Development of an Equine Herpesvirus Type 4-Specific Enzyme-Linked Immunosorbent Assay Using a B-Cell Epitope as an Antigen

Abstract

The equine herpesvirus type 4 (EHV-4)-specific region of glycoprotein G has served as an antigen for serodiagnosis and seroepizootic studies of EHV-4 infection (B. S. Crabb and M. J. Studdert, J. Virol. 67:6332-6338, 1993; S. Yasunaga, K. Maeda, T. Matsumura, K. Kai, H. Iwata, and T. Inoue, J. Vet. Med. Sci. 60:1133-137, 1998; S. Yasunaga, K. Maeda, T. Matsumura, T. Kondo, and K. Kai, J. Vet. Med. Sci. 62:687-691, 2000). Here we identified a major B-cell epitope in the type-specific region of EHV-4 and applied it as an antigen in enzyme-linked immunosorbent assays (ELISAs). A 24-amino-acid repeat sequence expressed as a glutathione S-transferase fusion protein specifically reacted as well as the type-specific region with sera from foals infected with EHV-4. Five synthetic peptides (12-mer peptides) in the repeat sequence were included as ELISA antigens. The results indicated that the 12-mer peptide MKNNPIYSEGSL contained a major B-cell epitope specific for EHV-4 infection. Inclusion of this 12-mer peptide in ELISAs for an epidemiological study specifically detected EHV-4 infection in the field. These results indicated that the 12-mer epitope was responsible for the type-specific antibody response and therefore is useful for seroepizootic studies and serodiagnosis of EHV-4 infection.

Equine herpesvirus type 4 (EHV-4) causes a pyrogenic respiratory disease in young horses that have been bred and reared in farms (3), and EHV-1 is a major cause of respiratory disease among racehorses during the winter in Japan (5, 7, 8). Since these viruses are serologically cross-reactive, horses infected with EHV-4 have antibodies that function against EHV-1 and are thus somewhat protected from EHV-1 infection. However, many EHV-4-infected racehorses in Japanese breeding and rearing farms have also become infected with EHV-1 in training centers during the winter. Such infection has caused fever and occasional disorders of the central nervous system (5, 7). Therefore, it is important to understand the pathogenesis of EHV-4 to prevent infections with both EHV-1 and EHV-4.

The C-terminal regions of EHV-1 and EHV-4 glycoprotein G (gG) are specifically recognized by sera from horses infected with the respective virus (2), and enzyme-linked immunosorbent assays (ELISAs) with type-specific regions as antigens can serologically distinguish infections with these viruses (1). A modification of a type-specific ELISA for Japanese isolates was useful for seroepizootic studies and for the serodiagnosis of infections with both types of EHV (8). This method did not detect any antibody induced by inactivated vaccine against EHV-1 but could distinguish between EHV-1 and EHV-4 infections in vaccinated and nonvaccinated horses (9). It was also reported that levels of immunoglobulin Ga (IgGa) antibody against the EHV-4-specific region increased as much as those of IgGb antibody after EHV-4 infection but decreased more rapidly than those of IgGb (6). Therefore, analysis of the antibody response to the EHV-4-specific region seems to be very important for understanding the equine immune response. However, since the purification of these antigens is labor-intensive and requires expression in Escherichia coli as well as affinity chromatography purification, a simpler procedure is desirable.

Here we identified a major B-cell epitope in the type-specific region of EHV-4 gG and applied it to ELISAs for seroepizootic studies and for the serodiagnosis of EHV-4 infection in the field.

MATERIALS AND METHODS

Virus and cells.

EHV-4 reference strain TH20p plaque purified from TH20 isolated from a colt with respiratory disease (4) was propagated and titrated in fetal horse kidney cells.

Serum samples.

Sera were collected at various times from three foals that were intranasally inoculated with EHV-4 strain TH20p. Foal 1 had been deprived of colostrum, and foals 2 and 3 had received colostrum after birth.

Field sera were collected during the acute and convalescent phases from foals or yearlings with pyrexia due to EHV-4 infection between 1988 and 1989. We also isolated EHV-4 from nasal swab samples from these horses during the acute phase. The controls were paired sera from 14 horses with pyrexia caused by EHV-1 infection between 1989 and 1996.

