• Open Access

Latent Equine Herpesvirus-1 in Trigeminal Ganglia and Equine Idiopathic Headshaking


Corresponding author: Dr Monica Aleman, Department of Medicine and Epidemiology, Tupper Hall 2108, One Shields Avenue, School of Veterinary Medicine, University of California Davis, Davis, CA 95616; e-mail: mraleman@ucdavis.edu.



Trigeminal neuralgia or neuropathic pain has been regarded as a putative cause of idiopathic headshaking in horses. Equine herpesvirus-1 (EHV-1) infection and resultant postherpetic pain have been suggested as a possible cause of such neuropathic pain.


To determine the presence of EHV-1 in the trigeminal ganglia of horses with idiopathic headshaking.


Nineteen horses: control (n = 11, 9 geldings, 2 mares, median age 11 years) and headshaking (n = 8, all geldings, median age 11.5 years) horses were sourced from the equine research herd and caseload at the Veterinary Medical Teaching Hospital.


Prospective study to determine the presence of EHV-1 latency in trigeminal ganglia of horses with idiopathic headshaking by real-time PCR detection of the glycoprotein B (gB) gene and the DNA polymerase (ORF 30) gene of EHV-1 in the absence of detectable late structural protein gene (gB gene) mRNA. Control horses were used for comparison. A house keeping gene (equine GAPDH) and positive and negative samples for EHV-1 were used for quality control.


All samples from control horses and 7 of 8 headshaking horses were negative for EHV-1. One headshaking horse tested positive for a single copy of EHV-1 gene.

Conclusions and Clinical Importance

This study does not support a role for EHV-1 infection and presumed postherpetic pain in the etiopathogenesis of equine headshaking.


equine herpesvirus-1


glycoprotein B gene

ORF 30

Open Reading Frame 30


polymerase chain reaction


Veterinary Medical Teaching Hospital

Idiopathic headshaking, also known as idiopathic trigeminal neuralgia, has been described for over 100 years and appears to affect most commonly mature horses (median 9 years old, range 1–30 years) and mainly geldings of Thoroughbred, Quarter Horse and related breeds, and Warmblood breeds.[1-3] Although the etiopathogenesis of headshaking remains elusive, trigeminal neuralgia is regarded as the likely explanation of observed clinical signs.[2, 3] Human sufferers of neuropathic pain describe burning, tingling, itching, and electric-shock-like sensations,[4] which might equate to the observed signs displayed by horses with idiopathic headshaking. One cause of chronic trigeminal neuropathic pain in humans is a postherpetic pain syndrome following reactivation of latent infection by the herpesvirus varicella-zoster virus within sensory neurons. This pain can manifest as a constant or intermittent spontaneous pain and mechanical allodynia of the affected body region, which can be on the side of the face.[5]

In a similar mechanism to varicella-zoster virus-induced postherpetic neuralgia, it is possible that herpesvirus latency and reactivation in the trigeminal ganglion are associated with neuropathic pain and are manifested as headshaking. Serological and PCR studies show that the majority of horses experience repeated reinfections or reactivations of equine herpesvirus-1 (EHV-1) throughout life, many of which are not associated with clinical disease.[6] Latent infection occurs in the trigeminal ganglion and lymphoreticular system.[6] Older horses might be more likely to have latent EHV-1 in trigeminal ganglia following repeated infections or reactivations,[6, 7] which would raise the possibility of EHV-1 being involved in the pathogenesis of idiopathic headshaking in mature horses.[3]

The aim of this study was to investigate the presence of EHV-1 in the trigeminal ganglia of horses diagnosed with idiopathic headshaking. Trigeminal ganglia from healthy horses (median age-matched) were also investigated for comparison. A high frequency of EHV-1 positive ganglia from headshaking horses, but not controls, might implicate a role for EHV-1 in the etiopathogenesis of idiopathic headshaking.

Materials and Methods

Prospective Study

All procedures were approved by the Institutional Animal Care and Use Committee of the University of California. Owner consent was gained for euthanasia and sample collection for all horses. Horses were euthanized by overdose of pentobarbital sodium (100 mg/kg IV).

