Loss of cytoplasmic incompatibility and minimal fecundity effects explain relatively low Wolbachia frequencies in Drosophila mauritiana

Abstract

Maternally transmitted Wolbachia bacteria infect about half of all insect species. Many Wolbachia cause cytoplasmic incompatibility (CI) and reduced egg hatch when uninfected females mate with infected males. Although CI produces a frequency-dependent fitness advantage that leads to high equilibrium Wolbachia frequencies, it does not aid Wolbachia spread from low frequencies. Indeed, the fitness advantages that produce initial Wolbachia spread and maintain non-CI Wolbachia remain elusive. wMau Wolbachia infecting Drosophila mauritiana do not cause CI, despite being very similar to CI-causing wNo from Drosophila simulans (0.068% sequence divergence over 682,494 bp), suggesting recent CI loss. Using draft wMau genomes, we identify a deletion in a CI-associated gene, consistent with theory predicting that selection within host lineages does not act to increase or maintain CI. In the laboratory, wMau shows near-perfect maternal transmission; but we find no significant effect on host fecundity, in contrast to published data. Intermediate wMau frequencies on the island of Mauritius are consistent with a balance between unidentified small, positive fitness effects and imperfect maternal transmission. Our phylogenomic analyses suggest that group-B Wolbachia, including wMau and wPip, diverged from group-A Wolbachia, such as wMel and wRi, 6-46 million years ago, more recently than previously estimated.

Keywords: Host-microbe interactions; WO phage; introgression; maternal transmission; mitochondria; spatial spread.

Figure 1

A) An estimated phylogram for various group-A (red) and group-B (blue) Wolbachia strains. All nodes have Bayesian posterior probabilities of 1. The phylogram shows significant variation in the substitution rates across branches, with long branches separating the A and B clades. B) An estimated chronogram for the same strains, with estimated divergence times and their confidence intervals at each node. To obtain these estimates, we generated a relative relaxed-clock chronogram with the GTR + Γ model with the root age fixed to 1, the data partitioned by codon position, and with a Γ(2,2) branch rate prior. We used substitution-rate estimates of Γ(7,7) × 6.87×10⁻⁹ substitutions/3^rd position site/year to transform the relative chronogram into an absolute chronogram.

Figure 2

All wMau variants share four large deletions, relative to sister wNo. Top panel) The normalized read depth for wMau R60 plotted across the four focal regions of the wNo reference genome; 10 kb of sequence surrounding regions are plotted on either side of each region. Bottom panel) The normalized read depth of wMau R60 plotted across the whole wNo reference genome. Regions that do not contain statistically significant CNVs are plotted in dark blue, and regions with significant CNVs are plotted in red. All wMau variants share the same CNVs, relative to wNo.

Figure 3

wMau infections do not influence D. mauritiana fecundity, regardless of host genomic background. A) Violin plots of the number of eggs laid by D. mauritiana females over five days when infected with their natural wMau variant (R31I and R41I), when infected with a novel wMau variant (R31^R41I and R41^R31I), and when uninfected (R31U, R41U, R31^R41U, and R41^R31U). Large black dots are medians, and small black dots are the eggs laid by each replicate over five days. B) The daily egg lay of these same infected (solid circles) and uninfected (open circles) R31 (aqua), R41 (red), R31^R41 (gray), and R41^R31 (dark blue) genotypes is reported. Large circles are means of all replicates, and small circles are the raw data. Only days where females laid at least one egg are plotted. Cytoplasm sources are denoted by superscripts for the reciprocally introgressed strains.

Figure 4

The mean number of eggs laid by infected R31 (R31I, large aqua dots) and uninfected R31-tet (large tan dots) genotypes are similar. Egg counts for each replicate are also plotted (small dots). Violin plots show egg lay across all ages for each genotype; large black circles are medians, and small black circles are the number of eggs laid by each replicate.

Figure 5

A) A nuclear relative chronogram. B) A mitochondrial relative chronogram with co-occurring Wolbachia strains listed in parentheses. See the text for an interpretation of the results, including the artifactual resolution of the phylogeny of the D. simulans clade.