CrossTalk opposing view: CNNM proteins are not Na⁺/Mg²⁺ exchangers but Mg²⁺ transport regulators playing a central role in transepithelial Mg²⁺ (re)absorption

Introduction

Magnesium (Mg²⁺) is indispensable for many physiological processes in the cell impacting organ function (de Baaij et al. 2015). Intracellular concentrations of free Mg²⁺ are maintained within a narrow range (0.5–1.2 mm) through tightly regulated Mg²⁺ influx and efflux mechanisms (Ebel & Gunther, 1980). Inhibition of Mg²⁺ efflux by substitution of extracellular Na⁺ by choline in mammalian cells indicates that Mg²⁺ extrusion is Na⁺ dependent (Guther et al. 1984). Thus, it is postulated that Mg²⁺ efflux occurs via a Na⁺/Mg²⁺ exchanger, this notion being corroborated in a large number of studies in different cell models (Romani, 2007).

Under the premise that differential gene expression is involved in the maintenance of cellular Mg²⁺ homeostasis, microarray analyses in epithelial cells exposed to low extracellular Mg²⁺ concentrations have been classically used to designate proteins involved in Mg²⁺ transport (Quamme, 2010). One of the protein families identified with this approach is the CNNM family. CNNM proteins have been proposed to facilitate epithelial Mg²⁺ extrusion since at least two members of this family, CNNM2 and CNNM4, localize in the basolateral membrane of epithelial cells where apical‐to‐basolateral Mg²⁺ transport occurs (Stuiver et al. 2011; Yamazaki et al. 2013). However, in this review of the literature, we will demonstrate that the evidence supporting this hypothesis is controversial and several findings suggest that CNNMs cannot function as Na⁺/Mg²⁺ exchangers.

Topology

The CNNM family (CNNM1–4) shares a high homology with the bacterial CorC protein, which is related to the maintenance of Mg²⁺ and Co²⁺ homeostasis (Hmiel et al. 1989). However, CorC requires CorA to function as a cation transporter and does not transport Mg²⁺ itself (Gibson et al. 1991). Within the CNNM family, membrane topology studies have demonstrated that CNNM2 is composed of dimers of three transmembrane domains (Fig. 1) (de Baaij et al. 2012). This number of transmembrane domains is significantly lower than that in typical Mg²⁺ channels/transporters such as TRPM6/M7 (tetramers of 6 transmembrane domains), or SLC41A1/A3 (dimers of 11 transmembrane domains) (Voets et al. 2004; Bates‐Withers et al. 2011; Kolisek et al. 2012; de Baaij et al. 2016). Additionally, a putative pore facilitating cation transport is unlikely to be present in CNNM2 since the re‐entrant loop in its tertiary structure does not completely span the cell membrane and therefore does not facilitate the formation of a typical pore (Fig. 1) (de Baaij et al. 2012).

Figure 1. Membrane topology and subcellular localization of CNNM2.

The left panel indicates the membrane topology of CNNM2 showing three full membrane‐spanning domains and a large intracellular C‐terminus containing two CBS modules that bind Mg‐ATP. Reported mutations have been indicated by purple dots. The right panel shows the basolateral localization of CNNM2 in the distal convoluted tubule kidney cell.

A common feature of CNNM proteins is the presence of intracellular cystathionine β‐synthase (CBS) domains (de Baaij et al. 2012). The CBS domains are the main site of regulation, as has recently been demonstrated by a number studies showing the interaction with phosphatases of the regenerating liver (PRL) proteins (Hardy et al. 2015; Gulerez et al. 2016; Kostantin et al. 2016). CBS domains regulate protein function upon binding Mg‐ATP (Corral‐Rodriguez et al. 2014; Hirata et al. 2014). Most intracellular Mg²⁺ is bound to ATP. Intracellular Mg‐ATP changes in concentration can be sensed by CNNM proteins and may regulate their function(s). Although the exact role between ATP production and Mg²⁺ concentrations is unknown, an interesting hypothesis is that CNNM proteins function as indirect sensors of intracellular Mg²⁺ concentrations.

