Skip to main content
PMC home page
eLife logoLink to eLife
. 2020 Aug 18;9:e58622. doi: 10.7554/eLife.58622

The dynamic nature of the human origin recognition complex revealed through five cryoEM structures

Matt J Jaremko 1,2,3,†, Kin Fan On 1,2,3,†, Dennis R Thomas 1,3, Bruce Stillman 3, Leemor Joshua-Tor 1,2,3,✉
Editors: Sriram Subramaniam4, John Kuriyan5
PMCID: PMC7467728  PMID: 32808929

Abstract

Genome replication is initiated from specific origin sites established by dynamic events. The Origin Recognition Complex (ORC) is necessary for orchestrating the initiation process by binding to origin DNA, recruiting CDC6, and assembling the MCM replicative helicase on DNA. Here we report five cryoEM structures of the human ORC (HsORC) that illustrate the native flexibility of the complex. The absence of ORC1 revealed a compact, stable complex of ORC2-5. Introduction of ORC1 opens the complex into several dynamic conformations. Two structures revealed dynamic movements of the ORC1 AAA+ and ORC2 winged-helix domains that likely impact DNA incorporation into the ORC core. Additional twist and pinch motions were observed in an open ORC conformation revealing a hinge at the ORC5·ORC3 interface that may facilitate ORC binding to DNA. Finally, a structure of ORC was determined with endogenous DNA bound in the core revealing important differences between human and yeast origin recognition.

Research organism: Human

Introduction

DNA replication is essential to all forms of life. In eukaryotes, replication is initiated from multiple start sites on chromosomes called replication origins (reviewed in Bell and Labib, 2016; Bleichert et al., 2017; Leonard and Méchali, 2013; On et al., 2018; Prioleau and MacAlpine, 2016). The very first step of this initiation process is accomplished by DNA association with the Origin Recognition Complex (ORC), a six-subunit protein that forms a partial ring around origin DNA (Li et al., 2018; Yuan et al., 2017). In Saccharomyces cerevisiae, this is a sequence-specific binding event, however, in metazoans the manner in which origins are identified is not yet clear. The bromo-adjacent homology (BAH)-domain at the N-terminus of ORC1 facilitates binding of ORC to DNA through nucleosomes which contain specific histone modifications that influence ORC1 affinity to the chromatin complex (Eaton et al., 2010; Hossain and Stillman, 2016; Kuo et al., 2012; Müller et al., 2010; Noguchi et al., 2006; Rivera et al., 2014). Once bound to origins, the ORC-related AAA+ ATPase CDC6 enters the complex to complete the ORC-CDC6 ring around the DNA (Yuan et al., 2017). Subsequent recruitment of the licensing factor CDT1 bound to the replicative helicase MCM hexamer complex then leads to the formation of MCM double-hexamers assembled around dsDNA in a head-to-head association (Evrin et al., 2009; Miller et al., 2019; Remus et al., 2009). During origin licensing, ORC must perform several tasks sequentially: bind DNA, find origin sites, and dissociate from DNA upon completion of the MCM double-hexamer loading and assembly. In fact, single molecule studies have shown that ORC is released from DNA immediately following stable association of two MCM molecules with the DNA substrate (Ticau et al., 2015). These dynamic events likely require significant conformational changes in ORC.

High resolution structures of ORC from Saccharomyces cerevisiae (ScORC), Drosophila melanogaster (DmORC), and Homo sapiens (HsORC) have revealed the detailed architecture of the complex (Bleichert et al., 2015; Li et al., 2018; Tocilj et al., 2017; Yuan et al., 2017). The complex consists of six subunits (ORC1-6), with subunits ORC1-5 forming a partial ring capable of encircling DNA (Li et al., 2018; Tocilj et al., 2017; Yuan et al., 2017). All five subunits contain a winged-helix domain (WHD) and a AAA+ domain which consists of the RecA-fold and lid domains (Figure 1a), although the ORC2 and ORC3 RecA-folds have diverged slightly and do not contain a lid domain. Adjacent ORC subunits associate in a double-layer arrangement through interfaces between the WHDs in one layer and the AAA+ domains in another with ATP-binding sites sandwiched between the ORC1·4, ORC4·5, and ORC5·3 AAA+ interfaces (Figure 1g; Tocilj et al., 2017). The ORC1·4 and ORC4·5 sites represent canonical AAA+ ATPase sites (Enemark and Joshua-Tor, 2008; Erzberger and Berger, 2006), while the ORC5·3 interface is considerably more flexible relative to the other ATP interfaces. Of the three ATP-binding sites, only the ORC1·4 site exhibits ATPase activity and this activity is essential for cell viability (Bowers et al., 2004; Klemm et al., 1997; Randell et al., 2006; Tocilj et al., 2017). In addition, several mutations at the ORC1·4 ATPase site are associated with the developmental disorder Meier-Gorlin Syndrome (Bicknell et al., 2011a; Bicknell et al., 2011b; de Munnik et al., 2012a; de Munnik et al., 2012b; Guernsey et al., 2011). However, the functional role for ATP hydrolysis is still unknown.

Figure 1. Overall architecture of human ORC.

(a) Cartoon representation of the general structure of an ORC subunit with a winged-helix [WH], and a AAA+ domain containing a lid and RecA-fold, as illustrated by ORC1 from ORCO1AAA. Human ORC structures were determined in five conformations: (b) ORC2-5, (c) ORCO2WH, (d) ORCO1AAA, (e) ORCOPEN, and (f) ORCDNA. Each panel contains a density map and model and is color coded by subunit as illustrated in panel (b). ATP carbon atoms are colored in black and the remaining atoms, including Mg2+ (green spheres), are colored by the CPK and Jmol element colors. (g) Close-up of the three ORC ATP sites in ORCO1AAA with density maps overlaying the model. Residues important for coordination of the ATP molecules and Mg2+ are labelled. DNA is colored red/grey and all displayed atoms are colored by the CPK and Jmol element colors.

Figure 1.

Figure 1—figure supplement 1. ATP sites of each ORC conformation.

Figure 1—figure supplement 1.

Close-up of the three ORC ATP sites in (a) ORCO2-5, (b) ORCO2WH, (c) ORCOPEN, and (d) ORCDNA with mesh density maps overlaying the model. Residues important for coordination of the ATP molecules and Mg2+ (shown as green spheres) are labelled.

Figure 1—figure supplement 2. CryoEM workflow of ORC1-5.

Figure 1—figure supplement 2.

Motion correction, averaging, CTF estimation, and particle picking were performed on movies real-time via the WARP software package. To determine an ab initio reconstruction, 2D classification and ab initio reconstruction was performed in Cryosparc v2. Motion correction and CTF estimation were also performed in Relion to allow convenient use of the Bayesian Polishing feature in the final post-processing Relion steps. The particle coordinates from WARP and the ab initio map from Cryosparc were both utilized in an initial 3D classification in Relion. All further processing steps were performed in Relion, including global and focused refinement and classification. The in silico purification resulted in four overall maps: ORC(O2WH) (3.2 Å), ORC(O1AAA) (3.7 Å), ORCOPEN (4.0 Å), and ORCDNA (4.3 Å). The mask used for focused 3D classification of the ORC2 WHD is highlighted in red. The focused maps of the ORC2 WHD focused classification are expanded to show two alternative conformations of the domain. See Methods section for details.

Figure 1—figure supplement 3. CryoEM workflow of ORC2-5.

Figure 1—figure supplement 3.

The workflow of the ORC2-5 data collection was the same as the ORC1-5 data collection (see Figure 1—figure supplement 1). The processing resulted in a 3.5 Å map (ORCO2-5). The mask used for focused 3D classification of the ORC2 WHD is highlighted in red.

Figure 1—figure supplement 4. CryoEM map resolutions of ORC1-5.

Figure 1—figure supplement 4.

(a) Fourier Shell Correlation of the ORC1-5 global maps ORC(O2WH) (blue), ORC(O1AAA) (red), ORCOPEN (green), and ORCDNA (yellow). Solid lines and dotted lines refer to the masked and unmasked FSCs, respectively. (b–e) Resolutions maps were calculated using ResMap (Kucukelbir et al., 2014) for ORC(O2WH) (b), ORC(O1AAA) (c), ORCOPEN (d), and ORCDNA (e). (f–i) Euler angle distribution plots of the ORC(O2WH) (f), ORC(O1AAA) (g), ORCOPEN (h), and ORCDNA (i) particles used in the final reconstruction maps.

Figure 1—figure supplement 5. CryoEM map resolution of ORC2-5.

Figure 1—figure supplement 5.

(a) Fourier Shell Correlation of the ORC2-5 global map ORCO2-5. Solid lines and dotted lines refer to the masked and unmasked FSCs, respectively. (b) Resolutions maps were calculated using ResMap (Kucukelbir et al., 2014) for ORCO2-5. (c) Euler angle distribution plots of the ORCO2-5 particles used in the final reconstruction maps.

Figure 1—figure supplement 6. 3DFSC analysis of the ORC conformations.

Figure 1—figure supplement 6.

The histograms and directional FSC plots for (a) ORCO2-5, (b) ORCO2WH, (c) ORCO1AAA, and (d) ORCOPEN, and (e) ORCDNA are plotted. The sphericity and global resolution values are reported within the graph windows.

Figure 1—figure supplement 7. Resolution improvement of ORC2 and ORC3 near the ORC gap opening of ORCO1AAA.

Figure 1—figure supplement 7.

Global (a) and focused (b) refinement of ORCO1AAA highlighting the regions of ORC2 and ORC3 near the ORC gap opening.

Figure 1—figure supplement 8. The intertwine between ORC subunits 2 and 3.

Figure 1—figure supplement 8.

The structure (a) and map (b) of human ORCO1AAA and the structues of (c) ScORC, and (d) DmORC are oriented to illustrate the tight interaction between subunits 2 and 3. ORC2 contains a N-terminal helix that positioned between ORC3 and ORC5, and ORC3 contains an N-terminal β-strand that forms a β-sheet with ORC2 β-strands.
Open in a new tab

Based on the previous structures, ORC exhibits two main conformations differing primarily in the positions of the ORC1 AAA+- and ORC2 WH-domains: an open state (ScORC and HsORC) (Li et al., 2018; Tocilj et al., 2017; Yuan et al., 2017) and a closed, auto-inhibited state (DmORC) (Bleichert et al., 2015). In the open state, the ORC1 AAA+-domain binds to the ORC4 AAA+-fold to form the ORC1·4 ATPase active site and ORC2 WHD is adjacent to the ORC3 WHD. The ring-shaped core of ORC in the open state is accessible to solvent and DNA substrates. In the closed state, the ORC1 AAA+ motif is dramatically rotated by ~80° towards the ORC2 WHD position and ORC2 WHD is collapsed into the open ring-shaped core of the complex. The collapsed ORC1 AAA+-domain and ORC2 WHD in the closed state completely obstruct the core of ORC and could not accommodate DNA. In addition, the ORC1·4 AAA+ interface is disrupted and, therefore, the main ATPase site is disrupted. Initially, the closed state was only seen in DmORC, but recent cryoEM studies on HsORC suggested that the ORC1 AAA+ domain also adopts the auto-inhibited conformation (Bleichert et al., 2018). Further structural analysis is necessary to determine the prevalence and function of the auto-inhibited state in other species.

As mentioned, origin recognition, which is conferred by ORC binding to DNA, differs between yeast and metazoans. In the yeast Saccharomyces cerevisiae, ORC specifically recognizes and binds to a class of DNA sequences, called the autonomously replicating sequences (ARSs), which act as the replication initiation sites in vivo (Bell and Stillman, 1992; Brewer and Fangman, 1987; Eaton et al., 2010; Huberman, 1987; Linskens and Huberman, 1988; Marahrens and Stillman, 1992; Rao and Stillman, 1995; Theis et al., 1999). Previous ScORC structures revealed multiple regions of Orc1, Orc2, Orc4 and Orc5 that interact with double stranded DNA, including a lysine-rich region in an unstructured domain of Orc1, as well as two motifs within the AAA+ domain in each subunit that interact with elements of ARS DNA: the initiation-specific motifs (ISM) in the AAA+-domains and β-hairpin wings in the WHDs (Bleichert et al., 2017; Li et al., 2018; On et al., 2018; Yuan et al., 2017). Of particular importance are the ISM of Orc2 and an α-helix inserted into the β-hairpin wing of Orc4. The Orc2 ISM binds to the DNA minor groove with the W396 sidechain tucked into the hydrophobic minor groove at the T-rich region within the ARS consensus DNA sequence (Li et al., 2018). Interestingly, the α-helix insertion in the β-hairpin wing of Orc4 extends deep into the DNA major groove and likely plays an important role in DNA sequence-specific interactions with the ARS sequence (Hu et al., 2020). The β-hairpin wing of Orc4 is unresolved in the DmORC and HsORC structures and the importance of this region is unknown in metazoans. In metazoans, notably, ORC lacks sequence specificity for origin recognition and may rely on interactions with nucleosomes to establish DNA origin sites (Noguchi et al., 2006; Remus et al., 2004; Vashee et al., 2003).

