Isolation of a capsid protein of bluetongue virus that induces a protective immune response in sheep

VIROLOGY

157, 172-179

(1987)

Isolation of a Capsid Protein of Bluetongue Virus That Induces a Protective Immune Response in Sheep H. HUISMANS,’ Department

N. T. VAN DER WALT, M. CLOETE, AND B. J. ERASMUS

of Biochemistry,

Veterinary

Research institute, Onderstepoort

Received May 12. 1986; accepted

October

0 1 10, Republic of South Africa

15, 1986

A method to purify the neutralization specific antigen of bluetongue virus P2 in large amounts has been developed. The purified protein is free from virus-specified or cellular contaminants and its immunological specificity has been preserved. The purification is based on the observation that protein P2 can be dissociated from the virion by treatment with monovalent or divalent salts. The salt concentration required to solubilize the outer capsid proteins is pH dependent and in general decreases with a decrease in pH. P2 purified by extraction from polyacrylamide gels does not induce immune-precipitating or neutralizing antibodies. The response against P5, on the other hand, is much less conformational dependent and P5 purified from gels readily induces PS-precipitating antibodies in rabbits. These antibodies do not neutralize the virus. Purified P2, immunoabsorbed with anticore serum to remove trace amounts of P7, was injected into sheep. An initial dose of 50 ag of P2 was sufficient to induce P2-precipitating antibodies as well as neutralizing and hemagglutination-inhibiting antibodies. These sheep were fully protected against challenge with a virulent strain of the same BTV serotype. Lower doses of P2 still provided a significant level of protection even though no neutralizing antibodies could be detected. o 1987 Academic PWSS. I~C.

INTRODUCTION

In view of the potentially important role of single protein subunit vaccines (Brown, 1984; Hilleman, 1985) we have investigated the possibility of purifying the two outer capsid proteins of BTV in large amounts. As a basis for this purification we have used the observation by Verwoerd et al. (1972) that the BTV outer capsid is unstable on CsCl gradients at a low pH. In preliminary experiments reported by Huismans et al. (1983) it was also found that P2 can be dissociated from the virion by treatment with MgC& at a low pH. The solubilized P2 is able to induce neutralizing antibodies in sheep. However, the P2 was still contaminated by P7 and lowmolecular-weight cellular contaminants. This paper describes the purification of the two outer capsid proteins of BTV and compares their ability to induce neutralizing antibodies. The potential of P2 as a single protein subunit vaccine for sheep was also investigated.

The double-layered protein capsid of bluetongue virus (BlV) is composed of seven structural proteins (Verwoerd et a/., 1972). Treatment of the virus with chymotrypsin and MgCi2 removes the outer capsid layer (van Dijk and Huismans, 1980) which consists of the two major proteins P2 and P5. These two proteins account for about 38%1 of the protein content of the virus (Verwoerd et a/., 1972) and their removal converts the virion into a core particle which contains two major proteins, P3 and P7, and three minor proteins, Pl, P4, and P6. Protein P7 is the major group-specific antigen (Huismans and Erasmus, 1981; Gumm and Newman, 1982; Hiibschle and Yang, 1983). This protein is present in the core in about twice the amount of the other major core protein, P3, and it is the major protein constituent of the capsomeres on the surface of BTV core particles (Huismans et a/., 1987). Protein P2 in the outer capsid layer of BTV is the main determinant of serotype specificity and the neutralization-specific immune response (Huismans and Erasmus, 1981). Monoclonal antibodies against P2 are able to neutralize the virus (Appleton and Letchworth, 1983; N. T. van der Walt, unpublished results) and can also provide passive immunity and protection against challenge with a pathogenic strain of the same serotype (Letchworth and Appleton, 1983).

MATERIALS

0042.6822187

$3.00

0 1987 by Academic Press. Inc. of reproduction in any form resewed.

METHODS

Virus and cells Most of the experiments were carried out with BTVlOA, an egg-adapted vaccine strain. The virus was propagated in BHK cells in roller bottles as described (Huismans, 1979) and purified by either one of two methods: the modified Freon extraction method described by Huismans (1979) or a much shorter TX-l 00 purification method. In this method the virus-infected cells were resuspended in cold TX-100 buffer (0.5% TX-l 00 in 4 mM Tris, pH 8.8) at a concentration of 5

’ To whom requests for reprints should be addressed.

