Cohesin mediates DNA loop extrusion and sister chromatid cohesion by distinct mechanisms.

Cohesin connects CTCF-binding sites and other genomic loci in cis to form chromatin loops and replicated DNA molecules in trans to mediate sister chromatid cohesion. Whether cohesin uses distinct or related mechanisms to perform these functions is unknown. Here, we describe a cohesin hinge mutant that can extrude DNA into loops but is unable to mediate cohesion in human cells. Our results suggest that the latter defect arises during cohesion establishment. The observation that cohesin's cohesion and loop extrusion activities can be partially separated indicates that cohesin uses distinct mechanisms to perform these two functions. Unexpectedly, the same hinge mutant can also not be stopped by CTCF boundaries as well as wild-type cohesin. This suggests that cohesion establishment and cohesin's interaction with CTCF boundaries depend on related mechanisms and raises the possibility that both require transient hinge opening to entrap DNA inside the cohesin ring.

Testing pseudotopological and nontopological models for SMC-driven DNA loop extrusion against roadblock-traversal experiments.

DNA loop extrusion by structural-maintenance-of-chromosome (SMC) complexes has emerged as a primary organizing principle for chromosomes. The mechanism by which SMC motor proteins extrude DNA loops is still unresolved and much debated. The ring-like structure of SMC complexes prompted multiple models where the extruded DNA is topologically or pseudotopologically entrapped within the ring during loop extrusion. However, recent experiments showed the passage of roadblocks much bigger than the SMC ring size, suggesting a nontopological mechanism. Recently, attempts were made to reconcile the observed passage of large roadblocks with a pseudotopological mechanism. Here we examine the predictions of these pseudotopological models and find that they are not consistent with new experimental data on SMC roadblock encounters. Particularly, these models predict the formation of two loops and that roadblocks will reside near the stem of the loop upon encounter-both in contrast to experimental observations. Overall, the experimental data reinforce the notion of a nontopological mechanism for extrusion of DNA.

CTCF is a DNA-tension-dependent barrier to cohesin-mediated loop extrusion.

In eukaryotes, genomic DNA is extruded into loops by cohesin1. By restraining this process, the DNA-binding protein CCCTC-binding factor (CTCF) generates topologically associating domains (TADs)2,3 that have important roles in gene regulation and recombination during development and disease1,4-7. How CTCF establishes TAD boundaries and to what extent these are permeable to cohesin is unclear8. Here, to address these questions, we visualize interactions of single CTCF and cohesin molecules on DNA in vitro. We show that CTCF is sufficient to block diffusing cohesin, possibly reflecting how cohesive cohesin accumulates at TAD boundaries, and is also sufficient to block loop-extruding cohesin, reflecting how CTCF establishes TAD boundaries. CTCF functions asymmetrically, as predicted; however, CTCF is dependent on DNA tension. Moreover, CTCF regulates cohesin's loop-extrusion activity by changing its direction and by inducing loop shrinkage. Our data indicate that CTCF is not, as previously assumed, simply a barrier to cohesin-mediated loop extrusion but is an active regulator of this process, whereby the permeability of TAD boundaries can be modulated by DNA tension. These results reveal mechanistic principles of how CTCF controls loop extrusion and genome architecture.

SMC complexes can traverse physical roadblocks bigger than their ring size.

Ring-shaped structural maintenance of chromosomes (SMC) complexes like condensin and cohesin extrude loops of DNA. It remains, however, unclear how they can extrude DNA loops in chromatin that is bound with proteins. Here, we use in vitro single-molecule visualization to show that nucleosomes, RNA polymerase, and dCas9 pose virtually no barrier to loop extrusion by yeast condensin. We find that even DNA-bound nanoparticles as large as 200 nm, much bigger than the SMC ring size, also translocate into DNA loops during extrusion by condensin and cohesin. This even occurs for a single-chain version of cohesin in which the ring-forming subunits are covalently linked and cannot open to entrap DNA. The data show that SMC-driven loop extrusion has surprisingly little difficulty in accommodating large roadblocks into the loop. The findings also show that the extruded DNA does not pass through the SMC ring (pseudo)topologically, hence pointing to a nontopological mechanism for DNA loop extrusion.

MCM complexes are barriers that restrict cohesin-mediated loop extrusion.

