Contagious equine metritis is a highly contagious sexually transmitted disease of equids caused by the gram-negative bacterium Taylorella equigenitalis. Although the organism is primarily transmitted via natural mating, transmission through indirect venereal contact with contaminated fomites can also contribute to the spread of this disease. The resulting infection is characterized by endometritis, vaginal discharge, and a severe adverse effect on fertility of mares. Affected mares often recover without treatment and have only short-term infertility, although some mares may harbor the organism in their genital tract for several months or longer without clinical signs; thus, they may serve as a reservoir for infection.1 Similar to other venereally transmitted pathogens, stallions are subclinical carriers of the causal agent (ie, have no clinical signs); without treatment, they can remain carriers for many months or even years. Recovery of T equigenitalis has been reported from placental tissue2 and from the genitalia of colts and fillies, with the organism acquired in utero or through contamination in the birth canal during parturition.3
The disease was first reported in England and Ireland in 19774,5 and was introduced into the United States in September of the same year through the importation of 2 carrier stallions. This resulted in a widespread outbreak of CEM in Thoroughbreds in Kentucky in the 1978 breeding season.6 The disease can be a source of severe economic loss to the equine breeding industry; it has also resulted in major restrictions in the international trade and transportation of Thoroughbreds and other horse breeds.
The United States, through rigorously enforced import regulations, has largely been free of this disease since the initial outbreak in 1978. However, in late 2006, a group of 3 Warmblood stallions imported in August 2004 had positive results when tested via routine culturing for T equigenitalis. More recently, a domestic Quarter Horse stallion had positive results during routine testing for this bacterium in December 2008 prior to export of its semen to Europe.7 Extensive epidemiological investigation and testing of 1,005 horses that had been directly or indirectly exposed to this and certain other stallions revealed an additional 22 stallions and 5 mares in the United States that were carriers of T equigenitalis.7 In contrast to outbreaks of the disease in Thoroughbreds (which are bred by natural mating), mares bred by AI to the stallions that were confirmed carriers during the last 2 CEM events reportedly had few if any clinical signs or effects on fertility.8,9 The stallions involved were used exclusively to provide semen for AI.
Semen used for AI is routinely extended in a semen extender that contains an antimicrobial or a combination of antimicrobials. We hypothesized that the inclusion of antimicrobials in semen extender prevents growth of T equigenitalis in extended semen and therefore reduces the risk of CEM transmission. The objective of the study reported here was to compare the growth of T equigenitalis in raw and extended semen from a stallion that was a CEM carrier and to compare the outcome of inseminating mares with raw, extended, or cryopreserved CEM-positive semen. As a corollary, PCR primers specific for T equigenitalis were developed and evaluated for use in testing of semen.
Materials and Methods
Animals—Twenty-one adult mares and 1 sexually mature stallion that were free of CEM were included in the study. In addition, semen was obtained from a stallion that was a naturally infected CEM carrier. All animal procedures were completed with the approval of the Institutional Animal Care and Use Committee at the University of Kentucky.
Comparison of growth of T equigenitalis in semen obtained from the carrier stallion—Semen was obtained from a naturally infected stallion that had positive results for T equigenitalis when tested via bacterial culture and PCR assay. Semen was stored frozen at −80°C until use. It was thawed and extended (1:3 mixture of semen and commercially available semen extender containing amikacin sulfate and penicillin G potassiuma). The final concentrations of amikacin and penicillin G potassium in the extender were 1 and 0.63 g/L, respectively. Aliquots of extended semen then were incubated for 30 minutes or 6 hours at room temperature (approx 21°C) and for 24 hours in a cooling container at 4°C.b After incubation, semen samples were cultured to determine the number of CFUs of T equigenitalis. A raw semen sample (ie, not extended) was analyzed to obtain baseline CFU counts for T equigenitalis, and a raw semen sample obtained from a stallion free of CEM was used as a negative control specimen. An aliquot of each semen sample was frozen at −20°C for subsequent PCR analysis. The number of CFUs of T equigenitalis per milliliter was determined for the raw semen samples and for each of the 3 extended semen samples. Colony-forming unit counts were determined by use of serial dilutions (1:10, 1:100, and 1:1,000) in tryptic soy broth and plating of each dilution on 4 chocolate Eugon agar plates (2 without antimicrobials and 2 with modified Timoney-Shin medium).10 Plates were incubated for 7 days at 37°C in 8% CO2, and the number of CFUs were then counted. Experiments were performed in triplicate with 3 aliquots of frozen semen. The number of CFUs in unextended semen and in extended semen incubated for 3 time periods was compared via an ANOVA with statistical software.c
AI of mares with raw, extended, or cryopreserved CEM-positive semen—The sexually mature stallion was exposed to T equigenitalis. The T equigenitalis used to experimentally infect this stallion was derived from the CEM-positive semen obtained from the same stallion as in the first part of the study. The stallion was teased by use of mares in estrus to induce an erection. Microlesions then were created by brushing the penile surface for 1 minute; this was followed by application of a concentrated suspension of T equigenitalis to the penile shaft, fossa glandis, and urethra, as described elsewhere.11 The procedures were repeated 48 hours later. Seven days after the second set of procedures, samples were obtained from the penile shaft, urethra, and fossa glandis and submitted for bacterial culture.
