Cryo-EM structure of the human cardiac myosin filament.

Pumping of the heart is powered by filaments of the motor protein myosin that pull on actin filaments to generate cardiac contraction. In addition to myosin, the filaments contain cardiac myosin-binding protein C (cMyBP-C), which modulates contractility in response to physiological stimuli, and titin, which functions as a scaffold for filament assembly1. Myosin, cMyBP-C and titin are all subject to mutation, which can lead to heart failure. Despite the central importance of cardiac myosin filaments to life, their molecular structure has remained a mystery for 60 years2. Here we solve the structure of the main (cMyBP-C-containing) region of the human cardiac filament using cryo-electron microscopy. The reconstruction reveals the architecture of titin and cMyBP-C and shows how myosin's motor domains (heads) form three different types of motif (providing functional flexibility), which interact with each other and with titin and cMyBP-C to dictate filament architecture and function. The packing of myosin tails in the filament backbone is also resolved. The structure suggests how cMyBP-C helps to generate the cardiac super-relaxed state3; how titin and cMyBP-C may contribute to length-dependent activation4; and how mutations in myosin and cMyBP-C might disturb interactions, causing disease5,6. The reconstruction resolves past uncertainties and integrates previous data on cardiac muscle structure and function. It provides a new paradigm for interpreting structural, physiological and clinical observations, and for the design of potential therapeutic drugs.

Structural basis of human transcription-DNA repair coupling.

Transcription-coupled DNA repair removes bulky DNA lesions from the genome1,2 and protects cells against ultraviolet (UV) irradiation3. Transcription-coupled DNA repair begins when RNA polymerase II (Pol II) stalls at a DNA lesion and recruits the Cockayne syndrome protein CSB, the E3 ubiquitin ligase, CRL4CSA and UV-stimulated scaffold protein A (UVSSA)3. Here we provide five high-resolution structures of Pol II transcription complexes containing human transcription-coupled DNA repair factors and the elongation factors PAF1 complex (PAF) and SPT6. Together with biochemical and published3,4 data, the structures provide a model for transcription-repair coupling. Stalling of Pol II at a DNA lesion triggers replacement of the elongation factor DSIF by CSB, which binds to PAF and moves upstream DNA to SPT6. The resulting elongation complex, ECTCR, uses the CSA-stimulated translocase activity of CSB to pull on upstream DNA and push Pol II forward. If the lesion cannot be bypassed, CRL4CSA spans over the Pol II clamp and ubiquitylates the RPB1 residue K1268, enabling recruitment of TFIIH to UVSSA and DNA repair. Conformational changes in CRL4CSA lead to ubiquitylation of CSB and to release of transcription-coupled DNA repair factors before transcription may continue over repaired DNA.

Cannabinoid receptor interacting protein 1a interacts with myristoylated Gαi N terminus via a unique gapped ß-barrel structure.

Cannabinoid receptor interacting protein 1a (CRIP1a) modulates CB1 cannabinoid receptor G-protein coupling in part by altering the selectivity for Gαi subtype activation, but the molecular basis for this function of CRIP1a is not known. We report herein the first structure of CRIP1a at a resolution of 1.55 Å. CRIP1a exhibits a 10-stranded and antiparallel ß-barrel with an interior comprised of conserved hydrophobic residues and loops at the bottom and a short helical cap at the top to exclude solvent. The ß-barrel has a gap between strands ß8 and ß10, which deviates from ß-sandwich fatty acid-binding proteins that carry endocannabinoid compounds and the Rho-guanine nucleotide dissociation inhibitor predicted by computational threading algorithms. The structural homology search program DALI identified CRIP1a as homologous to a family of lipidated-protein carriers that includes phosphodiesterase 6 delta subunit and Unc119. Comparison with these proteins suggests that CRIP1a may carry two possible types of cargo: either (i) like phosphodiesterase 6 delta subunit, cargo with a farnesyl moiety that enters from the top of the ß-barrel to occupy the hydrophobic interior or (ii) like Unc119, cargo with a palmitoyl or a myristoyl moiety that enters from the side where the missing ß-strand creates an opening to the hydrophobic pocket. Fluorescence polarization analysis demonstrated CRIP1a binding of an N-terminally myristoylated 9-mer peptide mimicking the Gαi N terminus. However, CRIP1a could not bind the nonmyristolyated Gαi peptide or cargo of homologs. Thus, binding of CRIP1a to Gαi proteins represents a novel mechanism to regulate cell signaling initiated by the CB1 receptor.

