi sequences, 0.0204 within the reference sequences, and 0.0202 between the Iraqi and reference groups. Tajima’s D was −1.0503 (Table 6).

The present study demonstrated serological and molecular evidence of T. equigenitalis among horses in Iraq, with 33.12% of the examined horses testing positive by ELISA and 14.65% testing positive by PCR. The difference between serological and molecular positivity highlights the distinct biological information provided by the two diagnostic approaches. Previous studies have reported considerable variation in the detection of T. equigenitalis among horse populations. Parlevliet et al. [22] reported PCR positivity in 54 (50.47%) of 107 asymptomatic mares in the Netherlands, including 14/36 (38.89%) PCR-positive horses of Icelandic origin. In Slovenia, Zdovc et al. [23] examined 980 genital swabs collected from 245 stallions and detected 17 (6.94%) culture-positive stallions, with three additional animals detected by PCR. In a small Croatian survey involving 12 horses, culture results were negative, whereas PCR detected Taylorella spp. in three (25%) horses; subsequent sequencing confirmed T. equigenitalis in one horse [24].

Considerably lower occurrence has been reported in some large-scale surveillance programs. A South African national screening program using quantitative PCR (qPCR) with culture confirmation detected T. equigenitalis in 39/3820 (1.02%) horses, comprising 36/3703 (0.97%) stallions and 3/117 (2.56%) mares distributed across 12 properties in three provinces [8]. In contrast, surveillance of Thoroughbred horses in Korea detected T. equigenitalis by qPCR in 71/526 (13.5%) horses [14]. In a Swedish retrospective investigation, Al-Kass et al. [25] reported T. equigenitalis in 53/25,512 (0.21%) examined samples, including 11/2308 (0.48%) stallions. In Iran, Aalinasab et al. [9] detected T. equigenitalis in 4/240 (1.70%) infertile mares. More recently, Solbach et al. [7] detected the bacterium by qPCR in 52/361 (14.40%) Icelandic horses located in southern Germany and Austria. Differences between these studies and the present findings may reflect variation in study populations, sampling strategies, diagnostic methods, anatomical sampling sites, sample handling, geographic setting, management practices, and study periods. Direct comparison should therefore be interpreted cautiously because the cited investigations differed substantially in design and diagnostic approach.

The present study identified different positivity rates using ELISA and PCR. Among the 157 horses, 21 (13.38%) were positive by both methods, 31 (19.74%) were ELISA-positive/PCR-negative, two (1.27%) were ELISA-negative/PCR-positive, and 103 (65.61%) were negative by both methods. Thus, most PCR-positive horses were also seropositive, whereas a substantially larger group had detectable antibodies without simultaneous PCR detection.

This discordance should be interpreted in the context of the different targets of the assays. ELISA detects antibodies and may therefore identify previous or ongoing exposure, whereas PCR detects bacterial DNA in the sampled material at the time of collection. Serological findings can be influenced by the timing and magnitude of the antibody response, cross-reactivity, nonspecific antibody binding, antigen characteristics, cutoff selection, and individual immunological variation [26, 27]. Conversely, PCR-negative results in exposed or infected animals may occur when the bacterial load is low, shedding is intermittent, the selected sampling site does not contain detectable organisms, sampling is suboptimal, DNA deteriorates during processing, or amplification is affected by inhibitors present in clinical specimens [28, 29]. Consequently, the higher ELISA positivity observed in this study should not be interpreted as evidence of lower PCR performance or reduced ELISA specificity without direct diagnostic accuracy evaluation against an appropriate reference standard.

Clinical findings demonstrated a significant association between serological and molecular positivity and reproductive status in mares, whereas PCR positivity was not significantly associated with the recorded clinical categories in stallions. Among mares, the highest ELISA positivity occurred in those with a history of abnormal vaginal discharge (68.42%), whereas the highest PCR positivity occurred among mares classified as infertile (50%). In stallions, PCR-positive results were detected only among apparently healthy animals, supporting the epidemiological importance of clinically inapparent carriage. Similar differences in the clinical expression of T. equigenitalis infection between mares and stallions have been described previously [6, 9, 13, 22, 30–33].

The reproductive abnormalities detected among serologically or molecularly positive mares are biologically compatible with the known involvement of T. equigenitalis in the female reproductive tract. Infection-associated inflammatory changes within the endometrium and reproductive tract may affect uterine function and reproductive performance [20, 34]. Previous studies have similarly described transient reproductive dysfunction in naturally infected mares [35, 36]. In contrast, stallions may harbor T. equigenitalis without overt clinical manifestations, thereby complicating recognition and control of infection [12, 37, 38]. The presence of clinically inapparent carriers is particularly relevant to CEM surveillance because apparently healthy breeding animals may contribute to maintenance and dissemination of the organism. International movement and trade of horses have also contributed historically to the geographic dissemination of CEM and underscore the importance of appropriate surveillance and biosecurity measures [39].

