Sustaining cardiac claudin-5 levels prevents functional hallmarks of cardiomyopathy in a muscular dystrophy mouse model.

Identification of new molecular targets in heart failure could ultimately have a substantial positive impact on both the health and financial aspects of treating the large heart failure population. We originally identified reduced levels of the cell junction protein claudin-5 specifically in heart in the dystrophin/utrophin-deficient (Dmd(mdx);Utrn(-/-)) mouse model of muscular dystrophy and cardiomyopathy, which demonstrates physiological hallmarks of heart failure. We then showed that at least 60% of cardiac explant samples from patients with heart failure resulting from diverse etiologies also have reduced claudin-5 levels. These claudin-5 reductions were independent of changes in other cell junction proteins previously linked to heart failure. The goal of this study was to determine whether sustaining claudin-5 levels is sufficient to prevent the onset of histological and functional indicators of heart failure. Here, we show the proof-of-concept rescue experiment in the Dmd(mdx);Utrn(-/-) model, in which claudin-5 reductions were originally identified. Expression of claudin-5 4 weeks after a single administration of recombinant adeno-associated virus (rAAV) containing a claudin-5 expression cassette prevented the onset of physiological hallmarks of cardiomyopathy and improved histological signs of cardiac damage. This experiment demonstrates that claudin-5 may represent a novel treatment target for prevention of heart failure.

Early treatment with lisinopril and spironolactone preserves cardiac and skeletal muscle in Duchenne muscular dystrophy mice.

BACKGROUND: Nearly universal cardiomyopathy in Duchenne muscular dystrophy (DMD) contributes to heart failure and death. Because DMD patients show myocardial fibrosis well before functional impairment, we postulated that earlier treatment using drugs with antifibrotic effect may be beneficial. METHODS AND RESULTS: Three groups of 10 utrn(+/-);mdx, or "het" mice, deficient for dystrophin and haploinsufficient for utrophin with skeletal myopathy and cardiomyopathy that closely mimics clinical DMD were studied. One het group received spironolactone and lisinopril starting at 8 weeks of life (het-treated-8); a second received the same starting at 4 weeks of life (het-treated-4), and the third het group was untreated. At 20 weeks, all mice had normal ejection fractions though circumferential strain rate was abnormal (-0.21±0.08) in untreated hets. This improved to -0.40±0.07 in het-treated-8 mice (P=0.003) and further improved to -0.56±0.10 in het-treated-4 mice (P=0.014 for het-treated-4 versus het-treated-8). Treated mice showed less cardiomyocyte damage, with a 44% reduction in intracardiomyocyte serum immunoglobulin G localization in het-treated-8 mice (P<0.0001) and a further 53% reduction in het-treated-4 mice (P=0.0003 versus het-treated-8); matrix metalloproteinases were similarly reduced. Cardiac, limb, and diaphragm function by ex vivo muscle testing remained at 80% of normal with early treatment compared to a decline to 40% of normal skeletal muscle function without treatment. CONCLUSIONS: These findings offer clinically available medications with proven antifibrotic effect as a new therapeutic strategy in DMD. Early initiation greatly attenuated myocardial disease and, for the first time with these drugs, improved skeletal myopathy. Thus, early initiation of such agents warrants further clinical evaluation to maintain ambulatory, respiratory, and cardiac function for patients with DMD and related myopathies.

Claudin-5 levels are reduced in human end-stage cardiomyopathy.

Claudin-5 is a transmembrane cell junction protein that is a component of tight junctions in endothelial cell layers. We have previously shown that claudin-5 also localizes to lateral membranes of murine cardiomyocytes at their junction with the extracellular matrix. Claudin-5 levels are specifically reduced in myocytes from a mouse model of muscular dystrophy with cardiomyopathy. To establish whether claudin-5 is similarly specifically reduced in human cardiomyopathy, we compared the levels of claudin-5 with other cell junction proteins in 62 cardiomyopathic end-stage explant samples. We show that claudin-5 levels are reduced in at least 60% of patient samples compared with non-failing controls. Importantly, claudin-5 reductions can be independent of connexin-43, a gap junction protein previously reported to be reduced in failing heart samples. Other cell junction proteins including alpha-catenin, beta-catenin, gamma-catenin, desmoplakin, and N-cadherin are reduced in only a small number of failing samples and only in combination with reduced claudin-5 or connexin-43 levels. We also show that reduced claudin-5 levels can be present independently from dystrophin alterations, which are known to be capable of causing and resulting from cardiomyopathy. These data are the first to show alterations of a tight junction protein in human cardiomyopathy samples and suggest that claudin-5 may participate in novel mechanisms in the pathway to end-stage heart failure.

