Human skeletal muscles contain approximately 35% of total human body magnesium, which is an essential cofactor in a number of cell reactions. Magnesium ions influence the equilibria of many reactions involved in cellular bioenergetics by interacting with phosphorylated molecules and interfere with the kinetics of ion transport across plasma membranes 1. There is considerable evidence that Mg²⁺ is actively transported and regulated, although the mechanisms are still largely unknown 2. In skeletal muscle variations of cytosolic pH, phosphocreatine (PCr) and inorganic phosphate (Pi) concentrations influence the complex multi-equilibrium system of the molecular species binding magnesium ions. As a consequence [Mg²⁺] changes considerably in different metabolic conditions such as rest, exercise and recovery, showing an increase matched by a decrease of intracellular pH during exercise and recovery 3.

We assessed the cytosolic pH and the [Mg²⁺] by ³¹P MRS at rest, during exercise and post-exercise recovery in the calf muscle of two patients with McArdle's disease with muscle glycogen phosphorylase deficiency (McArdle), and two brothers affected by Tarui's disease with muscle phosphofructokinase deficiency (PFK).

These two type of glycogenosis, being characterized by almost absent activity of enzymes involved in glycogenolysis (McArdle) and glycolysis (PFK) pathways, show in general limited/absent production of intracellular lactic acid, depending on the degree of enzyme deficit 45. As consequence, patients with McArdle's and Tarui's disease, typically show a decrease or a lack of intracellular acidification during muscle exercise when studied by ³¹P MRS 67. We used these diseases as natural experimental models to study the pattern of free Mg²⁺ during exercise and recovery in the absence of intracellular acidification to understand to what extent homeostasis of intracellular free Mg²⁺ is linked to pH.

³¹P MRS spectra of human skeletal muscle typically show the peak signal of: phosphomonoesters (PME) which represents the phosphorylated monosaccharide intermediates of glycogenolysis, inorganic phosphate (Pi), phosphocreatine (PCr) and the three phosphate groups α, β, γ of ATP.

Figure 1 shows ³¹P MRS spectra acquired at the end of exercise in the calf muscle of McArdle and PFK patients compared to that of a control subject reaching a similar level of PCr depletion. End exercise spectra of both PFK patients show a marked increase in the PME peak.

Table 1 reports the rest and end-exercise values of cytosolic [Mg²⁺] and pH in patients' calf muscle, compared to the mean values obtained in a control group of ten subjects with a comparable end-exercise PCr depletion. At rest all patients showed a cytosolic pH not different from control values. Resting cytosolic [Mg²⁺] in PFK II patient was 0.45 mM, higher than the control value (0.32 + 0.04 mM) 3, while in PFK I and both McArdle's patients was normal. Both McArdle and PFK I patients reached a similar PCr depletion just above 50%, while PFK II patient stopped at a lower degree of PCr depletion. All patients displayed a lack of intracellular acidosis during exercise, showing pH values higher than controls at the end of exercise. Cytosolic free [Mg²⁺] at the end of exercise was lower in both PFK patients compared to control values. The variation of free [Mg²⁺] from rest to end-exercise (Δ[Mg²⁺]) was negative in both PFK patients and in McArdle II patient.

Figure 2 reports the patterns of cytosolic free [Mg²⁺] and pH obtained in patients during exercise (panel A and B) and recovery (panel C and D) compared with typical patterns from a healthy volunteer with comparable PCr depletion. During exercise and recovery the McArdle patients did not show any significant change (McArdle II) or small change (McArdle I) in free [Mg²⁺], while both PFK patients showed decreased free [Mg²⁺] during exercise. On the other hand, during recovery the pattern of free [Mg²⁺] was different in the two PFK patients, with PFK II showing a moderate increase during early recovery.

Figure 3 report the PME pattern of the two PFK patients during exercise and recovery. PFK I patient shows a slower rate of both PME accumulation and recovery compared to PFK II patient.

