Crystal structure of an archaeal CorB magnesium transporter

Chen, Yu Seby; Kozlov, Guennadi; Moeller, Brandon E.; Rohaim, Ahmed; Fakih, Rayan; Roux, Benoît; Burke, John E.; Gehring, Kalle

doi:10.1038/s41467-021-24282-7

Download PDF

Article
Open access
Published: 29 June 2021

Crystal structure of an archaeal CorB magnesium transporter

Nature Communications volume 12, Article number: 4028 (2021) Cite this article

4559 Accesses
22 Citations
8 Altmetric
Metrics details

Subjects

Abstract

CNNM/CorB proteins are a broadly conserved family of integral membrane proteins with close to 90,000 protein sequences known. They are associated with Mg²⁺ transport but it is not known if they mediate transport themselves or regulate other transporters. Here, we determine the crystal structure of an archaeal CorB protein in two conformations (apo and Mg²⁺-ATP bound). The transmembrane DUF21 domain exists in an inward-facing conformation with a Mg²⁺ ion coordinated by a conserved π-helix. In the absence of Mg²⁺-ATP, the CBS-pair domain adopts an elongated dimeric configuration with previously unobserved domain-domain contacts. Hydrogen-deuterium exchange mass spectrometry, analytical ultracentrifugation, and molecular dynamics experiments support a role of the structural rearrangements in mediating Mg²⁺-ATP sensing. Lastly, we use an in vitro, liposome-based assay to demonstrate direct Mg²⁺ transport by CorB proteins. These structural and functional insights provide a framework for understanding function of CNNMs in Mg²⁺ transport and associated diseases.

Cryo-EM structures of human zinc transporter ZnT7 reveal the mechanism of Zn2+ uptake into the Golgi apparatus

Article Open access 08 August 2023

Dissecting the conformational complexity and mechanism of a bacterial heme transporter

Article Open access 24 April 2023

Cryo-EM structures of human magnesium channel MRS2 reveal gating and regulatory mechanisms

Article Open access 08 November 2023

Introduction

Magnesium (Mg²⁺), the most abundant divalent cation inside cells, is essential for a wide variety of biochemical processes, such as energy metabolism, maintenance of genomic stability, protein synthesis, and over 600 enzymatic reactions¹. CNNMs (CBS-pair domain divalent cation transport mediators) are a conserved family of integral membrane proteins implicated in Mg²⁺ homeostasis and divalent cation transport². CNNM2 and CNNM4 are found abundantly in the basolateral membrane of kidney and colon epithelial cells, where renal/intestinal (re)absorption of Mg²⁺ occurs^3,4. In humans, mutations in CNNM proteins are linked to two genetic diseases: CNNM2 mutations cause hypomagnesemia^3,5 while mutations in CNNM4 are associated with Jalili syndrome⁶. CNNM2-knockout mice are embryonic lethal⁷, while loss of CNNM4 leads to male infertility⁸ and susceptibility to cancer^9,10. CNNMs are additionally implicated in hypertension⁷, non-alcoholic steatohepatitis¹¹, and schizophrenia¹².

CNNMs are also known as ancient conserved domain proteins (ACDPs) because their core domains are evolutionarily conserved in essentially all organisms, from animals, plants, fungi, and archaea to bacteria (Fig. 1a)¹³. The bacterial CorB protein was first identified in a screen for cobalt-resistant mutants in Salmonella typhimurium¹⁴. CorB was proposed to mediate Mg²⁺ efflux together with CorC and CorD, in which CorC is a soluble protein that shares high sequence similarity to the cytosolic domains of CorB. Subsequently, many other CorB orthologs have been implicated in Mg²⁺ transport. For example, the Staphylococcus aureus ortholog, MpfA, is thought to function as a Mg²⁺ exporter as deletion mutants are unable to grow in the presence of high concentrations of magnesium^15,16. Disruption of the homologous gene (yhdP) in Bacillus subtilis leads to increased cellular Mg²⁺ content, again supporting a role in Mg²⁺ efflux¹⁷.

**Fig. 1: Overall structure and domain organization of CNNM/CorB Mg²⁺ transporters.**

Despite the clear association with Mg²⁺ transport, the field has not yet reached a consensus on the molecular mechanism of CNNM/CorB proteins^18,19. Some papers report direct efflux of Mg²⁺ through Na⁺-coupled exchange⁴, others observed direct Mg²⁺ influx^20,21, while still others propose an indirect mechanism through regulation of other Mg²⁺ channels (e.g., TRPM6/7)⁵. As all the studies to date have used measurements in cells^{3,4,5,15,21,22,23}, which preclude differentiation of direct or indirect mechanisms, it remains unclear whether CNNM/CorB proteins are themselves Mg²⁺ transporters or regulators of other Mg²⁺ transporters/channels.

Structurally, CNNM/CorB proteins are defined by a conserved core consisting of a transmembrane domain (TMD) and a cytosolic cystathionine-β-synthase (CBS)-pair domain (Fig. 1b)²⁴. The TMD constitutes the largest family of protein domains of unknown function (DUF21) in the Pfam database²⁵. The CBS-pair domains are composed of tandem CBS motifs and found in a wide variety of proteins, including the bacterial Mg²⁺ channel MgtE, where they mediate dimerization and Mg²⁺-ATP binding²⁶. The CBS-pair domains of CNNMs have been characterized structurally and shown to undergo conformational changes upon Mg²⁺-ATP binding^27,28. Outside of the core, eukaryotic CNNM and prokaryotic CorB diverge. Eukaryotes have an extracellular domain and a cytosolic cyclic-nucleotide binding homology (CNBH) domain²⁹. Prokaryotes lack the extracellular domain and have a distinct C-terminal CorC domain. Eukaryotic CNNMs contain an additional loop in the CBS-pair domain that allows binding of oncogenic phosphatase of regenerating liver (PRL) to regulate CNNM activities^30,31,32. Only the soluble domains of CNNM/CorB have been characterized structurally until recently, during the preparation of this manuscript, the TMD fragment of a bacterial CorB ortholog (TpCorB) was reported²³.

Here, we determine the crystal structure of an archaeal CorB ortholog in two distinct conformations (apo and Mg²⁺-ATP-bound). The TMD exists in an inward-facing conformation with a bound Mg²⁺ ion coordinated by a conserved π-helix. Major structural rearrangements in the cytosolic domains are observed upon Mg²⁺-ATP binding, suggesting a mechanism of Mg²⁺-ATP sensing. We employ a liposome-based transport assay to demonstrate the direct Mg²⁺ transport function of CorB proteins. Together, these structural and functional insights provide a framework for understanding the function of human CNNMs and associated diseases.

Results

Structure determination

To identify CNNM/CorB orthologs with suitable properties for structural studies, we performed small-scale screening of 20 prokaryotic CorB orthologs from diverse species (Supplementary Fig. 1). Each ortholog was expressed as a GFP-fusion protein and purified in six different detergents (DDM, LMNG, OGNG, LDAO, C12E9, and OG). Among them, two orthologs showed good expression and gel-filtration profiles: MtCorB from a thermophilic archaeon (Methanoculleus thermophilus) and TtCorB from a thermophilic, gram-negative bacterium (Tepidiphilus thermophilus). MtCorB and TtCorB share 26 and 25% sequence identities to the conserved core of human CNNM2, respectively (Supplementary Fig. 2). Attempts in crystallizing the full-length proteins yielded no suitable crystals. Thus, several deletion mutants were generated. MtCorB with a C-terminal CorC domain deletion (MtCorBΔC) purified in DDM yielded crystals in presence of Mg²⁺-ATP. However, upon structural elucidation, the unit cell only contained the soluble CBS-pair domain (Table 1).

Table 1 Statistics of data collection and refinement.

Full size table

After extensive construct and detergent optimizations, we ultimately obtained diffracting crystals of MtCorBΔC with an internal loop deletion (MtCorBΔC_Δloop) purified in UDM in complex with Mg²⁺-ATP (Supplementary Fig. 1d). The crystals were obtained by vapor diffusion in presence of 0.1 M sodium citrate pH 5.5, 0.1 M Li₂SO₄, 0.1 M NaCl, 20 mM MgCl₂, 34% PEG400 and 5 mM ATP. Initial phases were determined by molecular replacement using MtCNNM_CBS structure solved earlier, which represented 40% of the crystallized construct (residues 199–322). Two α-helices (residues 166–198) preceding the CBS-pair domain were built-in manually, then PHENIX AutoBuild³³ was used to place idealized helical fragments in the missing sections (residues 1–166). The resulting phases were of great quality that allowed tracing of most of the molecule, guided by strong densities of bulky residues (Supplementary Fig. 3a). The final structure, consisting of chain A (residues 5–322) and chain B (residues 4–154 and 160–322), is refined to 3.25 Å with an R-free of 0.27 (Table 1).

Architecture and domain organization

MtCorBΔC crystallizes as a homodimer with each protomer consisting of three distinct regions: a TMD, an acidic helical bundle (AHB), and a CBS-pair domain (Fig. 1c, d). The TMD and the AHB represent previously unknown structural regions that required de novo model building. Since we were able to visualize the linker connecting TMD and AHB, we were able to assign the two chains with confidence, with the right TMD connecting to the right CBS-pair domain. Our result shows a discrepancy with the hypothetical model generated by combining isolated TMD and CBS-pair domain structures of bacterial TpCorB, in which the right TMD was connected to the left CBS-pair domain²³.

Although MtCorB is present as a dimer, it does not show C2 symmetry (Supplementary Fig. 3b). The individual TMD, AHB, and CBS-pair domains form symmetric dimers, but their arrangement is not. Overlaying the TMD of two polypeptide chains reveals asymmetry in the positions of the cytosolic domains, primarily due to differences in the linker between the TMD and AHB (Supplementary Fig. 3c). Together with previous observations that cytosolic domains of CNNMs form symmetric dimers^{27,28,29,30,32,34}, these results suggest the interface between the transmembrane and cytosolic domains is dynamic, and the current asymmetric arrangement likely arises from crystal packing (Supplementary Fig. 3d).

