Summary PhD thesis Henri van Kruistum: The genomics of placenta evolution in livebearing fish

All life that we know develops according to instructions within its DNA. As organisms evolve, changes in an their body plan should therefore be reflected in associated changes in their genome. However, finding the mutations that cause complex phenotypic change is still a great challenge. In this thesis, I aim to find the genomic changes associated with the evolution of the placenta in the livebearing fish family Poeciliidae. For this, I sequenced, assembled, and then compared the genomes of 26 poeciliid species, aiming to find consistent genomic differences between placental and non-placental species.

In chapter 2, I re-assembled the genome of the placental livebearer Heterandria formosa, using existing sequencing data, after which I compared it to the publicly available genomes of three non-placental relatives. I show that a number of genes in the genome of H. formosa that are related to placenta formation show signs of positive selection, while evolving under regular constraint in the non-placental species. Additionally, I identify a small number of gene duplications that are unique to H. formosa.

Chapter 3 describes the genome assemblies of two poeciliid species for which the
genome sequence was previously unknown: the placental Poeciliopsis retropinna
and the non-placental Poeciliopsis turrubarensis. With the third-generation
sequencing techniques that were used, assemblies of excellent quality could be
generated. I used these assemblies to reliably identify structural variations between the two species, with the tandem duplication of the vtg1 gene in P. retropinna being a particularly interesting result.

In chapter 4 I compare the genomes of 26 poeciliid species, both publicly available as well as assembled in this chapter or in previous chapters. I show that placenta species within this group display accelerated evolution of genes involved with transporter- and vesicle-related functions, providing first evidence for genomic convergence associated with placenta evolution. Additionally, I observed differences in the presence of regulatory elements around developmental genes that were not observed in non-placental species, indicating that regulatory change is also a part of the evolution of the poeciliid placenta.

In Chapter 5 I investigate the occurrence of gene duplications and deletions across the genomes of twelve poeciliid species, three of which are placental. For this, I develop a new pipeline that can simultaneously identify structural variants in the genomes of multiple related species. According to this analysis, placenta evolution in the Poeciliidae is not associated with gene duplications, but instead gene deletions were found in the same molecular pathways for the three placental species: the calcineurin-NFAT signaling cascade and the BMP signaling pathway. In non-placental species, these deletions did not occur.

Finally, Chapter 6 describes a new approach to detect allele-specific methylation
based on Oxford Nanopore sequencing. I apply this approach to four individuals of
the non-placental Poeciliopsis gracilis: two parents, and a male and a female
offspring. I show that allele-specific methylation is widespread in the genome of P.
gracilis. Additionally, the inheritance of methylated alleles is not always random, but instead depends on parent-of-origin. The genes that are in the vicinity of regions that are affected by parent-specific methylation are located predominantly in the brain. These results lead to the hypothesis that genomic imprinting due to intralocus sexual conflict is the cause of the observed allele-specific methylation.