Outward-oriented sites within clustered CTCF boundaries are key for intra-TAD chromatin interactions and gene regulation

Ge, Xiao; Huang, Haiyan; Han, Keqi; Xu, Wangjie; Wang, Zhaoxia; Wu, Qiang

doi:10.1038/s41467-023-43849-0

Download PDF

Article
Open access
Published: 07 December 2023

Outward-oriented sites within clustered CTCF boundaries are key for intra-TAD chromatin interactions and gene regulation

Xiao Ge^1,2^na1,
Haiyan Huang ORCID: orcid.org/0009-0008-8326-4533^1,2^na1,
Keqi Han^1,2,
Wangjie Xu³,
Zhaoxia Wang³ &
…
Qiang Wu ORCID: orcid.org/0000-0003-3841-3591^1,2

Nature Communications volume 14, Article number: 8101 (2023) Cite this article

1878 Accesses
18 Altmetric
Metrics details

Subjects

Abstract

CTCF plays an important role in 3D genome organization by adjusting the strength of chromatin insulation at TAD boundaries, where clustered CBS (CTCF-binding site) elements are often arranged in a tandem array with a complex divergent or convergent orientation. Here, using Pcdh and HOXD loci as a paradigm, we look into the clustered CTCF TAD boundaries and find that, counterintuitively, outward-oriented CBS elements are crucial for inward enhancer-promoter interactions as well as for gene regulation. Specifically, by combinatorial deletions of a series of putative enhancer elements in mice in vivo or CBS elements in cultured cells in vitro, in conjunction with chromosome conformation capture and RNA-seq analyses, we show that deletions of outward-oriented CBS elements weaken the strength of long-distance intra-TAD promoter-enhancer interactions and enhancer activation of target genes. Our data highlight the crucial role of outward-oriented CBS elements within the clustered CTCF TAD boundaries in developmental gene regulation and have interesting implications on the organization principles of clustered CTCF sites within TAD boundaries.

In vivo dissection of a clustered-CTCF domain boundary reveals developmental principles of regulatory insulation

Article Open access 11 July 2022

Promoter-proximal CTCF binding promotes distal enhancer-dependent gene activation

Article 04 January 2021

Enhancer accessibility and CTCF occupancy underlie asymmetric TAD architecture and cell type specific genome topology

Article Open access 02 July 2019

Introduction

The interphase genome is organized into highly dynamic structures including chromosome territory, compartment, topologically associated domain (TAD), and chromatin loop¹. CTCF (CCCTC-binding factor) is a key architecture protein for interphase 3D genome organization and CBS (CTCF-binding site) elements throughout the mammalian genomes function as topological insulators to block aberrant enhancer activation and ensure proper gene expression, or to mark the boundaries between euchromatin and heterochromatin^2,3,4. A prevailing model for interphase chromosome folding, known as loop extrusion, posits that the cohesin complex extrudes chromatin fibers bidirectionally into expanding loops until blocked asymmetrically by orientated CBS elements^5,6,7. Specifically, continuous ATP-driven active loop extrusions lead to gradually larger-sized chromatin loops which are stabilized by pairs of convergent or opposite CBS elements. This model explains TAD formation and is consistent with the observations that the two flanking boundaries of each TAD are associated with mostly convergent forward-reverse CBS elements^8,9. However, it is puzzling that most TAD boundaries in mammalian genomes contain clustered CBS elements with complex orientations of both convergence and divergence.

TAD boundaries emerged as insulator elements that restrain inter-TAD chromatin interactions. Their dysfunction leads to abnormal development and progressive oncogenesis^10,11,12. Mammalian clustered CTCF TAD boundaries are more evolutionarily conserved than other genomic regions^13,14,15. High insulation score of TAD boundaries is associated with strong CTCF enrichment, and clustered enhancers, known as super-enhancers, tend to be insulated by strong boundaries^16,17. Divergent CBS elements are enriched at TAD boundaries; however, a divergent orientation signature is not strictly required for effective insulation. For example, the characteristics of specific CBS elements can outweigh CBS number and orientation¹⁸. Finally, insulation potency depends on the local context of CBS elements and robust enhancer-promoter communications can bypass CTCF insulation regardless of the intervening CBS strength^19,20,21,22.

As a paradigm for investigating mechanisms of 3D genome folding, the mouse clustered protocadherins (cPcdh) comprise three sequential gene clusters, Pcdh α, β, and γ, encoding 58 isoforms and spanning ~1 million bps (Fig. 1a)²³. The Pcdh α and γ clusters consist of 14 and 22 variable first exons, respectively, each of which is cis-spliced to a single set of cluster-specific constant exons; whereas the Pcdhβ cluster consists of 22 single variable exons (β1-β22) with no constant exon. The Pcdh promoter choice is determined by CTCF/cohesin-mediated long-range enhancer-promoter looping interactions anchored by convergent or opposite CBS elements^9,24,25. Their combinatorial expression patterns in single neurons generate enormous cell-surface molecular diversity for assembling complex networks of neuronal connectivity in the brain^26,27. In addition, precise expression patterns of members of the HoxD gene cluster are central for limb development^28,29. Here, using Pcdh and HOXD as model genes, we find that outward-oriented CBS elements within clustered CTCF boundaries are crucial for intra-TAD promoter-enhancer interactions and gene regulation. Our data shed interesting light on the organizational principles of the clustered CTCF TAD boundaries in 3D genome architecture and gene regulation.

**Fig. 1: The downstream *cPcdh* superTAD boundary comprises clustered enhancers for *Pcdhβγ* gene regulation.**

Results

The downstream boundary of cPcdh superTAD comprises a cluster of CTCF sites with mixed orientations

During mouse neocortical development, the 58 cPcdh genes are organized into a large megabase-sized superTAD comprised of the Pcdhα and Pcdhβγ TADs (Fig. 1a; Supplementary Fig. 1a–c)^9,30. At the telomeric boundary of the cPcdh superTAD, there is a complex array of eight CTCF sites or CBS elements (CBS a-h, Fig. 1a–c). Among these eight CBS elements, CBSa-e, CBSg, and CBSh are in the reverse orientation (Supplementary Fig. 1d). By contrast, a single CBSf is in the forward orientation (Supplementary Fig. 1e). We performed chromatin immunoprecipitation followed by high-throughput sequencing (ChIP-seq) experiments with a specific antibody against CTCF using microdissected mouse neocortical tissues and found that CTCF is enriched at these elements (Fig. 1c). Interestingly, the CBS b and c elements form a single peak and CBS f and g elements form a single peak (Fig. 1c). To see whether these elements are anchors of cohesin loop extrusion, we performed ChIP-seq experiments with a specific antibody against Rad21, a subunit of cohesin, and found that cohesin is co-localized with CTCF at all of these CBS elements, suggesting that cohesin loop extrusion from this boundary may regulate cPcdh gene expression (Fig. 1b, c).

We first comprehensively mapped expression patterns of each member of the three Pcdh gene clusters at a series of 12 developmental stages by collecting embryos every other day and new pubs every other 2 days. RNA-seq of microdissected neocortical tissues showed that each member of the three Pcdh gene clusters is expressed dynamically and that the cPcdh genes are most abundantly expressed at the newborn stage (Supplementary Fig. 2). Thus, we used the newborn mouse neocortical tissues in the following gene regulation studies.

