The molecular epidemiology of equine herpesvirus 1 (equine abortion virus) in Australasia 1975 to 1989

Vaccination of foals with a modified live, equid herpesvirus-1 gM deletion mutant (RacHΔgM) confers partial protection against infection

Equid herpesvirus-1 (EHV-1) causes respiratory and neurological disease and late gestation abortion in pregnant mares. Current vaccines contain either inactivated or live EHV-1, but fail to provide complete clinical or virological protection, namely prevention of nasopharyngeal shedding and cell-associated viraemia. Thus, the development of novel products, such as modified live virus (MLV) vaccines which stimulate virus-specific, humoral and cell mediated immune responses more effectively remains a priority. Two groups of weaned foals (n = 6 each group) were used in a longitudinal, prospective, experimental study to evaluate immune responses elicited by two vaccinations with a glycoprotein M (gM) deletion mutant of EHV-1 (RacHdeltagM). Following two concurrent intranasal and intramuscular inoculations six weeks apart, vaccinated (8.4 ± 0.2 months old) and control foals (6.2 ± 0.4 months) were challenge infected intranasally with EHV-1 Ab4/8 four weeks after the second vaccination and clinical signs and virological replication measured. Vaccination caused no adverse events, but did stimulate significantly higher complement fixing and virus neutralizing antibodies in serum compared with control foals at either equivalent or pre-vaccination time points. Virus-specific nasopharyngeal antibody levels and cytotoxic T lymphocyte responses were not significantly different between the groups. Following challenge infection, these immune responses were associated with a reduction in clinical signs and virological replication in the vaccinated foals, including a reduction in duration and magnitude of pyrexia, nasopharyngeal shedding and cell-associated viraemia. We conclude that the RacHΔgM MLV primed EHV-1-specific humoral immune responses in weaned foals. However, complete virological protection by vaccination against EHV-1 requires further research.

Genetic diversity of equine herpesvirus 1 isolated from neurological, abortigenic and respiratory disease outbreaks

Equine herpesvirus 1 (EHV‐1) causes respiratory disease, abortion, neonatal death and neurological disease in equines and is endemic in most countries. The viral factors that influence EHV‐1 disease severity are poorly understood, and this has hampered vaccine development. However, the N752D substitution in the viral DNA polymerase catalytic subunit has been shown statistically to be associated with neurological disease. This has given rise to the term “neuropathic strain,” even though strains lacking the polymorphism have been recovered from cases of neurological disease. To broaden understanding of EHV‐1 diversity in the field, 78 EHV‐1 strains isolated over a period of 35 years were sequenced. The great majority of isolates originated from the United Kingdom and included in the collection were low passage isolates from respiratory, abortigenic and neurological outbreaks. Phylogenetic analysis of regions spanning 80% of the genome showed that up to 13 viral clades have been circulating in the United Kingdom and that most of these are continuing to circulate. Abortion isolates grouped into nine clades, and neurological isolates grouped into five. Most neurological isolates had the N752D substitution, whereas most abortion isolates did not, although three of the neurological isolates from linked outbreaks had a different polymorphism. Finally, bioinformatic analysis suggested that recombination has occurred between EHV‐1 clades, between EHV‐1 and equine herpesvirus 4, and between EHV‐1 and equine herpesvirus 8.

Analysis of equid herpesvirus 1 strain variation reveals a point mutation of the DNA polymerase strongly associated with neuropathogenic versus nonneuropathogenic disease outbreaks