TMD in an inward-facing conformation with a negatively charged cavity

The TMD is a dimer with each chain composed of three transmembrane helices (TM1–3), a pair of short helices exposed on the intracellular side, and a juxtamembrane (JM) helix that forms a belt-like structure (Fig. 2a). The homodimeric TMD exists in an inward-facing conformation with dimerization predominantly formed by hydrophobic contacts of TM2 and TM3 of each protomer. Electrostatic surface potential analysis reveals a large, negatively charged cavity that has a maximum diameter of approximately 10 Å and is lined with conserved polar residues (Fig. 2b, c and Supplementary Fig. 2). A well-defined electron density is observed in the cavity (Fig. 2d), and the density was modeled as an Mg²⁺ ion based on the TpCorB structure²³. The Mg²⁺ is coordinated by hydroxyl groups of Ser21, Ser25, and Ser71; carboxyl group of Glu111; and the main-chain carbonyl groups of Ser21 and Gly110. Residues contributing to ion interactions are strongly conserved across species, suggesting these are conserved features of the CNNM/CorB proteins (Supplementary Fig. 2).

**Fig. 2: TM domain in an inward-facing conformation with a negatively charged cavity.**

A second conserved feature is a π-helical turn in TM3, in which the helix is composed of an i + 5 instead of i + 4 configurations, resulting in a bulge in the helix (Fig. 2e and Supplementary Fig. 3a). The residues surrounding the helical turn are highly conserved with Glu111, Ile112, and Pro114 completely invariant from archaea to humans (Fig. 2f). Glu111 points toward the negative cavity and is involved in the coordination of the Mg²⁺ ion, whereas Pro114 acts as a helix-breaker that allows the shift in hydrogen bonding of the π-helical turn. The positively charged Lys115 after Pro114 is also highly conserved across species.

The archaeal MtCorB shares sequence identify of 37% with the bacterial TpCorB. The bacterial TMD also exists as a homodimer in an inward-facing conformation with similar TM rearrangements (Supplementary Fig. 4a). Although it was not pointed out in the bacterial study²³, the bacterial structure also contains the acidic cavity and the conserved π-helical turn (Supplementary Fig. 4b, c). The polar residues in the acidic cavity and Mg²⁺-binding site are highly conserved with all five Mg²⁺-interacting residues being conserved except for Ser71, which is substituted by asparagine in bacteria.

A re-entrant JM helix encircles TM helices

The re-entrant JM helix following TM3 wraps around TM1 and TM2 like a belt through several hydrophobic contacts (Fig. 3a, b). The curvature is imposed by a conserved Pro142 that makes a kink in the helix (Supplementary Fig. 2). The outer surface of the JM helix is amphipathic and rich in aromatic residues. Surrounding the JM helix, we observed several electron densities, which we modeled as UDM detergent molecules from the protein buffer. In total, we observed ten UDM molecules, which likely correspond to conserved lipid-binding sites important for assisting conformational changes during transport. Structural comparison to the bacterial JM helix shows that the bacterial one is longer and makes more extensive contacts with TM3 of the other protomer (Supplementary Fig. 4d). This discrepancy could be species-specific or caused by the absence of the following domains in the bacterial TMD structure²³.

Underneath the JM helix resides the helix-turn-helix motif (α1 and α2) that connects TM1 and TM2 (Fig. 3c). The solvent-exposed portions contain several basic residues, which may be involved in binding phospholipid headgroups. Supporting this, a strong density was observed in this region and assigned as a sulfate molecule due to its presence in the crystallization buffer.

Acidic helical bundle

An unexpected feature of the MtCorB structure is the existence of the four-helical AHB domain following the TMD (Fig. 3d). The linker between the JM helix and AHB differs in the two protomers. It is extended and partially disordered in one and coiled in the other. The AHB itself is a symmetric dimer composed of two helices from each protomer and highly negatively charged. Thirteen of the 31 residues are acidic with three stretches of three or more consecutive glutamic acids, which likely represent cation binding sites. The negative charge of the AHB is well conserved across evolution with the exception of yeast and roundworms (Fig. 3e).

Structural basis of Mg²⁺-ATP binding by the CBS-pair domain

The isolated MtCorB_CBS structure from the in situ cleavage of MtCorBΔC allowed us to visualize the detailed interactions of Mg²⁺-ATP binding at higher resolution (Fig. 4a and Table 1). As observed in the structure with the TMD, two molecules of Mg²⁺-ATP are bound to the central cavity of the CBS-pair dimer. The adenine bases are sandwiched between Phe233 and Ile311 in a hydrophobic pocket comprising Met298, Phe237, Ile236, Val212, and Val213. The ATP ribose forms hydrogen bonds with the side chains of Thr207, Asp316, and Thr313. The phosphate groups are stabilized with Arg235 and Ser234 with the bound Mg²⁺ coordinated in an octahedral arrangement with three phosphates and three water molecules.

**Fig. 4: Mg²⁺-ATP binding to CBS-pair domain.**

Measurements of the affinity of adenosine nucleotides by isothermal titration calorimetry (ITC) showed MtCorBΔC has the highest affinity for ATP, followed by ADP and AMP (Fig. 4b and Supplementary Fig. 5). While the addition of Mg²⁺ has a relatively large effect on the binding of ATP to the isolated CBS-pair domain in human CNNMs^27,35, the effect was more muted with MtCorB. In presence of Mg²⁺, the affinity for ATP was increased three-fold with MtCorBΔC and ten-fold with the isolated MtCorB_CBS domain (Supplementary Fig. 5). These affinities are comparable to those measured from the isolated CBS-pair domains of TpCorB²³.

We used analytical ultracentrifugation (AUC) to characterize the effect of ATP binding on MtCorB (Fig. 4c and Supplementary Fig. 6). As observed for human CNNMs²⁷, dimerization of the CBS-pair domain was tightly coupled with Mg²⁺-ATP binding. While the MtCorB_CBS in the absence of ATP sedimented as a monomer, the addition of adenine nucleotides triggered dimerization. The addition of Mg²⁺ alone had no effect on the sedimentation of MtCorB_CBS.

Next, we turned to hydrogen–deuterium exchange mass spectrometry (HDX-MS) to detect conformational changes in full-length MtCorB upon ATP binding and deletion of the CorC domain. MtCorB and MtCorBΔC were first reconstituted into MSP1D1 nanodiscs (Supplementary Fig. 7a), and hydrogen exchange was monitored in the presence and absence of Mg²⁺-ATP. While peptide coverage of the hydrophobic TMD was limited, coverage of the cytosolic domains was excellent and revealed large conformational changes (Fig. 4d and Supplementary Fig. 7b, c). The addition of Mg²⁺-ATP significantly reduced the exchange in the CBS-pair domain dimerization interface and ATP-binding site. Similar changes were observed with both the full-length and truncated protein, which is consistent with the absence of a role of the C-terminal domain in Mg²⁺-ATP binding. This is different from human CNNMs where the C-terminal CNBH domain promotes CBS-pair domain dimerization and Mg²⁺-ATP binding²⁷. Comparison of full-length and truncated MtCorB showed changes in the CBS2 motif, suggesting this is the primary site of interaction of the CorC domain.

Major structural rearrangements in MtCorB upon loss of Mg²⁺-ATP binding

Further insight into the conformation in the absence of bound ATP came from the crystal structure of a MtCorBΔC R235L mutant, homologous to a Jalili syndrome mutant (R407L) that is unable to bind ATP in human CNNM4²⁷. Crystals of MtCorBΔC R235L mutant purified in UDM were obtained by vapor diffusion in presence of 0.1 M sodium citrate pH 5.5, 50 mM Li₂SO₄, and 11% PEG4000, but they diffracted poorly to 6 Å. Subsequent additives screening improved the resolution to 3.8 Å with the addition of 10 mM MgSO₄ (Table 1). The structure was determined by molecular replacement using individual TM and CBS-pair domains, while the AHB was rebuilt manually.

In the apo structure, the TMD remains in the same inward-facing conformation, but the cytosolic domains undergo major structural rearrangements (Fig. 5a). The CBS-pair domain has completely dissociated from the disc-like dimeric configuration to a more elongated, open conformation with different domain-domain contacts. To our knowledge, this dimeric configuration is not observed in any available CBS-pair domain structures²⁶. In agreement with the AUC results, the loss of ATP binding disrupted the dimerization of the CBS-pair domain, which then makes contact with the helix-turn-helix motif from TMD (Fig. 5b). In addition, the dimeric AHB dissociates and binds to the CBS2 motif of the neighboring molecule, competing for the Mg²⁺-ATP binding site (Fig. 5c).

**Fig. 5: Apo conformation of MtCorBΔC captured by R235L mutant.**

The morphing of the two conformations allowed us to visualize the conversion of the apo to the Mg²⁺-ATP bound form (Supplementary Movie 1). The two conformations are interchangeable by a 90° rotation of both protomers (Fig. 5d). Despite the large dimeric rearrangement, the individual protomer in the two states is similar with differences limited to the AHB and α8 helix (Fig. 5e). Our result shows a discrepancy to the isolated CBS-pair domain structures of the bacterial TpCorB, in which minimal structural changes were observed upon nucleotide binding²³. Perhaps, this was due to the absence of TMD and one of the AHB helices in the bacterial structures.

