Cloning, sequencing and expression of the gene that encodes the major neutralisation-specific antigen of African horsesickness virus serotype 9

https://doi.org/10.1016/S0166-0934(99)00176-7Get rights and content

Abstract

A marked improvement in the efficiency of cloning the large double stranded RNA (dsRNA) genome segments of African horsesickness virus (AHSV) was achieved when the dsRNA polyadenylation step was carried out with undenatured rather than strand-separated dsRNA. It is a prerequisite to use dsRNA of very high purity because in the presence of even trace amounts of single stranded RNA, the dsRNA appears to be poorly polyadenylated as judged by its effectiveness as a template for oligo-dT-primed cDNA synthesis. The full-length VP2 gene of AHSV-9, cloned by this approach, was sequenced and it was found to show the highest percentage identity (60%) to VP2 of AHSV-6, providing an explanation of why these two serotypes show some cross protection. The VP2 protein was also expressed in Spodoptera frugiperda (Sf9) cells by means of a baculovirus recombinant. The yield of the expressed VP2 was high, but the protein was found to be largely insoluble. Nine smaller, truncated VP2 peptides were subsequently expressed in insect cells, but no significant improvement in solubility of the peptides, as compared to that of the full-sized protein, was observed. A western immunoblot analysis of the overlapping peptides indicated the presence of a strong linear epitope located within a large hydrophilic domain between amino acids 369 and 403.

Introduction

African horsesickness virus (AHSV) is the aetiological agent of African horsesickness, a severe and often fatal viral disease of Equidae that is endemic to sub-Saharan Africa. The virus, of which nine different serotypes have been identified, is classified in the Orbivirus genus of the family Reoviridae and transmitted by arthropods of the genus Culicoides (Verwoerd et al., 1979). As in bluetongue virus (BTV), the prototype member of the Orbivirus genus, the AHSV genome is composed of ten double stranded RNA (dsRNA) segments (Bremer, 1976) encoding seven capsid proteins (VP1–7) and three nonstructural proteins (NS1–3). The segments are encapsidated in an icosahedral core particle surrounded by an outer capsid layer composed of the two major proteins VP2 and 5 (Roy et al., 1994). VP2, encoded by the second largest genome segment of the virus, is the most variable of the different capsid proteins (Bremer et al., 1990) and it is the main determinant of a serotype-specific immune response. Antibodies against VP2 are protective in vivo (Burrage et al., 1993), indicating the important role of the protein in any vaccine development strategy.

The vaccination strategy against AHS currently involves the use of polyvalent live attenuated virus vaccines. However, there are a number of concerns about the efficacy and safety of these vaccines, such as the possibility of the reversion of an avirulent virus to a virulent phenotype when subjected to backpassaging in susceptible hosts (House et al., 1992). To counteract these problems, the development of a non-replicating, VP2-based subunit vaccine for orbiviruses has been the focus of a number of investigations. In the case of BTV, VP2 isolated from purified virus (Huismans et al., 1987) or expressed by means of baculovirus recombinants (Roy et al., 1990) has been demonstrated to elicit protection in sheep against virulent viral challenge. In the case of AHSV-4 protection against AHSV was demonstrated by both inoculation with recombinant VP2 (Roy et al., 1996) and by use of a recombinant vaccinia virus expressing AHSV-4 VP2 (Stone-Marschat et al., 1996).

Most of the studies to locate VP2 neutralisation-specific domains have until recently been focused on BTV. Amino acids 328–335 of BTV-1 VP2 were identified as important serotype-specific determinants (Gould et al., 1988) whereas amino acids in the region of 208–402 were associated with serotype specificity of BTV-10 (DeMaula et al., 1993) and BTV-17 (Pierce et al., 1995). With respect to AHSV-4 the analysis of Escherichia coli expressed truncated VP2 peptides identified a linear region within amino acids 253–413 that was able to elicit neutralising antibodies in rabbits and mice (Martinez-Torrecuadrada and Casal, 1995). No similar studies with other AHSV serotypes have as yet been carried out. However, the sequences of the VP2 genes of AHSV-4 (Iwata et al., 1992b, Sakamoto et al., 1994), AHSV-3 (Vreede and Huismans, 1994) and AHSV-6 (Williams et al., 1998) have been determined. VP2 sequence information of other serotypes has been delayed mainly due to the technical difficulty of cloning genes as large as 3000 bp from virus-specified cDNA.

