Soluble proteins were purified from bacterial lysates by glutathione-affinity chromatography

as previously described [29], then analysed by sodium-dodecyl-sulphate (SDS) polyacrylamide gel electrophoresis (PAGE). GST-fused proteins from inclusion bodies (insoluble fraction) were dissolved in a CAPS buffer (CAPS 50 mM, DTT 1 mM and Sarkosyl 0.3%), hence denaturing the proteins [30]. The dissolved and denatured protein was dialyzed overnight against 20 mM Tris–HCl pH 8.5. Insoluble proteins dissolved in CAPS buffer/dialysed are referred find more to as ‘CAPS-denatured proteins’ throughout the text. Purified proteins were quantified by two different methods: (i) a Bradford assay at 595 nm and (ii) UV spectrophotometry at 280 nm (extinction coefficient determined from aa sequences of each fusion protein). Concentration measurements were consistent using both methods. Relative amounts of proteins to be injected were based on copy number considerations in a BTV particle, as determined by X-ray crystallography (780 copies for VP7, 360 copies for VP5 and 180 copies for VP2 [1]). Seven

groups of six Balb/c mice were injected subcutaneously at days 0, 14 and 28 with 100 μl of soluble protein/Montanide ISA 50V emulsion (Table 1). Three groups of six Balb/c mice were injected subcutaneously at days 0, 14 and 28 with 100 μl of CAPS-denatured protein/Montanide ISA 50V emulsion (Table 1). A group of six Balb/c mice were injected subcutaneously at days 0, 14 and 28 each with 100 μl of Zulvac-4® Bovis. Sera were used for normalisation of ELISA results. A group of six control Balb/c mice which were not immunised with any of the antigens was PFI-2 also included. Six groups of six IFNAR−/− mice were injected subcutaneously at days 0, 14 and 28 with: a mixture of VP2 Idoxuridine domain 1 (VP2D1) and VP2 domain 2 (VP2D2) in Montanide, then challenged with (i) BTV-4 or (ii)

BTV-8; or a mixture of VP2D1 + VP2D2 + VP5Δ1–100/Montanide, then challenged with (iii) BTV-4 or (iv) BTV-8; or a mixture VP2D1 + VP2D2 + VP5Δ1–100 + VP7/Montanide, then challenged with (v) BTV-4 or (vi) BTV-8 (Table 1). Blood samples were collected at day 0 and day 28. The mice received an intravenous lethal [31] challenge on day 40, with 103 pfu of BTV-4-italy03 (homologous-challenge), or 10 pfu of BTV-8-28 (heterologous-challenge). Blood was collected on the day of challenge (day 40), then at days 2, 3, 4, 5, 7, 10 and 12 p.i. Sera were tested for anti-VP2, anti-VP5 and anti-VP7 antibodies by ELISA and immunofluorescence and for NAbs by PRNT. Two groups of six IFNAR−/− mice were injected subcutaneously with VP5Δ1–100 on days 0, 14 and 28. These groups were not challenged with BTV-4 or BTV-8. Two additional groups of six IFNAR−/− mice were immunised with VP7 on days 0, 14 and 28, then challenged at day 40 with either BTV-4 or BTV-8. Two groups of non-immunised mice were used as positive controls, to confirm lethality of BTV-4 or BTV-8 challenge-strains.