Genotypic and phenotypic diversity of a baculovirus population within an individual insect host

Abstract

It is becoming increasingly apparent that many pathogen populations, including those of insects, show high levels of genotypic variation. Baculoviruses are known to be highly variable, with isolates collected from the same species in different geographical locations frequently showing genetic variation and differences in their biology. More recent studies at smaller scales have also shown that virus DNA profiles from individual larvae can show polymorphisms within and between populations of the same species. Here, we investigate the genotypic and phenotypic variation of an insect baculovirus infection within a single insect host. Twenty four genotypically distinct nucleopolyhedrovirus (NPV) variants were isolated from an individual pine beauty moth, Panolis flammea, caterpillar by in vivo cloning techniques. No variant appeared to be dominant in the population. The PaflNPV variants have been mapped using three restriction endonucleases and shown to contain three hypervariable regions containing insertions of 70–750 bp. Comparison of seven of these variants in an alternative host, Mamestra brassicae, demonstrated that the variants differed significantly in both pathogenicity and speed of kill. The generation and maintenance of pathogen heterogeneity are discussed.

Introduction

Variation in human pathogens and the emergence of novel strains during epidemics are well-documented (e.g., Macleod et al., 1999, Paul et al., 1999, Santos et al., 1998), although the mechanisms behind these changes and their dynamics during disease outbreaks are not well understood. The diversity of disease-causing organisms in other animals, particularly invertebrates, has received far less attention. However, investigating the consequences of infection diversity for disease severity and epidemiology is likely to be crucial if we are to understand the ecology and evolution of host–pathogen interactions. This is not only important for controlling disease outbreaks but is also relevant for the production, application, and long-term impact of both natural and recombinant microbial pest control agents in the field. The first step in this process is to identify the degree of pathogen variation in natural populations and establish how it varies at different temporal and spatial scales.

Insects and baculoviruses represent an excellent system for the study of molecular epidemiology and pathogen variation. There are many examples of disease epizootics in natural insect populations and insect hosts can often be reared in large numbers allowing the execution of manipulative experiments. The widespread use of restriction endonucleases for baculovirus characterization has clearly demonstrated that genotypic variation within baculovirus populations is ubiquitous. Nucleopolyhedroviruses (NPVs) and granuloviruses (GVs) isolated from the same host species in different geographical regions frequently show restriction fragment length polymorphisms (RFLPs) (e.g., Crook et al., 1985, Gettig and McCarthy, 1982, Kislev and Edelman, 1982, Shapiro et al., 1991, Vickers et al., 1991). More recent studies at geographically smaller scales, have shown that restriction endonuclease (REN) profiles of NPVs and GVs isolated from individual caterpillars also vary both within and between populations of the same host species (Cooper et al., 2003, Laitinen et al., 1996, Parnell et al., 2002). Many of the DNA profiles of these field-collected baculovirus isolates show evidence of heterogeneity, indicating that this variation potentially exists within individual hosts. Genotypic variants of baculoviruses have been separated using in vitro plaque assay techniques (e.g., Knell and Summers, 1981, Lee and Miller, 1978, Lynn et al., 1993, Maeda et al., 1990, Smith and Summers, 1978). Restriction mapping of such variants has shown that the differences resulted from alterations in restriction sites and insertions and deletions (e.g., Smith and Crook, 1988), in some cases relating to the presence or absence of units of tandemly repeated sequences (Garcia-Maruniak et al., 1996; Muñoz et al., 1999). Although cloning in vitro has demonstrated that considerable genotypic variation is often present within field-collected baculovirus isolates, it may not necessarily provide an accurate picture of the genotypes originally present in the wild type virus. The cloning process will select strains that will grow in vitro and there is some evidence that passage of baculoviruses in cell culture can result in the generation of variants with sizeable deletions (Pijlman et al., 2001). Thus a better picture of a virus population is likely to be obtained if techniques based on in vivo cloning are used (Smith and Crook, 1988). This technique, and variations on it using dilution and cloning from haemolymph, although more laborious, are likely to produce a more accurate picture of baculovirus population structure.

Pine beauty moth, Panolis flammea (D. & S.) (Lepidoptera: Noctuidae), is a pest on the introduced lodgepole pine (Pinus contorta) in Scotland, although it also occurs, but does not cause outbreaks, on the native Scots pine (Pinus sylvestris) in the UK. An NPV was isolated from an outbreak of P. flammea in the 1970s and this virus has been successfully developed as a control agent (Cory and Entwistle, 1990, Doyle and Entwistle, 1988). It was evident from laboratory amplification studies and the analysis of infected larvae after application of the virus in the field, that PaflNPV contained more than one genotype and that the proportions of genotypes within a mixture could vary under different conditions in the field (Entwistle et al., 1993; J.S. Cory, unpublished data). Despite the frequent demonstration of baculovirus variation in natural populations there has been little systematic study of baculovirus diversity or its role and importance in insect–pathogen dynamics and evolution. This led to the initiation of a project which focussed on the genotypic and phenotypic nature of baculovirus variation and the mechanisms that maintain it in natural populations. As part of this project we wanted to assess whether variation extended to individual larvae. In this paper, we address two issues; the first is whether multiple NPV genotypes are found in a single, infected caterpillar, and the second, if these genotypes exist, whether they differ in key phenotypic features.