In this study, we have determined cryoEM structures of the human Origin Recognition Complex in various conformational states and at higher resolution than previous studies. Four cryoEM maps revealed dynamic movements of ORC without DNA. The ORC1 AAA+ domain and the ORC2 WHD seem to repel each other to create dynamic movement between two extreme states. In addition, a hinge was identified at the ORC3·5 interface that creates prominent twist and pinch movements of the entire complex. Lastly, a structure of ORC bound to co-purified, endogenous DNA (ORCDNA) was determined that revealed several ORC regions in close contact with DNA. The ORC2 WH- and ORC1 AAA+-domains are both visible in this ORC conformation and bind DNA directly. The various ORC conformations provide insight into the dynamics of ORC during DNA replication initiation.

Results and discussion

Overall HsORC structure and comparison to ORC from Saccharomyces cerevisiae and Drosophila melanogaster

Two HsORC constructs were generated for recombinant expression based on our previous study (Tocilj et al., 2017) with the following changes: ORC2 was extended to full-length, the StrepTag was moved from the N-terminus of ORC1 to the N-terminus of ORC3 to improve sample purity during affinity purification, and the Sumo tag on ORC1 was removed. One construct consisted of full-length ORC1, but after expression and purification, ORC1 was absent yielding a complex of ORC2-5 (henceforth labelled ORCO2-5). Another construct included a truncated version of ORC1 [aa471-861] that yielded a purified complex of ORC1-5 with multiple conformations. We utilized cryo-electron microscopy (cryoEM) to investigate ORC due to the various conformations and potential flexibility seen in previous ORC structures (Bleichert et al., 2015; Li et al., 2018; Tocilj et al., 2017; Yuan et al., 2017). The cryoEM analysis lead to a 3.5 Å structure of the ORC2-5 complex (ORCO2-5) and four distinct structures of the ORC1-5 complex: ORCO1AAA (3.2 Å), ORCO2WH (3.7 Å), ORCOPEN (4.0 Å), and ORCDNA (4.3 Å) (Figure 1—figure supplement 1–4 and Supplementary file 1). The four structures determined from ORC1-5 illustrate the flexibility and dynamic movements of ORC when ORC1 is present in the complex.

The ORCO2-5 structure that lacks the ORC1 subunit consists of subunits ORC2-5 in a compact and substantially closed conformation (Figure 1b). The ORC2 WHD of ORCO2-5 is collapsed into the ring-shaped core of the complex with the β-hairpin wing facing the ORC4 WHD (Figure 2c). The ORC2 WHD is significantly deeper in the core and rotated ~135° compared to the auto-inhibited state of ORC2 WHD in the DmORC structure (Figure 2d and Video 1; Bleichert et al., 2015). Once ORC1 is bound to form ORC1-5, the complex opens up into several related conformations to accommodate ORC1 (Figure 1c–f). The first of these, ORCO2WH, consists of a dynamic ORC1 AAA+ domain that is unresolved in the refined map (Figure 1c). Interestingly, in this structure, the ORC2 WHD was stably bound inside the ring-shaped core of ORC in a conformation that is similar to the auto-inhibited state observed in the DmORC structure (Figure 2d,f; Bleichert et al., 2015). In contrast to ORCO2WH, ORCO1AAA consists of a dynamic ORC2 WHD that is unresolved in the refined map (Figure 1d). The ORC1 AAA+ domain in this structure is ordered and forms the important interface with ORC4 with an ATP coordinated between the two subunits, an ATPase active conformation seen in previous ScORC and HsORC structures (Li et al., 2018; Tocilj et al., 2017; Yuan et al., 2017). The flexibility of ORC2 WHD is also observed in ScORC and in the DNA-bound ScORC, where the domain was located in several states proximal to DNA (Li et al., 2018), and different from its position in the yeast OCCM structure (Yuan et al., 2017). Thus, the dynamic movements of the domain are prevalent across species and likely influence binding to DNA and CDC6. The subunits of the third conformational state of the ORC1-5 structure, ORCOPEN, are arranged in a more open configuration relative to the previous two structures where the overall spiral of the complex is widened (Figure 1e and Figure 3a). The conformation of ORCOPEN resembles the ScORC conformation bound to Cdc6 (Yuan et al., 2017). It was noted previously from low resolution cryo-EM structures comparing ScORC on DNA with ScORC-Cdc6 on DNA that substantial changes in the location of parts of the Orc1 subunit occurred upon binding of Cdc6 to ORC (Sun et al., 2012). Finally, the cryoEM map of ORCDNA consists of DNA bound in the core of the complex (ORC1-5) with several ORC subunits making contacts to the DNA (Figure 1f, Figure 4). The ORC1 AAA+ and ORC2 WHD of ORCDNA most likely envelop DNA predominantly through electrostatic interactions. The ORCOPEN and ORCDNA maps suffered from the existence of a preferred orientation which likely contributed to the lower resolutions of these maps (Figure 1—figure supplements 4 and 6). In addition, the nominal resolution of these maps, determined from their optimal direction is higher than their overall appearance merits. Nonetheless, these maps provided valuable information about the architecture and dynamics of the complex (see below).

Figure 2. Dynamic states of ORC1 AAA+ and ORC2 WH domains.

3D classification ORCO2WH was performed with the density map low-pass filtered to 20 Å and an E-step resolution limit of 20 Å. Two 3D classes from ORCO2WH contained densities for ORC1 AAA+ domain. The ORC1 AAA+ domain was fit into these densities which revealed the domain approaching the ORC1·4 position (a) and the auto-inhibited position (b). (c-f) In the ORC core, the ORC2 WHD was observed in two conformations: an alternate state, called the inactive state (c, e), and the auto-inhibited state (f), similar to the DmORC2 WHD (PDB ID: 4xgc) (d). The ORCO2WH inactive state (e) was determined by focused classification of the ORCO2WH with a tight mask around ORC2 WH.

Figure 2.

Figure 2—figure supplement 1. The dynamic states of ORCO2WH and ORCO1AAA.

Figure 2—figure supplement 1.

3D classification of ORCO1AAA (a) and ORCO2WH (b) were performed with each map low-pass filtered to 20 Å and an E-step resolution limit of 20 Å. Two 3D classes from ORCO1AAA (left) illustrate dynamic movements of ORC2 WHD in place, while two 3D classes from the ORCO2WH (right) illustrate dynamic movements of ORC1 AAA+ approaching the ORC1·4 position and the auto-inhibited position.
Open in a new tab

Figure 3. Multibody refinement of ORCO1AAA.

(a) Overlay of ORCOPEN and ORCO1AAA illustrates a large conformational change of ORC2 and ORC3. ORC2·3·(5 WH) and ORC1·4·(5 AAA+) were designated as two bodies for Relion’s multibody refinement. (b) Principal component analysis revealed two prominent movements of ORCO1AAA highlighted in red and blue. (c) Component 1 undergoes twist movement of ORC2·3 across the ORC1 AAA+ domain. (d) Component 2 undergoes a pinch movement of ORC2·3 towards the ORC1 AAA+ domain. Multibody Refinement results of ORCO2WH and ORCOPEN are available in Figure 3—figure supplement 1. To validate the Multibody Refinement, Cryosparc’s 3D Variability analysis was performed and yielded the same prominent movements. The movements from Multibody Refinement and 3D variability can be visualized in Video 3.

Figure 3.

Figure 3—figure supplement 1. Relion’s Multibody Refinement (Nakane et al., 2018) of ORC conformations.

Figure 3—figure supplement 1.

Principal component analysis results for, ORCO2WH (a), and ORCOPEN (b). The most prominent movements for each conformation are twist and pinch motions.
Figure 3—figure supplement 2. Extension of ORC3 N-terminal α-helix in ORCOPEN.

Figure 3—figure supplement 2.

The N-terminal region of ORC3 in the ORCOPEN structures extends into a long α-helix compared to the other ORC conformations. The long α-helix likely contributes to the open conformation of the structure.
Open in a new tab

Figure 4. Several regions of ORC contact the endogenous DNA in ORCDNA.

(a) ORC subunits 1–5 contain regions with positively-charged residues proximal to the DNA. (b) The ORC1 RecA-fold contains several positively-charged residues in close proximity to DNA. (c) The ORC2 WHD, shown as a density map due to the lower resolution of this domain, moves to interact with the DNA in the core of the complex. (c) The β-hairpin loops of ORC4 and ORC5 contain positively-charged residues proximal to the DNA. The DNA depicted consists of non-specific sequence. (d) The ORC4 β-hairpin loop of S. cerevisiae contains an α-helix that inserts into the major groove of DNA making sequence-specific interactions with DNA bases. (e) The β-hairpin loop in ORCDNA does not contain an α-helix. However, ORC4 R390 in humans may interact with the DNA backbone similar to ORC4 R478 in S. cerevisiae. (f) The superimposition of the ScORC and HsORC β-hairpin loops illustrates the significant difference between the regions.

Figure 4.

Figure 4—figure supplement 1. DNA density observed in the ORC core.

Figure 4—figure supplement 1.

Close-up of the DNA in ORCDNA with mesh density map overlaying the structure from the DNA axis (a) and side view (b). The top panel shows the mesh density map overlaying the structure and the bottom panel shows the surface density map with DNA in red and subunits in close proximity highlighted.
Figure 4—figure supplement 2. Endogenous Sf9 DNA co-purifies with ORC1-5.

Figure 4—figure supplement 2.

(a) After strep-tag purification of ORC1-5, the sample was applied to a 1% agarose gel. The ORC sample contained DNA of less than 0.5 kb. (b) Post 2D classification of the ORCDNA particle population. A strong signal is observed in the core of the complex.
Figure 4—figure supplement 3. The DNA in the core of ORCDNA adopts multiple conformations.

Figure 4—figure supplement 3.

Further 3D classification of ORCDNA particles revealed the DNA in multiple conformations in the core.
Figure 4—figure supplement 4. ORC1 sequence alignment with a focus on the region in close proximity to DNA in the ORCDNA structure.

Figure 4—figure supplement 4.

Sequences are ordered by similarity to H. sapiens. Positively-charged residues in close proximity to DNA in the human ORCDNA structure are highlighted in yellow. The sequence alignment was generated using MUSCLE (Edgar, 2004).
Figure 4—figure supplement 5. ORC4 sequence alignment with a focus on the β-hairpin loop region near DNA.

Figure 4—figure supplement 5.

Sequences are ordered by similarity to H. sapiens. The β-hairpin loop region is highlighted in yellow. ORC4 R390 in humans may interact with the DNA backbone similar to ORC4 R478 in S. cerevisiae. These arginines are highlighted in red. The sequence alignment was generated using MUSCLE (Edgar, 2004) and then structurally aligned.
Open in a new tab

Video 1. The conformations of ORC2 WHD in ORCO2-5 and ORCO2WH.

Download video file (3.7MB, mp4)
Open in a new tab

The ORC2 WHD in ORCO2WH is in a similar conformation to the autoinhibited state observed in DmORC (Bleichert et al., 2015). The ORC2 WHD must undergo a significant ~135°rotation to bind in the conformation observed in ORCO2-5.

