Skip to main content
Dryad logo

Molecular evolution of the sex peptide network in Drosophila

Citation

Findlay, Geoffrey; McGeary, Meaghan (2020), Molecular evolution of the sex peptide network in Drosophila, Dryad, Dataset, https://doi.org/10.5061/dryad.5x69p8d0j

Abstract

Successful reproduction depends on interactions between numerous proteins beyond those involved directly in gamete fusion.   While such reproductive proteins evolve in response to sexual selection pressures, how networks of interacting proteins arise and evolve as reproductive phenotypes change remains an open question.  Here, we investigated the molecular evolution of the “sex peptide network” of Drosophila melanogaster, a functionally well-characterized reproductive protein network.  In this species, the peptide hormone sex peptide (SP) and its interacting proteins cause major changes in female physiology and behavior after mating.  In contrast, females of more distantly related Drosophila species do not respond to SP.  In spite of these phenotypic differences, we detected orthologs of all network proteins across 22 diverse Drosophila species and found evidence that most orthologs likely function in reproduction throughout the genus.  Within SP-responsive species, we detected the recurrent, adaptive evolution of several network proteins, consistent with sexual selection acting to continually refine network function.  We also found some evidence for adaptive evolution of several proteins along two specific phylogenetic lineages that correspond with increased expression of the SP receptor in female reproductive tracts or increased sperm length, respectively.  Finally, we used gene expression profiling to examine the likely degree of functional conservation of the paralogs of an SP network protein that arose via gene duplication.  Our results suggest a dynamic history for the SP network in which network members arose before the onset of robust SP-mediated responses and then were shaped by both purifying and positive selection.

Methods

We obtained the protein sequence for each SP network protein in D. melanogaster from FlyBase. For species for which protein annotations were available on FlyBase we obtained orthologous protein-coding DNA sequences using the FlyBase Orthologs feature. For species with sequenced genomes that lacked FlyBase protein annotations, we manually searched for gene orthologs using tBLASTn and the D. melanogaster protein sequence as the query. For genes expected to have introns, we looked in the unannotated species for the approximate location of the D. melanogaster intron, and used known intron border consensus sequences and six-frame translation to identify predicted intron borders and remove intronic sequences.  We used MUSCLE as implemented in MEGA 6.06 to align amino acid sequences, then visually checked and edited each alignment for accuracy. Amino acid alignments were then back-translated in MEGA to obtain the cDNA alignment. (branch_alignments.txt)
To detect recombination, we used GARD as implemented in DataMonkey 2.0 Genes were partitioned at breakpoints evaluated as significant by the Kishino-Hasegawa test.  (Partitioned_alignments.txt)
To infer a Drosophila consensus phylogeny based on all SP network proteins, we concatenated the amino acid alignments of all SP network proteins within each of the 22 species (master_alignment.phy). We used PROML in Phylip to infer an unrooted maximum-likelihood phylogeny (with random input order, slow analysis, and all other default parameters).

Usage Notes

Gene alignments are in MEGA sequential format, for use in HyPhy. To use with PAML, simply move the trees to a separate file. 
 
Branch_alignments.txt contains all the alignments for each gene evaluated in this study. Genes are denoted by ##.
 
Partitioned_alignments.txt contains all the alignments for genes that were partitioned based on positive evidence of recombination.
 
Master_alignment.phy contains the concatenated amino acid alignments of all SP network proteins within each of the 22 species, in PHYLIP format. Gaps in the alignment were used in cases in which a protein was not present in a particular species. Two branches were evaluated for lineage-specific adaptive evolution in this study denoted by 'branch1', which leads to D. ananassae and D. bipectinata; and 'branch2', leading to the melanogaster group.

Funding

National Science Foundation, Award: 1652013