Epidemiological study.

Paired sera were collected during October 1997 and May 1998 from 40 racehorses (9) that had been inoculated three times with inactivated EHV-1 vaccine between October 1997 and February 1998.

DNA extraction.

Fetal horse kidney cells infected with EHV-4 TH20p were incubated with 1% sodium dodecyl sulfate (SDS) and proteinase K (0.1 mg/ml) in 0.1 M Tris-HCl- 0.1 M NaCl- 5 mM EDTA (pH 9.0) at 37°C overnight. Following extraction with phenol and chloroform-isoamyl alcohol (24:1), ethanol-precipitated DNA was dissolved in distilled water.

Expression of GST fusion proteins.

The type-specific region (egG4P) of strain TH20p was expressed as a fusion protein with glutathione S-transferase (GST) as described previously (8). A repeat sequence encoding 24 amino acids (egG4R) was amplified from TH20p by PCR with primers gG4-977F (5′-GAATTCATGAAGAATAACCCTATT-3′) and gG4-1050R (5′-GAGCTCCACACCTACGACAGTAG-3′). The amplified fragment was cloned into TA vector pCRII (Invitrogen), and the nucleotide sequence was confirmed. Plasmids containing the amplified fragment were digested with EcoRI and XhoI and cloned into the EcoRI and XhoI sites of pGEX-5X1. Recombinant plasmids were isolated from E. coli, and GST fusion proteins were purified as described previously (8).

SDS-PAGE and immunoblotting.

Polypeptides were separated by SDS- 15% polyacrylamide gel electrophoresis (PAGE) and stained with 0.006% Coomassie brilliant blue R-250 (Nacalai Tesque, Osaka, Japan) in 10% acetic acid.

Electrophoretically separated polypeptides were electrotransferred to polyvinylidene difluoride membranes (Immobilon; Millipore) and incubated for 2 h at room temperature with diluted horse serum infected with EHV-1. The membranes were washed three times and incubated with peroxidase-conjugated anti-horse immunoglobulins (EY Laboratories) at room temperature for 1 h. The reaction was visualized with the substrate diaminobenzidine-hydrogen peroxidase.

ELISAs.

ELISAs with GST fusion proteins were carried out as described previously (8). When synthetic peptides were used as the antigens, 100 μl of phosphate-buffered saline containing a peptide (10 μg/ml) was applied to wells (Maxisorp; Nunc) and incubated at 4°C overnight. The control was phosphate-buffered saline without a peptide. All reactions were carried out in duplicate, and the average absorbance was calculated.

Synthetic peptides.

Five 12-mer peptides (G1 to G5) were synthesized on the basis of the amino acid sequence of strain TH20p gG. Peptides overlapping by 8 or 10 residues were constructed by using an ACT 350 multiple-peptide synthesizer (Advanced ChemTech) as described by the manufacturer.

Statistics.

Values for the mean and the standard deviation were calculated and analyzed by Student's t test. Significance was established at a P value of <0.05.

RESULTS

Expression of one region of the repeat sequence.

The molecular masses of the EHV-4-specific region (egG4P; 93 amino acids) and a portion of the repeat sequence (egG4R; 24 amino acids) in the EHV-4-specific region expressed as fusion proteins with GST (GST-egG4P and GST-egG4R, respectively) were 36 and 30 kDa, respectively (Fig. 1A). Immunoblotting with serum from the foal infected with EHV-4 TH20p (foal 1) detected both GST-egG4P and GST-egG4R but not GST (Fig. 1B).

FIG. 1.

Expression of the type-specific region and a stretch of the repeat sequence in the type-specific region of EHV-4 TH20p. (A) A type-specific region (GST-egG4P) and a repeat sequence (GST-egG4R) were expressed as GST fusion proteins in E. coli. Whole bacterial proteins were separated by SDS-PAGE and stained with Coomassie brilliant blue. (B) Whole bacterial proteins were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes. Proteins were detected with diluted serum from a horse experimentally infected with EHV-4 (horse 1) and peroxidase-conjugated anti-horse IgG. Molecular masses are shown at the left.