Copyright All rights

AND

172

BLUETONGUE

VIRUS CAPSID

10’ cells/ml. After 10 min on ice the cells were disrupted with 10 strokes of a tight-fitting Dounce homogenizer and nuclei were removed by centrifugation at 5000 rpm for 10 min. The nuclei were washed twice with half the original volume ofTX-100 buffer. The virus particles in the combined supernatants were collected by centrifugation through a 5-ml cushion of 40% sucrose in low Tris buffer (LTB) (2 mlVl Tris, pH 8.8) for 90 min at 27,000 rpm in a Beckman SW27 rotor. The pellets were resuspended in LTB and centrifuged on 1O-300/0 sucrose gradients in LTB for 50 min at 27,000 rpm in a SW27 rotor. The visible virus bands were pooled and the pellets were resuspended in half the original volume of LTB and centrifuged on sucrose gradients as before. The second sucrose gradient step yielded at least 20% of the total virus yield. The virus was concentrated by centrifugation at 27,000 rpm for 90 min in a SW27 rotor. The pellet was resuspended in 2 ml\/l Tris, pH 8.8, at a concentration of about 2.0 mg/ml and stored at 4”. The total virus yield was in the order of 1 .O mg virus/5 X 1O8 cells. The purity of the virus was routinely verified by gel electrophoresis. x

Preparation of labeled soluble protein fractions from BTV-infected cells BHK cells, grown as monolayer cultures in Roux flasks, were infected with BTV at a multiplicity of 40 plaque-forming units (PFU)/cell. After incubation for 10 hr at 37” the cells were rinsed with methioninefree Eagle’s medium and then incubated for 150 min with 5 ml of this medium containing 100 ~Cilml [35S]methionine (Amersham). An excess of complete Eagle’s medium was added to the cells and after 30 min incubation at 37” the cells were harvested and divided into 10 equal portions. The cells were collected by low-speed centrifugation and the cell pellets were stored at -70”. Cell lysates were prepared by addition of 5 ml of a buffer containing 0.15 M KCI, 10 mMTris-HCI, pH 7.8, and 1% TX-l 00. The cell lysates were centrifuged for 90 min at 45,000 rpm in a Beckman SW50.1 rotor. The supernatants (Sl 00 fraction) were used for the immune precipitation tests. Immune precipitation

test

To 200 ~1 of an Sl 00 fraction, 100 ~1 serum was added and the solution was incubated overnight at 4”. After addition of 1 ml 0.01 STE buffer (0.01 l\/ITris, pH 7.4, 1 mNI EDTA, 0.01 M NaCI) the immune complexes were collected by centrifugation at 3000 rpm for 15 min. The precipitates were washed twice with 0.01 STE and then resuspended in the loading buffer for SDS-polyacrylamide gel electrophoresis (PAGE).

PROTEIN

173

ISOLATION

Plaque reduction

neutralization

test

Plaque reduction tests were carried out as described (Huismans and Erasmus, 1981). Antibody titers are expressed as the reciprocal of the serum dilution causing a 50% plaque reduction. Hemagglutination-inhibition

test

The tests were carried out as described by van der Walt (1980) using purified BTV as antigen and sheep red blood cells. The tests were read after 3 hr at room temperature. The hemagglutination-inhibition (HI) titer was defined as the highest dilution that completely inhibited 4 units of hemagglutination. The clinical

reaction

index

The severity of clinical bluetongue after challenge with virulent virus was numerically expressed as a clinical reaction index (CRI) which was obtained by adding the following 3 scores (a + b + c): (a) the fever score-the cumulative total of fever readings above 40” on Days 3 to 14 after challenge (maximum score about 12); (b) the lesion score-lesions of the mouth, nose, and feet were each scored on a scale of O-4 and added together (maximum score 12); (c) the death score-an additional 4 points were added to the sum of a and b if death occurred within 14 days after challenge. The relative reaction (RR) = CRI of test sheep expressed as a percentage of that of the control. The percentage protection = 100 - RR. Analysis

