Magnesium Counteracts Vascular Calcification: Passive Interference or Active Modulation?

In healthy individuals, serum Mg²⁺ concentrations are carefully balanced between 0.7 and 1.1 mmol/L by the coordinate action of the intestine, bone, and kidney.¹¹ Approximately 30% of the dietary Mg²⁺ intake is absorbed in the small intestine and colon.¹² The bone serves as the body’s Mg²⁺ store as 60% of the total Mg²⁺ is embedded at the surface of the hydroxyapatite crystals.¹³ The kidney is the main organ controlling systemic Mg²⁺ homeostasis, where transport is highly regulated by hormonal and intrarenal factors, including epidermal growth factor, insulin, pH, ATP, and estrogens.^14–18 Daily, 95% of the filtered Mg²⁺ is reabsorbed along the nephron.¹¹ The largest amount of Mg²⁺ (50%–70%) is reabsorbed paracellularly in the thick ascending limb of the loop of Henle.¹⁹ Fine-tuning of Mg²⁺ reabsorption is achieved in the distal convoluted tubule, where transient receptor potential melastatin type 6 (TRPM6) cation channels mediate apical Mg²⁺ uptake and solute carrier family 41 members 1 and 3 (SLC41A1/A3). Na⁺/Mg²⁺-exchangers facilitate basolateral Mg²⁺ extrusion.^20–22

When renal function declines, the fractional excretion of Mg²⁺ is increased to maintain normal serum Mg²⁺ concentrations. Therefore, patients with CKD stages 1 to 3 (glomerular filtration rate >30 mL/min) generally have normal Mg²⁺ concentrations.²³ As renal function further deteriorates during CKD stages 4 and 5, raising fractional excretion eventually fails to compensate for reduced glomerular filtration causing hypermagnesemia, especially if glomerular filtration rate drops <10 mL/min.²⁴ In a recent cohort of 365 hemodialysis patients, a mean Mg²⁺ concentration of 0.98 mmol/L was measured, which is in the high-normal range of normal serum Mg²⁺ concentrations.²⁵

In dialysis patients, the serum Mg²⁺ concentration is largely dependent on the dialysate Mg²⁺ concentration.²⁶ Dialysates for both peritoneal dialysis and hemodialysis normally contain 0.75 mmol/L Mg²⁺. Given that 30% of serum Mg²⁺ is protein bound, a dialysate Mg²⁺ concentration of 0.75 mmol/L generally results in mild hypermagnesemia (1.0–1.2 mmol/L).²⁷ The protein-binding properties of Mg²⁺ may cause misinterpretation of measured serum concentrations.¹¹ The development of acidosis in ESRD potentially decreases the fraction of Mg²⁺ bound to proteins, which in CKD may result in an increased ionized serum Mg²⁺ concentration as renal compensatory mechanisms fail.²⁸ Measurements of ionized Mg²⁺ therefore provide a more reliable estimation of Mg²⁺ bioavailability; however, this is clinically largely unavailable.²⁹ Other factors such as diet, diabetes mellitus, and medication may greatly affect serum Mg²⁺ concentrations in CKD patients. For instance, the use of proton pump inhibitors to treat gastric acid production hampers intestinal Mg²⁺ reabsorption and, therefore, has been associated with increased risk of hypomagnesemia.^30,31 Hypomagnesemia is associated with the progression to ESRD in patients with diabetes mellitus type 2 and in patients with diabetic nephropathy.^32,33

Hypomagnesemia (serum Mg²⁺ concentration <0.7 mmol/L) is a well-established risk factor for cardiovascular disease, events, and mortality in the general population and in CKD patients.^25,34–37 In the general population, dietary Mg²⁺ intake is associated with all-cause mortality, reduced risk of stroke, heart failure, and diabetes mellitus.³⁸ Moreover, serum Mg²⁺ concentration is inversely associated with a 66% and a 36% increased risk for death from heart failure (<0.7 mmol/L) and coronary heart disease (<0.8 mmol/L), respectively.^39,40 To assess whether Mg²⁺ status is linked to cardiovascular disease, a detailed overview of studies on the association between the circulating Mg²⁺ concentration and cardiovascular disease risk in both healthy and hemodialysis cohorts is provided in Tables 1 and 2, respectively. For both tables, our aim was to assess available evidence on associations between circulating Mg²⁺ concentration and cardiovascular disease outcome. Accordingly, studies on effects of dietary Mg²⁺ and associations between serum Mg²⁺ and indirect measures for cardiovascular disease, such as carotid intima-media thickness and hypertension, were excluded. Our overview of the available clinical association studies, as well as 2 previously published meta-analyses, indicates that serum Mg²⁺ concentration is inversely associated with cardiovascular risk in both healthy cohorts and hemodialysis cohorts.^67,68

