Marfan syndrome, magnesium status and medical prevention of cardiovascular complications by hemodynamic treatments and antis

ARTICLE

Auteur(s) : S. Igondjo-Tchen¹, N. Pagès², P. Bac³, G. Godeau¹, J. Durlach⁴

¹Faculté de Chirurgie dentaire, Paris V, 92210 Montrouge, France; ²Faculté de Pharmacie, Strasbourg, 67400 Illkirch, France; ³Faculté de Pharmacie , Paris XI, 92240 Chatenay-Malabry, France; ⁴SDRM, UPMC, Paris VI, 75252 Paris cedex 5, France

Introduction

The Marfan Syndrome (MFS) is the most common genetic disorder of the connective tissue with an estimated prevalence of up to 1 in 10000. MFS is not infrequent being almost as prevalent in the population as cystic fibrosis and muscular dystrophy. Approximately 80 % of individuals with MFS inherited the mutation from a parent, with the remaining 20 % of cases arising as a result of new mutations [1, 2]. Clinical variability of expression within families but also between families is the hallmark of the disorder [3]. The involvement of the cardiovascular system is one of the most important pronostic data for MFS: a direct relationship exists between aortic root dimension and complications of MFS [5]. Since simple aortic root dilatation precedes catastrophic aortic events over a long period of time the main management of patients with MFS can rely on the early medical prevention of the major aortic complications by hemodynamic treatments (β adrenergic antagonists mainly) and possibly, in the future, on an etiologic antisense gene therapy [1-6].

Hemodynamic treatments

It has been postulated that medical treatment which diminishes the forcefulness of cardiac contractions and reduces blood pressure may retard aortic root dilatation and can prevent severe aortic disorders [3, 6].

β-adrenergic receptor antagonists

In 1983, Pyeritz reported for the first time the use of β-adrenergic receptor antagonists to retard the rate of aortic root dilatation in patients with MFS. Since then their use has become ubiquitous [4, 5].
In 1994, Shores et al. published a landmark study that tested the hypothesis that the long-term administration of oral β-adrenergic receptor antagonists would reduce the rate of aortic dilatation in patients with MFS. The rate of increase in aortic ratio (measured aortic diameter divided by expected diameter) predicted for age, gender and body surface area-adjusted nomograms in patients receiving propranolol was 0.023 per year compared with 0.084 in untreated patients [3, 5].
Data from other studies are consistent with these results [3-8].
Propranolol was used initially, but it seems better to use β-blockers without intrinsic sympathomimetic activity such as atenolol or nadolol [3, 5, 8].
– No firm recommendations can be made with regard to dose or type of β-blocker, but common sense suggests that dose titration to a heart rate less than 60 would provide optimal therapy [5, 8].
– Retardation of aortic dilatation in these patients seems to be mediated by improvement in aortic root compliance [5]. However there was a heterogeneous response in the aortic root elastic properties after long-term treatment with β-blocker in asymptomatic patients with MFS. Stiffness index and distensibility are more likely to respond when the baseline end-diastolic aortic root diameter is < 40 mm [9].
Successful acceptance of β-adrenergic receptor antagonist therapy may be limited by adverse effects including bradycardia, hypotension, fatigue, dizziness, weakness and impotence. Contra indications to β-blockers include asthma, severe peripheral vascular disease and, occasionally, diabetes [5]. As with other long-term prophylactic therapies, the use of these agents can be challenging, especially in asymptomatic patients. One third of patients with MFS treated with β-blockers developed at least one side effect [10].
Several other hypotensive agents may represent a therapeutic alternative to patients who cannot or will not take adrenergic receptor antagonists: mainly calcium channel antagonists [3, 5, 7].

Other hypotensive agents

– Calcium channel inhibitors including verapamil and diltiazem may be used (long acting forms once or twice dayly), or occasionally an angiotensin-converting enzyme (ACE) inhibitor [3, 5, 7].
– However, no prospective data have been published comparing the rate of aortic ratio in patients with MFS given a Ca channel antagonist. A multicenter international trial ADAMS (Aortic Dilatation And Marfan Syndrome) has been initiated and is following its course [5].

