Neurology -- Correspondence for Cagliani et al., 61 (11) 1513-1519

Correspondence published:

A CAV3 microdeletion differentially affects skeletal muscle and myocardium: Josef Finsterer, Claudia Stoellberger (8 January 2004)

Reply to Finsterer: Rachele Cagliani, Giacomo P. Comi (8 January 2004)

A CAV3 microdeletion differentially affects skeletal muscle and myocardium 8 January 2004

Josef Finsterer
Postfach 348, 1180 Vienna, Austria,
Claudia Stoellberger
Send Correspondence to journal:
Re: A CAV3 microdeletion differentially affects skeletal muscle and myocardium

duarte{at}aonmail.at Josef Finsterer, et al.

In their article, Cagliani et al observed a 40% reduction of caveolin-3 in the myocardium of a single patient and concluded that this mutation affects skeletal muscle and myocardium differentially.[1] The findings raise concerns.
Was cardiac involvement (CI) absent in the patients because there were no comprehensive cardiac investigations? At least history, clinical cardiologic examination, ECG, echocardiography, and 24-hour ECG are necessary to assess CI. Only rudimentary data of these investigations were provided for patients III-1, III-4, and IV-1. Since CI may also develop during the disease course, it is important to regularly follow up. Reduced myocardial caveolin-3 in patient III-1 may be independent of the underlying mutation and could be also influenced by coronary heart disease or extracorporal circulation during surgery. In rabbits, chronic myocardial hypoxia increased nitric oxide synthase and simultaneously reduced caveolin-3.[2]
Was myocardial biopsy taken from a region supplied by a stenosed or normal coronary artery? Was the microdeletion also detected in the myocardium? Possibly hyper-CK-emia in patient III-2 erroneously led to the diagnosis myocardial infarction, although the patient died from CI in caveolinopathy. Did patient III-1 have myocardial thickening from arterial hypertension? Since caveolin-3 reduction manifests in the skeletal muscle with various different phenotypes, CI may be also heterogeneous, even within a single family. How do the authors explain that caveolin-3 knockout mice develop severe cardiomyopathy with hypertrophic cardiomyocytes, while humans with caveolin-3 mutations seem to show few cardiac abnormalities?[3] How is it explained that increased nitric oxide synthase results in decrease of caveolin-3 and the development of hypertrophic cardiomyopathy in mice but not in humans?[4] That reduction of caveolin-3 leads to cardiomyopathy is supported by inhibition of hypertrophy of rat cardiomyocytes by adenovirus -mediated over-expression of caveolin-3.[5] On the contrary, over- expression of caveolin-3 induces severe cardiomyopathy in mice. Which are the modifying factors that led to muscle but not to cardiac disease? Figure 4 shows the muscle of a control subject. Did all control subjects undergo muscle biopsy? Atypical absences in patient IV-1 may be associated with the expression of caveolin-3 also in endothelial cells and astrocytes of the brain. Besides skeletal muscle, myocardium, and brain, caveolin-3 also occurs in the smooth muscle. Did the authors find involvement also of smooth muscle-containing organs?
To demonstrate CI in caveolinopathies, thorough cardiologic examination, regular follow-ups, and investigations not only of a single patient but all mutation carriers are required.
References
1 Cagliani R, Bresolin M, Prelle A et al. A CAV3 microdeletion differentially affects skeletal muscle and myocardium. Neurology 2003;61:1513-1519.
2 Shi Y, Pritchard KA Jr, Holman P et al. Chronic myocardial hypoxia increases nitric oxide synthase and decreases caveolin-3. Free Radic Biol Med 2000;29:695-703.
3 Park DS, Woodman SE, Schubert W et al. Caveolin-1/3 double-knockout mice are viable, but lack both muscle and non-muscle caveolae, and develop a severe cardiomyopathic phenotype. Am J Pathol 2002;160:2207-2217.
4. Ohsawa Y, Toko H, Katsura M et al. Overexpression of P104L mutant caveolin-3 in mice develops hypertrophic cardiomyopathy with enhanced contractility in association with increased endothelial nitric oxide synthase activity. Hum. Mol. Genet. 2004 13: 151-157.
5. Koga A, Oka N, Kikuchi T et al. Adenovirus-mediated overexpression of caveolin-3 inhibits rat cardiomyocyte hypertrophy. Hypertension 2003;42:213-219.

Reply to Finsterer 8 January 2004

Rachele Cagliani,
I.R.C.C.S. E. Medea, Associazione La Nostra Famiglia
Via Don Luigi Monza, 20, Bosisio Parini (LC) Italy,
Giacomo P. Comi
Send Correspondence to journal:
Re: Reply to Finsterer

rcagliani{at}bp.lnf.it Rachele Cagliani, et al.

In general, we agree with Dr. Finsterer and the statement that to rule out cardiac involvement in caveolinopathies regular follow-up of affected individuals and complete cardiological investigation are essential.
However, the raised criticisms led to alternative hypotheses that appear to us as mutually exclusive: either relative decrease of caveolin-3 is a consequence of heart hypoxic damage, therefore caveolin-3 does not cause cardiac involvement or caveolin-3 deficiency contributes to clinical phenotype and the ischemic heart disease is an incidental finding. We also find that some specific points and comments are inappropriate.
As clearly shown in Fig. 5 a, the patient�s III-1 biopsied heart tissue does not show histological signs of ischemia. In addition, no sign of ischemic damage was observed elsewhere in this tissue sample. We consider that a 40% cav3 reduction in heart tissue can hardly be ascribed to factors such as hypoxia, extracorporeal circulation or coronary heart disease and much more so in the presence of a cav3 mutated subject. The only reason to test for the presence of the mutation in the heart tissue would be to suppose the presence of a somatic mosaicism, a hypothesis ruled out by the genealogic tree. Furthermore, patients III-8, III-9, IV-1 and IV-7 have now completed a re- evaluation after one year of follow-up, including neurological and cardiologic examination, ECG, echocardiography, and 24-hour ECG, without any evidence of heart involvement so far.
We would like also point out that, in our opinion, mouse strains (either knock out or transgenic) developed up to now [3,4] cannot be recognized as faithful models for the human pathology. As far as the knock out is concerned, we consider that the absence of any translated product cannot be compared to the production of a mutated polypeptide with a dominant effect on the wt protein. With respect to the recently described transgenic mice [4], the mutated allele was expressed at much higher levels than the wild type transcript, again resulting in a considerable difference as compared to patients carrying CAV3 mutations. We consider that these observations might well suffice to explain the different cardiac phenotypes in mouse models and CAV3 patients. The great majority of subjects carrying CAV3 mutations have not developed cardiac symptoms although it cannot be ruled out that cardiac involvement might represent an unusual feature in LGMD1C patients.
As we discussed, the molecular basis allowing partial caveolin -3 localization in cardiomyocytes are unknown; in any case the presence of a considerable protein amount (60%) in the patient�s heart might explain the absence of cardiac involvement in patients carrying Phe97del mutation in the CAV3 gene.