Equid herpesvirus 1 (EHV-1) can cause a wide spectrum of diseases ranging from inapparent respiratory infection to the induction of abortion and, in extreme cases, neurological disease resulting in paralysis and ultimately death. It has been suggested that distinct strains of EHV-1 that differ in pathogenic capacity circulate in the field. In order to investigate this hypothesis, it was necessary to identify genetic markers that allow subgroups of related strains to be identified. We have determined all of the genetic differences between a neuropathogenic strain (Ab4) and a nonneuropathogenic strain (V592) of EHV-1 and developed PCR/sequencing procedures enabling differentiation of EHV-1 strains circulating in the field. The results indicate the occurrence of several major genetic subgroups of EHV-1 among isolates recovered from outbreaks over the course of 30 years, consistent with the proposal that distinct strains of EHV-1 circulate in the field. Moreover, there is evidence that certain strain groups are geographically restricted, being recovered predominantly from outbreaks occurring in either North America or Europe. Significantly, variation of a single amino acid of the DNA polymerase is strongly associated with neurological versus nonneurological disease outbreaks. Strikingly, this variant amino acid occurs at a highly conserved position for herpesvirus DNA polymerases, suggesting an important functional role.

Natural recombinant between equine herpesviruses 1 and 4 in the ICP4 gene

Equine herpesvirus 1 (EHV-1) is a pathogen causing rhinopneumonia in young horses, abortion in mares, and myeloencephalitis in adult horses. Two types, EHV-1 P and EHV-1 B, have recently been dominant among 16 electropherotypes. EHV-1 P and EHV-1 B viruses were compared by long and accurate polymerase chain reaction (LA-PCR) and restriction fragment length polymorphism (RFLP) analysis. Differences in restriction sites were found to be focused in ORF64, which encodes the infected cell protein 4 (ICP4), and downstream of the ICP4 gene. The 3 ' -end and downstream of ICP4 gene of EHV-1 B were found to be replaced by the corresponding region of EHV-4, indicating that EHV-1 B is a naturally occurring recombinant virus between progenitors of EHV-1 P and EHV-4. This is the first report showing a natural interspecies recombinant in alphaherpesviruses.

Detection of equine herpesvirus type 1 using a real-time polymerase chain reaction

Equid herpesvirus 1 (EHV1) is a major disease of equids worldwide causing considerable losses to the horse industry. A variety of techniques, including PCR have been used to diagnose EHV1. Some of these PCRs were used in combination with other techniques such as restriction enzyme analysis (REA) or hybridisation, making them cumbersome for routine diagnostic testing and increasing the chances of cross-contamination. Furthermore, they involve the use of suspected carcinogens such as ethidium bromide and ultraviolet light. In this paper, we describe a real-time PCR, which uses minor groove-binding probe (MGB) technology for the diagnosis of EHV1. This technique does not require post-PCR manipulations thereby reducing the risk of cross-contamination. Most importantly, the technique is specific; it was able to differentiate EHV1 from the closely related member of the Alphaherpesvirinae, equid herpesvirus 4 (EHV4). It was not reactive with common opportunistic pathogens such as Escherichia coli, Klebsiella oxytoca, Pseudomonas aeruginosa and Enterobacter agglomerans often involved in abortion. Similarly, it did not react with equine pathogens such as Streptococcus equi, Streptococcus equisimilis, Streptococcus zooepidemicus, Taylorella equigenitalis and Rhodococcus equi, which also cause abortion. The results obtained with this technique agreed with results from published PCR methods. The assay was sensitive enough to detect EHV1 sequences in paraffin-embedded tissues and clinical samples. When compared to virus isolation, the test was more sensitive. This test will be useful for the routine diagnosis of EHV1 based on its specificity, sensitivity, ease of performance and rapidity.

Genomic diversity among equine herpesvirus-4 field isolates

Infection with equine herpesvirus-4 (EHV-4) is a major cause of respiratory tract disease, equine rhinopneumonitis, in horses. Although the full sequence of EHV-4 has been reported, genomic differences among EHV-4 field isolates have not yet been characterized. In this study, the genomic diversity between 23 Japanese EHV-4 isolates was analyzed by digestion with restriction endonucleases (BamHI, BgIII, EcoRI, SacI, and SalI) and polymerase chain reaction (PCR). The restriction endonuclease digestion patterns of the EHV-4 field isolates showed distinct differences which included mobility shifts of some fragments as well as loss and/or gain of fragments. Two EHV-4 genes containing repeat sequences, ORFs 24 and 71, were amplified by PCR and the amplified fragments were compared among the field isolates. The sizes of the amplified fragments varied among epizootiologically unrelated isolates, while the fragments of related isolates had the same size. The observed genomic diversity among EHV-4 field isolates may be a useful tool for epidemiological study of equine rhinopneumonitis by EHV-4 infection.