Interaction between hemodynamic drugs and magnesium status

A balanced magnesium status is necessary for the health of the cardiovascular apparatus and for the best efficiency of the used hemodynamic prophylactic treatments of patients with MFS.
Magnesium deficiency appears to act as a cardiovascular risk factor through multivarious noxious mechanisms: indirectly on lipid profile, on connective tissue particularly [11, 12] and on nervous, endocrine – and even directly on cardiovascular – systems [13-15]. Mitral valve prolapse appears as a common manifestation of magnesium deficit [13-18] as well as of MFS [3, 5, 8]. Magnesium deficit may induce mitral disorders either through papillary cardiomyopathy or through connective alterations in the valve [13-18].
The various hemodynamic drugs used by patients with MFS for the prevention of cardiovascular complications may have been considered as “partial magnesium analogues”. A balanced magnesium status is necessary for their best efficiency and their least toxicity [13, 15].
The close relationships between β-receptor and magnesium status were also illustrated during tocolysis. Efficiency and tolerance of β-mimetics are highly increased by nutritional doses of magnesium, while high parenteral magnesium doses increase the cardiovasotoxic effects of β-mimetics by adding their own stressful effects on the cardiovascular targets [13, 15].
It is advisable to ensure a balanced magnesium intake (6 mg/kg/day) [19] in patients with MFS to obtain the best efficiency and tolerance of the long-term prophylactic treatments of their cardiovascular complications.

Gene therapy might constitute an etiologic specific treatment of MFS

This autosomal dominant disorder is due to a mutation in the gene for fibrillin 1 – (a major component of the extracellular microfibrils) – on chromosome 15: FBN1. More than 100 different FBN1 mutations have been identified in individuals with MFS: the FBN1 mutations responsible for MFS cause defects in fibrillin synthesis, secretion and incorporation into the extracellular matrix. Fibrillin mutations in MFS act via a dominant-negative mechanism. The demonstration of a dominant-negative mechanism supports suppression of expression of the mutant FBN1 allele as a valid therapeutic approach for MFS [1-3, 5, 6].

FBN1-RZ1 antisense ribozyme

Ribozyme catalytic RNA molecules have been widely touted for their potential to suppress expression of specific gene products: there has been tremendous interest from the biotechnology industry in the application of antisense ribozyme [20].
Hammerhead ribozymes are a class of genetically engineered RNA molecules that can cleave other RNA sequences by means of an intrinsic enzyme-like activity [21]. These molecules have two important domains:
– The catalytic “hammerhead” portion that cleaves the target RNA sequence by an enzyme-like mechanism that is dependent on the presence of divalent cations, preferably Mg [22-25].
– Flanking antisense sequences that confer specificity of binding of the ribozyme to the desired region of a target of a RNA molecules [22, 24].
Hammerhead ribozymes might therefore be utilized to specifically target FBN1 transcripts in individuals with MFS.
The possibility that the reduction of the amount of product from the mutant FBN1 allele might be a valid therapeutic approach for MFS has been tested first in vitro, then in cultured fibroblasts [2, 26, 27].
FBN1-RZ1 ribozyme efficiently cleaves its target in vitro.
A radiolabelled 86 base FBN1 mRNA fragment is cleaved by this transacting hammerhead ribozyme whereas the control ribozyme leaves the FBN1 mRNA fragment intact. The specificity of action of FBN1-RZ1 is assessed by determination of its effect on total fibroblast RNA.
Ribozyme activity requires the presence of Mg [21] The observed cleavage of FBN1 mRNA is actually Mg-dependent. Ribozyme treatment in the absence of magnesium shows no cleavage of the 86 base fragment whereas treament in the presence of 10-40 mM MgCl₂ leads to the production of the expected 56 and 30 base subfragments at both 37°C and 50°C [2].
FBN1-RZ1 specifically reduces FBN1 mRNA and fibrillin production by cultured fibroblasts.
Transfection with FBN1-RZ1 specifically reduces FBN1 mRNA levels and dramatically decreases the amount of fibrillin produced.
To determine whether the FBN1-RZ1 mediated down regulation of fibrillin production was specific for fibrillin, the effect of delivery of FBN1-RZ1 on the production of an unrelated matrix protein fibronectin by cultured fibroblasts was also examined. In contrast to fibrillin production, delivery of FBN1-RZ1 did not reduce the amount of fibronectin [26].
The use of this transacting hammerhead ribozyme targeted to the 5'-end of the FBN1 mRNA is a valid approach to develop a therapy based on the ablation of a mutant gene product. This ribozyme might specifically down regulate the mutant FBN1 allele in fibroblasts derived from individuals with MFS or combine the ablation of endogenous FBN1 expression on MFS individuals with the delivery and expression of normal fibrillin [2, 26, 27]. But a major obstacle to be overcome is the efficient and effective delivery of the ribozyme to its target.

