Synergy between regulatory elements can render cohesin dispensable for distal enhancer function

Reliance on cohesin can arise at much shorter genomic separation than previously appreciated from synthetic loci or artificial enhancer activation systems.

“Our observations also clarify that reliance on cohesin can arise at much shorter genomic separation than previously appreciated from synthetic loci or artificial enhancer activation systems (19, 20).”

• cites [19] Rinzema NJ, Sofiadis K, Tjalsma SJD, Verstegen MJAM, Oz Y, Valdes-Quezada C, Felder A-K, Filipovska T, van der Elst S, de Andrade dos Ramos Z, Han R, Krijger PHL, de Laat W, 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). 35710842 PMC9205769 sourcePartly supported

  The Introduction and the section on cohesin-dependent gene activation (pages not provided) report mild cohesin reliance at 47 kb, strong reliance at 100 kb or more, and none at 11 kb or less. This supports the possibility of reliance at an intermediate separation, but the paper does not establish that this is much shorter than previously appreciated from synthetic loci or artificial enhancer activation systems. (LLM, conf=0.83)

  Experimental — accuracy not yet measured
• cites [20] Kane L, Williamson I, Flyamer IM, Kumar Y, Hill RE, Lettice LA, Bickmore WA, Cohesin is required for long-range enhancer action at the Shh locus. Nat. Struct. Mol. Biol 29, 891– 897 (2022). 36097291 PMC7613721 sourcePartly supported

  The Discussion reports that cohesin is required for Shh activation from enhancers more than 400 kb away, but dispensable for an enhancer 100 kb away; the Introduction identifies the study as using synthetic transcriptional activators. The paper therefore supports an example of cohesin reliance at a specified separation, but does not establish the claim’s comparison that this is much shorter than previously appreciated from synthetic loci or artificial activation systems. (LLM, conf=0.83)

  Experimental — accuracy not yet measured

Loss of long-range interactions in NIPBL-depleted ESCs stems from a lower rate of cohesin loop extrusion and shorter DNA loops.

“loss of the long-range interactions that normally give rise to TADs stems from a lower rate of cohesin loop extrusion and shorter DNA loops (26).”

• cites [26] Shah R, Tortora MMC, Louafi N, Rahmaninejad H, Hansen KL, Anderson EC, Wen D, Giorgetti L, Fudenberg G, Nora EP, Dosage sensitivity of the loop extrusion rate confers tunability to genome folding while creating vulnerability to genetic disruption. bioRxiv [Preprint] (2025). 10.1101/2025.08.14.667581. sourceSupported

These findings explain how strong CTCF insulators can be bypassed at the Sox2 locus.

“Moreover, they explain how strong CTCF insulators can be bypassed at the Sox2 locus (31, 40, 41).”

• cites [31] Huang H, Zhu Q, Jussila A, Han Y, Bintu B, Kern C, Conte M, Zhang Y, Bianco S, Chiariello AM, Yu M, Hu R, Tastemel M, Juric I, Hu M, Nicodemi M, Zhuang X, Ren B, CTCF mediates dosage- and sequence-context-dependent transcriptional insulation by forming local chromatin domains. Nat. Genet 53, 1064– 1074 (2021). 34002095 PMC8853952 sourceSupported
• cites [40] Chakraborty S, Kopitchinski N, Zuo Z, Eraso A, Awasthi P, Chari R, Mitra A, Tobias IC, Moorthy SD, Dale RK, Mitchell JA, Petros TJ, Rocha PP, Enhancer-promoter interactions can bypass CTCF-mediated boundaries and contribute to phenotypic robustness. Nat. Genet 55, 280– 290 (2023). 36717694 PMC10758292 sourceSupported
• cites [41] Taylor T, Sikorska N, Shchuka VM, Chahar S, Ji C, Macpherson NN, Moorthy SD, de Kort MAC, Mullany S, Khader N, Gillespie ZE, Langroudi L, Tobias IC, Lenstra TL, Mitchell JA, Sexton T, Transcriptional regulation and chromatin architecture maintenance are decoupled functions at the Sox2 locus. Genes Dev. 36, 699– 717 (2022). 35710138 PMC9296009 sourceSupported

Depleting RAD21 or CTCF also down-regulated Car2.