of solubilized

proteins

Buffer solutions were prepared as follows: Solutions atpH 5.0,5.5,6.0,6.5,and8.0werebufferedwithO.l M 4-morpholineethanesulfonic acid (Mes). Since Mes is a weak buffer at 8.0, buffers with 0.1 M Hepes and 0.1 NI Tris, pH 8.0, were also investigated. No difference between these and the Mes buffer, however, were observed. The MgCI, and CsCl concentrations in the buffers were adjusted so that after the addition of the virus suspension the final concentrations were as indicated in the text. The effect of the buffers on BTV was investigated by adding 50 ~1 of purified BTV (2.3 mg/mI) to 350 ~1 of each buffer in 0.4-ml Eppendorf tubes. After 90 min at 4” the 4.5-cm-long tubes were placed in a bucket of a Beckman SW27 rotor that had been filled with water (maximum four tubes per bucket) and centrifuged at 27,000 rpm for 60 min at 4”. The supernatants were collected and stored at 4’. Protein analysis was carried

174

HUISMANS

out as follows: To a 100-~1 sample of the supernatant, SDS and KCI were added to give final concentrations of 0.5% and 0.3 M, respectively. The samples were kept on ice for 15-30 min and the SDS precipitate together with the solubilized proteins were collected by centrifugation at 11,000 rpm in a microfuge. The precipitates were resuspended in electrophoresis loading buffer and analyzed by PAGE.

ET AL.

The protein bands were visualized for excision by KCI precipitation of the SDS-protein complexes, and the protein was extracted after homogenization of the gel pieces in a tight-fitting glass Dounce in an elution buffer containing 0.01 M STE and 0.1% SDS. Proteins were concentrated by reducing the volume of the elution buffer in a dialysis bag by polyethylene glycol extraction. RESULTS

Isolation

of P2

A cold suspension of 2 mg/ml of purified BlV in 2 mR/ITris-HCI, pH 8.8, was mixed with an equal volume solubilization buffer to give a final concentration of 0.5 M MgCI, in 0.1 M Mes buffer, pH 5.0. After 15-30 min on ice the solubilized proteins were isolated from the particulate virus material by centrifugation at 50,000 rpm for 1 hr at 4’ in a Beckman SW50.1 rotor. The supernatant was collected and centrifuged a second time through a l-ml cushion of 40% sucrose in 0.1 M Mes buffer, pH 5.0, containing 0.2 M MgC12. The P2 concentration in the supernatant was determined by measuring the absorbance at 280 nm and comparing it to an albumin standard. These concentrations were usually confirmed by a modification of the spectrophotometric protein determination method of Ehresmann eta/. (1973) as described by Oellermann (1974). The purity of the P2 preparations was determined by SDS-PAGE. lmmunoadsorption of P2 extracts to remove trace amounts of P7 was carried out by mixing equal volumes of P2 and a hyperimmune anticore serum. After 2 hr at 4’ immune complexes were removed by centrifugation for 30 min at 7000 rpm in a Beckman JS-21 centrifuge. The supernatant was divided in aliquots containing 5 to 100 pg protein and stored at -70”. Polyactylamide

gel electrophoresis

Protein electrophoresis was carried out in SDSPAGE gels using a discontinuous gel system (Laemmli, 1970). A 5% stacking gel and a 15% separating gel were used. Before electrophoresis samples were heated for 5 min at 94” in a loading buffer that contained 0.1 M Tris-HCI, pH 6.8, 0.3 M 2-mercaptoethanol, 1o/oSDS, 10% glycerol, and bromphenol blue as a dye marker. Electrophoresis was carried out at 200 V for 16 hr and gels were stained in 0.2% Servablue in 50% methanol and 10% acetic acid. They were destained at 50” in 4% acetic acid. Isolation

of BTV proteins

from PAGE gels

The procedure used was the same as that described by Hager and Burgess (1980). As much as 2 mg of purified BTV was loaded onto preparative PAGE gels.