Hydrogel coated angioplasty balloon

The effective use of ribozymes as therapeutic agents will depend on efficient delivery of the ribozyme to its target, but the various proposed available vectors raise problems of tolerance and efficiency which still remain unsolved [6, 27-29].
Among alternative approaches can be envisioned tissue specific delivery. A hydrogel angioplasty balloon might be a possible vector specifically for delivering an antisense hammerhead ribozyme in the aortic wall [6]. Since simple aortic root dilatation precedes dissection and rupture over a long period of time, specifically aiming at treating the genetic disorder of fibrillin 1 by providing the antisense hammerhead ribozyme in the aortic wall deserves consideration [6].
Ribozymes may be taken up by tissue upon local application in animal models without carriers. This agrees with the in vivo application of antisense oligodeoxynucleotides which is also achieved without carriers [24].
Further research might study ex vivo the actions of local application of antisense hammerhead ribozyme possibly on human (normal and from patients with MFS) aortic wall before assessing in vivo the efficiency and tolerance of the local vectorization of this small catalytic RNA molecule.
There has been tremendous interest from the biotechnology departments in the application of antisense ribozyme. The efficacy of intracellular binding of hammerhead ribozyme to its cleavage site in target RNA is a major requirement for its use as a therapeutic agent. Among several factors the length of the ribozyme antisense arms is important. Hammerhead ribozymes with short antisense arms have a high turnover when compared to their counterparts with long arms [28].

Interaction between ribozyme and magnesium status

Ribozyme activity requires the presence of magnesium [21] and hammerhead ribozyme particularly [23-25]. FBN1-RZ1 ribozyme is actually Mg-dependent: in the absence of magnesium no cleavage of FBN1 mRNA is observed while the expected cleavage is produced in the presence of 10-40 mM MgCl₂ [26].
One of the roles of magnesium requirement is to support the formation of the catalytically competent structure: the multiple conformational states of the hammerhead ribozyme are Mg²⁺-dependent [23-25]. Mg²⁺ may also act as an acid-base catalyst [24].
Around 20 per cent of the population consume less than two-thirds of the RDA for Mg. It is necessary to palliate the chronic primary magnesium deficiency by atoxic nutritional supplementation. Dietetic supplements should be achieved using a high magnesium density nutrient with the best possible availability, magnesium in water for example [19]. It is logical to obtain a balanced magnesium intake (6 mg/kg/day) through a daily magnesium supplement lower or equal to 300 mg of Mg per day, Mg supplementation under UL (tolerable upper intake level) for Mg [30] which defines the threshold of toxicity for a magnesium oral supplement [19]. Today the indications of pharmacological magnesium therapy in patients with MFS are not established. However it is so well advised to establish a balanced magnesium intake through an atoxic nutritional magnesium supplement that the environment should not be overlooked even in monogenic disease [31] such as MFS.

Conclusion

It is always necessary to ensure a balanced magnesium intake in patients with MFS. In particular it seems well advised to obtain the best efficiency and tolerance of the pharmacologic long term prophylactic treatment for cardiovascular complications and perhaps in the future for etiologic antisense gene therapy.

References

1. Pyeritz RE. The Marfan Syndrome. Ann Rev Med 2000; 51: 481-510.

2. Kilpatrick MW, Phylactou LA. Towards and RNA-based therapy for Marfan syndrome. Mol Med Today 1998; 4: 376-81.

3. Tsipouras P, Devereux RB. Marfan Syndrome: genetic basis and clinical manifestations. Seminars in Dermatology 1993; 12: 219-28.

4. Pyeritz RE. Effectiveness of beta-adrenergic blockade in the Marfan syndrome: experience over 10 years. Am J Med Genet 1989; 32: 245-52.

5. Tsipouras P, Silverman DI. The genetic basis of aortic disease. Cardiol Clin North Am 1999; 17: 683-96.

6. Durlach J. A possible advance in arterial gene therapy for aortic complications in the Marfan Syndrome by local transfer of an antisense Mg-dependent hammerhead ribozyme. Magnes Res 2001; 14: 65-7.

7. Rossi-Foulkes R, MJ, Rosen SE, Kramer-Fox R, Ehlers KH, O'Loughlin JE, Davis JG, Devereux RB. Phenotypic features and impact of beta-blockers or Ca-antagonist therapy on aortic human in the Marfan syndrome. Am J Cardiol 1999; 83: 1364-8.

8. Nienaber CA, Von Kodolitsch Y. Therapeutic management of patients with Marfan Syndrome. Cardiology in Review 1999; 7: 332-41.

9. Rios AS, Silber EN, Bavishi N, Varga P, Burton BK, Clark WA, Denes P. Effect of long-term β-blockade on aortic root compliance in patients with Marfan syndrome. Am Heart J 1999; 137: 1057-61.