“Depleting RAD21 or CCCTC-binding factor (CTCF) also down-regulated Car2 (fig. S12C) (7, 30)”

• cites [7] Liu NQ, Maresca M, van den Brand T, Braccioli L, Schijns MMGA, Teunissen H, Bruneau BG, Nora EP, de Wit E, WAPL maintains a cohesin loading cycle to preserve cell-type-specific distal gene regulation. Nat. Genet 53, 100– 109 (2021). 33318687 PMC7610352 sourceCould not verify

  The Results section on cohesin and pluripotency-specific gene expression reports that RAD21 depletion decreased expression of Sik1, Hsf2bp and Rrp1b, but does not mention Car2. A search for Car2 found no passage in our copy, and the passages on CTCF do not report Car2 expression after CTCF depletion. Page numbers are unavailable in our copy. (LLM, conf=0.76)

  Experimental — accuracy not yet measured
• cites [30] Nora EP, Goloborodko A, Valton A-L, Gibcus JH, Uebersohn A, Abdennur N, Dekker J, Mirny LA, Bruneau BG, Targeted Degradation of CTCF Decouples Local Insulation of Chromosome Domains from Genomic Compartmentalization. Cell 169, 930– 944.e22 (2017). 28525758 PMC5538188 sourcePartly supported

  In Results, ‘CTCF and transcriptional regulation’ (page unavailable), the paper reports that CTCF depletion down-regulated a subset of genes in mouse embryonic stem cells. Our copy does not identify Car2 among them, so it does not establish the gene-specific claim. The paper also does not report RAD21 depletion; with two citations attached to the sentence, that part may be attributed to the other source. (LLM, conf=0.78)

  Experimental — accuracy not yet measured

These findings clarify previous reports suggesting that 3D genome folding may be largely decoupled from long-range regulation, which were based on studying Sox2.

“These findings clarify previous reports suggesting that 3D genome folding may be largely decoupled from long-range regulation, which were based on studying Sox2 (39, 41).”

• cites [39] Alexander JM, Guan J, Li B, Maliskova L, Song M, Shen Y, Huang B, Lomvardas S, Weiner OD, Live-cell imaging reveals enhancer-dependent Sox2 transcription in the absence of enhancer proximity. eLife 8, e41769 (2019). 31124784 PMC6534382 sourceSupported
• cites [41] Taylor T, Sikorska N, Shchuka VM, Chahar S, Ji C, Macpherson NN, Moorthy SD, de Kort MAC, Mullany S, Khader N, Gillespie ZE, Langroudi L, Tobias IC, Lenstra TL, Mitchell JA, Sexton T, Transcriptional regulation and chromatin architecture maintenance are decoupled functions at the Sox2 locus. Genes Dev. 36, 699– 717 (2022). 35710138 PMC9296009 sourceSupported

GSEA underscored disruption of developmentally relevant pathways without highlighting cell cycle terms in NIPBL-depleted ESCs, in contrast to cohesin-depleted ESCs.

“Gene set enrichment analysis (GSEA) underscored the disruption of developmentally relevant pathways without highlighting cell cycle terms, in contrast to cohesin-depleted ESCs (fig. S5) (7).”

