Structure of the Saccharolobus solfataricus GINS tetramer

DNA replication is tightly regulated to ensure genomic stability and prevent several diseases, including cancers. Eukaryotes and archaea partly achieve this regulation by strictly controlling the activation of hexameric minichromosome maintenance (MCM) helicase rings that unwind DNA during its replication. In eukaryotes, MCM activation critically relies on the sequential recruitment of the essential factors Cdc45 and a tetrameric GINS complex at the onset of the S-phase to generate a larger CMG complex. We present the crystal structure of the tetrameric GINS complex from the archaeal organism Saccharolobus solfataricus (Sso) to reveal a core structure that is highly similar to the previously determined GINS core structures of other eukaryotes and archaea. Using molecular modeling, we illustrate that a subdomain of SsoGINS would need to move to accommodate known interactions of the archaeal GINS complex and to generate a SsoCMG complex analogous to that of eukaryotes.

1. Introduction

DNA replication is the fundamental life process of all organisms where genetic material is duplicated in preparation for cell division. This process is tightly regulated to ensure that DNA is fully and faithfully replicated only one time each cell division (Bell & Labib, 2016; Stillman, 2005). Disruptions that over-replicate, under-replicate or inaccurately replicate DNA lead to several serious diseases, including cancers (Bell & Labib, 2016; Stillman, 2005). One way that cells maintain this strict control is via the cell cycle, which temporally determines when cells begin to synthesize new DNA and when it safe to segregate the copied DNA to daughter cells (Bell & Labib, 2016; Stillman, 2005).

The molecular basis for the initiation of DNA replication involves local melting of DNA where some base pairs are opened to allow the establishment of an active helicase–DNA species that becomes competent to unwind and separate the two strands of the DNA double helix (Bell & Labib, 2016). The separated strands are used by DNA polymerases in the synthesis of new DNA. Bacteria and some viruses separate the two strands and then load a six-membered ring helicase to encircle the exposed single-stranded DNA (ssDNA) to initiate DNA unwinding (Kaguni, 2011; Bell & Kaguni, 2013). In contrast, eukaryotes and archaea load a six-membered ring helicase, the minichromosome maintenance (MCM) complex, to encircle DNA and subsequently activate this MCM–DNA species to become competent for unwinding DNA (Bell & Labib, 2016). In the case of yeast (Saccharomyces cerevisiae, Sc), and likely all eukaryotes, this activation critically relies on the sequential recruitment of Cdc45 and the tetrameric GINS complex (Moyer et al., 2006). Cdc45 and GINS factors interact directly with the MCM complex and with each other to form a larger CMG complex (Cdc45–MCM–GINS; Yuan et al., 2016). Following the recruitment of both Cdc45 and GINS, initial DNA unwinding is observed biochemically (Douglas et al., 2018) and structurally (Lewis et al., 2022).

Archaea also use a six-membered MCM ring to unwind DNA (Kelman et al., 2020), and archaea possess GINS and Cdc45 homologs (Marinsek et al., 2006; Makarova et al., 2012) that stimulate the activities of MCM in biochemical DNA-unwinding assays (Yoshimochi et al., 2008; Xu et al., 2016; Lang & Huang, 2015). Crystal structures of the archaeal proteins confirmed that archaeal RecJ (Oyama et al., 2016; Li et al., 2017; Oki et al., 2022) and archaeal GINS (Oyama et al., 2011) are indeed structural homologs of eukaryotic Cdc45 (Simon et al., 2016) and GINS (Choi et al., 2007; Kamada et al., 2007; Chang et al., 2007), respectively, suggesting fundamentally conserved mechanisms of action. Several structures have been determined of eukaryotic CMG from multiple organisms and in multiple configurations to illuminate a wealth of mechanistic detail (Yuan et al., 2016; Georgescu et al., 2017; Eickhoff et al., 2019; Rzechorzek et al., 2020; Baretić et al., 2020; Yuan et al., 2020; Lewis et al., 2022; Xia et al., 2023; Jenkyn-Bedford et al., 2021; Cvetkovic et al., 2023; Henrikus et al., 2024). Structures from archaeal organisms have been limited, and no structural details of potential interactions between archaeal MCM and either GINS or RecJ have been elucidated. As a stepping stone towards elucidating a potential archaeal CMG complex, we have determined the crystal structure of the tetrameric GINS complex from the archaeal organism Saccharolobus solfataricus (Sso), an organism for which we have determined multiple MCM structures in several states using X-ray crystallography (Miller et al., 2014; Meagher et al., 2019) and cryo-EM (Meagher et al., 2022). The core of the SsoGINS structure is highly conserved with other GINS structures determined from eukaryotic and archaeal organisms. One subdomain adopts an apparently stable conformation that would need to rearrange to permit the assembly of a CMG complex analogous to that of eukaryotes.

