Molecular testing for equine herpesviruses 1 (EHV-1) in healthy postpartum broodmares

Abstract

Objective

Our objective was to determine whether equine herpesviruses 1 (EHV-1) viral nucleic acids could be detected immediately after foaling from nasal and vaginal swabs, whole blood, and placental tissue of healthy mares.

Animals, procedure, and results

Nasal and vaginal swabs, EDTA blood, and placental tissue (296 samples) were collected from 74 clinically healthy postpartum broodmares within 24 h after giving birth to live, clinically healthy foals. All samples were tested (PCR) for nucleic acids of neuropathogenic and non-neuropathogenic strains of EHV-1, and all were negative.

Conclusion and clinical relevance

As EHV-1 was not detected in the immediate postpartum period in healthy mares with uncomplicated foaling, we inferred that EHV-1-positive samples from aborting mares and/or EHV-1 detection in fetal membranes indicate EHV-1-associated abortion.

Equine herpesviruses 1 (EHV-1) infection during pregnancy in broodmares can result in substantial reproductive losses (1). Reproductive EHV infections commonly manifest as 3rd-trimester abortions or neonatal foal deaths. Abortions are, economically, the most important result of EHV infections (1,2).

Typically, EHV causes lesions in several organs of the aborted fetus (3); however, abortion of EHV-1-negative fetuses in experimentally infected ponies has been documented (4). Further, EHV-associated abortions of virologically negative fetuses were confirmed in mares with naturally acquired infection (5). Conversely, pregnant mares can become infected without abortion, with similar EHV-1 viral loads in the chorioallantois of aborted fetuses and foals born clinically healthy from experimentally infected pony mares (6,7). Abortion associated with EHV-1 infection is initiated by endometrial thromboses and infarction due to viral-associated vasculitis, leading to placental separation and subsequent expulsion of the fetus and placenta (5).

Currently, detection of EHV-1 in tissues from an aborted fetus and/or the placenta is considered confirmatory for the etiological cause of the abortion. Nevertheless, whether clinically healthy broodmares giving birth to healthy foals could shed EHV-1 virus due to stress associated with foaling has not been investigated. Therefore, our objective was to determine if viral nucleic acids could be detected immediately after foaling from nasal and vaginal swabs, whole blood, and placental tissue of healthy mares. We hypothesised that EHV-1 would not be detected by molecular methods at the time of foaling in healthy broodmares giving birth to healthy foals.

Broodmare population and study design

Stud-farm veterinarians were asked to collect nasal and vaginal swabs, EDTA blood, and 2 × 2-centimeter pieces of placental tissue from clinically healthy broodmares within 24 h after the mares gave birth to live, clinically healthy foals. In each case, the placenta was inspected and gross pathological abnormalities recorded. A single, 2 × 2-centimeter piece of placental tissue from the pregnant horn was dissected and placed in 70% ethanol in a sterile container. A venous blood sample was collected into BD Vacutainer EDTA Tubes (BD Biosciences, Mississauga, Ontario), and nasal and vaginal swabs were taken using 13.3-centimeter plastic-shafted rayon-tipped nasal swabs and stored dry. Samples were refrigerated and submitted to the laboratory within 5 to 7 d after collection. The mare’s breed, age, parity, delivery, history of abortions, history of medical issues during pregnancy, and vaccination history for EHV-1 were collected. All procedures were approved by the Animal Care Committee of the University of Guelph (Protocol # 4390) and conformed to the standards of the Canadian Council on Animal Care.

Samples were processed and tested at the University of Guelph Animal Health Laboratory (Guelph, Ontario). A total of 296 samples from 74 case submissions were included in this study. Each EDTA blood sample was centrifuged at 3000 × g for 10 min, and 1 mL of the buffy coat cell layer was transferred into a fresh, 2.0-milliliter screw-cap vial. Swab samples were mixed by vortexing and clarified by centrifugation at 13 200 × g for 2 min. Tissue samples were stab-swabbed, and swabs were then immersed into 2-milliliter vials with 1 mL sterile PBS, mixed by vortexing, and clarified by centrifugation at 13 200 × g for 2 min, with 50-microliter aliquots used for nucleic acid extraction.

Nucleic acid extraction and polymerase chain reaction (PCR)

Total nucleic acids were extracted using the MagMAX-96 Viral RNA Isolation Kit in a MagMAX Express-96 Magnetic Particle Processor (Thermo Fisher, Waltham, Massachusetts, USA). Armored RNA Enterovirus (Asuragen, Austin, Texas, USA) was added to the kit lysis buffer to serve as an internal control. Nucleic acids extracted from samples and enterovirus internal control RNA were reverse-transcribed and amplified with a real-time PCR assay using Ag Path-ID One-Step RT-PCR Kit (Thermo Fisher), with primers and probes described for enterovirus (8) and EHV-1 (9). Reverse transcription, PCR amplification, and detection were performed in 25-microliter reactions in a Light Cycler 480 (Roche, Indianapolis, Indiana, USA). Samples with cycle threshold values < 37 were considered positive.

The study population consisted of 74 broodmares residing on 6 farms in southern Ontario. Within this population, 58 horses were standardbred, 14 were thoroughbred, 1 was a quarter horse, and 1 was a warmblood. Median age was 8 y (range: 4 to 21 y; age not recorded for 2 mares) and median parity was 2 (range: 1 to 13 events; data not recorded for 3 mares). Foaling was assisted in 45 mares and unassisted in 29. No previous abortions were reported for 49 mares, abortion history was unknown for 23 mares, and 2 mares had a history of abortion of undetermined cause. Forty-two mares were vaccinated against EHV-1 3× during pregnancy; however, the vaccine product was not recorded. Thirty mares were unvaccinated and 2 mares had an unknown vaccination status. Three standardbred mares were reported to have had placentitis during pregnancy and 71 mares had no history of issues during pregnancy. A total of 296 samples (4 specimens per mare) were tested for the presence of the neuropathogenic and non-neuropathogenic strains of EHV-1, for which all PCR reactions were negative.

