Dietary magnesium supplementation improves lifespan in a mouse model of progeria

Magnesium improves mitochondrial ATP synthesis in Lmna ^G609G/+ VSMCs

Previous studies have demonstrated mitochondrial dysfunction in progeria (Rivera‐Torres et al, 2013; Villa‐Bellosta et al, 2013; Aliper et al, 2015). Both oxygen consumption ratio (OCR) and ATP synthesis were significantly lower (by 41% and 39%, respectively) in Lmna ^G609G/+ VSMCs than in wild‐type cells (Fig 1A and B; Appendix Table S2). Moreover, Lmna ^G609G/+ VSMCs had significantly lower mitochondrial membrane potential (ΔΨ_m; 37%), assessed using the red‐to‐green ratio of JC‐10 fluorescence, than wild‐type cells (Fig 1C; Appendix Table S2). Notably, Lmna ^G609G/+ VSMCs showed significant higher ΔΨ_m (32%), OCR (37%), and mitochondrial ATP synthesis (31%) when incubated in a magnesium‐enriched medium.

Open in a separate window

Figure 1

Magnesium improves ATP synthesis in Lmna ^G609G/+ VSMCs

• A–C
  (A) Oxygen consumption ratio, (B) mitochondrial ATP synthesis, and (C) mitochondrial membrane potential (MP), in the indicated VSMC types. Results are presented as the mean ± SD of three independent experiments (four wells per experiment). One‐way ANOVA and Tukey's multiple comparisons post hoc test were used for statistical analysis. *P < 0.05; **P < 0.01; ***P < 0.001.

Source data are available online for this figure.

Magnesium ameliorates mitochondrial oxidative stress in Lmna ^G609G/+ VSMCs

Mitochondrial reactive oxygen species (ROS)‐mediated cell damage has been implicated in progeria (Richards et al, 2011; Sieprath et al, 2015; Kadoguchi et al, 2020). To evaluate the antioxidant properties of magnesium, ROS concentration was measured using the cell permeant reagent 2′,7′‐dichlorofluorescin diacetate (DCFDA), a fluorogenic dye that can be used to quantify hydroxyl, peroxyl, and other ROS activities within the cell. Lmna ^G609G/+ VSMCs showed significantly higher (3‐fold) ROS content than wild‐type cells (Fig EV2A; Appendix Table S3). In addition, the concentrations of two specific ROSs were also assessed. Mitochondrial superoxide (O₂ ⁻) and hydrogen peroxide (H₂O₂) were present in significantly higher (1.6‐fold and 2.3‐fold, respectively) concentrations in Lmna ^G609G/+ VSMCs than in wild‐type cells (Fig EV2B; Appendix Table S3). Notably, this overproduction of ROS was significantly reduced (by 69% for ROS, by 43% for H₂O₂, and by 29% for O₂ ⁻) in Lmna ^G609G/+ VSMCs incubated in magnesium‐enriched medium.

Open in a separate window

Figure EV2

Magnesium ameliorates oxidative stress in Lmna ^G609G/+ VSMCs

• A, B
  (A) Reactive oxygen species, and (B) superoxide and hydrogen peroxide radicals generated by the indicated VSMC types.
• C, D
  (C) Total antioxidant capacity and (D) total glutathione (which includes reduced ‐GSH- and oxidized ‐GSSG- glutathione), reduced glutathione (GHS), and the ratio of reduced and oxidized glutathione (GSSG) in the indicated cell types.
• E, F
  (E) Glutathione reductase (GR) activity, and (F) NADPH:NADP⁺ ratio and total NADPH (NADPH + NADP⁺) in the indicated VSMC types.
• G
  The boxed scheme shows the NADPH‐coupled glutathione redox systems, H⁺‐coupled ATP synthesis by mitochondrial ATP synthase, and H⁺‐coupled synthesis of NADPH by mitochondrial NADPH transhydrogenase (NNT). ΔΨ_m: mitochondrial membrane potential.

Data information: Results are presented as the mean ± SD of three independent experiments (four wells per experiment). One‐way ANOVA and Tukey's multiple comparisons post hoc test were used for statistical analysis. *P < 0.05; **P < 0.01; ***P < 0.001. Source data are available online for this figure.

The rate of ROS generation and the cellular defenses against ROS toxicity (which include enzymes, small molecules, and proteins) contribute to the overall level of oxidative stress. The total antioxidant capacity (TAC) can be considered a cumulative index of antioxidant status. To evaluate the overall cellular capacity to counteract ROS, TAC was assessed using a Cu²⁺ reduction assay. Lmna ^G609G/+ VSMCs showed significantly lower TAC (38%) than wild‐type VSMCs (Fig EV2C; Appendix Table S3). This reduction was significantly ameliorated (by 27%) in Lmna ^G609G/+ VSMCs incubated in magnesium‐enriched medium.

