Structural alterations of the vascular wall in magnesium‐deficient mice. A possible role of gelatinases A (MMP‐2) and B (MMP‐9)

ARTICLE

Auteur(s) : N. Pagès¹, B. Gogly², G. Godeau², S. Igondjo-Tchen², P. Maurois³, J. Durlach⁴, P. Bac³

¹Faculté de Pharmacie, Route du Rhin, Strasbourg, 76400 Illkirch, France ; ² Faculté de Chirurgie dentaire, Paris V, 92210 Montrouge, France ; ³Faculté de Pharmacie, Paris XI, Rue J.B. Clément. 92290 Châtenay Malabry, France ; ⁴SDRM, Université Pierre et Marie Curie, 75252 Paris, France

Introduction

Magnesium ions are known to protect the cardiovascular system by preventing both calcium accumulation and connective damages, whereas magnesium deficiency induces cardiovascular damages [1, 2]. The vascular lesions of magnesium deficiency are characterized in various animal species by a severe degradation of the extra cellular matrix (ECM) of the connective tissue i.e. by edema, hypertrophy and hyperplasia of the intima, by a thinning and a fragmentation of the internal elastica, by edema, necrosis and hyperplasia of the media. These lesions are accompanied by calcification [3-5].
Connective tissue constitutes a large portion of the body, contributing to organ shape and volume, playing a role in metabolic processes, influencing cell proliferation, differenciation and apoptosis, and serving as a repository for biologically active growth factor. It is separated from epithelia by the basement membrane (BM) and is composed of the stromal elements: ECM, blood and lymph vessels, and cellular components, including fibroblasts and macrophages. The structural proteins of the ECM and BM are varied and include fibrillar proteins (e.g. collagens and elastin), proteoglycans, and multidomain glycoproteins (e.g. fibronectin and laminin) [6].
The integrity of the ECM involves a balance between the synthesis and organization of its structural constituents and their degradation by metalloendopeptidases belonging to the matrixin family [7, 8]. The most important of these endopeptidases belong to the matrix metalloproteinase (MMP) family which, to date, comprises at least 26 members. Among them, the gelatinases MMP-2 and MMP-9, are potent in their ability to cleave gelatins, denatured collagens (IV, V, VII, X), elastin, fibronectin, and TNF-α [6]. A recent report has revealed that MMP-2 can also cleave native collagen I [9] and collagen III [10]. However, it seems that the MMP are capable of degrading all the components of the ECM [11]. The activities of these extracellular metalloproteinases are controlled by specific naturally occurring inhibitors (TIMPs) [7, 11].
The aim of the present paper was to identify the mechanisms involved in the alterations of the connective tissue induced by magnesium deficiency in mice.

Material and methods

Animals and treatment

Two groups (n = 5/group) of Swiss OF1 mice were fed (i) a magnesium-deficient diet (50 ppm ± 5 ppm) for 42 days or (ii) a standard diet (1700 ± 100 ppm magnesium). At the end of the deprivation period, the mice were killed under chloral anesthesia (7 ml/kg b.w. of a 5 % chloral saline solution) and the thoracic aorta was removed and cultured in Hanks medium at 37°C. Care and treatment of animals was according to the guidelines for animal care.

Plasma magnesium concentrations

Magnesium concentrations were determined by atomic spectrophotometry in plasma and expressed in mg/mL [12].

Materials

Cell culture medium, additives and fetal bovine serum were purchased from Gibco unless otherwise specified. Electrophoresis supplies were obtained from Biorad. Calf skin collagen type 1 and porcine skin gelatin were from Sigma Chemical Co. MMP-2 and MMP-9 were purchased from Valbiotech.

Cell culture

The thoracic aorta was rinsed three times with DMEM supplemented with 400 µg/mL of penicillin, 400 µg/mL of streptomycin, and 4 µg/mL of fungizone and cut into small pieces. Primary cultures were established in 25 cm² culture flasks in DMEM containing 20 % foetal bovine serum, penicillin (100 µg/mL), streptomycin (100 µg/mL), and fungizone (2 ng/mL). Monolayer cultures were maintained in 5/100 CO₂/air v/v and cell culture medium was changed every 48 hr. After passage, the cells were routinely maintained in 10 % foetal bovine serum containing DMEM.