After it was confirmed that the stallion had positive results for culture of CEM, semen was collected 3 times within 1 week and frozen at −80°C without addition of a semen extender or antimicrobials. To confirm unpublished observations of the authors regarding the effect of freezing and thawing on T equigenitalis, CFU counts were performed on semen from one of the ejaculates over 2 freeze-thaw cycles. Semen from the 3 collections were thawed, pooled, divided into 10-mL aliquots, and refrozen at −80°C until further use. The semen was pooled and divided into aliquots to ensure that all mares received the same amount of T equigenitalis organisms; a CFU count was performed after the aliquots were thawed. A sample of the pooled semen was obtained and submitted for bacterial culture to confirm the growth of T equigenitalis. Following completion of the study, the stallion was treated and tested in accordance with USDA guidelinesd to confirm that it was free from infection with T equigenitalis.
All mares included in the study were screened for T equigenitalis. Swab specimens were obtained twice (interval of at least 14 days between specimen collections) from the uterus, clitoral fossa, and clitoral sinuses. In addition, a blood sample was obtained from each mare and submitted for CF testing. The 21 mares were assigned via a randomization table to insemination with raw, extended, or cryopreserved CEM-positive semen (n = 6 mares/group) and insemination with CEM-negative semen (control group [3]).
Each mare received 10 mg of prostaglandin F2α,e IM. After induction of luteolysis, the mares were examined to evaluate follicular development, uterine edema, and cervical and uterine tone. When there was a lack of a distinct corpus luteum, the presence of a follicle at least 35 mm in diameter, pronounced uterine edema, and diminished uterine and cervical tone, each mare received 2,500 U of human chorionic gonadotropin,f IV, as an ovulation-inducing agent and was simultaneously inseminated. Mares inseminated with raw semen received a 10-mL aliquot of the raw CEM-positive semen from the experimentally infected stallion. Semen was thawed at room temperature, and mares were inseminated with the thawed semen within 1 hour after thawing. Mares inseminated with extended semen received a 10-mL aliquot of CEM-positive semen from the experimentally infected stallion that was thawed at room temperature 24 hours before use, extended with 30 mL of semen extender,a and stored in a cooling containerb at 4°C for the 24-hour period before AI. Mares inseminated with cryopreserved semen received 8 straws of semen from the naturally infected carrier stallion. Semen had been centrifuged in an extender without antimicrobials and cryopreserved in accordance with a standard commercial procedure at a local equine hospital in semen extender containing amikacin sulfate (0.5 g/L) and penicillin G potassium (1,000 U/mL) at a concentration of 100 × 106 spermatozoa/straw (200 × 106 spermatozoa/mL). Cryopreserved semen was stored in liquid nitrogen for at least 9 months prior to use. The semen was thawed at 37°C and inseminated within 5 minutes after thawing. The CEM-positive status of the cryopreserved semen was confirmed via PCR assay. Mares inseminated with CEM-negative semen received 10 mL of raw semen from a stallion free of CEM, as determined on the basis of bacterial culture.