Purification

of P2

Isolation of the outer capsid polypeptides of BTV was based on the observation that the outer capsid layer is unstable at a low pH in the presence of high concentrations of salt. In order to optimize conditions for purification a comparison was made of the stability of the outer protein layer at different pH values in the presence of different concentrations of either CsCl or MgC12. The results in Fig. 1 indicate that the best condition for the selective isolation of P2 without detectable solubilization of P5 is treatment of purified virus at pH 5.0 with MgCI, concentrations of between 0.2 and 0.5 M. The same result can be achieved with CsCl at pH 5.5 but as much as 1.25 M CsCl or higher is necessary to solubilize P2 quantitatively. To solubilize both outer capsid proteins a higher pH and a higher salt concentration are necessary. At pH 6.5 a MgC12 concentration of 0.5 M will solubilize P2 and P5 equally well. In the CsCl experiment the only detectable solubilization of P5 was observed at a salt concentration of 2.5 M and a pH of 6.5. At pH 8.0 no solubilization of either of the two outer capsids is observed even at a CsCl concentration of up to 3.75 M. The same results were obtained with the other BTV serotypes investigated, namely, BTV-3, BTV-4, and BTV-13 (results not shown). For the large-scale purification of P2 the purified virus was treated at pH 5.0 with 0.5 M MgC12. The purified protein induced a type-specific immune response in rabbits in contrast to P2 purified by excision from SDSPAGE gels. It was, however, observed that virus purified in the standard Freon extraction method of Huismans (1979) was almost always contaminated with a number of small cellular proteins which ranged in size from 17,000 to 14,000 Da. These proteins are also found in the nuclear fraction of uninfected cells and do not appear to be virus specified, even though they remain virus associated after several cycles of purification on sucrose gradients. These proteins are solubilized under the same conditions as protein P2. To eliminate this contamination an alternative method of virus purification that did not disrupt the cell nuclei was investigated. The TX-100 method described under Materials and

BLUETONGUE

pH 8.0 0IvABCDE

VIRUS CAPSID

175

ISOLATION

pH 6.5 BlVA

pH 5.0 BNABCDE

BCDE

pH 8.0 BTVABCOE

PROTEIN

pH 5.5

pH 6.5 BTVABCDE

BWABCDE

FIG. 1. Comparison of the stability of the outer protein layer of EITV at different pH values in the presence of different concentrations of MgCI, and CsCI. The effect of MgClz is shown in the top row, and the effect of CsCl is shown in the bottom row. Buffers were prepared as described under Materials and Methods and 350 ~1 of each buffer added to 100 pg purified EITV in 50 ~1 2 mM Tris, pH 8.0. After 90 min at 4” the suspensions were centrifuged at 27,000 rpm for 60 min. A 1 00-~1 of each supernatant was analyzed by SDS-PAGE as indicated under Materials and Methods: In the top row lanes A, 6, C, D. and E indicate the proteins solubilized at MgClp concentrations of 0, 0.2, 0.5, 1 .O, and 1.5 M, respectively, under each of the three different pH conditions investigated. In the bottom row lanes A, B, C, D, and E indicate the proteins solubilized at CsCl concentrations of 0, 0.5, 1.25, 2.5, and 3.75 M under each of the three pH conditions investigated. The BTV controls represent an equivalent 30.pg sample of untreated virus.

Methods was found to be very satisfactory. Results in Fig. 2 show that the low-molecular-weight contaminants were eliminated. When P2 was injected into rabbits and sheep it was found that almost all P2 preparations also induced antibodies against P7. This result suggested contamination of P2 with trace amounts of highly immunogenic P7 that could not be detected by SDS-PAGE. To eliminate this contamination P2 preparations were treated with hyperimmune anticore serum as described under Materials and Methods. This method effectively removed the P7 contamination.

Immune response against the major capsid proteins The aim of the experiment was to compare the immune response against purified P2 and P5. P2 was purified by the MgCI, treatment as described above as well as by extraction from SDS-PAGE gels. Attempts to purify P5 from mixtures of solubilized P2 and P5 without the use of detergents such as SDS were unsuccessful and the protein could only be purified as a single protein by the gel extraction method. Preliminary experiments, however, indicated that P5 differed from the other major capsid polypeptides in that its immu-

176

HUISMANS

ET AL.

conditions and their BTV susceptible status was verified by testing their sera with an immunodiffusion test. Two weeks before the commencement of the experiment the sheep were transferred to an insect-proof isolation stable where they were kept for the duration of the experiment. Two sheep were each injected subcutaneously with 2.5, 10, 50, and 100 pg, respectively, of immunoadsorbed purified P2 emulsified in incomplete Freund’s adjuvant. Two sheep were also injected with 100 pg of P2 that was not immunoadsorbed with core serum, and a further two sheep received phosphate buffer plus incomplete Freund’s adjuvant. Two booster inoculations containing half the amount of P2 were given at 1 and 4 weeks after the first injection. The last booster was administered without adjuvant. Serum samples were collected periodically throughout the course of the experiment. Thirty-four days after the last booster injection all the immunized sheep as well as the two nonvaccinated FIG. 2. A comparison of P2 isolated from BTV purified by the Freon extraction method and the shorter TX-1 00 purification procedure. Purification methods were carried out as indicated under Materials and Methods: Lanes A and C represent SDS-PAGE electrophoresis of a 30.pg sample of BTV purified by the Freon and TX-l 00 methods, respectively. Lanes B and D show proteins solubilized by treatment of Freon and TX-l 00 purified BTV, respectively. Conditions for treatment were 0.5 ILI MgC& at pH 5.5 as indicated in the legend to Fig. 1.