• cites [7] Liu NQ, Maresca M, van den Brand T, Braccioli L, Schijns MMGA, Teunissen H, Bruneau BG, Nora EP, de Wit E, WAPL maintains a cohesin loading cycle to preserve cell-type-specific distal gene regulation. Nat. Genet 53, 100– 109 (2021). 33318687 PMC7610352 sourcePartly supported

  The Results section, “WAPL is required for maintaining the pluripotent transcriptional state” (page unavailable), reports GSEA enrichment for developmental processes and no strong enrichment for cell-cycle categories, but these findings concern WAPL-depleted, not NIPBL-depleted, mouse ESCs. The Results section, “Cohesin is required for maintaining pluripotency-specific gene expression” (page unavailable), reports RAD21 depletion and similar expression changes to WAPL depletion; it does not establish the claimed GSEA contrast with NIPBL-depleted ESCs. (LLM, conf=0.83)

  Experimental — accuracy not yet measured

The paper's observations expand the list of recently discovered long-range regulatory mechanisms that work by potentiating enhancer activity, such as facilitation and range extension.

“Instead, our observations expand the list of recently discovered long-range regulatory mechanisms that work by potentiating enhancer activity, such as facilitation (60) and range extension (61).”

• cites [60] Blayney JW, Francis H, Rampasekova A, Camellato B, Mitchell L, Stolper R, Cornell L, Babbs C, Boeke JD, Higgs DR, Kassouf M, Super-enhancers include classical enhancers and facilitators to fully activate gene expression. Cell 186, 5826– 5839.e18 (2023). 38101409 PMC10858684 sourceSupported
• cites [61] Bower G, Hollingsworth EW, Jacinto SH, Alcantara JA, Clock B, Cao K, Liu M, Dziulko A, Alcaina-Caro A, Xu Q, Skowronska-Krawczyk D, Lopez-Rios J, Dickel DE, Bardet AF, Pennacchio LA, Visel A, Kvon EZ, Range extender mediates long-distance enhancer activity. Nature 643, 830– 838 (2025). 40604280 PMC12267059 sourceSupported

Deleting the SCR reduced Sox2 expression to ~10%.

“deleting the SCR reduced Sox2 expression to ~10% (Fig. 4F).”

Deleting ~80 kb of intervening sequence between Car2 and its enhancers boosted expression ~twofold.

“deleting ~80 kb of intervening sequence between Car2 and its enhancers boosted expression ~twofold (fig. S15B)”

When driven by proximal SRR2 alone, Car2 was entirely insensitive to NIPBL depletion.

“When driven by proximal SRR2 alone, Car2 was entirely insensitive to NIPBL depletion, as expected given the short genomic separation (Fig. 6A and fig. S33A).”

NIPBL depletion only down-regulated Sox2 expression by ~25%, even after 4 days.

“NIPBL depletion only down-regulated Sox2 expression by ~25%, even after 4 days (Fig. 4B and fig. S19B).”

Inverting the Car2 transcription unit on this allele, bringing the promoter only 5 kb away from SRR2, was sufficient to elicit protection against the loss of loop extrusion.

“However, simply inverting the Car2 transcription unit on this allele, which brings the promoter only 5 kb away from SRR2, was sufficient to elicit protection against the loss of loop extrusion (Fig. 6B and fig. S33B).”

As SRR2 was moved further downstream of the TSS (17 and 41 kb), its ability to buffer against disrupting loop extrusion decreased.

“However, as SRR2 was moved further downstream of the TSS (17 and 41 kb), its ability to buffer against disrupting loop extrusion decreased (Fig. 5F and fig. S32, C and D).”

The distance cutoff at which loop extrusion becomes necessary to support distal enhancer function differs: >100 kb for Sox2 and ~20 kb for Car2.

“Another distinction between the Sox2 and Car2 loci is the different distance cutoff at which loop extrusion becomes necessary to support distal enhancer function of the SCR: >100 kb for Sox2 and ~20 kb for Car2 (Fig. 4D and fig. S24).”

When the SCR was located within 17 kb of the Sox2 TSS, expression remained entirely insensitive to NIPBL depletion even after removing SRR2.

“When the SCR was located within 17 kb of the Sox2 TSS, expression remained entirely insensitive to depleting NIPBL, even after removing SRR2 (Fig. 5B and fig. S29A).”