Therefore, the current reference range of 0.7 to 1.1 mmol/L for blood Mg²⁺ concentration is under debate. An international team of Mg²⁺ researchers proposed that the reference values for normal Mg²⁺ concentration may be too low and should be reconsidered because the current range was derived from population studies from the 1970s.^69,70 Mg²⁺ intake is generally insufficient, and Mg²⁺ deficiency-related clinical complications may already arise in low-normal Mg²⁺ values, suggesting that a higher blood Mg²⁺ concentration is beneficial.⁶⁹ This notion is supported by data from CKD patients; in the CONTRAST study (Convective Transport Study), the relative risk for mortality in patients with serum Mg²⁺ concentrations <1.14 mmol/L was significantly increased compared with patients with lower serum concentrations.²⁵ Although Mg²⁺ concentration was negatively associated with cardiovascular risk in a recent Japanese cohort study, it is important to note that concentrations >1.27 mmol/L were found to be associated with increased risk.³⁴ Interestingly, similar trends were observed in heart failure patients as serum Mg²⁺ concentrations ≥1.05 mmol/L were associated with increased cardiovascular mortality.⁷¹ These studies suggest that depending on the population and the disease state, the optimal Mg²⁺ concentration may be in the range of 0.9 to 1.2 mmol/L. However, this hypothesis should be further supported by studies defining the optimal Mg²⁺ concentration based on clinical outcomes. This is essential to set a novel clinically relevant reference range for serum Mg²⁺ concentrations.

Determining a clear upper level is important as hypermagnesemia (currently set at >1.1 mmol/L) may result in nausea and vomiting, flushing, and headaches. Severe hypermagnesemia (>3.0 mmol/L) may lead to cardiac complications, such as bradycardia and hypotension.¹¹ However, the positive association of high serum Mg²⁺ with survival found in CKD patients suggests that a state of mild hypermagnesemia is predominantly protective in this population, possibly through the impact of Mg²⁺ on vascular function. In the following section, we will briefly review the available data on the role of Mg²⁺ in common cardiovascular diseases.

Moderate-to-severe Mg²⁺ deficiency is associated with arrhythmia and atrial fibrillation.^11,52 Reduction in cytosolic Mg²⁺ associated with hypomagnesemia can cause significant alterations in the myocardial action potential.⁷² In patients with normal cardiac conduction maintenance Mg²⁺ infusion resulted in prolongation of the electrocardiography P-R interval, A-H interval, atrioventricular refractory period, and sinoatrial conduction time.⁷³ Mg²⁺ has been widely considered as treatment for arrhythmic disorders, and success of Mg²⁺ treatment has been shown to largely depend on arrhythmia type. For example, Mg²⁺ is beneficial in torsades de pointes and is currently the first line of therapy.⁷⁴ Ventricular fibrillation and tachycardia do not respond to Mg²⁺.⁷⁵ Although a meta-analysis did not demonstrate beneficial effects of Mg²⁺ treatment on acute atrial fibrillation,⁷⁶ a recent editorial calls attention to limitations in sample size, patient selection, and follow-up of the current available studies and therefore emphasizes the need for further trial data to accurately assess a role for Mg²⁺ in improving the management of atrial fibrillation.⁷⁷

Hypomagnesemia is associated with an increased risk for coronary artery disease and carotid atherosclerosis.^40,78 Coronary artery calcification associated with atherosclerosis is a strong predictor of cardiovascular events in the general and the CKD population.^79,80 In CKD, intimal calcifications associated with atherosclerosis are prevalent.⁸¹ Recently, Mg²⁺ status was found to be inversely associated with coronary artery calcification density in ESRD patients, particularly those with high serum Pi concentrations (>1.40 mmol/L).⁸² Associations between serum Mg²⁺ concentration and subclinical markers of atherosclerosis and the presence of vascular calcification in CKD patients have been reported extensively.^{59,60,62,63,82–84} Although the potential mechanisms remain largely unclear and are beyond the scope of this review, low intracellular Mg²⁺ in vitro is linked with a proinflammatory and a proatherogenic vascular phenotype through increased production of reactive oxygen species, activation of NF-κB (nuclear factor kappa-beta) and cytokines, and proteasome activity in endothelial cells.^78,85,86