Within the Pcdhα TAD, members of the Pcdhα cluster are activated by a super-enhancer comprised of HS5-1 and HS7 (Fig. 1b and Supplementary Fig. 3a, b)^24,31,32. However, how members of the Pcdhβγ clusters are regulated within the Pcdhβγ TAD is not clear. To this end, we profiled the chromatin regulatory landscape by assaying for transposase-accessible chromatin with high-throughput sequencing (ATAC-seq) and identified six highly accessible regions at the telomeric boundary of the Pcdhβγ TAD, corresponding to the previously identified DNaseI hypersensitive sites of HS7L, HS5-1aL, HS5-1bL, HS18, HS19-20, and HS21 (Fig. 1c and Supplementary Fig. 3c, d)^24,33. Finally, we performed H3K27ac and H3K4me1 ChIP-seq and found, except the last one (HS21), all of the other five clustered HS elements are marked by H3K27ac and H3K4me1, suggesting that they form a super-enhancer for members of the Pcdhβγ gene clusters (Fig. 1c).

The Pcdhβγ TAD boundary comprises a super-enhancer in mice in vivo

We dissected the putative enhancers in mice in vivo by CRISPR/Cas9-based DNA-fragment editing system (Supplementary Fig. 4a–c)⁹ to test their functionality. We first deleted the HS7L element (Supplementary Fig. 4b, c) and found, surprisingly, it does not lead to decreased levels of cPcdh gene expression in neocortical and olfactory tissues (Fig. 1d, Supplementary Fig. 4e–g, and Supplementary Fig. 5a). Because HS5-1aL is in the Pcdhγ constant coding region and its deletion will destroy the transcript integration, we mutated the CBSa element within HS5-1aL by homologous recombination using CRISPR with an ssODN donor. We found that CBSa mutation or deletion abolished CTCF enrichments (Supplementary Fig. 4d). Interestingly, it only results in a moderate decrease of expression levels of members of the Pcdhβγ, but not α, clusters (Fig. 1e, Supplementary Figs. 4e, 5b), suggesting a limited role of CBSa in tethering HS5-1aL to Pcdhβγ.

We next deleted the entire region covering all of the other three putative enhancers (from HS5-1bL to HS19-20, CBSb-e) in mice and found that, in contrast to HS7 and HS5-1aL, this large deletion leads to a significant decrease of levels of Pcdhβγ gene expression, suggesting that enhancers for members of Pcdhβγ clusters reside in this region (Fig. 1f and Supplementary Fig. 5c). We then dissected this region in great details and found that deletion of the large DNA fragment covering both HS5-1bL and HS18 (CBSb-d) results in a significant decrease of levels of Pcdhβγ gene expression (Fig. 1g and Supplementary Fig. 5d). In addition, deletion of the DNA fragment covering both HS18 and HS19-20 (CBSde) also results in a significant decrease of levels of Pcdhβγ gene expression (Fig. 1h and Supplementary Fig. 5e).

Finally, we deleted each of these three elements individually in mice and found that, surprisingly, deletion of HS5-1bL (CBSbc) has no effect on Pcdhβγ gene expression (Fig. 1i and Supplementary Fig. 5f). By contrast, deletion of either HS18 (CBSd) or HS19-20 (CBSe) result in a significant decrease of levels of Pcdhβγ gene expression (Fig. 1j, k and Supplementary Fig. 5g, h). These data suggest that these two H3K27ac-enriched elements play a leading role in activating Pcdhβγ genes in vivo in the mouse brain. Consistently, 4C experiments revealed close contacts between Pcdhβγ and HS18-20 (Supplementary Fig. 6a, b). Unlike HS5-1, which is brain-specific, HS18-20 are enriched with H3K27ac and H3K4me3 marks in multiple tissues (Supplementary Fig. 6c, d), explaining why members of the Pcdhβγ gene clusters are expressed in both neural and non-neural tissues. Taken together, these data suggest that the downstream boundary of the Pcdhβγ TAD is a super-enhancer comprising a cluster of enhancers among which HS18 and HS19-20 are the most important.

HS5-1bL is a Pcdhγc3-specific insulator in mice in vivo

Pcdhγc3 is the most abundantly expressed isoform in the mouse neocortex and has a unique role in dendrite arborization (Supplementary Fig. 2e)³⁴. However, the underlying mechanism remains unclear. To this end, we performed QHR-4C experiments and found that Pcdhγc3 is in close contacts with HS5-1bL, HS18, and HS19–20 (Fig. 2a). Consistently, deletion of HS18 and/or HS19-20 results in a significant decrease of levels of Pcdhγc3 gene expression, suggesting that HS18 and HS19-20 are enhancers for Pcdhγc3 (Fig. 2b–d). Surprisingly, deletion of HS5-1bL in mice in vivo leads to a significant increase of levels of Pcdhγc3, but not γc4 or γc5, gene expression (Fig. 2e), suggesting that HS5-1bL is a putative insulator specifically blocking the activation of Pcdhγc3 by the HS18 and HS19-20 enhancers. To this end, we performed 4 C experiments with Pcdhγc3 as an anchor using neocortical tissues of the HS5-1bL deleted mice and found a significant increase of long-distance chromatin interactions between Pcdhγc3 and HS18–20 (Fig. 2f). To confirm this aberrantly increased chromatin contacts upon deletion of HS5-1bL, we performed 4 C with HS18 or HS19–20 as an anchor and indeed observed increased chromatin contacts with Pcdhγc3 (Fig. 2g). In conjunction with no expression increase of all other members of the Pcdhβγ gene clusters upon deletion of HS5-1bL (Fig. 1i), we conclude that HS5-1bL is a γc3-specific insulator.

**Fig. 2: *HS5-1bL* is an insulator specific for *Pcdhγc3*.**

Clustered Pcdhs are expressed from H3K9me3-enriched facultative heterochromatins of superTAD

We next mapped the chromatin landscape of a wide variety of histone modifications in the cPcdh superTAD and found that this region is enriched for both the repressive mark of H3K9me3 and the active mark of H3K36me3 (Fig. 3a), suggesting that cPcdhs are expressed from facultative heterochromatin region of the superTAD. In addition, expression of each member of the three Pcdh gene clusters is strongly correlated with active marks of H3K4me3 and H3K9ac (Fig. 3b and Supplementary Figs. 7, 8). Moreover, this region is also marked by the repressive histone marks of H3K9me2 in the mouse neocortex (Fig. 3c and Supplementary Fig. 9). Specifically, the coding region of each member of the Pcdh clusters is correlated with a peak of H3K9me2 (Supplementary Fig. 9) but not H3K9me3 (Supplementary Fig. 10). To further investigate mechanisms of cPcdh gene regulation, we analyzed methylated DNA immunoprecipitation sequencing (MeDIP-seq) data and found that each member of the H3K9me3-enriched cPcdhs is hypermethylated on the body of each variable exon (Fig. 3c, Supplementary Fig. 11a–d). We noted that Pcdhβ1 differs from other cPcdh members in that its DNA-methylation peak is much closer to the transcription start site (Supplementary Fig. 11e), explaining why it is silenced in the mouse neocortex (Fig. 3b).

**Fig. 3: The *cPcdh* genes are expressed from H3K9me3/H3K36me3-enriched facultative heterochromatin domains.**

Outward-oriented CBSf is the key boundary CTCF site

All of the HS5-1aL, HS5-1bL, HS18, and HS19-20 enhancers contain reverse-oriented CBS elements (CBSa-e, Fig. 1c) that are enriched for cohesin and are in close contacts with the forward-oriented CBS elements of the Pcdhβγ promoters via CTCF/cohesin-mediated chromatin loops. In addition, there are two more reverse-oriented CBS elements, CBSg and CBSh, located immediately downstream of this super-enhancer. However, there is a single forward-oriented CBSf element located between the super-enhancer and the two reverse-oriented CBS g and h elements (Fig. 4a). To this end, we generated a CBSf deletion mouse line (Fig. 4a and Supplementary Fig. 12a) and confirmed the abolishment of CTCF and cohesin enrichments at the location of CBSf but not CBSg (Fig. 4a, b). In addition, CTCF and cohesin binding at other CTCF sites within this boundary is not altered by CBSf deletion (Supplementary Fig. 12b).