Control Horses

Control horses were sourced from research herd horses with a normal physical and neurologic examination used for terminal studies, and clinical cases from the William R. Pritchard Veterinary Medical Teaching Hospital (VMTH) euthanized for nonneurologic causes such as colic. All control horses had no previous history or current clinical signs consistent with headshaking.

Headshaking Horses

Headshaking horses were sourced from the VMTH referral population. Inclusion criteria consisted of horses with a clinical diagnosis of idiopathic headshaking based on consistent behavior observed by the authors as previously described[3] and exclusion of other causes of headshaking. Owners were given specific questions with regards to the type of headshaking (vertical, horizontal, mild, violent movements, sudden) and associated signs (eg, snorting, rubbing nose, trotting with head low in the ground, apparent anxiety), frequency of behavior, severity, seasonality, and environmental factors including exercise at the time of the observed behavior. In addition to a thorough physical examination, a detailed examination of the oral cavity, nasal passages, eyes, ears, sinuses, and guttural pouches was performed to rule out headshaking because of other causes. Bridles and tack were also excluded as possible sources of headshaking.

Trigeminal Ganglia Collection and Processing

Whole trigeminal ganglia from both sides (right and left) were harvested immediately postmortem and flash frozen by immersion in liquid nitrogen. Samples were stored at −80°C until further processing. Preparation of the trigeminal ganglia for genomic DNA (gDNA) extraction was performed according to the Real-Time PCR Research and Diagnostics Core Facility from our institution as previously described.[8] In brief, 500 μL of stabilization solution1 was added to each sample containing 50–100 mg of trigeminal ganglion tissue. Proteinase K and 2 grinding beads (4-mm diameter, stainless steel beads2) were added and the samples homogenized in a grinder3 for 2 minutes at 1,000 strokes per minute. Following storage of the resulting lysate at −20°C for at least 1 hour to reduce foam, proteinase K digestion was performed at 56°C for 30 minutes. Total nucleic acid was extracted using a QIAxtractor4 according to the manufacturer's instructions.

Genomic DNA was precipitated with 100 μL total gDNA, 6 μL 5 M NaCl, 10 μL 5 mg/mL glycogen,5 and 300 μL absolute ethanol. The mixture was inverted in a 1.5-mL microcentrifuge tube several times and stored at −20°C overnight. Samples were centrifuged, washed with 70% ethanol, and resuspended in 20 μL water. A gene-specific, PCR-based preamplification method similar to Dolganov et al (2001)[9] was utilized. This method consists of 2 steps: a PCR-based, gene-specific preamplification step with DNA and a mixture of primers targeting the EHV-1 gB gene, equine GAPDH, and EHV-1 neurotropic and nonneurotropic strains. The 2nd step consists of a single-plex TaqMan reaction to quantify each preamplified pathogen.

The Real Time TaqMan PCR6 assays used for EHV1 neurotropic and nonneurotropic strains (NC_00149167.1) were designed by AB Primer Express 3 by the Real-Time PCR Diagnostic Core Facility, University of California, Davis, with sequences from listed Genbank accession numbers. Each PCR reaction contained 20× primers and probe for the respective TaqMan6 system with a final concentration of 400 nM for each primer and 80 nM for the TaqMan6 probe and commercially available PCR mastermix7 containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 5 mM MgCl2, 2.5 mM deoxynucleotide triphosphates, 0.625U AmpliTaq Gold DNA polymerase per reaction, 0.25 U AmpErase UNG per reaction, and 5 μL of the total nucleic acid sample in a final volume of 12 μL. Samples were placed in a 384 well plate and amplified in an automated fluorometer.8 Standard amplification conditions were used: 2 minutes at 50°C, 10 minutes at 95°C, 40 cycles of 15 seconds at 95°C, and 60 seconds at 60°C. Fluorescent signals were collected during the annealing temperature, and Ct values were exported with a threshold of 0.1 and a baseline of 2–5 for all genes of interest. Banked tissue samples from horses positive for EHV-1 and water were used as positive and negative controls, respectively.


Control Horses

Eleven horses comprised the control group (9 geldings and 2 mares) with a median age of 11 years (range 2–17 years). The breeds included Quarter Horse and related breeds (n = 5), Thoroughbred (n = 2), Warmblood (n = 2), and Arabian and Sport horse, one each.