nological specificity appeared to be very well preserved after gel electrophoresis. This was demonstrated directly by the use of PAGE-purified P5 as antigen but it was also evident from immune blots of BTV proteins separated by SDS-PAGE. In these experiments P5 was the only protein that reacted strongly with antiserum against BTV (results not shown). P2, P5, and purified P7 were injected into rabbits as indicated in Fig. 3. The sera were analyzed by immune precipitation and a plaque reduction test. The results in Fig. 3 show that both P2 and P5 induced in rabbits the synthesis of antibodies that will recognize and immune precipitate the corresponding soluble proteins. P2 purified by the gel extraction method did not induce antibodies that could be detected by immune precipitation. A similar result was obtained with P7. The result suggests that P5 is less conformation dependent than the other capsid proteins. Of the various capsid proteins only P2 purified by the MgCI, treatment induced neutralizing antibodies. Response against purified

P2 in sheep

Twelve l-year-old Merino wethers were used. The sheep were bred and raised under relatively insect-free

A

B

1024

C

<4

D

E

F

<4

C4

1024

FIG. 3. Immune precipitation (IP) of ?Yabeled BTV proteins with rabbit sera after injection with purified P2, P5, and P7. Lane A, BTV control. Lane B, IP with serum obtarned from rabbits injected with Pi! purified by treatment of purified BTV with 0.5 A/ MgCI, at pH 5.0. The solubilized P2 was treated with anticore serum as described under Materials and Methods. The other lanes are immune precipitations obtained with rabbit sera after injection of proteins purified by extraction from SDS-PAGE gels. Lane C, protein P2; lane D. protein P5; lane E, protein P7. Lane F, serum from a rabbit injected with purified BTV. The rabbits were injected with 100 rg purified protein in the presence of complete Freund’s adjuvant. Booster injections containing the same amount were given 1, 4, and 7 weeks after the first injection. The sera used for immune precipitations were collected 1 week after the last injection. The immune precipitates were analyzed by electrophoresis and autoradrography. The plaque neutralization titers of the sera used for IP were determined and are shown at the bottom of each lane. Titers are the reciprocal of the dilution that caused a 50% plaque reduction.

BLUETONGUE

VIRUS CAPSID

control sheep were challenged by subcutaneous inoculation of 1.0 ml infective sheep blood containing 1O5 PFU of virulent BW-10. Rectal temperatures were recorded twice daily and the sheep were carefully examined for clinical manifestation of bluetongue disease which was expressed numerically by the CRI. This was calculated as indicated under Materials and Methods. Sera were analyzed by immune precipitation and plaque reduction tests. The immune precipitation results are shown in Fig. 4. Sheep injected with an initial dose of as little as 10 pg of purified P2 developed P2-precipitating antibodies without any detectable response against P7 or any of the other capsid polypeptides except nonstructural protein NS2, whose precipitation was considered to be nonspecific. The nonspecific precipitation of NS2 was particularly evident with sheep sera. It was found in subsequent experiments (results not shown) that nonspecific precipitation of NS2 could be avoided if immune precipitation is carried out in the presence of 0.8 M KCI. The P2 sample that was not treated with anticore serum precipitated P7 very strongly. The protection results are summarized in Table 1. Also shown are the plaque reduction titers of the sera collected 1 week after the last booster injection. The

A

BCDE

F

G

H

-NS2 -6 -7

FIG. 4. Immune precipitation of 35S-labeled BTV proteins with sera from sheep injected with various amounts of purified protein P2. P2 was purified by treatment of BTV with 0.5 M MgCI, at pH 5.0. The protein was immunoabsorbed with antrcore serum and injected into sheep. Lane A, sheep sera from uninfected control; lane B, sheep sera from sheep injected with 2.5 pg P2; lane C, 10 pg; lane D, 50 pg; and lane E. 100 pg. The serum in F is from a sheep injected with 100 pg P2 that was not treated with anticore serum. Booster injections containing half amounts were given 1 and 4 weeks after the first injection. Sera collected 1 week after the last injection were used for precipitating soluble ‘?S-labeled proteins. Immune precipitates were analyzed by electrophoresis and autoradrographyas described under Materials and Methods: Lane G is a BTV control and lane H is immune precipitation with BTV antiserum.