Three ATP nucleotides were well resolved in the ORC maps (Figure 1g and Figure 1—figure supplement 1), consistent with previous observations (Tocilj et al., 2017). Although not provided in the sample buffer, magnesium ions were also resolved in the maps and were likely retained from the protein expression lysate. Only in the ORCO2WH structure was an ATP absent at the ORC1·4 site due to the displacement of the ORC1 AAA+-domain (Figure 1c, Figure 2a,b and Video 2) and consequential disruption of the ATP binding site. The architectures of the ORC1·4 and ORC4·5 sites are characteristic of AAA+ ATPase proteins, where the Walker-A and Walker-B motifs of the RecA-fold of one subunit coordinate the ATP molecule on one face and three conserved basic residues, one from the same subunit and two from the other subunit coordinate the other face (Figure 1g). At the ORC1·4 interface, for example, these three basic residues are R720 of the ORC1 lid domain, acting as the trigger (or sensor-II), the ORC4 finger R209, and the ORC4 tether R205 coordinating the other face (Figure 1g). As mentioned, the only site that has been shown to have ATPase activity is the ORC1·4 site, and mutations at this site are associated with the developmental disorder Meier-Gorlin Syndrome (Tocilj et al., 2017). In contrast to ORC1·4 and ORC4·5, the ORC5·3 interface appears considerably more flexible and, in fact, ORC3 does not seem to participate in ATP coordination (Figure 1g). Nonetheless, we do observe density for an ORC3 region proximal to the ATP site in low-pass filtered maps which indicate dynamic movements of the unresolved residues 86–94.

Video 2. The dynamic conformations of ORC1 AAA+ in ORCO2WH.

Download video file (3.2MB, mp4)
Open in a new tab

A low-pass filtered 3D classification of ORCO2WH revealed an ORC1 AAA+ position near the ORC1·4 interface that forms the ATP binding site and an ORC1 AAA+ position near the auto-inhibited state disrupting the ORC1·4 ATP binding site.

The structures of ORC2, ORC3, and the WHD of ORC5 have significantly improved resolution compared to the previous published HsORC structure (Tocilj et al., 2017). ORCO1AAA was especially improved in this region due to the 3D focused refinement we applied (Figure 1—figure supplement 7). The close intertwine between ORC2 and ORC3 in ScORC and DmORC structures is also apparent in the structures of HsORC presented here (Figure 1—figure supplement 8). The N-terminal regions of ORC2 and ORC3 wrap around each other contributing to their tight binding. In fact, ORC2 and ORC3 localize to centromeres in mitosis and may function independently of the other ORC subunits (Craig et al., 2003; Prasanth et al., 2004; Prasanth et al., 2010). The solvent-accessible surface area buried between ORC2 and ORC3 (3296 Å2) is significantly larger than the next closest interacting subunits ORC4 and ORC5 (2707 Å2). In the ScORC structure, the ORC2 N-terminal region wraps around the WHD of ORC3 and forms a helix on the opposite end (Li et al., 2018). In the HsORC structures, the ORC2 N-terminal helix (aa239-249) is present in the ORCO1AAA and ORCO2WH maps, while the loop (aa250-267) wrapping around ORC3 is dynamic and thus unresolved (Figure 1—figure supplement 8). In addition, the ORC3 N-terminal region (residues 10–14) loops around the RecA-like fold of ORC2 and forms a β-strand that adheres to one of the ORC2 RecA-like β-strands (aa335-339).

Dynamic states of ORC1 AAA+ and ORC2 WH

The ORCO1AAA and ORCO2WH structures illustrate the dynamic movements of ORC1 AAA+ and ORC2 WH domains relative to each other (Figure 1c,d). Other than the ORC1 AAA+ and ORC2 WHD, the architectures of the two conformations are similar. In both maps, density for the missing domains can be seen in 3D classifications with a 20 Å low-pass filtered template and E-step resolution limit of 20 Å (Figure 2a,b, Figure 2—figure supplement 1 and Video 2). In the ORCO1AAA structure, the ORC2 WHD undergoes conformational changes but does not change its overall position in the complex (Figure 2—figure supplement 1a). In contrast, in the ORCO2WH structure, the ORC1 AAA+ rotates ~90° from a position proximal to the ORC4 AAA+ domain where it completes the major ATP binding site in the ORCO1AAA structure, to a position proximal to the ORC1 and ORC5 WHDs (Figure 2a,b, Figure 2—figure supplement 1 and Video 2), disrupting the ORC1·4 ATP binding site. The latter position resembles the auto-inhibited state observed in the DmORC structure (Bleichert et al., 2015). This state was also observed in 2D classifications of the human ORC (Bleichert et al., 2018). While ORC1 AAA+ is locked in the auto-inhibited state in DmORC, the ORC1 AAA+ domain in human ORC appears to be dynamic along a continuum of positions between two extreme states: the auto-inhibited state observed in the DmORC and when it forms the ORC1·4 interface with ORC4 as observed in ORCO1AAA. As mentioned, the ATPase site between the ORC1 and ORC4 is the only site that exhibits activity and mutations at the site are associated with the developmental disorder Meier-Gorlin Syndrome, but the functional role of ATP hydrolysis is unknown. The ORC1 AAA+ domain’s dynamic movements may be a consequence of ORC1·4 ATP hydrolysis, but further investigation is necessary. On the other hand, the ORC4·5 and ORC5·3 subunit interfaces are stable, and do not appear to be disrupted, which may be a consequence of their inability to hydrolyze ATP. Based on this idea and previous studies using mutant ORC4 and ORC5 subunits (Siddiqui and Stillman, 2007), we conclude that ATP promotes the assembly and stability of the human ORC, which, unlike yeast ORC, is assembled and disassembled throughout the cell division cycle. Overall, the ORCO1AAA and ORCO2WH populations reveal a dynamic switch between the ORC1 AAA+ and ORC2 WHD where the locking into position of one domain within the ORC ring-shaped core kicks the other domain into a dynamic state. These domains are accommodated simultaneously only in the ORCOPEN and ORCDNA complexes, two conformations that separate the ORC1 AAA+ and ORC2 WH domains by widening the gap between ORC1 and ORC2 and by mutual electrostatic attraction to DNA, respectively (see below).

A minor ORC2 WHD conformation was observed in further 3D classifications of the ORCO2WH particles (Figure 2e). Secondary structure elements can be seen in the ORC2 WHD region that are clearly different from the auto-inhibited conformation. In fact, this conformation resembles the completely collapsed ORC2 WHD conformations in the ORCO2-5 structure where the domain is further tucked into the core of the complex compared to the auto-inhibited state (Figure 2c). Although the density region of the ORC2 WHD alternative state in ORCO2WH is of low resolution, we were confidently able to fit this region once we modelled the ORC2 WHD inactive state in the well-resolved ORCO2-5 structure (Figure 2c). Notably, this alternative conformation does not resemble any of the conformations of ORC2 WHD in any previously reported structures of the various ORCs (Bleichert et al., 2015; Bleichert et al., 2018; Li et al., 2018; Tocilj et al., 2017; Yuan et al., 2017). The buried ORC2 WHD appears to stabilize ORC2-5 complex prior to ORC1 binding, hence we termed this structure of the ORC2 WHD the inactive state (Figure 1b and Figure 2c,e). The binding of ORC1 seems to facilitate the release of the O2 WHD from this tucked-in state, generating a more labile structure for downstream events. In both the yeast OCCM and ORC-DNA structures (Li et al., 2018; Yuan et al., 2017), the ORC2 WHD interacts with the origin DNA and is far away from the gap between the ORC1 and ORC2 AAA+ domains, forming a collar of ORC WHDs that surround the DNA. Thus the movement of the ORC2 WHD seems to be a universal feature of controlling DNA access to the ring-shaped core of ORC and hence binding to DNA.

Hinge movement arising from the ORC3·5 interface

The ORCOPEN structure displays a conformation similar to what was reported previously for HsORC and ScORC (Figure 1e). When aligned to ORCO1AAA, a hinge in ORCOPEN is evident and originates from the somewhat loose ORC3·5 interface, allowing the ORC2·3 subunits (including the ORC5 WH) to rotate (22° rotation angle) away from the ORC1·4·5 motor module in a hinge motion (Figure 3a). In fact, these two portions of ORC can exist as two separate stable complexes, as evident by the isolation of these stable sub-complexes previously (Tocilj et al., 2017). In addition, the ORC3 N-terminal region of ORCOPEN consists of an extended, straightened α-helix (aa42-86) that is bent in the other ORC conformations by 32° (Figure 1e, Figure 3—figure supplement 2). This long helix in ORCOPEN may provide structural stability to the open conformation. The ORC3 straight helix is also present in the ScORC structure (Li et al., 2018; Yuan et al., 2017), while the helix is bent in the DmORC structure (Bleichert et al., 2015). Interestingly, both the ORC1 AAA+ and O2 WHD are ordered in the open conformation. The extra space in the ORC ring opening reduces clashes between the ORC2 WHD and ORC1 AAA+ domain and allows both domains to adopt a fixed orientation relative to their mutually-exclusive static positioning observed in the more closed ORCO1AAA and ORCO2WH structures (Figure 2a,b, Figure 2—figure supplement 1).

To assess the prominent movements around the ORC3·five interface, we performed multibody refinement in Relion on the ORCO1AAA, ORCO2WH, and ORCOPEN particle populations (Figure 3b–d and Video 3). Subunits ORC2·3 (including ORC5 WHD) and ORC1·4·5 of HsORC were designated as two separate masked bodies in Relion's multibody refinement to determine their independent movements (Nakane et al., 2018) (ORC1 was isolated as a third masked body in multibody refinement, but the experiment did not provide conclusive results, most likely due to the relatively small size of ORC1). The components of the masked bodies and ORC movements were further validated by performing Cryosparc’s unbiased 3D Variability Analysis on the ORCO1AAA particles (Video 3; Punjani and Fleet, 2020). Compared to Multibody Refinement, 3D Variability Analysis does not rely on user input to determine regions that move independently of one another (Punjani and Fleet, 2020). Both Multibody Refinement and 3D Variability Analysis revealed the same prominent movements in the ORCO1AAA particle population. For all three ORC populations, multibody refinement generated two components (specific movements of ORC) that contributed significantly more to the overall variance (total movement of ORC) (Figure 3—figure supplement 1b and Video 3). These movements contributed between 17% to 23% of the total variance of each particle population. The two movements can be described as twisting and pinching, respectively, both emanating from the hinge at the ORC3·5 interface (Figure 3c,d and Video 3). The ORC twisting motion involves ORC1 and ORC2 sliding over each other, while the pinching motion involves ORC1 and ORC2 narrowing/widening the ORC ring opening between them (Figure 3c,d and Video 3). The twisting and pinching movements provide potential flexibility to ORC during replication initiation events, including DNA binding, movement along DNA, and/or complex release after MCM double hexamer formation.

Video 3. The ORC3·5 interface acts as a flexible hinge.

Download video file (1.1MB, mp4)
Open in a new tab

Relion’s Multibody Refinement and Cryosparc’s 3D Variability Analysis revealed the same two prominent movements which resemble a twist and pinch motion, respectively.

In the global refinement of ORC, the ORC2·3 regions lining the gap with ORC1 suffered the most in terms of alignment. In refinements, the center of mass for alignments is located on ORC1·4·5 and, since ORC2·3 moves independently of ORC1·4·5, the edges of ORC2 and ORC3 were misaligned. However, the multibody/focused refinement significantly improved the ORC2·3 region of the map (Figure 1—figure supplement 7). The N-terminus of ORC3, which wraps around ORC2 and extends the β-sheet of the RecA-like fold of ORC2 with an additional strand, was clearly resolved in the focused-refined map. In addition, the density of an ORC2 loop region (residues 361–368), which comes into close proximity with bound DNA in the ORCDNA map (see below), is significantly improved. The improved density maps of ORC further validates the multibody refinement parameters and, therefore, the flexibility at the ORC3·5 interface.