Isolation of equine herpesvirus-1 lacking glycoprotein C from a dead neonatal foal in Japan

We isolated a variant equine herpesvirus-1 (EHV-1), strain 5089, from the lung of a dead neonatal foal in Japan and characterized the biological nature of the virus. The virus spread in cultured cells mainly by cell-to-cell infection, unlike wild-type EHV-1, which spreads efficiently as a cell-free virus. The virus titer in cultured supernatant and the intracellular virus titer were low compared to those of wild-type EHV-1. Heparin treatment of the virus had no effect on viral infectivity in cell culture. Glycoprotein C (gC) was not detected by Western blotting and fluorescent antibody tests in 5089 virions and 5089-infected cells, respectively. RT-PCR analysis revealed that the expression level of 5089 gC mRNA was reduced considerably compared to that of wild-type EHV-1. Sequencing analysis of the 5089 gC coding region showed a point mutation in the promoter region of the gC open reading frame. However, the mutation did not affect the promoter activity. These results suggested that the lack of gC in 5089 virions might be one of the reasons for spread of the virus by cell-to-cell infection and that gC mRNA expression might not be activated efficiently due to factors other than the mutation in the gC promoter region.

Outbreak of equine herpesvirus type 1 myeloencephalitis: new insights from virus identification by PCR and the application of an EHV-1 -specific antibody detection ELISA

Five of 10 pregnant, lactating mares, each with a foal at foot, developed neurological disease. Three of them became recumbent, developed complications and were euthanased; of the two that survived, one aborted an equine herpesvirus type 1 (EHV-1)-positive fetus 68 days after the first signs were observed in the index case and the other gave birth to a healthy foal on day 283 but remained ataxic and incontinent. The diagnosis of EHV-1 myeloencephalitis was supported by postmortem findings, PCR identification of the virus and by serological tests with an EHV-1-specific ELISA. At the time of the index case, the 10 foals all had a heavy mucopurulent nasal discharge, and PCR and the ELISA were used to detect and monitor EHV-1 infection in them. The status of EHV-1 infection in the five in-contact mares was similarly monitored. Sera from three of the affected mares, taken seven days after the index case were negative or had borderline EHV-1-specific antibody titres. In later serum samples there was an increase in the titres of EHV-1-specific antibody in two of the affected mares. In contrast, sera from the five unaffected in-contact mares were all EHV-1-antibody positive when they were first tested seven or 13 days after the index case.

Polymorphism of open reading frame 71 of equine herpesvirus-4 (EHV-4) and EHV-1

Open reading frame (ORF) 71 genes of both equine herpesvirus-1 (EHV-1) and EHV-4 encode a unique glycoprotein, which has been described to vary in molecular mass from 200 to 450 kDa. Using PCR and nucleotide sequence analysis, it was shown that the ORF 71 genes of EHV-1 and EHV-4 are polymorphic due to a variable number of reiterated sequences in two regions, designated regions A and B. Region A was threonine-rich and was located near the N terminus. Region B comprised a 38 amino acid repeat near the C terminus that expanded following cell culture adaptation. Western blot analysis of viruses showed that EHV-4 gp2 was modified by glycosylation and that variation in region A resulted in the marked differences in the molecular mass of EHV-4 gp2.

Identification of equine herpesviruses 1 and 4 by polymerase chain reaction

OBJECTIVE

To develop and validate specific, sensitive and rapid (< 8 hour) diagnostic tests using polymerase chain reaction (PCR) for the diagnosis of abortion and respiratory disease caused by equine herpesvirus 1 (EHV1; equine abortion virus) and EHV4 (equine rhinopneumonitis virus).