Glutamate receptors localize postsynaptically at neuromuscular junctions in mice.

Dlg (Discs Large) is a multidomain protein that interacts with glutamate receptors and potassium channels at Drosophila neuromuscular junctions (NMJs) and at mammalian central nervous system synapses. Dlg also localizes postsynaptically at cholinergic mammalian NMJs. We show here that alpha-amino-3-hydroxy-5-methylisoxazole-4-proprionate (AMPA) receptor subunits, together with glutamate, are present at the mammalian NMJ. Both AMPA and NMDA (N-methyl-D-aspartate) glutamate receptor subunits display overlapping postsynaptic localization patterns with Dlg at all NMJs examined in normal mice. Kir2 potassium channels also localize with Dlg and glutamate receptors at this synapse. Localization of the components of a glutamatergic system suggests novel mechanisms at mammalian neuromuscular synapses.

Truncated CASK does not alter skeletal muscle or protein interactors.

CASK (Ca2+, calmodulin-associated serine/threonine kinase) is an essential mammalian cell junction protein and is also crucial at Drosophila neuromuscular synapses. We have shown that CASK is present in mammalian skeletal muscle at the postsynaptic membrane of the neuromuscular junction. CASK interacts biochemically with channels at central synapses, and studies in cultured cells have led to proposed functions for CASK. However, in vivo functions of CASK in skeletal muscle remain unknown. To test hypotheses of CASK functions, we generated two lines of transgenic mice, which overexpress full-length and truncated CASK protein in skeletal muscle. Extensive analyses showed that overexpression of CASK protein did not affect the morphology or physiology of skeletal muscle, the morphology of the neuromuscular junction, or the levels or distribution of protein interactors. These results contrast with previous cell culture experiments and emphasize the importance of in vivo analysis of protein function.

CASK localizes to nuclei in developing skeletal muscle and motor neuron culture models and is agrin-independent.

Membrane-associated guanylate kinases (MAGUKs) are cytoplasmic multi-domain proteins that serve as scaffold proteins at cell junctions and synapses. Calmodulin-associated serine/threonine kinase (CASK) stabilizes the integrity of synapses in the brain. Additionally, CASK is capable of acting as a transcriptional co-activator and localizes to neuronal nuclei in the developing brain. We have recently described CASK localization to both the pre- and post-synaptic membranes of the neuromuscular junction (NMJ), where it forms a complex with discs large (Dlg). CASK also localizes to some, but not all nuclei in adult mouse skeletal muscle. To begin to dissect the roles of CASK in the cellular components of the NMJ, we investigated the localization of CASK during differentiation in cell culture models of skeletal muscle and motor neurons. We demonstrate that CASK localizes to the nucleus in undifferentiated myoblasts, but is pre-dominantly in the cytoplasm in differentiated myotubes of the C2C12 myogenic cell line. We also show nuclear localization of both CASK and Dlg in a motor neuron-neuroblastoma hybrid cell line, MN-1, suggesting a role for CASK and Dlg in nuclei of neurons in the peripheral nervous system. Finally, we demonstrate that CASK and Dlg do not co-cluster with acetylcholine receptors in C2C12 myotubes in response to agrin or laminin treatment, suggesting a novel mechanism of recruitment to the NMJ that is independent of acetylcholine receptor and utrophin complexes. These studies delineate important developmental characteristics of CASK and Dlg, and suggest dual roles for these proteins in both the skeletal muscle and motor neuron components of the NMJ.