PROTEIN

177

ISOLATION TABLE 1

RESPONSEOF SHEEPTO INJECTIONWITH PURIFIEDBTV CAPSID PROTEINP2

Sheep No.

Initial dose of P2 injected (rgY

Immune precipitation of P2

1 2 3 4 5 6 7 8 9 10 11 12

0 0 2.5 2.5 10.0 10.0 50.0 50.0 100.0 100.0 100.0 100.0

+ + + + + + .+ +

Plaque reduction titerb

CRI”

Protection (%)”

<4 <4 <4 14 <4 <4 16 14 128 128 128 128

8.1 7.0 3.3 7.0 1.9 1.3 0 5.7 0 0 0 0

60 14 76 84 100 30 100 100 100 100

a The initial dose was followed by two booster injections containing half the indicated amount at 1 and 4 weeks after the first injection. Sheep Nos. 1 1 and 12 were injected with P2 which was not treated with anticore serum. ’ Reciprocal of the dilution that caused a 50% plaque reduction. c Calculated as indicated under Materials and Methods.

plaque reduction titers indicated that the minimum initial dose required for the induction of detectable neutralizing antibodies was 50 pg. The plaque reduction titer after inoculation with a first dose of 100 pg was 128. One of the sheep inoculated with 50 pg gave only a very weak response against P2. The protection studies indicated that in one of the sheep injected with an initial dose of 2.5 pg P2, a degree of protection evaluated at about 60% was observed. Injection with 10 pg P2 gave 80% protection. The only symptom observed under these conditions was a weak febrile reaction 6-7 days after challenge. The reaction lasted for not more than 24 hr. This level of protection is normally considered sufficient in terms of the level achieved by a particular vaccine strain. No neutralizing antibodies could be detected after inoculation with this dose, but P2-precipitating antibodies were observed (Fig. 4). At higher doses of P2, neutralizing antibodies were induced and the sheep were fully protected. No viremia occurred in the protected sheep after challenge. The antibodies induced by P2 only neutralized virus of the homologous serotype. The kinetics of the immune response in sheep inoculated with 100 pg P2 is shown in Table 2. The first P2-precipitating antibodies were found 14 days after the first inoculation. No neutralizing antibodies were detected before the third inoculation. This also applied to the hemagglutination-inhibiting antibodies. The highest plaque reduction titer and HI titer responses

178

HUISMANS TABLE 2 KINETICSOF THE ANTIBODY RESPONSETO A SHEEP INJECTEDWITH PURIFIEDPROTEINP2

Days postinjections 0 7 14 21 28 31 34 37 40 42

Immune precipitation of P2 + + + + + + + +

Plaque reduction titerb

HI titer=

<4 <4 <4 <4 <4 <4 32 64 128 128

<4 <4 14 <4 <4 <4 32 64 64 64

BAt Day 0 the sheep was injected with 100 pg purified P2 followed by booster injections with half this amount at 14 and 28 days postinjection. b Reciprocal of the dilution that caused a 50% plaque reduction. c Reciprocal of the highest dilution of serum which gives complete inhibition of 4 hemaglutination units of BTV.

against P2 were obtained about 40 days after the first inoculation.

DISCUSSION Purified Bn/ outer capsid polypeptide P2 was used to induce neutralizing antibodies in sheep that fully protected the animals against challenge with virulent virus. The results confirm the previous observations that P2 is the main determinant of the neutralizationspecific immune response (Huismans and Erasmus, 1981; Appleton and Letchworth, 1983). The minimum dose required to produce at least 80% protection was about 10 pg of protein followed by two inoculations of 5 pg each. No detectable neutralizing antibodies were induced after inoculation with this dose, suggesting the possibility that the animals were immunologically primed to respond very quickly after challenge. The only symptoms shown by these sheep were a very weak febrile reaction lasting for not more than 24 hr. When the P2 inoculation dose in the case of BTV is increased to 50 pg or more, neutralizing antibodies can be detected in the sheep sera and the animals are fully protected. This was seen by the absence of any febrile reaction or viremia after challenge. Even though the dose required for protection appears to be somewhat lower than has been reported for foot-and-mouth disease virus (Brown, 1984) the minimum dose is still at least a 1OO-fold more than the dose required with an inoculum of inactivated virus. Campbell et a/. (1985)