Comparison of antibody responses to egG4P and egG4R.

We performed ELISAs with GST-egG4P and GST-egG4R as antigens together with serum samples that were serially collected from three foals experimentally infected with EHV-4 TH20p (foals 1, 2, and 3). The ELISAs with egG4P and egG4R were almost equally sensitive (Fig. 2). These findings suggested that egG4R contains a major immunogenic epitope in the type-specific region (egG4P).

FIG. 2.

Reactivities of purified GST-egG4P and GST-egG4R and synthetic peptide G1 against sera. A type-specific region (egG4P) and a repeat sequence (egG4R) were expressed as GST fusion proteins in E. coli and purified by chromatography on glutathione-Sepharose 4B. Purified proteins and synthetic peptide G1 served as ELISA antigens. Primary antibodies were serum samples that were serially collected from EHV-4-infected horses 1, 2, and 3 (left, middle, and right panels, respectively) and diluted 1:800 for the ELISA with GST-egG4P (filled circles) and GST-egG4R (open circles) or 1:500 for the ELISA with G1 (filled triangles). The ELISA with G1 included sera collected at days 0, 7, 10, 14, and 28 postinfection. OD405nm, optical density at 405 nm.

Identification of a major epitope in the repeat region.

We performed ELISAs with 12-mer peptides (G1 to G5) overlapping by 8 or 10 amino acids to identify the major B-cell epitope in egG4R. We found that peptide G1 (MKNNPIYSEGSL) significantly reacted with sera from EHV-4-infected foals (foals 1, 2, and 3) (P < 0.05), whereas the other four peptides (G2 to G5) did not (Fig. 3).

FIG. 3.

Reactivities of 12-mer synthetic peptides against EHV-4-infected horse sera. ELISAs were carried out with 12-mer peptides (G1 to G5) with 8 or 10 overlapping amino acids and GST-egG4R as antigens. Sera collected from three foals (1, 2, and 3) that were experimentally infected with EHV-4 TH20p were diluted 1:1,000. Peptide reactivity is shown as the mean absorbance at 405 nm; horizontal error bars represent the standard deviation of the mean. Repeat sequences are shown by bold letters, and the sequence EG in G5 is a spacer sequence between repeat sequences.

ELISAs with sera from foals experimentally infected with EHV-4 TH20p.

To determine whether peptide G1 is recognized by sera from horses experimentally infected with EHV-4, we performed ELISAs with sera from foals 1, 2, and 3. We detected antibodies specific for G1 in all horses, and the antibody titers were significantly increased at 14 days after EHV-4 infection (Fig. 2 and Table 1). The antibody response was more effective and faster in samples from foal 1, which did not receive colostrum after birth, than in samples from foals 2 and 3, which did (Fig. 2 and Table 1).

TABLE 1.

ELISA titers of sera serially collected from three EHV-4-infected foals

Horsea	Antigen	Titerb at the following day after infection:
		0	7	10	14	28
1	egG4P	<1:200	1:400	1:6,400	1:12,800	1:12,800
	egG4R	<1:200	1:800	1:12,800	1:12,800	1:12,800
	G1	<1:500	1:500	1:4,000	1:4,000	1:4,000
2	egG4P	1:800	1:800	1:400	1:3,200	1:3,200
	egG4R	1:800	1:800	1:400	1:3,200	1:3,200
	G1	<1:125	<1:125	<1:125	1:500	1:500
3	egG4P	1:400	1:200	1:200	1:1,600	1:6,400
	egG4R	1:200	1:200	1:200	1:1,600	1:6,400
	G1	<1:125	<1:125	<1:125	1:250	1:1,000

^aThree foals (1, 2, and 3) were experimentally infected with EHV-4 strain TH20p. One foal (foal 1) was deprived of colostrum, and two foals (foals 2 and 3) received colostrum after birth.

^bSera were serially diluted twofold and then added to antigen-coated wells. The ELISA titer was expressed as the maximum dilution that showed a significant difference in the absorbances (over 0.1) between antigen-coated and control wells.

Application of ELISAs with peptide G1 to the diagnosis of EHV-4 infection.