**Fig. 4: Outward-oriented *CBSf* regulates *Pcdhβγ* by blocking their aberrant looping from the outside of the TAD boundary.**

Remarkably, deletion of the forward-oriented CBSf element results in a significant decrease of expression levels of members of the Pcdhβγ gene clusters (Fig. 4c and Supplementary Fig. 13a), demonstrating that the single forward-oriented CBSf plays a crucial role in the regulation of Pcdhβγ gene expression. To investigate the underlying mechanism, we performed a series of QHR-4C experiments with members of the Pcdhβγ gene clusters as anchors and found a significant decrease of long-distance chromatin contacts between the super-enhancer and members of the Pcdhβγ gene clusters (Fig. 4d). In particular, we observed a significant increase of long-distance chromatin contacts with CBSh with a sharp edge at the location of CBSf (Fig. 4e). To confirm this, we performed reciprocal capture experiments with CBSh as an anchor and observed a significant increase of long-distance chromatin contacts with members of the Pcdhβγ gene clusters (Supplementary Fig. 12c). In addition, we profiled the histone modifications and found decrease in enrichments of repressive marks, H3K9me3 and H3K9me2, but not of active marks, H3K36me3, H3K4me3, and H3K9ac (Supplementary Figs. 14–17).

Finally, combined deletion of both CBSf and CBSg elements (Supplementary Fig. 12d) results in a similar phenotype (Supplementary Figs. 18, 12e, 13b). As a control, single deletion of CBSg results in a less decrease in Pcdhβγ expression (Supplementary Figs. 12f–h and 19). Taken together, we conclude that a single outward-oriented CBSf element within the clustered CTCF TAD boundary plays a key role in Pcdhβγ gene regulation and long-distance intra-TAD chromatin interactions.

Outward-oriented CBS elements within a clustered CTCF TAD boundary are crucial for HOXD13 expression

To see whether it is a general phenomenon for a key role of the outward-oriented boundary CBS element in intra-TAD chromatin interactions and gene regulation, we investigated the HOXD left boundary, which has a complex array of six conserved clustered CBS elements (CBS1-6, Supplementary Fig. 20a–d), of the human centromeric TAD (C-DOM) (Fig. 5a, b)^28,29. The human HOXD locus is located at the boundary between C-DOM and T-DOM (telomeric TAD) (Supplementary Fig. 21a, b). It contains nine HOXD genes, of which only HOXD13 is associated with a cluster of reverse-oriented CBS elements (CBS8-12, Fig. 5c). By contrast, HOXD8 and HOXD9 have two forward-oriented CBS elements (CBS13,14) and HOXD10-12 and HOXD1-4 have no CBS element (Fig. 5a–c and Supplementary Fig. 20a). We analyzed Hi-C and found that there are strong long-distance chromatin interactions between the human HOXD13 and the left boundary of C-DOM, consistent with the convergent rule of forward-reverse CBS elements (Supplementary Fig. 21a, b). Interestingly, within the left boundary, there are two outward-oriented CBS elements (CBS3 and CBS5, Fig. 5b and Supplementary Fig. 21b).

**Fig. 5: Double knockout of outward-oriented *CBS3* and *CBS5* elements downregulates *HOXD13* gene expression.**

We screened single-cell clones for deletion of both CBS3 and CBS5 elements and obtained three homozygous double-knockout cell clones (ΔCBS3 + 5, Supplementary Fig. 20g). We first confirmed the loss of CTCF enrichments at both CBS3 and CBS5 (Fig. 5a, b). We then performed 4C with HOXD13 as an anchor and found that there is a significant increase of chromatin interactions with the left boundary upon deletions of the two outward-oriented CBS elements (Fig. 5d). These increased chromatin interactions were confirmed by reciprocal 4C experiments with CBS2 or CBS4 as an anchor (Fig. 5e, f). By contrast, there is a significant decrease of chromatin interactions with proximal sequences (Fig. 5d), which is confirmed by reciprocal 4C experiments with GT2 or LNPK as an anchor (Supplementary Fig. 21c, d). In addition, we performed RNA-seq experiments and found that deletion of both outward-oriented CBS elements results in a significant decrease of expression levels of HOXD13 but not HOXD12-1, indicating that the two outward-oriented CBS elements play a key role in HOXD13 gene regulation (Fig. 5g). However, different from the Pcdhβγ genes which are associated with CBS elements, there is no CBS element associated with the LNPK gene within C-DOM and its expression is not altered upon deletion of outward-oriented CBS elements (Fig. 5g). Finally, single deletions of the CBS3 or CBS5 element demonstrated a cooperation role between the two outward-oriented CBS elements in HOXD13 gene regulation (Supplementary Fig. 22).

Discussion

The enormous connectivity of billions of neurons is intricately related to the precise cell-specific expression patterns of cPcdh genes in the brain. Specifically, the isoform-promiscuous cis hetero-dimerization and isoform-specific trans homo-dimerization of combinatorially expressed cPcdhs mediate cell-specific recognition between neuronal membranes, resulting in ingenious dendrite self-avoidance and axonal coexistence^{27,35,36,37,38,39,40,41,42}. Intriguingly, chromatin establishes a stable cPCDH expression pattern for single neurons in the brain during human fetal development⁴³. In addition, the proper spatiotemporal expression of the HoxD genes is central for limb development^29,44. Here, we first comprehensively mapped cPcdh expression patterns in the neocortex during mouse brain development. We then dissected, through sequential and combinatorial deletion of candidate enhancer elements, how cPcdhs are regulated by a downstream clustered CTCF boundary super-enhancer. In addition, we further deleted outward-oriented CBS elements within Pcdh and HOXD clustered CTCF TAD boundaries, which are associated with dozens of human pathogenic copy number variations (CNVs) (Supplementary Data 1). In conjunction with chromosome conformation and gene expression analyses, we found that outward-oriented CBS elements within clustered CTCF TAD boundaries are crucial for intra-TAD promoter-enhancer interactions and proper gene regulation (Fig. 5h). Unlike inward-oriented CBS elements, which directly make intra-TAD spatial contacts, such as between enhancers and promoters, the outward-oriented CBS elements may block extruding cohesin sliding from the outside of the TAD to prevent aberrant inter-TAD chromatin interactions (Fig. 5h).

The characteristics of the clustered CTCF TAD boundaries and their exact nature of insulation activities are under intense investigation^{13,18,20,22,29,44,45,46,47,48}. The CBS elements within a clustered CTCF TAD boundary between Pax3 and its distal enhancer cooperate redundantly for robust insulation¹⁸. In addition, various combinations of clustered CBS elements in distinct orientations could function as topological insulators when inserted between distal enhancers and their target promoters^20,22,49. Moreover, the CBS elements within a clustered CTCF TAD boundary located between the C-DOM and T-DOM (telomeric domain) of HoxD have various functions for controlling its spatiotemporal expression patterns during limb development^29,44. Here, we show that the outward-oriented CBS elements of the left boundary of C-DOM are crucial for maintaining intra-TAD chromatin interactions and HOXD13 gene expression (Fig. 5). Finally, consistent with the crucial role of outward-oriented CBS elements, an inversion leading to increased numbers of outward-oriented CBS elements within a clustered CTCF TAD boundary results in enhanced insulation activity and altered gene expression⁵⁰.