Even with the SCR at 265 or 365 kb and prolonged depletion, Sox2 expression did not drop more than ~50 to 60% after 3 days.

“Even with the SCR at these distal positions and prolonged depletion, Sox2 expression did not drop more than ~50 to 60% after 3 days.”

RNA-seq in NIPBL-FKBP ESCs revealed hundreds of dysregulated genes within 24 hours of depletion.

“RNA-seq in NIPBL-FKBP ESCs revealed hundreds of dysregulated genes within 24 hours of depletion (Fig. 1D).”

Deleting SRR2 homozygously in cells without any other alteration at the Sox2 locus caused cells to differentiate upon NIPBL depletion.

“Notably, deleting SRR2 homozygously in cells without any other alteration at the Sox2 locus (no reporters, SCR at 100 kb) caused cells to differentiate upon NIPBL depletion (Fig. 4G)”

Car2 expression is much stronger when either enhancer is positioned close to the promoter.

“We observed much stronger Car2 expression when either enhancer is positioned close to the promoter (Fig. 3G and fig. S15A).”

Replacing the Car2 promoter with the Sox2 promoter did not affect the dependence on loop extrusion.

“Replacing the Car2 promoter with the Sox2 promoter did not affect the dependence on loop extrusion (fig. S21)”

Car2 was not significantly down-regulated upon depleting NIPBL in NPCs, where the CE1 and CE2 enhancers are inactive.

“Car2 was not significantly down-regulated upon depleting NIPBL in NPCs, where the CE1 and CE2 enhancers are inactive (fig. S18).”

Depleting NIPBL from NPCs also dysregulated hundreds of genes within 24 hours.

“Depleting NIPBL from neural progenitor cells (NPCs) also dysregulated hundreds of genes within 24 hours (Fig. 2A and B and fig. S8)”

Depleting NIPBL for 24 hours decreased GFP and RFP expression from tagged Car2 alleles by nearly 60% after 24 hours and 80% after 3 days.

“Depleting NIPBL for 24 hours decreased GFP and RFP expression by nearly 60% after 24 hours and 80% after 3 days”

Sox2 expression was lower when the SCR was replaced with Car2 CE2.

“Sox2 expression was lower when we replaced the SCR with the Car2 CE2 (Fig. 4F and fig. S23A)”

Hi-C and 4C interactions between the Car2 TSS and its distal H3K27ac peaks are reduced upon NIPBL depletion.

“Hi-C and circularized chromosome conformation capture (4C) interactions between the Car2 TSS and these H3K27ac peaks are reduced upon NIPBL depletion (fig. S12, D and E, and fig. S13).”

Unlike at Car2, basal Sox2 expression does not increase drastically when the enhancer is at proximal positions.

“Yet, unlike what we observed at Car2, basal Sox2 expression does not increase drastically when the enhancer is at these proximal positions.”

Removing the SRR2 element rendered Sox2 expression highly reliant on NIPBL.

“removing the SRR2 element rendered Sox2 expression highly reliant on NIPBL (Fig. 4F and fig. S26A).”

Upon NIPBL or RAD21 depletion, the overall interaction profiles remained largely identical with or without the SRR2 deletion.

“Upon NIPBL or RAD21 depletion, the overall interaction profiles remained largely identical with or without the SRR2 deletion (Fig. 4I and fig. S28).”

SRR2 was required to maintain Sox2 expression both at baseline and after NIPBL depletion when distal regulation was attenuated-either by intrinsic weakness of the distal enhancer (Car2 E2 at 100 kb) or increased genomic distance (SCR at 365 kb).

“Furthermore, SRR2 was required to maintain Sox2 expression both at baseline and after NIPBL depletion when distal regulation was attenuated—either because of intrinsic weakness of the distal enhancer (using Car2 E2 at 100 kb) or increased genomic distance (placing SCR at 365 kb) (fig. S29B).”