The vasoprotective properties of Mg²⁺ are reinforced by multiple in vivo studies. In low-density lipoprotein receptor^−/− and ApoE^−/− transgenic mouse models of atherosclerosis, Mg²⁺ supplementation reduced cholesterol and triglyceride levels and atherogenesis in the aortic sinus.^87–89 Endothelial dysfunction in aortas of inbred low serum Mg²⁺ mice has been associated with reduced TRPM7 expression levels, illustrating a potential link between intracellular Mg²⁺ and onset of atherosclerosis.⁹⁰ A Mg²⁺-deficient diet in rats led to increased oxidative stress, reduced superoxide dismutase, catalase, and increased collagen synthesis in the arterial wall.⁹¹ Moreover, Mg²⁺-deficient mice demonstrated aortic thinning and structural alterations in collagen and elastin fibers, possibly related to matrix metalloprotease expression and activity.⁹² In studies using Abcc^−/− and Enpp1^asj mice, which develop extensive vascular calcification, Mg²⁺ restriction and supplementation experiments demonstrated a preventive role for Mg²⁺ in the development of ectopic and connective tissue calcification.^93–95 The effects of Mg²⁺ on vascular calcification are discussed in the next section of this review.

Hypertension is an important contributor to the development of cardiovascular events and is common in CKD because it develops in >80% of patients during stages 4 and 5.⁹⁶ The antihypertensive properties of Mg²⁺ are likely attributed to its Ca²⁺ antagonistic properties.⁹⁷ Alternative vasodilatory actions of Mg²⁺ are the associated increased production of prostaglandin I₂ and nitric oxide in endothelial cells.⁹⁸

Although the role of Mg²⁺ in hypertension has been controversial, a recent meta-analysis of randomized double-blind placebo-controlled trials revealed a significant causal antihypertensive effect of Mg²⁺ supplementation.⁹⁹ However, given the modest effect size of 2 mm Hg, the clinical relevance of this effect is questionable. Although the mechanisms are poorly understood, Mg²⁺ supplementation is worldwide the first line of treatment for preeclampsia that is widely advocated by the World Health Organization to prevent early childhood mortality.¹⁰⁰ Despite these results, in the context of CKD, it should be noted that in 14 hemodialysis patients treated with low dialysate Ca²⁺ (1.25 mmol/L), increased Mg²⁺ dialysate from 0.25 to 0.75 mmol/L paradoxically prevented blood pressure drops associated with dialysis.¹⁰¹ In these patients, postdialysis Mg²⁺ concentrations fell by 35%, whereas intracellular Ca²⁺ fell by 7.7%. The authors propose that given the Ca²⁺-blocking properties of Mg²⁺, subnormal levels of Mg²⁺ in combination with lower extracellular Ca²⁺ may have resulted in reduced cardiovascular contractility, which was reversed by increasing dialysate Mg²⁺ concentration.¹⁰¹

Diabetes mellitus is an established and well-known risk factor for cardiovascular disease.¹⁰² Insulin resistance is the main cause of diabetes mellitus type 2 and has been found to be associated with the presence and severity of coronary artery disease.¹⁰³ In fact, patients with diabetes mellitus often present with more severe atherosclerosis, characterized by larger and more inflammatory necrotic cores and more extensive lesion calcification.¹⁰⁴ Diabetes mellitus is strongly associated with hypomagnesemia, of which the potential mechanisms have been reviewed in detail previously.³³ In addition, dietary Mg²⁺ intake was associated with type 2 diabetes mellitus in a recent dose–response meta-analysis.³⁸ However, any causal relationship between hypomagnesemia and the incidence of cardiovascular disease in diabetes mellitus has yet to be identified.

The link between Mg²⁺ status and cardiovascular disease in humans and the impact of Mg²⁺ interventions on vascular disease in animal models illustrate that Mg²⁺ supplementation should be considered as potential strategy to counteract vascular disease. The field of cardiovascular research now faces the challenge to move forward from association studies toward experimental studies. The many positive effects that were shown in association studies (Table 1) warrant further clinical investigations to elucidate the treatment potential of Mg²⁺. This may be of particular interest for patients with CKD because these patients suffer from disturbed mineral homeostasis and increased cardiovascular risk. In this review, we will further focus on the mechanisms underlying beneficial effects of Mg²⁺ in vascular calcification.