DNA bound to HsORC

One class from the ORC1-5 sample contained clear density for DNA in the core of the complex and yielded a 4.3 Å map into which DNA could be fit into the ORC1-5 structure (ORCDNA) (Figure 1f, Figure 4, Figure 4—figure supplement 1). DNA was detected in the affinity-purified, recombinant HsORC expressed in Sf9 insect cells and the presence of this DNA species was only detected when ORC was expressed (Figure 4—figure supplement 2a). Furthermore, a post 2D classification of the DNA-bound HsORC yielded classes with strong density in the core of the complex (Figure 4—figure supplement 2b). The bright appearance of this density in the 2D classification likely represents the heavy phosphorous atoms in the DNA backbone since heavy atoms scatter electrons to a greater extent and to higher angles in TEM (Kellenberger et al., 1986). Since DNA-binding of HsORC is not sequence specific, the DNA is likely to have multiple orientations along its axis in the core, consistent with further 3D classification of the DNA-bound ORC populations (Figure 4—figure supplement 3). In other words, the DNA fragment visualized is a result of a superposition of sequences along the DNA fragment. The overall conformation of ORCDNA resembles the more closed conformations of ORCO1AAA and ORCO2WH, albeit with ORC2 WHD in a different conformation (see below). The structure shows DNA contacts by all five subunits of ORC, with the most significant interactions occurring with the ORC1 RecA-fold and ORC2 WHD (Figure 4a–c). In the ORC1 RecA-fold, two loops (residues 593–596 and 626–630, respectively) at the N-terminal regions of two α-helices are in close contact to the DNA. Importantly, there are several positively-charged and aromatic residues in the loops (K591, N595, H596 W626, H628, K629) that appear to be involved in binding the negatively-charged DNA backbone (Figure 4b). Sequence alignments show that while most of these residues are conserved in vertebrate species, they differ in insects, plants and fungi (Figure 4—figure supplement 4). The ORC2 WHD is also in close contact with the DNA and this domain undergoes significant secondary structure rearrangement, which prevented precise fitting of modelled ORC2 WHD into the region, and the resolution is not high enough for unambiguous model building of this region (Figure 4c). The ORC2 WHD may adopt multiple conformations in this structure similar to the multiple ORC2 WHD states observed in DNA-bound ScORC (Li et al., 2018). Nevertheless, clear secondary structure is visible in the ORC2 WHD region of the map, in which, a long helix has significant contact with the DNA density (Figure 4c). In addition, the initiation specific motif (ISM) of ORC2 (residues 364–368) in the RecA-like fold is near the DNA substrate. Specifically, the positively-charged ORC2 residue R367 emanating from this loop, appears to mediate the interaction with DNA. The ORC2 ISM in humans may play an important role in binding DNA similar to the ORC2 ISM in ScORC where W396 is buried in the minor groove of DNA (Li et al., 2018; Yuan et al., 2017). The β-hairpin loop in ORC5 of HsORC also consists of a positively-charged residue H398 in close proximity to DNA (Figure 4a,f). Similarly, the β-hairpin loop in ORC5 of ScORC contains positively-charged residues (N438, K439, N440, K447, R449) extending into the major groove of DNA (Li et al., 2018).

The Orc4 WHD plays an important role in origin-specific binding in S. cerevisiae (Hu et al., 2020; Li et al., 2018; Yuan et al., 2017). In S. cerevisiae, this domain contains an α-helix insertion in the β-hairpin wing (residues 481–489) that inserts into the major groove of the ARS-sequence DNA and is a key contributor to conferring sequence specificity for origin recognition (Figure 4d). The homologous ORC4 region in HsORC is visible in the ORCO1AAA cryoEM density map allowing us to now model this region (Figure 4e). This region in HsORC is 20 residues shorter than the ScORC segment and does not contain an α-helix (Figure 4f and Figure 4—figure supplement 5; Hu et al., 2020). The missing helix in HsORC likely reduces its ability to bind specifically to a DNA sequence resulting in the requirement of other mechanisms to establish stable origin interactions. This loss of specificity also contributes to the multiple conformations of DNA observed in the core of human ORC (Figure 4—figure supplement 3). Despite the missing helix, the shorter ORC4 loop contacts the DNA in the ORCDNA structure proximal to R390 (Figure 4e), a positively-charged residue that may participate in DNA-backbone interactions similar to R478 in ScORC4 (Figure 4d). Based on sequence alignments (Figure 4—figure supplement 5), the ORC4 helix insertion has likely evolved in a small clade of Saccharomyces-related budding yeast species in tandem with a loss of RNA interference (RNAi)-mediated transcriptional gene silencing, while the region is absent in most other eukaryotic organisms, including fungi, plants, and animals, which use RNAi for silencing (Hu et al., 2020). Since HsORC does not display DNA sequence specific DNA binding (Vashee et al., 2003; Vashee et al., 2001), the significant difference in ORC4 between budding yeast and human ORC contributes to the sequence-dependent and -independent origin binding in these species (Figure 4f). HsORC binds to specific locations in chromosomes (Miotto et al., 2016) and unlike ScORC, it is most likely targeted to these sites by binding to specific histones and their modifications (Hossain and Stillman, 2016; Kuo et al., 2012; Long et al., 2020).

Concluding remarks

DNA replication initiation is a highly dynamic, multi-complex process that requires essential coordination by ORC. Our findings uncover several dynamic conformations of HsORC which have implications in facilitating initiation events including DNA engagement, CDC6 recruitment, and MCM complex assembly onto DNA. We determined four structures of ORC that demonstrate the previously unappreciated flexibility of the complex. The ORCO1AAA and ORCO2WH consists of two dynamic domains, ORC2 WH and ORC1 AAA+, respectively, that repel each other from adopting their fixed positions. The domain dynamics are possibly linked to the catalytic activity at the ORC1·4 ATPase site, warranting further analysis to determine the molecular basis and the biological role of these domain movements. While the ORCOPEN structure largely resembles the ScORC conformation in the budding yeast OCCM complex and the previously reported HsORC structure, it unexpectedly reveals a significant hinge point at the ORC3-5 interface that generates a twist and pinch motion. The twist motion may allow ORC to move on DNA with a scanning mechanism to establish origin sites and the pinch motion may be involved in facilitating ORC to accommodate DNA into its central cavity via the partial ring opening and the subsequent incorporation of CDC6 into the ORC-DNA complex. The ATPase activity of ORC may promote the dynamic movement of the ORC1 and ORC2 subunits during assembly or prior to origin recognition. Lastly, the ORCDNA structure contains endogenous DNA bound in the core of ORC and reveals regions in the core that are in close proximity to the bound DNA. These observations along with previous studies on the dynamic, ATP-dependent assembly of ORC (Siddiqui and Stillman, 2007) suggest a model in which a pre-assembled ORC2-5 complex binds to ORC1 and promotes the opening of the ORC core to facilitate DNA binding. Since ORC1 appears to be the first ORC subunit to bind to chromosomes during mitosis (Kara et al., 2015; Okuno et al., 2001), ORC1 may recruit the other subunits via the pathway outlined in Figure 5. These interactions between ORC1 and ORC2-5 may be facilitated by intrinsically disordered regions (IDRs) the predicted N-terminal regions of ORC1 and ORC2 forming condensates via phase transitions (Parker et al., 2019). Proteins containing IDRs are difficult to express and by their nature contain large unstructured regions. We therefore removed the ORC1 N-terminal region from our construct and we were not able to resolve the ORC2 N-terminal region. Due to the absence of these regions, IDRs appear unnecessary for ORC assembly. Finally, it is known that the C-terminus of yeast Mcm3 binds to and activates the ATPase activity of the ORC-Cdc6 complex on DNA, possibly preventing futile loading of incomplete Mcm2-7 hexamers (Frigola et al., 2013). It is also possible that if the C-terminus of Mcm3 stimulates the ATPase activity of ORC, it may induce movement of the ORC1 AAA+ and a switch from the active to the inactive state of ORC after complete pre-RC assembly. Comparing the human and yeast ORC DNA contacts provides insights into the different sequence specificity requirements of ORC-DNA binding between the species. Our study establishes the foundation for further exploration of the critical determinants in metazoan replication origin establishment.

Figure 5. Proposed model for ORC dynamics.

Figure 5.

Open in a new tab

The binding of ORC1 to ORC2-5 opens the complex and facilitates dynamic movements of ORC1 AAA+ and ORC2 WH at the ring opening and a hinge motion at the ORC3·5 interface. The binding of DNA facilitates positioning of ORC1 AAA+ and ORC2 WHD on DNA for interaction to establish replication origin recognition.

Materials and methods

Protein preparation

Codon optimized human Origin Recognition Complex (HsORC) synthetic genes [NP_004144.2 HsORC subunit 1 (HsORC1), NP_006181.1 HsORC subunit 2 (HsORC2), NP_862820.1 HsORC subunit 3 (HsORC3), NP_859525.1 HsORC subunit 4 (HsORC4), NP_002544.1 HsORC subunit 5 (HsORC5)] were cloned into the MultiBac baculovirus expression system (Bieniossek et al., 2008). The ORC1 subunit was truncated at the N-terminus to facilitate stability and monodispersity of human ORC based on the previous study (Tocilj et al., 2017), while the remaining subunits were expressed as full-length proteins.

For recombinant expression of the hetero-pentameric HsORC1-5, gene-coding sequences were cloned into the pFL vector (pH promoter HsORC1 residues 471–861), pUCDM vector (pH promoter HsORC4 residues 1–436 (full-length), p10 promoter HsORC3 residues 1–712 (full-length) and pSPL vector (pH promoter HsORC2 residues 1–577 (full-length), p10 promoter HsORC5 residues 1–435 (full-length)). A StrepTag was introduced to the N-terminus of ORC3 for affinity purification. Sf9 insect cells were infected with baculovirus in Hyclone CCM3 media (GE Healthcare Life Sciences, Pittsburgh, PA) for 48 hr. For recombinant expression of HsORC2-5, the baculovirus system was identical to that of HsORC1-5, except that the HsORC1 was full-length (residues 1–861). However, after affinity purification of HsORC2-5, ORC1 was not present in the protein sample. The purification steps of HsORC1-5 and HsORC2-5 were identical.

All purification steps were performed at 4°C. Cell pellets were thawed on ice and resuspended in lysis buffer (50 mM HEPES-NaOH (pH 7.5), 200 mM KCl, 6.5 mM dithiothreitol (DTT), 1.7 mM adenosine triphosphate (ATP)). Following sonication, the lysate was centrifuged for 45 min at 143,000 g and the supernatant was loaded onto 5 ml of StrepTactin Super-flow resin (IBA, Goettingen, Germany). The beads were washed and recombinant HsORC1-5 complex was eluted with lysis buffer, which contained 5 mM desthiobiotin. The elution was concentrated and loaded onto a Superdex 200 Increase 10/300 GL size exclusion column (GE Healthcare Life Sciences, Pittsburgh, PA) equilibrated with minimal buffer (25 mM HEPES-NaOH (pH 7.5), 80 mM KCl, 3.3 mM dithiothreitol (DTT), 0.8 mM adenosine triphosphate (ATP)).

Cryo-electron microscopy sample preparation

CryoEM sample preparations for HsORC1-5 and HsORC2-5 were identical. ORC samples were diluted to 0.8 mg/ml and Lauryl Maltose Neopentyl Glycol (LMNG, Anatrace, Maumee, OH) were added to 0.05% w/v to reduce preferred orientations and promote thin layers of ice over grid holes. For cryoEM grid preparations, we applied 4 µl of ORC at a final concentration of 0.8 mg/ml to a glow-discharged Quantifoil R 2/1 300 mesh copper grid, incubated for 10 s at 25°C and 90% humidity, blotted for 2.9 s, and then plunged the grid into liquid ethane using a Leica Automatic Plunge Freezer EM GP2 (Buffalo Grove, IL).

Cryo-electron microscopy data acquisition

Cryo-EM data were collected on a FEI/ThermoFisher Titan Krios transmission electron microscope (TEM) operating at 300 keV. Dose-fractionated movies were collected using a Gatan K2 Summit direct electron detector operating in electron counting mode. For HsORC1-5, movies were collected with 30 frames during a 6 s exposure at an exposure rate of 2.2 e-/A2/frame, resulting in a cumulative exposure of 66 e-/Å2. For HsORC2-5, 35-frame movies were collected during a 7 s exposure at an exposure rate of 2.2 e-/A2/frame, resulting in a cumulative exposure of 77 e-/Å2. The EPU data collection software (ThermoFisher, Hillsboro, OR) was used to collect 9068 micrographs for HsORC1-5 and 5549 for HsORC2-5 micrographs by moving to hole centers and image shifting to five positions within the hole to acquire five exposures at 130,000x nominal magnification (1.07 Å/pixel at the specimen level), with a nominal defocus range of −1.0 to −2.4 μm. Data collection information is summarized in Supplementary file 1.

Image processing and 3D reconstruction

WARP was used to perform real-time image pre-processing (motion correction, CTF estimation, particle picking) during cryo-EM data collection for both HsORC1-5 and HsORC2-5 (Tegunov and Cramer, 2019). Particle picking was performed with the BoxNet pretrained neural network bundle employed in WARP and implemented in TensorFlow. A particle diameter of 160 Å and a threshold score of 0.6 yielded 2,112,391 particle coordinates for HsORC1-5 and 494,812 particle coordinates for HsORC2-5. These particles were initially subjected to a 2D classification in Cryosparc v2 (Punjani et al., 2017). A subset of particles were selected for ab initio reconstruction in Cryosparc v2 to generate an initial model for 3D classification in Relion 3.0 (Figure 1—figure supplements 1–2; Zivanov et al., 2018). Relion 3.0 was used for all remaining processing steps, including reprocessing of the micrographs (Relion’s implementation of motion correction), CTF estimation (Gctf) (Zhang, 2016; Zivanov et al., 2018).