NIPBL depletion decreased Hi-C and 4C interactions between Sox2 and the SCR.

“NIPBL depletion decreased Hi-C and 4C interactions between Sox2 and the SCR (fig. S19, D and E, and fig. S20).”

Lines with enhancer insertions within 11 kb of the Car2 promoter were not down-regulated when extrusion was disrupted.

“lines with enhancer insertions within 11 kb of the Car2 promoter were not down-regulated when extrusion was disrupted.”

When distal enhancers were present on the same allele, SRR2 synergized with them to boost basal Car2 expression.

“When distal enhancers were present on the same allele, SRR2 synergized with them to boost basal Car2 expression (Fig. 6A and fig. S33A)”

Despite the SCR having stronger activity than the Car2 enhancers, Car2 expression remained equally sensitive to NIPBL depletion beyond the same 10- to 19-kb threshold.

“Despite the SCR having stronger activity than the Car2 enhancers, Car2 expression remained equally sensitive to NIPBL depletion beyond the same 10- to 19-kb threshold (Fig. 4D and fig. S16A).”

Few genes were dysregulated in both ESCs and NPCs upon NIPBL depletion.

“Indeed, few genes were dysregulated in both cell types”

Wild-type untagged cells were largely unaffected by dTAG treatment.

“Wild-type (WT) untagged cells were largely unaffected by dTAG treatment (fig. S4G).”

Deleting both H3K27ac clusters abolished Car2 expression.

“deleting both abolished Car2 expression (Fig. 3E and fig S14, E and F).”

SRR2 reduced the reliance of Car2 on loop extrusion.

“SRR2 reduced the reliance of Car2 on loop extrusion.”

Removing all auxiliary elements in Sox2 reporter cells by five heterozygous deletions left basal Sox2 expression largely stable across all deletions.

“Whereas basal Sox2 expression remained largely stable across all deletions”

Sox2 remained largely insensitive to NIPBL depletion when the SCR was replaced with Car2 CE2.

“Yet Sox2 remained largely insensitive to NIPBL depletion (Fig. 4F and fig. S23A)”

The vast majority of genes sensitive to NIPBL depletion in ESCs remained stable in NPCs.

“the vast majority of the genes sensitive to NIPBL depletion in ESCs remained stable in NPCs (Fig. 2C).”

By 4 days of NIPBL depletion, Car2 expression had dropped by ~80%.

“By this time, Car2 expression had dropped by ~80% (fig. S12B).”

Inserting the weaker Car2 E2 at 17 kb did not drive higher Sox2 expression than at 100 kb.

“Inserting the weaker Car2 E2 at 17 kb did not drive higher expression than at 100 kb either”

When the SCR was >40 kb from Sox2, SRR2 became increasingly essential to maintain SCR-mediated regulation of Sox2 upon NIPBL depletion.

“Yet when the SCR was >40 kb from Sox2, SRR2 became increasingly essential to maintain SCR-mediated regulation of Sox2 upon NIPBL depletion (Fig. 5B and fig. S29A).”

Car2 expression is similar when driven by either CE1 or CE2 from the same distal positions.

“we observed similar expression of Car2 when driven by either CE1 or CE2 from the same distal positions (Fig. 3G and fig. S15A).”

Inserting SRR2 at Car2 with the same intervening CTCF sites enabled partial bypass of insulation.

“There, SRR2 again enabled partial bypass of insulation (Fig. 6D and fig. S35).”

Depleting either NIPBL or RAD21 markedly reduced contacts between the Sox2 promoter and SCR.

“Depleting either NIPBL or RAD21 markedly reduced contacts between the Sox2 promoter and SCR (Fig. 4H and fig. S20).”

Down-regulated genes upon NIPBL depletion tended to have a higher density of H3K27ac and NIPBL peaks within ~200 kb.