2. Materials and methods

2.1. Macromolecule production

The two genes encoding the GINS proteins of S. solfataricus were amplified from genomic DNA (ATCC 35092D-5) by polymerase chain reaction (PCR) and cloned into the two expression sites of pETDuet (Novagen). The gene encoding SsoGINS23 (NCBI AZF71192.1) was cloned between the BamHI and SalI sites of multiple cloning site 1, which provided an N-terminal hexahistidine tag. The gene encoding SsoGINS51 (AAK41311.1) was cloned between the NdeI and XhoI sites of multiple cloning site 2. The hexahistidine tag was genetically deleted from the construct by PCR amplification of the plasmid (see Table 1), including one phosphorylated primer, followed by circular ligation and transformation to yield the expression plasmid used in this study (pEE093).

Table 1
Macromolecule-production information

Source organism	Saccharolobus solfataricus
DNA source	Genomic DNA (ATCC 35092D-5)
Forward-gins51 (NdeI)	gcagggcatatgTTAGATGAATTGGTTAAAAAGGAATTATCAGAGGAAGAAATTACTGAAATTAAGCTAGAGGAAATAATCAAGT
Reverse-gins51 (XhoI)	ccgaggctcgagTCATAATTGTTCTTCAATATCTATCTTATAGGGAGTTAAATAACTTGCTATTATTAATGGCAAGGCTTCTCT
Forward-gins23 (BamHI)	gcagggggatccgATGATCGAAGTAAAACTCAGAGCTATTAAAAGATTGTCAAATGTTTACACTCGCC
Reverse-gins23 (SalI)	ccgagggtcgacCTAGGAATTTCCAATTATATCACCATATAATTCCTTAATAAGTTGTTTTATTGTCTTATAGATTAAAAGCTCTTCTTCTGTCATTCC
pETDuet-MCS1-R	[PHOS]GGTATATCTCCTTCTTAAAGTTAAACAAAATTATTTCTAGAGGGGA
pSSOGINS23-F1	ATGATCGAAGTAAAACTCAGAGCTATTAAAAGATTGTCAAATG
Cloning vector	pETDuet
Expression vector	pETDuet
Expression host	Escherichia coli BL21(DE3)-RIPL
Complete amino-acid sequence of the construct produced
GINS51	MLDELVKKELSEEEITEIKLEEIIKYITLIKKSKTFVSSEIRKEELKFLSELAESLFELRLSKVLEGKVGKGFDEFIFDIFKILKQFYVDLLTGRYIIYNDKIYCIVQKPLIYNDHRVNEGDVLVLPMREALPLIIASYLTPYKIDIEEQL
GINS23	MIEVKLRAIKRLSNVYTRRVMIIEDWNGSSITTGNIELVKGSENQLPQWLAIILEGKKVAKIEDKISIEDLGRILFQERQNMNTPASLVPLGKDFTSRVQLYLETLRKDNNVESLEKLRKSIGILNEIIKIRLRKLIQLAFLNIDDQNLINGMTEEELLIYKTIKQLIKELYGDIIGNS

