The specific ubiquitin – Proteasome system in the sarcomere
The ubiquitin-proteasome system (UPS) recognizes specific proteins and target them for degradation. Ubiquitin ligases recognize proteins to be degraded and interact with E1 (activating) and E2 (conjugating) enzymes to create poly-ubiquitin chains on the substrate. Poly-ubiquitin chains are then recognized by the 20S proteasome prior to degradation. Several ubiquitin ligases integrated to the sarcomere have been identified: the MuRF family proteins (MuRF1, MuRF2 and MuRF3), MAFbx / atrogin-1 and MDM2.
MuRF1 is found mainly in the M-line of the sarcomere where it interacts with the giant protein titin. MuRF1 specifically recognizes and degrades troponin I. MuRF1 and MuRF2 are reported to interact with troponin T, myosin light chain 2, myotilin and telethonin. While single deletion of either MuRF1 or MuRF2 allows a normal development, cardiac hypertrophy develop in mice lacking both genes (Witt et al., 2008). MuRF1 and MuRF3 also interact together with the E2 enzyme to degrade P / slow myosin heavy chains in the heart. Mice lacking both proteins develop a hypertrophic cardiomyopathy and skeletal muscle myopathy (Fielitz et al., 2007). MuRF1 may preferentially poly-ubiquitinate oxydized proteins as previous results suggest it (Zhao et al., 2007). Therefore, MuRF1, in cooperation with MuRF2 and MuRF3 may ensure protein quality control by detecting damaged proteins, to allow continuous optimal functions of the cardiac sarcomere (reviewed in Willis et al., 2009).
Additional ubiquitin ligases are also playing important roles in sarcomere maintenance: CHIP, which plays a role as co-chaperone of Hsp70 and as ubiquitin ligase has been recognized as an important factor involved in ischemia / reperfusion injury in the heart (Zhang et al., 2005).
MAFbx / atrogin-1 was first identified as an ubiquitin ligase involved in skeletal muscle atrophy (Bodine et al., 2001), possibly by regulating cardiac hypertrophy genes (Li et al., 2007).
MDM-2 is a critical regulator of apoptosis through its ubiquitin-dependent degradation of ARC (Apoptosis Repressor with Caspase recruitment domain). It interacts and mediates telethonin degradation in a proteasome-dependent manner (Tian et al., 2006). Its specific role in the maintenance of sarcomere has to be further explored.
Calpain degradation
The highly integrative organization of the sarcomere does not allow direct degradation of the integrated proteins. Predigestion by proteases such as calpains which are embedded in the sarcomeric structure is therefore required. Calpains are a group of calcium-dependent, non-lysosomal cystein proteases expressed ubiquitously in all cells. Calpains are involved in a variety of cellular processes such as cell-cycle control and cell fusion (Goll et al., 2003). Calpain 1 has been found tightly associated to titin in a calcium-dependent manner. It is required to allow the dissociation of sarcomere proteins from the assembled myofibrils before the ubiquitin proteasome system is able to degrade them (Dargelos et al., 2008; Jackman & Kandarian, 2004). When calpains are inhibited in the heart, protein aggregation occur, ubiquitin ligases such as MuRF1 and MAFbx / atrogin-1 are no longer efficient in mediating proteasome-dependent degradation, and autophagy is increased. However, increased levels of calpain 1 in cardiomyocytes of transgenic models lead to cell lysis, cardiac hypertrophy, inflammation and ultimately heart failure (Galvez et al., 2007). These findings are consistent with a critical role of the calpain system in protein quality control in the heart.
Sarcomere maintenance and autophagy
Autophagy is a process for protein degradation that uses lysosomal hydrolases working at acidic pH. Autophagy begins with the formation of isolation membrane (phagophore). The phagophore then elongates and engulfs a portion of the cytoplasm to form a mature autophagosome, which then fuses with lysosomes to form autolysosomes. Autophagy removes damaged organelles such as mitochondria and aggregates of misfolded or damaged proteins (Beau et al., 2011). Autophagy-mediated degradation also contributes to the maintenance of the sarcomere (Portbury et al., 2011). In cardiomyocytes, autophagy occurs at the basal level and can be further induced by physiological or pathological conditions such as starvation, haemodynamic stress, ischemia / reperfusion, proteotoxicity, and toxins (Gustafsson & Gottlieb, 2008). Inhibition of autophagy leads to global disorganization of the sarcomere, mitochondrial aggregation, with ventricular dilation, cardiac hypertrophy and contractile dysfunction in adult animal models (Nakai et al., 2007). When autophagy is inactivated, there is an increase in poly-ubiquitinated proteins, as well as in proteasome activity, which suggest that despite its compensatory increase in activity, the UPS may be rapidely overwhelmed by the accumulation of toxic proteins (Bennett et al., 2005). Conversely, inhibition of proteasome activity leads to the accumulation of poly-ubiquitinated proteins and activation of autophagy (Tannous et al., 2008b). These results suggest, therefore, an essential role of cooperativeness between UPS and autophagy in maintaining protein quality control in the heart. Cell signaling pathways, such as PI3K / Akt, and transcription factors like FOXO3 have just begun to be involved as essential mediators in coordinating the proteasomal and lysosomal systems. In addition, sHSP and co-chaperones, like the Hsp22 – BAG3 complex activate autophagy, while BAG1 / BAG3 ratio is important to the balance between proteasomal to autophagic degradation (reviewed in Willis et al., 2009).