At 3, 5, 7, 14, and 21 days after AI, the mares were observed for signs of vaginal discharge via inspection of the perivaginal area and vaginoscopy with a disposable speculum. Inflammatory signs were subjectively assessed on the basis of redness of the vaginal and cervical mucosa and the amount of discharge detected at the time of examination. Transrectal ultrasonography was used to detect the presence of uterine fluid. At 7, 14, and 21 days after AI, swab specimens were obtained from the uterus, clitoral fossa, and clitoral sinuses and submitted for bacterial culture. Swab specimens were placed in individual tubes containing Amies transport medium with charcoal and plated within 4 hours after collection. If vaginal discharge was observed at the time of any scheduled collection of swab specimens, samples of the discharge were obtained and submitted for bacterial culture. A blood sample was obtained from each mare 21 days after insemination and submitted for CF testing. A Fisher exact test was used to compare the outcome after AI (ie, positive or negative culture results).
Bacterial culture of swab specimens—Each swab specimen was plated on chocolate Eugon agar without antimicrobials and on a selective medium, and plates were incubated for a minimum of 7 days at 37°C in 8% CO2. Colonies with morphological and growth characteristics of T equigenitalis were tested for catalase and cytochrome oxidase activity and Gram stain reaction. Cellular morphology and motility were assessed via darkfield microscopy. Colonies of small, gram-negative, nonmotile coccobacilli or bacilli that had strong positive results for catalase and cytochrome oxidase were subcultured onto chocolate Eugon agar plates to confirm that they were microaerophilic and to obtain a pure growth of the colony for additional testing. Final identification as T equigenitalis was made on the basis of a positive reaction in a commercially available slide agglutination test.g
Development and evaluation of a T equigenitalis–specific PCR assay—Taylorella equigenitalis–specific PCR primers were designed on the basis of the T equigenitalis gene for 16S rRNA (GenBank accession No. AB069660) by use of a PCR primer design program.h Forward and reverse primers were designed to include base pair positions 434 to 454 in the resulting PCR product because these 18 residues differ between T equigenitalis and Taylorella asinigenitalis.12 Amplification was performed with 2 U of a DNA polymerasei in a total volume of 50 μL that contained 3mM MgCl2, 0.2 mM of each deoxynucleoside triphosphate, and 1μM of each forward and reverse primer. Thermal cycling was performed on a thermal cycler.j Samples were subjected to initial denaturation at 95°C for 5 minutes, which was followed by 35 cycles of denaturation at 95°C for 30 seconds, annealing at 55°C for 30 seconds, and elongation at 72°C for 1 minute to amplify the PCR template. To test specificity of the primers, DNA was isolated from T equigenitalis, T asinigenitalis, Streptococcus equi subsp equi, and Streptococcus equi subsp zooepidemicus with a DNA isolation kitk in accordance with the manufacturer's instructions. The DNA was isolated from 50 μL of semen via the DNA isolation kitk; DNA concentration was measured with a UV spectrophotometer.l A 2-μL volume of eluted DNA was used as the template for PCR amplification. The PCR products were developed via agarose gel electrophoresis, for which 5 μL of the PCR mixture was mixed with 1 μL of loading dyem and separated on a 1% agarose gel.
Results
Effect of freezing and thawing of semen on T equigenitalis—Semen from the CEM-positive stallion contained 92,500 CFUs/mL following the first freeze-thaw cycle and 94,500 CFUs/mL following the second freeze-thaw cycle. This confirmed unpublished observations made by our laboratory group that freezing and thawing of semen has a minimal effect on T equigenitalis.
Comparison of T equigenitalis infectivity in raw and extended CEM-positive semen samples—Raw semen from the carrier stallion yielded a heavy growth of T equigenitalis, with an estimated count of 42,167 CFUs/mL. Semen extended in an extender containing amikacin and penicillin G potassium significantly (P < 0.001) reduced the CFU count, compared with that of the raw semen, irrespective of the duration of incubation (30 minutes, 6 hours, or 24 hours; Figure 1).
Number of CFUs per milliliter for raw (without extender) semen from a CEM-positive stallion and extended semen incubated for 30 minutes and 6 hours at room temperature (approx 21 °C) and for 24 hours in a cooling container at 4°C. Values with different letters differ significantly (P < 0.001).