Molecular interactions between C9ORF72 and SMCR8: A local energetic frustration perspective.

The hexanucleotide repeat expansion in C9orf72 represents a major cause of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). C9orf72, together with SMCR8 and WDR41, can form a stable complex that regulates autophagy and membrane trafficking. Very recently, the cryo-EM structure of C9orf72-SMCR8-WDR41 helps in understanding the structure-function relationship of C9orf72. This protein complex is indispensable to several cellular processes and is strongly linked to familial ALS and FTD. Understanding the molecular basis of the C9orf72-SMCR8 protein-protein interaction is thus important to comprehend their function. To establish a basis for understanding the relationships between sequence, structure, and function of the C9orf72, this study reports a local frustration analysis on the C9orf72-SMCR8 complex structure. An analysis of local frustration profiles indicated that (1) the structural domains in C9orf72 are minimally-frustrated and relatively conserved, (2) high frustration patches on the protein-protein interface (3) increased frustration in the C-terminal helices involved in the dimerization of C9orf72 structures.

Autophagic defects observed in fibroblasts from a patient with ß-propeller protein-associated neurodegeneration.

Beta-propeller protein-associated neurodegeneration (BPAN) is associated with mutations in the autophagy gene WDR45. The aim of this study was to demonstrate autophagic defects in a patient with BPAN. We assayed autophagic markers using western blot analysis and immunocytochemistry and applied transmission electron microscopy (TEM) to visualize the autophagic structures in fibroblasts from a 7-year-old Korean female with WDR45 splice-site mutation (c.977-1G>A; NM_007075.3). The protein and mRNA expression levels of WDR45 gene were decreased in the patient-derived fibroblasts. The amount of increase in LC3-II upon treatment with an autophagy inducer and inhibitor was reduced in mutant cells compared to control cells, suggesting decreased autophagic flux. TEM showed the accumulation of large vacuoles in mutant cells with a decrease of autophagosomes. Our study demonstrated that the WDR45 mutation in this patient impaired autophagy and provided additional insight into ultrastructural changes of autophagic structures.

Protein haploinsufficiency drivers identify MYBPC3 variants that cause hypertrophic cardiomyopathy.

Hypertrophic cardiomyopathy (HCM) is the most common inherited cardiac disease. Variants in MYBPC3, the gene encoding cardiac myosin-binding protein C (cMyBP-C), are the leading cause of HCM. However, the pathogenicity status of hundreds of MYBPC3 variants found in patients remains unknown, as a consequence of our incomplete understanding of the pathomechanisms triggered by HCM-causing variants. Here, we examined 44 nontruncating MYBPC3 variants that we classified as HCM-linked or nonpathogenic according to cosegregation and population genetics criteria. We found that around half of the HCM-linked variants showed alterations in RNA splicing or protein stability, both of which can lead to cMyBP-C haploinsufficiency. These protein haploinsufficiency drivers associated with HCM pathogenicity with 100% and 94% specificity, respectively. Furthermore, we uncovered that 11% of nontruncating MYBPC3 variants currently classified as of uncertain significance in ClinVar induced one of these molecular phenotypes. Our strategy, which can be applied to other conditions induced by protein loss of function, supports the idea that cMyBP-C haploinsufficiency is a fundamental pathomechanism in HCM.

Hippocampal AMPA receptor assemblies and mechanism of allosteric inhibition.

AMPA-selective glutamate receptors mediate the transduction of signals between the neuronal circuits of the hippocampus1. The trafficking, localization, kinetics and pharmacology of AMPA receptors are tuned by an ensemble of auxiliary protein subunits, which are integral membrane proteins that associate with the receptor to yield bona fide receptor signalling complexes2. Thus far, extensive studies of recombinant AMPA receptor-auxiliary subunit complexes using engineered protein constructs have not been able to faithfully elucidate the molecular architecture of hippocampal AMPA receptor complexes. Here we obtain mouse hippocampal, calcium-impermeable AMPA receptor complexes using immunoaffinity purification and use single-molecule fluorescence and cryo-electron microscopy experiments to elucidate three major AMPA receptor-auxiliary subunit complexes. The GluA1-GluA2, GluA1-GluA2-GluA3 and GluA2-GluA3 receptors are the predominant assemblies, with the auxiliary subunits TARP-γ8 and CNIH2-SynDIG4 non-stochastically positioned at the B'/D' and A'/C' positions, respectively. We further demonstrate how the receptor-TARP-γ8 stoichiometry explains the mechanism of and submaximal inhibition by a clinically relevant, brain-region-specific allosteric inhibitor.

