On April 8th, Rohit Kolora led the discussion on the paper by Jay et al. (2018) titled “supergene evolution triggered by introgression of a chromosomal inversion”. Then in the following week (April 15th), Greg Owens led the follow-up study by Jay (BioRxiv). Since the two discussions are highly related, I summarize them in synthesis. These two discussions led to an interesting insight on how inversion supergene variants underlying complex alternative phenotypes were initiated and maintained. In contrast to the notion that frequency-dependent selection being the predominant force maintaining supergene polymorphisms (Schwander et al. 2014), Jay et al. (BioRxiv) claimed that for “many complex polymorphisms, instead of representing adaptations to the existence of alternative ecological optima, could be maintained primarily because chromosomal rearrangements are prone to carrying recessive harmful mutations.”
Jay et al. (2018) demonstrated that the supergene in Heliconius numataI (Hn) related to Heliconius wing patterning gene, cortex, was an inverted region called P1 introgressed from H. pardalinus (Hp)(Figure on the left). Even more interestingly, there were two other inversions: P2 and P3 right next to P1, and the three inversions maintain wing pattern polymorphisms in this Heliconius species complex.
Jay et al. (BioRxiv) further looked into the mechanism that maintains inversion
polymorphisms and found several interesting things: (1) p1 is the oldest inversion, followed by its adjacent P2, and P3 is the youngest (see Figure below); (2) insertions within inversions are mostly transposable elements (TE) and the insertion TEs are younger than the rest of the TEs within the inversions; (3) for all three inversions, the inverted haplotypes demonstrate signatures of negative selection and tend to carry more nonsynonymous polymorphisms than their ancestral haplotypes (suggesting relaxed purifying selection in the inverted haplotypes) (see Figure below);
(4) within P1, most of the non-synonymous mutations arose before introgression got fixed in both Hn and Hp, but new mutation accumulations after introgression remained polymorphic. Jay et al. (BioRxiv) claimed this alternative way in which inversion polymorphisms can be maintained: the formation of an inversion implies a bottleneck, which fixes deleterious mutations that get locked together due to recombination suppression.
If so, the inverted haplotypes carrying the supergene would mutate towards a “dead-end”. This is different than the long-term stable polymorphism situation as we imagined. It’s shocking that such ephemeral gene block heavily loaded with deleterious mutation would persist for more than three million years… Maybe the story doesn’t simply end here, there could still be some evolutionary force 'actually maintaining' inversions? Going back to the complex alternative phenotypes underpined by inversion polymorphisms in various systems, how many of them were ephemeral as described here? At least for some of the inversion supergene polymorphism, there has been evidence of negative frequency dependent selection maintaining the polymorphism.
On the technical side, some concerns were raised about Jay et al. (BioRxiv) genome assembly. The diagonal line in Figure 1A
seems too wabbly in the uninverted regions, casting doubts on the quality of the assembly. Quality assembly is especially important when TE-calling, haplotype dating, and synonymous vs non-synonymous polymorphisms are involved in the inference.
Reference
Jay, P., M. Chouteau, A. Whibley, H. Bastide, V. Llaurens, H. Parrinello, and M. Joron. 2019. Mutation accumulation in chromosomal inversions maintains wing pattern polymorphism in a butterfly. bioRxiv, doi: 10.1101/736504.
Jay, P., A. Whibley, L. Frézal, M. Á. Rodríguez de Cara, R. W. Nowell, J. Mallet, K. K. Dasmahapatra, and M. Joron. 2018. Supergene Evolution Triggered by the Introgression of a Chromosomal Inversion. Curr. Biol. 28:1839–1845.
Schwander, T., R. Libbrecht, and L. Keller. 2014. Supergenes and complex phenotypes. Curr. Biol. 24:R288-294.