The impact of transposable elements on mammalian development

Abstract

Despite often being classified as selfish or junk DNA, transposable elements (TEs) are a group of abundant genetic sequences that have a significant impact on mammalian development and genome regulation. In recent years, our understanding of how pre-existing TEs affect genome architecture, gene regulatory networks and protein function during mammalian embryogenesis has dramatically expanded. In addition, the mobilization of active TEs in selected cell types has been shown to generate genetic variation during development and in fully differentiated tissues. Importantly, the ongoing domestication and evolution of TEs appears to provide a rich source of regulatory elements, functional modules and genetic variation that fuels the evolution of mammalian developmental processes. Here, we review the functional impact that TEs exert on mammalian developmental processes and discuss how the somatic activity of TEs can influence gene regulatory networks.

Keywords: Endogenous retrovirus; Genetic variation; Genome regulation; LINE-1; Retrotransposon.

Figure 1. The structures of key classes of TEs found in mammalian genomes.

Different types of TEs (DNA transposons; LINE-1, long interspersed element class 1; SINE, short interspersed element; ERV, endogenous retrovirus) found in mammalian genomes are represented. The percentage of the human and mouse genomes occupied by each TE is indicated in blue. Abbreviations: IR, inverted repeat; UTR, untranslated region; EN, endonuclease; RT, reverse transcriptase; LTR, long terminal repeat; ORF, open reading frame.

Figure 2. Modes of TE mobilization in mammalian genomes

(A) The LTR retrotransposition cycle (adapted from (Mager and Stoye, 2015)). The ERV RNA is transcribed from the 5’ LTR and transported to the cytoplasm, where Gag and Pol are translated and processed into mature proteins including protease, integrase and reverse transcriptase (not shown in the figure but see (Mager and Stoye, 2015) for further details). These ERV proteins and RNAs are then packaged into a virus-like particle (VLP; blue opal surrounded by Gag molecules) and reverse-transcribed (RT) by Pol. The resulting ERV dsDNA is then processed by the integrase activity of Pol to generate a pre-integration complex (PIC), which is imported to the nucleus. Here, the integrase activity of Pol inserts the ERV dsDNA into the genome. New ERV insertions are often flanked by target site duplications (blue or green arrowheads). Host factors (HF; pink circles) also participate in several steps of the retrotransposition process. (B) The non-LTR retrotransposition cycle (adapted from (Macia et al., 2015)). The full-length active LINE-1 RNA is transcribed and transported to the cytoplasm where ORF1p and ORF2p are translated (Alisch et al., 2006; Dmitriev et al., 2007). These proteins preferentially bind to their encoding mRNA, generating a ribonucleoprotein particle (RNP), which is imported into the nucleus. The EN activity of ORF2p generates a single strand (SS) break in genomic DNA that is used by the RT activity of ORF2p to generate the first strand cDNA (red arrow). How second strand synthesis occurs is not well understood. New LINE-1 insertions are often flanked by TSDs (blue or green arrowheads) and are also often 5’-truncated (not shown). As is the case for the LTR retrotransposition cycle, host factors (pink circles) are involved in several steps of the non-LTR retrotransposition process.

Figure 3. The impact of TEs on mammalian genomes.

(A) Deleterious effects of TE insertions on host gene expression. A cartoon of a host gene containing an upstream promoter (black arrow), exons (blue tubes) and introns (grey lines) is shown. TEs are inserted in sense or antisense orientations (blue arrows). Red arrows on TEs denote promoter sequences. Various RNA transcripts (wavy lines) can be produced, based on the location/orientation of the TE and the promoter used. The left side indicate the type of mechanism that is responsible for the generation of each type of transcript. Asterisk indicates a premature termination codon or frameshift. For full details see main text. (B) De novo TE insertions (grey) can impact genes (blue) in various ways. Shown are examples where a new TE insertion can act as: i) an enhancer; ii) a TE insertion within the gene body can lead to exonization of the TE and this can lead to a new protein product with an alternative function; iii) additionally, exonized TE sequences can induce premature termination of translation (PTC, premature termination codon); and iv) a full-length TE insertion within the gene can generate a shorter gene transcript when the TE promoter is used (new transcript) that could have an alternative function. t, denotes evolutionary time. (C) LINE-1 elements can generate processed pseudogenes. Shown are a cellular gene (green) and an active LINE-1 (blue). Upon transcription (mRNAs, green and blue lines) and translation, the LINE-1 RNP can generate a new insertion by conventional cis retrotransposition. However, the cellular RNA could be occasionally transferred to the LINE-1 RNP, generating a gene RNP that can be inserted into the genome by trans retrotransposition. The potential consequences are indicated below: i) using a nearby promoter (red arrow), a new transcript (grey/green wavy line) can be generated and this can have an alternative function or could also interfere with the regulation of the parental gene; ii) the pseudogene can acquire mutations (red lines and dots in the pseudogene) over evolution and thus acquire a new cellular function. t, denotes evolutionary time. (D) TE insertions can impact gene regulation and chromosome architecture. The cartoon shows the possible outcomes of a gene (A, purple) that accumulated two new TE insertions. The upstream TE, if full-length, could generate antisense transcripts (grey wavy line) that act as long non coding RNAs (New lncRNAs) or that could be translated into a protein product as recently described (New protein (LINE-1 ORF0) (Denli et al., 2015)). Note that these antisense RNAs could also alter chromosome architecture and gene regulation by interfering with: gene A (cis interference) or with a gene located elsewhere in the genome (gene B, green, trans interference). Additionally, the upstream TE can generate new chimeric RNAs (grey/purple wavy line) that can induce deregulation of gene A expression (altered expression) or that could induce expression of gene A in a different cell type (altered cell type expression). Similarly, the TE inserted within the gene could also generate antisense RNAs (purple wavy line) that can interfere with gene A regulation (antisense expression and gene regulation). Finally, the TE inserted within the gene can generate a chimeric RNA using the TE promoter (grey/purple wavy line) that could have an alternative function (chimeric RNA and alternative function).

Figure 4. LINE-1 expression and activity in humans: from tissues to cells.

(A) TE and LINE-1 (L1) expression and activity during embryogenesis and in adult humans. Pre-implantation, post-implantation embryos and a scheme of an adult human are shown. (B) LINE-1 (L1) expression and activity in different types of cultured human stem cells are shown. The developmental relationships between these cell types are also shown. In A and B, each blue box indicate the level of TE/L1 expression. On top of each box, it is indicated if new insertions in each tissue/cell type can be transmitted to the next generation (Heritable) or not (Somatic). Data on embryogenesis has been mostly analysed using mouse models, while expression in adult tissues has been analysed using mostly human samples.