Cryptic genetic variation enhances primate L1 retrotransposon survival by enlarging the functional coiled coil sequence space of ORF1p

L1-ORF1p nucleoprotein can rapidly assume distinct conformations and simultaneously bind more than one nucleic acid

LINE-1 (L1) is a parasitic retrotransposable DNA element, active in primates for the last 80-120 Myr. L1 has generated nearly one-third of the human genome by copying its transcripts, and those of other genetic elements (e.g. Alu and SVA), into genomic DNA by target site-primed reverse transcription (TPRT) and remains active in modern humans. L1 encodes two proteins that bind their encoding transcript (cis preference) to form an L1 ribonucleoprotein (RNP) that mediates retrotransposition. ORF2p provides reverse transcriptase and endonuclease activity. ORF1p, its major component, is a homo-trimeric phospho-protein that binds single-stranded nucleic acid (ssNA) with high affinity and exhibits nucleic acid (NA) chaperone activity. We used optical tweezers to examine ORF1p binding to individual single-stranded DNA (ssDNA) molecules and found that the arrangement of ORF1p on the ssDNA depends on their molar ratio. When the concentration of ORF1p is just sufficient to saturate the entire NA molecule, the nucleoprotein (NP) is compact and stable. However, additional ORF1p binds and destabilizes the compacted NP, allowing it to engage a second ssDNA. Our results suggest that ORF1p displaced from its RNA template during TPRT could bind and destabilize remaining downstream L1 RNP, making them susceptible to hijacking by non-L1 templates, and thereby enable retrotransposition of non-L1 transcripts.

Investigating the phylogenetic history of toxin tolerance in mushroom-feeding Drosophila

Understanding how and when key novel adaptations evolved is a central goal of evolutionary biology. Within the immigrans-tripunctata radiation of Drosophila, many mushroom-feeding species are tolerant of host toxins, such as cyclopeptides, that are lethal to nearly all other eukaryotes. In this study, we used phylogenetic and functional approaches to investigate the evolution of cyclopeptide tolerance in the immigrans-tripunctata radiation of Drosophila. First, we inferred the evolutionary relationships among 48 species in this radiation using 978 single copy orthologs. Our results resolved previous incongruities within species groups across the phylogeny. Second, we expanded on previous studies of toxin tolerance by assaying 16 of these species for tolerance to α-amanitin and found that six of them could develop on diet with toxin. Finally, we asked how α-amanitin tolerance might have evolved across the immigrans-tripunctata radiation, and inferred that toxin tolerance was ancestral in mushroom-feeding Drosophila and subsequently lost multiple times. Our findings expand our understanding of toxin tolerance across the immigrans-tripunctata radiation and emphasize the uniqueness of toxin tolerance in this adaptive radiation and the complexity of biochemical adaptations.

Co-expression of distinct L1 retrotransposon coiled coils can lead to their entanglement

L1 (LINE1) non-LTR retrotransposons are ubiquitous genomic parasites and the dominant transposable element in humans having generated about 40% of their genomic DNA during their ~ 100 million years (Myr) of activity in primates. L1 replicates in germ line cells and early embryos, causing genetic diversity and defects, but can be active in some somatic stem cells, tumors and during aging. L1 encodes two proteins essential for retrotransposition: ORF2p, a reverse transcriptase that contains an endonuclease domain, and ORF1p, a coiled coil mediated homo trimer, which functions as a nucleic acid chaperone. Both proteins contain highly conserved domains and preferentially bind their encoding transcript to form an L1 ribonucleoprotein (RNP), which mediates retrotransposition. However, the coiled coil has periodically undergone episodes of substantial amino acid replacement to the extent that a given L1 family can concurrently express multiple ORF1s that differ in the sequence of their coiled coils. Here we show that such distinct ORF1p sequences can become entangled forming heterotrimers when co-expressed from separate vectors and speculate on how coiled coil entanglement could affect coiled coil evolution.