Eukaryotic genomes are compacted into loops and topologically associating domains (TADs)1-3, which contribute to transcription, recombination and genomic stability4,5. Cohesin extrudes DNA into loops that are thought to lengthen until CTCF boundaries are encountered6-12. Little is known about whether loop extrusion is impeded by DNA-bound machines. Here we show that the minichromosome maintenance (MCM) complex is a barrier that restricts loop extrusion in G1 phase. Single-nucleus Hi-C (high-resolution chromosome conformation capture) of mouse zygotes reveals that MCM loading reduces CTCF-anchored loops and decreases TAD boundary insulation, which suggests that loop extrusion is impeded before reaching CTCF. This effect extends to HCT116 cells, in which MCMs affect the number of CTCF-anchored loops and gene expression. Simulations suggest that MCMs are abundant, randomly positioned and partially permeable barriers. Single-molecule imaging shows that MCMs are physical barriers that frequently constrain cohesin translocation in vitro. Notably, chimeric yeast MCMs that contain a cohesin-interaction motif from human MCM3 induce cohesin pausing, indicating that MCMs are 'active' barriers with binding sites. These findings raise the possibility that cohesin can arrive by loop extrusion at MCMs, which determine the genomic sites at which sister chromatid cohesion is established. On the basis of in vivo, in silico and in vitro data, we conclude that distinct loop extrusion barriers shape the three-dimensional genome.

Cornelia de Lange syndrome mutations in NIPBL can impair cohesin-mediated DNA loop extrusion.

Cornelia de Lange syndrome (CdLS) is a developmental multisystem disorder frequently associated with mutations in NIPBL. CdLS is thought to arise from developmental gene regulation defects, but how NIPBL mutations cause these is unknown. Here we show that several NIPBL mutations impair the DNA loop extrusion activity of cohesin. Because this activity is required for the formation of chromatin loops and topologically associating domains, which have important roles in gene regulation, our results suggest that defects in cohesin-mediated loop extrusion contribute to the etiology of CdLS by altering interactions between developmental genes and their enhancers.

Cohesin mediates DNA loop extrusion by a "swing and clamp" mechanism.

Structural maintenance of chromosomes (SMC) complexes organize genome topology in all kingdoms of life and have been proposed to perform this function by DNA loop extrusion. How this process works is unknown. Here, we have analyzed how loop extrusion is mediated by human cohesin-NIPBL complexes, which enable chromatin folding in interphase cells. We have identified DNA binding sites and large-scale conformational changes that are required for loop extrusion and have determined how these are coordinated. Our results suggest that DNA is translocated by a spontaneous 50 nm-swing of cohesin's hinge, which hands DNA over to the ATPase head of SMC3, where upon binding of ATP, DNA is clamped by NIPBL. During this process, NIPBL "jumps ship" from the hinge toward the SMC3 head and might thereby couple the spontaneous hinge swing to ATP-dependent DNA clamping. These results reveal mechanistic principles of how cohesin-NIPBL and possibly other SMC complexes mediate loop extrusion.

Genome folding through loop extrusion by SMC complexes.

Genomic DNA is folded into loops and topologically associating domains (TADs), which serve important structural and regulatory roles. It has been proposed that these genomic structures are formed by a loop extrusion process, which is mediated by structural maintenance of chromosomes (SMC) protein complexes. Recent single-molecule studies have shown that the SMC complexes condensin and cohesin are indeed able to extrude DNA into loops. In this Review, we discuss how the loop extrusion hypothesis can explain key features of genome architecture; cellular functions of loop extrusion, such as separation of replicated DNA molecules, facilitation of enhancer-promoter interactions and immunoglobulin gene recombination; and what is known about the mechanism of loop extrusion and its regulation, for example, by chromatin boundaries that depend on the DNA binding protein CTCF. We also discuss how the loop extrusion hypothesis has led to a paradigm shift in our understanding of both genome architecture and the functions of SMC complexes.

DNA loop extrusion by human cohesin.