Histopathologic examination

The thoracic aorta wall structure was examined by a standard method, under microscopic examination (× 40) after staining with eosin-hematoxylin. Fibrous macromolecule structures were studied also by light microscopy by means of specific colorations. Collagens were revealed by Red Sirius F3ba [13] and elastin by (+) catechin fuchsin [14].

Determination of MMP secreted into culture medium by zymographic analyse of gelatinase

Gelatinase levels were determined by mean of a zymographic assay in which gelatin (collagen type I) was incorporated into polyacrylamide gel. A 0.5 mg/mL gelatin concentration within gels was found optimal for the detection of gelatinase and its activated form. Briefly, conditioned culture medium was mixed with Laemmli sample buffer containing 0.1 % SDS without reducing agent and electrophoresed without boiling, under nondenaturing conditions. Following electrophoresis, SDS was eluted from the gel by 3 successive washings in 2.5 % Triton X 100 for 20 min to allow collagen fibrillation and protein to renature. Gels were then immersed for 16-48 hr in 100 mM of Tris, 5 mM of CaCl₂, 2 µM of ZnCl₂ pH 8.0, and stained with Coomassie blue R.250 to reveal zones of lysis [15].

Results

At the end of the deprivation period, the mouse magnesium plasma level was 5.0 ± 0.3 vs 21.0 ± 1.5 mg/mL at the beginning of the experiment and in controls. In the magnesium-deficient mice a severe vascular lesion appeared as compared to controls (figure 1). The lesion was characterized by an aortic wall thinning, a disorganization of smooth muscle cells and a hyaline deposit. In controls, the intima of aorta consisted of a layer of endothelial cells which lay on the internal elastic membrane of the media. The media consisted of 4 to 6 thick, wavy and parallel elastic lamellae. Between each lamella were smooth muscle cells which were surrounded by a scant ground substance. In magnesium-deficient mice, profound changes occurred in the elastic framework consisting in disorganization, disorientation, fragmentation or intense destruction of the fibers. Large pools of ground substance which stained very weakly were observed. Elastin coloration showed an important thinning and even breakdowns of the elastic laminas (figure 2) whereas collagen coloration showed a wall refinement and punctiform settlings of collagens (instead of a continuum in controls) (figure 3). Finally, zymography showed that MMP-9 and MMP-2, which were present as zymogens (inactive forms) in controls were supposed to be present in their active and inactive forms in magnesium-deficient mice. Indeed, it appeared two gelatinolytic new bands presenting a MMP-9 lower molecular weight that could be one or two active forms of this zymogen. The same type activation seems also to be present for MMP-2 but the new active band is less clearly appearent. This could explain the collagen and elastin alterations previously described (figure 4). The set up of additional gelatinolytic activities in magnesium deficiency has not been reported so far and would explain, at least in part, the sensitivity of magnesium-deficient mice to various stress or xenobiotics.