The L1-ORF1p coiled coil enables formation of a tightly compacted nucleic acid-bound complex that is associated with retrotransposition

Long interspersed nuclear element 1 (L1) parasitized most vertebrates and constitutes ∼20% of the human genome. It encodes ORF1p and ORF2p which form an L1-ribonucleoprotein (RNP) with their encoding transcript that is copied into genomic DNA (retrotransposition). ORF1p binds single-stranded nucleic acid (ssNA) and exhibits NA chaperone activity. All vertebrate ORF1ps contain a coiled coil (CC) domain and we previously showed that a CC-retrotransposition null mutant prevented formation of stably bound ORF1p complexes on ssNA. Here, we compared CC variants using our recently improved method that measures ORF1p binding to ssDNA at different forces. Bound proteins decrease ssDNA contour length and at low force, retrotransposition-competent ORF1ps (111p and m14p) exhibit two shortening phases: the first is rapid, coincident with ORF1p binding; the second is slower, consistent with formation of tightly compacted complexes by NA-bound ORF1p. In contrast, two retrotransposition-null CC variants (151p and m15p) did not attain the second tightly compacted state. The C-terminal half of the ORF1p trimer (not the CC) contains the residues that mediate NA-binding. Our demonstrating that the CC governs the ability of NA-bound retrotransposition-competent trimers to form tightly compacted complexes reveals the biochemical phenotype of these coiled coil mutants.

Somatic retrotransposition in the developing rhesus macaque brain

The retrotransposon LINE-1 (L1) is central to the recent evolutionary history of the human genome and continues to drive genetic diversity and germline pathogenesis. However, the spatiotemporal extent and biological significance of somatic L1 activity are poorly defined and are virtually unexplored in other primates. From a single L1 lineage active at the divergence of apes and Old World monkeys, successive L1 subfamilies have emerged in each descendant primate germline. As revealed by case studies, the presently active human L1 subfamily can also mobilize during embryonic and brain development in vivo. It is unknown whether nonhuman primate L1s can similarly generate somatic insertions in the brain. Here we applied approximately 40× single-cell whole-genome sequencing (scWGS), as well as retrotransposon capture sequencing (RC-seq), to 20 hippocampal neurons from two rhesus macaques (Macaca mulatta). In one animal, we detected and PCR-validated a somatic L1 insertion that generated target site duplications, carried a short 5' transduction, and was present in ∼7% of hippocampal neurons but absent from cerebellum and nonbrain tissues. The corresponding donor L1 allele was exceptionally mobile in vitro and was embedded in PRDM4, a gene expressed throughout development and in neural stem cells. Nanopore long-read methylome and RNA-seq transcriptome analyses indicated young retrotransposon subfamily activation in the early embryo, followed by repression in adult tissues. These data highlight endogenous macaque L1 retrotransposition potential, provide prototypical evidence of L1-mediated somatic mosaicism in a nonhuman primate, and allude to L1 mobility in the brain over the past 30 million years of human evolution.

Reconstruction of full-length LINE-1 progenitors from ancestral genomes

Sequences derived from the Long INterspersed Element-1 (L1) family of retrotransposons occupy at least 17% of the human genome, with 67 distinct subfamilies representing successive waves of expansion and extinction in mammalian lineages. L1s contribute extensively to gene regulation, but their molecular history is difficult to trace, because most are present only as truncated and highly mutated fossils. Consequently, L1 entries in current databases of repeat sequences are composed mainly of short diagnostic subsequences, rather than full functional progenitor sequences for each subfamily. Here, we have coupled 2 levels of sequence reconstruction (at the level of whole genomes and L1 subfamilies) to reconstruct progenitor sequences for all human L1 subfamilies that are more functionally and phylogenetically plausible than existing models. Most of the reconstructed sequences are at or near the canonical length of L1s and encode uninterrupted ORFs with expected protein domains. We also show that the presence or absence of binding sites for KRAB-C2H2 Zinc Finger Proteins, even in ancient-reconstructed progenitor L1s, mirrors binding observed in human ChIP-exo experiments, thus extending the arms race and domestication model. RepeatMasker searches of the modern human genome suggest that the new models may be able to assign subfamily resolution identities to previously ambiguous L1 instances. The reconstructed L1 sequences will be useful for genome annotation and functional study of both L1 evolution and L1 contributions to host regulatory networks.