Plasmid pEE093 was freshly transformed into Escherichia coli BL21(DE3)-RIPL cells and seeded into a 100 ml overnight culture in LB medium supplemented with 2% glucose and 50 mg l⁻¹ ampicillin. Following overnight growth, the 100 ml culture was distributed into 6 l LB medium supplemented with 0.4% glucose and 50 mg l⁻¹ ampicillin. The cultures were grown in a shaker/incubator at 37°C until the OD reached 0.4, and the temperature was reduced to 18°C. When the OD of the culture had reached 0.6–0.9, isopropyl β-D-1-thiogalactopyranoside (IPTG) was added to a 0.4 mM concentration to induce expression of the two proteins. After 18 h at 18°C, the cells were harvested by centrifugation and were resuspended in buffer 1 (50 mM Tris pH 8.3, 250 mM NaCl, 10% glycerol, Roche EDTA-free protease inhibitor). The cells were lysed on ice with one pass through a microfluidizer at 62 MPa. The cellular lysate was heated in a water bath for 15 min at 80°C and the soluble fraction was isolated by centrifugation. The sodium chloride concentration was increased to 1 M and 10% polyethylenimine–HCl was added to a 0.3% final concentration to precipitate nucleic acids. The soluble fraction was isolated by centrifugation and subjected to ammonium sulfate precipitation at 4°C by adding 43.6 g ammonium sulfate per 100 ml of solution. Precipitated proteins were isolated by centrifugation, resuspended in buffer 1 and purified by size-exclusion chromatography on an S200 column. The NaCl concentration was decreased by dilution with buffer 2 (20 mM Tris pH 8.3, 50 mM NaCl, 5% glycerol, 2 mM β-mercaptoethanol) and the solution was applied onto a Mono Q column for anion-exchange chromatography. The sample was eluted with a linear gradient from buffer 2 to buffer 3 (20 mM Tris pH 8.3, 1 M NaCl, 5% glycerol, 2 mM β-mercaptoethanol). SDS–PAGE of the fractions indicated two co-eluting bands of similar intensity. Pooled fractions (2 ml) were dialyzed against 20 mM Tris pH 8.3, 100 mM NaCl and concentrated to 11 mg ml⁻¹ for crystallization (based on absorbance at 280 nm; tetramer ɛ₂₈₀ = 57 760 M⁻¹ cm⁻¹; tetramer molecular weight of 76 870.34).

2.2. Crystallization

Crystallization screening was performed with an SPT Labtech Mosquito LCP drop-setting robot using the hanging-drop method with 200 nl protein solution mixed with 200 nl well solution. Crystals grew in 96-well plates over a few days using multiple commercial screens in a Formulatrix plate incubator/hotel at 18°C. Crystals from JCSG solution G1 [0.18 M triammonium citrate, 20%(w/v) PEG 3350] provided suitably large crystals for harvesting and subsequent collection of diffraction data (Fig. 1, Table 2).

Table 2
Crystallization

Method	Vapor diffusion, hanging drop
Plate type	96-well
Temperature (K)	291
Protein concentration (mg ml^−1)	11 (based on absorbance at 280 nm)
Buffer composition of protein solution	20 mM Tris pH 8.3, 100 mM NaCl
Composition of reservoir solution	0.18 M triammonium citrate, 20%(w/v) PEG 3350.
Volume and ratio of drop	200 nl:200 nl
Volume of reservoir (µl)	100

Figure 1 Crystals of SsoGINS imaged using a Formulatrix Rock Imager 1000 plate incubator/hotel. The crystallization solution consisted of 0.18 M tri­ammonium citrate, 20%(w/v) PEG 3350.

2.3. Data collection and processing

Single crystals were harvested from crystallization drops, quickly passed through 1:3 ethylene glycol:well solution and flash-cooled in liquid nitrogen. Cooled crystals were shipped to SER-CAT beamline 22-ID at the Advanced Photon Source (APS) in a dry shipper for evaluation and data collection. Data were collected from two positions of one crystal using an EIGER 16M detector. Each collection consisted of 360° of total oscillation in 0.25° oscillation widths and 0.1 s exposures with 13.531% beamline transmission. The two data sets were processed, scaled and merged using HKL-2000 (Otwinowski & Minor, 1997) in primitive hexagonal point group 6 with a high-resolution limit of 2.3 Å and 37-fold overall average redundancy (Table 3). The systematic absence of the odd 00l reflections strongly suggested space group P6₃.