CTCF plays a key role in the formation of insulation at TAD boundaries, where in many cases multiple CBS elements are arranged in a clustered array with complex and mixed orientations, such as the telomeric downstream boundary of the Pcdhβγ TAD and the centromeric upstream boundary of the C-DOM of the HOXD cluster. According to the convergent rule, the inward-oriented CBS elements within the clustered CTCF TAD boundaries are essential for intra-TAD chromatin interactions^8,9 (Fig. 5h). Thus, chromatin contacts such as long-distance enhancer-promoter interactions primarily occur within the TADs^51,52. However, recent studies provide increasing evidence for inter-TAD communications between neighboring TADs⁵⁰. Although TADs are likely a statistical model from population cells, there is clear evidence of clustering of CBS elements at TAD boundaries^13,15,48. On one hand, the outward-oriented CBS elements could be crucial for intra-TAD chromatin interactions. On the other hand, the outward-oriented CBS elements function as crucial insulators blocking aberrant chromatin interactions from neighboring TADs as we observed a sharp edge for increased chromatin interactions beyond the outward CBS element (Fig. 4d, e). In the case of clustered CTCF TAD boundary of Pcdhβγ, the extruding cohesin complexes from the downstream TAD cannot pass through the forward CBSf element. However, upon deletion of the outward CBSf element, these upward-extruding cohesin complexes pass through the entire TAD boundary and eventually anchored at variable promoters, leading to an increase of chromatin interactions between Pcdhβγ genes and CBSh region.

The structure of TADs and loops is closely related to gene expression. Their formation is thought to be resulted from CTCF-anchored active cohesin “loop extrusion”. Specifically, cohesin-extruded loops anchored between convergent forward-reverse CBS elements can enhance or stabilize enhancer-promoter spatial contacts, which may or may not activate gene expression^53,54,55,56. In addition, both the spatial and linear distances between distal enhancers and target promoters could be crucial for gene regulation^57,58,59. Our study on Pcdhγc3 regulation is consistent with this idea. Pcdhγc3, the most abundantly expressed cPcdh isoform in the mouse neocortex (Supplementary Fig. 2e), plays a crucial role in dendrite arborization³⁴. It is puzzling why HS5-1bL deletion leads to a significant increase of expression levels of only γc3, but not other members, of cPcdh genes, suggesting that it is a γc3-specific insulator (Figs. 1 i, 2e and Supplementary Fig. 4e). It has been shown that gene activation is not only dependent on promoter-associated CBS elements, but also on the density of enhancers of the genomic context^32,57. Consistently, only γc3, but not γc4 and γc5, is associated with two forward-oriented CBS elements (Fig. 2a) despite the fact that, compared with other members of Pcdhβγ clusters, all of the three Pcdhγc genes are located close to the downstream distal enhancers of HS18-20, which contain two reverse-oriented CBS elements (Fig. 1b). A “double clamp” interaction mechanism of the two forward-reverse CBS pairs, similar to the activation of members of the Pcdhα gene cluster²⁴, may be utilized to activate γc3. This explains why only γc3, but not γc4 and γc5, expression levels are enhanced upon deletion of the HS5-1bL (CBSbc) insulator.

HOXD13 showed strong interactions with two H3K27ac-marked regions of LNPK and GT2. We observed a significant decrease of chromatin interactions with LNPK and GT2 upon removal of outward-oriented CBS elements (Fig. 5d), explaining the decreased levels of HOXD13 gene expression. This suggests that the outward-oriented CBS boundary elements of the C-DOM may ensure proper interactions with LNPK and GT2 for precise control of HOXD13 expression patterns. Our data highlight a significant role of the outward-oriented CBS elements in providing optimal insulation for restraining aberrant inter-TAD interactions and maintain proper intra-TAD promoter-enhancer interactions in the Pcdh and HoxD clusters. We analyzed genome-wide CBS patterns at TAD boundaries and found that 85.7% have outward-oriented CBS elements (Supplementary Data 2, 3). However, whether it could be generalized to the entire genome awaits further studies.

We observed that H3K9me3 marks correlate with the monoallelic cPcdh gene expression and that H3K9me2 marks correlate with both monoallelic and biallelic cPcdh gene expression²⁶. The correlation between H3K9me3 enrichments and DNA methylation levels in both monoallelic and biallelic cPcdh genes may be coordinated by UHRF1⁶⁰. Interestingly, almost every member of the cPcdh genes is marked by both repressive (H3K9me3 and H3K9me2) and active (H3K36me3) marks (Fig. 3), suggestive of a facultative heterochromatic state related to monoallelic cPcdh gene expression. Whether the overlap of H3K9me3 and H3K36me3 is resulted from allelic differences of the epigenome and/or admixtures of distinct subpopulation cells awaits further exploration.

Methods

Cell culture

Human HEK293T cells were cultured in DMEM (Hyclone) supplemented with 10% (v/v) FBS (Gibco) and 10 U/ml penicillin-streptomycin (Gibco, 15140122). Human HEC-1-B cells were cultured in MEM (Hyclone) supplemented with 200 mM L-glutamine, Earle’s balanced salt solution (Hyclone, SH30024.01), 1 mM sodium pyruvate (Sigma, SLBP4879V), 10% fetal bovine serum, and 10 U/ml penicillin-streptomycin. All cells were incubated at 37 °C and 5% CO₂ in a humidified incubator, and passaged every 3 days.

CRISPR screening of CBS-deletion single-cell clones

CBS3 and CBS5 within the clustered CTCF TAD boundary of the HOXD locus were sequentially knocked out by two steps of CRISPR/Cas9-mediated DNA fragment editing (Supplementary Data 4)^9,61. In the first step, CBS3 or CBS5 was deleted in HEK293T cells to generate homozygous CBS3 or CBS5 knockout cell clones. In the second step, a CBS3 knockout cell clone was used to delete CBS5 and a CBS3 knockout cell clone was used to delete CBS3, to generate cell clones homozygous for CBS3 and CBS5 double knockout (ΔCBS3 + 5).

For each step of CRISPR editing, cells were grown to ~80% confluence in 6-well tissue culture plates and transfected with 1.5 μg of pcDNA3.1-Cas9 and 1.5 μg dual sgRNA expression plasmids per well by Lipofectamine 3000 (Invitrogen, L3000015). Two days after transfection, 2 μg/ml puromycin was added to the cell growth medium and maintained for 5 days to select the transfected cells. The transfected cells were suspended into single-cell solutions, diluted and plated into 96 well plates. Two weeks later, single-cell clones were picked and genotyped by PCR with specific primers (Supplementary Data 5). Positive clones were Sanger sequenced.

The sgRNA expression plasmids were constructed as previously described^9,61 with paired oligonucleotides (Supplementary Data 5). For each sgRNA, a forward oligonucleotide with the (ACCG-5’) overhang and a reverse complementary oligonucleotide with the (AAAC-5’) overhang were annealed and cloned into the Bsa I site of pGL3-U6-sgRNA-PGK-Puro vector for sgRNA transcription⁶². All sgRNA-expressed plasmids were confirmed by Sanger sequencing.

