Zhao et al. Epigenetics & Chromatin (2019) 12:27 https://doi.org/10.1186/s13072-019-0272-y Epigenetics & Chromatin RESEARCH Open Access Profling of chromatin accessibility and identifcation of general cis-regulatory mechanisms that control two ocular lens diferentiation pathways Yilin Zhao1, Deyou Zheng1,2 and Ales Cvekl1,3* Abstract Background: Promoters and enhancers are cis-regulatory DNA sequences that control specifcity and quantity of transcription. Both are rich on clusters of cis-acting sites that interact with sequence-specifc DNA-binding transcrip- tion factors (TFs). At the level of chromatin, these regions display increased nuclease sensitivity, reduced nucleosome density, including nucleosome-free regions, and specifc combinations of posttranslational modifcations of core histone proteins. Together, “open” and “closed” chromatins represent transcriptionally active and repressed states of individual genes, respectively. Cellular diferentiation is marked by changes in local chromatin structure. Lens mor- phogenesis, regulated by TF Pax6, includes diferentiation of epithelial precursor cells into lens fbers in parallel with diferentiation of epithelial precursors into the mature lens epithelium. Results: Using ATAC-seq, we investigated dynamics of chromatin changes during mouse lens fbers and epithelium diferentiation. Tissue-specifc features of these processes are demonstrated via comparative studies of embryonic stem cells, forebrain, and liver chromatins. Unbiased analysis reveals cis-regulatory logic of lens diferentiation through known (e.g., AP-1, Ets, Hsf4, Maf, and Pax6 sites) and novel (e.g., CTCF, Tead, and NF1) motifs. Twenty-six DNA-binding TFs, recognizing these cis-motifs, are markedly up-regulated in diferentiating lens fbers. As specifc examples, our ATAC-seq data uncovered both the regulatory regions and TF binding motifs in Foxe3, Prox1, and Mip loci that are consistent with previous, though incomplete, experimental data. A cross-examination of Pax6 binding with ATAC-seq data demonstrated that Pax6 bound to both open (H3K27ac and P300-enriched) and closed chromatin domains in lens and forebrain. Conclusions: Our study has generated the frst lens chromatin accessibility maps that support a general model of stage-specifc chromatin changes associated with transcriptional activities of batteries of genes required for lens fber cell formation. Analysis of active (or open) promoters and enhancers reveals important cis-DNA motifs that estab- lish the molecular foundation for temporally and spatially regulated gene expression in lens. Together, our data and models open new avenues for the feld to conduct mechanistic studies of transcriptional control regions, reconstruc- tion of gene regulatory networks that govern lens morphogenesis, and identifcation of cataract-causing mutations in noncoding sequences. Keywords: ATAC-seq, Diferentiation, Lens, “Open” chromatin, RNA-seq, Tissue specifcity *Correspondence: [email protected] 1 The Departments of Genetics, Albert Einstein College of Medicine, Bronx, NY 10461, USA Full list of author information is available at the end of the article © The Author(s) 2019. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Zhao et al. Epigenetics & Chromatin (2019) 12:27 Page 2 of 23 Introduction into the lens epithelium while its posterior part difer- Cellular diferentiation is driven by the coordinated entiates into the “primary” lens fbers to fll up the space expression of batteries of genes that encode proteins by E14.5 of mouse embryogenesis (Fig. 1a). Te bulk of required for cellular specialization and function. Difer- the mature lens is formed by highly elongated lens fb- entiation is mostly regulated at the level of transcrip- ers that represent cells at the terminal diferentiation tion. Tissue specifcity of transcription is primarily state. From E14.5, lens growth is driven by a proliferat- regulated by a combinatorial action of sequence-spe- ing subpopulation of the lens epithelium (“germinative cifc DNA-binding transcription factors and their inter- zone”) that exits the cell cycle at the lens equator and actions with promoters and distal enhancers [1, 2]. Both generates “secondary” lens fbers [8, 9]. Te newborn promoters and enhancers of transcriptionally active (P0.5) mouse lens is comprised of approximately 30,000 genes display increased sensitivity to nuclease digestion epithelial and over 140,000 lens fber cells (Fig. 1a) [9, [3] and are located to “open” chromatin domains. Open 