Evolutionary Dynamics of the Pericentromeric Heterochromatin in Drosophila virilis and Related Species

Abstract

Pericentromeric heterochromatin in Drosophila generally consists of repetitive DNA, forming the environment associated with gene silencing. Despite the expanding knowledge of the impact of transposable elements (TEs) on the host genome, little is known about the evolution of pericentromeric heterochromatin, its structural composition, and age. During the evolution of the Drosophilidae, hundreds of genes have become embedded within pericentromeric regions yet retained activity. We investigated a pericentromeric heterochromatin fragment found in D. virilis and related species, describing the evolution of genes in this region and the age of TE invasion. Regardless of the heterochromatic environment, the amino acid composition of the genes is under purifying selection. However, the selective pressure affects parts of genes in varying degrees, resulting in expansion of gene introns due to TEs invasion. According to the divergence of TEs, the pericentromeric heterochromatin of the species of virilis group began to form more than 20 million years ago by invasions of retroelements, miniature inverted repeat transposable elements (MITEs), and Helitrons. Importantly, invasions into the heterochromatin continue to occur by TEs that fall under the scope of piRNA silencing. Thus, the pericentromeric heterochromatin, in spite of its ability to induce silencing, has the means for being dynamic, incorporating the regions of active transcription.

Highly contiguous assemblies of 101 drosophilid genomes

“…While a number of studies have already used long reads to assemble the genomes of one or a few drosophilid species (Bracewell et al, 2019;Chakraborty et al, 2021;Comeault et al, 2020;Flynn et al, 2020;Hill et al, 2020;Mai et al, 2020;Miller et al, 2018;Paris et al, 2020;Rezvykh et al, 2021;Solares et al, 2018), a sequencing and genome assembly project at a scale similar to that of the large genome assembly consortia, especially without similar resources and funding, remains challenging even with the benefits of long reads. Yet, there continue to be rapid improvements to long-read sequencing that may alleviate some of these logistical challenges.…”

“…Long reads have proved to be a way to quickly generate affordable yet high-quality genomes, in fact the cost of a highly contiguous and complete Drosophila assembly based on long-read sequencing was recently estimated to be about $1,000 US dollars ( Miller et al, 2018 ; Solares et al, 2018 ), orders of magnitude less than the first D. melanogaster genome. While a number of studies have already used long reads to assemble the genomes of one or a few drosophilid species ( Bracewell et al, 2019 ; Chakraborty et al, 2021 ; Comeault et al, 2020 ; Flynn et al, 2020 ; Hill et al, 2020 ; Mai et al, 2020 ; Miller et al, 2018 ; Paris et al, 2020 ; Rezvykh et al, 2021 ; Solares et al, 2018 ), a sequencing and genome assembly project at a scale similar to that of the large genome assembly consortia, especially without similar resources and funding, remains challenging even with the benefits of long reads. Yet, there continue to be rapid improvements to long-read sequencing that may alleviate some of these logistical challenges.…”

Paramutation-like Epigenetic Conversion by piRNA at the Telomere of Drosophila virilis

“…A new D. virilis TE library was generated using RepeatModeler2 [ 28 ] on a PacBio assembly of cdi A strain 160 [ 29 ] and a nanopore assembly of cdi I strain 9 [ 22 ]. Additional nanopore reads were further obtained for the cdi A strain using the same approach as in [ 30 ].…”

“…piRNA sequences were obtained from a previous study [ 27 ] and mapping was performed with BWA aln, randomly assigning multiple mappers [ 33 ]. Mapping was performed to whole genome assemblies from cdi A strain 160 [ 29 ] and cdi I strain 9 [ 22 ]. In the case of the cdi A strain, the artifictual telomeric TART array (with only a fragment remaining) was removed from the assembly image in Figures 2 and 3 based on being present in only one read among all individual PacBio and nanopore reads and that read most likely being a duplex read.…”

“…Why the cdi gene has adopted piRNA cluster status is not known, though it should be noted that the strain carrying the cdi dual-strand cluster is also the strain that induces dysgenesis. This strain 160 is the only P-like strain in the D. virilis syndrome of dysgenesis and is noted by a large excess of TEs that are seemingly absent or in low abundance in other strains [ 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ]. The polymorphic nature of this piRNA cluster (where the active piRNA cluster allele is designated cdi A and the inactive piRNA cluster allele is designated cdi I ) was used to investigate the manner in which maternal piRNAs could trigger cluster piRNA biogenesis in the next generation [ 16 ].…”