A single amino acid polymorphism in natural Metchnikowin alleles of Drosophila results in systemic immunity and life history tradeoffs

Abstract

Antimicrobial peptides (AMPs) are at the interface of interactions between hosts and microbes and are therefore expected to be rapidly evolving in a coevolutionary arms race with pathogens. In contrast, previous work demonstrated that insect AMPs tend to evolve more slowly than the genome average. Metchikowin (Mtk) is a Drosophila AMP that has a single amino acid residue that segregates as either proline (P) or arginine (R) in populations of four different species, some of which diverged more than 10 million years ago. These results suggest that there is a distinct functional importance to each allele. The most likely hypotheses are driven by two main questions: does each allele have a different efficacy against different specific pathogens (specificity hypothesis)? Or, is one allele a more potent antimicrobial, but with a host fitness cost (autoimmune hypothesis)? To assess their functional differences, we created D. melanogaster lines with the P allele, R allele, or Mtk null mutation using CRISPR/Cas9 genome editing and performed a series of life history and infection assays to assess them. In males, testing of systemic immune responses to a repertoire of bacteria and fungi demonstrated that the R allele performs as well or better than the P and null alleles with most infections. Females show some results that contrast with males, with Mtk alleles either not contributing to survival or with the P allele outperforming the R allele. In addition, measurements of life history traits demonstrate that the R allele is more costly in the absence of infection for both sexes. These results are consistent with both the specificity hypothesis (either allele can perform better against certain pathogens depending on context), and the autoimmune hypothesis (the R allele is generally the more potent antimicrobial in males, and carries a fitness cost). These results provide strong in vivo evidence that differential fitness with or without infection and sex-based functional differences in alleles may be adaptive mechanisms of maintaining immune gene polymorphisms in contrast with expectations of rapid evolution. Therefore, a complex interplay of forces including pathogen species and host sex may lead to balancing selection for immune genotypes. Strikingly, this selection may act on even a single amino acid polymorphism in an AMP.

Fig 1. Mtk^R allele flies have lower egg hatch rates and adult longevity times.

(a) Overview of the experimental design, where individual pairs of males and females of a given genotype were crossed. They were allowed to lay eggs for 24 hours and those embryos were counted. They were then monitored for hatching rates and the larvae were moved into fly food vials. The pupae were counted each day, along with adult male and female emergence, and those adults were tracked over time until death. Figure was created with BioRender.com. (b) The egg-to-larvae hatch rate was counted for each family. Each dot represents the offspring of a single male and female of the indicated genotype. Each dot represents the proportion of all eggs that hatched from one dish (mean eggs per family was 41). The boxes indicate the interquartile range. Outer edges of the box indicate 25^th (lower) and 75^th (upper) percentiles and the middle line indicates 50^th percentile (median). Whiskers represent maximum and minimum ranges of data within 1.5 times the interquartile range of the box. Letters indicate statistical significance groups, based on a logistic regression and Tukey post hoc test. (c) Lines represent adult longevity beginning at emergence from pupae. Each genotype contains an average of 3879 total flies (males plus females). Statistics based on an ANOVA with Tukey post-hoc test (S2 Table). The entire experiment was performed twice, and the graph represents a combination of data from both experiments.

Fig 2. Mtk^R male flies survive as well or better than other genotypes against a variety of filamentous fungal and yeast infections.

Infections were performed with the indicated microbes, using either fungal spores (green underline) or yeast cultures (orange underline). (a) Fusarium oxysporum, (b) Beauveria bassiana, (c) Aspergillus fumigatus, (d) Aspergillus flavus, (e) Candida glabrata, (f) Galactomyces psuedocandidus. Each dot represents the survival 21 days after infection for a vial starting with 10 males. Each set of data represents two independent experiments combined. Corresponding survival curves and controls for this experiment are shown in S4 Fig. The boxes indicate the interquartile range. Outer edges of the box indicate 25^th (lower) and 75^th (upper) percentiles and the middle line indicates 50^th percentile (median). Whiskers represent maximum and minimum ranges of data within 1.5 times the interquartile range of the box. Letters indicate statistical significance groups, based on a logistic regression and Tukey post hoc test (S2 Table).

Fig 3. Mtk alleles are associated with differential survival after infection with some bacterial infections in males.

Infections were performed with the indicated microbes, using bacteria (purple underline). (a) Bacillus thuringiensis, and (c) Enterococcus faecalis, are Gram-positive. (b) Providencia rettgeri, and (d) Serratia marcescens, are Gram-negative. Each dot represents the survival 7 days after infection for a vial starting with 10 males. Each set of data represents two independent experiments combined. Corresponding survival curves and controls for this experiment are shown in S6 Fig. The boxes indicate the interquartile range. Outer edges of the box indicate 25^th (lower) and 75^th (upper) percentiles and the middle line indicates 50^th percentile (median). Whiskers represent maximum and minimum ranges of data within 1.5 times the interquartile range of the box. Letters indicate statistical significance groups, based on a logistic regression and Tukey post hoc test (S2 Table).

Fig 4. Side-by-side male and female fungal spore infections demonstrate key differences in phenotypes.

Infections were performed in females and males with the indicated microbes, using two representative sporulating fungal species (green underline). (a) Beauveria bassiana, (b) Aspergillus fumigatus. Each dot represents survival 21 days after infection for a vial starting with 10 flies. Each set of data represents two independent experiments combined. Females are shown in the lighter colors on the left of each genotype pair, and males are shown in the darker colors on the right of each genotype pair. Corresponding survival curves and controls for this experiment are shown in S9 Fig. The boxes indicate the interquartile range. Outer edges of the box indicate 25^th (lower) and 75^th (upper) percentiles and the middle line indicates 50^th percentile (median). Whiskers represent maximum and minimum ranges of data within 1.5 times the interquartile range of the box. Letters indicate statistical significance groups, based on logistic regression and Tukey post-hoc test (S2 Table). Non-bolded letters indicate females, and bolded letters indicate males.

Fig 5. Pathogen load differences suggest that greater pathogen resistance may underlie higher survival of Mtk^R males with certain infections.

Infections were performed in males with the indicated microbes, using bacteria (purple underline) or yeast (orange underline). (a) Enterococcus faecalis, (b) Candida glabrata, (c) Providencia rettgeri. Each dot represents pathogen load 24 hours after infection for 3 pooled flies. Each set of data represents three independent experiments combined. The boxes indicate the interquartile range. Outer edges of the box indicate 25^th (lower) and 75^th (upper) percentiles and the middle line indicates 50^th percentile (median). Whiskers represent maximum and minimum ranges of data within 1.5 times the interquartile range of the box. Letters indicate statistical significance groups, based on a logistic regression and Tukey post hoc test (S2 Table).