Skip to main content
European Commission logo print header

Genomics of sexual isolation and reinforcement in the house mouse

Final Report Summary - SPECIATIONGENOMICS (Genomics of sexual isolation and reinforcement in the house mouse)

One main topic in evolutionary biology is this unresolved and fascinating riddle: how do new species arise? Understanding the drivers of speciation and the ingredients for the evolution of reproductive isolation is critical to interpreting patterns of biological diversity and to predict how environmental factors will affect biodiversity. When populations evolve for a sufficient amount of time in allopatry, i.e. geographically separated, reproductive barriers can evolve relatively easily as a by-product of divergence through time. However, when populations are still in contact and can exchange genes through hybridisation, gene flow tends to impede divergence and the evolution of reproductive isolation. In the last two decades, theoretical work has shown that selective processes could play a major role in speciation, in particular when the evolution of reproductive isolation is impeded by on-going gene flow. A major current challenge in speciation research is therefore to gain insights into the mechanisms underlying the evolution of reproductive isolation in the face of gene flow, a situation that appears to be widespread in nature. The project SPECIATION GENOMICS took advantage of the dramatic period of development in the field of evolutionary genetics as it absorbs the new possibilities of the post-genomic era to tackle the genomics of speciation and adaptation in the house mouse, a model species for the study of speciation, and therefore dissect the factors favouring speciation. In this species, it was hypothesized that sexual isolation between two European subspecies, Mus musculus musculus and M. m. domesticus, could have evolved specifically in the hybrid zone as a selective response to the production of maladaptive hybrids, a process called ‘reinforcement’ which places selection at the centre of the speciation process. As expected under a reinforcement hypothesis, sexual isolation was shown to be restricted to the hybrid zone. Similarly, one can expect that genes underlying sexual isolation will show strong level of divergence in the hybrid zone as compared to more distant locations (allopatric populations) and, remarkably, will display the signature of an evolution under positive selection. This prediction can be used to test for the role of selection in favouring the evolution of reproductive isolation and to identify the genes involved in sexual isolation. SPECIATION GENOMICS took up this challenge to determine the genetic bases of sexual isolation between the two hybridising mouse subspecies and test for the role of selection in the evolution of sexual barriers by seeking for its signature in the genome. To this end, SPECIATION GENOMICS combined candidate gene approaches, targeting olfactory receptors and protein pheromones known to play a role in mate recognition in mice, and genome-wide approaches to identify the genes showing the strongest sign of divergence under selection, that could consequently be considered as good candidate ‘speciation genes’. Furthermore, SPECIATION GENOMICS exploited the status of model species of the house mouse to dedicate Next-Generation Sequencing technologies and available genomic resources to these key research objectives.

To achieve its objectives, SPECIATION GENOMICS has produced genomic data and research results of high quality that have been and will continue to be published in top international conferences and journals:

* The first study consisted in carrying out a scan for selection based on microsatellites to detect the potential signature of selection and reinforcement in candidate genomic regions (Vomeronasal Receptors and Major Urinary Proteins). The strategy consisted in analysing a large number of markers (more than 1,200 microsatellite loci) that were typed on pooled samples representing hybrid zone populations and allopatric populations either distant or close to the hybrid zone. The study revealed a handful of loci showing the expected pattern of reduced variability specific to hybrid zone populations, which appear in the vicinity of Vomeronasal Receptors genes of family 1 and 2. The first key result of this study was the evidence of the expected signature of reinforcement at the genetic level: this provided additional arguments in favour of a reinforcement process in the house mouse, and validated the approach developed in SPECIATION GENOMICS to tackle reinforcing selection using genetic data. The study also pointed at a few receptors that might evolve under selection in the hybrid zone, and which can therefore be considered as good candidates for sexual isolation.

* The second study tackled the hypothesis that regulatory changes could also underlie, alternatively or in combination with protein-coding changes, sexual isolation and reinforcement in the house mouse. Using RNAseq technology, SPECIATION GENOMICS investigated patterns of differential expression between hybrid and allopatric populations in order to reveal changes in expression at some vomeronasal genes that could underlie the evolution of assortative mate preference in the hybrid zone. To this end, we deeply sequenced the transcriptome of the vomeronasal organ of wild female and male adult mice from allopatric and hybrid zone populations in order to compare levels of expression at vomeronasal genes and potentially detect differentially expressed genes. When comparing populations of the two subspecies from the hybrid zone, known to be quite divergent, we found a huge number of differentially expressed genes. In contrast, we only found a bit more than 200 differentially expressed genes when comparing allopatric and hybrid zone samples within a given subspecies, among which five vomeronasal receptor genes, which therefore appear as strong candidates for sexual isolation. Interestingly, the vomeronasal receptors that were differentially expressed between allopatric and hybrid zone samples only pertain to family 2 of vomeronasal receptors, and even more strikingly four out of these five receptors belong to the same phylogenetic subclade. This result strongly suggests that the evolution of sexual isolation in the house mouse could have involved the specialisation of a group of closely related vomeronasal receptors, a result consistent with other recent findings on the molecular and functional characterisation of vomeronasal receptors in mice.

* The third and last major experiment investigated the genetic basis of sexual isolation in the hybrid zone by means of whole-genome resequencing in allopatric and hybrid zone populations. This experiment complemented the candidate gene approach carried out in the two first experiments by analysing the whole set of protein-coding genes as well as inter-genic and regulatory regions. We adopted a pooling strategy to deeply sequence samples representative of several populations from each of the subspecies and geographical areas (allopatry versus hybrid zone) and obtain high quality population genetics estimates. The analysis under progress should reveal genomic islands of differentiation, potentially involved in reproductive isolation, as well as regions of the genome potentially subject to copy number variation between allopatric and hybrid zone populations, and therefore help to gain insights into the respective role of regulatory, protein and structural changes in the evolution of behaviour barriers and in speciation more generally.

By making use of cutting-edge sequencing and population genomics tools and by combining expression and sequence analyses, SPECIATION GENOMICS sheds light on the genetic bases of speciation and on the molecular footprint of adaptive speciation by reinforcement in the house mouse. It opens new perspectives to understand the origin of biodiversity on Earth.