Liposome-based Mg²⁺ transport assays

In order to address the most debated question in the field: whether CNNM/CorB proteins mediate direct or indirect Mg²⁺ transport, we carried out an in vitro, liposome-based Mg²⁺ transport assay using a modified version of assays used for CorA proteins^36,37. We reconstituted CorB proteins into liposomes and encapsulated mag-fura-2, a ratiometric Mg²⁺indicator (Fig. 6a). Upon addition of Mg²⁺, we observed a time-dependent change in the ratio of fluorescence from excitation at two wavelengths (330/369 nm), which could be used to calculate the Mg²⁺ concentration inside the liposomes. CorB liposomes showed a time-dependent increase in the fluorescence ratio while no fluorescence change was observed for liposomes devoid of protein. Assays with the bacterial CorB (TtCorB) showed more transport activity than the archaeal protein (MtCorB), possibly due to better compatibility with the lipid composition of the proteoliposomes. Thus, we used the bacterial protein for further functional studies. By varying the protein/lipid ratio and the concentration of Mg²⁺, we showed that the TtCorB-mediated Mg²⁺ transport is protein- and magnesium-dependent (Supplementary Fig. 8a, b).

**Fig. 6: Liposome-based assay shows direct Mg²⁺ transport by CorB proteins.**

Next, to test whether the TtCorB-mediated Mg²⁺ transport is electrogenic, we used valinomycin, a potassium-selective ionophore, to generate an inward negative membrane potential (Fig. 6b). We used TmCorA, a well-studied Mg²⁺ channel, as a positive control. As shown previously, TmCorA showed increased Mg²⁺ accumulation in presence of valinomycin^36,37. In contrast, we did not observe an increase in Mg²⁺ transport for TtCorB. These results suggest TtCorB mediates Mg²⁺ transport through an electroneutral antiporter mechanism. We also examined the ion selectivity of TtCorB toward different divalent cations. Due to the differences in the affinity and response of mag-fura-2 to different ions, it is difficult to directly compare ions, but we could compare the ion selectivity of TtCorB with TmCorA. In our experimental conditions, TtCorB was more selective and showed less transport of Ca²⁺ and Zn²⁺ than TmCorA (Supplementary Fig. 8e). TtCorB and TmCorA showed similar transport activity with Mg²⁺.

Lastly, to probe the importance of the conserved residues lining the negatively charged cavity in TMD (Fig. 6c), we tested the effect of several cavity mutations on the Mg²⁺ transport activity. During purification, some mutants, especially ones involved in Mg²⁺ coordination, have abnormal gel-filtration profiles compared to wild-type proteins as shown by the earlier elution volumes (Supplementary Fig. 8c), suggesting the importance of these residues in maintaining the structural integrity of the protein. Despite that, all mutants showed high purity and incorporated well into liposomes (Supplementary Fig. 8d). Mg²⁺ transport assays were performed on all the mutants (Fig. 6d, e). Some mutants (E33A and N80A) have near-wildtype transport activity, while others, such as the Mg²⁺-binding site mutants (S26A, S30A, and N76A), have increased transport activity. On the other hand, mutations of the four conserved residues (S117A, E118A, P121G, and K122A) near the π-helical turn consistently reduced the Mg²⁺ transport activity. These results demonstrate the functional importance of the conserved residues in the Mg²⁺-binding site and π-helical turn.

Huang et al.²³ had also tested the effect of several cavity mutations with an indirect, cell-based Mg²⁺ efflux assay through the expression of chimeric CNNM4/TpCorB proteins in HEK293 cells. In comparison, we have similar results for S117A and E118A equivalent mutants, which showed reduced Mg²⁺ efflux activity. In contrast, divergent results were obtained for S30A and E80A equivalent mutants. Both mutants had reduced efflux activity in the bacterial TpCorB, while our S30A mutant increased activity, and the N80A mutant has near-wildtype activity. Huang et al. did not measure transport activity with mutations equivalent to S26A, E33A, N76A, P121G, and K122A.

Homology modeling of human CNNM

We used homology modeling to improve our understanding of mutations in human CNNM proteins (HsCNNMs) that cause diseases. Previous attempts to predict the structure of HsCNNM2 were confounded by the hydrophobicity of the HsCNNM2 JM helix³⁸ (Supplementary Fig. 9a). Using the Mg²⁺-ATP bound structure of MtCorBΔC, we generated a homology model of HsCNNM2, which showed good geometry and allowed TM helices to be reliably positioned (Fig. 7a and Supplementary Fig. 9). The main differences lie in the length of linkers connecting the TM helices. The HsCNNM2 model reproduced the negatively charged cavity, π-helix in TM3, the Mg²⁺ ion binding site, and the presence of buried polar residues (e.g., Thr331, Asp335, and Ser348), supporting the importance of these features for CNNM function (Fig. 7b, c).

We used the model to understand the molecular basis for human disease mutations (Fig. 7d). Nine missense mutations in HsCNNM2 are responsible for hypomagnesemia^3,5,39,40. T568I in the CBS-pair domain prevents ATP binding and Mg²⁺ transport^27,35. L418P in the dimerization interface of AHB likely disrupts the helical bundle. In the TMD, S269W, and L330F both likely disrupt TM helix packing, and E357K is a charge reversal mutation of the invariant glutamate that coordinates the Mg²⁺. D335N near the extracellular surface may be important in the outward-facing conformation. The high degree of sequence conservation (84% identity) between HsCNNM2 and HsCNNM4 also allowed us to use the model to interpret HsCNNM4 mutations that cause Jalili syndrome^{41,42,43,44,45,46}. Of particular interest are mutations that occur in the TMD. S196P and S200Y both map to the negatively charged cavity. HsCNNM4 S196P affects the same residue as CNNM2 S269W, confirming the importance of this position for function. R236Q, S245L, and L324P are located at the inner surface of the membrane, in TM3, and the JM helix, respectively.

Molecular dynamics simulations

We used molecular dynamics simulations to study asymmetry and conformational changes in the MtCorBΔC structure. Unbiased MD simulations of the MtCorBΔC structure were first carried out (Supplementary Movie 2). As expected from the asymmetry in the structure, the interface between the TMD and cytosolic domains was flexible and large movements were observed. At the end of the simulation, the conformations of the two protomers were essentially switched while the individual domains all retained their symmetry. In addition, we also performed unbiased MD simulations of symmetric MtCorBΔC dimer (Supplementary Movie 3). In this case, the cytosolic domains appear more rigid than that of asymmetric dimer. We also observed an upward movement of the AHB towards the TMD, suggesting entrapment and stabilization by the helix-turn-helix motif of TMD.

The current structure resembles a transporter fold in the inward-facing conformation as supported by the hydration analysis of the structure (Supplementary Fig. 10a). In order to simulate the outward-facing conformation, we pulled an Mg²⁺ ion through the TMD dimerization interface while keeping a distance constraint with Glu111 of both protomers. This in turn causes the two helices to flip open, generating a possible outward-facing conformation (Fig. 7e). We then performed targeted MD to gain insight into the conformational changes from the inward to the outward-facing conformation (Supplementary Movie 4 and Supplementary Fig. 10b, c). The conserved residues Glu10, Ser80, and Thr83, which are buried in the membrane in the inward-facing conformation, become solvent-exposed and available for ion binding. The conservation of the π-helix and buried polar residues suggest that the mechanism of ion transport is conserved across all CNNM/CorB family members.

Discussion

Mg²⁺ is the most abundant divalent cation inside cells and essential for a wide variety of biochemical and enzymatic reactions¹. Despite this, our structural knowledge of Mg²⁺ ion transporter is limited to three proteins: MgtE⁴⁷, CorA^48,49,50, and TRPM7⁵¹. The structure of CorB proteins defines a different class of Mg²⁺ transporter, highly conserved and present in all organisms.

CNNM/CorB proteins had previously been suggested to be too small in ion transport due to the small number of predicted TM helices¹⁸. CorA is a pentamer with a total of 10 TM helices, MgtE is a dimer with 10, and TRPM7 has 24. Our structure shows that 6 TM helices suffice to form a large cavity (~1500 Å³) with ion binding sites deep within the membrane (Fig. 2). In fact, the current structure resembles a transporter fold in an inward-facing conformation with solvent accessible cavity toward the cytosol. This suggests that CNNM/CorB proteins might transport Mg²⁺ through a rocker-switch mechanism, in which the protein may exist in other conformational states, such as the occluded or outward-facing conformations. We have attempted to simulate the outward-facing conformation (Fig. 7e), but further studies are needed to validate and capture CNNM/CorB proteins in other conformational states.

The biggest controversy in the field surrounds the question: whether CNNMs/CorBs are Mg²⁺ transporters themselves or Mg²⁺ sensors that regulate other transporters? The dispute arises because all transport assays performed so far were done through overexpression of CNNMs/CorBs in cells^{3,4,5,15,21,22,23}, which preclude the differentiation of direct or indirect Mg²⁺ transport. With our in vitro, liposome-based assay, we demonstrated direct Mg²⁺ transport of CorB proteins (Fig. 6). In addition, we showed that the transport likely occurs through an electroneutral antiporter mechanism, similar to that proposed in CNNMs⁴. Together with the structural insights, our results clearly demonstrate CNNM/CorB proteins are magnesium transporters themselves.