For HsORC2-5, a total of 494,812 particles (from WARP/BoxNet coordinates) were initially extracted from micrographs using an unbinned box size of 320 pixels binned to a box size of 80 pixels (4 x bin) for 3D classification in Relion. The particles were classified into five classes using the scaled and low-pass filtered ab initio reconstruction from Cryosparc v2 as a reference map and a regularization parameter of four. The 209,479 particles contributing to the best 3D class average containing well-defined structural features were further unbinned, classified, and refined to yield a 3.7 Å map of 168,444 particles. To further improve the resolution, we performed Bayesian polishing and CTF refinement, twice each, to re-refine the particle stacks, which yielded a map with a reported resolution of 3.3 Å. The ORC2 WHD in the refined map was significantly unresolved, therefore, a 3D classification (3 classes, no alignment, regularization parameter of 100) with a mask encompassing the ORC2 WHD was performed that yielded a map with more complete density in the region. The 53,009 particles from the classification were further refined to yield a final map of 3.5 Å (ORCO2-5).

For HsORC1-5, a total of 2,112,391 particles (from WARP/BoxNet coordinates) were initially extracted from micrographs using an unbinned box size of 320 pixels binned to a box size of 80 pixels (4 x bin) for 3D classification in Relion. The particles were classified into five classes using the scaled and low-pass filtered ab initio reconstruction from Cryosparc v2 as a reference map and a regularization parameter of four. The 855,373 particles contributing to the best 3D class average containing well-defined structural features were further unbinned, classified, and refined to yield four prominent density maps at resolutions of ~3.8 Å (ORCO1AAA),~4.0 Å (ORCO2WH),~4.0 Å (ORCopen), and ~4.3 Å (ORCDNA). The observable structural features of the ORCopen and ORCDNA maps were not consistent with the reported resolutions and appeared to suffer from issues with flexibility and misalignment, and preferred orientation and were, therefore, not further refined. In addition, the program 3DFSC (Tan et al., 2017) revealed that these maps also suffered from preferred orientation (Figure 1—figure supplement 6). The ORCO1AAA and ORCO2WH maps contained well-defined features consistent with their respective resolutions, but were not sufficiently resolved for atomic modeling. To further improve the resolution, we performed CTF refinement and Bayesian polishing to re-refine the particle stacks, which yielded reconstructed maps with a reported resolution of 3.2 Å (ORCO1AAA) and 3.7 Å (ORCO2WH). Several regions of the ORCO1AAA map (particular the regions of ORC2 and ORC3 that line the gap in the ORC open-ring structure) did not contain structural features that are consistent with 3.2 Å cryo-EM density. Considering several 3D classes consisted of ORC2 and ORC3 in different orientations relative to the rest of the complex, we performed multibody refinement (Nakane et al., 2018) using two masked regions: ORC2·3·5 WHD and ORC1·4·5 AAA+. The chosen locations of the masks were further validated by performing Cryosparc’s unbiased 3D Variability Analysis (Punjani and Fleet, 2020) on the ORCO1AAA particle population which revealed the same movements as the Multibody Refinement. The multibody refinement improved the resolution and map quality of the region of ORC2·3 mentioned above (Figure 3—figure supplement 2). A final composite ORCO1AAA map of the focused and non-focused refinements was generated for atomic model building and refinement using the ‘vop max’ operation in UCSF Chimera (Goddard et al., 2007). Final cryoEM density map information is summarized in Supplementary file 1.

Model building

The atomic model of HsORC (PDB ID: 5ujm) (Tocilj et al., 2017) was used as a template for model building in ORCO1AAA and map. Density in the ORC2 and ORC3 region of the previous structure were of low resolution and relied significantly on the Drosophila melanogaster ORC2 and ORC3 model as a template (Bleichert et al., 2015). In this work, the high-resolution map of ORCO1AAA allowed for precise modelling of each residue in Coot which resulted in regions that were rebuilt, and regions built for the first time (especially in ORC2 and ORC3). Subunits of the ORCO1AAA model were used as templates for model building into the ORCO2-5, ORCO2WH, ORCOPEN, and ORCDNA maps. The ORC2 WHD of the ORCO2WH map was modelled initially using the ORC2 WHD from the DmORC atomic model (Bleichert et al., 2015). Each subunit of the models was rigid body fit using the ‘fit in map’ function in UCSF Chimera (Goddard et al., 2007). The resulting model, including nucleotides and coordinating metal ions, was subject to further model building using the COOT software package (Emsley and Cowtan, 2004), and then real-space refined with the PHENIX package (Afonine et al., 2012). The rotation angle calculations at the O3·5 hinge was determined using the ‘align’ function in UCSF ChimeraX (Goddard et al., 2018). The ORCO1AAA and ORCOPEN structures were aligned at ORC4, followed by alignment of ORC2·3 which generates a report containing the rotation angle between the aligned ORC2·3 subunits. UCSF ChimeraX (Goddard et al., 2018) was also used to generate the figures and buried surface areas were determined using the ‘measure buriedarea’ function in ChimeraX). Model statistics are summarized in Supplementary file 1.

Acknowledgements

We thank members of the Joshua-Tor laboratory for helpful comments and suggestions. We thank the CSHL Mass Spectrometry Shared Resource, which is supported by Cancer Center Support Grant 5P30CA045508. This work was supported by a Ruth L Kirschstein National Research Service Awards National Institutes of Health (NIH) fellowship F32GM129923 (to MJJ) and by NIH grant GM45436 (to BS). LJ is an investigator of the Howard Hughes Medical Institute.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Contributor Information

Leemor Joshua-Tor, Email: leemor@cshl.edu.

Sriram Subramaniam, University of British Columbia, Canada.

John Kuriyan, University of California, Berkeley, United States.

Funding Information

This paper was supported by the following grants:

  • Howard Hughes Medical Institute to Leemor Joshua-Tor.

  • National Institutes of Health F32GM129923 to Matt J Jaremko.

  • National Academies of Sciences, Engineering, and Medicine GM45436 to Bruce Stillman.

Additional information

Competing interests

Reviewing editor, eLife.

No competing interests declared.

Author contributions

Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing.

Conceptualization, Data curation, Formal analysis, Validation, Investigation, Visualization, Methodology, Writing - original draft, Writing - review and editing.

Data curation, Formal analysis.

Conceptualization, Investigation, Writing - review and editing.

Conceptualization, Formal analysis, Supervision, Funding acquisition, Validation, Methodology, Writing - original draft, Project administration, Writing - review and editing.

Additional files

Supplementary file 1. CryoEM and model statistics of all ORC structures.
elife-58622-supp1.docx (32.6KB, docx)
Transparent reporting form

Data availability

All coordinates and cryoEM maps have deposited in the PDB and EMDB: ORC-O1AAA: PDB code: 7JPO EMDB code: EMD-22417 ORC-O2WH: PDB code: 7JPP EMDB code: EMD-22418 ORC-O2-5: PDB code: 7JPQ EMDB code: EMD-22419 ORC-OPEN: PDB code: 7JPR EMDB code: EMD-22420 ORC-DNA: PDB code: 7JPS EMDB code: EMD-22421.

The following datasets were generated:

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O1AAA. RCSB Protein Data Bank. 7JPO

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O1AAA. Electron Microscopy Data Bank. EMD-22417

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2WH. RCSB Protein Data Bank. 7JPP

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2WH. Electron Microscopy Data Bank. EMD-22418

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2-5. RCSB Protein Data Bank. 7JPQ

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2-5. Electron Microscopy Data Bank. EMD-22419

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-OPEN. RCSB Protein Data Bank. 7JPR

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-OPEN. Electron Microscopy Data Bank. EMD-22420

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-DNA. RCSB Protein Data Bank. 7JPS

Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-DNA. Electron Microscopy Data Bank. EMD-22421

