Magnesium Research (1994) 7, 1, 11-16
Summary: The effects of MgCl2 and MgSO4 are different on the total transfer through the human amniotic membrane: MgCl2 at low concentration (1 mM) decreases the total conductance Gt and increases it at high concentration (4 mM) on the fetal side (FS) and on the maternal side (MS), while MgSO4 has no effect on the MS and increases Gt on the FS. Moreover, whatever the concentration, MgCl2 increases the flux ratio F1/F2 while MgSO4 decreases it to reach a value near to 1. Gt is the sum of various components: three paracellular components (Gp) and nine cellular components (Gc constituted from channels, exchangers, antiporters and cotransporters). All components of Gt on the two faces, are decreased by 1 mM MgCl2 and increased at 4 mM. MgCl2 also has an effect on all ionic exchangers across the membrane. In contrast, on the MS, MgSO4 (1mM) decreases GpNa, increases GpK and the antiport Na/H component and has no effect on any of the other components, while at 4 mM, MgSO4 has no effect. On the FS, MgSO4 (1 mM) increases GpNa and GpK, but does not modify the other components. At 4 mM, the effect is the same, except for an increase of GpCl. These data show the importance of the anion-cation association in the ionic exchanges through a membrane: MgCl2 and MgSO4 have a different action --MgCl2 interacts with all the exchangers, while the effect of MgSO4 is limited to paracellular components without interaction with cellular components excepted the antiport Na/H.
Key words: Amniotic membrane, conductance components, exchangers, MgCl2, MgSO4.
Among numerous magnesium salts, MgCl2 and MgSO4 are the ones most commonly used in magnesium therapy. Curiously, the most widely used preparations are those containing magnesium sulphate. However, among the soluble salts of magnesium, this salt has the least advantageous pharmacological properties. Its absorption, cellular penetration, membrane effects and anti-hypoxic effects are poor1-3. Comparative studies between MgCl2 and MgSO4 indicate classically a better absorption for MgCl2 than for MgSO44-8 and a better retention for MgCl23,9,10. Moreover, sulphate infusion leads to a considerable increase in the urinary excretion of magnesium11, calcium and potassium 12,13. The membranous effects of magnesium salts may e evaluated by the study of monovalent fluxes. It has already been shown that the effects of MgCl2 and MgSO4 are different on the total transfer through the human amniotic membrane14-18: MgCl2 at low concentration (1 mM) decreases the total conductance, Gt, and increases it at high concentration (4 mM) on the fetal side (FS) and on the maternal side (MS), while MgSO4 has no effect on the MS and increases Gt on the FS. Moreover, whatever the concentration, MgCl2 increases the fluxes ratio F1/F2, while MgSO4 decreases it to reach a value near to 1. Gt is the sum of various components: three paracellular components (Gp) and nine cellular components (Gc, constituted from ionic channels, exchangers, antiporters and cotransporters), and the aim of this work was to study the comparative effects of MgCl2 and MgSO4 on all components of amniotic transfer to identify the targets of the deleterious action of MgSO4.
Strips of human amnion, isolated from the amniotic sac, were obtained after 12 normal full term deliveries. The amnion was immersed into Hanks' solution (pH 7.4 and 37 ± 1° C). A circular area of the amnion was sampled and placed between two acrylic chambers containing Hanks' solution according to the Bara and Guiet-Bara device19.
The cellular conductance (Gc) in the maternal-to-fetal and in the fetal-to-maternal directions was measured by injecting a constant current square wave pulse through a double-barrelled microelectrode and analysing the recorded transient voltage on an oscilloscope. The transamniotic ionic conductance, Gt, was measured by observing the transamniotic potential difference when a direct current was passed across the whole tissue. The transamniotic potential difference was recorded with two agar-agar salt bridges placed 1.3-1.5 mm from each side of the tissue, while an electrical current was passed across the tissue by means of Ag/AgCl electrodes and agar-agar salts and measured on a Schlumberger electrometer. The paracellular conductance Gp = Gt - Gc.
