Equine herpesvirus 1 (EHV-1) is a serious threat to horses worldwide (1). EHV-1 outbreaks have emphasized the potential effects of this virus including: (1) severe respiratory disease, neurological disease, abortion, or death; (2) the highly contagious nature of the virus; (3) incomplete protection from vaccination and antiviral therapy; and (4) substantial economic costs for the equine industry (2–11). The prevalence of EHV-1 among 4,228 horses with fever, respiratory signs, and/or neurological deficits was 2.7% in the USA between 2008 and 2014 (12). The incidence of EHV-1-induced neurological disease was reported increased over the last decade internationally (13–15). Therefore, the global equine industry is compelled to identify more effective EHV-1 control measures and treatments.

Equine herpesvirus 1 infection is initiated by inhalation of infectious virus, whereupon the virus enters epithelial cells of the upper respiratory tract and replicates, leading to viral shedding via nasal discharge, and further spreading of the infection (16). Fever and respiratory clinical signs may ensue, although some horses experience subclinical shedding (5). EHV-1 upper respiratory tract entry facilitates infection of leukocytes, enabling the virus to circulate and infect endothelial cells of the central nervous system, leading to myeloencephalopathy; or reach the pregnant uterus, causing late-term abortion (17–19).

Although much effort has been invested in improving EHV-1 vaccines to prevent disease and curtail virus spread, current vaccines offer limited protection from infection or reactivation of the virus. Vaccines presently on the market have been shown to suppress EHV-1 disease and shedding but may not limit viral load. The level of protection against EHV-1-induced neurological disease by vaccination is unclear, and not claimed by any of the vaccine manufacturers (20–22). Further, natural exposure induces immune protection that can be as short as 3–6 months (20–23). Care of infected animals includes administration of antiviral drugs and supportive therapies (6, 24, 25). During EHV-1 infection, antiviral treatment using synthetic nucleoside analogs shows promising efficacy in cell culture models, but treatment of ponies with an experimental EHV-1 infection has been less effective (25–28). Currently used antiviral drugs are only active during lytic infection as they interfere with replication of the viral genome and, as such, they have no effect on latent EHV-1.

After EHV-1 infects the horse cell, the EHV-1 DNA genome is released into the nucleus to eventually produce more infectious EHV-1. A cost of this strategy is that the EHV-1 DNA is now subjected to the host cell’s gene regulation machinery. Recent work with human herpes simplex virus (HSV1) has demonstrated that the viral genome becomes susceptible to host-mediated epigenetic regulation, including the assembly and modulation of host-derived histones on the viral DNA genome (29–31). Subsequently, posttranslational modifications of the histone tails either permit or repress viral gene expression (29, 32, 33). Remarkably, maintaining a repressive epigenetic state of the HSV1 DNA suppressed viral gene expression during lytic infection, and suppressed reactivation from latency in vitro and in vivo (33–35). In these studies, the repressive state was preserved by preventing the removal of methyl groups from lysine 9 of histone 3 (H3K9) with the use of compounds that block the activity of the lysine-specific demethylase 1 (LSD1) protein [e.g., tranylcypromine, a monoamine oxidase inhibitor (MAOI); or a novel selective LSD1 inhibitor, OG-L002] (33–35). Because LSD1 is not the only protein that modulates histone methylation, LSD1 inhibition is not expected to have global effects. Further, the use of a histone demethylase inhibitor in conjunction with standard antiviral therapy in experimental HSV1 infection exerted a synergistic reduction of active infection, and limited reactivation from viral latency (33).

Regulating EHV-1 by maintaining a repressive epigenetic state would offer a new strategy to combat both lytic and latent EHV-1 infections. Although the extent of epigenetic regulation of the EHV-1 genome has not been determined to date, recent work showed that enforcing a permissive epigenetic state accelerates EHV-1 protein expression and induces productive infection (36). This finding suggests that EHV-1 DNA is subject to epigenetic regulation in horse cells and EHV-1 gene expression can be modulated by altering histone modifications. Given these data, we hypothesize that maintaining a repressive epigenetic state of the EHV-1 genome in the host equine cell would decrease viral load during lytic infection. To test this hypothesis, we investigated how histone tail hypermethylation altered EHV-1 lytic infection of permissive equine cells in vitro.

This study was carried out in accordance with the recommendations of Institutional Animal Care and Use Committee for the use of vertebrates in research, and the protocol approved by the Cornell University Center for Animal Resources and Education.

The objective of this work was to measure the effect of inhibition of the histone demethylase LSD1 on EHV-1 lytic infection in vitro. Adherent primary EFKCs or leukocyte cultures were treated and infected, and the resulting EHV-1 viral load and gene expression quantified.