10. Shores J, Berger KR, Murphy EA, Pyeritz RE. Progression of aortic dilatation and the benefit of long-term beta-adrenergic blockade in Marfan syndrome. New Engl J Med 1994; 330: 1335-41.

11. Pagès N, Gogly B, Godeau G, Igondjo-Tchen S, Maurois P, Durlach J, Bac P. Structural alterations of the vascular wall in Mg-deficient mice. A possible role of gelatinases A (MMP2) and B (MMP9). Magnes Res 2003; 16: 43-48.

12. Benjamin IJ. Matrix metalloproteinases: from biology to therapeutic strategies in cardiovascular disease. J Invest Med 2001; 49: 381-97.

13. Durlach J. (1988): Magnesium in clinical practice. pp 360. London-Paris: John Libbey publ.

14. Durlach J, Bara M, Guiet-Bara A. (1989): Magnesium level in drinking water: its importance in cardiovascular risk. In: Magnesium in health and disease, Y. Itokawa and J. Durlach eds., pp 173-182. London: John Libbey.

15. Durlach J, Bara M. (2000): Le Magnésium en biologie et en Médecine. Tec. et Doc. Ed., pp 403. Paris.

16. Durlach J, Lutfalla G, Poenaru S, Reba A, Henrotte JG, Fabiani F, de Vernejoul F. (1985): Latent tetany and mitral valve prolapse due to chronic primary magnesium deficit. in Halpern M.J. and Durlach J. Mg deficiency: physiopathology and treament implications on Mg (Lisbon oct.1983), Karger ed., pp 102-112. Basel-Munchen-Paris-London-N.Y.-Sydney-Tokyo.

17. Durlach J, Durlach V. Idiopathic mitral valve prolapse and Mg. State of the art. Mg-Bull 1986; 8: 156-169.

18. Durlach J. Primary mitral valve prolapse: a clinical form of primary magnesium deficit. Magnes Res 1994; 7: 339-40.

19. Durlach J. (2001): Importance and clinical forms of chronic primary magnesium deficiency in human beings. In: Advances in Magnesium Research: Nutrition and Health, Y. Rayssiguier, A. Mazur and J. Durlach eds. pp 13-20. John Libbey & Company Ltd.

20. James HA, Gibson I. The therapeutic potential of ribozymes. Blood 1998; 91: 371-82.

21. Cech TR, Bass B. Biological catalysis by RNA. Ann Rev Biochem 1986; 55: 599-629.

22. Dorai T, Olsson CA, Katz AE, Buttyan R. Development of a hammerhead ribozyme against bcl-2. I. Preliminary evaluation of a potential gene therapeutic agent for hormone-refractory human prostate cancer. Prostate 1997; 32: 246-58.

23. Menger M, Tuschl T, Porschke D. Mg²⁺ dependent conformational changes in the hammerhead ribozyme. Biochemistry 1996; 35: 14710-6.

24. Eckstein F, Bramlage B. The hammerhead ribozyme. Biopolymers (Nucleic Acid Science) 1999; 52: 147-52.

25. Menger M, Eckstein F, Porschke, D. Multiple conformational states of the hammerhead ribozyme, broad time range of relaxation and topology of dynamics. Nucleic Acids Res 2000; 28: 4428-34.

26. Kilpatrick MW, Phylactou LA, Godfrey M, Wu CH, Wu GY, Tsipouras P. Delivery of a hammerhead ribozyme specifically down-regulates the production of fibrillin-1 by cultured dermal fibroblasts. Hum Mol Gen 1996; 5: 1939-44.

27. Phylactou LA, Tsipouras P, Kilpatrick MW. Hammerhead ribozymes targeted to the FBN1 mRNA can discriminate a single base mismatch between ribozyme and target. Biochem Biophys Res Comm 1998; 249: 804-10.

28. Sioud M. Effects of variations in length of hammerhead ribozyme antisense arms upon the cleavage of longer RNA substrates. Nucleic Acids Res 1997; 25: 333-8.

29. Brasselet C, Lafont A, Lemarchand P, Metz D. Intérêts de la thérapie génique en pathologie cardiovasculaire: principe et exemples d'applications. Courrier ARCOL 1999; 1: 185-7.

30. National Research Council (1997): Magnesium (chap. VI). In: Dietary Reference Intakes for Ca, P, Mg, vitamine D and Fluoride. Washington DC: IOM National Academy Press.

31. Kaprio J. Commentary: role of other genes and environment in monogenic disease. Br Med J 2001; 322: 1023.