Reduced glutathione (GSH) is the major detoxifying redox buffer in cells and participates in the defense against ROS and the repair of mitochondrial oxidative damage, by being both a potent antioxidant itself and a substrate for antioxidant enzymes, including the glutathione reductase redox systems. Notably, total glutathione, which includes GSH and oxidized glutathione (GSSG), and glutathione reductase (GR) activity were significant lower in Lmna ^G609G/+ VSMCs than in wild‐type cells (Fig EV2D and E; Appendix Table S3). In addition, the ratio of reduced glutathione to oxidized glutathione (GSH:GSSG) was measured to assess the oxidative profile of the cells. Lmna ^G609G/+ VSMCs showed a significantly lower (51%) GSH:GSSG ratio than wild‐type VSMCs (Fig EV2D). Notably, this reduction was significantly ameliorated (by 51%) in treated Lmna ^G609G/+ VSMCs, although the GR activity and total glutathione concentration were similar in treated and untreated Lmna ^G609G/+ VSMCs.

GR uses reduced nicotinamide adenine dinucleotide phosphate (NADPH) to maintain the GSH redox state. Notably, although both types of VSMCs contained similar amounts of total nicotinamide adenine dinucleotide phosphate (NADPH and its oxidized form, NADP⁺), Lmna ^G609G/+ VSMCs had a significantly lower (48%) NADPH:NADP⁺ ratio than wild‐type cells (Fig EV2F and G; Appendix Table S3). However, the NADPH:NADP⁺ ratio was significantly improved (by 45%) by magnesium treatment of Lmna ^G609G/+ VSMCs.

Magnesium ameliorates acidification‐induced mitochondrial calcium overload

An increase in glycolysis that compensates for the loss of mitochondrial ATP synthesis has previously been shown in patient cells (Rivera‐Torres et al, 2013). Lmna ^G609G/+ VSMCs showed higher cytosolic ATP synthesis (1.8‐fold, Fig 2A; Appendix Table S4), lactate production (1.9‐fold, Fig 2B; Appendix Table S4), and extracellular acidification (2.1‐fold, Fig 2C; Appendix Table S4) than wild‐type cells. However, Lmna ^G609G/+ VSMCs incubated in a magnesium‐enriched medium showed significantly lower intracellular lactate concentration (19%) and extracellular acidification (14%). In contrast, cytosolic ATP synthesis was 21% higher in treated Lmna ^G609G/+ VSMCs than in untreated cells. This result is consistent with the notion that magnesium increases the activities of the ATP‐coupled glycolytic enzymes hexokinase, phosphofructokinase, phosphoglycerate kinase, and pyruvate kinase (Pilchova et al, 2017).

Open in a separate window

Figure 2

Magnesium reduces acidification‐induced mitochondrial calcium overload

• A
  Cytosolic ATP synthesis in the indicated VSMC types, measured by incorporation of phosphate‐32 (³²Pi) into ADP. [³²Pi]‐ATP was separated from ³²Pi using the molybdate method, as explained in the Materials and Methods section.
• B, C
  (B) Intracellular lactate concentration and (C) external acidification in the indicated VSMC types.
• D
  Calcium accumulation in mitochondria after 24 h of incubation in MEM containing 10 μCi/ml calcium‐45 (⁴⁵Ca²⁺) as a radiotracer.
• E
  Magnesium concentration in isolated mitochondria.
• F
  The boxed scheme describes the mitochondrial calcium overload hypothesis. Lactic acidosis forces the Na⁺/H⁺ exchanger (NHX) to import Na⁺, resulting in cytosolic Na⁺ overload. Subsequently, the Na⁺/Ca²⁺ exchanger (NCX) is forced into reverse mode to dispose of excess Na⁺, resulting in cytosolic calcium overload. This Ca²⁺ is then taken up by mitochondria, resulting in mitochondrial calcium overload. 2‐DG (2‐deoxyglucose; 50 mM) blocks glycolysis through competitive hexokinase inhibition, whereas oligomycin (10 μM) inhibits mitochondrial ATP synthase. G‐6-P: glucose‐6-phosphate.

Data information: Results are presented as the mean ± SD of three independent experiments (four wells per experiment). One‐way ANOVA and Tukey's multiple comparisons post hoc test were used for statistical analysis. *P < 0.05; **P < 0.01; ***P < 0.001. Source data are available online for this figure.

Intracellular acidification can lead to cytosolic and mitochondrial calcium overload, which depolarizes ΔΨ_m to limit ATP production and stimulates mitochondrial ROS generation and permeability transition (Brookes et al, 2004; Görlach et al, 2015; Santulli et al, 2015). Lmna ^G609G/+ VSMCs incubated with ⁴⁵Ca²⁺ as a radiotracer showed significantly higher (55%) mitochondrial calcium than wild‐type cells, which was significantly reduced (by 21%) in treated Lmna ^G609G/+ VSMCs (Fig 2D). In addition, Lmna ^G609G/+ VSMCs showed significantly lower (35%) mitochondrial magnesium than wild‐type cells, which was significantly increased (by 37%) in treated Lmna ^G609G/+ VSMCs (Fig 2E and F).