In vitro synthesis of Cas9 mRNA and sgRNAs for micro-injection

A T7 promoter-driven Cas9 expression plasmid⁶³ was linearized by Xba I, purified and in-vitro transcribed using the T7 Ultra Kit (ThermoFisher, AM1345). The transcribed Cas9 mRNA was treated with Turbo DNase to remove template DNA and added with poly-A tail. The sgRNA template was generated by PCR amplification with specific primers (Supplementary Data 5) to contain a T7 promoter, a 20 nt target sequence, and a scaffold region⁴⁹. The amplified product was purified and transcribed using the MEGAshortscript T7 Kit (ThermoFisher, AM1354). The transcribed sgRNA was treated with Turbo DNase. Both sgRNA and Cas9 mRNA were purified by the MEGAclear Kit (ThermoFisher, AM1908), eluted into elution buffer and stored at −80 °C before usage.

Generation of CBS-deletion mice

Mice were maintained in an SPF mouse facility at 23 °C, with a humidity between 40 and 60%, with a 12 h (7:00–19:00) light/12 h (19:00–7:00) dark cycle. All the mouse experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Shanghai Jiao Tong University (Protocol#: 1602029).

C57BL/6J mice were used as embryo donors and ICR mice were used as foster mothers. 6-week-old C57BL/6J female mice were super-ovulated and mated with the sterilized C57BL/6J male mice to produce enough embryos. Zygotes were collected from the oviducts of the super-ovulated C57BL/6J female mice, washed with 200 μg/ml hyaluronidase (Sigma, V900833) and incubated in the M2 medium (Sigma, M7167) in a 5% CO₂ incubator at 37 °C for 1 h. Viable embryos were then microinjected with a solution containing 100 ng/μl of Cas9 mRNA and 50 ng/μl each of two sgRNAs targeting each fragment and recovered in a 5% CO₂ incubator at 37 °C for 1 h. The injected live embryos were transplanted into the oviducts of the pseudo-pregnant ICR female mice under a stereoscopic microscope, with each ICR mouse receiving 25–30 embryos. Post-surgery mice were maintained at a 37 °C heating plate for 1 h for recovery before being transferred back for regular housing until the birth of the F0 mice. Mice exhibiting signs of dystocia at day 20 received C-section rescue.

The chimeric F0 mice were screened 10 days later for desired deletions by PCR genotyping with specific primers (Supplementary Data 5) and the chimeric mutant mice were confirmed by Sanger sequencing. The F0 mice of 2-month old with desired deletions were mated with wild-type C57BL/6 J mice to generate heterozygous F1 mice. The F1 mice were genotyped by PCR. The targeted F1 male and female mice were crossed to generate the F2 homozygous mice, which were genotyped (deletions of HS7L, CBSf, CBSfg, or CBSg were shown in Supplementary Figs. 4, 12, the other mouse lines were genotyped previously⁴⁹) and used for downstream experiments. The wild-type F2 littermates were used as controls.

ATAC-seq

ATAC-seq was performed following the omni-ATAC procedure⁶⁴ with some modifications. Briefly, the neocortex was microdissected and digested with 0.00625% collagenase followed by filtration through a 100 μm cell strainer to generate single-cell suspensions. The suspended cells were washed twice with PBS solution and then counted. 10⁵ cells were aliquoted and resuspended with 50 μl of the resuspension buffer (RSB) supplemented with 0.1% NP40, 0.1% Tween-20, and 0.01% digitonin followed by incubation on ice for 3 min for lysis. The nuclei were then washed with 1 ml of RSB supplemented with only 0.1% Tween-20 and resuspended in 50 µl of transposition mix, which includes 25 µl of 2× TD buffer, 3 μl transposase, 16 μl PBS, 0.5 μl 1% digitonin, 0.5 μl 10% Tween-20 and 5 μl H₂O, followed by incubation at 37 °C for 30 min in a shaker at 1000 rpm. Transposition products were immediately cleaned up with 1× Ampure XP Beads (Beckman, A63881), and the purified DNA was used for library construction by the TruePrep DNA Library Prep Kit V2 for Illumina (Vazyme Biotech, TD502). The ATAC-seq libraries were pooled and sequenced at the 2 × 150 bp mode using an Illumina NovaSeq 6000 platform.

RNA-seq

Total RNA was extracted from cultured cells or the microdissected mouse neocortex using the Trizol reagent (ThermoFisher, 15596026). For each sample, 1 μg of total RNA was used to enrich mRNA using Poly(A) Magnetic Isolation Module (NEB, E7600). The mRNA was purified twice, eluted into the fragmentation buffer with random primers for fragmentation at 94 °C for 15 min, and reverse-transcribed into the first strand cDNA. The second strand was then synthesized by the second-strand DNA polymerase. The generated double-stranded cDNA was purified with Ampure XP beads (Beckman, A63881), end repaired, 3’-adenylated, and 5’-phosphorylated. The Illumina sequencing adapters were then ligated to both ends of cDNA and excess adapters were removed using Ampure XP beads (Beckman). The purified cDNA was amplified by PCR with barcoding to generate RNA-seq libraries. Pooled multiplexed libraries were sequenced at the 2 × 150 bp mode on an Illumina NovaSeq 6000 platform. All experiments were performed with at least two biological replicates.

Fetal sex is determined based on the expression of Xist and Ddx3y. For expression mapping at different embryonic stages during cortical development, mouse neocortices were isolated from both males (Xist⁺, Ddx3y^-) and females (Xist^-, Ddx3y⁺). As we did not observe sex-specific differences in cPcdh expression at any developmental stages (Supplementary Fig. 2f, g), we did not discriminate between male and female mouse neocortices for all the following experiments.

ChIP-seq

Mouse neocortical tissues was microdissected and dissociated into single cells using 0.00625% collagenase. 2 × 10⁶ of mouse neocortical or cultured human cells were aliquoted and cross-linked using 1% formaldehyde in PBS containing 10% FBS for 10 min at room temperature followed by two washes with PBS containing proteinase inhibitors (Roche, 04693132001). Cells were then resuspended and lysed twice with 1 ml of prechilled lysis buffer containing proteinase inhibitors (Roche) for 10 min with slow rotations. For cortical cells, additional SDS was added to the lysis solution to adjust the final SDS concentration to 0.4% for better sonication. Cells were then sonicated at 25% power for a train of 15 s ON and 30 s OFF for 20 cycles with the Bioruptor system to break the genomic DNA into 0.1–10 kb fragments. After sonication, the SDS-free lysis buffer was added to cortical samples to dilute the SDS concentration to 0.1% to allow the binding of antibodies to proteins. After removal of the insoluble debris by centrifuging at 14,000 g for 10 min, the lysate was pre-cleaned with protein-A agarose beads (Millipore, 16–157) for 1 h with slow rotations to remove non-specific binding and then incubated with specific antibody (Supplementary Data 6) overnight with slow rotations at 4 °C. The antibody-protein-DNA complex was isolated by adding protein A-agarose beads and incubated for 4 h with slow rotations followed by sequential washes with the low-salt buffer, high-salt buffer, salt-free buffer, LiCl buffer, and TE buffer. The beads were eluted in the elution buffer (0.1 M NaHCO₃, 1% SDS) and de-cross-linked by proteinase K at 65 °C overnight. The DNA was finally purified and used for the construction of DNA library using the VAHTS Universal DNA Library Prep Kit for Illumina V3 (Vazyme Biotech, ND607). ChIP-seq libraries were sequenced at the 2 × 150 bp mode on an Illumina NovaSeq 6000 platform.

