Epigenetics and island-mainland divergence in an insectivorous small mammal

Abstract

Geographically isolated populations, including island-mainland populations, tend to exhibit phenotypic variation in many species. The so-called island syndrome occurs when different environmental pressures lead to insular divergence from mainland populations. This phenomenon can be seen in an island population of Nova Scotia masked shrews (Sorex cinereus), which have developed a specialized feeding habit and digestive enzyme compared to their mainland counterparts. Epigenetic modifications, such as DNA methylation (DNAm), can impact phenotypes by altering gene expression without changing the DNA sequence. Here, we used a de novo masked shrew genome assembly and epigenome-wide assay to investigate morphological and DNA methylation patterns between island and mainland populations. Island shrews were morphologically and epigenetically different than their mainland counterparts, exhibiting a smaller body size; remarkably, the gene ontology analyses mirrored these phenotypes including genes for the digestive enzyme phenotypes. Moreover, island shrews appeared to age faster than their mainland counterparts. This study provides an example counter to the prevailing island syndrome morphological patterns, along with novel epigenomic insight into the drivers of island-mainland divergence in mammals.

Comparative genomics of the world’s smallest mammals reveals links to echolocation, metabolism, and body size plasticity

“…Nuclear genome completeness was assessed for both assemblies using Benchmarking universal single-copy orthologues (BUSCO) version 3.0.2 (Simão et al, 2015) by comparing the genomes to highly conserved genes in mammals. The smoky shrew mitochondrial genome was assembled using the PacBio subreads and the mitochondrial genome of a previous smoky shrew assembly (NCBI accession: GCA_026122425.1) (Cossette et al, 2023) as a backbone with MitoHifi version 3.2 (Allio et al, 2020;Uliano-Silva et al, 2023). The mitochondrial genome for the maritime shrew was assembled using MitoZ (Meng et al, 2019) with the function "mitoz all" on the fastq reads and parameters --genetic_code 2 --clade Chordata --kmers_megahit 59 63 79 99 119 141 --requiring_taxa Chordata and was manually curated.…”

“…The smoky shrew genome annotation was generated by NCBI using their eukaryotic genome annotation pipeline the integrated RNAseq data (SRA RNA-Seq accession: SRX20204431, SRX20204430) from Cossette et al (2023). We generated the maritime shrew annotation inhouse via GenSAS version 6.0 (Humann et al, 2019) as it was too fragmented for the NCBI pipeline.…”

“…Repeat regions were identified using RepeatModeler version 2.0.1 (Smit & Hubley, 2023b) and RepeatMasker version 4.1.1 (Smit & Hubley, 2023a). Here, we had to use the smoky shrew liver and heart RNA reads from Cossette et al (2023) and mapped them to the genome using HISAT2 version 2.2.1 (Kim et al, 2019). The resulting BAM files were used by Augustus version S3.4.0 (Stanke et al, 2006) for gene prediction.…”

“…For the smoky shrew assembly (Genbank assembly accession: GCA_029834395.2), the five cells produced 115,806,042,303 bp of HiFi data, or ~43X coverage from the estimated 2.7 GB genome (Cossette et al, 2023). The mean length of the HiFi reads was 14,652 bp.…”