Table 3
Data collection and processing

Values in parentheses are for the outer shell.

Diffraction source	SER-CAT beamline 22-ID, APS
Wavelength (Å)	1.0
Temperature (K)	100
Detector	EIGER 16M
Crystal-to-detector distance (mm)	300
Rotation range per image (°)	0.25
Total rotation range (°)	360 × 2
Exposure time per image (s)	0.1
Space group	P63
a, b, c (Å)	180.24, 180.24, 51.20
α, β, γ (°)	90, 90, 120
Mosaicity (°)	0.107–0.283
Resolution range (Å)	50–2.30 (2.34–2.30)
Total No. of reflections	1581846
No. of unique reflections	42463 (1957)
Completeness (%)	99.2 (91.5)
Multiplicity	37.3 (19.0)
〈I/σ(I)〉	34.5 (1.0)†
Rr.i.m.	0.013 (0.224)
Overall B factor from Wilson plot (Å^2)	52.69

†〈I/σ(I)〉 is 1.5 in the 2.38–2.34 Å bin, 1.5 in the 2.43–2.38 Å bin and 4 in 2.48–2.43 Å bin.

2.4. Structure solution and refinement

The structure was solved by molecular replacement with Phaser (McCoy et al., 2007), which placed two copies of a monomer of SsoGINS23 and two copies of a monomer of SsoGINS51 in space group P6₃ (Table 4). The monomeric search models used for molecular replacement were taken from an AlphaFold-Multimer (Evans et al., 2021) prediction of an (SsoGINS23)₂(SsoGINS51)₂ tetramer. The structure was iteratively refined in Phenix (Liebschner et al., 2019; Afonine et al., 2012) and rebuilt in Coot (Casañal et al., 2020; Emsley et al., 2010; Emsley & Cowtan, 2004). The final refinement included 24 autogenerated TLS groups (four chain A, six chain B, eight chain C and six chain D) and secondary-structure restraints. The final structure has an r.m.s.d. of 2.193 Å (2844 matched atoms) from the AlphaFold-Multimer (Evans et al., 2021) model that provided the molecular-replacement search models as calculated in PyMOL (Schrödinger).

Table 4
Structure refinement

Values in parentheses are for the outer shell.

Resolution range (Å)	31.22–2.30 (2.36–2.30)
Completeness (%)	91.1
σ Cutoff	All data
No. of reflections, working set	37080 (2256)
No. of reflections, test set	1904 (112)
Final Rcryst	0.220 (0.3043)
Final Rfree	0.243 (0.3716)
Cruickshank DPI	n.d.
No. of non-H atoms
Protein	5304
Ligand	0
Water	5
R.m.s. deviations
Bond lengths (Å)	0.002
Angles (°)	0.386
Average B factors (Å^2)
Protein	79.6
Ramachandran plot
Most favored (%)	98.43
Allowed (%)	1.57

3. Results and discussion

3.1. Overall architecture

The structure adopts a tetrameric architecture, with two copies of GINS51 and two copies of GINS23 in an approximately twofold-symmetric complex. The structure is highly consistent with other GINS structures (Fig. 2) despite relatively low sequence identity. SsoGINS51 has 21.7% sequence identity to Thermococcus kodakarensis (Tk) GINS51, 16% identity to ScSld5 and 15.6% identity to ScPsf1. SsoGINS23 has 21.2% sequence identity to TkGINS23, 21.1% identity to ScPsf2 and 18.8% identity to ScPsf3. Each subunit has two subdomains, with a larger α-helical subdomain (A) and a smaller subdomain that contains β-strands (B), consistent with the structures of other archaeal and eukaryotic GINS protein structures (Oyama et al., 2011; Choi et al., 2007; Kamada et al., 2007; Chang et al., 2007; Yuan et al., 2016). The order of the A and B subdomains is reversed between the two subunits, with the A subdomain first in SsoGINS51 and the B subdomain first in SsoGINS23 (Fig. 2). These subdomain orders are consistent with those of other eukaryotic and archaeal GINS structures (Oyama et al., 2011; Choi et al., 2007; Kamada et al., 2007; Chang et al., 2007; Yuan et al., 2016) and with the A/B topologies predicted for the SsoGINS subunits (Xu et al., 2016).