DESIGN

Primer sets based on nucleotide sequences encoding glycoprotein H (gH) of EHV1 and gB of EHV4 were designed and used in single round and second round (seminested) PCRs, and in a multiplex PCR for the diagnosis of EHV1 and EHV4 infections.

METHODS

Oligonucleotide primers were designed for each virus, PCR conditions were defined and the specificity and sensitivity of the assays were determined. The tests were applied to tissue samples from aborted equine foetuses and to nasopharyngeal swabs from horses with acute febrile respiratory disease.

RESULTS

Individual single round and a second round (seminested) EHV1 and EHV4 PCRs were specific in that EHV1 primers amplified all (n = 30) EHV1 isolates and did not amplify EHV4. Similarly EHV4 primers amplified all (n = 6) EHV4 isolates and did not amplify EHV1. Both PCRs were sensitive in that the first round EHV1 PCR detected 1220 molecules of EHV1 plasmid DNA and the first round EHV4 PCR detected 7280 molecules of EHV4 plasmid DNA. The EHV1 second round PCR was 100 times more sensitive in that it detected 12 molecules of EHV1 DNA and the EHV4 second round PCR was 1000 times more sensitive in that it detected 8 molecules of EHV4 DNA. There was a high correlation between detection of EHV1 by virus isolation and PCR when tissue samples from 71 aborted foetuses were examined; all samples positive by virus isolation were positive by PCR. Similarly the EHV4 PCR was at least as sensitive as virus isolation when applied to nasaopharyngeal swabs from horses with respiratory disease in that all samples positive by virus isolation were also positive by PCR.

CONCLUSION

Individual single round and second round (seminested) PCRs and a seminested multiplex PCR were developed that enabled reliable, rapid detection of EHV1 and EHV4 in aborted foetal tissues and nasopharyngeal swab samples.

Differentiation and genomic and antigenic variation among fetal, respiratory, and neurological isolates from EHV1 and EHV4 infections in The Netherlands

Ten monoclonal antibodies (MAbs) were produced against equine herpes virus type 1 (EHV1). Two appeared type-specific, while the other eight were directed against epitopes common to both EHV1 and EHV4. Two MAbs directed against the glycoprotein gp2 recognized linear epitopes, as demonstrated by Western blotting. With pools of type-specific MAbs, 282 field isolates were typed in an immunoperoxidase monolayer assay (IPMA). From a total of 254 fetal or neonatal isolates, 244 (96%) were typed as EHV1, whereas 14 out of 15 (93%) respiratory tract isolates were typed as EHV4. Surprisingly, 3 out of 13 isolates (23%) originating from horses with neurological disease were typed as EHV4. No antigenic differences were found among 75 randomly selected EHV1 field isolates, using the panel of ten MAbs and six additional MAbs, directed against gp2, gB, or gC. Typing by restriction endonuclease analysis with BamHI corresponded completely with that of MAb analysis. There was a remarkable degree of uniformity in BamHI restriction patterns, with 90% of the investigated EHV1 isolates belonging to the 1P electropherotype. Among 30 randomly selected EHV1 isolates we could not identify the EHV1.1B electropherotype, which has been the predominant electropherotype in Kentucky since 1982. Mobility differences were seen in fragments originating from the repeat regions. These differences were not caused by heterologous cell passage, since all viruses were passaged in equine cell systems.