Magnesium prevents phosphate‐induced Lmna ^G609G/+ VSMC calcification

Previous studies show that calcification can occur without cellular activity, both in cultured devitalized aortas (Villa‐Bellosta, 2018) and in fixed smooth muscle cells (Villa‐Bellosta & Sorribas, 2009; Villa‐Bellosta et al, 2011). To determine the effect of magnesium on vascular calcification, treated and untreated Lmna ^G609G/+ VSMCs were incubated in 2 mM phosphate‐calcifying medium. Phosphate‐induced calcification was then assessed in both living and fixed cells (Fig 3A–G; Appendix Table S5). Untreated Lmna ^G609G/+ VSMCs showed 11‐fold higher (in live cells) and 17‐fold higher (in fixed cells) calcium deposition after 7 days of incubation in phosphate‐calcifying medium. However, treated living Lmna ^G609G/+ VSMCs showed significantly lower calcium accumulation (7.3‐fold), although the calcium content in fixed cells was similar in treated and untreated Lmna ^G609G/+ VSMCs (17‐fold). The addition of pyrophosphate or phosphonoformic acid (two known inhibitors of calcium phosphate crystal deposition) (Villa‐Bellosta & Sorribas, 2009) to the phosphate‐calcifying medium completely prevented calcium accumulation in both fixed/living and treated/untreated Lmna ^G609G/+ VSMCs. Notably, magnesium supplementation of the phosphate‐calcifying medium significantly reduced (by 38% in untreated and 51% in treated cells) calcium deposition in living cells. By contrast, magnesium supplementation did not reduce calcium deposition in either treated or untreated fixed VSMCs. Taken together, these results suggest that magnesium prevents calcium phosphate deposition by a cellular activity‐dependent mechanism, and not by direct binding to calcium phosphate crystals, preventing their formation and growth. Finally, the capacity to inhibit calcification (ΔCa²⁺) was calculated as the difference in calcium deposition in living versus fixed cells (Ca²⁺ in fixed cells − Ca²⁺ in living cells). The ΔCa²⁺ in treated cells was significantly higher than that in untreated cells (Fig 3G). Importantly, magnesium supplementation of the phosphate‐calcifying medium caused significant increases in ΔCa²⁺ in both treated and untreated cells.

Open in a separate window

Figure 3

Magnesium improves Lmna ^G609G/+ vascular smooth muscle cell calcification

• A
  Scheme showing the principle of the measurement. Lmna ^G609G/+ VSMCs were incubated in MEM (containing 0.8 mM magnesium; untreated) or in magnesium‐enriched MEM (containing 1.8 mM magnesium; treated) from passage 1 to passage 8. Then, cells were incubated overnight in MEM containing 0.1% FBS and some cells were fixed, as described in the Materials and Methods section. Then, cells were incubated in MEM (containing 0.1% FBS) with 1 or 2 mM phosphate, 1.8 mM magnesium (+Mg), 100 μM pyrophosphate (+PPi), or 500 μM phosphonoformic acid (+PFA). After 7 days of incubation, during which the media were replaced daily, the calcium content was measured as described in the Materials and Methods section.
• B
  Representative time‐course of 2 mM phosphate on calcification of Lmna ^G609G/+ VSMCs (up). Calcification was visualized with Alizarin red. Representative microscopic images (10x; scale bar: 100 μm) showing calcification of treated and untreated Lmna ^G609G/+ VSMCs (down).
• C–F
  Measures of calcium in treated living Lmna ^G609G/+ VSMCs (C), treated fixed Lmna ^G609G/+ VSMCs (D), untreated living Lmna ^G609G/+ VSMCs (E), and untreated fixed Lmna ^G609G/+ VSMCs (F).
• G
  The calcification inhibitory capacity was calculated as the difference in calcium deposition between living and fixed cells (ΔCa²⁺).

Data information: Results are presented as the mean ± SD of three independent experiments (four wells per condition). One‐way ANOVA and Tukey's multiple comparisons post hoc test were used for statistical analysis. **P < 0.01; ***P < 0.001. Source data are available online for this figure.

Magnesium prevents vascular calcification in HGPS mice

Clinically, plasma magnesium is usually measured despite the fact that less than 1% of magnesium exists extracellularly. Hence, plasma magnesium levels do not always accurately reflect total body magnesium stores. In fact, plasma magnesium levels can be normal despite depletion of the total body magnesium content. Notably, plasma magnesium levels were in the normal range in both wild‐type and Lmna ^G609G/+ mice, although they were significantly lower in 21‐ and 34‐week‐old Lmna ^G609G/+ mice than in wild‐type littermates (Table EV1).