Various metabolic inhibitors, which were used to observe the different components of the total transamniotic conductance, were20:
(1) Cellular conductance - Ouabain (0.1) mM); Na/K-ATPase (Ge1); amiloride (0.5 mM); Na+ channels (GlNa); bumetanide (0.1 mM); Na-K-2Cl cotransport (Ge3); barium (1 mM); K+ channels (G1K); diisothiocyanostilbene disulphonic acid (300 µM); Cl- channels (GlCl); acetamido-isothio-cyanostilbene-disulphonic acid (300 µM) Cl-HCO3 antiport (Ge4); manganese (1 mM); Na-Mg exchanger (Ge5); dinitrophenol (20 mM): coupling factor (Gcp).
(2) Paracellular conductance - Triaminopyrimidinium (10 mM); GpNa; protamines (100 mg/litre); GpK and GpCl.
The composition of the Hanks' solution used in the experiments was (mmol/litre): NaCl 150, KCl 6, CaCl2 1, MgCl2 0.5 MgSO4 0.5, Na2HPO4-KH2PO4-NaHCO3 1, glucose 5.5. The magnesium (MgCl2 or MgSO4) was added on the maternal or fetal sides of the amnion at two concentrations: 1 mM and 4 mM.
The results are expressed as means ±SD. Statistical comparisons were carried out by conventional paired data analysis.
Effects of MgCl2
On the maternal side (Fig. 1), the addition of 1 mM MgCl2 induced a decrease (P < 0.01) in all the paracellular components (GpK, GpNa, GpCl); at 4 mM, the three components were increased (P < 0.01). On the fetal side (Fig. 2), the addition of 1 mM MgCl2 decreased (P < 0.01) GpK, GpNa and GpCl; at 4 mM, the three components were all increased (P < 0.01).
On the maternal side (Figs. 3 and 4), the addition of 1 mM MgCl2 induced a decrease (P < 0.01) in all the cellular components; at 4 mM, all components were increased (P < 0.01). On the fetal side (Figs. 5 and 6), the addition of 1 mM MgCl2 decreased (P < 0.01) all the cellular components significantly; at 4 mM, there was an increase in all the components (P < 0.01).
Effects of MgSO4
On the maternal side (Fig. 1), the addition of 1 mM MgSO4 induced an increase (P < 0.01) in GpK, and a significant decrease in GpNa, but had no effect (P = 0.25) on GpCl; at 4 mM, GpK, GpNa and GpCl remained constant (P = 0.3). On the fetal side (Fig. 2), the addition of 1 MgSO4 induced an increase (P <0.01) in GpK and GpNa and had no action (P = 0.20) on GpCl; at 4 mM, the three components were all increased (P < 0.01).
On the maternal side (Figs. 3 and 4), the addition of 1 or 4 mM MgSO4 had no significant effect (P=0.25) on any of the cellular components except the antiport Na/H component (Ge2) which was increased (P < 0.01). On the fetal side (Figs. 5 and 6), the addition of 1 or 4 mM MgSO4 had no significant effect (P = 0.30) on any of the cellular components.
The biological effects of Cl- and SO4 2- are generally opposite. For example, following infusion of hypertonic sodium sulphate, increases in serum Na concentration and decreases in serum chloride, potassium, phosphate, calcium and magnesium were regularly observed12. In general, studies of the membrane effects of MgCl2 and MgSO4 have shown that the sulphate anion has unfavourable effects in comparison with other ions used in magnesium therapy. Our previous studies14-18 on the membrane effects of MgCl2 and MgSO4 on monovalent ion fluxes have shown that MgSO4 has deleterious actions in comparison with other magnesium salts, particularly MgCl2. These data may be compared to the effects of waters containing sulphuric acid or sulphate ions on cardiovascular diseases--the presence of SO4 in water may act as a cardiovascular risk factor21,22.