Restriction endonuclease analysis of equine herpesvirus-1 isolates recovered in Ontario, 1986-1992, from aborted, stillborn, and neonatal foals

Ninety-two equine herpesvirus type 1 isolates were recovered from aborted, stillborn, or neonatal foals from Ontario, Canada, from 1986 to 1992. From this total, 32 strains were randomly chosen for further study. Four or 5 isolates from each winter were selected, each from a different premises, and characterized by restriction enzyme analysis using BamHI, KpnI, BglII, HindIII, and EcoRI. Additional isolates from 2 premises and from a zebra foal were also assessed. For the strains isolated in 1986 and 1989-1992, the DNA pattern of 18 strains was similar to that of type 1P (Kentucky D) for BamHI and KpnI. None of the 32 strains studied could be differentiated by HindIII or EcoRI. Using BglII, an inconsistent fragment pattern and distribution were observed. Of the 8 strains isolated in 1987 and 1988, 7 were assigned into the 1B prototype group. The geographic distribution of 17 type 1P and 12 1B isolates was random across southern Ontario. These findings suggest that both electropherotypes can be recovered from horses in Ontario. The patterns of the additional equine isolates from the same premises were identical. The zebra isolate was different from the prototype equine herpesvirus type 1 and type 4 patterns and from all other equine isolates.

The nucleotide sequence of asinine herpesvirus 3 glycoprotein G indicates that the donkey virus is closely related to equine herpesvirus 1

The nucleotide sequence of the glycoprotein G (gG) homologue of asinine herpesvirus 3 (AHV3), a respiratory alphaherpesvirus of donkeys, was determined. The AHV3 gG gene consists of 1233 base pairs (bp) and codes for a predicted protein of 411 amino acids. This is identical in size to the equine herpesvirus 1 (EHV1) gG gene and 6 amino acids longer than the equine herpesvirus 4 (EHV4) gG gene. The predicted amino acid sequence of AHV3 gG has characteristics of a class 1 membrane protein. The amino acid sequence of AHV3 gG shows 92% and 60% identity to EHV1 gG and EHV4 gG respectively. Two regions within the gG amino acid sequences of EHV1 and EHV4 were previously defined, an N-terminal constant region and an immunodominant highly variable region located toward the C-terminus. In the corresponding constant region of AHV3 gG there was 96% and 75% amino acid identity with EHV1 and EHV4 gGs respectively. In the variable region, there was 73% and 24% identity respectively. Phylogenetic analyses using the gG nucleotide sequences indicated that AHV3 is much closer in evolutionary distance to EHV1 than either virus is to EHV4. These findings provide additional support for the view that AHV3, or another closely related virus, may be the progenitor of EHV1 and has adapted to horses in relatively recent times.

Assessment of statistical procedures used in papers in the Australian Veterinary Journal

One hundred and thirty-three papers (80 Original Articles and 53 Short Contributions) of 279 papers in 23 consecutive issues of the Australian Veterinary Journal were examined for their statistical content. Only 38 (29%) would have been acceptable to a statistical referee without revision, revision would have been indicated in 88 (66%), and the remaining 7 (5%) had major flaws. Weaknesses in design were found in 40 (30%), chiefly in respect to randomisation and to the size of the experiment. Deficiencies in analysis in 60 (45%) were in methods, application and calculation, and in the failure to use appropriate methods for multiple comparisons and repeated measures. Problems were detected in presentation in 44 (33%) of papers, with insufficient information about the data or its statistical analysis and presentation of statistics (appropriate missing or inappropriate shown) the main problems. Conclusions were considered to be inconsistent with the analysis in 35 (26%) of papers, due mainly to their interpretation of the results of significance testing. It is suggested that statistical refereeing, the publication of statistical guidelines for authors and statistical advice to Animal Experimentation Ethics Committees could all play a part in achieving improvement.

Development of PCR assays to detect genetic variation amongst equine herpesvirus-1 isolates as an aid to epidemiological investigation