Utrophin deficiency worsens cardiac contractile dysfunction present in dystrophin-deficient mdx mice.

The loss of dystrophin in patients with Duchenne muscular dystrophy (DMD) causes devastating skeletal muscle degeneration and cardiomyopathy. Dystrophin-deficient (mdx) mice have a much milder phenotype, whereas double knockout (DKO) mice lacking both dystrophin and its homolog, utrophin, exhibit the clinical signs observed in DMD patients. We have previously shown that DKO and mdx mice have similar severities of histological features of cardiomyopathy, but no contractile functional measurements of DKO heart have ever been carried out. To investigate whether DKO mice display cardiac dysfunction at the tissue level, contractile response of the myocardium was tested in small, unbranched, ultrathin, right ventricular muscles. Under near physiological conditions, peak isometric active developed tension (F(dev), in mN/mm2) at a stimulation frequency of 4 Hz was depressed in DKO mice (15.3 +/- 3.7, n = 8) compared with mdx mice (24.2 +/- 5.4, n = 7), which in turn were depressed compared with wild-type (WT) control mice (33.2 +/- 4.5, n = 7). This reduced Fdev was also observed at frequencies within the murine physiological range; at 12 Hz, Fdev was (in mN/mm2) 11.4 +/- 1.8 in DKO, 14.5 +/- 4.2 in mdx, and 28.8 +/- 5.4 in WT mice. The depression of Fdev was observed over the entire frequency range of 4-14 Hz and was significant between DKO versus mdx mice, as well as between DKO or mdx mice versus WT mice. Under beta-adrenergic stimulation (1 micromol/l isoproterenol), Fdev in DKO preparations was only (in mN/mm2) 14.7 +/- 5.1 compared with 30.9 +/- 8.9 in mdx and 41.0 +/- 4.9 in WT mice. These data show that cardiac contractile dysfunction of mdx mice is generally worsened in mice also lacking utrophin.

Claudin-5 localizes to the lateral membranes of cardiomyocytes and is altered in utrophin/dystrophin-deficient cardiomyopathic mice.

Remodeling of adherens junction, gap junction, and desmosomal proteins at the intercalated discs of cardiomyocytes in heart characterizes several animal models of cardiomyopathy, especially dilated cardiac myopathy (DCM). In this study, we show that the tight junction protein, claudin-5, is present in cardiac muscle and localizes to the lateral membranes of cardiomyocytes in normal mice. We further examined claudin-5 in utrophin/dystrophin-deficient (double knockout, dko) mice, a mouse model of muscular dystrophy with cardiomyopathy, and found that claudin-5 mRNA and protein levels are decreased in dko hearts as compared with normal. Intercalated disc cell junction proteins, and another tight junction protein, zonula occludens-1 (ZO-1), are not altered in the dko mouse. Ultrastructural data from dko hearts also shows that the lateral membranes of cardiomyocytes exhibit an abnormal wavy appearance. These data demonstrate that claudin-5 is specifically altered in dko hearts, suggesting that alterations of the lateral membranes of cardiomyocytes, rather than intercalated discs, are associated with cardiomyopathy in the dko mouse.

CASK and Dlg form a PDZ protein complex at the mammalian neuromuscular junction.

Membrane-associated guanylate kinases (MAGUKs) are modular adapter proteins that serve as scaffolding molecules and anchor channels and receptors via their PDZ (PSD-95, Dlg, Zo-1) domains. Calcium, calmodulin-associated serine/threonine kinase (CASK) is a MAGUK that is critical at synapses in the central nervous system and at cell-cell junctions because of its interactions with channels, receptors, and structural proteins. We show via confocal microscopy that CASK and another MAGUK, Discs Large (Dlg), are present at the mammalian neuromuscular junction in skeletal muscle. Immunoprecipitation data from mouse muscle show that CASK associates with Dlg, providing evidence of a MAGUK protein complex at this synapse. These data indicate that CASK and Dlg may act as a scaffold for organizing receptors and channels at the postsynaptic membrane of the neuromuscular junction.