Transposable Elements Contribute to the Regulation of Long Noncoding RNAs in Drosophila melanogaster

Abstract

Background: Transposable elements (TEs) and noncoding sequences are major components of the genome, yet their functional contributions to long noncoding RNAs (lncRNAs) are not well understood. Although many lncRNAs originating from TEs (TE-lncRNAs) have been identified across various organisms, their characteristics and regulatory roles, particularly in insects, remain largely unexplored. This study integrated multi-omics data to investigate TE-lncRNAs in D. melanogaster, focusing on the influence of transposons across different omics levels. Results: We identified 16,118 transposons overlapping with lncRNA sequences that constitute 2119 TE-lncRNAs (40.4% of all lncRNAs) using 256 public RNA-seq samples and 15 lncRNA-seq samples of Drosophila S2 cells treated with heavy metals. Of these, 67.2% of TE-lncRNAs contain more than one TE. The LTR/Gypsy family was the most common transposon insertion. Transposons preferred to insert into promoters, transcription starting sites, and intronic regions, especially in chromosome ends. Compared with lncRNAs, TE-lncRNAs showed longer lengths, a lower conservation, and lower levels but a higher specificity of expression. Multi-omics data analysis revealed positive correlations between transposon insertions and chromatin openness at the pre-transcriptional level. Notably, a total of 516 TE-lncRNAs provided transcriptional factor binding sites through transposon insertions. The regulatory network of a key transcription factor was rewired by transposons, potentially recruiting other transcription factors to exert regulatory functions under heavy metal stress. Additionally, 99 TE-lncRNAs were associated with m⁶A methylation modification sites, and 115 TE-lncRNAs potentially provided candidate small open reading frames through transposon insertions. Conclusions: Our data analysis demonstrated that TEs contribute to the regulation of lncRNAs. TEs not only promote the transcriptional regulation of lncRNAs, but also facilitate their post-transcriptional and epigenetic regulation.

Keywords: Drosophila; TE-lncRNA; heavy metal; long noncoding RNA; transposable element.

Figure 1

The pipeline of identifying TE-lncRNAs from RNA-seq data. (A) The pipeline of identifying TE-lncRNAs. (B) A total of 27,642 mRNAs, 5246 lncRNAs, and 39,051 TEs were identified in D. melanogaster. (C) The proportion of TE-lncRNAs in lncRNAs and TE-mRNAs in mRNAs.

Figure 2

The characteristics of the lncRNAs derived from TEs. (A) The length of TE-lncRNAs, Non-TE-lncRNAs, and coding genes. (B) Phylop scores among TE-lncRNA, Non-TE-lncRNA, and coding genes sequence. (C) The expression levels among TE-lncRNAs, Non-TE-lncRNAs, and coding genes. (D) The expression specificity of TE-lncRNAs, Non-TE-lncRNAs, and coding genes. Statistical significance: **: p < 0.01; ***: p < 0.001; NS.: p ≥ 0.05.

Figure 3

The contribution of TEs to lncRNAs. (A) The quantities of various types of TEs that overlap with genome, lncRNA, and CDS sequences. (B) Number of TEs overlapping with each TE-lncRNA. (C) The quantities of various types of TEs that overlap with lncRNAs. (D) Positional preference of TE insertions in lncRNAs. (E) Hotspots of TE insertion on chromosomes.

Figure 4

Impact of TE-lncRNAs on epigenetic levels in Drosophila. (A) ATAC peaks in lncRNAs including TE-lncRNA and Non-TE-lncRNA. (B) Expression of ATAC-TE-lncRNA and Non-ATAC-TE-lncRNA. (C) The correlation between TE coverage and ATAC peaks. (D) The distribution of TE and ATAC peaks in lncRNA. (E) IGV view of the ATAC peaks in TE-lncRNA. Statistical significance: ***: p < 0.001.

Figure 5

Co-expressed regulatory networks of TE-lncRNAs at the transcriptional level. (A) Differentially expressed genes under heavy metal conditions. (B) The quantities of TFBSs between TE-lncRNAs and Non-TE-lncRNAs. (C) The correlation between gene modules and heavy metal samples; each ME color represents a different expression module. (D) The TF regulatory networks mediated by transposons. Statistical significance: ***: p < 0.001.

Figure 6

Validation of genes expressed in response to heavy metal stress in Drosophila S2 cells. (A) GO enrichment analysis of coding genes from those regulatory networks correlated with TE-lncRNAs; grey indicates genes, yellow indicates functional descriptions, and the line between grey circles and yellow circles indicates relations between genes and functions. (B) Expression heatmap of TE-lncRNAs from the btn network. (C) Expression validation of btn using quantitative PCR (qPCR). (D) Expression validation of MSTRG.3783.1 using qPCR. Statistical significance: *: p < 0.05; **: p < 0.01.

Figure 7

Effects of TE-lncRNAs on post-translational regulation in Drosophila. (A) Number of ORFs in TE-lncRNAs and Non-TE-lncRNAs. (B) Expression level of TE-ORF-lncRNAs in different samples. (C) Statistics of translated TE-lncRNAs. (D) Peptides encoded by TE-lncRNAs. Statistical significance: ***: p < 0.001.