In the present study, we have shown that on the maternal and fetal sides MgCl2 decreases all components at 1 mM and increases them at 4 mM, according to the biphasic effect already observed on the total conductance Gt 23. In contrast, MgSO4 has a differential action on the two sides. On the MS, 1 mM MgSO4 decreases GpNa, increases GpK and the antiport Na/H, and has no effect on the other components, while at 4 mM, MgSO4 has no significant effect on any components except for an increase in the antiport Na/H exchange. On the FS, 1 mM MgSO4 increases GpNa and GpK and does not modify the other components. At 4 mM, the effect is the same except for an increase of GpCl. Also, the effects of MgSO4 are limited to the paracellular components and to the antiport Na/H.
MgSO4 does not act on the Na/K-ATPase component (Ge1). This result implies that the hypothesis of a direct contribution of MgSO4 to the membrane potential is inaccurate. On the other hand, MgCl2, which interacts with the Na/K-ATPase component, may participate directly in the formation of the membrane potential. Moreover, MgCl2, which increases the ionic channel conductances (G1Na, GlK, GlCl), regulates the viability of these channels and may decrease or increase the permeability of monovalent ions as a function of its concentration. This is not true for MgSO4. The principal effect of MgSO4 is on the paracellular conductances which are structurally situated in the intercellular space.
It is possible to explain the results obtained with MgCl2 and MgSO4 by their crystal structures24. MgCl2 forms a stable hexahydrate [Mg(H2O)6 Cl2] between -3.4 and +116° C; the crystals consist of dianions with the magnesium co-ordinated to the six water molecules as a complex, [Mg(H2O)6]2+, and two independent chloride anions, Cl-. MgSO4 forms a stable heptahydrate [Mg(H2O)7 SO4] at ambient temperature; the crystals again consist of dian-ions with the magnesium coordinated to six water molecules, while the seventh water molecule is associated with the sulphate anion: [Mg(H2O)6]2+ [SO4+H2O]2-. Therefore both salts are hexa-aqueous complexes of magnesium and the counter ions (Cl-,SO4 2-) do not form complexes with magnesium ions. Moreover, MgCl2 measures about 12 per cent magnesium, while MgSO4 measures about 10 per cent magnesium3.
These data confirm the important role of the anion and may explain the present results. The size of MgCl2 (a smaller molecule than MgSO4 should allow interactions with the various exchange components (paracellular and cellular), and the more hydrated MgSO4 molecule implicates bindings with external charges, that is, interactions with paracellular components in preference to cellular components. In the two crystals, magnesium has the same complex form ([Mg(H2O)6]2+). The different effects observed might also be attributed to the anions associated with magnesium.
This observation is important in clinical practice. For example, the hypocalcaemia observed during high dose magnesium therapy should not only be interpreted as an effect of magnesium overload on the calcium metabolism, but also in relation to the particular effects of the anion used on the calcium metabolism (SO4 increases the calcium urinary excretion 12,13.
MgCl2 and MgSO4 have differential effects on the ionic exchange through the human amniotic membrane (MgCl2 interacts with paracellular and cellular transfer; MgSO4 interacts with paracellular transfer only). These data show the importance of the anion-cation association. During magnesium therapy, it is important to consider the anion effect associated with magnesium.
1. Duhm, J., Deuticke, B. & Gerlach, E. (1968): Metabolism of 2,3-diphosphoglycerate and glycolysis in human red blood cells under the influence of dipyramidole and inorganic sulfur compounds. Biochim. Biophys. Acta 170, 452-454.
2. Morris, M.E. & Levy. G. (1983): Absorption of sulfate from orally administered magnesium sulfate in man. J. Toxicol. Clin. Toxicol. 20, 107-114.
3. Durlach, J. (1988): Magnesium in clinical practice, 360 pp. London, Paris: John Libbey.
4. Richard, A. & Hazard, R. (1943): Précis de Thérapeutique et de Pharmacologie. p. 3856. Paris: Masson.
5. Glénard, R. (1948): Le magnésium et la fibre lisse. In: Journées thérapeutiques de Paris 1947, pp. 313-320. Paris: Doin.
6. Classen, H.G.. Marquardt, P., Späth, M., Ebel, H. & Schumacher, K.A. (1973): Vergleichende tierexperimentelle Untersuchungen über die Resorption von Magnesium also Sulfate, Chlorid, Aspartat und Aspartat-Hydrochlorid aus dem Magen-Darm-Trakt. Arzneimittelforschung 23, 267-271.