Severe hyperphosphatemia in ESRD patients paradoxically leads to both bone demineralization and vascular calcification.¹⁰⁵ In the course of the disease, high Pi concentrations persistently elevate FGF23 (fibroblast growth factor 23) levels. The resulting defective inhibitory regulation of PTH (parathyroid hormonde) secretion and decreased 1,25[OH)]₂D₃ (1,25-dihydroxyvitamin D) synthesis results in reduced intestinal Ca²⁺ and Pi absorption and high bone turnover.^106–108 FGF23-specific signaling is regulated by the FGF receptor 1-klotho complex in the distal convoluted tubule of the nephron and the parathyroid.^109,110 However, because klotho expression levels decline over the course of CKD development as functional renal mass decreases, FGF23 signaling is compromised even further.¹¹¹ Administration of recombinant α-klotho effectively attenuated CKD progression and CKD-associated cardiac remodeling, highlighting the importance of klotho signaling pathways in CKD and cardiovascular health.¹¹²

Disturbances in these regulatory axes manifest as CKD-mineral bone disorder, which is characterized by severe hyperphosphatemia and hypercalcemic episodes providing a permissive milieu for vascular calcification (Figure 1). This CKD-induced calcification milieu is associated with loss of proteins that act as local and circulating inhibitors of soft tissue calcification such as fetuin-A and MGP (matrix gla protein).¹¹³ Reduced levels of these proteins are associated with vascular calcification in hemodialysis patients.^114,115 In vivo studies report extensive calcification in knockout mouse models of fetuin-A, MGP, and of local inhibitors native to VSMCs, such as osteoprotegerin and pyrophosphate.^116–119

The development of the calcification milieu and the loss of calcification inhibitors promote the formation of amorphous Ca²⁺-Pi particles (ACPs). Nucleation and maturation of these ACPs into hydroxyapatite crystals in the vessel wall initiate the process of vascular calcification. Ultimately, however, vascular calcification is an active cell-mediated process further potentiated by ACP phagocytosis and increased Pi uptake mediated by sodium-dependent phosphate transporters 1 and 2, during which VSMCs transdifferentiate from a contractile into an osteoblast-like phenotype.^120–123 VSMC transdifferentiation is typically characterized by the expression of genes that are normally restricted to bone tissue, such as BMP2 (bone morphogenetic protein 2), osterix, RUNX2 (runt-related transcription factor 2), and alkaline phosphatase.¹²⁴ Expression of these osteoinductive genes induces matrix remodeling and mineralization and is accompanied by decreased expression of VSMC lineage markers, such as transgelin and calponin.¹²⁵

Transdifferentiated VSMCs participate in the local spread of calcification by diminished synthesis of calcification inhibitors, the release of Ca²⁺-loaded exosomes (matrix vesicles) and apoptosis. All these factors contribute to the calcification by local Ca²⁺ release and providing ACP nucleation sites in the extracellular matrices of surrounding VSMCs.^126,127 In healthy VSMCs, exosomes are loaded with fetuin-A and MGP and are secreted to maintain vessel compliance.¹²⁸ In calcifying VSMCs, however, the presence of calcification inhibitors in these exosomes is depleted and replaced by a protein–lipid complex consisting of phosphatidyl serine and annexin A6, converting the exosome into a potent nucleation site.^129,130 Furthermore, in vitro studies have shown that exposure of VSMCs to artificial ACP and calciprotein particles (CPPs) similar to those found in uremic sera increased exosome secretion, which in turn enhanced calcification.^131,132 Because of the presence of fetuin-A in human serum, CPPs containing ACP have recently shown to form, rather than crystalline, hydroxyapatite.¹³³ These primary CPPs mature spontaneously into secondary CPPs containing crystalline Ca²⁺-Pi and were found in sera of CKD patients.¹³⁴ Exposure of secondary CPP to fixated cells did not result in calcification, demonstrating a role for VSMC in vascular calcification possibly related to exosome secretion.¹³¹

Similar to exosomes, Ca²⁺-loaded apoptotic bodies released from VSMCs undergoing apoptosis form larger nucleation sites and promote calcification in neighboring cells potentially by causing local Ca²⁺ spikes.¹³⁵ Apoptosis plays a role in the initiation and acceleration in VSMC calcification and is induced by an intracellular Ca²⁺ burst after excessively phagocytosed ACP undergoes lysosomal breakdown.¹³⁶ Altogether, the calcification milieu, loss of calcification inhibitors, and VSMC transdifferentiation illustrate the dynamic and complex nature of vascular calcification activating a vicious cycle of events amplifying the calcification process.