QHR-4C

For QHR-4C⁴⁹ experiments, mouse neocortices were microdissected and dissociated with 0.00625% collagenase to obtain single cells. Mouse neocortical or cultured human cells were cross-linked in 2% formaldehyde for 10 min at room temperature and quenched by adding excess pre-chilled glycine. The fixed cells were incubated on ice for 5 min, and permeabilized twice with the permeabilization buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 5 mM EDTA, 0.5% NP-40, 1% Triton X-100, and protease inhibitors), with slow rotations for 10 min at 4 °C. The permeabilized cells were then digested with Dpn II for 16 h at 37 °C while shaking at 900 rpm followed by incubation at 65 °C for 20 min to inactive Dpn II. Proximity ligation was then performed for 16 h at 16 °C by adding T4 DNA ligase in 1x T4 ligation buffer. The ligated product was de-cross-linked by proteinase K at 65 °C overnight, treated with RNase A at 37 °C for 45 min, and purified with phenol-chloroform followed by ethanol precipitation. DNA was sonicated to 0.2–1 kb fragments using the Bioruptor system at 33% power for a train of 15 s ON and 30 s OFF for four cycles.

The anchor fragments were linear-amplified using a specific 5’ biotin-tagged primer (Supplementary Data 5) with 120 cycles (95 °C 20 s for denaturing, 60 °C 20 s for annealing, and 72 °C 90 s for elongation). The amplified products were finally denatured at 95 °C for 5 min and then immediately chilled on ice to prevent reannealing. The biotin-tagged single-stranded DNA (ssDNA) was pulled down with Streptavidin Magnetic beads (ThermoFisher, 65001) and ligated with adapters which were generated by annealing two partially complementary single-stranded oligonucleotides (Adapter-U and Adapter-L, Supplementary Data 5) by T4 DNA ligase in 1x T4 ligation buffer containing PEG8000 at 16 °C for 12 h. The on-bead ligation products were washed twice with the B/W buffer (5 mM Tris-HCl pH 7.5, 1 M NaCl, and 0.5 mM EDTA) to remove excess adapters and used as the template for PCR amplification of the 4C libraries with barcoded anchor-specific P5-forward primers and indexed P7-reverse primers (Supplementary Data 5). Multiplexed libraries from the same anchor with different combinations of barcodes and indexes were pooled for purification with a PCR purification kit (Qiagen, 11732668001). 4C libraries were sequenced at the 2 × 150 bp mode on an Illumina NovaSeq 6000 platform.

In situ Hi-C

For in situ Hi-C⁸ experiments, 5 × 10⁶ cells were cross-linked with formaldehyde and incubated in a lysis buffer (10 mM Tris-HCl pH8.0, 10 mM NaCl, 0.2% NP-40) to obtain nuclei. Nuclei were permeabilized with SDS, quenched by adding Triton X-100, and digested with Mbo II(NEB, R0147M) overnight at 37 °C, followed by heat inactivation at 62 °C. The ends of restriction fragments were filled in with biotin-14-dATP (Thermo, 19524016), dCTP, dTTP, and dGTP using the Klenow fragment of DNA polymerase I (NEB, M0210L), and ligated using T4 DNA ligase (NEB, M0202S). After ligation, the nuclei were de-cross-linked overnight at 68 °C. The ligated DNA was precipitated with ethanol and fragmented by sonication. 300-500 bp DNA fragments were selected using AMPure XP beads (Beckman, A63881). The biotinylated DNA was pulled down with M-280 Streptavidin beads (Thermo, 11206D), washed, and used for library construction on bead. The DNA was end-repaired, tailed with dA using Klenow exo^- (NEB, M0212S), and ligated with Illumina U-type adapters, followed by adapter linearization with a USER enzyme (NEB, M5505S). The obtained DNA was used as templates for PCR amplification with 10-12 cycles. PCR products were purified with AMPure XP beads (Beckman, A63881) and sequenced at the 2 × 150 bp mode on an Illumina NovaSeq 6000 platform.

High-throughput sequencing data analysis

RNA-seq reads were aligned against the mouse (GRCm38) or human (GRCh38) genome using Hisat2⁶⁵ (version 2.0.4) to generate the sequence alignment map (SAM) files. SAM files were then converted into the binary versions (BAM files) using Samtools (version 1.15.1). BAM files were used as input for Cufflinks⁶⁶ (version 2.1.1) to calculate FPKM values.

ChIP-seq and ATAC-seq reads were aligned to the mouse (mm9) or human (hg19) genome using Bowtie2 (version 2.3.5) to generate SAM files, which were converted into BAM files using Samtools (version 1.15.1). BAM files were indexed and converted into BedGraph files.

For 4 C data, P7 reads were initially demultiplexed based on the unique barcode-index combinations. The anchor primer sequences were trimmed and PCR duplicate reads were removed using FastUniq (version 1.1) of the UNIX system. The obtained unique reads were aligned against the mouse (mm9) or human (hg19) genome using Bowtie2 (version 2.3.5) to generate SAM files, which were converted into BAM files. The BAM files were used as input for r3Cseq (version 1.20) in the R package (3.3.3) to calculate the reads per million (RPM) values.

Hi-C reads were pre-processed with HiC-Pro (v3.0.0)⁶⁷, including reads alignment to the human hg19 reference genome, interaction matrix generation, and ICE normalization. 10 kb ICE normalized interaction matrix was generated. TADs and TAD boundaries were called at a 10-kb resolution using the directionality index (DI) method⁵¹. Insulation scores were calculated at a 10-kb resolution as previously described⁶⁸.

For genome-wide CBS patterns at TAD boundaries, we analyzed in situ HiC data. CBS position and orientation at TAD boundaries were determined by FIMO. We refer to a CTCF motif as “forward” (>) when it is present on the forward strand of the chromosome and ‘reverse’ (<) when it is on the reverse strand. We defined TAD boundary regions as ±50 kb from the center of each boundary. TAD boundary regions were intersected with CTCF motifs using “bedtools intersect” to identify CBS numbers and orientations for each TAD boundary.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The raw high-throughput sequencing data and processed data generated in this study have been deposited in the GEO database under accession code GSE210817. Source data are provided with this paper.