Nevertheless, the clinical associations observed in the present study should not be interpreted as proof that T. equigenitalis caused each recorded reproductive disorder. The cross-sectional design identifies associations between diagnostic positivity and clinical history but does not establish temporal or causal relationships.

Age was significantly associated with both serological and molecular findings, although the patterns differed between diagnostic methods. Seropositivity was highest among horses aged >10 years (62.86%), whereas PCR positivity was highest among horses aged >5–6 years (38.10%). The higher seropositivity among older horses may reflect cumulative exposure over successive breeding seasons and greater lifetime opportunities for contact with infected animals. Repeated breeding activity may similarly increase the probability of previous exposure or prolonged colonization [11, 40]. Because ELISA detects antibodies rather than the organism itself, accumulation of previous exposure with increasing age offers a plausible explanation for the observed serological pattern.

In contrast, the highest molecular positivity in the present study occurred among horses aged 5–6 years. This finding may reflect differences in current exposure or bacterial detection across age groups; however, the study did not directly measure breeding frequency, mating activity, duration of colonization, or bacterial shedding. Therefore, specific behavioral or reproductive mechanisms underlying the age-associated PCR pattern cannot be established from the available data. Solbach et al. [7] similarly reported that advancing age was not associated with greater T. equigenitalis positivity, indicating that age-related patterns may vary among populations and epidemiological settings.

Sex was also significantly associated with both ELISA and PCR positivity. Females had substantially higher seropositivity (52.38%) than males (10.96%) and higher PCR positivity (25.00%) than males (2.74%). Higher positivity among mares has also been reported in other investigations [6, 14]. This finding is consistent with the more evident reproductive manifestations that may occur after colonization of the female reproductive tract. Nevertheless, the lower positivity detected among stallions in the present study does not diminish their potential epidemiological importance. Previous studies have demonstrated that stallions may harbor T. equigenitalis on external genital sites for prolonged periods while remaining clinically unaffected and can consequently contribute to maintenance and venereal transmission of CEM [7, 30].

Serological and molecular positivity varied among the four study provinces. Seropositivity was highest in Baghdad (39.71%) and Wasit (36.84%), followed by Babylon (30.00%) and Nineveh (9.52%). PCR positivity showed a narrower geographic range, with positivity of 16.18% in Baghdad, 15.79% in Wasit, 13.33% in Babylon, and 9.52% in Nineveh. These geographic differences may reflect variation in the populations sampled and local epidemiological conditions.

Differences in horse population structure, breeding management, animal movement, frequency of contact among breeding animals, and access to veterinary supervision have been proposed as factors that can influence transmission of equine infectious diseases [41–43]. However, these variables were not directly quantified in the present study. Therefore, the observed provincial differences should be regarded as epidemiological associations rather than evidence that any specific management or movement factor was responsible for the geographic distribution of T. equigenitalis.

Molecular characterization of the 23 Iraqi T. equigenitalis sequences demonstrated measurable genetic variation within the analyzed partial 16S rRNA gene region. Genetic diversity analysis identified 13 haplotypes, with a haplotype diversity of 0.909 and nucleotide diversity (π) of 0.0105. The mean genetic distance within the Iraqi sequences was 0.0046, compared with 0.0204 among the reference sequences and 0.0202 between the Iraqi and reference groups. These findings indicate that multiple sequence variants were represented among the PCR-positive Iraqi samples while overall nucleotide divergence within the Iraqi group remained relatively limited.

Sequence comparisons also demonstrated 97.18%–99.12% nucleotide identity between the Iraqi sequences and the Iranian Tay.Zahraei1 T. equigenitalis isolate (GenBank accession No. DQ840456.1), with GhIQf-Equine 10 showing the highest identity (99.12%). The high sequence similarity demonstrates genetic relatedness within the analyzed 16S rRNA gene region. Previous molecular studies have documented genetic variation and closely related lineages among T. equigenitalis isolates obtained from different geographic locations [21, 44, 45].

The presence of 32 polymorphic sites, including 12 parsimony-informative and 20 singleton variable sites, further demonstrates sequence variation within the analyzed dataset. However, the partial 16S rRNA sequence data used in the present study do not provide sufficient resolution to establish specific transmission lineages, recent clonal expansion, direction of transmission, common geographic origin, or direct epidemiological linkage between Iraqi and Iranian isolates. Therefore, the observed sequence similarity should be interpreted as molecular relatedness rather than evidence of cross-border transmission. Higher-resolution approaches, including multilocus sequence typing or whole-genome sequencing combined with detailed animal movement and breeding histories, would be required to resolve these epidemiological relationships more precisely [21, 44, 45].

The negative Tajima’s D value (−1.0503) was also observed in the analyzed dataset. However, because statistical significance for Tajima’s D was not reported and the analysis was based on partial 16S rRNA sequences, this finding alone does not establish recent population expansion, purifying selection, or another specific evolutionary process. Further genomic investigation with a larger and epidemiologically representative collection of isolates would be required to characterize the population structure and evolutionary dynamics of T. equigenitalis circulating among horses in Iraq.