Eukaryotic genomes are folded into loops and topologically associating domains, which contribute to chromatin structure, gene regulation, and gene recombination. These structures depend on cohesin, a ring-shaped DNA-entrapping adenosine triphosphatase (ATPase) complex that has been proposed to form loops by extrusion. Such an activity has been observed for condensin, which forms loops in mitosis, but not for cohesin. Using biochemical reconstitution, we found that single human cohesin complexes form DNA loops symmetrically at rates up to 2.1 kilo-base pairs per second. Loop formation and maintenance depend on cohesin's ATPase activity and on NIPBL-MAU2, but not on topological entrapment of DNA by cohesin. During loop formation, cohesin and NIPBL-MAU2 reside at the base of loops, which indicates that they generate loops by extrusion. Our results show that cohesin and NIPBL-MAU2 form an active holoenzyme that interacts with DNA either pseudo-topologically or non-topologically to extrude genomic interphase DNA into loops.

Protein engineering of a ubiquitin-variant inhibitor of APC/C identifies a cryptic K48 ubiquitin chain binding site.

Ubiquitin (Ub)-mediated proteolysis is a fundamental mechanism used by eukaryotic cells to maintain homeostasis and protein quality, and to control timing in biological processes. Two essential aspects of Ub regulation are conjugation through E1-E2-E3 enzymatic cascades and recognition by Ub-binding domains. An emerging theme in the Ub field is that these 2 properties are often amalgamated in conjugation enzymes. In addition to covalent thioester linkage to Ub's C terminus for Ub transfer reactions, conjugation enzymes often bind noncovalently and weakly to Ub at "exosites." However, identification of such sites is typically empirical and particularly challenging in large molecular machines. Here, studying the 1.2-MDa E3 ligase anaphase-promoting complex/cyclosome (APC/C), which controls cell division and many aspects of neurobiology, we discover a method for identifying unexpected Ub-binding sites. Using a panel of Ub variants (UbVs), we identify a protein-based inhibitor that blocks Ub ligation to APC/C substrates in vitro and ex vivo. Biochemistry, NMR, and cryo-electron microscopy (cryo-EM) structurally define the UbV interaction, explain its inhibitory activity through binding the surface on the APC2 subunit that recruits the E2 enzyme UBE2C, and ultimately reveal that this APC2 surface is also a Ub-binding exosite with preference for K48-linked chains. The results provide a tool for probing APC/C activity, have implications for the coordination of K48-linked Ub chain binding by APC/C with the multistep process of substrate polyubiquitylation, and demonstrate the power of UbV technology for identifying cryptic Ub-binding sites within large multiprotein complexes.

Rapid movement and transcriptional re-localization of human cohesin on DNA.

The spatial organization, correct expression, repair, and segregation of eukaryotic genomes depend on cohesin, ring-shaped protein complexes that are thought to function by entrapping DNA It has been proposed that cohesin is recruited to specific genomic locations from distal loading sites by an unknown mechanism, which depends on transcription, and it has been speculated that cohesin movements along DNA could create three-dimensional genomic organization by loop extrusion. However, whether cohesin can translocate along DNA is unknown. Here, we used single-molecule imaging to show that cohesin can diffuse rapidly on DNA in a manner consistent with topological entrapment and can pass over some DNA-bound proteins and nucleosomes but is constrained in its movement by transcription and DNA-bound CCCTC-binding factor (CTCF). These results indicate that cohesin can be positioned in the genome by moving along DNA, that transcription can provide directionality to these movements, that CTCF functions as a boundary element for moving cohesin, and they are consistent with the hypothesis that cohesin spatially organizes the genome via loop extrusion.

Dual RING E3 Architectures Regulate Multiubiquitination and Ubiquitin Chain Elongation by APC/C.

Protein ubiquitination involves E1, E2, and E3 trienzyme cascades. E2 and RING E3 enzymes often collaborate to first prime a substrate with a single ubiquitin (UB) and then achieve different forms of polyubiquitination: multiubiquitination of several sites and elongation of linkage-specific UB chains. Here, cryo-EM and biochemistry show that the human E3 anaphase-promoting complex/cyclosome (APC/C) and its two partner E2s, UBE2C (aka UBCH10) and UBE2S, adopt specialized catalytic architectures for these two distinct forms of polyubiquitination. The APC/C RING constrains UBE2C proximal to a substrate and simultaneously binds a substrate-linked UB to drive processive multiubiquitination. Alternatively, during UB chain elongation, the RING does not bind UBE2S but rather lures an evolving substrate-linked UB to UBE2S positioned through a cullin interaction to generate a Lys11-linked chain. Our findings define mechanisms of APC/C regulation, and establish principles by which specialized E3-E2-substrate-UB architectures control different forms of polyubiquitination.