This work assessed the potential of the LSD1 inhibitor OG-L002 to limit EHV-1 viral load and gene expression during lytic infection in vitro. When used individually, OG-L002 inhibited EHV-1 activity in EFKCs to levels similar to that of ganciclovir treatment. When used in combination, a synergistic effect of ganciclovir and OG-L002 was measured. This observation reinforces the premise that these drugs inhibit viral replication through independent mechanisms, and could be an effective strategy to limit EHV-1 infection. However, under the culture conditions described herein, neither ganciclovir nor OG-L002 treatment of leukocytes suppressed EHV-1 viral load. Studies on the effect of compounds that alter epigenetic modifications on leukocytes have not been published before, and drug dosage or cell penetrance may account for the cell-specific impacts of these compounds on fetal kidney cells relative to equine leukocytes. The lack of EHV-1 inhibition by ganciclovir treatment in leukocytes was also unexpected, particularly in contrast to the results of EFKC cultures. These findings further suggest that differences between the cell types may be due to either cell bioavailability and/or length of exposure in vitro. Studies demonstrating the efficacy of ganciclovir in controlling EHV-1 in equine embryonic lung cells, equine dermal cells, pig kidney cells, or rabbit kidney cells rather than leukocytes have been reported (28, 39–41). Of note, ganciclovir treatment of human lymphoblastoid cells in vitro showed minimal antiviral effect when used for less than 7 days, even at high doses of 20 µg/mL; and, significant effects of ganciclovir treatment were observed at 14 days (42). In addition, the efficacy of antiviral drugs in vivo has not been consistent with in vitro findings, partly due to bioavailability or differential cell sensitivity (43–45). Unlikely these findings were due to lower efficiency of viral replication in leukocytes in comparison to kidney cells, since viral IE expression increased at 24 hpi.

With respect to the cell type dependence of the LSD1 inhibitor OG-L002, there is also the potential that tissue-specific expression or activities of LSD1 isoforms may have resulted in the lack of impact in leukocytes. Several LSD1 isoforms are generated by alternative splicing and these can differ in substrate specificities (H3K4 versus H3K9) and coregulatory protein interactions, resulting in differential gene regulation (46). Nevertheless, inhibition of histone demethylases, including LSD1, have been shown to inhibit HSV, human cytomegalovirus, adenovirus type 5, and varicella zoster virus (VZV) in primary fibroblasts in vitro, and to suppress HSV lytic infection and reactivation in multiple animal models (33–35).

The principal hypothesis of this study was based on previous HSV studies, in which promoting or maintaining an epigenetic repression of the virus suppressed viral infection and reactivation from latency. Therefore, it was anticipated that epigenetic suppression of EHV-1 would decrease viral load during lytic infection and prevent reactivation. Because effects were seen on the transcription of the EHV-1 IE gene in EFKC, the histone demethylase inhibitor OG-L002 was likely to be functioning as expected. In support of this observation, we identified a putative enhancer core element in the EHV-1 genome that may serve as the binding site for the protein complex needed for LSD1 activity, with the motif TAATGAGAT at positions 119,838–119,846 of the EHV-1 genome sequence NC_001491. For HSV and VZV IE gene expression, cellular transcription factors and proteins in the virion recruit host cellular transcriptional coactivator host cell factor-1, LSD1, and JMJD2 to bind the enhancer core element upstream of the IE gene (33). The presence of the enhancer core motif upstream of EHV-1 ORF64 and near other transcription factor binding motifs (SP1, CCAAT, and TATA) supports the likelihood that epigenetic mechanisms similar to those described for HSV1 also regulate EHV-1.

Additional experiments are required to confirm the inhibition of histone demethylation in the equine system, and optimization of the type of drug and dosage for different cell types. Inhibitors of other proteins in the same pathway also exert antiviral activity, and it would be of interest to investigate their impact on EHV-1 infection (47). If testing other compounds would show an effect on viral load in equine leukocytes, the use of respective drug inhibitor controls and additional experiments testing mechanisms of action should be pursued. The use of compounds that alter epigenetic modifications in the host cell rather than targeting viral protein(s) is advantageous in that it is not likely to elicit resistant viral strains. Identifying treatments that simultaneously impair EHV-1 lytic infection and reactivation from latency through a mechanism complementary to current antiviral drugs would be of great value in gaining control of EHV-1 disease severity and spread.

In conclusion, this study showed a favorable effect of the histone demethylase inhibitor OG-L002 in controlling EHV-1 load and gene expression in equine kidney cells, and these results support our hypothesis that maintaining a repressive epigenetic state on the EHV-1 genome in the host equine cell decreases viral load during lytic infection. However, this same effect was not observed in EHV-1 infected equine leukocytes, and additional compounds that alter epigenetic status should be tested.

This study was carried out in accordance with the recommendations of Institutional Animal Care and Use Committee for the use of vertebrates in research, and the protocol approved by the Cornell University Center for Animal Resources and Education.

RT: designed and prepared grant proposal that funded study, including hypothesis, specific aims, and methods; performed molecular experiments; compiled data; performed statistical analyses; and prepared manuscript draft. EŽ: performed part of the molecular experiments; compiled data; performed statistical analyses; participated in discussions; and edited manuscript draft. GW and TK: provided expertise, input and editing of grant proposal preparation (coinvestigator), including methods and experimental design; discussed results and conclusions; and edited manuscript draft. MF: designed and prepared grant proposal that funded study, including hypothesis, specific aims, and methods; performed cellular experiments; compiled data; and edited manuscript draft.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.