Competitive outcome of multiple infections in a behavior-manipulating virus/wasp interaction

Abstract

Infections by multiple parasites are common in nature and may impact the evolution of host-parasite interactions. We investigated the existence of multiple infections involving the DNA virus LbFV and the Drosophila parasitoid Leptopilina boulardi. This vertically transmitted virus forces infected females to lay their eggs in already parasitized Drosophila larvae (a behavior called superparasitism), thus favoring its spread through horizontal transmission. Previous theoretical work indicated that the evolution of the level of the manipulation strongly depends on whether infected parasitoids can be re-infected or not. Here, we describe a strain of LbFV that differs from the reference strain by showing a deletion within the locus used for PCR detection. We used this polymorphism to test for the existence of multiple infections in this system. Viral strains did not differ on their vertical or horizontal transmission rates nor on the way they affect the parasitoid's phenotype, including their ability to manipulate behavior. Although already infected parasitoids were much less susceptible to new infection than uninfected ones, frequent coinfection was detected. However, following coinfection, competition between viral strains led to the rapid elimination of one strain or the other after a few generations of vertical transmission. We discuss the implications of these results for the evolution of the behavioral manipulation.

Keywords: Drosophila parasitoid; Leptopilina boulardi Filamentous Virus; coinfection; within‐host competition.

Figure 1

Example of data‐visualization of multiple infections using the R package Mondrian. Individuals are represented by horizontal lines, and each column represents one viral strain. Horizontal lines (representing individuals) are reordered to optimize the reading. LbFV1 is always represented in black, LbFV2 in gray, and uninfected individuals in white.

Figure 2

(A) Alignment of LbFV1 and LbFV2 molecular markers amplified with primers 500‐R and 102‐F (Accession Numbers: FM876312 and JX455824, respectively). Stars indicate homology. Dark and light gray bars indicate two regions showing high sequence similarity between them, suggesting a putative duplication. (B) Nucleotide (1) and protein (2) alignments of the putative duplications. (C) Agarose gel electrophoresis of PCR products obtained on individual parasitoids.

Figure 3

Frequency distribution of infection statuses among progenies of (A) singly infected females and (B) coinfected females. Black, gray, and white bars stand for LbFV1‐, LbFV2‐infected and uninfected offspring, respectively. The number of offspring tested is shown at the top of each graph. The proportion of a given viral strain is obtained by summing vertically the corresponding color. For coinfected lines, horizontal reading gives the proportion of their offspring sharing a given infection status.

Figure 4

Superparasitism intensity of the different infection statuses. (A) uninfected versus singly infected parasitoid females. (B) coinfected females. Different letters indicate significant differences (pairwise t‐test on log‐transformed data with Bonferroni correction for multiple comparisons, α = 0.008).

Figure 5

Distribution of infection statuses in horizontal transmission experiments. (A) experiment 1; horizontal transmission from singly infected parasitoid females. (B) experiment 2: horizontal transmission from singly infected and coinfected parasitoid females. Different colors stand for different infection statuses in emerged offspring of the recipient line: uninfected (white), infected with LbFV1 (black), or infected with LbFV2 (gray). The number of offspring tested is shown at the top of each graph. The proportion of a given viral strain is obtained by summing vertically the corresponding color. In case of infected recipient line (A) or when the donor line is coinfected (B), horizontal reading gives the proportion of offspring in the recipient line sharing a given infection status.

Figure 6

Expected distribution of coinfection after four generations of vertical transmission under the hypothesis of independent transmission of viral strains. The expected number of coinfected lines was inferred from a Markovian model (see methods). OBS, observed number of coinfected lines.