“we found that down-regulated genes tended to have a higher density of H3K27ac and NIPBL peaks within ~200 kb (Fig. 1E and fig. S7)”

Mutating the Car2 promoter to abolish CTCF binding did not alter the reliance of Car2 on cohesin loop extrusion.

“Mutating the Car2 promoter to abolish CTCF binding did not alter the reliance of Car2 on cohesin loop extrusion (fig. S17).”

The transcriptional response to NIPBL depletion was largely distinct between ESCs and NPCs.

“the transcriptional response to NIPBL depletion was largely distinct from that of ESCs (Fig. 2C).”

Reinserting the SCR closer to Sox2 (4, 17, or 43 kb downstream of the TSS) rendered the locus entirely insensitive to NIPBL depletion.

“Reinserting the SCR closer to Sox2 (4, 17, or 43 kb downstream of the TSS) rendered the locus entirely insensitive to NIPBL depletion (fig. S23B)”

The entire cell population uniformly down-regulates Car2 after depleting NIPBL.

“This single-cell analysis also revealed that the entire cell population uniformly down-regulates Car2 after depleting NIPBL (fig. S14B).”

In contrast to Sox2, the same CTCF cassette fully inhibited the Car2 enhancers.

“in contrast to Car2, where the same CTCF cassette fully inhibited the enhancers (Fig. 3E).”

NIPBL knockout ESCs are not viable.

“As NIPBL knockout ESCs are not viable (fig. S3)”

Deletion of SRR2 has no discernable impact on Sox2-SCR contacts before NIPBL depletion.

“Deletion of SRR2 has no discernable impact before NIPBL depletion (Fig. 4H and fig. S28).”

Inserting strong CTCF sites between Car2 and its enhancers abolished Car2 expression in cis.

“inserting strong CTCF sites between Car2 and its enhancers abolished Car2 expression in cis”

None of the other auxiliary element deletions exhibited the SRR2-like behavior.

“None of the other deletions exhibited this behavior (fig. S25B).”

Global H3K27ac and H3K27me3 CUT&Tag signal remained largely stable after 24 hours of NIPBL depletion.

“Global H3K27ac and H3K27me3 CUT&Tag (cleavage under targets and tagmentation) signal remained largely stable after 24 hours of NIPBL depletion (fig. S6)”

NIPBL depletion only caused ~15% down-regulation of Sox2 reporters within 24 hours.

“NIPBL depletion only caused ~15% down-regulation of Sox2 reporters within 24 hours (fig. S22, B and C)”

NIPBL depletion only dysregulated 8% of differentiation-dynamic genes after subtracting genes already dysregulated in ESCs.

“NIPBL depletion only dysregulated 8% of the differentiation-dynamic genes (after subtracting genes already dysregulated in ESCs; Fig. 2D and fig. S9, D and E)”

Long-range regulatory effects scaled distinctly with genomic distance between Sox2 and Car2.

“Long-range regulatory effects scaled distinctly with genomic distance between Sox2 and Car2, corroborating the importance of genomic context observed at other loci (56).”

Promoter-proximal regulatory elements are more effective at supporting distal enhancer action at Sox2 than at Car2.

“We noticed that promoter-proximal regulatory elements are more effective at supporting distal enhancer action at Sox2 than Car2.”

The paper's 4C experiments do not support a cohesin-independent tethering model for Sox2.

“Although our 4C experiments do not support a cohesin-independent tethering model for Sox2, we cannot rule out that promoter-proximal elements sustain higher-order nuclear microenvironments that are not readily detected by chromosome conformation capture.”

• AssumptionRests on an assumption: The dTAG-13 treatment specifically and completely depletes NIPBL-FKBP in the relevant cell types and time windows without confounding off-target effects.

SRR2 can support cohesin-independent long-range regulation outside of the Sox2 locus.

“SRR2 can therefore support cohesin-independent long-range regulation outside of the Sox2 locus.”

The contrasting behavior between Sox2 and Car2 arises from specificities of the Sox2 locus, not the SCR.