ET AL.

have shown that a dose of 1O7 PFU of inactivated BfV can induce a protective immune response in sheep. This corresponds to about 0.5 pg of purified virus or 0.1 pg of protein P2. We were able to confirm the complete protection of sheep after inoculation with inactivated BTV in a dose range of between 0.5 and 1 .O pg virus (B. J. Erasmus, unpublished results). Therefore, despite the relatively mild conditions under which P2 was isolated, it is clear that a very drastic loss in immunogenicity occurs when the protein is removed from the virion. This loss does not seem to apply to the major protein component of BTV cores, protein P7. This protein was found to induce a remarkably strong immune response in sheep after inoculation with quantities of P7 that were not detectable by SDS-PAGE. P7 is a group-specific antigen of BTV (Huismans and Erasmus, 1981) and it is also referred to as the so-called soluble antigen (Gumm and Newman, 1982). However, when P7 was extracted from SDS-PAGE gels it lost its immunological specificity and failed to induce P7-precipitating antibodies. It has never been previously investigated if the other major outer capsid protein BTV, P5, can also induce neutralizing antibodies. This protein was purified by extraction from SDS-PAGE gels by a method that did not appear to affect the ability of the protein to induce P5precipitating antibodies in rabbits. This result was in strong contrast to that obtained with other capsid proteins such as P2 and P7, which after denaturation on SDS-PAGE gels appear to loose the immunological specificity for inducing an antibody response that still recognizes the native soluble protein or virus particle. The P5-specific antibodies did not, however, neutralize the virus. The P2 purified by the nondenaturing method, on the other hand, induced neutralizing and HI antibodies on rabbits and sheep. The kinetics of the appearance of HI antibodies corresponded to those of the neutralizing antibodies. The method developed to purity P2 revealed some very interesting characteristics about the stability of the outer capsid layer of Bn/. Two factors, pH and salt concentration, were found to have a pronounced effect on this stability. We found that the virus was more stable in the presence of CsCl than MgC& even under conditions of equal ionic strength. At pH 8.0 even the highest concentration of CsCl investigated (3.75 n/l) did not dissociate the outer capsid layer, whereas at the same pH both P2 and P5 are solubilized at a MgC& concentration of 1 .O M. This method can be used to isolate core particles and also to activate the core-associated BTV transcriptase. Just as with CsCl the minimum concentration of MgC12 required to solubilize outer capsid polypeptide P2 decreases with a decrease in pH.

BLUETONGUE

VIRUS CAPSID

In the case of both MgCI, and CsCI, P2 is preferentially solubilized in the lower pH range, providing the basis for the selective purification of P2. Rotavirus is another member of the Reoviridae family with capsid proteins that can be solubilized by a high concentration of divalent cations. Bican et al. (1982) were able to show that at pH 8.0 treatment with 1 or 1.5 M CaCI, converted dense rotavirus particles into 40-nm core particles with the loss of VP39, the major external protein of dense particles. Despite the ease by which large amounts of purified P2 can be isolated from purified virus, it is very unlikely that a P2 subunit vaccine that is prepared in this way will ever replace the attenuated live virus vaccine currently in use, even though in specialized cases a complementary role could be envisaged. The reason for the limited usefulness of such a vaccine is that a relatively large quantity of purified virus is required for even one single protective dose. Furthermore, since the neutralizing antibodies that are induced by P2 are also serotype specific it seems likely that a subunit vaccine may have to include P2 from all or at least a number of the different serotypes against which protection is required. This would complicate the preparation and the usefulness of such a subunit vaccine even more. The most obvious future development of a subunit vaccine therefore lies in the direction of the production of P2 by molecular cloning in an alternate host (Brown, 1984). The approach of cloning such genes in virus vectors such as vaccinia virus seems particularly promising (Macket et a/., 1982; Panicali and Paoletti, 1982). These possibilities are currently being investigated. ACKNOWLEDGMENTS The technical assistance of P. Carter and L. M. Pieterse is gratefully acknowledged by the authors.