Reducing Pi load is an important therapeutic strategy to minimalize the risk of cardiovascular complications, including vascular calcifications.¹³⁷ Mg²⁺-based Pi binders have been shown to reduce serum Pi concentrations efficiently and were introduced in the early 1980s.^138,139 The introduction of Mg²⁺-based binders was mainly to replace Ca²⁺- or aluminum-based drugs, which can lead to vascular calcification, osteomalacia, dementia, and anemia.^138,140

Despite the promising first clinical trials testing the use of Mg²⁺-hydroxide and Mg²⁺-carbonate in dialysis patients, concerns rose about hypermagnesemia and gastrointestinal complications.^139,141 Instead, a combination of Ca²⁺-acetate and Mg²⁺-carbonate has been used since and showed similar efficacy in reducing serum Pi concentrations, which was demonstrated in 255 hemodialysis patients.¹⁴² Pi concentrations below the KDIGO (Kidney Disease: Improving Global Outcomes) target of 1.78 mmol/L or lower were achieved in the Ca²⁺-acetate and Mg²⁺-carbonate group after 16 days compared with 30 days in the conventional sevelamer group.¹⁴² Mild hypermagnesemia remained an issue as the serum Mg²⁺ concentration increased by 0.3 mmol/L. Close monitoring of serum Mg²⁺ concentrations in CKD patients and reduced dialysate Mg²⁺ in ESRD patients from 0.75 to 0.5, to 0.25 mmol/L is therefore proposed by the authors as an effective solution to decrease the probability of hypermagnesemia and its potential toxicity.¹⁴³ However, the clinical benefit of preventing hypermagnesemia by adjusting dialysate Mg²⁺ concentration in CKD patients is arguable as a negative Mg²⁺ balance increases cardiovascular risk potentially through calcification in this population, as discussed elsewhere in this review.

The promising effects of a Ca²⁺-acetate and Mg²⁺-carbonate binder compared with sevelamer on aortic medial calcification were demonstrated in uremic rats: Ca²⁺-acetate and Mg²⁺-carbonate prevented an increasing serum PTH and aortic calcium content more effectively.¹⁴⁴ Prevention of hyperphosphatemia and medial expression of osteogenic proteins such as BMP-2 and SRY-box 9 (sex-determining region Y box 9) in the media were achieved equally by both Ca²⁺-acetate and Mg²⁺-carbonate and sevelamer.¹⁴⁴ In hemodialysis patients, the use of a Ca²⁺-carbonate/Mg²⁺-carbonate combination correlated with reduced coronary artery calcification in a small clinical pilot study in 2009.¹⁴⁵ Although the size and design of the study are insufficient to admit clinical use, this study served as an indication that Mg²⁺ is an interesting novel and cost-effective treatment option. In addition to the Pi-binding effects of Mg²⁺ in the intestine, the concomitant increase in serum Mg²⁺ concentration may be protective for vascular calcification.¹⁴² Interestingly, the use of sevelamer itself has recently been found to be associated with increased serum Mg²⁺ concentrations.¹⁴⁶ The authors suggest that the beneficial effects of sevelamer on reduced inflammation, inhibition of vascular calcification, and decreased mortality might be partially explained by the higher serum Mg²⁺ concentrations. Follow-up studies should determine whether direct use of Mg²⁺-based Pi binders would be a more efficient treatment option in CKD patients.