Figure 2 (a) The structure of the SsoGINS tetramer is an approximately twofold-symmetric assembly of two GINS51 subunits (green and cyan) and two GINS23 subunits (magenta and yellow). Both subunit types consist of two subdomains; one is fully α-helical (A) and the other contains β-strands (B). These subdomains occur in reverse order for the two subunit types. The secondary, tertiary and quaternary structures are highly similar to those previously observed for the GINS complexes from the archaeon Thermococcus kodakarensis (b) (PDB entry 3anw; Oyama et al., 2011) and the eukaryotic yeast (c) (PDB entry 7z13; Lewis et al., 2022). The structurally conserved core consists of the full GINS23 subunits (yeast Psf2 and Psf3) and the A subdomains of GINS51 (yeast Sld5 and Psf1). The B subdomains of GINS51 (yeast Sld5 and Psf1) occupy variable positions, indicating that this subdomain may be mobile. Structure images were prepared with PyMOL (Schrödinger) and secondary-structure diagrams were prepared with the Google Colab implementation of SSDraw (Chen & Porter, 2023). The SsoGINS51A subdomains uniquely form a significant dimeric interaction.

The core tetrameric quaternary structure of the GINS complexes of eukaryotes and archaea (Fig. 2) is highly conserved. In the archaeal structures, the core consists of both complete GINS23 subunits and the two GINS51A sub­domains. In eukaryotic GINS, the core consists of the complete Psf2 and Psf3 subunits and the helical A subdomains of Sld5 and Psf1. The underlying subdomain structure of GINS51B, Sld5B and Psf1B is conserved, but its placement relative to the GINS tetramer core varies among the structures. In the case of SsoGINS, the two GINS51B subdomains have a unique and sizable dimeric interaction surface (at the top of the complex in the central panel of Fig. 2). This placement was also predicted by AlphaFold-Multimer (Evans et al., 2021), as illustrated in the discussion below.

3.2. The SsoGINS51B dimeric interface is incompatible with simultaneous polymerase interaction

The dimeric association of the two SsoGINS51B sub­domains was intriguing and was investigated further. In eukaryotes and archaea, the corresponding subdomain interacts structurally with the key replication factors Cdc45/RecJ and with DNA polymerase. In the archaeon T. kodakarensis, the GINS51B subdomain interacts with RecJ and with the DNA polymerase II small subunit (PDB entry 7e15; Oki et al., 2022). In the eukaryotic CMG structure, the GINS Psf1B subdomain similarly interacts with Cdc45 and with DNA polymerase epsilon subunit B (PDB entry 7z13; Lewis et al., 2022). The crystal structure of TkGINS51–RecJ–Pol was superimposed on one SsoGINS51B subdomain to investigate the compatibility of similar interactions within the SsoGINS tetramer (Fig. 3). This superposition suggested that a RecJ subunit could similarly interact with SsoGINS51B without clashes, but the polymerase subunit would fully clash with the dyad-related SsoGINS51B subdomain (magenta circle in Fig. 3). We therefore conclude that the dimeric interface of the SsoGINS51B subdomains of the crystal structure is not compatible with the simultaneous adoption of an interaction with polymerase as identified for TkGINS.

Figure 3 The dimeric interaction of the SsoGINS51A subdomains is incompatible with formation of the interaction with polymerase illustrated for TkGINS51B–RecJ–Pol. The GINS51B subunit (purple) of the TkGINS51B–RecJ–Pol structure (PDB entry 7e15; Oki et al., 2022) was superimposed on one SsoGINS51B subdomain of the SsoGINS crystal structure to investigate the compatibility of similar interactions. Following this superposition, the RecJ subunit (salmon) did not clash with any component of SsoGINS, but the DNA polymerase II small subunit severely clashed with the dyad-related SsoGINS51B subunit. Hence, the SsoGINS51B dimeric interface is not compatible with simultaneous interaction with DNA polymerase in the fashion of TkGINS–RecJ–Pol. Superpositions and figures were prepared in PyMOL (Schrödinger).