“This contrasting behavior arises from specificities of the Sox2 locus, not the SCR”

Only in the absence of such proximal elements, as in the Car2 locus, does cohesin extrusion prove essential for efficient regulatory communication between promoters and distal enhancers.

“Only in the absence of such proximal elements, as in the Car2 locus, does cohesin extrusion prove essential for efficient regulatory communication between promoters and distal enhancers (Fig. 3I).”

NIPBL depletion in mouse ESCs reduces NIPBL levels below 10% within hours of dTAG-13 exposure.

“NIPBL levels fell below 10% within hours of exposure to dTAG-13 (dTAG, fig. S1B)”

NIPBL is essential for maintaining transcriptional profiles.

“Altogether, these analyses indicate that NIPBL is essential for maintaining transcriptional profiles”

The mild but detectable down-regulation of Sox2 when the SCR is naturally positioned at 100 kb (<25%) arises from its distal location.

“demonstrating that the mild but detectable down-regulation of Sox2 when the SCR is naturally positioned at 100 kb (<25%) arises from its distal location.”

This distance corresponds to the genomic range where enhancers can function independently of cohesin.

“Notably, this distance corresponds to the genomic range where enhancers can function independently of cohesin (Figs. 3 and 4).”

SRR2 must lie less than 19 kb from the Car2 promoter to support loop extrusion-independent activation by distal enhancers.

“These observations further demonstrate that SRR2 must lie less than 19 kb from the Car2 promoter to support loop extrusion–independent activation by distal enhancers.”

SRR2 enables partial functional bypass of strong CTCF sites.

“SRR2 therefore enables partial functional bypass of strong CTCF sites.”

Cohesin loop extrusion is not generally more important for inducing transcriptional changes than steady-state maintenance.

“Altogether, our observations indicate that cohesin loop extrusion is not generally more important for inducing transcriptional changes than steady-state maintenance.”

Buffering against the disruption of loop extrusion is not a unique property of SRR2 but rather a degenerate feature of weak enhancers when located proximal to a promoter.

“These experiments demonstrate that buffering against the disruption of loop extrusion is not a unique property of SRR2 but rather a degenerate feature of weak enhancers when located proximal to a promoter.”

Elements such as SRR2 support cohesin-independent regulation by distal enhancers most effectively when located proximal (< ~17 kb) to the TSS.

“Therefore, elements such as SRR2 support cohesin-independent regulation by distal enhancers most effectively when located proximal (< ~17 kb) to the TSS.”

Promoters and enhancers do not themselves specify sensitivity to loop extrusion.

“Second, promoters and enhancers do not themselves specify sensitivity to loop extrusion.”

Loop extrusion can be important for the maintenance of transcriptional profiles, not simply for inducing new ones.

“Our observations in embryonic stem cells and neural precursors demonstrate that loop extrusion can be important for the maintenance of transcriptional profiles, not simply for inducing new ones”

The resilience of Sox2 to loss of loop extrusion is unlikely to arise from properties intrinsic to the SCR.

“The resilience of Sox2 to loss of loop extrusion is therefore unlikely to arise from properties intrinsic to the SCR.”

A mechanism exists through which promoter-proximal regulatory elements can bypass loop extrusion for distal regulatory communication.

“Third, we identified a mechanism through which promoter-proximal regulatory elements can bypass loop extrusion for distal regulatory communication.”

Loci across the genome exhibit pervasive shifts in cohesin sensitivity, in line with the highly cell type-specific nature of enhancers.

“we found that loci across the genome exhibit pervasive shifts in cohesin sensitivity, in line with the highly cell type–specific nature of enhancers.”

The limited distance dependence and loop extrusion dependence of the native Sox2 locus must arise from genetic elements other than the Sox2 promoter or the SCR.

“Therefore, the limited distance dependence and loop extrusion dependence of the native Sox2 locus must arise from genetic elements other than the Sox2 promoter or the SCR.”