In calcified vessels, hydroxyapatite (Ca₁₀(PO₄)₆(OH)₂) is the most abundant type of crystal.¹⁴⁷ Reduction or delay of hydroxyapatite formation by magnesium has been proposed as a mechanism to halt the calcification process. Mg²⁺ reduces ACP formation and maturation toward hydroxyapatite.^148–150 In aqueous solutions, Mg²⁺ delayed hydroxyapatite maturation with 20 hours, which was determined by the degree of crystallinity.¹⁵¹ Crystallization of ACP was prevented when the Mg²⁺/Ca²⁺ molar ratio exceeded 0.2 resulting simultaneously in reduced solubility of the crystal.¹⁵² Mechanistically, the stabilizing effect of Mg²⁺ on ACP has been attributed to the capacity of Mg²⁺ to form stronger complexes with Pi than Ca²⁺.¹⁵²

An alternative mechanism is that Mg²⁺ stabilizes extracellular ATP, which is otherwise hydrolyzed at the ACP surface enabling hydroxyapatite formation.¹⁵³ Mg²⁺ shields the ACP surface from ATP, thereby preventing its breakdown. Although the effect of Mg²⁺ on ATP has often been neglected, the role of extracellular ATP in vascular calcification has been studied because its hydrolysis is necessary for pyrophosphate synthesis, which is a direct inhibitor of hydroxyapatite formation.^154,155 As noted elsewhere in this review, Mg²⁺ protected against vascular calcification in Abcc6^−/− mice.⁹⁴ In this model of pseudoxanthoma elasticum, hepatic ABCC6 (ATP-binding cassette subfamily C member 6)–dependent–mediated cellular ATP secretion has been identified as the principal source of circulating pyrophosphate.¹⁵⁶ Pyrophosphate levels are 2.5-fold reduced in pseudoxanthoma patients where ABCC6 is dysfunctional, explaining the underlying mechanism in related mineralization disorders.¹⁵⁶

The stabilizing effects of Mg²⁺ on ACP nucleation and hydroxyapatite maturation in clinical setting have often been proposed in literature. However, this hypothesis has been poorly addressed in models of vascular calcification. Pasch et al¹³³ linked Mg²⁺ status to calcification propensity of hemodialysis patients, which is based on the intrinsic capacity of the serum to inhibit the maturation of primary CPP to secondary CPP and found that Mg²⁺ effectively delayed CPP maturation. Of note, secondary CPPs have been shown to induce calcification in vitro.¹³¹

It is often proposed that Mg²⁺ favors the formation of Mg²⁺-containing whitlockite (Ca₉Mg(HPO₄)(PO₄)₆) crystals rather than hydroxyapatite.¹⁵⁷ Whitlockite is smaller, more soluble, and less inflammatory compared with apatite and is only formed when Mg²⁺/Ca²⁺ ratios increase.^157–159 Formation of whitlockite after an increased serum Mg²⁺ concentration may therefore be a mechanism by which Mg²⁺ retards vascular calcification progression. However, Mg²⁺ supplementation to calcifying human VSMCs neither altered cellular apatite architecture nor resulted in the presence of whitlockite.¹⁶⁰ In addition, analysis of iliac arteries of dialysis patients showed the presence of both hydroxyapatite and whitlockite in calcified areas, colocalizing with calcification inhibitors.¹⁶¹ These findings combined suggest that preventive mechanisms of Mg²⁺ likely involve pathways alternative to the formation of whitlockite.

Multiple studies report that Mg²⁺ supplementation prevents the transcriptional changes in VSMC transdifferentiation and apoptosis, thereby halting the calcification process in both in vitro and ex vivo models of vascular calcification.^162–164 Mg²⁺ supplementation effectively counteracts expression of osteogenic transcription factors (BMP-2, RUNX2, Msh homeobox 2, SRY-box 9), bone proteins, and genes associated with matrix mineralization (osteocalcin and alkaline phosphatase).^162,165,166 Simultaneously, it was observed that Mg²⁺ prevents the loss of calcification inhibitors (BMP-7, MGP, and osteopontin) that protect against osteogenic conversion. These examples illustrate that Mg²⁺ is actively involved in the prevention of VSMC transdifferentiation to an osteogenic phenotype. However, whether Mg²⁺ directly modulates osteogenic gene expression remains under debate.

Because osteogenic gene expression is a convenient readout for vascular calcification in VSMCs, it has been widely exploited in in vitro studies. Given that inhibition of vascular calcification on any level may delay or even abrogate VSMC transdifferentiation, using osteogenic gene expression as readout is prone to misinterpretation of the mechanisms involved. VSMC calcification is often initiated by Pi- and Ca²⁺-enriched media and adding Mg²⁺ to calcifying VSMCs may have both extracellular and intracellular effects. However, when effective all will result in reduced VSMC transdifferentiation, calcification, and thus in lower osteogenic gene expression. Although this is poorly supported by direct evidence, the experimental bias of measuring osteogenic gene expression has resulted in the predominant hypothesis that intracellular Mg²⁺ reduces vascular calcification, overlooking potential extracellular effects.