Function

Since its identification in 2003, the function of CNNM proteins has been subject to debate (Wang et al. 2003). Because of their homology to cyclin proteins, CNNMs were initially hypothesized to have a role in cell cycle progression. However, a cyclin function was never demonstrated. For years, CNNM2 has been the main focus of functional CNNM experiments, as the protein was identified in renal distal convoluted tubule (DCT) cells (Goytain & Quamme, 2005; Stuiver et al. 2011; de Baaij et al. 2012). In the DCT, Mg²⁺ is actively reabsorbed and urinary Mg²⁺ excretion is determined. In human studies, TRPM6 has been established as the Mg²⁺ uptake mechanism in this nephron segment (Schlingmann et al. 2002). However, recent studies in mice show that TRPM7 may be equally important in determining renal Mg²⁺ excretion (Chubanov et al. 2016). The protein mediating Mg²⁺ extrusion still has to be identified.

As a result, several groups have examined the potential Mg²⁺ transport properties of CNNM2 using electrophysiology in a multitude of cell models. In 2005, Quamme and his team showed CNNM2‐mediated Mg²⁺ currents in Xenopus laevis oocytes with a K _m for Mg²⁺ of 0.56 ± 0.05, establishing CNNM2 as a Mg²⁺ transporter (Goytain & Quamme, 2005). In the DCT, luminal Mg²⁺ concentrations range between 0.2 and 0.7 mm Mg²⁺ (Dai et al. 2001), supporting CNNM2‐mediated Mg²⁺ uptake. However, in mammalian cells, these findings could not be reproduced by two independent research groups (Stuiver et al. 2011; Sponder et al. 2016), although a minor Mg²⁺‐dependent Na⁺ current was detected in HEK293 cells overexpressing CNNM2 (Stuiver et al. 2011; Yamazaki et al. 2013). From 2013 onwards, CNNM4 function has also been examined, as it was shown to be expressed in the colon and to be involved in cancer progression (Yamazaki et al. 2013; Funato et al. 2014). In electrophysiological examinations of CNNM4‐expressing cells, no Mg²⁺ currents were detected (Yamazaki et al. 2013). Because electrophysiology does not permit the detection of electroneutral Mg²⁺ transport, this led several authors to suggest that CNNM proteins might be Na⁺‐dependent Mg²⁺ exchangers at a 2:1 (Na⁺:Mg²⁺) ratio.

To tackle this hypothesis, alternative experiments were staged. Several groups have aimed to measure CNNM activity using fluorescent Mg²⁺ probes, resulting in a large pile of conflicting data. First, using the Mg²⁺‐sensitive fluorescent Magnesium Green indicator, the group of Miki showed that overexpression of CNNM4 or CNNM2 decreased cellular Mg²⁺ concentrations (Yamazaki et al. 2013; Funato et al. 2014). Their data suggest that the Mg²⁺ efflux is Na⁺ dependent, supporting the exchange theory. Second, Tremblay and colleagues demonstrated CNNM3‐dependent Mg²⁺ uptake using Mag‐Fura‐2 (Hardy et al. 2015; Kostantin et al. 2016). As their experiments are performed in a physiological 140 mm Na⁺‐containing buffer, this contradicts Na⁺ dependency. Third, Kolisek's group could not detect CNNM2‐mediated Mg²⁺ influx nor efflux, questioning the ability of CNNMs to transport Mg²⁺ (Sponder et al. 2016). Several important limitations of these experiments have to be considered. All experiments were performed in non‐physiological Mg²⁺ concentrations, including loading steps using 10 mm up to 40 mm Mg²⁺. Moreover, the Mg²⁺ probes are notoriously difficult to work with as they have dissociation constants above physiological Mg²⁺ concentrations, as well as sensitivity to Ca²⁺ (Szmacinski & Lakowicz, 1996). Additionally, Magnesium Green is a non‐ratiometric probe, which means that the experiments are extremely sensitive to osmolality changes.