Citation: Journal of the American Veterinary Medical Association 241, 7; 10.2460/javma.241.7.916
The PCR analysis revealed that the semen sample obtained from the stallion confirmed free of T equigenitalis had no visible band after agarose gel electrophoresis. All samples obtained from the stallion experimentally infected with the organism, whether they comprised raw semen or semen extended in an antimicrobial-containing semen extender, yielded positive results for the PCR assay.
AI of mares with raw, extended, or cryopreserved CEM-positive semen—Samples collected from the penile shaft, urethra, and fossa glandis 7 days after experimental exposure of the stallion to T equigenitalis yielded positive results on bacterial culture. All mares inseminated with raw CEM-positive semen had clearly detectable signs of vaginitis and cervicitis at each examination time point. Mild signs of vaginitis and cervicitis were evident 3 days after AI, and small amounts of discharge could be observed within the vaginal lumen during vaginoscopy. By day 5 after AI, signs of inflammation were more severe in all mares. The intensity of vaginitis and cervicitis did not change appreciably between days 5, 7, and 14 after AI. Signs were less intense in all mares by 21 days after AI. None of the mares had external signs of vaginal discharge at any of the time points after AI, and accumulation of fluid was not detected within the uterine lumen with transrectal ultrasonographic examination. None of the mares inseminated with extended or cryopreserved semen from the CEM-positive stallion or mares in the control group inseminated with raw CEM-negative semen developed signs of vaginitis or cervicitis or a vaginal discharge. The CEM-positive status of the cryopreserved semen was confirmed via PCR assay (Figure 2).
Photograph of an agarose gel after electrophoresis depicting the outcome of PCR amplification with Taylorella equigenitalis–specific primers and DNA isolated from semen with positive bacterial culture results for T equigenitalis and cryopreserved semen with negative bacterial culture results for T equigenitalis. Lanes were as follows: 1, DNA standard; 2 to 4, CEM-positive semen; and 5 to 7, CEM-negative semen).
Citation: Journal of the American Veterinary Medical Association 241, 7; 10.2460/javma.241.7.916
Swab specimens obtained from the uterine lumen, clitoral fossa, and clitoral sinuses of mares inseminated with raw CEM-positive semen had positive culture results for T equigenitalis on days 7, 14, and 21 after AI (Figure 3). None of the samples obtained from mares inseminated with CEM-positive semen extended with semen extender containing amikacin and penicillin G potassium, CEM-positive cryopreserved semen, or CEM-negative raw semen had positive culture results for T equigenitalis. The prevalence of positive culture results for T equigenitalis after insemination with raw semen was significantly (P = 0.002) higher than after insemination with extended or cryopreserved semen. Mares inseminated with raw CEM-positive semen had seroconverted (CF test) by day 21 after AI.
Number of mares with positive culture results for T equigenitalis 7 (gray bar), 14 (black bar), and 21 (white bar) days after insemination with raw CEM-negative semen, raw CEM-positive semen, extended CEM-positive semen, or cryopreserved CEM-positive semen.
Citation: Journal of the American Veterinary Medical Association 241, 7; 10.2460/javma.241.7.916
Development and evaluation of the T equigenitalis–specific PCR assay—The DNA isolated from T equigenitalis yielded a visible PCR product on agarose gels (Figure 4). However, DNA isolated from T asinigenitalis, S equi subsp equi, and S equi subsp zooepidemicus did not result in a visible PCR product.
Photograph of an agar gel after electrophoresis depicting the outcome of PCR amplification with T equigenitalis–specific primers and DNA isolated from various bacteria. Lanes were as follows: 1 and 10, DNA standard; 2, 6, and 8, Taylorella asinigenitalis; 3, Streptococcus equi subsp equi; 4, Streptococcus equi subsp zooepidemicus; and 5, 7, and 9, T equigenitalis.