We performed ELISAs with paired sera from seven horses with pyrexia due to EHV-4 infection in the field to determine whether peptide G1 can diagnose EHV-4 infection. The amount of antibodies specific for G1 significantly increased during the convalescent phase (P < 0.001) (Fig. 4A), but none of the sera from 14 horses with pyrexia due to EHV-1 infection reacted with G1 (Fig. 4B).

FIG. 4.

Reactivity of peptide G1 against sera from horses with pyrexia due to EHV-4 or EHV-1 infection. ELISAs were carried out with G1 as an antigen and paired sera collected during acute and convalescent phases from horses with pyrexia caused by EHV-4 (A) or EHV-1 (B) infection and diluted 1:125. OD405nm, optical density at 405 nm.

Application of ELISAs with peptide G1 to epidemiological studies of EHV-4 infection.

To determine whether ELISAs with peptide G1 could be useful for epidemiological studies of EHV-4 infection, paired sera from 40 racehorses were collected during October 1997 and May 1998. These horses had been inoculated three times with inactivated EHV-1 vaccine between October 1997 and February 1998. The results of the ELISAs indicated that the absorbance was significantly increased in samples from two of the racehorses (Fig. 5). These results agreed with previous findings, obtained with egG4P as an antigen, indicating that these two horses were infected with EHV-4 during the same period (9). The absorbance during October 1997 tended to be somewhat higher than that during May 1998 in the other horses, but the absorbance did not increase significantly despite EHV-1 vaccination during this observation period (Fig. 5).

FIG. 5.

Comparison of antibodies to peptide G1 in October 1997 and May 1998. Paired sera were collected from 40 racehorses in October 1997 and May 1998. Racehorses had been inoculated three times with inactivated EHV-1 vaccine between October 1997 and February 1998. Sera were diluted 1:125 for the ELISAs with G1 as an antigen. x and y axes show the absorbance at 405 nm of sera collected in October 1997 and May 1998, respectively. Open circles indicate two horses that were naturally infected with EHV-4 during this observation period. The trend line represents the relationship in terms of absorbance between October 1997 and May 1998.

DISCUSSION

We identified a major immunogenic B-cell epitope in the type-specific region of EHV-4 gG. The type-specific region of a laboratory strain (TH20p) contained tandem repeats, and one stretch of these (egG4R) was expressed and used as an ELISA antigen. The ELISAs with GST-egG4P and GST-egG4R were almost equally sensitive, indicating that one stretch of the repeat sequence (egG4R) is immunodominant in the type-specific region (Fig. 1 and 2 and Table 1). A more detailed investigation of this region showed that 12-mer peptide G1 was a major immunogenic B-cell epitope that reacted with sera from EHV-4-infected horses (Fig. 3). Although synthetic peptide G1 reacted with all sera from EHV-4-infected horses, the antibody response was a little less effective than that to either GST-egG4P or GST-egG4R (Fig. 2 and 3 and Table 1). These results suggested that although one stretch of the repeat sequence (egG4R) contains another B-cell epitope(s), the antibody response to G1 was sufficiently powerful to detect EHV-4 infection in horses.

Figure 2 and Table 1 show that the antibody responses of foals 2 and 3, which had received colostrum after birth, were slower and less effective than that of foal 1, which had been deprived of colostrum. In addition, foal 1 had EHV-4 viremia, whereas foals 2 and 3 did not (data not shown). These results indicated that maternal antibodies in equine colostrum suppress the spread of EHV-4 in horses.

Levels of antibodies specific for peptide G1 did not increase either in horses with pyrexia due to EHV-1 infection in the field (Fig. 4) or in racehorses inoculated with inactivated EHV-1 vaccine (Fig. 5). In addition, ELISAs with G1 as the antigen detected 2 horses infected with EHV-4 from among 40 that had been vaccinated with EHV-1 (Fig. 5). These results showed that peptide G1 specifically reacted with only EHV-4-specific antibodies and that ELISAs with this peptide could detect EHV-4 infection as well as ELISAs with egG4P in horses with circulating EHV-1 antibodies.

In conclusion, ELISAs with the EHV-4-specific epitope are useful for the diagnosis and epidemiological study of EHV-4 infections in horses.