Structural basis for p50RhoGAP BCH domain-mediated regulation of Rho inactivation.

Spatiotemporal regulation of signaling cascades is crucial for various biological pathways, under the control of a range of scaffolding proteins. The BNIP-2 and Cdc42GAP Homology (BCH) domain is a highly conserved module that targets small GTPases and their regulators. Proteins bearing BCH domains are key for driving cell elongation, retraction, membrane protrusion, and other aspects of active morphogenesis during cell migration, myoblast differentiation, and neuritogenesis. We previously showed that the BCH domain of p50RhoGAP (ARHGAP1) sequesters RhoA from inactivation by its adjacent GAP domain; however, the underlying molecular mechanism for RhoA inactivation by p50RhoGAP remains unknown. Here, we report the crystal structure of the BCH domain of p50RhoGAP Schizosaccharomyces pombe and model the human p50RhoGAP BCH domain to understand its regulatory function using in vitro and cell line studies. We show that the BCH domain adopts an intertwined dimeric structure with asymmetric monomers and harbors a unique RhoA-binding loop and a lipid-binding pocket that anchors prenylated RhoA. Interestingly, the ß5-strand of the BCH domain is involved in an intermolecular ß-sheet, which is crucial for inhibition of the adjacent GAP domain. A destabilizing mutation in the ß5-strand triggers the release of the GAP domain from autoinhibition. This renders p50RhoGAP active, thereby leading to RhoA inactivation and increased self-association of p50RhoGAP molecules via their BCH domains. Our results offer key insight into the concerted spatiotemporal regulation of Rho activity by BCH domain-containing proteins.

Probing the Structure and Function of the Cytosolic Domain of the Human Zinc Transporter ZnT8 with Nickel(II) Ions.

The human zinc transporter ZnT8 provides the granules of pancreatic ß-cells with zinc (II) ions for assembly of insulin hexamers for storage. Until recently, the structure and function of human ZnTs have been modelled on the basis of the 3D structures of bacterial zinc exporters, which form homodimers with each monomer having six transmembrane α-helices harbouring the zinc transport site and a cytosolic domain with an α,ß structure and additional zinc-binding sites. However, there are important differences in function as the bacterial proteins export an excess of zinc ions from the bacterial cytoplasm, whereas ZnT8 exports zinc ions into subcellular vesicles when there is no apparent excess of cytosolic zinc ions. Indeed, recent structural investigations of human ZnT8 show differences in metal binding in the cytosolic domain when compared to the bacterial proteins. Two common variants, one with tryptophan (W) and the other with arginine (R) at position 325, have generated considerable interest as the R-variant is associated with a higher risk of developing type 2 diabetes. Since the mutation is at the apex of the cytosolic domain facing towards the cytosol, it is not clear how it can affect zinc transport through the transmembrane domain. We expressed the cytosolic domain of both variants of human ZnT8 and have begun structural and functional studies. We found that (i) the metal binding of the human protein is different from that of the bacterial proteins, (ii) the human protein has a C-terminal extension with three cysteine residues that bind a zinc(II) ion, and (iii) there are small differences in stability between the two variants. In this investigation, we employed nickel(II) ions as a probe for the spectroscopically silent Zn(II) ions and utilised colorimetric and fluorimetric indicators for Ni(II) ions to investigate metal binding. We established Ni(II) coordination to the C-terminal cysteines and found differences in metal affinity and coordination in the two ZnT8 variants. These structural differences are thought to be critical for the functional differences regarding the diabetes risk. Further insight into the assembly of the metal centres in the cytosolic domain was gained from potentiometric investigations of zinc binding to synthetic peptides corresponding to N-terminal and C-terminal sequences of ZnT8 bearing the metal-coordinating ligands. Our work suggests the involvement of the C-terminal cysteines, which are part of the cytosolic domain, in a metal chelation and/or acquisition mechanism and, as now supported by the high-resolution structural work, provides the first example of metal-thiolate coordination chemistry in zinc transporters.

Structure and inhibition mechanism of the human citrate transporter NaCT.