Citation: Journal of the American Veterinary Medical Association 241, 7; 10.2460/javma.241.7.916
Discussion
The objectives of the present study were to investigate the effect of antimicrobial-containing semen extender on the growth of T equigenitalis in CEM-positive semen and to investigate the infectivity of CEM-positive semen (raw, extended with antimicrobial-containing semen extender, or cryopreserved with cryopreservation semen extender containing antimicrobials) in mares bred via AI. The experimental design was not designed to mimic a typical breeding situation; rather, it was designed to specifically address whether antimicrobials in a commercially available semen extendera would reduce the number of CFUs of T equigenitalis in infected semen, thereby reducing the risk of transmitting CEM in mares undergoing AI with extended semen. Semen was collected 3 times from an experimentally infected stallion, frozen, and subsequently thawed and pooled to standardize the dose of T equigenitalis among inseminations. Freezing and thawing did not alter the CFUs of T equigenitalis in semen, which was in agreement with previous observations by our laboratory group. Although there are no data to suggest that damage to the spermatozoa through freezing and thawing can affect the impact of antimicrobials on bacteria in infected semen, this cannot completely be ruled out and may require further examination. Although thawed semen typically would not be stored for 1 hour at room temperature before insemination in clinical situations, the conditions of the experimental design would likely not have resulted in reduced growth of T equigenitalis in semen but instead would have allowed for bacterial growth and allowed us to more rigorously test our hypothesis.13
Results of the study reported here confirmed that antimicrobial-containing semen extender greatly reduced the number of CFUs of T equigenitalis when CEM-positive semen was extended at a ratio of 1:3 and incubated for 30 minutes or 6 hours at room temperature or for 24 hours in a cooling container. Although incubation of semen for 30 minutes at room temperature was reflective of clinical conditions when semen is collected and extended for immediate insemination or for shipment via controlled cooled conditions, the 6-hour time point was selected to determine whether additional incubation time at room temperature would further reduce the amount of T equigenitalis in extended semen for the conditions of the present study. No significant difference was observed between the 2 time points. This is consistent with results of other studies14,n in which semen treated with an antimicrobial-containing semen extender reduced the number of colonies of T equigenitalis detected via culture. In the present study, we used semen from an experimentally infected stallion to test the effect of antimicrobial-containing extenders on the growth of T equigenitalis. Similar results have been reported when semen from a noninfected stallion was spiked in vitro with T equigenitalis.15 Results of the experimental challenge study reported here confirmed that 6 mares inseminated with raw CEM-positive semen had clinical signs of the disease (ie, vaginitis, cervicitis, and vaginal discharge) at each sampling point after AI. Clinical signs had decreased in intensity by day 21 after AI, which is in accordance with observations in other studies.15,16 The swab specimens obtained from the mares in this experimental group all had positive results when cultured for T equigenitalis. In contrast, none of the mares inseminated with CEM-positive semen extended with semen extender containing amikacin and penicillin G potassium, CEM-positive cryopreserved semen, or raw CEM-negative semen developed clinical signs of CEM. Furthermore, all of the swab specimens obtained from mares in these experimental groups were culture-negative for T equigenitalis. These findings strongly suggested that extending semen with an antimicrobial-containing semen extender may significantly reduce the risk of transmission of CEM by AI.
These results may help to explain the findings of the extensive epidemiological investigations that were conducted following the rediscovery of CEM in the United States in December 2008. Of 723 mares in that outbreak exposed primarily via fresh-cooled semen from stallions confirmed to be carriers of CEM, only 5 (0.7%) were culture-positive for T equigenitalis, 3 of which were inseminated with extended semen.7 This contrasts markedly with the high transmission rate between stallions and mares observed after natural breeding during the 1978 outbreak in Thoroughbreds in Kentucky.7 Mares inseminated in the present study with raw semen that contained a strain of T equigenitalis isolated from the 2008 outbreak developed typical signs of CEM and were culture-positive for T equigenitalis, which confirmed that this strain of T equigenitalis is capable of infecting mares and causing clinical expression of disease.
The PCR assay was found to have high sensitivity and specificity for detecting T equigenitalis in equine semen, regardless of whether it was raw, extended fresh-cooled, or cryopreserved semen. However, further studies are needed to confirm these promising results, especially with regard to screening cryopreserved semen for evidence of T equigenitalis.
For the study reported here, we concluded that antimicrobial-containing semen extender significantly reduced the number of T equigenitalis in CEM-positive semen and thereby may reduce the risk of transmission of CEM through AI. The study also confirmed the disease-inducing ability of the strain of T equigenitalis associated with the 2008 CEM outbreak in the United States. Finally, a PCR assay was developed that was able to detect T equigenitalis in semen.