A search for variable restriction sites has been carried out for equine herpesvirus-1 (EHV-1) in an attempt to develop markers which can be used to group epidemiologically related viruses into groups, and to learn more about the dynamics of EHV-1 disease. Crude viral DNA extracts of EHV-1, prepared by Hirt extraction, were digested with AluI, HaeIII, or RsaI, and Southern blotted following electrophoresis. DNA fingerprints, produced by probing the Southern blots with the EHV-1 EcoR1-I fragment, separated 56 isolates into 16 groups. The variable sites within the EcoR1-I fragment were mapped approximately using fragments from within EcoR1-I, and the precise location of the variable sites determined from the DNA sequence of this fragment. Oligonucleotide primers flanking the variable sites were synthesized, and used in PCR assays to detect variable fragments. The AluI variable fragment was found to result from the presence or absence of a single AluI site. In contrast, the variable bands seen with HaeIII and RsaI, resulted from variation in the copy number of two tandemly repeated sequences, one of which had not previously been recognized. In addition, HaeIII digests of EHV-1 isolates probed with the glycoprotein B (gB) gene of EHV-1 also separated isolates into two groups. The variable HaeIII site was mapped towards the 5'-end of the gB gene and a PCR assay established. The distribution of the variable AluI site within the EcoR1-I fragment and the HaeIII site within the gB gene were estimated on a large number of clinical isolates using PCR on unpurified viral tissue culture medium. Both sites had a good distribution and together with additional variable sites should provide the basis for the rapid DNA fingerprinting of EHV-1 isolates.

A type-specific serological test to distinguish antibodies to equine herpesviruses 4 and 1

We describe a type-specific ELISA, which distinguishes antibody to equine herpesvirus 4 (EHV4; equine rhinopneumonitis) and EHV1 (equine abortion virus) thereby identifying horses that have been infected with either or both of these antigenically related viruses. The antigens used are parts of the EHV4 and EHV1 glycoprotein G (gG) homologues expressed in E. coli as fusion proteins [Crabb and Studdert, 1993: J Virol 67: 6332-6338). The expressed proteins comprise corresponding regions of the gG molecules that are highly divergent and encompass strong, typespecific epitopes. Plasma samples from 97 Thoroughbred and 174 Standardbred horses were tested, all of which were unvaccinated. All horses were strongly EHV4 ELISA positive while 30% were EHV1 ELISA positive. The type-specificity of the EHV1 gG antigen was tested in cross-absorption experiments and it was found that 96% (66 of 69) of EHV1 ELISA positive horses were true EHV1 antibody positives. It was also shown that 100% (26 of 26) horses known to have been exposed to EHV1, either by infection or immunisation with EHV1, had significant levels of antibody against the EHV1 gG antigen (i.e., all horses recognised the EHV1 epitope(s) contained within this molecule). Maintenance of EHV1 gG antibody was examined by testing sera obtained from mares four years after confirmed EHV1 abortion. Seven out of 10 of these mares remained EHV1 ELISA positive. In summary, the ELISA is highly specific and is sufficiently sensitive to detect all horses previously infected with EHV4 and most previously infected with EHV1.

Epitopes of glycoprotein G of equine herpesviruses 4 and 1 located near the C termini elicit type-specific antibody responses in the natural host

Specific serological diagnosis of equine herpesvirus 4 (EHV4; equine rhinopneumonitis virus) and EHV1 (equine abortion virus) hitherto has not been possible because of extensive antigenic cross-reactivity between these two closely related but distinct viruses. Recently, we identified EHV4 glycoprotein G (gG) and characterized it as a type-specific, secreted glycoprotein (B. S. Crabb, H. S. Nagesha, and M. J. Studdert, Virology 190:143-154, 1992). This paper shows that EHV1 gG also possesses type-specific epitopes and describes the localization of strong, type-specific epitopes to the apparently corresponding and highly variable regions comprising amino acids 287 to 382 of EHV4 gG and 288 to 350 of EHV1 gG. Fusion proteins expressing these variable regions reacted strongly and type specifically with sera from four foals, three of which were colostrum-deprived, specific-pathogen-free foals, whose history with respect to exposure to EHV4 or EHV1 was well-defined. These antigens provided the basis for the development of a single-well diagnostic enzyme-linked immunosorbent assay to distinguish horses infected with EHV4, EHV1, or both. Such a type-specific test provides for the first time the opportunity to differentiate antibodies to these viruses, and it has, therefore, important implications for understanding the epidemiology of these equine pathogens. Evidence for the existence of EHV1 in Australia 10 years prior to the first confirmed case of EHV1 abortion is presented.