References

  1. Afonine PV, Grosse-Kunstleve RW, Echols N, Headd JJ, Moriarty NW, Mustyakimov M, Terwilliger TC, Urzhumtsev A, Zwart PH, Adams PD. Towards automated crystallographic structure refinement with phenix.refine. Acta Crystallographica Section D Biological Crystallography. 2012;68:352–367. doi: 10.1107/S0907444912001308. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bell SP, Labib K. Chromosome duplication in Saccharomyces cerevisiae. Genetics. 2016;203:1027–1067. doi: 10.1534/genetics.115.186452. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bell SP, Stillman B. ATP-dependent recognition of eukaryotic origins of DNA replication by a multiprotein complex. Nature. 1992;357:128–134. doi: 10.1038/357128a0. [DOI] [PubMed] [Google Scholar]
  4. Bicknell LS, Bongers EM, Leitch A, Brown S, Schoots J, Harley ME, Aftimos S, Al-Aama JY, Bober M, Brown PA, van Bokhoven H, Dean J, Edrees AY, Feingold M, Fryer A, Hoefsloot LH, Kau N, Knoers NV, Mackenzie J, Opitz JM, Sarda P, Ross A, Temple IK, Toutain A, Wise CA, Wright M, Jackson AP. Mutations in the pre-replication complex cause Meier-Gorlin syndrome. Nature Genetics. 2011a;43:356–359. doi: 10.1038/ng.775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bicknell LS, Walker S, Klingseisen A, Stiff T, Leitch A, Kerzendorfer C, Martin CA, Yeyati P, Al Sanna N, Bober M, Johnson D, Wise C, Jackson AP, O'Driscoll M, Jeggo PA. Mutations in ORC1, encoding the largest subunit of the origin recognition complex, cause microcephalic primordial dwarfism resembling Meier-Gorlin syndrome. Nature Genetics. 2011b;43:350–355. doi: 10.1038/ng.776. [DOI] [PubMed] [Google Scholar]
  6. Bieniossek C, Richmond TJ, Berger I. MultiBac: multigene baculovirus‐based eukaryotic protein complex production. Current Protocols in Protein Science. 2008;51:5.20.21–5.20.25. doi: 10.1002/0471140864.ps0520s51. [DOI] [PubMed] [Google Scholar]
  7. Bleichert F, Botchan MR, Berger JM. Crystal structure of the eukaryotic origin recognition complex. Nature. 2015;519:321–326. doi: 10.1038/nature14239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bleichert F, Botchan MR, Berger JM. Mechanisms for initiating cellular DNA replication. Science. 2017;355:eaah6317. doi: 10.1126/science.aah6317. [DOI] [PubMed] [Google Scholar]
  9. Bleichert F, Leitner A, Aebersold R, Botchan MR, Berger JM. Conformational control and DNA-binding mechanism of the metazoan origin recognition complex. PNAS. 2018;115:E5906–E5915. doi: 10.1073/pnas.1806315115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bowers JL, Randell JC, Chen S, Bell SP. ATP hydrolysis by ORC catalyzes reiterative Mcm2-7 assembly at a defined origin of replication. Molecular Cell. 2004;16:967–978. doi: 10.1016/j.molcel.2004.11.038. [DOI] [PubMed] [Google Scholar]
  11. Brewer BJ, Fangman WL. The localization of replication origins on ARS plasmids in S. cerevisiae. Cell. 1987;51:463–471. doi: 10.1016/0092-8674(87)90642-8. [DOI] [PubMed] [Google Scholar]
  12. Craig JM, Earle E, Canham P, Wong LH, Anderson M, Choo KH. Analysis of mammalian proteins involved in chromatin modification reveals new metaphase centromeric proteins and distinct chromosomal distribution patterns. Human Molecular Genetics. 2003;12:3109–3121. doi: 10.1093/hmg/ddg330. [DOI] [PubMed] [Google Scholar]
  13. de Munnik SA, Bicknell LS, Aftimos S, Al-Aama JY, van Bever Y, Bober MB, Clayton-Smith J, Edrees AY, Feingold M, Fryer A, van Hagen JM, Hennekam RC, Jansweijer MC, Johnson D, Kant SG, Opitz JM, Ramadevi AR, Reardon W, Ross A, Sarda P, Schrander-Stumpel CT, Schoots J, Temple IK, Terhal PA, Toutain A, Wise CA, Wright M, Skidmore DL, Samuels ME, Hoefsloot LH, Knoers NV, Brunner HG, Jackson AP, Bongers EM. Meier-Gorlin syndrome genotype-phenotype studies: 35 individuals with pre-replication complex gene mutations and 10 without molecular diagnosis. European Journal of Human Genetics. 2012a;20:598–606. doi: 10.1038/ejhg.2011.269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. de Munnik SA, Otten BJ, Schoots J, Bicknell LS, Aftimos S, Al-Aama JY, van Bever Y, Bober MB, Borm GF, Clayton-Smith J, Deal CL, Edrees AY, Feingold M, Fryer A, van Hagen JM, Hennekam RC, Jansweijer MC, Johnson D, Kant SG, Opitz JM, Ramadevi AR, Reardon W, Ross A, Sarda P, Schrander-Stumpel CT, Sluiter AE, Temple IK, Terhal PA, Toutain A, Wise CA, Wright M, Skidmore DL, Samuels ME, Hoefsloot LH, Knoers NV, Brunner HG, Jackson AP, Bongers EM. Meier-Gorlin syndrome: growth and secondary sexual development of a microcephalic primordial dwarfism disorder. American Journal of Medical Genetics Part A. 2012b;158A:2733–2742. doi: 10.1002/ajmg.a.35681. [DOI] [PubMed] [Google Scholar]
  15. Eaton ML, Galani K, Kang S, Bell SP, MacAlpine DM. Conserved nucleosome positioning defines replication origins. Genes & Development. 2010;24:748–753. doi: 10.1101/gad.1913210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Edgar RC. MUSCLE: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinformatics. 2004;5:113. doi: 10.1186/1471-2105-5-113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Emsley P, Cowtan K. Coot: model-building tools for molecular graphics. Acta Crystallographica. Section D, Biological Crystallography. 2004;60:2126–2132. doi: 10.1107/S0907444904019158. [DOI] [PubMed] [Google Scholar]
  18. Enemark EJ, Joshua-Tor L. On helicases and other motor proteins. Current Opinion in Structural Biology. 2008;18:243–257. doi: 10.1016/j.sbi.2008.01.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Erzberger JP, Berger JM. Evolutionary relationships and structural mechanisms of AAA+ proteins. Annual Review of Biophysics and Biomolecular Structure. 2006;35:93–114. doi: 10.1146/annurev.biophys.35.040405.101933. [DOI] [PubMed] [Google Scholar]
  20. Evrin C, Clarke P, Zech J, Lurz R, Sun J, Uhle S, Li H, Stillman B, Speck C. A double-hexameric MCM2-7 complex is loaded onto origin DNA during licensing of eukaryotic DNA replication. PNAS. 2009;106:20240–20245. doi: 10.1073/pnas.0911500106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Frigola J, Remus D, Mehanna A, Diffley JF. ATPase-dependent quality control of DNA replication origin licensing. Nature. 2013;495:339–343. doi: 10.1038/nature11920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Goddard TD, Huang CC, Ferrin TE. Visualizing density maps with UCSF chimera. Journal of Structural Biology. 2007;157:281–287. doi: 10.1016/j.jsb.2006.06.010. [DOI] [PubMed] [Google Scholar]
  23. Goddard TD, Huang CC, Meng EC, Pettersen EF, Couch GS, Morris JH, Ferrin TE. UCSF ChimeraX: meeting modern challenges in visualization and analysis. Protein Science. 2018;27:14–25. doi: 10.1002/pro.3235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Guernsey DL, Matsuoka M, Jiang H, Evans S, Macgillivray C, Nightingale M, Perry S, Ferguson M, LeBlanc M, Paquette J, Patry L, Rideout AL, Thomas A, Orr A, McMaster CR, Michaud JL, Deal C, Langlois S, Superneau DW, Parkash S, Ludman M, Skidmore DL, Samuels ME. Mutations in origin recognition complex gene ORC4 cause Meier-Gorlin syndrome. Nature Genetics. 2011;43:360–364. doi: 10.1038/ng.777. [DOI] [PubMed] [Google Scholar]
  25. Hossain M, Stillman B. Opposing roles for DNA replication initiator proteins ORC1 and CDC6 in control of cyclin E gene transcription. eLife. 2016;5:e12785. doi: 10.7554/eLife.12785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hu Y, Tareen A, Sheu Y-J. Evolution of DNA replication origin specification and gene silencing mechanisms. bioRxiv. 2020 doi: 10.1101/2020.07.04.187286. [DOI] [PMC free article] [PubMed]
  27. Huberman JA. Eukaryotic DNA replication: a complex picture partially clarified. Cell. 1987;48:7–8. doi: 10.1016/0092-8674(87)90347-3. [DOI] [PubMed] [Google Scholar]
  28. Kara N, Hossain M, Prasanth SG, Stillman B. Orc1 binding to mitotic chromosomes precedes spatial patterning during G1 phase and assembly of the origin recognition complex in human cells. Journal of Biological Chemistry. 2015;290:12355–12369. doi: 10.1074/jbc.M114.625012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kellenberger E, Carlemalm E, Villiger W, Wurtz M, Mory C, Colliex C. Z-contrast in biology A comparison with other imaging modes. Annals of the New York Academy of Sciences. 1986;483:202–228. doi: 10.1111/j.1749-6632.1986.tb34522.x. [DOI] [PubMed] [Google Scholar]
  30. Klemm RD, Austin RJ, Bell SP. Coordinate binding of ATP and origin DNA regulates the ATPase activity of the origin recognition complex. Cell. 1997;88:493–502. doi: 10.1016/S0092-8674(00)81889-9. [DOI] [PubMed] [Google Scholar]
  31. Kucukelbir A, Sigworth FJ, Tagare HD. Quantifying the local resolution of cryo-EM density maps. Nature Methods. 2014;11:63–65. doi: 10.1038/nmeth.2727. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kuo AJ, Song J, Cheung P, Ishibe-Murakami S, Yamazoe S, Chen JK, Patel DJ, Gozani O. The BAH domain of ORC1 links H4K20me2 to DNA replication licensing and Meier-Gorlin syndrome. Nature. 2012;484:115–119. doi: 10.1038/nature10956. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Leonard AC, Méchali M. DNA replication origins. Cold Spring Harbor Perspectives in Biology. 2013;5:a010116. doi: 10.1101/cshperspect.a010116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Li N, Lam WH, Zhai Y, Cheng J, Cheng E, Zhao Y, Gao N, Tye BK. Structure of the origin recognition complex bound to DNA replication origin. Nature. 2018;559:217–222. doi: 10.1038/s41586-018-0293-x. [DOI] [PubMed] [Google Scholar]
  35. Linskens MH, Huberman JA. Organization of replication of ribosomal DNA in Saccharomyces cerevisiae. Molecular and Cellular Biology. 1988;8:4927–4935. doi: 10.1128/MCB.8.11.4927. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Long H, Zhang L, Lv M, Wen Z, Zhang W, Chen X, Zhang P, Li T, Chang L, Jin C, Wu G, Wang X, Yang F, Pei J, Chen P, Margueron R, Deng H, Zhu M, Li G. H2A.Z facilitates licensing and activation of early replication origins. Nature. 2020;577:576–581. doi: 10.1038/s41586-019-1877-9. [DOI] [PubMed] [Google Scholar]
  37. Marahrens Y, Stillman B. A yeast chromosomal origin of DNA replication defined by multiple functional elements. Science. 1992;255:817–823. doi: 10.1126/science.1536007. [DOI] [PubMed] [Google Scholar]
  38. Miller TCR, Locke J, Greiwe JF, Diffley JFX, Costa A. Mechanism of head-to-head MCM double-hexamer formation revealed by cryo-EM. Nature. 2019;575:704–710. doi: 10.1038/s41586-019-1768-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Miotto B, Ji Z, Struhl K. Selectivity of ORC binding sites and the relation to replication timing, fragile sites, and deletions in cancers. PNAS. 2016;113:E4810–E4819. doi: 10.1073/pnas.1609060113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Müller P, Park S, Shor E, Huebert DJ, Warren CL, Ansari AZ, Weinreich M, Eaton ML, MacAlpine DM, Fox CA. The conserved bromo-adjacent homology domain of yeast Orc1 functions in the selection of DNA replication origins within chromatin. Genes & Development. 2010;24:1418–1433. doi: 10.1101/gad.1906410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Nakane T, Kimanius D, Lindahl E, Scheres SH. Characterisation of molecular motions in cryo-EM single-particle data by multi-body refinement in RELION. eLife. 2018;7:e36861. doi: 10.7554/eLife.36861. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Noguchi K, Vassilev A, Ghosh S, Yates JL, DePamphilis ML. The BAH domain facilitates the ability of human Orc1 protein to activate replication origins in vivo. The EMBO Journal. 2006;25:5372–5382. doi: 10.1038/sj.emboj.7601396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Okuno Y, McNairn AJ, den Elzen N, Pines J, Gilbert DM. Stability, chromatin association and functional activity of mammalian pre-replication complex proteins during the cell cycle. The EMBO Journal. 2001;20:4263–4277. doi: 10.1093/emboj/20.15.4263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. On KF, Jaremko M, Stillman B, Joshua-Tor L. A structural view of the initiators for chromosome replication. Current Opinion in Structural Biology. 2018;53:131–139. doi: 10.1016/j.sbi.2018.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Parker MW, Bell M, Mir M, Kao JA, Darzacq X, Botchan MR, Berger JM. A new class of disordered elements controls DNA replication through initiator self-assembly. eLife. 2019;8:e48562. doi: 10.7554/eLife.48562. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Prasanth SG, Prasanth KV, Siddiqui K, Spector DL, Stillman B. Human Orc2 localizes to Centrosomes, centromeres and heterochromatin during chromosome inheritance. The EMBO Journal. 2004;23:2651–2663. doi: 10.1038/sj.emboj.7600255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Prasanth SG, Shen Z, Prasanth KV, Stillman B. Human origin recognition complex is essential for HP1 binding to chromatin and heterochromatin organization. PNAS. 2010;107:15093–15098. doi: 10.1073/pnas.1009945107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Prioleau MN, MacAlpine DM. DNA replication origins-where do we begin? Genes & Development. 2016;30:1683–1697. doi: 10.1101/gad.285114.116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Punjani A, Rubinstein JL, Fleet DJ, Brubaker MA. cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nature Methods. 2017;14:290–296. doi: 10.1038/nmeth.4169. [DOI] [PubMed] [Google Scholar]
  50. Punjani A, Fleet DJ. 3d variability analysis: directly resolving continuous flexibility and discrete heterogeneity from single particle cryo-EM images. bioRxiv. 2020 doi: 10.1101/2020.04.08.032466. [DOI] [PubMed]
  51. Randell JC, Bowers JL, Rodríguez HK, Bell SP. Sequential ATP hydrolysis by Cdc6 and ORC directs loading of the Mcm2-7 helicase. Molecular Cell. 2006;21:29–39. doi: 10.1016/j.molcel.2005.11.023. [DOI] [PubMed] [Google Scholar]
  52. Rao H, Stillman B. The origin recognition complex interacts with a bipartite DNA binding site within yeast replicators. PNAS. 1995;92:2224–2228. doi: 10.1073/pnas.92.6.2224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Remus D, Beall EL, Botchan MR. DNA topology, not DNA sequence, is a critical determinant for Drosophila ORC-DNA binding. The EMBO Journal. 2004;23:897–907. doi: 10.1038/sj.emboj.7600077. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Remus D, Beuron F, Tolun G, Griffith JD, Morris EP, Diffley JF. Concerted loading of Mcm2-7 double hexamers around DNA during DNA replication origin licensing. Cell. 2009;139:719–730. doi: 10.1016/j.cell.2009.10.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Rivera C, Gurard-Levin ZA, Almouzni G, Loyola A. Histone lysine methylation and chromatin replication. Biochimica Et Biophysica Acta (BBA) - Gene Regulatory Mechanisms. 2014;1839:1433–1439. doi: 10.1016/j.bbagrm.2014.03.009. [DOI] [PubMed] [Google Scholar]
  56. Siddiqui K, Stillman B. ATP-dependent assembly of the human origin recognition complex. Journal of Biological Chemistry. 2007;282:32370–32383. doi: 10.1074/jbc.M705905200. [DOI] [PubMed] [Google Scholar]
  57. Sun J, Kawakami H, Zech J, Speck C, Stillman B, Li H. Cdc6-induced conformational changes in ORC bound to origin DNA revealed by cryo-electron microscopy. Structure. 2012;20:534–544. doi: 10.1016/j.str.2012.01.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Tan YZ, Baldwin PR, Davis JH, Williamson JR, Potter CS, Carragher B, Lyumkis D. Addressing preferred specimen orientation in single-particle cryo-EM through tilting. Nature Methods. 2017;14:793–796. doi: 10.1038/nmeth.4347. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Tegunov D, Cramer P. Real-time cryo-electron microscopy data preprocessing with warp. Nature Methods. 2019;16:1146–1152. doi: 10.1038/s41592-019-0580-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Theis JF, Yang C, Schaefer CB, Newlon CS. DNA sequence and functional analysis of homologous ARS elements of Saccharomyces cerevisiae and S. carlsbergensis. Genetics. 1999;152:943–952. doi: 10.1093/genetics/152.3.943. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Ticau S, Friedman LJ, Ivica NA, Gelles J, Bell SP. Single-molecule studies of origin licensing reveal mechanisms ensuring bidirectional helicase loading. Cell. 2015;161:513–525. doi: 10.1016/j.cell.2015.03.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Tocilj A, On KF, Yuan Z, Sun J, Elkayam E, Li H, Stillman B, Joshua-Tor L. Structure of the active form of human origin recognition complex and its ATPase motor module. eLife. 2017;6:e20818. doi: 10.7554/eLife.20818. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Vashee S, Simancek P, Challberg MD, Kelly TJ. Assembly of the human origin recognition complex. Journal of Biological Chemistry. 2001;276:26666–26673. doi: 10.1074/jbc.M102493200. [DOI] [PubMed] [Google Scholar]
  64. Vashee S, Cvetic C, Lu W, Simancek P, Kelly TJ, Walter JC. Sequence-independent DNA binding and replication initiation by the human origin recognition complex. Genes & Development. 2003;17:1894–1908. doi: 10.1101/gad.1084203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Yuan Z, Riera A, Bai L, Sun J, Nandi S, Spanos C, Chen ZA, Barbon M, Rappsilber J, Stillman B, Speck C, Li H. Structural basis of Mcm2-7 replicative helicase loading by ORC-Cdc6 and Cdt1. Nature Structural & Molecular Biology. 2017;24:316–324. doi: 10.1038/nsmb.3372. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Zhang K. Gctf: real-time CTF determination and correction. Journal of Structural Biology. 2016;193:1–12. doi: 10.1016/j.jsb.2015.11.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Zivanov J, Nakane T, Forsberg BO, Kimanius D, Hagen WJ, Lindahl E, Scheres SH. New tools for automated high-resolution cryo-EM structure determination in RELION-3. eLife. 2018;7:e42166. doi: 10.7554/eLife.42166. [DOI] [PMC free article] [PubMed] [Google Scholar]