12]. Recent transcriptome analysis of both lens epithe- chromatin regions display lower nucleosomal den- lium and fbers by RNA-seq revealed thousands of dif- sity or even nucleosome-free regions [2]. In addition, ferentially expressed genes (DEGs) between these two active chromatin is marked by a specifc combination compartments, including a number of highly lens-spe- of modifed core histone proteins (H3K27ac, H3K4me1, cifc genes [13]. Analysis of transcription factors (TFs) and H3K4me3) and by the presence of H2A.Z and H3.3 and their DNA-binding motifs among the diferentially histone variants [4–6]. In contrast, transcriptionally expressed genes confrmed pivotal roles of many known repressed genes are often organized within “closed” TFs and implicated novel TFs that may help drive lens chromatin domains, marked by diferent histone modi- diferentiation. A major question in the feld is to link fcations (e.g., H3K27me3 and H3K9me3) [2, 7]. Hence, the expression data with changes in chromatin struc- distinct cell types as well as diferentiation cascades ture. In turn, the analysis of chromatin accessibility yielding diferentiated cells display unique chroma- would allow genome-wide identifcation of promoter tin structure. Studies of chromatin dynamics, such as regions and distal enhancers and unbiased analysis of changes in open chromatin regions during diferentia- the cis-regulatory grammar [2] of the lens-specifc gene tion and between diferent cell types, thus provide criti- expression programs. cal insights into the molecular mechanisms of tissue Te entire process of lens diferentiation, from the birth specialization and diferentiation. of individual lens progenitor cells to the terminal dif- Te ocular lens is a highly specialized tissue that ferentiation of lens fbers and massive up-regulation of is formed from a single lens progenitor cell type. Te crystallin gene expression, is directly or indirectly regu- progenitors give rise to the lens vesicle, comprised of lated by the transcription factor Pax6 [10, 14, 15]. Pax6 lens precursor cells that ultimately diferentiate into acts as a dual transcriptional activator and repressor [16] lens epithelial and lens fber cells [8, 9]. Tus, lens dif- through interactions with and recruitment of diferent ferentiation is an excellent model to study early (i.e., chromatin remodeling complexes [17]. Other well-stud- formation of lens progenitors), middle (i.e., formation ied transcription factors of lens diferentiation include of lens fbers and lens epithelium), and late stages of c-Maf, FoxE3, Gata3, Hsf4, Mab21l1, Msx2, Pitx3, Prox1, diferentiation (maturation of epithelium and terminal Sox1, Sox2, Sox11, Tfap2a (AP-2α), and Zeb2 (Sip1) [11]. diferentiation–maturation of lens fbers including deg- More recent studies point to the involvement of tran- radation of their nuclei) [10]. Mouse lens diferentiation scription factors downstream of BMP, FGF, Hippo-Yap, is the leading model for understanding the complexity integrin, and Notch signaling [11]. Nevertheless, the reg- of these processes [11]. Te lens precursor cells form a ulatory mechanisms and genome-wide targets of these transitional polarized structure, the hollow lens vesicle transcription factors remain poorly understood in lens in E11.5 mouse embryos. Its anterior part diferentiates diferentiation. (See fgure on next page.) Fig. 1 ATAC-seq analysis of lens fbers and lens epithelium. a Schematic diagram of embryonic E14.5 and newborn (P0.5) lens morphology. b Circos plot of global chromatin accessibility in all ATAC-seq samples (mean read counts inside peaks from biological replicates normalized by all read numbers inside all peaks). The arrows mark some genes with highest peaks in E14.5 fbers. c ATAC-seq signal tracks of the genes marked in panel b. d Pie charts show the genomic distributions of peaks. e Principal component analysis of the top 500 peaks with the biggest variance across 8 samples. The arrows show lens fber cell diferentiation (E14.5 epithelium (epi) E14.5 fbers P0.5 fbers) and epithelium maturation → → (E14.5 epi P0.5 epi) paths. f Heat maps show the read densities of lens, forebrain, liver, and ESCs ATAC-seq data within 5 kb from the centers of → ± diferential accessible regions (DARs) which are from the pairwise comparison. Epi and fber represent epithelial and fber cells Zhao et al. Epigenetics & Chromatin (2019) 12:27 Page 3 of 23 b Cryga a P ax6 E14.5 epithelium E14.5 lens E14.5 fiber ATAC-seq Cryaa RNA-seq C P0.5 epithelium rybb1