To overcome these hurdles, we have recently developed a stable isotope assay using ²⁵Mg²⁺ and mass spectrometry. This set‐up measures cellular Mg²⁺ uptake using the physiological condition of 1 mm Mg²⁺. Our ²⁵Mg²⁺ uptake experiments, performed in HEK293 cells overexpressing CNNM2, demonstrate that CNNM2 increases cellular Mg²⁺ uptake rather than Mg²⁺ efflux (Arjona et al. 2014). This conclusion is illustrated by the fact that inhibition of Mg²⁺ efflux through Quinidin, an inhibitor of Na⁺/Mg²⁺ exchange, did not affect CNNM2‐mediated ²⁵Mg²⁺ cellular accumulation. Interestingly, this uptake is independent of Na⁺, as replacing Na⁺ with NMDG had no effect, nor did ouabain, an inhibitor of the Na⁺/K⁺‐ATPase (Arjona et al. 2014). However, the CNNM2‐dependent Mg²⁺ uptake was completely abolished by using 2‐aminoethoxydiphenyl borate (2‐APB), which is a known inhibitor of TRPM7 (Chokshi et al. 2012; Arjona et al. 2014).

Altogether, the functional experiments in a wide range of models and techniques show that CNNM2 is not a Na⁺/Mg²⁺‐exchanger, as the Mg²⁺ transport is independent of Na⁺ and can be bi‐directional (influx and efflux). Moreover, the discrepancy between the different models shows that CNNM2 function is largely dependent on other proteins present in the cell. This is suggestive of a regulating function rather than a transport function. Given that the CNNM2‐dependent ²⁵Mg²⁺ uptake could be blocked by 2‐APB, TRPM7, a channel involved in Mg²⁺ homeostasis, is an interesting candidate for regulation by CNNM2 (Paravicini et al. 2012).

In an in vivo context, dysfunctional CNNM2 has been linked to defects in renal Mg²⁺ reabsorption in mice and zebrafish (Arjona et al. 2014; Funato et al. 2017). In addition, CNNM4 knockout in mice results in impaired intestinal Mg²⁺ absorption (Yamazaki et al. 2013). Thus, though CNNM proteins are not Mg²⁺ transporters, they regulate the body Mg²⁺ balance.

Conclusion and perspective

Hence, the following evidence supports that CNNM proteins are not Na⁺/Mg²⁺ exchangers: (i) CNNM proteins do not comply with the consensual topological attributes of a typical Mg²⁺ transporter, i.e. abundant transmembrane domains and a cation pore formed by membrane‐spanning domains; (ii) conflicting functional data of CNNM‐mediated Mg²⁺ extrusion between different cellular models indicate that the Mg²⁺ extrusion function of CNNM proteins depends on the Mg²⁺ channels/transporters specifically expressed by each cellular model; and (iii) CNNM2‐mediated Mg²⁺ extrusion in epithelial cells is not dependent on Na⁺ in a series of experiments. Altogether, this shows that CNNM proteins are not Na⁺/Mg²⁺ exchangers. Future research should focus on the identification of the protein partners of CNNMs to elucidate the CNNM‐dependent regulation of Mg²⁺ transport.

Call for comments

Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief (250 word) comment. Comments may be submitted up to 6 weeks after publication of the article, at which point the discussion will close and the CrossTalk authors will be invited to submit a ‘LastWord’. Please email your comment, including a title and a declaration of interest, to jphysiol@physoc.org. Comments will be moderated and accepted comments will be published online only as ‘supporting information’ to the original debate articles once discussion has closed.

Additional information

Competing interests

None declared.

Author contributions

Both authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

This work was financially supported by grants from the Netherlands Organization for Scientific Research (NWO VENI 016.186.012) and the Dutch Kidney Foundation (Kolff 14OKG17).

Biographies

Francisco J. Arjona (1980) is a postdoctoral researcher at the department of Physiology in the Radboud University Medical Centre (Nijmegen, The Netherlands). His current research focuses on unravelling the proteins responsible for Mg²⁺ handling in the kidney, providing their functional characterization in health and disease. As a molecular and animal physiologist, he combines in vitro and in vivo approaches with the aim of providing adequate translational and mechanistic insights.

Jeroen H. F. de Baaij (1987) is a physiologist interested in the molecular and genetic origin of renal electrolyte disorders, including hypomagnesaemia. He currently works as assistant professor at the Radboud University Medical Centre in Nijmegen, The Netherlands, where he studies the genetic origin of disease in patients with magnesium disturbances and the function of genes involved in renal Mg²⁺ reabsorption.