To assess the effect of supplemental magnesium on Lmna ^G609G/+ mice, their drinking water was supplemented with MgCl₂. Thereafter, the consumption of food and water was measured in the mice between 8 and 34 weeks of age. The median food and water consumption of untreated and treated Lmna ^G609G/+ mice was similar (3.46 ± 0.77 versus 3.53 ± 0.72 g/day/mouse and 3.96 ± 0.62 versus 4.01 ± 0.73 ml/day/mouse, respectively). Therefore, the total magnesium intake by treated Lmna ^G609G/+ mice was significantly higher (4.6‐fold) than that by untreated Lmna ^G609G/+ mice (976.2 ± 261.7 versus 213.9 ± 45.0 mg/day/kg; Fig 4A, see Materials and Methods section). Notably, the plasma magnesium concentration was significantly higher in treated Lmna ^G609G/+ mice than in untreated Lmna ^G609G/+ mice (1.02 ± 0.06 versus 0.96 ± 0.05 mM; Fig 4B; Appendix Table S6). Finally, the total calcium content of aortas obtained from treated Lmna ^G609G/+ mice was significantly lower than that of aortas obtained from untreated Lmna ^G609G/+ mice (401.5 ± 77.7 versus 741.9 ± 101.6 μg/g aorta; Fig 4C; Appendix Table S6).

Open in a separate window

Figure 4

Oral magnesium treatment improves the longevity of Lmna ^G609G/+ mice

• A, B
  (A) Magnesium intake and (B) plasma magnesium concentration in 34‐week-old untreated and treated Lmna ^G609G/+ mice (n = 16).
• C
  Calcium content of aortas obtained from 34‐week-old wild‐type mice and untreated and treated Lmna ^G609G/+ mice (n = 20).
• D
  Body masses of 34-week‐old untreated and treated Lmna ^G609G/+ mice.
• E
  Kaplan–Meier graph for untreated and treated Lmna ^G609G/+ mice (n = 16).
• F
  Representative photographs of 40‐wk-old wild‐type, untreated, and treated Lmna ^G609G/+ mice.

Data information: Results are presented as the mean ± SD. Statistical analyses were performed using Student's t‐test (A, B, D), log‐rank test (E), or one‐way ANOVA and Tukey's multiple comparison post hoc test (C). **P < 0.05, **P < 0.01, ***P < 0.001. Source data are available online for this figure.

Magnesium improves the longevity of HGPS mice

The body mass of 34‐wk‐old treated Lmna ^G609G/+ mice was significantly higher (10%) than that of untreated Lmna ^G609G/+ mice (26.5 ± 1.1 versus 24.0 ± 2.4 g; Fig 4D). Moreover, the median survival time of treated Lmna ^G609G/+ mice was extended from 38.2 weeks to 42.9 weeks (Fig 4E and F).

Magnesium improves the antioxidant status of HGPS mice

Liver homogenates from Lmna ^G609G/+ mice had 42% lower TAC (Fig EV3A; Appendix Table S7), 38% lower total glutathione (Fig EV3B; Appendix Table S7), 48% lower GSH:GSSG ratio (Fig EV3C; Appendix Table S7), 55% lower NADPH:NAD⁺ ratio (Fig EV3D; Appendix Table S7), and 43% lower GR activity (Fig EV3E; Appendix Table S7) than wild‐type mice, implying the presence of an impairment in the NADPH‐coupled GR redox system. Notably, treated Lmna ^G609G/+ mice showed significant improvements in TAC (26%), GSH:GSSG ratio (52%), and NADPH:NAD ratio (45%) compared with untreated Lmna ^G609G/+ mice. However, total glutathione and GR activity were not significantly better in treated Lmna ^G609G/+ mice versus untreated Lmna ^G609G/+ mice.

Open in a separate window

Figure EV3

Magnesium improves the NADPH‐coupled glutathione redox status in Lmna ^G609G/+ mice

Liver homogenates were obtained from 34‐week‐old wild‐type, untreated, or treated Lmna ^G609G/+ mice.

• A–C
  (A) Total antioxidant capacity, (B) total glutathione (which includes reduced and oxidized glutathione), and (C) the ratio of reduced (GSH) and oxidized (GSSG) in the indicated experimental mouse groups.
• D, E
  (D) The NADPH:NADP⁺ ratio and (E) glutathione reductase (GR) activity.

Data information: Results are presented as mean ± SD (n = 16). One‐way ANOVA and Tukey's multiple comparisons post hoc test were used for statistical analysis. *P < 0.05; ***P < 0.001. Source data are available online for this figure.

Magnesium improves ATP synthesis in HGPS mice

Liver homogenates from untreated Lmna ^G609G/+ mice showed significantly lower (55%) intracellular ATP, which was 65% higher in treated mice (Fig 5A; Appendix Table S8). Moreover, isolated mitochondria showed 89% higher calcium content in untreated Lmna ^G609G/+ mice relative to wild‐type mice, but this was 34% lower in treated mice (Fig 5B; Appendix Table S8). In contrast, isolated mitochondria showed 33% lower magnesium content in untreated Lmna ^G609G/+ mice relative to wild‐type mice, but this was 35% higher in treated mice (Fig 5C; Appendix Table S8). Moreover, the activities of complexes I, III, IV, and V were significantly lower in untreated Lmna ^G609G/+ than wild‐type mice, but these defects were significantly ameliorated in treated Lmna ^G609G/+ mice (Fig 5D and E; Appendix Table S8), Notably, the subunits of these mitochondrial complexes are encoded by mitochondrial DNA.

