Transposon insertion site profiling chip (TIP-chip)

Abstract

Mobile elements are important components of our genomes, with diverse and significant effects on phenotype. Not only can transposons inactivate genes by direct disruption and shuffle the genome through recombination, they can also alter gene expression subtly or powerfully. Currently active transposons are highly polymorphic in host populations, including, among hundreds of others, L1 and Alu elements in humans and Ty1 elements in yeast. For this reason, we wished to develop a simple genome-wide method for identifying all transposons in any given sample. We have designed a transposon insertion site profiling chip (TIP-chip), a microarray intended for use as a high-throughput technique for mapping transposon insertions. By selectively amplifying transposon flanking regions and hybridizing them to the array, we can locate all transposons present in a sample. We have tested the TIP-chip extensively to map Ty1 retrotransposon insertions in yeast and have achieved excellent results in two laboratory strains as well as in evolved Ty1 high-copy strains. We are able to identify all of the theoretically detectable transposons in the FY2 lab strain, with essentially no false positives. In addition, we mapped many new transposon copies in the high-copy Ty1 strain and determined its Ty1 insertion pattern.

Fig. 1.

TIP-chip workflow. (A) Choosing restriction enzyme combinations for parallel digests of the yeast genome. The enzymes cut the DNA into overlapping pieces, so that each nucleotide of the yeast genome is contained in three separate restriction fragments, one from each enzyme. More than 96% of the yeast genome is contained in at least one fragment >1 kb and <10 kb; these somewhat arbitrary limits were chosen based on previous experience with PCR amplification and the proposed array design. (B) The Ty1 element, with LTRs shown as arrowheads. The small arrow at the 5′ end of the Ty1 denotes the position of the Ty1-specific primer used (JB8784; see supporting information). (C) Preparation of genomic DNA for hybridization to the TIP-chip. Genomic DNA is digested in three parallel reactions, with three restriction enzymes with 6-base recognition sequences. The digested fragments are ligated to digest-specific vectorettes and amplified by using vectorette PCR. Longer amplicons may not amplify well and may be underrepresented in the resulting mixture. The amplicons are then pooled and digested in three parallel reactions with three enzymes with 4-base recognition sequences. The resulting fragments are heat-inactivated, labeled, and hybridized to the microarray.

Fig. 2.

Typical hybridization of FY2 amplicons to a TIP-chip. Each putative transposon flank appears as a line on the array. The bound features are numbered; these numbers correspond to Table 1. The numbers are placed so that they are nearest the endpoint of the linear signal closest to the Ty1 element and thereby indicate the orientation of the Ty1 element. Ty1 hybridization controls (features spanning the LTRs) in the middle of the array produce the “TY” pattern. Interruptions in the lines of spots represent intervening hybridization negative controls.

Fig. 3.

Two Ty1 insertion sites from the FY2 strain, shown as graphs of log normalized intensity versus chromosomal coordinates. The top graph, from chromosome 12, shows a new Ty1 insertion site (line 26 in Fig. 2), in which the Ty1 lies on the right side of the line (downstream in chromosome coordinates, confirmed by PCR), giving the line a positive slope. The bottom graph displays the same information for two known tail-to-tail Ty1 insertions on chromosome 16 (line 37 in Fig. 2). The gap in the line is due to masking of the 6-kb Ty1 elements; there are no features spanning this region. Arrowheads mark positions of confirmed and known Ty1 elements. Blue brackets mark regions for which the Z score is >2.5 for each spot (P value ≈ 0.01 for each spot, therefore much lower for the entire line).

Fig. 4.

Histogram showing confirmed new Ty1 insertion positions relative to transcription start sites of tRNA genes (at 0). Bin size is 150 nucleotides; orientation of Ty1s is indicated.