LINE-1 Retrotransposition in Human Genomic Variation.

Abstract

Long interspersed element-1 (LINE-1 or L1) is a ubiquitous mobile element in mammalian genomes. There are ~500,000 copies of L1 throughout the human genome, comprising ~17% of our DNA. Interestingly, though the majority of these elements are inactive due to 5' truncation or mutations in the two L1-encoded open reading frames (ORF1 and ORF2), a small number of L1s (~80-100 per human genome) are capable of mobilization by the copy and paste mechanism of retrotransposition termed target-site primed reverse transcription (TPRT). As L1s rely upon their encoded proteins (ORF1p and ORF2p) for mobility, only intact, full-length L1s are potentially active. Previous analysis of the human genome reference sequence (HGR) showed that 90 L1s contained intact ORFs, and that 6 of these were responsible for >80% of the retrotransposition activity. These 6 highly active L1s were polymorphic in humans. Therefore, I hypothesized that low allele frequency L1s comprise the majority of activity in human populations. I used fosmid libraries developed from six geographically diverse individuals to identify 68 full-length L1s absent from the HGR. Approximately 55% (37/68) were highly active when tested in a cultured cell assay. To determine the allele frequency of the L1s in the population, 26 of the 68 elements were examined in genotyping panels. Four of the 26 were either private or African specific when typed on the H952 subset of the human genome diversity panel. The sequences of the 68 L1s showed that 53 (37 highly active and 16 low-level or inactive L1s) contained intact ORF1 and ORF2. The sixteen L1s were then examined for changes that led to their inability to retrotranspose with high efficiency in cell culture. Using sequence comparisons and functional assays, I identified a novel amino acid change in the ORF2-encoded protein of one element and a splicing mutation in the 5' untranslated region of another L1. My examination of individual genomes readily identified highly active, rare L1s, and suggests that L1 activity in humans is more prevalent than previously appreciated. The sequences of these elements present future opportunities for the elucidation of L1 biology.