Open in a separate window

Figure 5

Magnesium improves mitochondrial ATP synthesis in Lmna ^G609G/+ mice

• A
  ATP concentration in liver homogenates obtained from 34‐week-old wild‐type, untreated, or treated Lmna ^G609G/+ mice.
• B
  Mitochondrial calcium measured in liver mitochondria isolated from the indicated experimental mouse groups.
• C
  Magnesium concentration in isolated mitochondria.
• D
  Activities of the indicated mitochondrial complexes (I, II, III, IV, and V) in the absence or presence of rotenone (2 μM), 2‐thenoyltrifluoroacetone (TTFA; 1 mM), antimycin A (10 μM), potassium cyanide (KCN; 1 mM), or oligomycin (10 μM).
• E
  The boxed scheme shows the five mitochondrial complexes involved in the electron transport chain and their known inhibitors. Data information: Results are presented as mean ± SD (n = 16). One‐way ANOVA and Tukey's multiple comparisons post hoc test were used for statistical analysis. *P < 0.05; **P < 0.01; ***P < 0.001.

Source data are available online for this figure.

Magnesium improves mitochondrial ATP synthesis

ATP synthesis in isolated mitochondria was significantly lower in untreated Lmna ^G609G/+ mice than in wild‐type mice, in media containing either 0.1 mM magnesium (109.9 ± 29.6 versus 244.8 ± 78.8 nmol/min/mg protein, respectively) or 1 mM magnesium (226 ± 75.1 versus 503.7 ± 104.4 nmol/min/mg protein, respectively; Fig EV4A; Appendix Table S9). In both media, ATP synthesis was significantly higher (57% and 54%, respectively) for treated Lmna ^G609G/+ mice than for untreated Lmna ^G609G/+ mice. Notably, ATP synthesis in all the experimental groups was 2‐fold higher when mitochondria were assessed in incubation media containing 1 mM magnesium compared to incubation media containing 0.1 mM magnesium, which implies that ATP synthase is simulated by magnesium independently of the effect of treatment.

Open in a separate window

Figure EV4

Magnesium increases the activity of mitochondrial ATP synthase and extramitochondrial NADH oxidation

• A
  ATP synthase activity was measured in mitochondria isolated from the livers of 34-week‐old wild‐type, untreated, or treated Lmna ^G609G/+ mice in media containing 0.1 mM magnesium or 1 mM magnesium. The isolated mitochondria were also incubated in the absence or presence of oligomycin (10 μg/ml). The incubation media contain 5 mM succinate and 2 μM rotenone. The isolated mitochondria were also incubated in the absence or presence of oligomycin (10 μM).
• B
  Extramitochondrial NADH oxidation was measured in mitochondrial isolates from the livers of wild‐type, untreated, or treated Lmna ^G609G/+ mice in media containing rotenone (2 μM), antimycin A (10 μM), and NADH (1 mmo/L). The isolated mitochondria were also incubated in the absence or presence of MgCl₂ (5 mM), cytochrome C (10 μM), or KCN (1 mM).
• C, D
  The boxed scheme shows the five mitochondrial complexes involved in the electron transport chain and their known inhibitors used in the experiment.

Data information: Results are presented as mean ± SD (n = 16). One‐way ANOVA and Tukey's multiple comparisons post hoc test were used for statistical analysis. **P < 0.01; ***P < 0.001. Source data are available online for this figure.

Aorta isolation

Lmna ^G609G/+ mice were euthanized via carbon dioxide inhalation and thoracic aorta tissue was perfused with saline and removed according to previously published protocol (Villa‐Bellosta & Hamczyk, 2015).

Quantification of aortic calcification

To quantify the calcium content, mice aortas were dried, weighed, and treated with 0.6 M HCl 24 h. Then, calcium was quantified using a colorimetric QuantiChrom Calcium Assay Kit (BioAssay Systems, Hayward, CA).

Plasma magnesium levels

Blood was collected in heparin‐containing tubes and separated into plasma by centrifugation at 5,000 g for 10 min at 4°C. Mg was determined with the QuantiChrom Magnesium Assay Kit (BioAssay System) by the manufacturer's instructions.