3.3. AlphaFold structure prediction of SsoGINS–RecJ

The superposition of the structure of TkGINS51–RecJ–Pol onto the crystal structure of SsoGINS suggested that the SsoGINS structure, including the GINS51B subdomain positions, is compatible with placing SsoRecJ analogous to the structure of TkGINS51–RecJ–Pol (Oki et al., 2022). In yeast, GINS and Cdc45 interact with each other following their sequential recruitment (Douglas et al., 2018) to the MCM ring to form a CMG complex (Moyer et al., 2006; Yuan et al., 2016). To explore the SsoGINS–RecJ interaction that potentially exists within an assembled CMG complex, the SsoGINS–RecJ structure was predicted with AlphaFold-Multimer (Evans et al., 2021). The prediction used two subunits of SsoGINS51, two subunits of SsoGINS23 and two subunits of SsoRecJ (NCBI WP_009990587.1). It used a database cutoff of 20 January 2023. The predicted structure (Fig. 4) was high confidence in the underlying structures and in their relative placements (ptm 0.831219297780319; iptm 0.81859624; rank 0.8211208546427338) and had strong positional correlation between the GINS complex and RecJ based on the predicted alignment error plot. The GINS portion of the structure strongly resembles the present crystal structure (compare with Fig. 2). In the predicted structure, the RecJ subunits interact with SsoGINS51B, occupying positions that are very similar to those observed when TkGINS51–RecJ–Pol was superimposed on the SsoGINS51B subdomain of the crystal structure (Fig. 3). The interaction between SsoGINS51A and SsoRecJ in the predicted structure is very similar to the interactions in other archaeal and eukaryotic structures. All involve the addition of strands of the GINS B subdomain to extend the β-sheet of the first domain of Cdc45/RecJ. Hence, the SsoGINS–RecJ interaction of the AlphaFold-Multimer model is consistent with the interactions observed in both archaea and eukaryotes and would not sterically clash with the conformation of the SsoGINS crystal structure.

Figure 4 The predicted structure of SsoGINS–RecJ shows interaction between SsoGINS51B similar to the interactions observed in TkGINS–RecJ and ScGINS. (a) The AlphaFold-Multimer (Evans et al., 2021) prediction of the SsoGINS2RecJ2 structure shows an interaction between RecJ and SsoGINS51B in a complex that is similar to the superposition in Fig. 3. (b) The mode of interaction is similar to that of archaeal and eukaryotic structures. For each case, the two-stranded antiparallel sheet of the B subdomain (light green) forms an extended sheet with the first domain of Cdc45/RecJ (rainbow). Superpositions and figures were prepared in PyMOL (Schrödinger).

3.4. Modeled changes for compatibility with eukaryotic Mcm2-7–Cdc45–GINS

To assess the interactions among the components of a potential SsoCMG complex, an AlphaFold-Multimer prediction was generated using six copies of the SsoMCM N-terminal domain, two copies of SsoGINS51, two copies of SsoGINS23 and two copies of SsoRecJ. The prediction used a database cutoff of 20 January 2023. The MCM N-terminal domain was used rather than full-length MCM to save computational time and because Mcm2-7 interacts with Cdc45–GINS exclusively via the Mcm2-7 N-terminal domains in the yeast CMG complex. The predicted model (ptm 0.5662312904705491; iptm 0.5284666; rank 0.5360195239327329) is highly similar to ScCMG in the interactions between MCM and GINS (Fig. 5a). In the predicted model, the SsoGINS51B–RecJ positions match those of the SsoGINS–RecJ prediction (see Fig. 4), which places the RecJ subunit at the opposite end of GINS from the MCM ring and not adjacent.