Mechanism of APC/CCDC20 activation by mitotic phosphorylation.

Chromosome segregation and mitotic exit are initiated by the 1.2-MDa ubiquitin ligase APC/C (anaphase-promoting complex/cyclosome) and its coactivator CDC20 (cell division cycle 20). To avoid chromosome missegregation, APC/C(CDC20) activation is tightly controlled. CDC20 only associates with APC/C in mitosis when APC/C has become phosphorylated and is further inhibited by a mitotic checkpoint complex until all chromosomes are bioriented on the spindle. APC/C contains 14 different types of subunits, most of which are phosphorylated in mitosis on multiple sites. However, it is unknown which of these phospho-sites enable APC/C(CDC20) activation and by which mechanism. Here we have identified 68 evolutionarily conserved mitotic phospho-sites on human APC/C bound to CDC20 and have used the biGBac technique to generate 47 APC/C mutants in which either all 68 sites or subsets of them were replaced by nonphosphorylatable or phospho-mimicking residues. The characterization of these complexes in substrate ubiquitination and degradation assays indicates that phosphorylation of an N-terminal loop region in APC1 is sufficient for binding and activation of APC/C by CDC20. Deletion of the N-terminal APC1 loop enables APC/C(CDC20) activation in the absence of mitotic phosphorylation or phospho-mimicking mutations. These results indicate that binding of CDC20 to APC/C is normally prevented by an autoinhibitory loop in APC1 and that its mitotic phosphorylation relieves this inhibition. The predicted location of the N-terminal APC1 loop implies that this loop controls interactions between the N-terminal domain of CDC20 and APC1 and APC8. These results reveal how APC/C phosphorylation enables CDC20 to bind and activate the APC/C in mitosis.

Characterization of a DNA exit gate in the human cohesin ring.

Chromosome segregation depends on sister chromatid cohesion mediated by cohesin. The cohesin subunits Smc1, Smc3, and Scc1 form tripartite rings that are thought to open at distinct sites to allow entry and exit of DNA. However, direct evidence for the existence of open forms of cohesin is lacking. We found that cohesin's proposed DNA exit gate is formed by interactions between Scc1 and the coiled-coil region of Smc3. Mutation of this interface abolished cohesin's ability to stably associate with chromatin and to mediate cohesion. Electron microscopy revealed that weakening of the Smc3-Scc1 interface resulted in opening of cohesin rings, as did proteolytic cleavage of Scc1. These open forms may resemble intermediate states of cohesin normally generated by the release factor Wapl and the protease separase, respectively.

Mechanism of polyubiquitination by human anaphase-promoting complex: RING repurposing for ubiquitin chain assembly.

Polyubiquitination by E2 and E3 enzymes is a predominant mechanism regulating protein function. Some RING E3s, including anaphase-promoting complex/cyclosome (APC), catalyze polyubiquitination by sequential reactions with two different E2s. An initiating E2 ligates ubiquitin to an E3-bound substrate. Another E2 grows a polyubiquitin chain on the ubiquitin-primed substrate through poorly defined mechanisms. Here we show that human APC's RING domain is repurposed for dual functions in polyubiquitination. The canonical RING surface activates an initiating E2-ubiquitin intermediate for substrate modification. However, APC engages and activates its specialized ubiquitin chain-elongating E2 UBE2S in ways that differ from current paradigms. During chain assembly, a distinct APC11 RING surface helps deliver a substrate-linked ubiquitin to accept another ubiquitin from UBE2S. Our data define mechanisms of APC/UBE2S-mediated polyubiquitination, reveal diverse functions of RING E3s and E2s, and provide a framework for understanding distinctive RING E3 features specifying ubiquitin chain elongation.

Cohesin's ATPase activity couples cohesin loading onto DNA with Smc3 acetylation.