VSMCs isolation and culture

VSMCs were obtained from three independents wild‐type or Lmna ^G609G/+ mice aorta pools (8–10 aortas per pool) by double digestion with collagenase method (Villa‐Bellosta & Hamczyk, 2015). VSMCs were grown in minimum essential medium Eagle (MEM) supplemented with 1 mM l‐glutamine, 100 IU/ml penicillin, 100 μg/ml streptomycin, and 10% fetal bovine serum at 37°C in a humidified atmosphere of 5% CO₂. All cell culture reagents were obtained from Invitrogen (Paisley, UK). MEM containing 0.8 mM Mg²⁺ was supplemented with 1 mM MgCl₂ (Sigma‐Aldrich; final concentration of 1.8 mM) to obtain magnesium‐enriched medium. After first trypsinization (passage 1), cells were incubated in MEM containing 0.8 or 1.8 mM Mg²⁺ to passage 8. Cells were grown considering 1:3 splitting during trypsinization. VSMCs were used directly from same passage (8–9), without a quiescent intermediate step. Figure 1B shows the growth rate after the 8^th trypsinization (passage 8).

Cell proliferation

The replicative incorporation of 5‐bromodeoxyuridine (BrdU) was measured using BrdU Cell Proliferation ELISA Kit (Abcam, ab126556) by the manufacturer's instructions.

Cell viability

The cell viability was assessed measuring the mitochondrial activity by using tetrazolium salt which in cleavage to formazan by cellular mitochondria dehydrogenase, (WST‐1 Assay Kit, ab65475, Abcam). Water‐soluble WST‐1 was added to each well, and the absorbance was measured using a scanning multiwell microplate according to the manufacturer's protocol.

β‐Gal activity

Senescence‐associated β‐galactosidase activity was measured in cell lysates by plate reader using a fluorescent probe (β‐gal Activity Assay Kit, BioVision), according to the manufacturer's protocols. Beta‐galactosidase hydrolyzes a non‐fluorescent substrate to generate a strong fluorescent product, which was measured (Ex/Em = 480/520) in two time points (0 and 60 min).

ATP quantification

ATP was measured by a coupled luciferin/luciferase reaction with an ATP Determination Kit (Invitrogen). Cells were treated with lysis buffer (50 Tris–HCl mM, 150 NaCl mM, 1% Triton X‐100 containing inhibitor cocktail, pH 7.4). VSMCs or liver lysates (intracellular ATP) and ATP standards were measured, according to the manufacturer's instructions (Villa‐Bellosta, 2019). For mitochondrial ATP measurement, VSMCs were previously incubated with or without oligomycin (10 μM) for 15 min. Mitochondrial ATP was calculated by the subtraction of intracellular ATP levels (with oligomycin) from total ATP (without oligomycin).

Mitochondrial membrane potential (ΔΨ_m) measurement

The ΔΨ_m was assessed by plate reader using a fluorescent probe (JC‐10; ab112134, Abcam) by following the manufacturer's instruction. When mitochondria are polarized electrically, JC‐10 forms J‐aggregates that emit orange‐red fluorescence. J‐monomers, indicating depolarized mitochondria, emit green fluorescence. The ΔΨ_m was calculated by a ratio of red/green fluorescence, indicating mitochondria depolarization with smaller ratio.

Oxygen consumption rate

Oxygen consumption rate (OCR) by VSMCs (5 × 10⁴ cells/well; Fig 2) or isolated mitochondria (Fig EV4) was measured on standard fluorescence plate reader using the Extracellular O₂ Consumption Assay (Abcam, ab197243) according to the manufacturer's protocol. OCR was measured using fluorescence microplate reader (excitation/emission wavelength of 380/650 nm).

Liver mitochondria were isolated by the standard method of homogenization followed by low (700 × g, 10 min) and high (12,000 × g, 15 min) centrifugation using the isolation medium composed of 250 mM sucrose, 2 mM EGTA, and 5 mM MOPS‐KOH (pH 7.4). Measurement buffer contained 60 mM sucrose, 0.5 mM EGTA, 5 mM MOPS‐KOH (pH 7.4), and 1 mM NADH.

ROS detection

Reactive oxygen species (ROS) were measured using the cell permeant reagent 2′,7′‐dichlorofluorescin diacetate (DCFDA) according to the manufacturer's protocol (Abcam, ab113851). DCFDA is deacetylated by cellular esterases and oxidized by ROS into a highly fluorescent compound which was measured using a fluorescence microplate reader (excitation/emission wavelength of 488/535 nm). Amplex^® Red reagent (10‐acetyl‐3,7‐dihydroxyphenoxazine) was used to detect hydrogen peroxide (H₂O₂). In the presence of peroxidase, the Amplex^® Red reagent reacts with H₂O₂ to produce the red‐fluorescent oxidation product, resorufin, which was measured using fluorescence microplate reader (excitation/emission wavelength of 430/590 nm), according to the manufacturer's protocol (A22188, Invitrogen). To detect production of mitochondrial superoxide radical in live cells, the Mitochondrial Superoxide Detection Kit (ab219943, Abcam) was used, according to the manufacturer's protocols. Superoxide was measured using fluorescence microplate reader (excitation/emission wavelength of 540/590 nm).

Total antioxidant capacity

Total antioxidant capacity (TAC) was determined using a commercially available assay kit (Abcam, ab65329) which utilizes the conversion of Cu²⁺ ions to Cu⁺ through endogenous protein and small molecule antioxidants, standardized to Trolox equivalents. VSMCs and liver were used for analysis of TAC according to the manufacturer's protocols. Colorimetric activity was measured by optimal density at 570 nm.

