Genome-wide analysis of Agrobacterium T-DNA insertion sites in the Arabidopsis genome generated under non-selective conditions.

摘要

Agrobacterium-mediated transformation is an unique phenomenon of horizontal gene transfer between prokaryotes and eukaryotes. This process occurs by transferring a DNA (T-DNA) segment from the Ti plasmid of A. tumefaciens into host plant chromosomes with the aid of several virulence proteins. Although T-DNA can be found any place in the plant genome, a number of previous studies suggested that T-DNA integrated preferentially into promoter regions, A/T rich regions, or gene-rich regions of the genome. However, these interpretations were derived from transformation libraries generated under selective conditions which required expression of marker transgenes. Such expression might cause a bias in the selection of transformants.; I therefore performed an analysis of T-DNA insertion sites in the Arabidopsis genome using a library of T-DNA/plant DNA junctions generated under non-selective conditions. The frequency with which T-DNA insertions in this library mapped to genie and intergenic regions closely resembled their respective proportions of the Arabidopsis genome. Contrary to what others previously found using selective conditions, I found a relatively high frequency of T-DNA insertions in repetitive sequences. Analysis using a randomly generated in silico Arabidopsis library indicated that T-DNA insertion sites generated under non-selective conditions were not significantly biased toward any chromosomal region, whereas T-DNA insertion sites recovered using selective conditions were significantly enriched in 5' upstream regions.; In an attempt to elucidate whether transcriptional activity of the chromatin affects selection of T-DNA integration sites, I investigated transcriptional activity of T-DNA integration sites prior to infection. I used a microarray and custom-made DNA array containing 1 kbp genomic fragments surrounding T-DNA integration sites generated from non-selected and selected libraries and random Arabidopsis genomic fragments as a control. Hybridization analysis with cDNA and nuclear run-on transcript probes indicated that the expression level of junction sequences recovered after selection was significantly higher than that of random Arabidopsis sequences, whereas the expression level of genomic sequences targeted by T-DNA under non-selective conditions was similar to that of randomly selected Arabidopsis sequences.; In conclusion, my results suggest that T-DNA insertion into the genome is random and does of show a bias toward transcriptionally active regions. Selection pressure might shift the recovery of T-DNA insertions into gene-rich or transcriptionally active regions of chromatin.