Decision letter

Editor: Sriram Subramaniam1
Reviewed by: Tim Grant2

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

The structures that you report of multiple states of the human ORC complex, an essential replication complex in free and DNA-bound states, is a major advance. The resolution is significantly improved compared to what was reported previously and the newly described conformational states provide further mechanistic insight into the significance of protein movements that are relevant for DNA binding.

Decision letter after peer review:

Thank you for submitting your article "The dynamic nature of the human Origin Recognition Complex revealed through five cryoEM structures" for consideration by eLife. Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by Sriram Subramaniam as the Reviewing Editor and John Kuriyan as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Tim Grant (Reviewer #3).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

Jaremko et al. describe structures of multiple states of the human ORC complex, an essential replication complex in free and DNA-bound states. The structures contribute a major advance, as the resolution is significantly improved compared to the previous HsORC structure and describe conformational states that provide further mechanistic insight into HsORC function. Altogether these results are important and should be published.

Essential revisions:

1) Validation and analysis:

Supplementary file 1 and validation reports don't match. Supplementary file 1 lacks info.

– FSC (model-to-map) for final models should be provided.

– Maps should be displayed next to the model or a table indicating which conformation includes which subunits/domains/residues in the corresponding model should be provided to give the reader more clarity about which parts of the complex were modeled in each conformation and how the conformations differ from each other.

– Show EM maps for all ATPs sites, not only 1 conformation (in Figure 1—figure supplement 2).

2) Revision of Figures 1, 2, 4 and Video 1:

Figure 1:

– Sideview does not add much new info, rather show a cryoEM map next to the model to visualize which parts of the map have been modeled.

– Panel J) Show close-up of the β-341 hairpin loop of ORC4O1AAA and show corresponding area in ScORC as well as cartoon of WHD domains from HsORC4 and ScORC4 side by side either in 1J or as part of Figure 4. The difference in this critical DNA binding area is a major finding of the manuscript and needs to be visualized for the reader.

Show intertwining of ORC2 and ORC3 from HsORC and comparison to DmORC.

Figure 2:

– Label subunits again. Panel 2B does not show structural change, needs a sideview or arrow to show the 90° tilt of the subunit in the bottom panel.

– Panel C-F: Display tilt angle of inactive versus autoinhibited conformation.

Figure 4:

– EM-Map hides DNA and it is not clear which residues interact with DNA. Show cartoon representation of structural models throughout this figure. Show top down view on dsDNA at appropriate zoom to visualize all interactions in 1 image.

Video 1:

– Use colorzone command or encircle WHD domain to highlight WHD domain unambiguously. Label red map, blue map and don't change color at the end of the video?

3) A deeper discussion of findings and the mechanistic model:

– The authors find shortening of an α-helix in the WHD domain of HsORC4 compared to ScORC that could account for sequence specificity of ScORC. Is there more data to support this? Which organisms have or don't have this helix? Does the presence of the helix correlate with sequence specificity of Orc? Are there prior sequencing data of corresponding ScORC4-WHD domain mutants that support this? This is an important finding, but is explored very little in the manuscript.

– The authors speculate that Arg89 of ORC3 could be involved in ATP coordination (subsection “Overall HsORC structure and comparison to ORC from Saccharomyces cerevisiae and Drosophilamelanogaster”). Ideally, they'd mutate the residue and test function. Absent that, is this residue conserved?

– The authors also speculate that HsORC may detect intracellular levels of ATP. Is there data to back this up? What is the Kd of ATP? Is it reasonable for acting as a sensor? If not, then this needs to be removed.

– The authors propose an interesting model for HsORC activation and DNA binding. Is there biochemical data in support of this model? Could comment a bit more on the role of the conserved WHD domain of ORC2 and the inactive state: Does the WHD domain in inactive state prevent nonspecific DNA binding of ORC2-5? Is there available DNA binding data on deletion mutants of WHD domain of ORC2?

– Discussion in regard to IDRs and phase separation: The text reads as if IDRs are present in several subunits – in which subunits and where are the IDRs? Are they predicted? Why are they not present in structures? Are they not resolved (likely owing to flexibility) or were they not included in expression constructs? Please clarify in discussion.

– Comment more on the characteristic structural changes from closed to open to DNA-bound (subunit to subunit translation, helical radius, etc). What does "open" refer to? Widening of the spiral?

Report on buried surface interaction area between ORC subunits to confirm exceptional intertwining of ORC2 and ORC3 from HsORC and DmORC.

– How do the authors come to the conclusion that the loop in ORC3 (Figure 1—figure supplement 5) is related to DNA binding, is this area close to DNA or DNA binding site? Is this area resolved in all conformations?

4) Multibody refinement:

– To solidify the ORC3·5 hinge, verify that hinge point is not an artefact of mask choice. Run multibody refine with a mask boundary between ORC1·4 and ORC2·3·5 and report on results.

– Figure legend is in discrepancy with text: Legend indicates that for multibody refinement particles from only 1 conformation were used, text states particles from 3 conformations were included.

– The text is also not very quantitative in description of the MBR.

5) DNA-bound structure:

It makes sense that the DNA sequence would not be in register because HsORC isn't sequence specific, but is the major/minor groove positioning in register? I would think that the positioning of the double helix is essentially fixed relative to the major/minor grooves. Can the authors make out DNA helix clearly? Clearer figures would help.

6) Sample preparation details:

In the sample preparation, addition of ATP in the buffer is mentioned. Were there any divalent cations, Mg2+ or Mn2+, included? What is the rationale for the ATP addition instead of ATPγS, AMPPNP or ADP·BFx? What are the green spheres shown next to the ATP molecules in Figure 1G-I? Why is ATP not hydrolyzed in ORC1·4 or ORC4·5?

7) In Figure 1—figure supplement 3D and E, the maps look of much lower resolution than 3.7 Å (ORCOPEN) or 4.3 Å (ORCDNA). Were these maps unsharpened? The authors may consider showing representative EM map regions or 3D-FSC.

8) DNA density:

For the ORCDNA complex, it will be important to show an EM map covering the DNA portion. Although base sequences are mixed, the DNA backbones should still be recognizable.

9) Resolution anisotropy:

The resolution is extremely directionally dependent, and the true resolution of the ORC open and ORC DNA maps is low in all but one direction. Based on the direction plots shown in Figure 1—figure supplement 3 all of the samples suffered somewhat from preferential orientation. I believe the authors should mention this, and perform a direction dependent resolution analysis of their maps (e.g. using the 3D FSC server – https://3dfsc.salk.edu/) and report the results.

eLife. 2020 Aug 18;9:e58622. doi: 10.7554/eLife.58622.sa2

Author response

Essential revisions:

1) Validation and analysis:

Supplementary file 1 and validation reports don't match. Supplementary file 1 lacks info.

– FSC (model-to-map) for final models should be provided.

We have included the FSC (model-to-map) values at 0.143 and 0.5 in Supplementary file 1. Any other difference between the table and validation reports are due to the different validation software packages utilized. The statistics provided in Supplementary file 1 were generated from Phenix/Molprobity. Note – the structures have been further refined since the first submission and the geometry has greatly improved.

– Maps should be displayed next to the model or a table indicating which conformation includes which subunits/domains/residues in the corresponding model should be provided to give the reader more clarity about which parts of the complex were modeled in each conformation and how the conformations differ from each other.

We removed the sideview and added the density map next to each model in Figure 1. In addition, we have labelled each structure with their key feature (missing O1AAA+/O2WH, DNA present, open conformation) to better aid the reader.

– Show EM maps for all ATPs sites, not only 1 conformation (in Figure 1—figure supplement 2).

We added Figure 1—figure supplement 1 to include all the remaining ATP sites of all models with their corresponding EM densities.

2) Revision of Figures 1, 2, 4 and Video 1:

Figure 1:

– Sideview does not add much new info, rather show a cryoEM map next to the model to visualize which parts of the map have been modeled.

We removed the sideview and added the density map next to each model in Figure 1.

– Panel J) Show close-up of the β-341 hairpin loop of ORC4O1AAA and show corresponding area in ScORC as well as cartoon of WHD domains from HsORC4 and ScORC4 side by side either in 1J or as part of Figure 4. The difference in this critical DNA binding area is a major finding of the manuscript and needs to be visualized for the reader.

Show intertwining of ORC2 and ORC3 from HsORC and comparison to DmORC.

We added a figure of the ScORC4 loop in Figure 4D and moved the HsORC4 loop to Figure 4E with the overlaid density map removed for better visual comparison. We also show a superposition of the two loops in Figure 4F.

We included a figure of the ORC2-3 intertwine in our structure, ScORC, and DmORC for comparison in Figure 1—figure supplement 8, and calculated buried surface areas for a more quantitative comparison.

Figure 2:

– Label subunits again. Panel 2B does not show structural change, needs a sideview or arrow to show the 90° tilt of the subunit in the bottom panel.

We replaced the maps in Figure 2A, B with models to better visualize the movement of O1AAA+ domain. The maps and process to generate the maps were moved to Figure 2—figure supplement 1 to better layout the method of visualizing the dynamic regions. We also changed Video 2 of the cryoEM map morphing to model morphing to better visualize the movement discussed in the text.

– Panel C-F: Display tilt angle of inactive versus autoinhibited conformation.

The change in conformation isn’t a simple tilt, but we have included Video 1 which clearly illustrates the different conformations of O2WH in the inactive and autoinhibited conformations.

Figure 4:

– EM-Map hides DNA and it is not clear which residues interact with DNA. Show cartoon representation of structural models throughout this figure. Show top down view on dsDNA at appropriate zoom to visualize all interactions in 1 image.

We added a top down view of the DNA with important residues included in Figure 4A. We also included a figure of the DNA overlaid with the density map and a color-coded density map in Figure 4—figure supplement 1.

Video 1:

– Use colorzone command or encircle WHD domain to highlight WHD domain unambiguously. Label red map, blue map and don't change color at the end of the video?