Figure 5 Generation of a model for SsoCMG. (a) A model for the structure of the SsoMCM N-terminal domain hexamer bound to SsoGINS and SsoRecJ predicted by AlphaFold-Multimer (Evans et al., 2021) is illustrated on the left. The ScCMG portion of PDB entry 7z13 (Lewis et al., 2022) is illustrated on the right and the overlay of the two is shown in the center. The MCM subunits are colored purple, blue, green, yellow, orange and red. The GINS subunits are colored yellow, magenta, green and cyan in the same scheme as the other figures. RecJ/Cdc45 subunits are in salmon and light gray as in the other figures. The MCM–GINS portions are consistent between the two structures. The SsoRecJ subunit interacts with GINS51B as in Fig. 4(a), which places it opposite the MCM ring rather than adjacent as in ScCMG. (b) A model for SsoCMG was generated from the structures in (a) by superimposing one of the SsoGINS51B–RecJ units (green and salmon) on the Psf1B and Cdc45 subunits of ScCMG and regularizing the SsoGINS51 linker. The ATPase domains of SsoMCM were added by superimposing the structure of the nearly full-length SsoMCM hexamer crystal structure (PDB entry 6mii; Meagher et al., 2019). Superpositions and figures were prepared in PyMOL (Schrödinger).

Superpositions with existing structures generated a potential model for SsoCMG that included the MCM ATPase domains and placed the RecJ unit adjacent to the MCM ring as in ScCMG. The SsoCMG prediction (above) was first aligned with ScCMG (PDB entry 7z13; Lewis et al., 2022) based on all six OB-folds of the MCM subunits, which generated the superposition in Fig. 5(a). Next, one Sso51B–RecJ unit was superimposed on the Psf1–Cdc45 unit of ScCMG. The linker between the 51A and 51B subdomains was regularized in Coot (Casañal et al., 2020; Emsley et al., 2010; Emsley & Cowtan, 2004) to illustrate that the linker was sufficiently long to allow the envisioned repositioning of Sso51B–RecJ. Lastly, a nearly full-length crystal structure of SsoMCM (PDB entry 6mii; Meagher et al., 2019) was superimposed on the AlphaFold model based on all six OB-folds to provide positions of the ATPase domains. The final model (Fig. 5b) represents our best prediction for a potential SsoCMG complex analogous to that of eukaryotes. The AlphaFold-predicted SsoCMG model of Fig. 5(a) was morphed to the eukaryotic-like SsoCMG model of Fig. 5(b) with PyMOL (Schrödinger) using its rigimol routine to help illustrate the molecular difference of the Sso51B–RecJ positions (Supplementary Video S1). The 51B subdomain center of mass in the form in Fig. 5(a) differs from that in Fig. 5(b) by 34.4 Å. Center of masses were calculated in PyMOL (Schrödinger). Although it is not known whether an SsoCMG structure would closely match eukaryotic CMG structures, this analysis shows that the 51B subdomain would need to move a large distance from its crystal structure position to accommodate a eukaryote-like CMG structure.

4. Concluding remarks

The large movement described above for the 51B sub­domain is currently speculative because the specific structural form of a potential SsoCMG complex is not known. The two predicted structural arrangements strongly differ in the proximity of two 51B subdomains, which provides a basis for a fluorescence-quenching assay similar to those used to study ring opening in DNA sliding clamps (Paschall et al., 2011; Thompson et al., 2012). For such an assay, a fluorescent label could be incorporated into the 51B subdomain close to the twofold-symmetric interface of the crystal structure. In the twofold-symmetric form (as in Fig. 5a) of the crystal structure (Fig. 2), the fluorescent groups of the two 51B subdomains are expected to self-quench due to their close proximity. In a eukaryotic-like CMG form (as in Fig. 5b), measurable fluorescence is anticipated because the two 51B subdomains would be far apart. This loss-of-quenching assay could potentially allow dissection of the mobility of 51B subdomains and the biochemical events that lead to mobility, and is potentially adaptable to a single-molecule format.

Footnotes

‡Present address: Cancer Research UK Cambridge Institute, University of Cambridge, Li Ka Shing Centre, Cambridge CB2 0RE, United Kingdom.