NADPH‐coupled glutathione redox system assay

Reduced/oxidized glutathione (GSH/GSSG) ratio, NADPH/NADP ratio, and glutathione reductase (GR) activity in both VSMCs and liver lysates were measured with commercials kit (Abcam, ab138881, ab176724, and ab83461, respectively), using a 96‐well plate reader, according to the manufacturer's protocols. Liver samples were measured in triplicate on the same plate, and fluorescence/colorimetric values were normalized to micrograms of protein loaded in the assay per sample. Protein was measured with the Pierce BCA Protein Assay Kit (Thermo Scientific, Rockford, USA), according to the manufacturer's protocols.

GSH and total glutathione were determined by changes in fluorescence intensity (excitation/emission wavelength of 490/520 nm). 20 mg of liver tissue or VSMCs (5 × 10⁶) were homogenized/lysed in 400/100 μl cold lysis buffer, respectively. Homogenates/lysates were centrifuged at top speed for 15 min at 4°C. Supernatants were deproteinized using Deproteinizing Sample Preparation Kit‐TCA (Abcam, ab204708) according to the manufacturer's protocol.

NADPH and total nicotinamide adenine dinucleotide phosphate (NADPH⁺NADP⁺) were determined by changes in fluorescence intensity (excitation/emission wavelength of 540/590 nm). 20 mg of liver tissue or VSMCs (5 × 10⁶) were homogenized/lysed in 400/100 μl lysis buffer, respectively. Homogenates/lysates were centrifuged at 2,500/1,500 rpm, respectively, for 5 min at RT.

GR activity was measured by optimal density at 405 nm. 20 mg of liver tissue or 1 × 10⁶ VSMCs were homogenized/lysed in 200 μl cold assay buffer and centrifuged at 10,000 × g for 15 min at 4°C. Supernatant was pre‐treated to destroy GSH before the assay, according to the manufacturer's protocol.

Mitochondrial and cytosolic ATP synthesis

Liver mitochondria were isolated by the standard procedure of differential centrifugation using the isolation medium composed of 250 mM sucrose, 2 mM EGTA, and 5 mM MOPS‐KOH (pH 7.4). Mitochondria were washed and finely suspended in the medium composed of 120 mM KCl, 20 mM MOPS, and 0.5 mM EGTA (KME medium). Mitochondria protein was measured with the Pierce BCA Protein Assay Kit (Thermo Scientific, Rockford, USA), according to the manufacturer's protocols. Mitochondria (1 mg of protein/ml) were incubated in KME medium containing 5 mM succinate, 2 μM rotenone, 10 mM NaCl, 800 pmol of A23187/mg of protein, and 5 mM phosphate (10 μCi/ml of ³²Pi as radiotracer) at 37°C. After 5 min, 1.2 mM ADP was added, and the reaction was stopped 30 s later by addition of 200 μl of 30% (w/v) cold trichloroacetic acid. Orthophosphate (³²Pi) was separated from ATP ([γ³²P]ATP) by molybdate method, as previously described (Villa‐Bellosta, 2018, 2019). Briefly, 20 μl of sample was mixed with 400 μL of ammonium molybdate (to bind the orthophosphate) and 0.75 M sulfuric acid. Samples were then extracted with 800 μl of isobutanol/petroleum ether (4:1) to separate the phosphomolybdate from pyrophosphate and ATP. Next, 400 μl of the aqueous phase containing ATP was removed and subjected to radioactivity counting (Tri‐Carb 2810TR, PerkinElmer).

For cytosolic ATP synthesis, VSMCs (1 × 10⁶) were incubated in MEM containing 10 μM oligomycin and 10 μCi/ml of phosphate‐32 (³²Pi) as radiotracer, at 37°C. The reaction was stopped by addition of 200 μl of 30% (w/v) cold trichloroacetic acid. Orthophosphate (³²Pi) was separated from ATP ([γ³²P]ATP) by molybdate method.

Glycolysis assay

Extracellular acidification and lactate were measured with fluorometric kits (Abcam, ab197244 and ab169557, respectively), using a 96‐well plate reader, according to the manufacturer's protocols. Extracellular acidification was determined in VSMCs (5 × 10⁵ cells/well) by changes in fluorescence intensity (excitation/emission wavelength of 380/615 nm), using a water‐soluble and cell‐impermeable pH‐sensitive reagent. L‐lactate was determined by changes in fluorescence intensity (excitation/emission wavelength of 535/587 nm). VSMCs (1 × 10⁶) were homogenized with 110 μl cold lactate assay buffer on ice and centrifuged at 14,000 g for 5 min. Supernatant was measured in duplicate on the same plate.