We added Video 1 of the ORC2 WHD that illustrates the movement of the domain more clearly. In addition, we added Figure 2—figure supplement 1 which lays out the low-pass filter method to visualize the ORC2 WHD and ORC1 AAA+ domain dynamics.

3) A deeper discussion of findings and the mechanistic model

– The authors find shortening of an α-helix in the WHD domain of HsORC4 compared to ScORC that could account for sequence specificity of ScORC. Is there more data to support this? Which organisms have or don't have this helix? Does the presence of the helix correlate with sequence specificity of Orc? Are there prior sequencing data of corresponding ScORC4-WHD domain mutants that support this? This is an important finding, but is explored very little in the manuscript.

We removed the short comments from section 1, and expanded the discussion in the “DNA bound to HsORC” section.

We also added the really intriguing observation that the ORC4 helix insertions, along with sequence specificity, have likely evolved in Saccharomyces cerevisae budding yeast and other fungi coincident with loss of RNA-interference mediated silencing, while this region is absent in other organisms, including fungi, plants, and animals that do have RNAi. This is discussed in much greater detail in a paper that just came out in bioRxiv (Hu, et al., 2020) and that we now refer to in this manuscript.

– The authors speculate that Arg89 of ORC3 could be involved in ATP coordination (subsection “Overall HsORC structure and comparison to ORC from Saccharomyces cerevisiae and Drosophila melanogaster”). Ideally, they'd mutate the residue and test function. Absent that, is this residue conserved?

R89 is not conserved and after further inspection this claim has been removed from the text. However, we are confident in the remaining residues involvement in ATP coordination that are discussed.

– The authors also speculate that HsORC may detect intracellular levels of ATP. Is there data to back this up? What is the Kd of ATP? Is it reasonable for acting as a sensor? If not, then this needs to be removed.

We agree that this is probably too speculative and thus removed this from the text because we have no current data to validate the claim.

– The authors propose an interesting model for HsORC activation and DNA binding. Is there biochemical data in support of this model? Could comment a bit more on the role of the conserved WHD domain of ORC2 and the inactive state: Does the WHD domain in inactive state prevent nonspecific DNA binding of ORC2-5? Is there available DNA binding data on deletion mutants of WHD domain of ORC2?

As observed from the structures, the position of the ORC2 WHD in ORC2-5 occupies the same space as the DNA occupies in the ORC-DNA complex. Hence, we speculated that it obstructs DNA from binding at the central cavity in ORC2-5. ORC2-5 exhibits DNA binding activity (data not shown) but its exact mode of DNA interaction and how this interaction is related to the ORC-DNA conformation in the current study is still under investigation. Since metazoan ORC does not exhibit the sequence-specific DNA interaction observed in S. cerevisiae, we do not expect ORC2-5 or the ORC in the ORCDNA conformation to discriminate between specific and non-specific DNA-binding in our current study.

– Discussion in regard to IDRs and phase separation: The text reads as if IDRs are present in several subunits – in which subunits and where are the IDRs? Are they predicted? Why are they not present in structures? Are they not resolved (likely owing to flexibility) or were they not included in expression constructs? Please clarify in discussion.

We have further clarified the discussion of IDRs in the text: “These interactions between ORC1 and ORC2-5 may be facilitated by intrinsically disordered regions (IDRs) within the predicted N-terminal regions of ORC1 and ORC2 forming condensates via phase transitions (Parker, et al., 2019). Proteins containing IDRs are difficult to express and structurally resolve, therefore we removed the ORC1 N-terminal region from our construct and we were not able to resolve the ORC2 N-terminal region. Due to the absence of these regions, IDRs are likely not necessary for ORC assembly. “

– Comment more on the characteristic structural changes from closed to open to DNA-bound (subunit to subunit translation, helical radius, etc). What does "open" refer to? Widening of the spiral?

Report on buried surface interaction area between ORC subunits to confirm exceptional intertwining of ORC2 and ORC3 from HsORC and DmORC.

We added to the discussion (subsection “Overall HsORC structure and comparison to ORC from Saccharomyces cerevisiae and Drosophila melanogaster”) describing widening of the spiral in the ORCOPEN conformation and added Figure 3—figure supplement 2 which illustrates the change in conformation of a long α-helix in ORC that likely contributes to the stability of the open conformation.

In addition, we have utilized the ‘measure buriedarea’ function in ChimeraX to determine the solvent-accessible surface area buried between ORC subunits and determined that buried area between ORC2 and ORC3 (3296 Å2) is significantly larger than the next closest interacting subunits ORC4 and ORC5 (2707 Å2). This is discussed in the subsection “Overall HsORC structure and comparison to ORC from Saccharomyces cerevisiae and Drosophila melanogaster”. We have also added a figure in Figure 1—figure supplement 8 that focuses on the intertwine between O2 and O3.

– How do the authors come to the conclusion that the loop in ORC3 (Figure 1—figure supplement 5) is related to DNA binding, is this area close to DNA or DNA binding site? Is this area resolved in all conformations?

The cross-reference was incorrectly placed here. We also removed this information in the text.

4) Multibody refinement:

– To solidify the ORC3·5 hinge, verify that hinge point is not an artefact of mask choice. Run multibody refine with a mask boundary between ORC1·4 and ORC2·3·5 and report on results.

To avoid mask bias, we performed Cryosparc 3D Variability to complement the Multibody Refinement results. Compared to Multibody Refinement, 3D Variability Analysis does not rely on user input to determine regions that move independently of one another. In addition, though ORC1 may be too small for a single body in Multibody Refinement, this is not the case for 3D Variability. Both analyses revealed the same prominent movements of the ORCO1AAA particle population. We included the movements of both analyses in Video 3.

– Figure legend is in discrepancy with text: Legend indicates that for multibody refinement particles from only 1 conformation were used, text states particles from 3 conformations were included.

– The text is also not very quantitative in description of the MBR.

We thank the reviewers for finding this mistake. There are 3 conformations and we have corrected this. The Multibody Refinement results for ORCO2WH and ORCOPEN are available in Figure 3—figure supplement 1, which shows the principle component analysis results. We have included text in the Figure 3 legend to clarify this.

In addition, we’ve added more text discussing the Multibody Refinement variance results and added Cryosparc’s 3D Variability Analysis to further validate the movements we observed (subsection “Hinge movement arising from the ORC3·5 interface”). We have added Video 3 that illustrates the movements generated from Multibody Refinement and 3D Variability Analysis.

5) DNA-bound structure:

It makes sense that the DNA sequence would not be in register because HsORC isn't sequence specific, but is the major/minor groove positioning in register? I would think that the positioning of the double helix is essentially fixed relative to the major/minor grooves. Can the authors make out DNA helix clearly? Clearer figures would help.

We can roughly see the backbone but the resolution is too low. Even though we collected 2M particles we cannot resolve the various DNA conformations by 3D classification because of the number of particles remaining (see Figure 4—figure supplement 3). We therefore decided to report conservatively on our conclusions of the ORC-DNA binding. We have added a figure of the EM map covering the DNA in Figure 4—figure supplement 1 to better clarify the model and map of the DNA region.

6) Sample preparation details:

In the sample preparation, addition of ATP in the buffer is mentioned. Were there any divalent cations, Mg2+ or Mn2+, included?

We did not include divalent cations in our purification buffers. However, the Mg2+ observed likely came from the lysate during purification. We included this information in the text (subsection “Overall HsORC structure and comparison to ORC from Saccharomyces cerevisiae and Drosophila melanogaster”) and Figure 1 legend.

What is the rationale for the ATP addition instead of ATPγS, AMPPNP or ADP·BFx?

Based on our previous study (Tocilj et al., 2017), we assumed ATP would engage ORC in an active state and we assumed this stabilized the complex. We attempted a data collection on an ORC sample that included ATPγS, but our best dataset was, in fact, with ATP.

What are the green spheres shown next to the ATP molecules in Figure 1G-I?

The green spheres are our interpretation of the density near the ATP and agrees with the positioning of Mg2+ as seen in many AAA+ ATPases. We have clarified this in the text (subsection “Overall HsORC structure and comparison to ORC from Saccharomyces cerevisiae and Drosophila melanogaster”) and Figure 1 legend.

Why is ATP not hydrolyzed in ORC1·4 or ORC4·5?

We presume that some of the ATP at the ORC1·4 site is hydrolyzed and the ORCO2WH conformation is, in fact, the result of this hydrolysis, but not the others. It should be noted that these four conformations come from the same sample so some particles might have hydrolyzed ATP and some haven’t. The ATPase at the ORC4·5 interface is not functional based on our previous work (see Tocilj et al.), and would therefore not hydrolyze ATP. We also note, from work not yet published that the ATP is tightly bound at the ORC4·5 and is still clearly present from lysates even when ATP is not provided in the sample.

7) In Figure 1—figure supplement 3D and E, the maps look of much lower resolution than 3.7 Å (ORCOPEN) or 4.3 Å (ORCDNA). Were these maps unsharpened? The authors may consider showing representative EM map regions or 3D-FSC.

We also noticed the maps resolution appears lower than the reported values. Therefore, these maps were not sharpened. We performed 3DFSC and included plots in Figure 1—figure supplement 6.

8) DNA density:

For the ORCDNA complex, it will be important to show an EM map covering the DNA portion. Although base sequences are mixed, the DNA backbones should still be recognizable.

The DNA backbone is somewhat recognizable. The resolution is low in this region due to multiple conformations of loosely bound DNA (Figure 4—figure supplement 3) and we cannot isolate individual conformations by 3D classification because the number of particles (<40k) in each class are too low to perform further classification.

We have added a figure of the EM map covering the DNA in Figure 4—figure supplement 1.

9) Resolution anisotropy:

The resolution is extremely directionally dependent, and the true resolution of the ORC open and ORC DNA maps is low in all but one direction. Based on the direction plots shown in Figure 1—figure supplement 3 all of the samples suffered somewhat from preferential orientation. I believe the authors should mention this, and perform a direction dependent resolution analysis of their maps (e.g. using the 3D FSC server – https://3dfsc.salk.edu/) and report the results.

We performed 3DFSC and included plots in Figure 1—figure supplement 6. We also mentioned the analysis in the manuscript and referenced the work.

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Data Citations

    1. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O1AAA. RCSB Protein Data Bank. 7JPO
    2. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O1AAA. Electron Microscopy Data Bank. EMD-22417
    3. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2WH. RCSB Protein Data Bank. 7JPP
    4. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2WH. Electron Microscopy Data Bank. EMD-22418
    5. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2-5. RCSB Protein Data Bank. 7JPQ
    6. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2-5. Electron Microscopy Data Bank. EMD-22419
    7. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-OPEN. RCSB Protein Data Bank. 7JPR
    8. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-OPEN. Electron Microscopy Data Bank. EMD-22420
    9. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-DNA. RCSB Protein Data Bank. 7JPS
    10. Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-DNA. Electron Microscopy Data Bank. EMD-22421

    Supplementary Materials

    Supplementary file 1. CryoEM and model statistics of all ORC structures.
    elife-58622-supp1.docx (32.6KB, docx)
    Transparent reporting form
    elife-58622-transrepform.docx (245.8KB, docx)

    Data Availability Statement

    All coordinates and cryoEM maps have deposited in the PDB and EMDB: ORC-O1AAA: PDB code: 7JPO EMDB code: EMD-22417 ORC-O2WH: PDB code: 7JPP EMDB code: EMD-22418 ORC-O2-5: PDB code: 7JPQ EMDB code: EMD-22419 ORC-OPEN: PDB code: 7JPR EMDB code: EMD-22420 ORC-DNA: PDB code: 7JPS EMDB code: EMD-22421.

    The following datasets were generated:

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O1AAA. RCSB Protein Data Bank. 7JPO

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O1AAA. Electron Microscopy Data Bank. EMD-22417

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2WH. RCSB Protein Data Bank. 7JPP

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2WH. Electron Microscopy Data Bank. EMD-22418

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2-5. RCSB Protein Data Bank. 7JPQ

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-O2-5. Electron Microscopy Data Bank. EMD-22419

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-OPEN. RCSB Protein Data Bank. 7JPR

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-OPEN. Electron Microscopy Data Bank. EMD-22420

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-DNA. RCSB Protein Data Bank. 7JPS

    Jaremko MJ, On KF, Thomas DR, Stillman B, Joshua-Tor L. 2020. ORC-DNA. Electron Microscopy Data Bank. EMD-22421

    Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd

    RESOURCES