The only studies convincingly supporting an intracellular role of Mg²⁺ are the ones that target Mg²⁺ channels. In VSMCs, Mg²⁺ homeostasis is mainly maintained by TRPM7 cation channels, which have been shown to be downregulated in calcification conditions.^165,167 Reduced TRPM7 activity using nonselective inhibitor 2-APB (aminoethoxydiphenyl borate) or a specific siRNA resulted in progressive VSMC transdifferentiation, illustrating a crucial role for intracellular Mg²⁺ in this context.^165,166 Furthermore, angiotensin-2 supplementation prevented osteoinductive expression and calcification in VSMCs by increasing Mg²⁺ influx. This effect was abrogated by blocking Mg²⁺ channel TRMP7 using 2-APB.¹⁶⁸

Several mechanisms have been proposed by which increased intracellular Mg²⁺ concentrations facilitated by TRPM7 activity could prevent osteoinductive gene expression. First, Mg²⁺ effectively abolished Pi-induced Wnt/β-catenin signaling, which is involved in osteoblast maturation and exercises its osteoinductive effects through increasing RUNX2 expression.^169,170 Second, Mg²⁺ has been implicated in the regulation of miRNAs involved in vascular homeostasis, a variety of which were recently found to be compromised in CKD.^171,172 Mg²⁺ successfully abrogated and even improved deteriorated expression profiles of microRNA-30b, microRNA-133a, and microRNA-223 that regulate RUNX2, Smad1, and osterix expression in calcifying VSMCs.¹⁷³ Third, Mg²⁺ is implicated in the modulation of VSMC calcium handling and the activation of the Ca²⁺-sensing receptor (CaSR) important for MGP function, which will be discussed below.

To identify additional mechanisms by which Mg²⁺ prevents calcification, it is relevant to learn from other calcification models. For instance, Mg²⁺ prevented SaOS-2 differentiation into mature osteoblasts in high concentrations (5 mmol/L), as reflected by matrix mineralization and alkaline phosphatase activity.¹⁷⁴ Importantly, however, these results were not reproducible in normal human osteoblasts. Furthermore, in tendon-derived stem cells, Mg²⁺ prevented matrix mineralization, a process that highly resembles that of VSMCs.¹⁷⁵ The authors proposed a role for Mg²⁺ in mitochondrial export of Ca²⁺ and Pi by the inhibition of mitochondrial transition pores, preventing transmembrane depolarization and matrix mineralization. However, application of these findings to VSMC calcification has not been evaluated to date.

The studies targeting TRPM7 support an intracellular effect of Mg²⁺ and reject the hypothesis that Mg²⁺-dependent regulation of calcification genes is only secondary to extracellular Pi binding and ACP stabilization. However, there is a lack of data on intracellular Mg²⁺ concentrations limiting conclusive confirmation on an active role for Mg²⁺ in this context. Additional studies measuring intracellular Mg²⁺ concentrations are necessary, but are hampered by the poor availability of selective fluorescent Mg²⁺ probes.

Excessive intracellular Ca²⁺ causes VSMC death and subsequent release of apoptotic bodies, which contribute to matrix calcification by providing ACP nucleation sites.^135,176,177 As a natural Ca²⁺ channel antagonist, Mg²⁺ has the capacity to block Ca²⁺ channels in VSMCs and prevent Ca²⁺ overload.^97,178 As a consequence of Ca²⁺ channel blocking, Mg²⁺ has excellent vasodilatory properties, which in arterioles and venules is already effective at 0.01 to 0.1 mmol/L concentrations and reduces myogenic tone.^97,179 Therefore, a role for Mg²⁺ in preventing intracellular Ca²⁺ bursts, and subsequent apoptosis has been identified as a potential mechanism of action in preventing VSMC calcification.¹⁰