Discussion

The magnesium-deficient diet used for 42 days in Swiss OF1 mice in the present assay, led as usual to a severe magnesium deficiency, since magnesium plasma levels were 4-fold decreased as compared to controls [16]. These conditions induced severe alterations of the 2 main connective fibers, elastin and collagens, as described previously by others [4, 5, 17].
It is now well known that the connective tissue disorder may be responsible for severe pathological damage including cancer invasion and metastatis, arthritis, autoimmune diseases, periodontitis, tissue ulceration, atherosclerosis, aneurysm and heart failure [1, 7, 18] and would favor atherosclerosis [5]. Indeed, organ cell alignment and tissue structural integrity are maintained through interactions with the ECM. Tissue remodeling requires the degradation of ECM through the action of metalloproteinases (MMPSs) [19]. The MMPs are upregulated and/or activated during inflammation or physiological remodeling processes in response to specific stimuli. In general, MMPs are secreted as zymogens (inactive or latent proforms of these enzymes) [18] and are activated by proteolytic removal of an amino-terminal domain [20].
Magnesium deficiency experiments in rats showed that magnesium acts directly on heart and vessel muscular and endothelial cells [21] inducing aging-type damage [1]. Fibrosis during magnesium deficiency has long been known [22]. More precisely, magnesium deficiency results in a process referred to as a vascular wall alteration implicating collagens and elastin and resulting in intimal thickening [17, 23]. The normal fibrillar proteins have a structure which usually ensures that the hydrophobic groups and potential points of enzymic attack are hidden [5]. But, in magnesium deficiency, vascular lesions appear corresponding mainly to thinning and fragmentation of the elastic membranes [4]. These changes were attributed to elastin structural changes resulting in a higher susceptibility of its peptide bonds to proteolytic attack [5]. In other respects, in normal conditions of connective tissue remodeling, both fibrillar proteins can undergo extensive breakdown leading to rapid changes in tissue mass and function [6]. A slowing of collagen resorption has been demonstrated during magnesium deficiency in different experimental systems [9, 24, 25] which can be reverted by magnesium supplementation [26]. This less readily degradation of collagen was attributed either to changes in collagen conformation or to an inhibiting effect of magnesium on the collagenases involved in the collagen degradation [5].
In the present work, we observed, in agreement with previous findings, in magnesium-deficient mice an overall vascular wall damage. Specific coloration of elastin showed thinning and fragmentation of the elastic membranes. Specific coloration of collagen showed also a wall refinement and punctiform settlings of collagens instead of a continuum in controls. Interestingly, it appeared by zymography that the MMP-2 and MMP-9, implicated in the elastin and collagen degradation, which were under a zymogen form in controls were supposed to be under an active form in magnesium deficient mice. This abnormal activation of MMP-2 and -9 would greatly damage the physiological structure of their corresponding substrates i.e. elastin and collagens. The reason for such an activation in magnesium deficiency remains to be elucidated. It may be postulated that the specific tissue inhibitors of metalloproteinases (TIMPs) are unefficient in severe magnesium deficiency. The normalization of the connective structure would be linked to a restoration of TIMP activity, resulting in an inactivation of MMP, thus allowing a connective tissue normalization.

Conclusion

Magnesium deficiency seems responsible, in mice, for a severe alteration of vascular macromolecular components. These macromolecular changes would induce an increase in the vessel permeability. If the same type of macromolecular alteration occurs at the connective component level in cerebral capillaries, then we could hypothesize a transient or othervise hyperpermeability of the blood-brain barrier. However, further investigations, using Western and Dot blots are needed to confirm the present results.

References

1. Durlach J, Bara, M. (2000): Le Magnésium en Biologie et en Médecine, Cachan, Ed. Méd. Inter.

2. Kumar BP, Shivakumar K, Kharta CC. Mg deficiency-related changes in lipid peroxidation and collagen metabolism in vivo in the rat heart. Int J Biochem Cell Biol 1997; 29: 129-34.

3. Britton WM, Stokstad ELR. Aorta and other soft tissue calcification in the magnesium deficient rat. J Nutr 1970; 100: 1501-6.

4. Seelig MS. (1980) Magnesium deficiency in the pathogenesis of disease. Avioli L.V. ed., Plenum Medical Book Company, New York-London, 1980.

5. Rayssiguier Y. (1981): Magnesium and lipids interrelationships in the pathogenesis of vascular diseases. Proceed. 3rd International Symposium on Magnesium, Baden-Baden.

6. Hulboy DL, Rudolph LA, Matrisian LM. Matrix metalloproteinases as mediators of reproductive function. Mol Hum Reprod 1997; 3: 27-45.

7. Woessner JF. Matrix matalloproteinases and their inhibitors in connective tissue remodeling. FASEB J 1991; 5: 2145-54.

8. Birkedal-Hansen H. Proteolytic remodeling of extracellular matrix. Curr Opin Cell Biol 1995; 7: 718-735.

9. Aimes R., Quigley JP. Matrix metalloproteinase-2 is an interstitial collagenase. Inhibitor-free enzymes catalyzes the cleavage of collagen fibrils and soluble native type I collagen generating the specific and length fragments. J Biol Chem 1995; 270: 587-276.

