Table 1 Host factors that regulate L1 mobilization

Several proteins inhibit L1 transcription. MeCP2 binds methylated cytosines in the CpG island core of the L1 promoter [27, 47, 75, 105]. MeCP2 occupancy prevents cytosine hydroxymethylation and L1 de-repression by the activator TET1, and facilitates the recruitment of methyltransferases affixing the repressive chromatin mark H3K9me3 [104, 285, 286]. Other factors such as KAP1, the HUSH complex and MORC2 bind and silence full-length L1s, including those located in euchromatic genomic regions, again via deposition of repressive marks [96, 103, 106, 285]. Another key repressor, SOX2, is a transcription factor that inhibits neuronal gene and L1 expression during development. Neuronal maturation requires SOX2 down-regulation, which may explain the potential specificity of L1 mobilization in neurons [20, 22]. By contrast, the transcription factors RUNX3 and YY1 assist L1 transcription and retrotransposition [69, 207]. Although the mechanism for L1 activation by RUNX3 is unclear, YY1 appears to direct transcriptional initiation to the correct (+ 1) L1 start site, and may also support loops involving the L1 promoter and enhancer elements [69, 287]. Numerous factors [112] also repress L1 at the post-transcriptional level, and may each do so in multiple ways. For example, the adenosine deaminase ADAR1 inhibits L1 mobilization via editing dependent and independent activities, which could involve binding to the L1 RNP [57, 245, 288, 289]. The exonuclease TREX1 has been shown to inhibit retrotransposition in vitro by depleting L1 ORF1p and altering its subcellular localization [247], whereas SAMHD1 inhibits L1 mobilization by limiting the availability of intracellular nucleotides and other mechanisms [244, 246]. Finally, during L1 integration, TPRT intermediate DNA-RNA hybrids are targeted by host factors, such as APOBEC3A, which deaminates transiently exposed single-stranded DNA [139].