Development and application of transposable element technology

Eunice Kennedy Shriver National Institute Of Child Health & Human Development

Investigators

Linked publications & trials

Abstract

The human genome contains over 500,000 copies of L1, a non-LTR retrotransposon that accounts for 17% of the genome. A small number of these L1s are transposition competent and are responsible for the majority of de novo insertions including those causing over 124 cases of human disease. The majority of genomic L1s (>95 %) are grossly truncated at their 5' end. Full-length copies of L1 are 6 kb and encode its RNA Pol II promoter, an RNA binding protein called ORF1, and a combination endonuclease and reverse transcriptase designated ORF2. DNA vectors that express engineered L1s in cultured cells reveal many features of L1 transposition including the roles of ORF2 in making a nick at insertion sites and in reverse transcription of the 1st strand using the nick as primer. Painstaking analysis of individual de novo insertions confirmed the high frequency of 5' truncation, which renders the elements nonfunctional for transposition; however, the mechanism of the 5' truncations is unknown. The hypothesis that the RT activity of ORF2 has poor processivity is unlikely as in vitro assays demonstrate high processivity. What seems more likely is that 5' truncation is the outcome of genetic conflict with host factors, that serve as defenders of the genome. We aim to identify the molecular mechanism of the frequent 5' truncation of newly transposed L1 elements. Engineered vectors expressing L1 containing antisense reporter genes such as GFP allow de novo integration in cultured cells (HeLa or HEK293T) to be measured. The reporter genes are expressed only after transposition because in the vector they are disrupted by a sense strand intron (antisense for the reporter mRNA) that is spliced before reverse transcription. Unfortunately, these reporter genes cannot be used to measure 5' truncation because they are placed in the 3' UTR to preserve L1 coding function. To detect 5' truncations we designed antisense reporter genes inserted into the 5' UTR of L1 engineered vectors. To prevent placing ATGs upstream of the L1 ORF1, all AUGs in the L1 coding strand of the reporter were removed by making synonymous codons encoding the reporter protein. With this recoding, antisense reporters with a sense intron still encode a functional product following reverse transcription but do not introduce start codons that would block ORF1 and ORF2 translation. There are many variables that could prevent these new L1 vectors from measuring transposition. A reporter gene in the 5' UTR could disrupt critical functions of L1 such as promoter activity, ORF1 translation, or reverse transcription. To identify active L1 vectors we chose a total of 12 designs with various positions in the 5' UTR and four versions of RFP and mCherry reporter genes. We have now identified a functional 5' UTR reporter that can determine overall levels of 5' truncation in cell lines by determining what fraction of cells express GFP and the red fluorescent protein. These are the cells with full length L1 integration. We have used this dual reporter system in HEK293T cells and have visualized the presence of both red and green fluorescence with microscopy. Dual reporter L1 with mutations in ORF1 and in ORF2 demonstrate that the florescence is due to de novo integration. Importantly, integration frequencies of dual reporter L1 as determined by GFP expression are equivalent to L1 lacking the 5' mCherry. This demonstrates that the reporter in the 5' UTR does not reduce integration activity. Comparing the ratios of red+green to green reveals full-length insertions represent approximately 7% of all integration. We are currently sequencing full-length and truncated insertions produced by the dual reporter L1. Previous studies identified two resides in L1 that appeared to alter the ratio of 5' truncation to full-length. We found that mutations of these residues lowered the ratio of full-length to 5' truncated insertions by 2-fold. Future experiments with dual reporter L1 include the identification of host factors that affect the percentage of full-length insertions.