Throughout evolution, the CBS-pair domain of CNNM/CorB proteins remains associated with the TMD. The ATP-binding site is a disease mutational hotspot (Fig. 7d), and disruption of ATP-binding results in loss of Mg²⁺ efflux activity^23,27,35. While the CBS-pair domain has no ATPase activity^23,27, ATP-binding likely plays a regulatory role as was observed for the bacterial Mg²⁺ channel MgtE⁵². Using a disease-mimicking mutant, we were fortunate to capture the apo state of the transporter (Fig. 5). Structurally, the loss of Mg²⁺-ATP binding disrupts the dimerization of both CBS-pair and AHB domains. Dramatic changes in the dimeric rearrangement of CBS-pair domains upon nucleotide-binding have been well-documented in other CBS domain-containing proteins²⁶ as well as human CNNMs^27,28. However, the current dimeric configuration involving AHB contact has not been observed before and seems CNNM/CorB specific. The acidic residues in the AHB are reminiscent of the acidic patches in the cytosolic domains of CorA and MgtE that are involved in Mg²⁺ sensing and regulation of channel activity^53,54. Disruption of the AHB dimer could play a regulatory role through the loss of Mg²⁺ binding sites. In addition, the CBS-TMD contact could result in an inactivation mechanism (Fig. 7f). For example, upon loss of ATP binding, the generation of the CBS-TMD contact would lock the TMD in the inward-facing conformation. This in turn inactivates the transporter as it is unable to change to other conformational states required for Mg²⁺ transport. Upon Mg²⁺-ATP binding, dimerization of the CBS-pair domain results in loss of CBS-TMD contact, and this, in turn, alleviates the inactivation of TMD, thereby allowing Mg²⁺ transport to occur.

The CNNM/CorB family of proteins is highly homologous in the conserved core from humans down to bacteria. Despite their common origin, eukaryotic CNNMs have evolved to be different. In animals, PRL phosphatases provide an additional level of regulation through binding the CNNM CBS-pair domain. PRLs are highly oncogenic and promote cancer metastasis through their suppression of CNNM Mg²⁺ efflux^9,10. PRL binding itself is regulated by PRL phosphorylation in response to Mg²⁺ levels^30,31. Although further studies are needed to fully understand the function of human CNNMs and their regulation by PRLs, the structural and functional insights from this study provide a framework for understanding CNNM function and the development of therapeutic approaches for regulating their activity.

Methods

Construction of phylogenetic tree

Amino acid sequences of various CNNM/CorB orthologs were aligned using MUSCLE⁵⁵. The phylogenetic tree was generated using the neighbor-joining method and bootstrapping of 1000 replications in MEGAX (Version 10.1.8)⁵⁶. The CNNM/CorB orthologs and their UniProt accession numbers are: CNNM2 (Homo sapiens; Q9H8M5), cnnm2a (Danio rerio; A2ATX7), CNNM4 (H. sapiens; Q6P4Q7), CNNM4 (Xenopus tropicalis; A0JPA0), CNNM1 (H. sapiens; Q9NRU3), CNNM3 (H. sapiens; Q8NE01), UEX (Drosophila melanogaster; A0A0B7P9G0), cnnm-1 (Caenorhabditis elegans; A3QM97), MAM3 (Saccharomyces cerevisiae; Q12296), CBSDUF1 (Arabidopsis thaliana; Q67XQ0), CBSDUFCH2 (Arabidopsis thaliana; Q84R21), MtCorB (M. thermophilus; A0A1G8XA46), StCorB (Salmonella typhimurium; Q8XFY3), TtCorB (T. thermophilus; A0A0K6IWT9), MpfA (S. aureus; A0A0H3JL60), and yhdP (B. subtilis; O07585).

Cloning of CorB orthologs

Codon-optimized cDNA of 20 CorB orthologs were synthesized (Bio Basic Inc., Markham, Canada) and sub-cloned into NcoI and XhoI sites of pCGFP-BC vector⁵⁷ with a C-terminal GFP-His8-tag for small-scale expression and detergent screening. Promising orthologs were subcloned into NdeI and XhoI sites of pET29a vector (Millipore Sigma) with a C-terminal His6-tag for large-scale expression. Constructs for M. thermophilus CorB (UniProt entry A0A1G8XA46): MtCorB (residues 1–426), MtCorBΔC (residues 1–322), MtCorBΔC_Δloop (residues 1–322 Δ259–262), and MtCorBΔC R235L. Construct for T. thermophilus CorB (UniProt entry A0A0K6IWT9): TtCorB (residues 1–445). For MtCorB constructs lacking TMD, MtCorB_CBS+CorC (residues 199–417), MtCorB_CBS (residues 199–322), and MtCorB_CorC (residues 333–417) were subcloned into BamHI and XhoI sites of pGEX-6P-1 vector (GE Healthcare) with an N-terminal GST-tag. QuikChange Lightning Site-Directed Mutagenesis Kit (Agilent) was used for the generation of point mutations. A list of mutagenic primers is included in Supplementary Table 1.

Small-scale expression and detergent screening of CorB orthologs

CorB orthologs cloned in pCGFP-BC were transformed into Escherichia coli strain C41 (DE3). Cells were grown in Luria Broth (LB) at 37 °C to an optical density of 0.6 and induced with 1 mM IPTG for 4 h at 30 °C. The cell pellet was obtained by centrifuging at 5000g for 10 min. The pellet was re-suspended in lysis buffer (50 mM HEPES, 500 mM NaCl, 5% glycerol, pH 7.5) supplemented with cOmplete^TM protease inhibitor cocktail (Roche) and split into 6 fractions. Lysis was performed using a 24-probe sonicator. Each fraction was solubilized with a different detergent (DDM, n-dodecyl-β-D-maltopyranoside; LMNG, lauryl maltose neopentyl glycol; OGNG, octyl glucose neopentyl glycol; LDAO, lauryldimethylamine-N-oxide; C12E9, dodecyl nonaethylene glycol ether; OG, n-octyl-β-d-glucopyranoside) to a final concentration of 1%, then purified by IMAC in the same detergent (3× CMC). Elutions were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and size-exclusion chromatography on SEPAX Zenic-C SEC-300 connected to fluorescence detector (Ex 480 nm/ Em 510 nm) in buffer (50 mM HEPES pH 7.5, 150 mM NaCl, 0.5 mM TCEP, 3× CMC of detergent).

Expression and purification of MtCorB, TtCorB, and TmCorA

MtCorB and TtCorB constructs were transformed into E. coli strain C41 (DE3), while TmCorA⁵⁰ was transformed into E. coli strain BL21 (DE3). MtCorB was grown in LB at 37 °C to an optical density of 0.6 and induced with 0.5 mM IPTG overnight at 18 °C. TtCorB and TmCorA were grown in LB at 37 °C to an optical density of 0.8 and induced with 0.5 mM IPTG for 4 h at 30 °C. The cell pellet was obtained by centrifuging at 5000g for 20 min. The pellet was re-suspended in lysis buffer (50 mM HEPES, 500 mM NaCl, 5% glycerol, pH 7.5) supplemented with cOmplete^™ protease inhibitor cocktail (Roche), 10 µg/mL DNAse I, 1 mM CaCl₂, and 0.1 mg/mL lysozyme. MtCorB cells were lysed by passing through Avestin Emulsiflex-C3 homogenizer (10,000–15,000 p.s.i.), while TtCorB and TmCorA cells were lysed by sonication. Cellular debris was removed by centrifugation at 27,000g for 10 min at 4 °C (this step is omitted for full-length MtCorB). Membranes were pelleted by ultracentrifugation at 150,000g for 1 h at 4 °C. The membrane fraction was collected, flash-frozen in liquid nitrogen, and stored at −80 °C for later use. The membrane fraction after thawing was solubilized in lysis buffer supplemented with 1% DDM (MtCorB & TmCorA) or UDM (TtCorB) for 1 h at 4 °C on a rotator, then ultracentrifuged at 150,000g for 30 min at 4 °C. The supernatant was loaded onto Qiagen Ni-NTA resin by batch binding, and incubated with gentle shaking for 1 h at 4 °C. The resin was then washed with wash buffer (50 mM HEPES, 500 mM NaCl, 5% glycerol, 30 mM imidazole, pH 7.5) containing 0.03% DDM or 0.05% UDM and eluted with elution buffer (50 mM HEPES, 500 mM NaCl, 5% glycerol, 300 mM imidazole, pH 7.5) containing 0.03% DDM or 0.05% UDM. The affinity-purified protein was further purified by size exclusion chromatography on a HiLoad 16/600 Superdex 200 pg column (GE Healthcare) in HPLC buffer (20 mM HEPES, 150 mM NaCl, pH 7.5) containing 0.03% DDM or 0.05% UDM. For TtCorB and TmCorA, the purification or HPLC buffers additionally contain 5 mM BME or 3 mM TCEP. The final purified proteins were concentrated using 50 kDa (MtCorBΔC) or 100 kDa (MtCorB, TtCorB, and TmCorA) cutoff concentrators (Amicon Ultra, Millipore), flash-frozen in liquid nitrogen, and stored at −80 °C for later use. The protein concentration is determined spectrophotometrically using Nanodrop, and purity is verified by SDS-PAGE.

Expression and purification of MtCorB constructs lacking TMD

Constructs were transformed into E. coli strain BL21 (DE3). Cells were grown in LB at 37 °C to an optical density of 0.8 and induced with 1 mM IPTG overnight at 20 °C. The cell pellet was obtained by centrifuging at 5000g for 20 min. The pellet was re-suspended in lysis buffer (50 mM HEPES, 500 mM NaCl, 5% glycerol, pH 7.5) supplemented with cOmplete^™ protease inhibitor cocktail (Roche), 10 µg/mL DNAse I, 1 mM CaCl₂, and 0.1 mg/mL lysozyme. Cells were lysed by sonication. Cellular debris was removed by centrifugation at 44,000g for 45 min at 4 °C. The supernatant was loaded onto Glutathione Sepharose resin (GE Healthcare), washed with lysis buffer, and eluted with lysis buffer containing 20 mM glutathione. The GST-tag was removed by overnight incubation with PreScission Protease, leaving an N-terminal Gly-Pro-Leu-Gly-Ser extension. The protein was further purified by size exclusion chromatography on a HiLoad 16/600 Superdex 75 pg column (GE Healthcare) in HPLC buffer (20 mM HEPES, 100 mM NaCl, pH 7.5). The protein was diluted to 5 µM, dialyzed overnight in dialysis buffer (20 mM HEPES, 100 mM NaCl, 5 mM EDTA, pH 7.5), and re-injected onto Superdex-75 to remove bound nucleotides. The final purified protein was concentrated to around 10 mg/mL (measured by NanoDrop), and the purity verified by SDS-PAGE.