References

Mirny, L. & Dekker, J. Mechanisms of chromosome folding and nuclear organization: their interplay and open questions. Cold Spring Harb. Perspect. Biol. 14, 1–20 (2022).
Article Google Scholar
Phillips, J. E. & Corces, V. G. CTCF: master weaver of the genome. Cell 137, 1194–1211 (2009).
Article PubMed PubMed Central Google Scholar
Hansen, A. S., Cattoglio, C., Darzacq, X. & Tjian, R. Recent evidence that TADs and chromatin loops are dynamic structures. Nucleus 9, 20–32 (2018).
Article CAS PubMed Google Scholar
Wu, Q., Liu, P. & Wang, L. Many facades of CTCF unified by its coding for three-dimensional genome architecture. J. Genet. Genom. 47, 407–424 (2020).
Article CAS Google Scholar
Nichols, M. H. & Corces, V. G. A CTCF code for 3D genome architecture. Cell 162, 703–705 (2015).
Article CAS PubMed PubMed Central Google Scholar
Sanborn, A. L. et al. Chromatin extrusion explains key features of loop and domain formation in wild-type and engineered genomes. Proc. Natl. Acad. Sci. USA 112, E6456–E6465 (2015).
Article CAS PubMed PubMed Central Google Scholar
Fudenberg, G. et al. Formation of chromosomal domains by loop extrusion. Cell Rep. 15, 2038–2049 (2016).
Article CAS PubMed PubMed Central Google Scholar
Rao, S. S. et al. A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping. Cell 159, 1665–1680 (2014).
Article CAS PubMed PubMed Central Google Scholar
Guo, Y. et al. CRISPR Inversion of CTCF sites alters genome topology and enhancer/promoter function. Cell 162, 900–910 (2015).
Article CAS PubMed PubMed Central Google Scholar
Lupianez, D. G. et al. Disruptions of topological chromatin domains cause pathogenic rewiring of gene-enhancer interactions. Cell 161, 1012–1025 (2015).
Article CAS PubMed PubMed Central Google Scholar
Flavahan, W. A. et al. Insulator dysfunction and oncogene activation in IDH mutant gliomas. Nature 529, 110–114 (2016).
Article ADS CAS PubMed Google Scholar
Kloetgen, A. et al. Three-dimensional chromatin landscapes in T cell acute lymphoblastic leukemia. Nat. Genet. 52, 388–400 (2020).
Article CAS PubMed PubMed Central Google Scholar
Kentepozidou, E. et al. Clustered CTCF binding is an evolutionary mechanism to maintain topologically associating domains. Genome Biol. 21, 5 (2020).
Article PubMed PubMed Central Google Scholar
Nanni, L., Ceri, S. & Logie, C. Spatial patterns of CTCF sites define the anatomy of TADs and their boundaries. Genome Biol. 21, 197 (2020).
Article CAS PubMed PubMed Central Google Scholar
McArthur, E. & Capra, J. A. Topologically associating domain boundaries that are stable across diverse cell types are evolutionarily constrained and enriched for heritability. Am. J. Hum. Genet. 108, 269–283 (2021).
Article CAS PubMed PubMed Central Google Scholar
Gong, Y. et al. Stratification of TAD boundaries reveals preferential insulation of super-enhancers by strong boundaries. Nat. Commun. 9, 542 (2018).
Article ADS PubMed PubMed Central Google Scholar
Whyte, W. A. et al. Master transcription factors and mediator establish super-enhancers at key cell identity genes. Cell 153, 307–319 (2013).
Article CAS PubMed PubMed Central Google Scholar
Anania, C. et al. In vivo dissection of a clustered-CTCF domain boundary reveals developmental principles of regulatory insulation. Nat. Genet. 54, 1026–1036 (2022).
Article CAS PubMed PubMed Central Google Scholar
Despang, A. et al. Functional dissection of the Sox9-Kcnj2 locus identifies nonessential and instructive roles of TAD architecture. Nat. Genet. 51, 1263–1271 (2019).
Article CAS PubMed Google Scholar
Huang, H. et al. CTCF mediates dosage- and sequence-context-dependent transcriptional insulation by forming local chromatin domains. Nat. Genet. 53, 1064–1074 (2021).
Article CAS PubMed PubMed Central Google Scholar
Taylor, T. et al. Transcriptional regulation and chromatin architecture maintenance are decoupled functions at the Sox2 locus. Genes Dev. 36, 699–717 (2022).
Article CAS PubMed PubMed Central Google Scholar
Chakraborty, S. et al. Enhancer-promoter interactions can bypass CTCF-mediated boundaries and contribute to phenotypic robustness. Nat. Genet. 55, 280–290 (2023).
Article CAS PubMed Google Scholar
Wu, Q. et al. Comparative DNA sequence analysis of mouse and human protocadherin gene clusters. Genome Res. 11, 389–404 (2001).
Article CAS PubMed PubMed Central Google Scholar
Guo, Y. et al. CTCF/cohesin-mediated DNA looping is required for protocadherin alpha promoter choice. Proc. Natl. Acad. Sci. USA 109, 21081–21086 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Allahyar, A. et al. Enhancer hubs and loop collisions identified from single-allele topologies. Nat. Genet. 50, 1151–1160 (2018).
Article CAS PubMed Google Scholar
Esumi, S. et al. Monoallelic yet combinatorial expression of variable exons of the protocadherin-alpha gene cluster in single neurons. Nat. Genet. 37, 171–176 (2005).
Article CAS PubMed Google Scholar
Lv, X. et al. Patterned cPCDH expression regulates the fine organization of the neocortex. Nature 612, 503–511 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Andrey, G. et al. A switch between topological domains underlies HoxD genes collinearity in mouse limbs. Science 340, 1234167 (2013).
Article PubMed Google Scholar
Amandio, A. R. et al. Sequential in cis mutagenesis in vivo reveals various functions for CTCF sites at the mouse HoxD cluster. Genes Dev. 35, 1490–1509 (2021).
Article CAS PubMed PubMed Central Google Scholar
Jiang, Y. et al. The methyltransferase SETDB1 regulates a large neuron-specific topological chromatin domain. Nat. Genet. 49, 1239–1250 (2017).
Article CAS PubMed PubMed Central Google Scholar
Ribich, S., Tasic, B. & Maniatis, T. Identification of long-range regulatory elements in the protocadherin-alpha gene cluster. Proc. Natl. Acad. Sci. USA 103, 19719–19724 (2006).
Article ADS CAS PubMed PubMed Central Google Scholar
Kehayova, P., Monahan, K., Chen, W. & Maniatis, T. Regulatory elements required for the activation and repression of the protocadherin-alpha gene cluster. Proc. Natl. Acad. Sci. USA 108, 17195–17200 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Yokota, S. et al. Identification of the cluster control region for the protocadherin-beta genes located beyond the protocadherin-gamma cluster. J. Biol. Chem. 286, 31885–31895 (2011).
Article CAS PubMed PubMed Central Google Scholar
Steffen, D. M. et al. A unique role for protocadherin gammaC3 in promoting dendrite arborization through an Axin1-eependent mechanism. J. Neurosci. 43, 918–935 (2023).
Article CAS PubMed PubMed Central Google Scholar
Schreiner, D. & Weiner, J. A. Combinatorial homophilic interaction between gamma-protocadherin multimers greatly expands the molecular diversity of cell adhesion. Proc. Natl. Acad. Sci. USA 107, 14893–14898 (2010).
Article ADS CAS PubMed PubMed Central Google Scholar
Lefebvre, J. L., Kostadinov, D., Chen, W. V., Maniatis, T. & Sanes, J. R. Protocadherins mediate dendritic self-avoidance in the mammalian nervous system. Nature 488, 517–521 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Thu, C. A. et al. Single-cell identity generated by combinatorial homophilic interactions between alpha, beta, and gamma protocadherins. Cell 158, 1045–1059 (2014).
Article CAS PubMed PubMed Central Google Scholar
Toyoda, S. et al. Developmental epigenetic modification regulates stochastic expression of clustered protocadherin genes, generating single neuron diversity. Neuron 82, 94–108 (2014).
Article CAS PubMed Google Scholar
Mountoufaris, G. et al. Multicluster Pcdh diversity is required for mouse olfactory neural circuit assembly. Science 356, 411–414 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Ing-Esteves, S. et al. Combinatorial effects of alpha- and gamma-protocadherins on neuronal survival and dendritic self-avoidance. J. Neurosci. 38, 2713–2729 (2018).
Article CAS PubMed PubMed Central Google Scholar
Brasch, J. et al. Visualization of clustered protocadherin neuronal self-recognition complexes. Nature 569, 280–283 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Canzio, D. et al. Antisense lncRNA transcription mediates DNA demethylation to drive stochastic protocadherin alpha promoter choice. Cell 177, 639–653.e15 (2019).
Article CAS PubMed PubMed Central Google Scholar
Almenar-Queralt, A. et al. Chromatin establishes an immature version of neuronal protocadherin selection during the naive-to-primed conversion of pluripotent stem cells. Nat. Genet. 51, 1691–1701 (2019).
Article CAS PubMed PubMed Central Google Scholar
Rodriguez-Carballo, E. et al. The HoxD cluster is a dynamic and resilient TAD boundary controlling the segregation of antagonistic regulatory landscapes. Genes Dev. 31, 2264–2281 (2017).
Article CAS PubMed PubMed Central Google Scholar
Narendra, V., Bulajic, M., Dekker, J., Mazzoni, E. O. & Reinberg, D. CTCF-mediated topological boundaries during development foster appropriate gene regulation. Genes Dev. 30, 2657–2662 (2016).
Article CAS PubMed PubMed Central Google Scholar
Barutcu, A. R., Maass, P. G., Lewandowski, J. P., Weiner, C. L. & Rinn, J. L. A TAD boundary is preserved upon deletion of the CTCF-rich Firre locus. Nat. Commun. 9, 1444 (2018).
Article ADS PubMed PubMed Central Google Scholar
Chang, L. H. et al. Multi-feature clustering of CTCF binding creates robustness for loop extrusion blocking and topologically associating domain boundaries. Nat Commun. 14, 5615 (2023).
Okhovat, M. et al. TAD Evolutionary and functional characterization reveals diversity in mammalian TAD boundary properties and function Preprint at https://doi.org/10.1101/2023.03.07.531534 (2023).
Jia, Z. et al. Tandem CTCF sites function as insulators to balance spatial chromatin contacts and topological enhancer-promoter selection. Genome Biol. 21, 75 (2020).
Article CAS PubMed PubMed Central Google Scholar
Galupa, R. et al. Inversion of a topological domain leads to restricted changes in its gene expression and affects interdomain communication. Development 149, dev200568 (2022).
Article CAS PubMed PubMed Central Google Scholar
Dixon, J. R. et al. Topological domains in mammalian genomes identified by analysis of chromatin interactions. Nature 485, 376–380 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Nora, E. P. et al. Spatial partitioning of the regulatory landscape of the X-inactivation centre. Nature 485, 381–385 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Alexander, J. M. et al. Live-cell imaging reveals enhancer-dependent Sox2 transcription in the absence of enhancer proximity. Elife 8, e41769 (2019).
Article PubMed PubMed Central Google Scholar
Benabdallah, N. S. et al. Decreased enhancer-promoter proximity accompanying enhancer activation. Mol. Cell 76, 473–484 e7 (2019).
Article CAS PubMed PubMed Central Google Scholar
Fulco, C. P. et al. Activity-by-contact model of enhancer-promoter regulation from thousands of CRISPR perturbations. Nat. Genet. 51, 1664–1669 (2019).
Article CAS PubMed PubMed Central Google Scholar
Hsieh, T. S. et al. Enhancer-promoter interactions and transcription are largely maintained upon acute loss of CTCF, cohesin, WAPL or YY1. Nat. Genet. 54, 1919–1932 (2022).
Article CAS PubMed PubMed Central Google Scholar
Kubo, N. et al. Promoter-proximal CTCF binding promotes distal enhancer-dependent gene activation. Nat. Struct. Mol. Biol. 28, 152–161 (2021).
Article CAS PubMed PubMed Central Google Scholar
Rinzema, N. J. et al. Building regulatory landscapes reveals that an enhancer can recruit cohesin to create contact domains, engage CTCF sites and activate distant genes. Nat. Struct. Mol. Biol. 29, 563–574 (2022).
Article CAS PubMed PubMed Central Google Scholar
Zuin, J. et al. Nonlinear control of transcription through enhancer-promoter interactions. Nature 604, 571–577 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Butz, S. et al. DNA sequence and chromatin modifiers cooperate to confer epigenetic bistability at imprinting control regions. Nat. Genet. 54, 1702–1710 (2022).
Article CAS PubMed PubMed Central Google Scholar
Li, J. et al. Efficient inversions and duplications of mammalian regulatory DNA elements and gene clusters by CRISPR/Cas9. J. Mol. Cell Biol. 7, 284–298 (2015).
Article CAS PubMed PubMed Central Google Scholar
Shen, B. et al. Efficient genome modification by CRISPR-Cas9 nickase with minimal off-target effects. Nat. Methods 11, 399–402 (2014).
Article CAS PubMed Google Scholar
Chang, N. et al. Genome editing with RNA-guided Cas9 nuclease in zebrafish embryos. Cell Res. 23, 465–472 (2013).
Article CAS PubMed PubMed Central Google Scholar
Corces, M. R. et al. An improved ATAC-seq protocol reduces background and enables interrogation of frozen tissues. Nat. Methods 14, 959–962 (2017).
Article CAS PubMed PubMed Central Google Scholar
Kim, D., Paggi, J. M., Park, C., Bennett, C. & Salzberg, S. L. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat. Biotechnol. 37, 907–915 (2019).
Article CAS PubMed PubMed Central Google Scholar
Trapnell, C. et al. Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat. Biotechnol. 28, 511–515 (2010).
Article CAS PubMed PubMed Central Google Scholar
Servant, N. et al. HiC-Pro: an optimized and flexible pipeline for Hi-C data processing. Genome Biol. 16, 259 (2015).
Article PubMed PubMed Central Google Scholar
Crane, E. et al. Condensin-driven remodelling of X chromosome topology during dosage compensation. Nature 523, 240–244 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We are grateful for technical supports from Jingwei Li, Leyang Wang, and Mo Zhang. This work was supported by grants from the National Natural Science Foundation of China (32330016), the National Key R&D Program of China (2022YFC3400200), and the Science and Technology Commission of Shanghai Municipality (21DZ2210200).

