key: cord-0332158-scqkv30n authors: David, Jean R.; Ferreira, Erina A.; Jabaud, Laure; Ogereau, David; Bastide, Héloïse; Yassin, Amir title: Evolution of assortative mating following selective introgression of pigmentation genes between two Drosophila species date: 2022-01-16 journal: bioRxiv DOI: 10.1101/2022.01.14.476347 sha: c89c988ae384b0743d65309a13dbf73ed8fdfa36 doc_id: 332158 cord_uid: scqkv30n Adaptive introgression is ubiquitous in animals but experimental support for its role in driving speciation remains scarce. In the absence of conscious selection, admixed laboratory strains of Drosophila asymmetrically and progressively lose alleles from one parental species and reproductive isolation against the predominant parent ceases after 10 generations. Here, we selectively introgressed during one year light pigmentation genes of D. santomea into the genome of its dark sibling D. yakuba, and vice versa. We found that the pace of phenotypic change differed between the species and the sexes, and identified through genome sequencing common as well as distinct introgressed loci in each species. Mating assays showed that assortative mating between introgressed flies and both parental species persisted even after four years (∼ 60 generations) from the end of the selection. Those results indicate that selective introgression of as low as 0.5% of the genome can beget morphologically-distinct and reproductively-isolated strains, two prerequisites for the delimitation of new species. Our findings hence represent a significant step towards understanding the genome-wide dynamics of speciation-through-introgression. In sexually reproducing organisms, speciation begins when extrinsic or intrinsic 33 barriers significantly reduce gene flow between populations and ends with the 34 evolution of pervasive phenotypic differences delimiting the nascent species (Coyne 35 and Orr 2004; The Marie Curie SPECIATION Network 2012; Kulmuni et al. 2020) . 36 The pace of this process can be dramatically accelerated if the diagnostic characters 37 also contribute, either directly or through genetic linkage, to reproductive isolation. 38 The search for such traits, which were dubbed 'magic', has been a 'holy grail' in 39 speciation genetics (Servedio et We also aimed to quantify subtle differences in pigmentation intensity between 192 the two strains that were sequenced in 2020, i.e. BCyak-2 and BCyak-3. For this, 193 flies were killed in 70% ethanol and wings and legs removed using a pair of forceps. 194 Each fly was then individually placed on its left side in 2 mL 70% ethanol solution in 195 an excavated glass block and photographed under a binocular Leica stereoscope 196 provided with a digital camera connected to a computer. Flies were photographed 197 and grey scale intensity was measured using ImageJ (Abramoff et al. 2004 Minimap2-generated SAM files were converted to BAM format using samtools 226 1.9 software (Li et al. 2009 ). The BAM files were then cleaned and sorted using 227 Picard v.2.0.1 (http://broadinstitute.github.io/picard/). We generated synchronized 228 files for the 20 D. y. yakuba lines using Popoolation 2. We then used a customized 229 Perl script to extrapolate allele frequencies to 2 diploid counts for each strain, after 230 excluding sites with less than 10 reads and alleles with frequencies less than 25% for 231 the total counts using a customized Perl script (cf. Ferreira et al. 2021). We also 232 excluded tri-allelic sites for each line. We then parsed the parental strains for 233 divergent sites, i.e. sites with distinct alleles fixed in each strain, and estimated the 234 ancestry proportion at each site in the two introgressed strains in 50 kb-long 235 windows. 236 237 We estimated precopulatory reproductive isolation between the two parental 239 and the two introgressed strains, Bcyak-2 and BCyak-3, using both no choice and Two-choice analyses were conducted for both males and females. For a given 249 sex, a virgin fly was introduced into an individual vial along with two virgin flies from 250 the opposite sex, with one being from the same strain as the tested fly and one from 251 another strain. Copulations were observed also for two hours, and once copulation 252 started flies were anesthetized under slight CO 2 , and the identity of the mating and 253 the un-mating flies identified. In some instances, e.g., those involving a D. santomea 254 male, no marking was needed. For most other cases, flies were individually left to 255 feed in vials with artificial food blue or red colorants (Sainte Lucie co., France) 24 256 hours before the start of the experiment as in Comeault and . A chi-257 squared test was then conducted for each strains combination to test the deviation 258 from parity between homo-and hetero-gamic successful matings. 