BACKGROUND: Cohesin mediates sister chromatid cohesion by topologically entrapping sister DNA molecules inside its ring structure. Cohesin is loaded onto DNA by the Scc2/NIPBL-Scc4/MAU2-loading complex in a manner that depends on the adenosine triphosphatase (ATPase) activity of cohesin's Smc1 and Smc3 subunits. Subsequent cohesion establishment during DNA replication depends on Smc3 acetylation by Esco1 and Esco2 and on recruitment of sororin, which "locks" cohesin on DNA by inactivating the cohesin release factor Wapl. RESULTS: Human cohesin ATPase mutants associate transiently with DNA in a manner that depends on the loading complex but cannot be stabilized on chromatin by depletion of Wapl. These mutants cannot be acetylated, fail to interact with sororin, and do not mediate cohesion. The absence of Smc3 acetylation in the ATPase mutants is not a consequence of their transient association with DNA but is directly caused by their inability to hydrolyze ATP because acetylation of wild-type cohesin also depends on ATP hydrolysis. CONCLUSIONS: Our data indicate that cohesion establishment involves the following steps. First, cohesin transiently associates with DNA in a manner that depends on the loading complex. Subsequently, ATP hydrolysis by cohesin leads to entrapment of DNA and converts Smc3 into a state that can be acetylated. Finally, Smc3 acetylation leads to recruitment of sororin, inhibition of Wapl, and stabilization of cohesin on DNA. Our finding that cohesin's ATPase activity is required for both cohesin loading and Smc3 acetylation raises the possibility that cohesion establishment is directly coupled to the reaction in which cohesin entraps DNA.

Mitotic lamin disassembly is triggered by lipid-mediated signaling.

Disassembly of the nuclear lamina is a key step during open mitosis in higher eukaryotes. The activity of several kinases, including CDK1 (cyclin-dependent kinase 1) and protein kinase C (PKC), has been shown to trigger mitotic lamin disassembly, yet their precise contributions are unclear. In this study, we develop a quantitative imaging assay to study mitotic lamin B1 disassembly in living cells. We find that CDK1 and PKC act in concert to mediate phosphorylation-dependent lamin B1 disassembly during mitosis. Using ribonucleic acid interference (RNAi), we showed that diacylglycerol (DAG)-dependent PKCs triggered rate-limiting steps of lamin disassembly. RNAi-mediated depletion or chemical inhibition of lipins, enzymes that produce DAG, delayed lamin disassembly to a similar extent as does PKC inhibition/depletion. Furthermore, the delay of lamin B1 disassembly after lipin depletion could be rescued by the addition of DAG. These findings suggest that lipins activate a PKC-dependent pathway during mitotic lamin disassembly and provide evidence for a lipid-mediated mitotic signaling event.

Coordination of kinase and phosphatase activities by Lem4 enables nuclear envelope reassembly during mitosis.

Mitosis in metazoa requires nuclear envelope (NE) disassembly and reassembly. NE disassembly is driven by multiple phosphorylation events. Mitotic phosphorylation of the protein BAF reduces its affinity for chromatin and the LEM family of inner nuclear membrane proteins; loss of this BAF-mediated chromatin-NE link contributes to NE disassembly. BAF must reassociate with chromatin and LEM proteins at mitotic exit to reform the NE; however, how its dephosphorylation is regulated is unknown. Here, we show that the C. elegans protein LEM-4L and its human ortholog Lem4 (also called ANKLE2) are both required for BAF dephosphorylation. They act in part by inhibiting BAF's mitotic kinase, VRK-1, in vivo and in vitro. In addition, Lem4/LEM-4L interacts with PP2A and is required for it to dephosphorylate BAF during mitotic exit. By coordinating VRK-1- and PP2A-mediated signaling on BAF, Lem4/LEM-4L controls postmitotic NE formation in a function conserved from worms to humans.

Deregulated replication licensing causes DNA fragmentation consistent with head-to-tail fork collision.

Correct regulation of the replication licensing system ensures that no DNA is rereplicated in a single cell cycle. When the licensing protein Cdt1 is overexpressed in G2 phase of the cell cycle, replication origins are relicensed and the DNA is rereplicated. At the same time, checkpoint pathways are activated that block further cell cycle progression. We have studied the consequence of deregulating the licensing system by adding recombinant Cdt1 to Xenopus egg extracts. We show that Cdt1 induces checkpoint activation and the appearance of small fragments of double-stranded DNA. DNA fragmentation and strong checkpoint activation are dependent on uncontrolled rereplication and do not occur after a single coordinated round of rereplication. The DNA fragments are composed exclusively of rereplicated DNA. The unusual characteristics of these fragments suggest that they result from head-to-tail collision (rear ending) of replication forks chasing one another along the same DNA template.