Whether a given gene depends on loop extrusion is dictated by cell type-specific processes.

“whether a given gene depends on loop extrusion is dictated by cell type–specific processes.”

Loop extrusion across the Car2 locus is absolutely needed to connect Car2 with its distal enhancers for proper activation.

“indicating that loop extrusion across the locus is absolutely needed to connect Car2 with its distal enhancers for proper activation (Fig. 3E and fig. S14, E and F).”

Cohesin loop extrusion can be essential to support communication between distal, but not proximal, enhancer-promoter pairs.

“our experiments establish that cohesin loop extrusion can be essential to support communication between distal, but not proximal, enhancer-promoter pairs (Fig. 3I).”

Synergy between proximal and distal regulatory elements plays a previously unappreciated role in the robustness of cohesin-independent long-range enhancer-promoter communication.

“These observations indicate that synergy between proximal and distal regulatory elements plays a previously unappreciated role in the robustness of cohesin-independent long-range enhancer-promoter communication.”

The independent, parallel contributions of cohesin loop extrusion and proximal regulatory elements can impart robustness to long-range enhancer action.

“Our observations illustrate how the independent, parallel contributions of cohesin loop extrusion and proximal regulatory elements can impart robustness to long-range enhancer action”

Proximal elements can also sustain cohesin-independent regulation across a strong CTCF insulator.

“This demonstrates that proximal elements can also sustain cohesin-independent regulation across a strong CTCF insulator.”

Only upon disrupting both pathways does Sox2 lose the ability to communicate with the distal SCR enhancer.

“Only upon disrupting both pathways does Sox2 lose the ability to communicate with the distal SCR enhancer.”

The cis-regulatory logic driving transcriptional dynamics and the emergence of new cell types appears to operate largely independently of the long-range interactions created by cohesin loop extrusion.

“The cis-regulatory logic driving transcriptional dynamics and the emergence of new cell types therefore appears to operate largely independently of the long-range interactions created by cohesin loop extrusion.”

Sensitivity to loop extrusion requires sufficiently distal and essential enhancers.

“First, sensitivity to loop extrusion requires sufficiently distal and essential enhancers.”

The resistance of Sox2 to the loss of loop extrusion stems from the promoter-proximal SRR2 element.

“Therefore, the resistance of Sox2 to the loss of loop extrusion stems from the promoter-proximal SRR2 element.”

The pattern of transcriptional dysregulation was reproducible with a time course using an independent NIPBL-AID ESC line.

“This pattern was reproducible with a time course using an independent ESC line generated with the auxin-inducible degron system (NIPBL-AID, fig. S4, A to F).”

Other context features in the extended Sox2 locus, beyond the sequence of the promoter and the main enhancer, must be at play.

“Thus, other context features in the extended Sox2 locus, beyond the sequence of the promoter and the main enhancer, must be at play.”

Promoter sequence alone cannot account for the resilience of Sox2 to loss of loop extrusion.

“indicating that promoter sequence alone cannot account for the resilience of Sox2 to loss of loop extrusion.”

The cohesin independence conferred by the SRR2 element at Sox2 constitutes a previously uncharacterized regulatory behavior.

“The cohesin independence conferred by the SRR2 element at Sox2 constitutes a previously uncharacterized regulatory behavior.”

At loci without such additional promoter-proximal supporting elements, such as Car2, 3D genome folding and long-range regulation remain tightly coupled.

“At loci without such additional promoter-proximal supporting elements, such as Car2, 3D genome folding and long-range regulation remain tightly coupled”

The mechanism by which SRR2 confers cohesin independence at Sox2 does not rely on loop extrusion.

“Unlike ultra-long-range regulation by CTCF (57-59), this mechanism does not rely on loop extrusion.”

• AssumptionRests on an assumption: The dTAG-13 treatment specifically and completely depletes NIPBL-FKBP in the relevant cell types and time windows without confounding off-target effects.