10. Berton A, Godeau G, Emonard M, Bqka K, Bellon P, Hornebeeta W., Bellon G. Analysis of the ex vivo specificity of human gelatinases A and B towards skin elastic fibers and collagen fibers by compterized morphometry. Mat Biol 2000; 19: 139-48.

11. Denis LJ, Verweij J. Matrix metalloproteinase inhibitors: Present achievements and future prospects. Invest. New Drugs 1997; 15: 175-85.

12. Rousselet F, Durlach J. (1971): Techniques analytiques et explorations pratiques du métabolisme magnésique en clinique humaine. In: Durlach J., 1er Symposium International sur le Déficit Magnésique en Pathologie Humaine. I. Vol. des Rapports, SGEMV ed. Vittel, 65-90.

13. Junqueira LCU, Bignolas G, Brentani RR. Picrosirius staining plus polarization microscopy, a specific method for collagen detection in tissue sections. Histochem J 1979; 11: 447-55.

14. Godeau G, Gonnord G, Jolivet O. A selective method for the quantitative estimation of elastin fibers. Analyt Quant Cytol 1986; 4: 321-5.

15. Gogly B, Groult N, Hornebeck W, Godeau G, Pellat B. Collagen zymography as a sensitive and specific technique for the determination of subpicogram levels of interstitial collagenase. Anal Biochem 1998; 255: 211-6.

16. Bac P, Pagès N, Herrenknecht C, Teste JF. (1996): Effects of pharmacological oral doses of several magnesium salts on a model of central nervous hyperexcitability due to magnesium deficiency in mice. In Current Research in Magnesium, (M. J. HALPERN and J. DURLACH, eds), pp 287-291. London: John Libbey.

17. Seelig MS, Haddy FJ. (1980): Magnesium and the arteries. I. Effects of magnesium deficiency on arteries and on the retention of sodium, potassium and calcium. In: Magnesium in Health and Disease. Cantin M.D. and Seelig M.S. eds., p605-638. New York-London: S.P. Medical and Scientific Books.

18. Johnson LL, Dyer R, Hupe DJ Matrix metalloproteinases. Chem Biol 1998; 2: 466-71.

19. Rayssiguier Y, Badinand F, Kopp J. Effect of magnesium deficiency on parturition and uterine involution in the rat. J Nutr 1979; 109: 2117-25.

20. Stetler-Stevenson WG, Krutzsch HC, Wacher MP, Margulies MK, Liotta L. The activity of human type IV collagenase proenzyme. J. Biol. Chem. 1989; 264: 1353-6.

21. Miyamoto A, Kanda J, Haruta T, Nishio A. Stability of free Mg2+ concentration and increased concentration of free Ca2+ in vascular smooth muscle cells during dietary magnesium deficiency in rat. Magnes Res 1996; 9: 101-8.

22. Gunther T, Ising H, Merker HJ. Electrolyt und Kollagengehalt im Rattenherzen bei chronischem Mg-Mangel und Stress. Z Klin Chem Klin Biochem 1973; 11: 87-92.

23. Mller W, Iffland R, Firching R. Relationship between magnesium and elastic fibres. Magnes Res 1993; 6: 215-22.

24. Chevalier F, Fabry R, Kopp J, Rayssiguier Y. (1980): John Lee Pratt Intern. Symp. Mg in animal nutrition, Blacksburg, VA Abstract.

25. Larvor P. (1981) Physiological and biochemical function of Mg in animals. In Fontenot J.P., Bunce G.E., Webb Jr K.E., Allen V.G. Role of Mg in animal nutrition. John Lee Pratt Animal Nutrition Program, Blacksburg, VA-USA.

26. Szelenyi I. (1971) Physiological interrelationships between Mg and heart. In: 1er Symp. Intern.Déf. Magn.Path. Hum. Durlach J., I. Vol. des Rapports, SGEMV ed. Vittel, pp 195-212.