Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2010 Dec 21;107(51):21966-72.
doi: 10.1073/pnas.1016382107. Epub 2010 Dec 3.

DNA Transposon Hermes Inserts Into DNA in Nucleosome-Free Regions in Vivo

Affiliations
Free PMC article

DNA Transposon Hermes Inserts Into DNA in Nucleosome-Free Regions in Vivo

Sunil Gangadharan et al. Proc Natl Acad Sci U S A. .
Free PMC article

Abstract

Transposons are mobile genetic elements that are an important source of genetic variation and are useful tools for genome engineering, mutagenesis screens, and vectors for transgenesis including gene therapy. We have used second-generation sequencing to analyze ≈2 × 10(5) unique de novo transposon insertion sites of the transposon Hermes in the Saccharomyces cerevisiae genome from both in vitro transposition reactions by using purified yeast genomic DNA, to better characterize intrinsic sequence specificity, and sites recovered from in vivo transposition events, to characterize the effect of intracellular factors such as chromatin on target site selection. We find that Hermes transposon targeting in vivo is profoundly affected by chromatin structure: The subset of genome-wide target sites used in vivo is strongly associated (P < 2e-16 by Fisher's exact test) with nucleosome-free chromatin. Our characterization of the insertion site preferences of Hermes not only assists in the future use of this transposon as a molecular biology tool but also establishes methods to more fully determine targeting mechanisms of other transposons. We have also discovered a long-range sequence motif that defines S. cerevisiae nucleosome-free regions.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.
Fig. 1.
Hermes transposition mechanism. Hermes insertions have 8-bp target site duplications. Our experimental method specifically retrieves adjoining genomic DNA sequence at the right end of Hermes to retain both position and orientation information.
Fig. 2.
Fig. 2.
Sequences of in vitro target sites. (A) Composition plot of target sites from the in vitro experiment. IUBMB nomenclature for bases is used. R, purine; Y, pyrimidine; W, A/T; M, A/C; and K, G/T. (B) Sequence composition.
Fig. 3.
Fig. 3.
Target site distribution in the neighborhood of all yeast ORFs. In haploid (blue line), diploid (red line), and in vitro (green line) datasets, the insertion site frequency is represented as a deviation from the MRC (yellow dotted line). Data upstream and downstream of each gene are presented as positions relative to the transcription start site or transcription termination site. Insertions are plotted in percentage intervals along the gene.
Fig. 4.
Fig. 4.
Hermes target NFRs near the borders of yeast ORFs. (A) Hermes insertions are strongly correlated with regions of lower nucleosome occupancy upstream of transcription start sites (TSS). All TSS in the yeast genome are aligned, and Hermes target sites in haploid (blue), diploid (red), in vitro (green), and MRC (dashed, orange) datasets are plotted (after normalization) with respect to each neighboring TSS (Upper). Nucleosome occupancy (purple) as in Lee et al. (15) (Lower). (B) Hermes insertions are correlated with regions of lower nucleosome occupancy at transcription termination sites (TTS). Colors are as in A. (C) Insertion site distribution at yeast tRNAs Hermes target sites in haploid (blue), diploid (red), in vitro (green), and MRC (dashed, orange) datasets are plotted (after normalization) with respect to each tRNA TSS (Upper). Nucleosome occupancy (purple) as in Lee et al. (15) (Lower).
Fig. 5.
Fig. 5.
Favored target sites are more common in NFRs. The percentages of target sites in each of the categories of the in vivo data that fall into NFRs are shown, along with the percent of in vitro and matched random control target sites in NFRs and the percentage of the yeast genome that is occupied by NFRs.
Fig. 6.
Fig. 6.
Extended target site motifs by nucleosome occupancy. (A) Nucleotide frequency distributions in 1-kb regions surrounding all in vivo and all in vitro target sites recovered in this study. (B) Nucleotide frequency distribution in the 500 bp flanking the midpoint of all Hermes target sites found in NFRs (Left), regions of intermediate nucleosome occupancy (IOR) (Center), and nucleosome occupied regions (NOR) (Right) as defined in Fig. S2. (C) Corresponding nucleotide frequency distribution of 500 bp defining NFRs, IORs, and NORs in the yeast genome.
Fig. 7.
Fig. 7.
Clusters of insertions. (A) Hotspots and clusters. (B) Distribution of target site clusters in NORs, IORs, and NFRs. (C) Sequence composition of 150-bp windows. Windows are centered at the midpoint of randomly chosen fragments of the yeast genome, MRC dataset, in vitro or in vivo cluster, and hotspots.

Similar articles

See all similar articles

Cited by 44 articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources

Feedback