Natural and Artificial Selection for Parasitoid Resistance in Drosophila melanogaster Leave Different Genetic Signatures.

Adaptation of complex traits depends on standing genetic variation at multiple loci. The allelic variants that have positive fitness effects, however, can differ depending on the genetic background and the selective pressure. Previously, we interrogated the Drosophila melanogaster genome at the population level for polymorphic positions and identified 215 single nucleotide polymorphisms (SNPs) that had significantly changed in frequency after experimental evolution for increased parasitoid resistance. In the current study, we follow up on 11 of these SNPs as putative targets of the experimental selection process (Jalvingh et al., 2014). We study the patterns of genetic variation for these SNPs in several European field populations. Furthermore, we associate the genetic variation of these SNPs to variation in resistance against the parasitoid Asobara tabida, by determining the individual phenotype and SNP genotype for 144 individuals from four Selection lines and four non-selected Control lines and for 400 individuals from 12 Field lines that differ in parasitoid resistance. For the Selection lines we additionally monitored the changes in allele frequencies throughout the five generations of experimental selection. For three genes, mbl (Zn-finger protein), mthl4 (G-protein coupled receptor) and CG17287 (protein-cysteine S-palmitoyltransferase) individual SNP genotypes were significantly associated with resistance level in the Selection and Control lines. Additionally, the minor allele in mbl and mthl4 were consistently and gradually favored throughout the five generations of experimental evolution. However, none of these alleles did appear to be associated to high resistance in the Field lines. We suggest that, within field populations, selection for parasitoid resistance is a gradual process that involves co-adapted gene complexes. Fast artificial selection, however, enforces the sudden cumulating of particular alleles that confer high resistance (genetic sweep). We discuss our findings in the context of local adaptation.

Bacterial Communities Differ among Drosophila melanogaster Populations and Affect Host Resistance against Parasitoids.

In Drosophila, diet is considered a prominent factor shaping the associated bacterial community. However, the host population background (e.g. genotype, geographical origin and founder effects) is a factor that may also exert a significant influence and is often overlooked. To test for population background effects, we characterized the bacterial communities in larvae of six genetically differentiated and geographically distant D. melanogaster lines collected from natural populations across Europe. The diet for these six lines had been identical for ca. 50 generations, thus any differences in the composition of the microbiome originates from the host populations. We also investigated whether induced shifts in the microbiome-in this case by controlled antibiotic administration-alters the hosts' resistance to parasitism. Our data revealed a clear signature of population background on the diversity and composition of D. melanogaster microbiome that differed across lines, even after hosts had been maintained at the same diet and laboratory conditions for over 4 years. In particular, the number of bacterial OTUs per line ranged from 8 to 39 OTUs. Each line harboured 2 to 28 unique OTUs, and OTUs that were highly abundant in some lines were entirely missing in others. Moreover, we found that the response to antibiotic treatment differed among the lines and significantly altered the host resistance to the parasitoid Asobara tabida in one of the six lines. Wolbachia, a widespread intracellular endosymbiont associated with parasitoid resistance, was lacking in this line, suggesting that other components of the Drosophila microbiome caused a change in host resistance. Collectively, our results revealed that lines that originate from different population backgrounds show significant differences in the established Drosophila microbiome, outpacing the long-term effect of diet. Perturbations on these naturally assembled microbiomes to some degree influenced the hosts' resistance against natural parasites.

Natural variation in differentiated hemocytes is related to parasitoid resistance in Drosophila melanogaster.

As a measure of parasitoid resistance, hemocyte load and encapsulation ability were measured in lines collected from natural populations of Drosophila melanogaster in Europe. Results show large geographic variation in resistance against the parasitoid wasp Asobara tabida among the field lines, but there was no clear correlation between resistance and total hemocyte load, neither before nor after parasitization. This was in contrast to the patterns that had been found in a comparison among species of Drosophila, where total hemocyte counts were positively correlated to encapsulation rates. This suggests that the mechanisms underlying between-species variation in parasitoid resistance do not extend to the natural variation that exists within a species. Although hemocyte counts did not correspond to encapsulation ability within D. melanogaster, the ratios of lamellocytes and crystal cells were very similar in lines with successful encapsulation responses. Apart from variation in the hemocytic response of the different hemocyte types, within-species variation was also observed for accurate targeting of the foreign body by the hemocytes. These results are discussed in the context of possible causes of variation in immune functions among natural populations.

The life of a dead ant: the expression of an adaptive extended phenotype.

Specialized parasites are expected to express complex adaptations to their hosts. Manipulation of host behavior is such an adaptation. We studied the fungus Ophiocordyceps unilateralis, a locally specialized parasite of arboreal Camponotus leonardi ants. Ant-infecting Ophiocordyceps are known to make hosts bite onto vegetation before killing them. We show that this represents a fine-tuned fungal adaptation: an extended phenotype. Dead ants were found under leaves, attached by their mandibles, on the northern side of saplings approximately 25 cm above the soil, where temperature and humidity conditions were optimal for fungal growth. Experimental relocation confirmed that parasite fitness was lower outside this manipulative zone. Host resources were rapidly colonized and further secured by extensive internal structuring. Nutritional composition analysis indicated that such structuring allows the parasite to produce a large fruiting body for spore production. Our findings suggest that the osmotrophic lifestyle of fungi may have facilitated novel exploitation strategies.