Some modifications are described to our previous dsRNA cloning strategy that significantly improved the efficiency of cloning genome segments of more than 2500 bp. A large number of full-length copies of the VP2 gene of AHSV-9 were cloned by this approach. The gene was then sequenced and expressed in Spodoptera frugiperda (Sf9) cells. Since the full-length VP2 protein was largely insoluble we expressed a number of smaller overlapping VP2 peptides which were analysed for solubility and the presence of linear epitopes by means of immunoblot analysis.

Section snippets

Viruses and cells

A South African isolate of AHSV serotype 9 (AHSV-9) was obtained from Onderstepoort Veterinary Institute (OVI), Pretoria, South Africa. The virus was propagated in chicken embryo reticulocyte (CER) cells. Sf9 cells were grown as monolayers or as suspension cultures in spinner flasks at 27°C in Grace’s insect medium containing 10% fetal calf serum (Highveld Biological).

Isolation and cloning of dsRNA

AHSV dsRNA was isolated from infected cells by SDS-phenol extraction (Huismans and Erasmus, 1981). Further purification was

Cloning of the VP2 gene of AHSV-9

Purified AHSV-9 dsRNA was separated from smaller single stranded (ss)RNA contaminants by two successive sucrose gradient fractionation steps. By selective pooling of the appropriate fractions, pools of the large-sized dsRNA fragments (2500–4000 bp) and medium-sized fragments (1000–2500 bp) were obtained which were free from tRNA and small ssRNA contaminants. When cDNA was prepared from these fractions by the standard oligo dT-primed reverse transcription, using a dsRNA template that had been

Discussion

In the process of cloning the VP2 gene of AHSV-9 we found that the yield of full-sized, large virus-specific cDNA fragments was significantly improved by using dsRNA templates that were polyadenylated without prior denaturation of the dsRNA. Although such an approach has been used previously (Asamizu et al., 1985) we found that we could only obtain consistent results when the dsRNA is highly purified. In the presence of even small amounts of ssRNA, the dsRNA polyadenylation appears to be

Acknowledgements

This work was supported by the Foundation for Research Development and the Agricultural Research Council. We would also like to thank the Onderstepoort Veterinary Institute for the dsRNA and immune sera that were provided and particularly so Dr Albie van Dijk for her support and valuable discussion throughout the investigations.

References (32)

Cited by (7)

  • Synthesis of empty african horse sickness virus particles

    2016, Virus Research
    Citation Excerpt :

    To construct pFBd9.2–9.3, the VP3 coding sequence was recovered from plasmid pBR-9.3 by digestion with BglII, treated with Klenow polymerase and then blunt-end cloned into the SmaI site of pFastBAC-dual to generate pFBd9.3. The VP2 coding sequence was isolated from plasmid pBS-9.2 (Venter et al., 2000) as a SalI/XbaI fragment and cloned into the corresponding sites of pFBd9.3. The resulting plasmid, pFBd9.2–9.3, contains the VP2 and VP3 coding sequences under the transcriptional control of the polyhedrin and p10 promoters, respectively.

  • The use of soluble African horse sickness viral protein 7 as an antigen delivery and presentation system

    2011, Virus Research
    Citation Excerpt :

    However, the insolubility of the VP2 proteins that are expressed in insect cells remains the main stumbling block (Scanlen et al., 2002). The possibility of using smaller epitope domains of AHSV VP2, instead of the full-length VP2, has been explored and a number of neutralization-specific domains were identified between VP2 amino acid residues 253 and 413 (Martinez-Torrecuadrada and Casal, 1995; Martinez-Torrecuadrada et al., 2001; Venter et al., 2000). However, expression of these peptides in bacterial or insect cells renders them largely insoluble and non-immunogenic (Venter et al., 2000).

View all citing articles on Scopus
View full text