Crystallization

Crystals of MtCorB_CBS with Mg²⁺-ATP were obtained by equilibrating 0.4 µL of protein (20.8 mg/mL MtCorBΔC purified in 0.03% DDM; 5 mM ATP added prior to crystallization) and 0.4 µL of reservoir solution (0.1 M MOPS, pH 7.0; 9% PEG 8000; 20 mM MgCl₂) in sitting-drop vapor diffusion system incubated at 22 °C. Rod-like crystals appeared after 2 weeks. The crystal was cryo-protected with reservoir solution supplemented with 5 mM ATP and 30% ethylene glycol, picked up in a nylon loop, and flash-cooled in an N₂ cold stream.

Crystals of MtCorBΔC_Δloop with Mg²⁺-ATP were obtained by equilibrating 1 µL of protein (19.5 mg/mL MtCorBΔC_Δloop purified in 0.05% UDM; 5 mM ATP added prior to crystallization) and 1 µL of reservoir solution (0.1 M sodium citrate, pH 5.5; 0.1 M Li₂SO₄; 0.1 M NaCl; 20 mM MgCl₂; 34% PEG400; 10 mM Na₂HPO₄) in hanging-drop vapor diffusion system incubated at 22 °C. Petal-like crystals appear after 1 week. The crystal was directly picked up in a nylon loop and flash-cooled in an N₂ cold stream.

Crystals of MtCorBΔC R235L were obtained by equilibrating 1 µL of protein (18.4 mg/mL MtCorBΔC R235L purified in HPLC buffer containing 20 mM HEPES, 500 mM NaCl, 0.05% UDM, pH 7.5) and 1 µL of reservoir solution (0.1 M sodium citrate, pH 5.5; 50 mM Li₂SO₄; 11% PEG4000; 10 mM MgSO₄) in hanging-drop vapor diffusion system incubated at 22 °C. Rod-like crystals appear after 1 day. The crystal was cryo-protected with reservoir solution supplemented with 25% glycerol and 0.6 M NaCl, picked up in a nylon loop, and flash-cooled in an N₂ cold stream.

Data collection and structure determination

The MtCorB_CBS dataset from a single crystal was collected using a Pilatus3 6 M detector at beamline 5.0.2 of Advanced Light Source (ALS). Data processing and scaling were performed with HKL-2000⁵⁸ with auto-corrections enabled. Initial phases were obtained by molecular replacement with Phaser⁵⁹ in PHENIX⁶⁰ using the CBS-pair domain structure of EcCorC (PDB: 5YZ2)⁶¹.

The MtCorBΔC_Δloop dataset from a single crystal was collected using a Pilatus3 6 M detector at beamline 08ID-1 of the Canadian Macromolecular Crystallography Facility of the Canadian Light Source (CLS). The dataset showed anisotropic diffraction up to 3.25 Å. Data processing and scaling were performed with HKL-2000⁵⁸ with auto-corrections enabled, in which ellipsoid truncation was performed automatically. Resolution cut-off is based on CC_1/2 = 0.3⁶². Resolution limits after ellipsoid truncation were a* = 4.07 Å, b* = 3.71 Å, and c* = 3.25 Å. Initial phases for the CBS-pair domain were obtained by molecular replacement with Phaser⁵⁹ in PHENIX⁶⁰ using MtCorB_CBS structure (determined in this study). Then AutoBuild³³ was used to build in the missing domains (TMD and AHB).

The MtCorBΔC R235L dataset from a single crystal was collected using Pilatus3 6 M detector at beamline 08B1-1 of the CLS. Data processing and scaling were performed with HKL-2000⁵⁸ with auto-corrections enabled. Initial phases were obtained by molecular replacement with Phaser⁵⁹ in PHENIX⁶⁰ using the CBS-pair and TMD structures (determined in this study).

After molecular replacement, the models were subsequently improved through iterative cycles of manual building with Coot⁶³ and refinement with phenix.refine⁶⁴. TLS parameters were included at later stages of the refinement⁶⁵. The final structures were validated with MolProbity⁶⁶. Crystallographic data collection and structure refinement statistics are shown in Table 1. Structural images were prepared with PyMOL, Version 2.3.4 (Schrödinger LLC, New York). Electrostatic surface potentials were calculated using the APBS plugin within PyMOL⁶⁷.

Isothermal titration calorimetry

ITC experiments were performed on a MicroCal VP-ITC titration calorimeter (Malvern Instruments Ltd.) at 20 °C with stirring at 310 rpm. TMD-containing constructs (15 µM final concentration) and ligands were prepared in HPLC buffer containing 0.05% UDM (±50 mM MgCl₂). Cytosolic constructs (30 µM final concentration) were prepared in HPLC buffer (±10 or 50 mM MgCl₂). The ligands were injected 19 times (5 µL for the first injection, 15 µL for subsequent injections), with 4 min intervals between injections. Results were analyzed using ORIGIN software (MicroCal) and fitted to a binding model with a single set of identical sites.

Analytical ultracentrifugation

Sedimentation velocity AUC experiments were performed at 20 °C using a Beckman Coulter XL-I Optima analytical ultracentrifuge and an An-60Ti rotor at 98,000g (35,000 rpm) for 18 h with scans performed every 60 s. A double-sector cell, equipped with a 12 mm Epon centerpiece and sapphire windows, was loaded with 380 and 400 µL of sample and HPLC buffer. MtCorB_CBS (100 µM) with various ligands were monitored using interference optics. The data were analyzed with Sedfit v1501b⁶⁸ using a continuous c(s) distribution. Numerical values for the solvent density, viscosity, and partial specific volume were determined using Sednterp⁶⁹. Buffer density and viscosity were calculated to be 1.0039 g/cm³ and 0.01026 mPa·s, respectively (20 mM HEPES, 100 mM NaCl, pH 7.5). Partial specific volume for MtCorB_CBS was calculated to be 0.7464 cm³/g. The frictional ratio (f/f₀) value for MtCorB_CBS was calculated using US-SOMO⁷⁰ to be 1.26. Residual and c(s) distribution graphs were plotted using GUSSI⁷¹.

Production and purification of MSP1D1

pMSP1D1 was a gift from S. Sligar (Addgene plasmid 20061). MSP1D1 production was carried out according to published protocols⁷². In brief, E. coli BL21 (DE3) cells transformed with pMSP1D1 were grown in LB at 37 °C to an optical density of 0.8 and induced with 1 mM IPTG for 4 h at 30 °C. MSP1D1 was purified by nickel affinity chromatography according to standard conditions described in⁷². The polyhistidine tag was removed by overnight incubation with TEV protease and further purified by Superdex-75 size-exclusion column (GE Healthcare) in HPLC buffer (20 mM HEPES, 150 mM NaCl, pH 7.5).

Reconstitution into nanodisc

MtCorB and MSP1D1 were mixed with soybean polar extract (Avanti) solubilized in 40 mM DDM at a MtCorB:MSP1D1:lipid molar ratio of 2:10:550 in HPLC buffer. Detergent was removed by incubation with Bio-Beads (Bio-Rad SM-2 Resin) at 4 °C overnight with constant rotation. Bio-beads were removed via filtration and the reconstitution mixture was re-loaded onto Qiagen Ni-NTA resin to remove empty nanodiscs. The resin was washed with wash buffer (20 mM HEPES, 200 mM NaCl, 20 mM imidazole, pH 7.5) and eluted with wash buffer with 300 mM imidazole. The eluted protein was further purified by size exclusion chromatography on a Superdex 200 Increase 10/300 GL column (GE Healthcare) in HPLC buffer (20 mM HEPES, 150 mM NaCl, pH 7.5).

Hydrogen deuterium exchange mass spectrometry

HDX-MS reactions were performed in a similar manner as described previously^73,74. In brief, HDX reactions for MtCorB and MtCorBΔC were conducted in a final reaction volume of 10 µL with a molar quantity of 20 pmol of MtCorB and MtCorBΔC. The reaction was started by the addition of 9.0 µL of D₂O buffer (100 mM NaCl, 20 mM HEPES pH 7.5, 94% D₂O (V/V)) to 1.0 µL of protein solution (final D₂O concentration of 84.9%). The reaction proceeded for 3, 30, 300, or 3000 s at 20 °C, before being quenched with ice-cold acidic quench buffer, resulting in a final concentration of 0.6 M guanidine–HCl and 0.9% formic acid post quench. All conditions and timepoints were created and run in triplicate. Samples were flash-frozen immediately after quenching and stored at −80 °C until injected onto the ultra-performance liquid chromatography (UPLC) system for proteolytic cleavage, peptide separation, and injection onto a QTOF for mass analysis, described below.

Protein samples were rapidly thawed and injected onto a UPLC system kept in a Peltier has driven cold box at 2 °C (LEAP). The protein was run over two immobilized pepsin columns (Trajan; ProDx Protease column PDX.PP01-F32) and the peptides were collected onto a VanGuard Precolumn trap (Waters). The trap was eluted in line with an ACQUITY 1.7 μm particle, 100 × 1 mm² C18 UPLC column (Waters), using a gradient of 5–36% B (Buffer A 0.1% formic acid, Buffer B 100% acetonitrile) over 16 min. MS experiments were performed on an Impact HD QTOF (Bruker) and peptide identification was done by running tandem MS (MS/MS) experiments run in data-dependent acquisition mode. The resulting MS/MS datasets were analyzed using PEAKS7 (PEAKS) and a false discovery rate was set at 1% using a database of purified proteins and known contaminants. HDExaminer Software (Sierra Analytics) was used to automatically calculate the level of deuterium incorporation into each peptide. All peptides were manually inspected for the correct charge state and presence of overlapping peptides. Deuteration levels were calculated using the centroid of the experimental isotope clusters. No correction was made for back exchange during analysis, so all deuterium incorporation values are relative. Differences in exchange in a peptide were considered significant if they met three of the following criteria: >5% change in exchange, >0.4 Da mass difference in exchange, a p value < 0.01 using a two-tailed Student’s t test, and change spanned by multiple peptides. The source data are provided as a Source Data file.

