Off-target piRNA gene silencing in Drosophila melanogaster rescued by a transposable element insertion

Abstract

Transposable elements (TE) are selfish genetic elements that can cause harmful mutations. In Drosophila, it has been estimated that half of all spontaneous visible marker phenotypes are mutations caused by TE insertions. Several factors likely limit the accumulation of exponentially amplifying TEs within genomes. First, synergistic interactions between TEs that amplify their harm with increasing copy number are proposed to limit TE copy number. However, the nature of this synergy is poorly understood. Second, because of the harm posed by TEs, eukaryotes have evolved systems of small RNA-based genome defense to limit transposition. However, as in all immune systems, there is a cost of autoimmunity and small RNA-based systems that silence TEs can inadvertently silence genes flanking TE insertions. In a screen for essential meiotic genes in Drosophila melanogaster, a truncated Doc retrotransposon within a neighboring gene was found to trigger the germline silencing of ald, the Drosophila Mps1 homolog, a gene essential for proper chromosome segregation in meiosis. A subsequent screen for suppressors of this silencing identified a new insertion of a Hobo DNA transposon in the same neighboring gene. Here we describe how the original Doc insertion triggers flanking piRNA biogenesis and local gene silencing. We show that this local gene silencing occurs in cis and is dependent on deadlock, a component of the Rhino-Deadlock-Cutoff (RDC) complex, to trigger dual-strand piRNA biogenesis at TE insertions. We further show how the additional Hobo insertion leads to de-silencing by reducing flanking piRNA biogenesis triggered by the original Doc insertion. These results support a model of TE-mediated gene silencing by piRNA biogenesis in cis that depends on local determinants of transcription. This may explain complex patterns of off-target gene silencing triggered by TEs within populations and in the laboratory. It also provides a mechanism of sign epistasis among TE insertions, illuminates the complex nature of their interactions and supports a model in which off-target gene silencing shapes the evolution of the RDC complex.

Fig 1. Annotation of TE structure and insertion positions for Mps1^A15 and Mps1^{A15.revertant} alleles within alt.

Transcript annotations and stranded mRNA-seq mappings from (blue[+] and green [–]density plots) from ovaries of mated females are from Flybase (www.flybase.org) and [56]. The Mps1^A15 Doc insertion is 5’ truncated by 827 bp and located within the 4th/5th exon of alt, in the sense direction of the alt transcript. The Mps1^A15.rev Hobo insertion is internally deleted, inserted in the sense direction of alt and located in the first intron, 1188 bp from the first alt TSS.

Fig 2. A Hobo insertion triggers germline de-silencing of two genes silenced by the Mps1^A15 Doc insertion.

Germline polyA mRNA TPM values from 0–2 hour embryos laid by homozygous mothers. yw/yw indicates the wildtype strain used for the original screen. The scheme above each graph describes which TE insertions (yellow arrowheads) were present in mothers of each experiment. Error bars are S.E. In the presence of only the Mps1^A15 Doc insertion, CG7524, Mps1 and alt are silenced in the germline of Mps1^A15 homozygous mothers. In Mps1^A15.rev homozygous mothers, the Mps1^A15.rev Hobo insertion restores germline expression of CG7524 and Mps1 in the presence of th Doc insertion, but expression of alt is not restored. Red indicates the names of the genes affected by the Doc insertion.

Fig 3. A Doc insertion within alt triggers the formation of genic piRNA biogenesis from both strands which is disrupted by a Hobo insertion.

A) Small RNAs derived from convergently transcribed Mps1 and alt are single stranded and in the same direction of the genic transcripts. B) Results across multiple libraries indicate that the Doc insertion of the Mps1^A15 allele triggers dual-strand piRNA biogenesis across multiple genes. C) The Hobo insertion of the Mps1^A15.rev allele disrupts Doc triggered dual-strand genic piRNA biogenesis. D) Quantification of mapped reads from combined libraries indicates piRNAs derived from both strands of Mps1 are reduced by approximately three-fold in Mps1^A15.rev compared to Mps1^A15 but maintained at similar levels in both from alt.

Fig 4. Silencing of Mps1 by the Doc insertion is in cis and dependent on deadlock.

A) TPM of polyA mRNA-seq on whole ovaries. Silencing of Mps1 and alt by the Doc insertion depends on functional deadlock. B) Deadlock dependent silencing of Mps1 by the Doc insertion occurs in cis. In the absence of functional Deadlock, SNP analysis of polyA mRNA-seq reads reveals the de-silencing of the suppressed allele in Mps1^A15 / + heterozygotes.

Fig 5. The insertion of the Hobo element alters local RNA expression of intronic and exonic sequence.

A) qPCR was performed on random hexamer cDNA generated from ovary total RNA. Values were normalized to rp49 and maximum value for each amplicon made equal to 1 to enable analysis of relative expression. Upstream and Downstream 1 PCR amplicons are intronic and immediately flank the Hobo insertion. Downstream 2 amplicon is derived from an exon sequence. The Hobo insertion alters local expression by increasing transcript abundance of intronic sequence immediately upstream of the insertion, but decreases transcript abundance downstream. B) Model for disruption of Doc triggered silencing of Mps1. The presence of the Doc element within the alt transcript triggers the formation of a dual-strand cluster that spreads into Mps1. The Hobo insertion blocks processivity of the alt transcript, presumably through transcriptional termination. In this case, the Doc insertion is no longer a target for cluster formation.

Fig 6. Relative expression (plate normalized) of Mps1 in ovaries is determined strictly by genotype and the dose of A15 alleles, indicating a strict zygotic effect on gene expression.

Reciprocal progeny of all three pairwise crosses show similar expression levels, indicating no maternal effect. Expression levels of Mps1 in Mps1^A15/+ and Mps1^A15/Mps1^A15.rev are intermediate between Mps1^A15/Mps1^A15 homozygotes and +/+ (or Mps1^A15.rev/Mps1^A15.rev) homozygotes, indicating a strict dose effect of Mps1^A15 on Mps1 expression. Relative expression levels are normalized to maximum 1.0.