Headshaking Horses

All 8 horses in this group displayed stereotypic behavior of idiopathic headshaking, and no apparent cause could be determined. All headshakers were geldings with a median age of 11.5 years (range 10–15 years) of Quarter Horse and related breeds (n = 5) and Thoroughbred, Warmblood, and Sport horse breeds (n = 1, each). These horses displayed severe clinical signs of headshaking making them unsafe to ride or handle and had been previously treated but remained unresponsive to multiple therapies such that the owner had elected for euthanasia on humane grounds.

None of the control or headshaking horses displayed clinical signs of active infection with EHV-1 such as acute respiratory or neurological disease during examination, hospitalization, or at time of euthanasia. These horses had no previous history of respiratory or neurologic disease.

EHV-1 DNA Detection

Trigeminal ganglia samples from all control and 7 of 8 headshaking horses tested negative for EHV-1. One headshaking horse tested positive for a single copy of EHV-1 gene. All positive and negative control samples tested as expected.


Based on the results from this study, EHV-1 does not appear to play a role in the etiopathogenesis of idiopathic headshaking in horses. All affected horses from this study had a history of headshaking in previous years and current signs compatible with headshaking behavior. Although the number of horses with idiopathic headshaking in this study was low, we postulate that if EHV-1 played any role in this disorder, latent virus would have been detected in the trigeminal ganglia of a greater proportion of horses with headshaking compared with a control population. Most affected horses (59%) have a seasonal onset and cessation of clinical manifestations, and 41% of horses have constant or sporadic episodes.[3] The majority of affected horses develop signs in the spring and early summer and cease in the late summer and fall.[3] A few cases develop signs in the fall and cease in the spring.[3] The horses presented here were examined during the spring. Seasonality should not affect the presence of latent virus.

Based on published prevalence of EHV-1 in the equine population,[6, 7] it was expected to find horses positive for EHV-1 in both groups of horses. The prevalence of EHV-1 latency in this study was lower than the 12.4% reported in a recent study[7] in trigeminal ganglia from horses undergoing routine postmortem examination, which could be because of the small sample size in the current study. Despite sample size, it is important to emphasize that only 1 of the 8 horses with idiopathic headshaking had latent EHV-1 in the trigeminal ganglia. The prevalence and distribution of EHV-1 biovars varies with breed, age, and tissue tested, with trigeminal ganglia being a site of tropism for latent EHV-1.[7] It has been suggested that older horses might be more likely to have EHV-1 latency in trigeminal ganglia, but median age of the current control and headshaking groups were similar to that of the horses (13 years) in which latent EHV-1 was found at postmortem examination.[7]

The current study was performed at the same referral veterinary facility with the same laboratory and methodology as previously described,[7] thereby reducing any influence of geographic region or laboratory methodology in variation of EHV-1 detection. To ensure analytical sensitivity, methodology included nucleic acid precipitation and preamplification steps which allows detection of even a single target gene in a tissue sample.[8] Latent EHV-1 appears evenly distributed within the tissue such that single samples are representative.[7] All samples passed quality control following precipitation and preamplification based on established equine GAPDH values (cycle threshold range 9–13) and were processed concurrently with positive and negative control samples for EHV-1.

In conclusion, trigeminal ganglia from 7 of 8 headshaking horses sampled in this study were negative for latent EHV-1, and therefore it appears highly unlikely that EHV-1 is involved in the etiopathogenesis of equine headshaking. However, a larger population of horses with idiopathic headshaking should be investigated and the results of this study interpreted with caution. Involvement of other causes of presumed trigeminal neuropathic pain warrants investigation in headshaking horses.


The study was supported by gifts from anonymous donors toward equine neurology research.

Conflict of interest: None.


  1. 1

    DX Binding Solution (DXB), Qiagen, Valencia, CA

  2. 2

    SpexCertiprep, Metuchen, NJ

  3. 3

    GenoGrinder 2000 (SpexCertiprep), Lebanon, NJ

  4. 4

    QIAExtractor, Qiagen

  5. 5

    Fermentas Inc, Glen Burnie, MD

  6. 6

    Real Time TaqMan PCR, Foster City, CA

  7. 7

    TaqMan Universal PCR Mastermix, Applied Biosystems, Carlsbad, CA

  8. 8

    7900 HT FAST Real Time PCR System, Applied Biosystems