Proteoliposome reconstitution and Mg²⁺ transport assay

Proteoliposome were made following⁷⁵ with modifications. A 3:1 mixture of 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoethanolamine (POPE) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (POPG) (Avanti Polar Lipids) were dried into a thin film, followed by overnight incubation in a vacuum chamber. Dried lipids were solubilized to 5 mg/mL in buffer A (20 mM HEPES-KOH, 100 mM KCl, 50 mM NaCl, 0.5 mM EDTA-KOH, pH 7.5) supplemented with 35 mM CHAPS and incubated at room temperature for 2 h. UDM-purified MtCorB and TtCorB or DDM-purified TmCorA were mixed with 100 µL of solubilized lipids (0.5 mg) at a protein/lipid ratio of 1:30, 1:90, 1:270, or 1:810 (wt:wt) and incubated for 20 min at room temperature. For the no-protein liposome control, the same volume of HPLC buffer containing 0.05% UDM was added. Proteoliposomes were formed by adding the protein/lipid sample to a partially dehydrated Sephadex G-50 column (1.5 mL) equilibrated in buffer A. Membrane impermeable magnesium indicator, mag-fura-2 (Thermo Fisher; 1 mM stock in H₂O) was added to a final concentration of 50 µM to the proteoliposomes. Dye encapsulation was performed through 1 freeze–thaw cycle (we observed reduced activity with an increased number of cycles) and subsequent extrusion through 0.8 µm polycarbonate filters. Extravesicular dye was removed by centrifuging the proteoliposomes through a partially dehydrated Sephadex G-50 column equilibrated in buffer A.

The transport assay was measured with a SpectraMax M5e fluorescence plate reader using SoftMax Pro 5 software. Totally, 10 µL of dye-encapsulated liposomes were diluted with 190 µL of reaction buffer (buffer A without EDTA) in a 96-well black bottom plate (Greiner) and baseline fluorescence was measured for 1 min at 25 °C. The uptake reaction was initiated by the addition of MgCl₂, CaCl₂, or Zn(CH₃CO₂)₂, and recorded for 10 min. For mag-fura-2-encapsulated liposomes, fluorescence was measured using two excitation wavelengths (330 and 369 nm) and one emission wavelength (509 nm) every 6 s including 1 s shaking between reads. Triton X-100 and EDTA were added to 0.1% and 12.5 mM to obtain the maximum and minimum mag-fura-2 signals for conversion to [Mg²⁺]. All measurements were performed in triplicates. To test for electrogenic transport, the NaCl and KCl in the reaction buffer were replaced with 150 mM NMDG-Cl and 1 µM valinomycin (Sigma; 1 mM stock in DMSO) to generate an outward potassium gradient. For no valinomycin condition, an equal volume of DMSO was added.

Homology modeling of TtCorB and HsCNNM2

The homology model of TtCorB and HsCNNM2 were constructed with MODELLER Version 9.25⁷⁶ using the MtCorBΔC structure as the template. The sequence alignment was performed using Clustal Omega⁷⁷. The model was evaluated using PROCHECK⁷⁸. Conservation coloring was done with ProtSkin (http://www.mcgnmr.mcgill.ca/ProtSkin/).

Molecular dynamics simulation

MtCorBΔC structure was used as the starting model for MD simulations. The transporter was embedded in a bilayer of 3POPC:1POPG lipids and solvated in 150 mM NaCl (neutral with 277 Na⁺ and 139 Cl⁻ ions) using the web service CHARMM-GUI^79,80. Most residues were assigned their standard protonation state at pH 7. The total number of atoms in the atomic model is on the order of 200,000. The all-atom CHARMM force field PARAM36 for protein^81,82,83, lipids⁸⁴, and ions⁸⁵ was used. Explicit water was represented with the TIP3P model⁸⁶. The model was refined using energy minimization for at least 2000 steps. All the simulations were performed under NPT (constant number of particle N, pressure P, and temperature T) conditions at 303 K and 1 atm, and periodic boundary conditions with electrostatic interactions were treated by the particle mesh Ewald method⁸⁷ and a real-space cutoff of 12 Å. The simulations use a time step of 2 fs, with bond distances involving hydrogen atoms fixed using the SHAKE algorithm⁸⁸. After minimization, positional restraints on all of the Cα atoms were gradually released after which a trajectory of 500 ns was generated using NAMD version 2.11⁸⁹. The same process was applied to the model of symmetric MtCorBΔC dimer, in which the simulation ran for 124 ns.

Targeted molecular dynamics simulation

For preparing the target (open) model, an Mg²⁺ ion was pulled through the interface helices of MtCorBΔC while keeping a distance restraint with Glu111 of both protomers. This caused the two helices to open. All forces were then removed and the whole system was equilibrated for ~8 ns. For the targeted MD simulation, the MtCorBΔC structure was used as the starting model. During the simulation, forces were applied on the backbone of the transmembrane helices with a force constant of 200 kcal mol⁻¹ A⁻² for a period of 10 ns. This decreases the RMSD value linearly to reach the final target conformation.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Data availability

Data supporting the findings of this manuscript are available from the corresponding author upon reasonable request. A reporting summary for this article is available as a Supplementary Information file. Atomic coordinates and structure factors for MtCorB_CBS with Mg²⁺-ATP, MtCorBΔC with Mg²⁺-ATP, and MtCorBΔC R235L mutant have been deposited to the Protein Data Bank (www.rcsb.org) under accession numbers 7MSU, 7M1T, and 7M1U, respectively. Source data are provided with this paper.