In the skeletal muscle variations of cytosolic pH, phosphocreatine and inorganic phosphate concentrations influence the complex multi-equilibrium system of the molecular species which bind magnesium ions. As a consequence free cytosolic [Mg²⁺] can change considerably in different metabolic conditions such as rest, exercise and recovery. It has been shown by ³¹P MRS that the increase of cytosolic free [Mg²⁺] occurring in skeletal muscle of healthy subjects during exercise and initial recovery is matched by a decrease in cytosolic pH, and the changes in cytosolic free [Mg²⁺] were mainly the result of the predominant effect of [H⁺]. 3. This result was attributed to mechanisms of binding competition existing between Mg²⁺ and H⁺ towards the molecules negatively charged present in the cell cytosol 3. However, the causal relationship between pH and [Mg²⁺] has not been proved yet, as it could be argued that muscular exercise per se elicits an increase in cytosolic free [Mg²⁺]. Therefore, we used patients with McArdle's and Tarui's disease as experimental models to study the pattern of [Mg²⁺] during exercise and recovery in the absence of intracellular acidification to understand to what extent homeostasis of intracellular free Mg²⁺ is linked to pH. Due to the rare nature of these disorders we were able to enrol just two patients for both diseases and therefore we had to deal with a small sample size. The results show that the increase in cytosolic [Mg²⁺] occurring in skeletal muscle during exercise is actually the consequence of the increase of H⁺ concentration and not of other mechanisms related to muscle contraction. In addition, we found that both PFK patients showed a reduction of [Mg²⁺] during exercise concomitant with the PME increase. The decrease of [Mg²⁺] also persisted during recovery in PFK I patient who displayed a slower PME recovery. The PME peak in the ³¹P MRS spectra corresponds to the phosphorylated monosaccharide intermediates of glycogenolysis. Therefore, due to the deficit of the phosphofructokinase activity in Tarui's disease, the PME accumulation shown by these patients is likely due to the increase of fructose- 6-phosphate, which represents an additional binding site for cytosolic Mg²⁺. As a consequence, we interpret the decrease of [Mg²⁺] concomitant with the PME increase as due to the binding of Mg²⁺ to fructose-6-phosphate. A previous study (6) reported that the abnormal PME accumulation of PFK patients during exercise was accompanied by a subnormal Pi accumulation. This finding was interpreted as a result of the incorporation of free Pi into phosphorylated glycolytic intermediates. However, both our patients did not show any Pi trap into PME, since we found that the sum of PCr and Pi was constant for the whole exercise duration, while the total phosphates signal increased proportionally to PME increase. Therefore, the [Mg²⁺] decrease found in PFK patients cannot be ascribed to a diminished Pi build-up.

Moreover, PFK II patient displayed a larger decrease of [Mg²⁺] from rest to end-exercise compared to PFK I patient. Accidentally, PFK II patient also had a smaller PCr breakdown, nevertheless, this cannot be the cause of the larger decrease of [Mg²⁺], as a lower [PCr] corresponds to lower calculated [Mg²⁺] (3).

Our results show that:

i) free [Mg²⁺] is strongly linked to pH in skeletal muscle homeostasis as previously suggested by a study in healthy volunteers 3, and by computer simulation on a chemical model mimicking muscle cell cytosol 13;

ii) the decrease of free [Mg²⁺] during exercise in both PFK patients suggests that [Mg²⁺] is influenced by the accumulation of fructose- 6-phosphate, an additional binding site for cytosolic Mg²⁺, as shown by the accumulation of the phosphomonoesters peak in the ³¹P MRS spectra of these patients.

iii) ³¹P MRS is a suitable tool for the in vivo assessment of free cytosolic [Mg²⁺] in human skeletal muscle during rest, exercise and recovery;

EM participated in the study design, performed the post-processing and statistical analysis, RL and CT participated in the study design and coordinated the data collection, AM participated in the study design, BB participated in the coordination of the study, SI participated in the study design, coordinated the study, and drafted the manuscript. All authors read and approved the final manuscript.