Equine herpesviruses 1 nucleic acids were not detected in any biological sample collected from any broodmare after foaling. Therefore, from a diagnostic viewpoint, it is reasonable to conclude that a positive test result from clinical specimens (i.e., placental tissues, whole blood, nasal swab) submitted from an aborting mare should be considered clinically relevant. In agreement with our findings, a study at a stud farm in South Africa investigating vertical transmission of EHV-1 to healthy foals at birth did not detect viral nucleic acids in nasal swabs from mares and foals, or in placental tissues collected from foaling mares (10).

In an experimental EHV-1 challenge study in pregnant mares, there were high viral loads in aborted fetal and placental tissues of aborting mares, but also comparable viral levels in placentae of 2 foals born clinically healthy (7). This observation has not been reported for healthy newborn foals or associated placental tissues, though screening placental tissues for the presence of EHV is not routinely done. The virulence of circulating EHV-1 strains was demonstrated to be variable and to affect some organs preferentially (abortogenic versus neuropathogenic), which could have direct effects on clinical manifestation and perhaps on disease severity (3). Of note, an abortogenic strain inoculated in pregnant mares resulted in high viral loads in the placenta of an aborted fetus, as expected, but also in placentae of healthy foals (7). Currently, detection of EHV-1 in placental tissue of an aborted fetus with concomitant viremia in the aborting mare would be considered evidence of EHV infection-induced abortion. Equine herpesvirus-associated abortions in mares with EHV-negative fetal or placental tissues have been reported during naturally occurring and experimentally induced abortions (5,11). Lesions and viral distribution throughout the placentae of aborted fetuses are not homogenously distributed (11); therefore, placental sampling site could affect test results. Based on those studies, even if placental or fetal tissues are EHV-negative, abortions could still be caused by EHV. However, interpretation of the opposite scenario is mostly unknown and deserves further investigation in a large population of clinically healthy broodmares.

The site of sample collection for virus detection is relevant for diagnostic purposes and results interpretation. There is surprisingly little information available regarding virus detection during naturally occurring abortions in broodmares. According to the American College of Veterinary Internal Medicine consensus statement, a conventional nonquantitative PCR result, positive for EHV-1 on a blood sample, indicates viremia probably resulting from an active infection, whereas a positive EHV-1 test result on a nasal swab sample should be interpreted as indicative of infectious virus shedding (12). The period of viremia and/or viral shedding during the postabortion period in mares with EHV-associated abortions remains largely unknown. Currently, the American Association of Equine Practitioners EHV control guidelines, for example, recommend maintaining isolation procedures (primary perimeter) for 28 d after the last suspected new infection (13).

The lack of positive tests in this group of mares suggests to us that EHV-1 shedding is not a common event in healthy broodmares, at least not in this region. Some stress factors, such as weaning, commingling, transportation, and concurrent infections, have been anecdotally associated with development of EHV-1 disease (12). Foaling and dystocia are considered stress factors that could lead to EHV-1 shedding without clinical relevance. Thus, in this study, samples were collected from healthy mares after foaling to test this hypothesis. Equine herpesvirus 1 was not detected in this study, nor by Brown et al (10). Therefore, at this time, this factor appears to have no effect on EHV-1 shedding or infection in the postpartum period.

The number of foaling events as a risk factor for an EHV-associated abortion is unknown. Any pregnant mare may be considered at risk of acquiring EHV-1 and virus-associated abortion, likely depending on the mare’s immune status against EHV. An EHV-associated abortion can occur in a mare of any age but is largely restricted to the last trimester of pregnancy (11). Virus was not detected in any member of this group of mares, regardless of the parity or age of the mare. Immune status against the virus was not investigated. There were only a few breeds represented in this study; however, breed has not been identified as a factor for EHV-1 infection/shedding or, more importantly, for EHV-1 associated abortions.

More than 1/3 of the tested mares were unvaccinated for EHV-1, though effects of vaccination on viral detection in the postpartum period was not a specific objective of this study. Nevertheless, commercially available killed and modified-live vaccines are available for prevention of EHV-associated abortions (12,14). Such vaccines are administered with the objective of helping reduce the viral load in the environment, minimize viral replication in the respiratory tract, and reduce nasal shedding and both magnitude and duration of viremia (14). A recent review article on the efficacy of vaccination against EHV-1 concluded that the reduction in viremia could not be evaluated based on the current literature, and thus the efficacy remains unknown (14). Equine herpesvirus 1-associated abortions occur in vaccinated broodmares, and those mares can shed virus for several weeks after aborting (15). The efficacy of vaccination against EHV for the prevention of abortions remains controversial and therefore requires further investigation.

Important limitations of this study included overrepresentation of the standardbred breed and the small number of farms sampled. Therefore, results were susceptible to bias due to data clustering. Factors such as farm management and preventive-medicine practices could have affected the observed results.

In summary, EHV-1 was not detected in the immediate postpartum period in healthy mares with uncomplicated foaling in southern Ontario. Despite the limited number of mares and farms in this study, we inferred that EHV-1-positive samples from aborting mares and/or EHV-1 detection in fetal membranes indicate EHV-1-associated abortion.

Funding Statement

The Ontario Animal Health Network (OAHN) provided the funding for this project and their support is gratefully acknowledged.