ABBREVIATIONS
| AI | Artificial insemination |
| CEM | Contagious equine metritis |
| CF | Complement-fixation |
| CFU | Colony-forming unit |
Equipro containing amikacin and potassium penicillin G, Minitube, Verona, Wis.
Equitainer, Hamilton Research Inc, South Hamilton, Mass.
SAS, version 9.2, SAS Institute Inc, Cary, NC.
USDA APHIS. FY 2009 CEM incident testing protocol for CEM investigations [version 4]; 2009. Available at: www.obpa.usda.gov/budsum/fy09budsum.pdf. Accessed Dec 5, 2009.
Lutalyse, Pfizer, New York, NY.
Chorulon, Intervet, Millsboro, Del.
Monotayl, Bionor Laboratories AS, NO-3702 Skien, Norway.
Jellyfish 3.3.1, Field Scientific LLC, Lewisburg, Pa.
FastStart Taq DNA polymerase, Roche Diagnostics Corp, Indianapolis, Ind.
Mastercycler ep gradient, Eppendorf, Westbury, NY.
ZR Genomic DNA I kit, Zymo Research, Orange, Calif.
NanoDrop ND-1000, Agilent Technologies, Palo Alto, Calif.
Blue/Orange loading dye, Promega, Madison, Wis.
Oliveri B, Love B, Rezabek G, et al. The effects of common semen extenders inoculated with various bacterial strains on semen quality in horses (abstr), in Proceedings. Am Assoc Vet Lab Diagn Annu Conf 2009;160.
References
1 Powell DG. Contagious equine metritis. Adv Vet Sci Comp Med 1981; 25:161–184.
2 Powell DG, Whitwell K. The epidemiology of contagious equine metritis (CEM) in England 1977–1978). J Reprod Fertil Suppl 1979; 27:331–335.
3 Timoney PJ, Powell DG. Isolation of the contagious equine metritis organism from colts and fillies in the United Kingdom and Ireland. Vet Rec 1982; 111:478–482.
4 Crowhurst RC. Genital infection in mares. Vet Rec 1977; 100:476.
5 Timoney PJ, Ward J, Kelly P. A contagious genital infection of mares. Vet Rec 1977; 101:103.
6 Swerczek TW. Contagious equine metritis in the USA. Vet Rec 1978; 102:512–513.
7 Erdman MM, Creekmore LH, Fox PE, et al. Diagnostic and epidemiologic analysis of the 2008–2010 investigation of a multi-year outbreak of contagious equine metritis in the United States. Prev Vet Med 2011; 101:219–228.
8 Hayna JH, Syverson CM, Dobrinsky JR. Embryo transfer during concurrent contagious equine metritis infection. Reprod Fertil Dev 2008; 20(suppl 1): 157–158.
9 Timoney PJ. Contagious equine metritis: an insidious threat to the US horse breeding industry. J Anim Sci 2011; 89:1552–1560.
10 Timoney PJ, Shin SJ, Jacobson RH. Improved selective medium for isolation of the contagious equine metritis organism. Vet Rec 1982; 111:107–108.
11 Platt H, Atherton JG. The experimental infection of ponies with contagious equine metritis. Equine Vet J 1978; 10:153–159.
12 Wakeley PR, Errington J, Hannon S, et al. Development of a real time PCR for the detection of Taylorella equigenitalis directly from genital swabs and discrimination from Taylorella asinigenitalis. Vet Microbiol 2006; 118:247–254.
13 Timoney PJ, Harrington A, McArdle J, et al. Survival properties of the causal agent of contagious equine metritis 1977. Vet Rec 1978; 102:152.
14 Timoney PJ, O'Reilly PJ, Harrington AM, et al. Survival of Haemophilus equigenitalis in different antibiotic-containing semen extenders. J Reprod Fertil Suppl 1979; 27:377–381.
15 Acland HM, Allen PZ, Kenney RM. Contagious equine metritis: distribution of organisms in experimental infection of mares. Am J Vet Res 1983; 44:1197–1202.
16 Acland HM, Kenney RM. Lesions of contagious equine metritis in mares. Vet Pathol 1983; 20:330–341.