Citrate is best known as an intermediate in the tricarboxylic acid cycle of the cell. In addition to this essential role in energy metabolism, the tricarboxylate anion also acts as both a precursor and a regulator of fatty acid synthesis1-3. Thus, the rate of fatty acid synthesis correlates directly with the cytosolic concentration of citrate4,5. Liver cells import citrate through the sodium-dependent citrate transporter NaCT (encoded by SLC13A5) and, as a consequence, this protein is a potential target for anti-obesity drugs. Here, to understand the structural basis of its inhibition mechanism, we determined cryo-electron microscopy structures of human NaCT in complexes with citrate or a small-molecule inhibitor. These structures reveal how the inhibitor-which binds to the same site as citrate-arrests the transport cycle of NaCT. The NaCT-inhibitor structure also explains why the compound selectively inhibits NaCT over two homologous human dicarboxylate transporters, and suggests ways to further improve the affinity and selectivity. Finally, the NaCT structures provide a framework for understanding how various mutations abolish the transport activity of NaCT in the brain and thereby cause epilepsy associated with mutations in SLC13A5 in newborns (which is known as SLC13A5-epilepsy)6-8.

Amino Acid Composition in Various Types of Nucleic Acid-Binding Proteins.

Nucleic acid-binding proteins are traditionally divided into two categories: With the ability to bind DNA or RNA. In the light of new knowledge, such categorizing should be overcome because a large proportion of proteins can bind both DNA and RNA. Another even more important features of nucleic acid-binding proteins are so-called sequence or structure specificities. Proteins able to bind nucleic acids in a sequence-specific manner usually contain one or more of the well-defined structural motifs (zinc-fingers, leucine zipper, helix-turn-helix, or helix-loop-helix). In contrast, many proteins do not recognize nucleic acid sequence but rather local DNA or RNA structures (G-quadruplexes, i-motifs, triplexes, cruciforms, left-handed DNA/RNA form, and others). Finally, there are also proteins recognizing both sequence and local structural properties of nucleic acids (e.g., famous tumor suppressor p53). In this mini-review, we aim to summarize current knowledge about the amino acid composition of various types of nucleic acid-binding proteins with a special focus on significant enrichment and/or depletion in each category.

Exome sequencing identifies novel missense and deletion variants in RTN4IP1 associated with optic atrophy, global developmental delay, epilepsy, ataxia, and choreoathetosis.

Inherited optic neuropathies (IONs) are neurodegenerative disorders characterized by optic atrophy with or without extraocular manifestations. Optic atrophy-10 (OPA10) is an autosomal recessive ION recently reported to be caused by mutations in RTN4IP1, which encodes reticulon 4 interacting protein 1 (RTN4IP1), a mitochondrial ubiquinol oxydo-reductase. Here we report novel compound heterozygous mutations in RTN4IP1 in a male proband with developmental delay, epilepsy, optic atrophy, ataxia, and choreoathetosis. Workup was notable for transiently elevated lactate and lactate-to-pyruvate ratio, brain magnetic resonance imaging with optic atrophy and T2 signal abnormalities, and a nondiagnostic initial genetic workup, including chromosomal microarray and mitochondrial panel testing. Exome sequencing identified a paternally inherited missense variant (c.263T>G, p.Val88Gly) predicted to be deleterious and a maternally inherited deletion encompassing RTN4IP1. To our knowledge, this is the first report of a non-single nucleotide pathogenic variant associated with OPA10. This case highlights the expanding phenotypic spectrum of OPA10, the association between "syndromic" cases and severe RTN4IP1 mutations, and the importance of nonbiased genetic testing, such as ES, to analyze multiple genes and variants types, in patients suspected of having genetic disease.

Structural snapshot of the mitochondrial protein import gate.

The translocase of the outer mitochondrial membrane (TOM) complex is the main entry gate for most mitochondrial proteins. The TOM complex is a multisubunit membrane protein complex consisting of a ß-barrel protein Tom40 and six α-helical transmembrane (TM) proteins, receptor subunits Tom20, Tom22, and Tom70, and regulatory subunits Tom5, Tom6, and Tom7. Although nearly 30 years have passed since the main components of the TOM complex were identified and characterized, the structural details of the TOM complex remained poorly understood until recently. Thanks to the rapid development of the cryoelectron microscopy (EM) technology, high-resolution structures of the yeast TOM complex have become available. The identified structures showed a symmetric dimer containing five different subunits including Tom22. Biochemical and mutational analyses based on the TOM complex structure revealed the presence of different translocation paths within the Tom40 import channel for different classes of translocating precursor proteins. Previous studies including our cross-linking analyses indicated that the TOM complex in intact mitochondria is present as a mixture of the trimeric complex containing Tom22. Furthermore, the dimeric complex lacking Tom22, and the trimer and dimer may handle different sets of mitochondrial precursor proteins for translocation across the outer membrane. In this Structural Snapshot, we will discuss possible rearrangement of the subunit interactions upon dynamic conversion of the TOM complex between the different subunit assembly states, the Tom22-containing core dimer and trimer.