References

de Baaij, J. H., Hoenderop, J. G. & Bindels, R. J. Magnesium in man: implications for health and disease. Physiol. Rev. 95, 1–46 (2015).
Article PubMed CAS Google Scholar
Funato, Y. & Miki, H. Molecular function and biological importance of CNNM family Mg2+ transporters. J. Biochem. 165, 219–225 (2019).
Article CAS PubMed Google Scholar
Stuiver, M. et al. CNNM2, encoding a basolateral protein required for renal Mg2+ handling, is mutated in dominant hypomagnesemia. Am. J. Hum. Genet. 88, 333–343 (2011).
Article CAS PubMed PubMed Central Google Scholar
Yamazaki, D. et al. Basolateral Mg2+ extrusion via CNNM4 mediates transcellular Mg2+ transport across epithelia: a mouse model. PLoS Genet. 9, e1003983 (2013).
Article PubMed PubMed Central CAS Google Scholar
Arjona, F. J. et al. CNNM2 mutations cause impaired brain development and seizures in patients with hypomagnesemia. PLoS Genet. 10, e1004267 (2014).
Article PubMed PubMed Central CAS Google Scholar
Daneshmandpour, Y., Darvish, H., Pashazadeh, F. & Emamalizadeh, B. Features, genetics and their correlation in Jalili syndrome: a systematic review. J. Med. Genet. 56, 358–369 (2019).
Article CAS PubMed Google Scholar
Funato, Y., Yamazaki, D. & Miki, H. Renal function of cyclin M2 Mg2+ transporter maintains blood pressure. J. Hypertens. 35, 585–592 (2017).
Article CAS PubMed Google Scholar
Yamazaki, D. et al. The Mg2+ transporter CNNM4 regulates sperm Ca2+ homeostasis and is essential for reproduction. J. Cell Sci. 129, 1940–1949 (2016).
CAS PubMed Google Scholar
Funato, Y. et al. Membrane protein CNNM4-dependent Mg2+ efflux suppresses tumor progression. J. Clin. Investig. 124, 5398–5410 (2014).
Article PubMed PubMed Central Google Scholar
Yamazaki, D. et al. Cnnm4 deficiency suppresses Ca(2+) signaling and promotes cell proliferation in the colon epithelia. Oncogene 38, 3962–3969 (2019).
Article CAS PubMed Google Scholar
Simon J. et al. Magnesium accumulation upon cyclin M4 silencing activates microsomal triglyceride transfer protein improving NASH. J. Hepatol. 75, 34–45 (2021).
Thyme, S. B. et al. Phenotypic landscape of schizophrenia-associated genes defines candidates and their shared functions. Cell 177, 478–491 (2019). e420.
Article CAS PubMed PubMed Central Google Scholar
Wang, C. Y. et al. Molecular cloning and characterization of a novel gene family of four ancient conserved domain proteins (ACDP). Gene 306, 37–44 (2003).
Article CAS PubMed Google Scholar
Gibson, M. M., Bagga, D. A., Miller, C. G. & Maguire, M. E. Magnesium transport in Salmonella typhimurium: the influence of new mutations conferring Co2+ resistance on the CorA Mg2+ transport system. Mol. Microbiol. 5, 2753–2762 (1991).
Article CAS PubMed Google Scholar
Armitano, J., Redder, P., Guimaraes, V. A. & Linder, P. An essential factor for high Mg(2+) tolerance of Staphylococcus aureus. Front. Microbiol. 7, 1888 (2016).
Article PubMed PubMed Central Google Scholar
Trachsel, E., Redder, P., Linder, P. & Armitano, J. Genetic screens reveal novel major and minor players in magnesium homeostasis of Staphylococcus aureus. PLoS Genet. 15, e1008336 (2019).
Article CAS PubMed PubMed Central Google Scholar
Akanuma, G. et al. Defect in the formation of 70S ribosomes caused by lack of ribosomal protein L34 can be suppressed by magnesium. J. Bacteriol. 196, 3820–3830 (2014).
Article PubMed PubMed Central CAS Google Scholar
Arjona, F. J. & de Baaij, J. H. F. CrossTalk opposing view: CNNM proteins are not Na(+) /Mg(2+) exchangers but Mg(2+) transport regulators playing a central role in transepithelial Mg(2+) (re)absorption. J. Physiol. 596, 747–750 (2018).
Article CAS PubMed PubMed Central Google Scholar
Funato, Y., Furutani, K., Kurachi, Y. & Miki, H. CrossTalk proposal: CNNM proteins are Na(+) /Mg(2+) exchangers playing a central role in transepithelial Mg(2+) (re)absorption. J. Physiol. 596, 743–746 (2018).
Article CAS PubMed PubMed Central Google Scholar
Hardy, S. et al. The protein tyrosine phosphatase PRL-2 interacts with the magnesium transporter CNNM3 to promote oncogenesis. Oncogene 34, 986–995 (2015).
Article CAS PubMed Google Scholar
Sponder, G., Svidova, S., Schweigel, M., Vormann, J. & Kolisek, M. Splice-variant 1 of the ancient domain protein 2 (ACDP2) complements the magnesium-deficient growth phenotype of Salmonella enterica sv. typhimurium strain MM281. Magnes. Res. 23, 105–114 (2010).
CAS PubMed Google Scholar
Goytain, A. & Quamme, G. A. Functional characterization of ACDP2 (ancient conserved domain protein), a divalent metal transporter. Physiol. Genomics 22, 382–389 (2005).
Article CAS PubMed Google Scholar
Huang Y, et al. Structural basis for the Mg²⁺ recognition and regulation of the CorC Mg²⁺ transporter. Science Advances 7, eabe6140 (2021).
Gimenez-Mascarell P, et al. Current Structural knowledge on the cnnm family of magnesium transport mediators. Int. J. Mol. Sci. 20, 1135 (2019).
El-Gebali, S. et al. The Pfam protein families database in 2019. Nucleic Acids Res. 47, D427–D432 (2019).
Article CAS PubMed Google Scholar
Ereno-Orbea, J., Oyenarte, I. & Martinez-Cruz, L. A. CBS domains: Ligand binding sites and conformational variability. Arch. Biochem. Biophys. 540, 70–81 (2013).
Article CAS PubMed Google Scholar
Chen, Y. S. et al. Mg(2+)-ATP sensing in CNNM, a putative magnesium transporter. Structure 28, 324–335 (2020). e324.
Article CAS PubMed Google Scholar
Corral-Rodriguez, M. A. et al. Nucleotide binding triggers a conformational change of the CBS module of the magnesium transporter CNNM2 from a twisted towards a flat structure. Biochem. J. 464, 23–34 (2014).
Article CAS PubMed Google Scholar
Chen, Y. S. et al. The cyclic nucleotide-binding homology domain of the integral membrane protein CNNM mediates dimerization and is required for Mg(2+) efflux activity. J. Biol. Chem. 293, 19998–20007 (2018).
Article CAS PubMed PubMed Central Google Scholar
Gulerez, I. et al. Phosphocysteine in the PRL-CNNM pathway mediates magnesium homeostasis. EMBO Rep. 17, 1890–1900 (2016).
Article CAS PubMed PubMed Central Google Scholar
Kozlov, G. et al. PRL3 pseudophosphatase activity is necessary and sufficient to promote metastatic growth. J. Biol. Chem. 295, 11682–11692 (2020).
Article CAS PubMed PubMed Central Google Scholar
Gimenez-Mascarell, P. et al. Structural basis of the oncogenic interaction of phosphatase PRL-1 with the magnesium transporter CNNM2. J. Biol. Chem. 292, 786–801 (2017).
Article CAS PubMed Google Scholar
Terwilliger, T. C. et al. Iterative model building, structure refinement and density modification with the PHENIX AutoBuild wizard. Acta Crystallogr. D Biol. Crystallogr. 64, 61–69 (2008).
Article CAS PubMed Google Scholar
Gimenez-Mascarell P, et al. Structural Insights into the intracellular region of the human magnesium transport mediator CNNM4. Int. J. Mol. Sci. 20, 6279 (2019).
Hirata, Y., Funato, Y., Takano, Y. & Miki, H. Mg2+-dependent interactions of ATP with the cystathionine-beta-synthase (CBS) domains of a magnesium transporter. J. Biol. Chem. 289, 14731–14739 (2014).
Article CAS PubMed PubMed Central Google Scholar
Payandeh, J. et al. Probing structure-function relationships and gating mechanisms in the CorA Mg2+ transport system. J. Biol. Chem. 283, 11721–11733 (2008).
Article CAS PubMed Google Scholar
Stetsenko, A. & Guskov, A. Cation permeability in CorA family of proteins. Sci. Rep. 10, 840 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
de Baaij, J. H. et al. Membrane topology and intracellular processing of cyclin M2 (CNNM2). J. Biol. Chem. 287, 13644–13655 (2012).
Article PubMed PubMed Central CAS Google Scholar
Garcia-Castano, A. et al. Novel variant in the CNNM2 gene associated with dominant hypomagnesemia. PLoS ONE 15, e0239965 (2020).
Article CAS PubMed PubMed Central Google Scholar
Franken, G. A. C. et al. The phenotypic and genetic spectrum of patients with heterozygous mutations in cyclin M2 (CNNM2). Hum. Mutat. 42, 473–486 (2021).
Article CAS PubMed PubMed Central Google Scholar
Parry, D. A. et al. Mutations in CNNM4 cause Jalili syndrome, consisting of autosomal-recessive cone-rod dystrophy and amelogenesis imperfecta. Am. J. Hum. Genet. 84, 266–273 (2009).
Article CAS PubMed PubMed Central Google Scholar
Polok, B. et al. Mutations in CNNM4 cause recessive cone-rod dystrophy with amelogenesis imperfecta. Am. J. Hum. Genet. 84, 259–265 (2009).
Article CAS PubMed PubMed Central Google Scholar
Hirji, N. et al. Jalili Syndrome: cross-sectional and longitudinal features of seven patients with cone-rod dystrophy and amelogenesis imperfecta. Am. J. Ophthalmol. 188, 123–130 (2018).
Article PubMed PubMed Central Google Scholar
Parveen, A. et al. A novel pathogenic missense variant in CNNM4 underlying Jalili syndrome: Insights from molecular dynamics simulations. Mol. Genet. Genom. Med. 7, e902 (2019).
Google Scholar
Lopez Torres L., Schorderet D., Valmaggia C. & Todorova M. A novel mutation in CNNM 4 (G492C) associated with Jalili Syndrome. Acta Ophthalmol. 93 (2015).
Prasad, M. K. et al. A targeted next-generation sequencing assay for the molecular diagnosis of genetic disorders with orodental involvement. J. Med. Genet. 53, 98–110 (2016).
Article CAS PubMed Google Scholar
Hattori, M., Tanaka, Y., Fukai, S., Ishitani, R. & Nureki, O. Crystal structure of the MgtE Mg2+ transporter. Nature 448, 1072–1075 (2007).
Article ADS CAS PubMed Google Scholar
Payandeh, J. & Pai, E. F. A structural basis for Mg2+ homeostasis and the CorA translocation cycle. EMBO J. 25, 3762–3773 (2006).
Article CAS PubMed PubMed Central Google Scholar
Eshaghi, S. et al. Crystal structure of a divalent metal ion transporter CorA at 2.9 angstrom resolution. Science 313, 354–357 (2006).
Article ADS CAS PubMed Google Scholar
Lunin, V. V. et al. Crystal structure of the CorA Mg2+ transporter. Nature 440, 833–837 (2006).
Article ADS CAS PubMed Google Scholar
Duan, J. et al. Structure of the mammalian TRPM7, a magnesium channel required during embryonic development. Proc. Natl Acad. Sci. USA 115, E8201–E8210 (2018).
Article CAS PubMed PubMed Central Google Scholar
Tomita, A. et al. ATP-dependent modulation of MgtE in Mg(2+) homeostasis. Nat. Commun. 8, 148 (2017).
Article ADS PubMed PubMed Central CAS Google Scholar
Payandeh, J., Pfoh, R. & Pai, E. F. The structure and regulation of magnesium selective ion channels. Biochim. Biophys. Acta 1828, 2778–2792 (2013).
Article CAS PubMed Google Scholar
Guskov, A. et al. Structural insights into the mechanisms of Mg2+ uptake, transport, and gating by CorA. Proc. Natl Acad. Sci. USA 109, 18459–18464 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Edgar, R. C. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32, 1792–1797 (2004).
Article CAS PubMed PubMed Central Google Scholar
Kumar, S., Stecher, G., Li, M., Knyaz, C. & Tamura, K. MEGA X: molecular evolutionary genetics analysis across computing platforms. Mol. Biol. Evol. 35, 1547–1549 (2018).
Article CAS PubMed PubMed Central Google Scholar
Kawate, T. & Gouaux, E. Fluorescence-detection size-exclusion chromatography for precrystallization screening of integral membrane proteins. Structure 14, 673–681 (2006).
Article CAS PubMed Google Scholar
Otwinowski, Z. & Minor, W. [20] Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 276, 307–326 (1997).
Article CAS PubMed Google Scholar
McCoy, A. J. et al. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674 (2007).
Article CAS PubMed PubMed Central Google Scholar
Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr. D Biol. Crystallogr. 66, 213–221 (2010).
Article CAS PubMed PubMed Central Google Scholar
Feng, N. et al. The C2′- and C3′-endo equilibrium for AMP molecules bound in the cystathionine-beta-synthase domain. Biochem. Biophys. Res. Commun. 497, 646–651 (2018).
Article CAS PubMed Google Scholar
Karplus, P. A. & Diederichs, K. Assessing and maximizing data quality in macromolecular crystallography. Curr. Opin. Struct. Biol. 34, 60–68 (2015).
Article CAS PubMed PubMed Central Google Scholar
Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D. Biol. Crystallogr. 66, 486–501 (2010).
Article CAS PubMed PubMed Central Google Scholar
Afonine, P. V. et al. Towards automated crystallographic structure refinement with phenix.refine. Acta Crystallogr. D. Biol. Crystallogr. 68, 352–367 (2012).
Article CAS PubMed PubMed Central Google Scholar
Painter, J. & Merritt, E. A. Optimal description of a protein structure in terms of multiple groups undergoing TLS motion. Acta Crystallogr. D Biol. Crystallogr. 62, 439–450 (2006).
Article PubMed CAS Google Scholar
Chen, V. B. et al. MolProbity: all-atom structure validation for macromolecular crystallography. Acta Crystallogr. D Biol. Crystallogr. 66, 12–21 (2010).
Article CAS PubMed Google Scholar
Jurrus, E. et al. Improvements to the APBS biomolecular solvation software suite. Protein Sci. 27, 112–128 (2018).
Article CAS PubMed Google Scholar
Schuck, P. Size-distribution analysis of macromolecules by sedimentation velocity ultracentrifugation and lamm equation modeling. Biophys. J. 78, 1606–1619 (2000).
Article ADS CAS PubMed PubMed Central Google Scholar
Laue, T. et al., eds. Analytical Ultracentrifugation in Biochemistry and Polymers Science. 90–125. (Royal Society of Chemistry, Cambridge, UK, 1992).
Brookes, E., Demeler, B. & Rocco, M. Developments in the US-SOMO bead modeling suite: new features in the direct residue-to-bead method, improved grid routines, and influence of accessible surface area screening. Macromol. Biosci. 10, 746–753 (2010).
Article CAS PubMed Google Scholar
Brautigam, C. A. Calculations and publication-quality illustrations for analytical ultracentrifugation data. Methods Enzymol. 562, 109–133 (2015).
Article CAS PubMed Google Scholar
Ritchie, T. K. et al. Chapter 11—Reconstitution of membrane proteins in phospholipid bilayer nanodiscs. Methods Enzymol. 464, 211–231 (2009).
Article CAS PubMed PubMed Central Google Scholar
Jenkins, M. L. et al. Structural determinants of Rab11 activation by the guanine nucleotide exchange factor SH3BP5. Nat. Commun. 9, 3772 (2018).
Article ADS PubMed PubMed Central CAS Google Scholar
Lucic, I. et al. Conformational sampling of membranes by Akt controls its activation and inactivation. Proc. Natl Acad. Sci. USA 115, E3940–E3949 (2018).
Article CAS PubMed PubMed Central Google Scholar
Nimigean, C. M. A radioactive uptake assay to measure ion transport across ion channel-containing liposomes. Nat. Protoc. 1, 1207–1212 (2006).
Article CAS PubMed Google Scholar
Webb, B. & Sali, A. Comparative protein structure modeling using MODELLER. Curr. Protoc. Bioinform. 54, 5 6 1–5 6 37 (2016).
Article Google Scholar
Madeira, F. et al. The EMBL-EBI search and sequence analysis tools APIs in 2019. Nucleic Acids Res. 47, W636–W641 (2019).
Article CAS PubMed PubMed Central Google Scholar
Laskowski, R. A., MacArthur, M. W., Moss, D. S. & Thornton, J. M. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291 (1993).
Article CAS Google Scholar
Jo, S., Kim, T., Iyer, V. G. & Im, W. CHARMM-GUI: a web-based graphical user interface for CHARMM. J. Comput. Chem. 29, 1859–1865 (2008).
Article CAS PubMed Google Scholar
Wu, E. L. et al. CHARMM-GUI membrane builder toward realistic biological membrane simulations. J. Comput. Chem. 35, 1997–2004 (2014).
Article CAS PubMed PubMed Central Google Scholar
MacKerell, A. D. et al. All-atom empirical potential for molecular modeling and dynamics studies of proteins. J. Phys. Chem. B 102, 3586–3616 (1998).
Article CAS PubMed Google Scholar
MacKerell, A. D. Jr., Feig, M. & Brooks, C. L. 3rd Improved treatment of the protein backbone in empirical force fields. J. Am. Chem. Soc. 126, 698–699 (2004).
Article CAS PubMed Google Scholar
Best, R. B. et al. Optimization of the additive CHARMM all-atom protein force field targeting improved sampling of the backbone phi, psi and side-chain chi(1) and chi(2) dihedral angles. J. Chem. Theory Comput. 8, 3257–3273 (2012).
Article CAS PubMed PubMed Central Google Scholar
Klauda, J. B. et al. Update of the CHARMM all-atom additive force field for lipids: validation on six lipid types. J. Phys. Chem. B 114, 7830–7843 (2010).
Article CAS PubMed PubMed Central Google Scholar
Beglov, D. & Roux, B. Finite representation of an infinite bulk system: solvent boundary potential for computer simulations. J. Chem. Phys. 100, 9050–9063 (1994).
Article ADS CAS Google Scholar
Jorgensen, W. L., Chandrasekhar, J., Madura, J. D., Impey, R. W. & Klein, M. L. Comparison of simple potential functions for simulating liquid water. J. Chem. Phys. 79, 926–935 (1983).
Article ADS CAS Google Scholar
Darden, T., York, D. & Pedersen, L. Particle mesh Ewald: an N· log (N) method for Ewald sums in large systems. J. Chem. Phys. 98, 10089–10092 (1993).
Article ADS CAS Google Scholar
Ryckaert, J.-P., Ciccotti, G. & Berendsen, H. J. Numerical integration of the cartesian equations of motion of a system with constraints: molecular dynamics of n-alkanes. J. Comput. Phys. 23, 327–341 (1977).
Article ADS CAS Google Scholar
Phillips, J. C. et al. Scalable molecular dynamics with NAMD. J. Comput. Chem. 26, 1781–1802 (2005).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank Katharina Dürr and her lab members for assistance with the initial screening of CorB orthologs; Alexei Gorelik for advice on crystallographic data processing; Nazzareno D’Avanzo for assistance with liposome transport assay; Satchal Erramilli for the TmCorA plasmid and nanodisc protocol; Martin Schmeing and Camille Fortinez for crystallographic data collection at ALS; Canadian Light Source (CLS) staff: Michel Fodje, Shaun Labiuk, Bulat Gabidullin, and Kathryn Janzen. CLS is supported by the Canada Foundation for Innovation, the Natural Sciences and Engineering Research Council (NSERC), the National Research Council, the Canadian Institutes of Health Research (CIHR), the Government of Saskatchewan, and the University of Saskatchewan. Y.S.C. is supported by a Canadian Institutes of Health Research Doctoral Research Award (GSD-167011). This work is supported by Natural Sciences and Engineering Research Council of Canada grant RGPIN-2020-07195 to K.G.

