LINE-1 Integration Preferences in Human Somatic Cells

[摘要] Long INterspersed Element-1 (LINE-1 or L1) is the only autonomously active transposable element in the human genome. The vast majority of L1s are inactive, but a small number (~80-100 per human genome) retain the ability to mobilize by a ;;copy and paste’ mechanism called retrotransposition. L1 encodes two proteins (ORF1p and ORF2p) required for retrotransposition. ORF2p is a 150kDa protein that has endonuclease (EN) and reverse transcriptase (RT) activities that are responsible for initiating L1 integration by a mechanism termed target-site primed reverse transcription (TPRT). During canonical TPRT, the L1 EN makes a single-strand endonucleolytic nick at a double-stranded genomic DNA target sequence (typically 5’-TTTT/A-3’ and variants of that sequence), to expose a 3’-hydroxyl group that is used as a primer by the L1 RT to reverse transcribe L1 messenger RNA. Different types of transposable elements (TEs) have evolved convergent strategies to target genomic ;;safe havens,’ where TE insertions are predicted to have relatively minimal effects on host fitness and gene expression. Whether L1 integrates into specific genomic regions requires elucidation. In this thesis, I have examined L1 integration preferences in four human cell lines that are proxies for in vivo cell types known to accommodate endogenous de novo L1 retrotransposition events. By combining cultured cell, molecular biological, the Pacific Bioscience sequencing platform, and computational approaches, I characterized 65,079 de novo engineered human L1 integration sites. I compared our L1 insertion dataset to a weighted random model, which assumes that L1 integration preferences are mediated solely by the presence of a degenerate L1 EN consensus cleavage site in the human genome. The data suggest that gene content, transcriptional activity, strand bias, epigenetic environment, and DNA replication status have minimal effects on L1 integration. Thus, L1 EN is the principal determinant of L1 integration.In contrast to canonical EN-dependent L1 retrotransposition, previous studies indicated that L1s could also integrate at sites of DNA damage, including dysfunctional telomeres, by an endonuclease-independent (ENi) mechanism in certain cultured cell lines that contain mutations in genes that render the non-homologous end-joining (NHEJ) pathway of DNA repair and p53 inactive. Here, we explored whether the disruption of other DNA repair pathways influence ENi L1 integration. We observed ENi retrotransposition in certain tissue culture cell lines containing defects in the Fanconi anemia (FA) DNA repair pathway. Since defects in the FA pathway can lead to the accumulation of inter-strand DNA crosslinks that, if left unrepaired, can interfere with DNA replication, we hypothesized that lesions arising at stalled DNA replication forks may provide substrates for enhanced ENi retrotransposition. Indeed, the examination of L1 EN mutant integration sites in FANCD2-deficient cells, suggests that a 3’-hydroxyl group present at Okazaki fragments and/or double-strand DNA breaks generated at collapsed DNA replication forks might be used as a primer to initiate ENi L1 retrotransposition. In sum, our results suggest that ENi L1 retrotransposition may represent an ancestral mobilization mechanism used by LINE-like retrotransposons prior to the acquisition of an endonuclease domain. Under this scenario, LINE-like elements were reliant upon genomic features (e.g., sites of genomic DNA damage, replication forks, and, less frequently, dysfunctional telomeres) to initiate TPRT in the absence of an endonuclease. Indeed, we posit that the acquisition of an endonuclease domain allowed L1 to autonomously insert throughout the genome and, as originally implied by its name, become an interspersed retrotransposon.

[发布机构] University of Michigan

[效力级别] Genetics

[关键词] Transposable Elements;Genetics;Science;Human Genetics