Abstract
We use data from the serial passage of co-occluded recombinant Autographa californica nuclear polyhedrosis virus (AcMNPV) to estimate the viral multiplicity of infection of cells within infected insects. Co-occlusion, the incorporation of wild-type and mutant virus genomes in the same occlusion body, has been proposed as a strategy to deliver genetically modified viruses as insecticides in a way that contains their spread in the environment. It may also serve as a means whereby naturally occurring mutant forms of NPVs can be maintained in a stable polymorphism. Here, a recombinant strain of AcMNPV was constructed with a deletion of its polyhedrin gene, rendering it incapable of producing occlusion bodies (i.e., occlusion negative). This was co-occluded with wild-type AcMNPV and used to infect fifth-instar Trichoplusia ni larvae. The fate of both genotypes was monitored over several rounds of insect infection. Levels of the occlusion-negative virus genome declined slowly over successive rounds of infection. We applied these data to a model of NPV population genetics to derive an estimate of 4.3 +/- 0.3 viral genomes per occlusion body-producing cell.