Author information

Authors and Affiliations

Department of Biochemistry & Centre de Recherche en Biologie Structurale, McGill University, Montréal, QC, Canada
Yu Seby Chen, Guennadi Kozlov, Rayan Fakih & Kalle Gehring
Department of Biochemistry and Microbiology, University of Victoria, Victoria, BC, Canada
Brandon E. Moeller & John E. Burke
Department of Biochemistry and Molecular Biology, University of Chicago, Gordon Center for Integrative Science, Chicago, IL, USA
Ahmed Rohaim & Benoît Roux
Department of Biophysics, Faculty of Science, Cairo University, Giza, Egypt
Ahmed Rohaim
Department of Biochemistry and Molecular Biology, The University of British Columbia, Vancouver, BC, Canada
John E. Burke

Authors

Yu Seby Chen
View author publications
You can also search for this author in PubMed Google Scholar
Guennadi Kozlov
View author publications
You can also search for this author in PubMed Google Scholar
Brandon E. Moeller
View author publications
You can also search for this author in PubMed Google Scholar
Ahmed Rohaim
View author publications
You can also search for this author in PubMed Google Scholar
Rayan Fakih
View author publications
You can also search for this author in PubMed Google Scholar
Benoît Roux
View author publications
You can also search for this author in PubMed Google Scholar
John E. Burke
View author publications
You can also search for this author in PubMed Google Scholar
Kalle Gehring
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

Y.S.C. designed experiments, cloned constructs, performed small-scale screenings, purified proteins, solved crystal structures, and performed liposome transport assays. G.K. assisted with crystal screening and performed ITC experiments. B.E.M. performed HDX-MS experiments. A.R. performed molecular dynamics simulations. R.F. performed AUC and ITC experiments of soluble domains. Y.S.C. and K.G. wrote the paper. K.G., J.E.B., and B.R. oversaw the research.

Corresponding author

Correspondence to Kalle Gehring.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Communciations thanks Luis Alfonso Martinez-Cruz and other, anonymous, reviewers for their contributions to the peer review of this work. Peer review reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Supplementary Movie 1

Supplementary Movie 2

Supplementary Movie 3

Supplementary Movie 4

Description of Additional Supplementary Files

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Chen, Y.S., Kozlov, G., Moeller, B.E. et al. Crystal structure of an archaeal CorB magnesium transporter. Nat Commun 12, 4028 (2021). https://doi.org/10.1038/s41467-021-24282-7

Download citation

Received: 11 February 2021
Accepted: 11 June 2021
Published: 29 June 2021
DOI: https://doi.org/10.1038/s41467-021-24282-7

This article is cited by

Hypomagnesaemia with varying degrees of extrarenal symptoms as a consequence of heterozygous CNNM2 variants
- Willem Bosman
- Gijs A. C. Franken
- Joost G. J. Hoenderop
Scientific Reports (2024)
Structural and functional comparison of magnesium transporters throughout evolution
- G. A. C. Franken
- M. A. Huynen
- J. H. F. de Baaij
Cellular and Molecular Life Sciences (2022)

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.