Spatial Structure and Activity of Synthetic Fragments of Lynx1 and of Nicotinic Receptor Loop C Models.

Lynx1, membrane-bound protein co-localized with the nicotinic acetylcholine receptors (nAChRs) and regulates their function, is a three-finger protein (TFP) made of three ß-structural loops, similarly to snake venom α-neurotoxin TFPs. Since the central loop II of α-neurotoxins is involved in binding to nAChRs, we have recently synthesized the fragments of Lynx1 central loop, including those with the disulfide between Cys residues introduced at N- and C-termini, some of them inhibiting muscle-type nAChR similarly to the whole-size water-soluble Lynx1 (ws-Lynx1). Literature shows that the main fragment interacting with TFPs is the C-loop of both nAChRs and acetylcholine binding proteins (AChBPs) while some ligand-binding capacity is preserved by analogs of this loop, for example, by high-affinity peptide HAP. Here we analyzed the structural organization of these peptide models of ligands and receptors and its role in binding. Thus, fragments of Lynx1 loop II, loop C from the Lymnaea stagnalis AChBP and HAP were synthesized in linear and Cys-cyclized forms and structurally (CD and NMR) and functionally (radioligand assay on Torpedo nAChR) characterized. Connecting the C- and N-termini by disulfide in the ws-Lynx1 fragment stabilized its conformation which became similar to the loop II within the 1H-NMR structure of ws-Lynx1, the activity being higher than for starting linear fragment but lower than for peptide with free cysteines. Introduced disulfides did not considerably change the structure of HAP and of loop C fragments, the former preserving high affinity for α-bungarotoxin, while, surprisingly, no binding was detected with loop C and its analogs.

Bridging of DNA breaks activates PARP2-HPF1 to modify chromatin.

Breaks in DNA strands recruit the protein PARP1 and its paralogue PARP2 to modify histones and other substrates through the addition of mono- and poly(ADP-ribose) (PAR)1-5. In the DNA damage responses, this post-translational modification occurs predominantly on serine residues6-8 and requires HPF1, an accessory factor that switches the amino acid specificity of PARP1 and PARP2 from aspartate or glutamate to serine9,10. Poly(ADP) ribosylation (PARylation) is important for subsequent chromatin decompaction and provides an anchor for the recruitment of downstream signalling and repair factors to the sites of DNA breaks2,11. Here, to understand the molecular mechanism by which PARP enzymes recognize DNA breaks within chromatin, we determined the cryo-electron-microscopic structure of human PARP2-HPF1 bound to a nucleosome. This showed that PARP2-HPF1 bridges two nucleosomes, with the broken DNA aligned in a position suitable for ligation, revealing the initial step in the repair of double-strand DNA breaks. The bridging induces structural changes in PARP2 that signal the recognition of a DNA break to the catalytic domain, which licenses HPF1 binding and PARP2 activation. Our data suggest that active PARP2 cycles through different conformational states to exchange NAD+ and substrate, which may enable PARP enzymes to act processively while bound to chromatin. The processes of PARP activation and the PARP catalytic cycle we describe can explain mechanisms of resistance to PARP inhibitors and will aid the development of better inhibitors as cancer treatments12-16.

Structure of the C9orf72 ARF GAP complex that is haploinsufficient in ALS and FTD.