Author information

These authors contributed equally: Xiao Ge, Haiyan Huang.

Authors and Affiliations

Center for Comparative Biomedicine, Ministry of Education Key Laboratory of Systems Biomedicine, State Key Laboratory of Systems Medicine for Cancer, Joint International Research Laboratory of Metabolic and Developmental Sciences, School of Life Sciences and Biotechnology, Institute of Systems Biomedicine, Shanghai Jiao Tong University, Shanghai, 200240, China
Xiao Ge, Haiyan Huang, Keqi Han & Qiang Wu
WLA Laboratories, Shanghai, 201203, China
Xiao Ge, Haiyan Huang, Keqi Han & Qiang Wu
Laboratory Animal Center, Instrumental Analysis Center, Shanghai Jiao Tong University, Shanghai, 200240, China
Wangjie Xu & Zhaoxia Wang

Authors

Xiao Ge
View author publications
You can also search for this author in PubMed Google Scholar
Haiyan Huang
View author publications
You can also search for this author in PubMed Google Scholar
Keqi Han
View author publications
You can also search for this author in PubMed Google Scholar
Wangjie Xu
View author publications
You can also search for this author in PubMed Google Scholar
Zhaoxia Wang
View author publications
You can also search for this author in PubMed Google Scholar
Qiang Wu
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

Q.W. conceived the research. X.G. and K.H. performed experiments. H.H. and X.G. analyzed data. Z.W. and W.X. supervised the mouse experiments. X.G., H.H., and Q.W. wrote the manuscript with inputs from all authors.

Corresponding author

Correspondence to Qiang Wu.

Ethics declarations

Competing interests

The authors declare no competing interests

Peer review

Peer review information

Nature Communications thanks Hongjie Yao and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Data 1

Supplementary Data 2

Supplementary Data 3

Supplementary Data 4

Supplementary Data 5

Supplementary Data 6

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as 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 images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Ge, X., Huang, H., Han, K. et al. Outward-oriented sites within clustered CTCF boundaries are key for intra-TAD chromatin interactions and gene regulation. Nat Commun 14, 8101 (2023). https://doi.org/10.1038/s41467-023-43849-0

Download citation

Received: 09 May 2023
Accepted: 21 November 2023
Published: 07 December 2023
DOI: https://doi.org/10.1038/s41467-023-43849-0

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.