6. Virus infectivity assays

Infectivity assays were used before sensitive molecular techniques were developed to detect low levels of virus in surveys (Bailey, 1976; Bailey et al., 1981; 1983b). These assays take advantage of the fact that most bee viruses when injected into adult bees or pupae multiply rapidly to high titres that can subsequently be detected by serology (Dall, 1987). Dilution series of the extracts provide a measure of quantitation. Different viruses develop titre and kill pupae at different rates, which can be detected by the ‘breaking’ of the eye-colour development in white-eyed pupae (Anderson and Gibbs, 1988; 1989). This can provide an early indication of which virus is being multiplied. Although labour intensive, infectivity assays can rival the most sensitive molecular tests available (Denholm, 1999). One serious drawback of honey bee infectivity assays is that often unapparent viruses present at very low levels in the assay pupae can also be amplified, sometimes by the mere injection of buffer (Bailey, 1967; Anderson and Gibbs 1988; 1989). Several important bee viruses (ABPV, KBV and SBPV) were discovered this way, as a by-product of the propagation of CBPV, AIV and BVX respectively (Bailey et al., 1963; Bailey and Milne, 1969; Bailey and Woods, 1974; 1977; Bailey and Ball, 1991), and the technique may yet prove useful for the discovery of other symptomless bee viruses.

In general terms, the procedures for virus infection infectivity assays are the same as for virus propagation. It is especially important is that the larvae and pupae are transferred to the incubation plate as carefully as possible, and that they are checked for vitality and survival before being used for infection experiments. Larvae and pupae should be checked under a stereo microscope for damage and vitality. In both cases it is advisable to incubate them for 12-24 hours prior to conducting the assay, and remove larvae or pupae that show signs of necrosis or low vitality. The infectivity assays should also include a number of additional methodological controls (effects of transfer, incubation, manipulation, feeding etc.), to facilitate the interpretation of the data.