Mitochondrial calcium and magnesium

For calcium accumulation by mitochondria, VSMCs were incubated in MEM containing 10 μCi/ml calcium‐45 (⁴⁵Ca²⁺) as a radiotracer. After 24 h, VSMCs were washed five times in MEM. Mitochondria were isolated from VSMCs by method of homogenization followed by low‐ and high‐speed centrifugation at 4°C. The homogenate was centrifuged at 1,000 × g for 10 min. Mitochondria were sedimented at 3,500 × g for 15 min. The pellet contains the isolated mitochondria was washed and centrifuged at 12,000 × g for 15 min. The isolation medium contained 250 mM sucrose, 0.5 mM EGTA, and 5 mM MOPS‐KOH (pH 7.4). EGTA was excluded from washing medium. Mitochondria pellets was resuspended in liquid scintillation counting (UltraGold, 6013329, PerkinElmer) and subjected to radioactivity counting (Tri‐Carb 2810TR, PerkinElmer).

Liver mitochondria were isolated by the method of homogenization followed by low (1,000 × g, 10 min)‐ and high (3,500 × g, 15 min)‐speed centrifugation at 4°C. Mitochondria were washed and finely suspended in dH₂O. Mitochondria protein was measured with the Pierce BCA Protein Assay Kit (Thermo Scientific, Rockford, USA), according to the manufacturer's protocols. Calcium was measured using a colorimetric QuantiChrom Calcium Assay Kit (BioAssay System, Hayward, CA) according to the manufacturer's protocols.

Mitochondrial magnesium concentration was measured in isolated mitochondria using mag‐fura 2‐AM (Thermo Fisher) as described previously (Kolisek et al, 2003). Briefly, mitochondria were loaded with 5 μM mag‐fura 2‐AM for 40 min at 25°C and washed twice to remove excess dye. Magnesium concentration was determined by measuring the fluorescence of the probe‐loaded mitochondria with excitation at 340 and 380 nm, and emission at 510 nm. Mitochondrial magnesium was calculated from the 340/380 nm ratio according to the formula of Grynkiewicz et al (Grynkiewicz et al, 1985). The minimum (R _min) and maximum (R _max) ratios were obtained at the end of each experiment. R _max was obtained by the addition of SDS (10% w/v) and MgCl₂ (25 mM). R _min was detected by addition of EDTA (50 mM, pH 8).

Calcification assay and quantification

VSMCs were used from same passage (8 or 9), considering 1:3 splitting during trypsinization. Cells were grown to confluence and used after a quiescent intermediate step (overnight in culture media containing 0.1% fetal bovine serum). Calcification assays were performed on cells incubated for 7 days in MEM supplemented with 1 mM l‐glutamine, 100 IU/ml penicillin, 100 μg/ml streptomycin, 0.1% fetal bovine serum, and 2 mM phosphate (phosphate‐calcifying medium), as described previously (Villa‐Bellosta, 2018; Villa‐Bellosta et al, 2011, p.; Villa‐Bellosta & Hamczyk, 2015). Phosphate‐calcifying medium was replaced every day. To quantify the calcium content of VSMC, wells were treated with 0.6 M HCl overnight at 4°C and analyzed using a colorimetric QuantiChrom Calcium Assay Kit (BioAssay System, Hayward, CA). Phosphate, magnesium chloride, phosphonoformic acid, and pyrophosphate were obtained from Sigma‐Aldrich. Cells were fixed as described previously (Villa‐Bellosta & Sorribas, 2009).

Mitochondrial complex I‐V activities

Mitochondria were isolated from mouse liver tissues using Mitochondria Isolation Kit for tissue (Cayman Chemical, #701010). The activities of electron transport chain complexes I, II, III, IV, and V were determined using complex I activity assay kit (Cayman Chemical, #700930), complex II activity assay kit (Cayman Chemical, #700940), complex II/III activity assay kit (Cayman Chemical, #700950), complex IV activity assay kit (Cayman Chemical, #700990), and complex V activity assay kit (Cayman Chemical, #701000), respectively, following the manufacturer's instruction. Rotenone, 2‐thenoyltrifluoroacetone (TTFA), antimycin A, potassium cyanide (KCN), and oligomycin were obtained from Sigma‐Aldrich.

Magnesium treatment

To assess the effect of supplemental magnesium on Lmna ^G609G/+ mice, their drinking water (containing 39 mg/l magnesium) was supplemented with 15 g/l MgCl₂. Thereafter, the consumption of food (containing 0.17% magnesium) and that of both untreated and treated water was measured in the mice between 8 and 34 weeks of age. Magnesium intake was measured twice a week. Average daily food and water intake was calculated per day and by weight of the mouse in each cage.

We thank Daniel Azpiazu, Ana Lobo, Ana de la Calle, Veronica, and Jose Luis for excellent technical assistance. We also thank the group of Prof. Carlos López‐Otín for the generous gift of the HGPS mouse model and the personnel of FIIS‐FJD Animal Facility for animals’ production and maintenance. This study was supported by grants from Progeria Research Foundation (PRF‐2016‐68) from USA and Spanish Ministerio de Economía y Competitividad (SAF‐2014‐60669‐JIN).