REFERENCES APPLETON.J. A., and LETCHWORTH,G. J. (1983). Monoclonal antibody analysis of serotype-restricted and unrestricted bluetongue viral antigenic determinants. Virology 124, 286-299. BICAN, P., COHEN, J., CHARPILIENNE,A., and SCHERRER,R. (1982). Purification and characterization of rotavirus cores. /. viral. 43, 1 1 131117. BROWN, F. (1984). Synthetic viral vaccines. Annu. Rev. Microbial. 3, 221-235. CAMPBELL, C. H., BARBER,T. L., KNUDSEN, R. C.. and SWANEY, L. M.

PROTEIN

ISOLATION

179

(1985). Immune response of mice and sheep to bluetongue virus inactivated by gamma irradiation. ln “Bluetongue and Related Orbiviruses,” pp. 639-647. Alan R. Liss, New York. EHRESMANN,B., IMBAULT,P., and WEIL, J. H. (1973). Spectrophotometric determination of protein concentration in cell extracts containing tRNA’s and rRNA’s. Anal. Biochem. 54, 454-463. GUMM. I. D., and NEWMAN, J. F. E. (1982). The preparation of purified bluetongue virus group specific antigen for use as a diagnostic reagent. Arch. Viol. 72, 83-93. HAGER, D. A., and BURGESS,R. R. (1980). Elution of proteins from sodium dodecyl sulfate-polyacrylamide gels. Removal of sodium dodecyl sulfate, and renaturation of enzymatic activity: Results with Sigma subunlt of Escherichia co/i RNA polymerase, wheat germ DNA topoisomerase, and other enzymes. Anal. Biochem. 109, 7686. HILLEMAN, M. R. (1985). Newer directions In vaccine development and utilization. /. Infect. Dis. 151, 407-419. HOBSCHLE,0. J. B., and YANG, C. (1983). Purification of the group specific antigen of bluetongue virus by chromatofocusing. /. l&o/. Methods6, 171-178. HUISMANS. H. (1979). Protein synthesis in bluetongue virus-infected cells. Virology 92, 385-396. HUISMANS, H., and ERASMUS, B. J. (1981). Identification of the serotypespecific antigens of bluetongue virus. Ondersfepoort J. Vet. Res. 48, 51-58. HUISMANS, H.. VAN DER WALT, N. T., CLOETE, M., and ERASMUS,B. J. (1983). The biochemical and immunological characterization of bluetongue virus outer capsid polypeptides in “Double-stranded RNA virus.” (R. W. Compans and D. H. L. Bishop, Eds.), pp. 165172. Elsevier. New York. HUISMANS, H., VAN DIJK. A. A.. and ELS, H. J. (1987). Uncoating of parental bluetongue virus to core and subcore particles in infected L cells. Virology 157, 180-l 88. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227, 680-685. LETCHWORTH,G. J., and APPLETON,J. A. (1983). Passive protection of mice and sheep against bluetongue virus by a neutralizing monoclonal antibody. Infect. lmmun. 39, 208-2 12. MACKE~, M., SMITH, G. L.. and Moss, B. (1984). A general method for the production and selection of vaccinia virus recominants expressing foreign genes. /. Viral. 49, 857-864. OELLERMANN,R. A. (1974). The elimination of ribonucleic acid interference in the spectrophotometric determination of protein concentration. Onderstepoort J. Vet. Res. 41, 22 l-224. PANICALI, D., and PAOLE~I, E. (1982). Construction of pox viruses as cloning vectors: Insertion of the thymidine kinase gene of herpes simplex virus into the DNA of infectious vaccinia virus. Proc. Nat/. Acad. Sci. USA 79,4927-4931. VAN DERWALT, N. T. (1980). A haemagglutination and haemagglutination inhibition test for bluetongue virus. Onderstepoortl. Vet. Res. 47,113-l 17. VAN DIJK, A. A., and HUISMANS, H. (1980). The in virro activation and further characterization of the bluetongue virus-associated transcnptase. Vjrology 104, 347-356. VERWOERD,D. W.. ELS, H. J., DE VILLIERS, E.-M. and HUISMANS, H. (1972). Structure of the bluetongue virus capsid. 1. Wol. 10, 783794.