259 Experimental hybridization led to sexually dimorphic, phenotypically distinct 261 The trajectories of pigmentation evolution during the two 5-year introgression 263 experiments are given in Figure 1E -H in terms of the PCA of pigmentation scores. 264 Interestingly, all the segments were quite similar, and the last one, i.e. the epigynium 275 or tergite 8, which is very dark in D. yakuba was the lightest in the introgressed 276 females (t = 23.24, P < 2.4 x 10 -9 ). The posterior segments of introgressed males 277 ( Figure 1F ) were lighter than D. yakuba (t-test for the sum of segments 5 and 6 = 278 9.25, P < 4.3 x 10 -6 ) but still much darker than D. santomea (t = 10.85, P < 1.8 x 10 -279 6 ). However, the last segment, i.e. the epandrium or tergite 9, became almost 280 completely light (t = 10.16, P < 1.7 x 10 -6 ), as in D. santomea (t = 1.00, P = 0.34). For 281 the "dark D. santomea" experiment, introgressed females ( Figure 1G ) at the end of 282 selection in 2016 were darker than the parental D. santomea (t-test for the sum of 283 segments 6 and 7 = 10.11, P < 3.3 x 10 -6 ), but not as dark as D. yakuba (t = 7.60, P 284 < 1.8 x 10 -5 ). The males ( Figure 1H ), on the other hand, had much darker posterior 285 abdomen (t-test for the sum of segments 5 and 6 = 21.34, P < 5.1 x 10 -9 ), yet still 286 lighter than D. yakuba (t = 10.96, P < 4.1 x 10 -7 ). The last segments in both sexes 287 were completely light as in D. santomea. Remarkably, introgressed females from 288 both experiments significantly differed (t = 9.46, P < 3.5 x 10 -6 ), but not introgressed 289 males (t = 1.99, P = 0.065). 290 After two years from the end of selection in 2016, both experiments tended 291 toward pigmentation values of the ancestral backcross parent, but at a much slower 292 rate. This was most pronounced in females of the "light yakuba" experiment (t = 2.79, 293 P = 0.021), but not in males (t = 1.02, P = 0.321), and in males of the "dark 294 santomea" experiment (t = 3.42, P < 0.004), but not in females (t = 1.63, P = 0.121). 295 For the second round of selection in the "light yakuba" experiment, starting in 2018, 296 the two strains BCyak CC and BCyak selD very slightly differed only for male 297 pigmentation of segments 5 and 6 in 2020 (t = 2.19, P = 0.042). This indicated that 298 selection has attained its limits very rapidly in 2016, but morphological differences 299 between introgressed flies and their parental species persisted for more than 60 300 generations after selection. The two strains were likely derived from the BCyak CC and BCyak selD strains, 331 which corresponded to the second round of selection in the "light yakuba" 332 experiment, and which by 2020 slightly differed in male pigmentation (see above). 333 However, the two sequenced strains, BCyak-2 and BCyak-3, did not show significant 334 difference in pigmentation, even when more numerical analyses were used to 335 quantify melanization (Figure 3) . Nonetheless, both strains showed significant 336 differences with the two parental species for females' segment 7 and males' segment 337 5, and from a single parent for females' segment 6 and males' segment 6, 338 resembling D. santomea for the former and D. yakuba for the later. 339 340 In no-choice experiments, homogamic mating occurred with almost the same 342 frequency between pairs belonging to the same strain/species (70-85%) ( Table 2) . 343 The two introgressed yakuba lines, BCyak-2 and BCyak-3, readily mated with each 344 other. However, a significant low mating success was observed in the cross between homogamic, regardless to the tested sex (Table 3) . However, sex-dependent 353 assortative mating was found for all crosses between D. yakuba and introgressed 354 strains. In all those crosses, females always showed a higher preference for 355 homogamic males, whereas no significant departure from parity was observed for 356 males. We declare no conflicts of interest. Hybridization and speciation Evolution of Darwin's finches and their beaks 600 revealed by genome sequencing Speciation driven by 603 hybridization and chromosomal