In VSMCs, Ca²⁺ influx could be regulated by a sensing mechanism. The CaSR is expressed in the parathyroid and the kidney, and there are indications that VSMCs also express functional CaSR.¹⁸⁰ This receptor plays an important role in mineral-bone homeostasis by regulating PTH secretion. In addition to Ca²⁺ channel blocking, Mg²⁺ has been implicated in CaSR activation, possibly functioning as calcimimetic and indirect gatekeeper of Ca²⁺ influx.¹⁸¹ In contrast to Ca²⁺, Mg²⁺ acts as a partial agonist and activates the CaSR 2 to 3× less potently.^181–183

Systemically, lower PTH after CaSR activation in the parathyroid results in decreased bone turnover and intestinal Ca²⁺ uptake, but promotes renal Pi reabsorption. In dialysis patients, higher Mg²⁺ concentrations indeed correlate with decreased PTH levels.¹⁸⁴ Although the presence and function of CaSR in VSMCs remain uncertain, vascular calcification has been associated with loss of functional CaSR and MGP in VSMCs.^185–187 In VSMCs, treatment with calcimimetics resulted in the activation of the CaSR, which led to reduced mineralization.¹⁸⁰ In aortas of uremic rats and in bovine VSMCs, the calcimimetic AMG641 decreased medial calcification and increased expression of MGP.¹⁸⁸ Recently, the first in vitro and in vivo evidence suggested that Mg²⁺ supplementation in VSMCs resulted in reduced Pi- and hydroxyapatite-induced calcification through restoring CaSR mRNA and protein levels.¹⁸⁹ However, this study did not examine parameters related to mineral-bone metabolism in response to Mg²⁺ treatment in the in vivo part of their study. Therefore, the role of Mg²⁺ in the regulation of hormones and receptors involved in CKD-mineral bone disorder in its protection against vascular calcification remain to be determined.

Final conclusions about the molecular effects of Mg²⁺ are seriously hampered by the basic experimental setup of many in vitro studies that suffice with simple Mg²⁺ supplementation to calcification medium. This setup does not distinguish between passive chemical and active cell-mediated mechanisms. However, because cellular entrance of Mg²⁺ via TRPM7 has been shown to be necessary for at least some of its protective effects, an active mechanism preventing VSMC transdifferentiation is likely. This review identified a substantial knowledge gap of the role of intracellular Mg²⁺, as the molecular targets linking Mg²⁺ with osteogenic gene expression are unknown. In addition, the effect of Mg²⁺ supplementation on intracellular VSMC Mg²⁺ concentration has never been studied and urgently requires attention. Basic studies toward intracellular Mg²⁺ homeostasis and the molecular players that regulate Mg²⁺ concentrations in VSMCs are lacking and are essential to drive further advances in this field. Several of the mechanisms that have been repeatedly suggested have never been thoroughly studied in the context of vascular calcification, including the relevance of Mg²⁺ on cellular Ca²⁺ fluxes, the role of the CaSR in VSMCs and in particular the chemical impact of Mg²⁺ on ACP maturation. Furthermore, this review highlights the potential experimental bias of measuring osteogenic gene expression as effective inhibition of mineralization by Mg²⁺ through both extracellular and intracellular pathways will all result in reduced VSMC transdifferentiation. Therefore, an additional challenge that the field now faces lies in determining the relative contribution of each effect to the prevention of vascular calcification.

In the general population, Mg²⁺ is inversely associated with cardiovascular outcome. Results of these studies strongly reinforce the hypothesis that the current clinical reference ranges (0.7–1.1 mmol/L) for serum Mg²⁺ should be reconsidered, as concentrations of <0.8 mmol/L are associated with increased risk for cardiovascular disease and mortality (Table 1).

In CKD population, the pronounced effects of Mg²⁺ in experimental models of vascular calcifications drive the hypothesis that Mg²⁺ protects against mortality in CKD through the prevention of vascular calcification. However, the clinical role of Mg²⁺ in CKD patients has only been studied in observational cohorts, which focus mostly on total cardiovascular risk (Table 2). The effects of Mg²⁺ supplementation on cardiovascular outcome aside from arrhythmia and preeclampsia have been poorly assessed. Currently, randomized controlled clinical trials using Mg²⁺ supplementation as treatment for vascular calcification are in progress, and their results are eagerly awaited. These large-scale clinical trials will determine the translational value of the many experimental model systems that show a preventive effect of Mg²⁺ on vascular calcification. Nevertheless, further elucidation of the molecular mechanisms may contribute to additional targeted therapeutic options improving Mg²⁺ homeostasis in CKD patients.