7. Classen, H.G. (1990): Magnesium and placebo effects in human medicine. In: Metal ions in biological systems, vol. 26, eds. H. Sigel & A. Sigel, pp. 597-609. New York: Marcel Dekker.
8. Malagelada. J. R., Holtermuller, K.H., McCall, J.T. & Vay Liang, W.G. (1978): Pancreatic gallbladder and intestinal responses to intraluminal magnesium salts in man. Am. J. Dig. Dis. 23, 481-485.
9. Franke, H. (1934): Magnesium und Kohlehydratstoffwechsel. Arch. Exp. Pathol. Pharmacol. 174, 727-740.
10. Naito, Y. (1980): Effect of Mg salts on urinary Mg excretion in rats fed hypomagnesic diet. Igaku to Seibutsugaku 101, 173-176.
11. Martin, H.E., Mehl, J. & Wertman, M. (1952): Clinical studies of Mg metabolism. Med. Clin. North Am. 36, 1157-1171.
12. Wolf, A.V. & Ball, S.M. (1950): Effect of intravenous sodium sulfate on renal excretion in the dog. Am. J. Physiol. 160, 353-360.
13. Walser, M. & Browder, A.A. (1959): Ion association. III. The effect of sulfate infusion on calcium excretion. J. Clin. Invest. 38, 1404-1411.
14. Bara, M. Guiet-Bara, A. & Durlach, J. (1984): Comparative effects of MgCl2 and MgSO4 on monovalent cations transfer across isolated human amnion. Magnes. Bull. 6, 36-40.
15. Guiet-Bara, A., Bara, M. & Durlach, J. (1985): Cellular and shunt conductances of human isolated amnion. II. Comparative effects of MgCl2 and MgSO4: electrophysiological studies. Magnes. Bull. 7, 16-19.
16. Bara, M.. Guiet-Bara. A. & Durlach, J. (1988): Modification of human amniotic membrane stability after addition of magnesium salts. Magnes. Res. 1, 23-28.
17. Bara, M., Guiet-Bara, A. & Durlach, J. (1988): Analysis of magnesium membraneous effects: binding and screening. Magnes. Res. 1, 29-33.
18. Durlach, J., Durlach, V., Bara, M., Guiet-Bara, A. (1992): A new method of in vitro prescreening evaluation of several Mg salts. Meth. Find. Exp. Clin. Pharmacol. 14, 305-310.
19. Bara. M. & Guiet-Bara. A. (1981): Détermination des constantes électriques de la membrane des cellules épithéliales amniotiques humaines in vitro. C.R. Soc. Biol. 176, 749-754.
20. Bara, M., Guiet-Bara. A. & Durlach. J. (1990): Comparative study of effects of magnesium and taurine on electrical parameters of natural and artificial membranes. VII. Effects on cellular and paracellular ionic transfer through isolated human amnion. Magnes. Res. 3, 249-254.
21. Durlach, J., Bara, M. & Guiet-Bara, A. (1985): Magnesium level in drinking water and cardiovascular risk factor: a hypothesis. Magnesium 4, 5-15.
22. Kobayashi, J.A. (1957): Geographical relationship between the chemical nature of river water and death rate from apoplexy. Ber. Ohara Inst. Lansw. Biol. 11, 12-21.
23. Bara, M., Guiet-Bara, A. & Durlach, J. (1989): A qualitative theory of the screening-binding effects of magnesium salts on epithelial cell membrane: a new hypothesis. Magnes. Res. 2, 243-247.
24. Theophanides, T., Angiboust, J.F..Anastassopoulou. J. & Manfait. M. (1990): Possible role of water structure In biological magnesium systems. Magnes. Res. 3, 5-13.
All articles by Dr. Durlach are copyrighted, and permission is granted to Web users only to make single hard copies for personal use. Additional reprints should be obtained from the originating journals. Excerpts may be used by the media with attribution to Dr. Durlach.
This page was first uploaded to The Magnesium Web Site on April 15, 1996