Mutation of C9orf72 is the most prevalent defect associated with amyotrophic lateral sclerosis and frontotemporal degeneration1. Together with hexanucleotide-repeat expansion2,3, haploinsufficiency of C9orf72 contributes to neuronal dysfunction4-6. Here we determine the structure of the C9orf72-SMCR8-WDR41 complex by cryo-electron microscopy. C9orf72 and SMCR8 both contain longin and DENN (differentially expressed in normal and neoplastic cells) domains7, and WDR41 is a ß-propeller protein that binds to SMCR8 such that the whole structure resembles an eye slip hook. Contacts between WDR41 and the DENN domain of SMCR8 drive the lysosomal localization of the complex in conditions of amino acid starvation. The structure suggested that C9orf72-SMCR8 is a GTPase-activating protein (GAP), and we found that C9orf72-SMCR8-WDR41 acts as a GAP for the ARF family of small GTPases. These data shed light on the function of C9orf72 in normal physiology, and in amyotrophic lateral sclerosis and frontotemporal degeneration.

Ligand-bound glutamine binding protein assumes multiple metastable binding sites with different binding affinities.

Protein dynamics plays key roles in ligand binding. However, the microscopic description of conformational dynamics-coupled ligand binding remains a challenge. In this study, we integrate molecular dynamics simulations, Markov state model (MSM) analysis and experimental methods to characterize the conformational dynamics of ligand-bound glutamine binding protein (GlnBP). We show that ligand-bound GlnBP has high conformational flexibility and additional metastable binding sites, presenting a more complex energy landscape than the scenario in the absence of ligand. The diverse conformations of GlnBP demonstrate different binding affinities and entail complex transition kinetics, implicating a concerted ligand binding mechanism. Single molecule fluorescence resonance energy transfer measurements and mutagenesis experiments are performed to validate our MSM-derived structure ensemble as well as the binding mechanism. Collectively, our study provides deeper insights into the protein dynamics-coupled ligand binding, revealing an intricate regulatory network underlying the apparent binding affinity.

Cryo-EM structure of VASH1-SVBP bound to microtubules.

The dynamic tyrosination-detyrosination cycle of α-tubulin regulates microtubule functions. Perturbation of this cycle impairs mitosis, neural physiology, and cardiomyocyte contraction. The carboxypeptidases vasohibins 1 and 2 (VASH1 and VASH2), in complex with the small vasohibin-binding protein (SVBP), mediate α-tubulin detyrosination. These enzymes detyrosinate microtubules more efficiently than soluble αß-tubulin heterodimers. The structural basis for this substrate preference is not understood. Using cryo-electron microscopy (cryo-EM), we have determined the structure of human VASH1-SVBP bound to microtubules. The acidic C-terminal tail of α-tubulin binds to a positively charged groove near the active site of VASH1. VASH1 forms multiple additional contacts with the globular domain of α-tubulin, including contacts with a second α-tubulin in an adjacent protofilament. Simultaneous engagement of two protofilaments by VASH1 can only occur within the microtubule lattice, but not with free αß heterodimers. These lattice-specific interactions enable preferential detyrosination of microtubules by VASH1.

Efficient screening of ligand-receptor complex formation using fluorescence labeling and size-exclusion chromatography.

Evidence of a complex formation is a crucial step in the structural studies of ligand-receptor interactions. Here we presented a simple and fast approach for qualitative screening of the complex formation between the chimeric extracellular domain of the nicotinic acetylcholine receptor (α7-ECD) and three-finger proteins. Complex formation of snake toxins α-Bgtx and WTX, as well as of recombinant analogs of human proteins Lynx1 and SLURP-1, with α7-ECD was confirmed using fluorescently labeled ligands and size-exclusion chromatography with simultaneous absorbance and fluorescence detection. WTX/α7-ECD complex formation also was confirmed by cryo-EM. The proposed approach could easily be adopted to study the interaction of other receptors with their ligands.

Cryo-EM structures of holo condensin reveal a subunit flip-flop mechanism.

Complexes containing a pair of structural maintenance of chromosomes (SMC) family proteins are fundamental for the three-dimensional (3D) organization of genomes in all domains of life. The eukaryotic SMC complexes cohesin and condensin are thought to fold interphase and mitotic chromosomes, respectively, into large loop domains, although the underlying molecular mechanisms have remained unknown. We used cryo-EM to investigate the nucleotide-driven reaction cycle of condensin from the budding yeast Saccharomyces cerevisiae. Our structures of the five-subunit condensin holo complex at different functional stages suggest that ATP binding induces the transition of the SMC coiled coils from a folded-rod conformation into a more open architecture. ATP binding simultaneously triggers the exchange of the two HEAT-repeat subunits bound to the SMC ATPase head domains. We propose that these steps result in the interconversion of DNA-binding sites in the catalytic core of condensin, forming the basis of the DNA translocation and loop-extrusion activities.