Comparison of the restriction patterns of equine herpesvirus (EHV-1) strains isolated for eight years in France

The equine herpes viruses strains (EHV) isolated from organs of aborted foetuses or from nasal swabs have been analysed by comparison of their restriction endonucleases patterns using two enzymes, Bam HI and Pst I. The majority of the clinical samples came from the west part of France ("Normandie") after abortions or respiratory disorders. All the viruses isolated were EHV-1 strains whose patterns show considerable homogeneity although some differences can be described. The genomic DNAs of the same twenty strains have been digested by the Pst I enzyme, which induced a great number of restriction fragments. It allows a more precise epidemiological study between strains isolated in the same studs with different Bam HI patterns or between strains with identical Bam HI profiles but with distinct respiratory or abortigenic pathogenicity. No strain isolated from aborted foetuses or nasal swabs presented the vaccinal pattern.

Cloning and restriction endonuclease mapping of the genome of an equine herpesvirus 4 (equine rhinopneumonitis virus), strain 405/76

Purified virion DNA of an Australian isolate of equine herpesvirus 4(EHV 4.405/76) was digested with restriction enzymes and the DNA fragments were cloned into pUC19. The resulting recombinant plasmid library, representing 92% of the virus genome, was used in hybridization analyses to construct restriction maps for BamHI, EcoRI, and SalI for the EHV4 genome. The results show that the genome of EHV 4.405/76 was approximately 145 kb and comprised a unique long (UL) region of 112 kb and a unique short (US) region of 12.4 kb. US is flanked by an internal and terminal repetitive sequence (IRS and TRS) of about 10.3 kb. The BamHI and EcoRI restriction maps are similar to those previously published for an English isolate EHV 4.1942 strain although some differences such as location of an additional fragment and changes in positions of two other small fragments were found.

Identification of equine herpesvirus 4 glycoprotein G: a type-specific, secreted glycoprotein

Equine herpesvirus 4 (EHV4) glycoproteins of M(r) 63K and 250K were identified in the supernatant of infected cell cultures. The 63K glycoprotein was type-specific; that is, it reacted with monospecific sera from horses that had been immunized or infected with EHV4, but not with monospecific sera from horses immunized or infected with EHV1, a closely related alphaherpesvirus. It was postulated that the secreted protein may be the homologue of similarly secreted glycoproteins of herpes simplex virus 2 glycoprotein G (HSV2 gG) and pseudorabies virus (PRV) gX, which is the homologue of HSV2 gG. The US region of the EHV4 genome, toward the internal repeat structure, was sequenced. Four open reading frames (ORFs) were identified of which ORF4 showed 52% similarity to the gene-encoding PRV gX in a 650-nucleotide region. ORF4 coded for a primary translational product of 405 amino acids which has a predicted size of 44K. The amino acid sequence of ORF4 showed 28% identity with PRV gX and 16% identity with HSV2 gG, although significantly greater identity was observed in the N-terminal region including the conservation of 4 cysteine residues. Accordingly, we designate ORF4 as EHV4 gG. The predicted amino acid sequence of the EHV4 gG showed characteristics of an envelope glycoprotein. Expression of the entire EHV4 gG gene in the bacterial expression vector pGEX-3X produced a type-specific fusion protein of M(r) 70K of which the gG portion composes 43K. Antibody that was affinity purified from selected portions of Western blots containing the 70K gG fusion protein reacted with the 63K secreted glycoprotein. Conversely, antibody affinity purified to the 63K secreted product reacted with the 70K gG fusion protein. These results showed that the EHV4 63K secreted glycoprotein was EHV4 gG, the third alphaherpesvirus gG homologue known to be, at least in part, secreted. The type-specificity of this glycoprotein provides, for the first time, the opportunity to differentiate between antibodies present in polyclonal sera from EHV4, EHV1, and dual-infected horses and this has important implications for understanding the epidemiology of these viruses.