High-resolution nucleosome mapping of targeted regions using BAC-based enrichment

Abstract

We report a target enrichment method to map nucleosomes of large genomes at unprecedented coverage and resolution by deeply sequencing locus-specific mononucleosomal DNA enriched via hybridization with bacterial artificial chromosomes. We achieved ∼10 000-fold enrichment of specific loci, which enabled sequencing nucleosomes at up to ∼500-fold higher coverage than has been reported in a mammalian genome. We demonstrate the advantages of generating high-sequencing coverage for mapping the center of discrete nucleosomes, and we show the use of the method by mapping nucleosomes during T cell differentiation using nuclei from effector T-cells differentiated from clonal, isogenic, naïve, primary murine CD4 and CD8 T lymphocytes. The analysis reveals that discrete nucleosomes exhibit cell type-specific occupancy and positioning depending on differentiation status and transcription. This method is widely applicable to mapping many features of chromatin and discerning its landscape in large genomes at unprecedented resolution.

Comprehensive nucleosome mapping of the human genome in cancer progression

“…The periodic occurrence of A/T containing dinucleotides at ~10 bp intervals was calculated from first principles and verified in several subsequent studies [23][24][25][26] . When we examined the dinucleotide frequency of 150 bp fragments, we found the acknowledged 10bp periodicity for A/T containing dinucleotides, comparable to the frequency patterns identified in other human studies (Supplementary Figure 1C) 5,26,27 . These results affirm that the high resolution maps provided by mTSS-seq are consistent with the major qualitative features of nucleosome distribution at the TSS, and with the sequence composition of bona fide nucleosomal particles.…”

“…A full comprehension of the relationship between chromatin structure and genome function in cancer necessitates genome-wide chromatin structural measurements at multiple points in time throughout cancer progression. Although there have recently been a handful of extremely important studies measuring nucleosome distribution in a variety of organisms, there have been no genome-wide nucleosome distribution maps in primary patient tumors compared to their matched normal tissue [5][6][7]9,11,12,27,41,43 . To meet this need, we developed an innovative approach, mTSS-seq, to comprehensively measure genome wide nucleosome distribution changes in the progression of cancer.…”

Poly-dA:dT Tracts Form an In Vivo Nucleosomal Turnstile

“…It is possible that such a difference could result from the different histone-DNA contacts of the two DNA-strands. However, mouse [9] and human [10] , which have nucleosomes very similar to those of yeast (84% identical in histone fold domains, between mouse and yeast), display a trend opposite to yeast ( Figure 4 ); poly-A/poly-T combinations tend to be more depleted than poly-T/poly-A combinations, two consecutive poly-As generally result in 3′-biased NFRs, and, overall, there appear to be a more robust nucleosome boundaries 3′ to poly-As (5' to poly-Ts). This observation suggests that specific factors (e.g.…”

“… In vivo nucleosome occupancy for ( A–C ) regions with available high-resolution nucleosome data from mouse Th1 cells [9] and ( D–F ) non-repetitive regions on chromosome 22, for human granulocytes [10] (heatmaps) surrounding all instances of ( A , D ) poly-A/poly-A, ( B , E ) poly-A/poly-T, and ( C , F ) poly-T/poly-A combinations. Gaussian smoothed between rows (SD = 10 and 50, for mouse and human, respectively).…”

“…For the data displayed in Figure 2 , Figure 4 ( D–E ), and Figure S2 and S3 in File S1 ( in vitro and in vivo yeast sequencing data [1] , [8] , and MNase-digested chromatin from human granulocytes [10] ), we smoothed the data within each locus (Gaussian, SD = 20 bp), while for the data in Figure 3 (yeast microarray data [7] ) and Figure 4 ( A–C ) (mouse Th1 sequencing data, representing the centres of 147 bp fragments isolated from native, MNase-digested chromatin [9] ), we performed no such smoothing. Smoothing the data in this way makes it correspond more closely to nucleosome occupancy by distributing the dyad occupancy (nucleosome centre position) over the area covered by a nucleosome.…”