plasticity in a wild yeast Relationships within the melanogaster 605 species subgroup of the genus Drosophila (Sophophora) -II. Phylogenetic 606 relationships between six species based upon polytene chromosome banding 607 sequences Élimination des formes mutantes dans les 610 populations de drosophile. Cas des drosophiles ebony Minimap2: pairwise alignment for nucleotide sequences The Sequence Alignment/Map format and SAMtools Genetic Network Mask the Contribution of Hox Gene Evolution Sequential adaptive 624 introgression of the mitochondrial genome in Drosophila yakuba and Drosophila 625 santomea An Anomalous Hybrid Zone in Genome-wide tests for 629 introgression between cactophilic Drosophila implicate a role of inversions during 630 speciation Patterns of Genomic Differentiation in 632 the Drosophila nasuta Species Complex What does Drosophila genetics tell us about speciation? Trends in The paradox behind the pattern of rapid 637 adaptive radiation: how can the speciation process sustain itself through an early 638 burst? Annual review of ecology, evolution, and systematics Distinct genetic 640 architectures underlie divergent thorax, leg, and wing pigmentation between Pigmentation Differences Within and Between Drosophila Species Current Topics in Developmental Biology Rapid and Predictable Evolution of Admixed Populations Between Two Drosophila 648 The Influence of Abdominal Pigmentation on 650 Desiccation and Ultraviolet Resistance in Two Species of Drosophila Temperature-Based Extrinsic Reproductive Isolation in Two Species of Drosophila Genetic Basis of Postzygotic Reproductive Isolation Between Drosophila santomea 656 and D. yakuba Due to Hybrid Male Sterility Genetic Basis of Prezygotic Reproductive Isolation Between Drosophila santomea 659 and D. yakuba Due to Mating Preference Correlated Evolution of two Sensory Organs via a Single Cis-Regulatory Nucleotide Change A genomic perspective on hybridization 664 and speciation Reveals Three Sources of Adaptive Variation during a Rapid Radiation Major Locus Controls a Genital Shape Difference Involved in Reproductive Isolation Between Drosophila yakuba and Drosophila santomea R: A language and environment for statistical computing. R 673 Foundation for Statistical Computing Evidence 675 for archaic adaptive introgression in humans Mating of Ebony Vestigial and Wild Type Drosophila 677 melanogaster in Light and Dark Searching for Sympatric 680 Speciation in the Genomic Era A survey of the trans-regulatory landscape for Drosophila melanogaster 683 abdominal pigmentation Supervised machine 685 learning reveals introgressed loci in the genomes of Drosophila simulans and D How Common Is Homoploid 688 Natural 691 selection interacts with recombination to shape the evolution of hybrid genomes Asymmetric introgression reveals the genetic 695 architecture of a plumage trait Magic 697 traits in speciation: 'magic' but not rare? The complex genetic architecture of male 699 mate choice evolution between Drosophila species A framework for comparing processes of 701 speciation in the presence of gene flow Widespread introgression across 705 a phylogeny of 155 Drosophila genomes Genetic Architecture and Functional 708 Characterization of Genes Underlying the Rapid Diversification of Male External 709 Genitalia Between Drosophila simulans and Drosophila mauritiana What do we need to know about 712 speciation? Evolution of mate choice and the so-called 714 magic traits in ecological speciation FlyBase 2.0: the next generation Fine scale mapping of genomic 724 introgressions within the Drosophila yakuba clade Introgression of regulatory alleles and a missense coding mutation drive 728 plumage pattern diversity in the rock pigeon Sexual selection and 730 genetic colour polymorphisms in animals 2020. The Past and Future of Experimental 732 Systematics in the (Post)genomic Era: A Look at the Drosophila in Systematics and the Exploration of Life Ancient balancing selection at tan underlies female colour dimorphism in Drosophila 737 erecta The pdm3 Locus Is a Hotspot for Recurrent Evolution of Female-Limited Color Dimorphism in Long-term evolution of and D. santomea, and two introgressed "light yakuba" strains. 20 copulating pairs 757 were tested per cross. For heterogamic crosses, significant deviation from the 758 homogamic D. yakuba cross cross, i.e. 17 successful crosses out of 20, was 759 estimated using chi-squared test: * < 0.05, ** < 0.01 and *** < 0.001.