Physiological signaling and stress signals
An essential component of cardiac signaling is a process termed mechanotransduction, which translates a mechanical stress into a transcriptional response, and may constitute one of the most important stimuli leading to cardiac hypertrophy. Z-disc-associated proteins appear to play a critical role in this process (Frank et al., 2008). Mechanosignaling results in increased rate of protein synthesis, alteration of cell shape and increased expression of genes that are normally expressed predominantly during fetal life. Nodal points of mechanotransduction are found along the cardiac sarcomere, notably in the Z-disc, I-band and M-band regions (reviewed in Frank et al., 2006). It has been found that separate directional pathways are implicated by static transverse and longitudinal forces applied to cardiomyocytes to activate distinct cell signaling pathways. The mains involved are focal adhesion kinases (FAK), proteins kinase C (PKC) and integrins:
FAK is the primary effector of integrin signaling, and is localized to costameres in myocytes (Tornatore et al., 2011). Mechanical stress leads to disruption of FAK interaction with myosin heavy chain at the A-band to Z-disc, costameres and nuclei. Cyclic stretch induces a FAK-mediated activation of JNK and c-jun as well as of MEF2 (Nadruz et al., 2005). FAK appears therefore to play a critical function in hypertrophy, triggered in biomechanical as well as pharmacological models.
PKCe is a modulator of cardiac hypertrophy that belongs to the groups of "unconventional" PKCs which do not require Ca2+ for their activation (Dorn & Force, 2005). PKCe translocate to the Z-disc in cardiomyocytes upon stimulation, in particular by mechanical stress, or in pressure overload (Gu & Bishop, 1994). PKCe is necessary for cardiac hypertrophy induced by G protein receptors coupled agonists (Iwata et al., 2005). CapZ also plays a role in PKC signaling, which subsequent effects on cardiac contractility. Anchors to the Z-disc for PKCe are also provided by ZASP which links them to a-actinin through binding either to PDZ or LIM domains.
Other molecules, including PDE5A (PhosphoDiEsterase 5A), PCAF (P300/CBP Associated Factor), Zyxin, myopodin and HDAC4, ArgBP2, localize to the Z-disc, A-bands and I-bands of myocardial tissue, were found to shuttle between the sarcomere and the nucleus, and modulate the signaling pathways involved in cardiomyopathies (rewieved in Frank & Frey, 2011).
Strains on cultured cardiomyocytes also increases FAK activity and PKCe, which lead to the activation of the Rho/ROCK GTPases/kinase pathway, for which substantial evidences support a role in myofibrillogenesis (Franchini et al., 2000; Torsoni et al., 2003). One of the target of PKCe is the muscle actin capping protein for the barbed end of the actin filament (Schafer et al., 1994). Myocyte contractility, through titin and T-cap may also regulate muscle LIM proteins (MLP) shuttling to and from the nucleus, that may play a further role in myocyte remodeling and hypertrophy, and is required for adaptation to hypertrophic stimuli (Iwata et al., 2005).
Integrins are transmembrane proteins which transduce signals from the extracellular matrix to the inner cell space and conversely. In the myocardium, integrin signalling plays an important role in mediating hypertrophic signals converging on the kinases Erk 1/2, PKC (Heidkamp et al., 2003), p38 MAPK (Aikawa et al., 2002) or JNK (Zhang et al., 2003). Among the proteins binding to integrins upon signaling, were identified the Integrin-Linked Kinase (ILK), FAK and Melusin.
ILK is a serine/threonine kinase. Its specific cardiac deletion in mice leads to dilated cardiomyopathy (DCM) and sudden death, while cardiac-restricted overexpression of ILK induces cardiac hypertrophy via activation of Erk 1/2 and p38 MAPK, indicating a function in hypertrophic signaling (Bendig et al., 2006; Lu et al., 2006).
Melusin is transiently upregulated in the heart upon pressure overload. Deletion of the Melusin gene show DCM and contractile dysfunction upon stress only, while overexpression leads to pathological hypertrophy (Brancaccio et al., 2003; De Acetis et al., 2005). The Calcineurin / NFAT pathway is modulated by proteins like CIB1/ MLP/Calsarcin/LMDC1 that play a pivotal role in the mediation of pathologic cardiac hypertrophy (Frey et al., 2004). Calcineurin is linked to the Z-disc via calsarcin and MLP, and directly binds to the L-Type Calcium Channel, the major mediator of calcium influx in cardiomyocytes. Calcineurin activates by dephosphorylation the NFAT (Nuclear Factor of Activated T cells) transcription factors family, one of the major mediators of cardiac hypertrophy and remodelling.
PKA and PKG (cGMP-dependent protein kinase-G) are bound to titin. PKA is stimulated by P-adrenergic stimulation by catecholamines, while PKG is stimulated by nitric oxide and the natriuretic peptide (Wong & Fiscus, 2010). PKA can phosphorylate troponin-I, myosin-binding protein C and titin (Yamasaki et al., 2002; Yang et al., 2001; Zakhary et al., 1999).
From healthy to failing heart : Etiology and studies of myofibrilar cardiomyopathies
Diagnosis
As MFM refers to a group of genetically distinct disorders, common morphologic features observed on muscles are determinant. The following clinical findings should direct the diagnosis to a myofibrillar myopathy.
Clinical signs
Most patients with MFM show progressive muscle weakness. A small proportion of patients show paresthesias, muscle atrophy, stiffness or aching, cramps, dyspnea or dysphagia, or mild facial weakness. In about one third of the cases, the weakness is predominantly distal, in another third it is more proximal than distal, and in the other third it is mixed. In some patients, however, the cardiomyopathy may precede the muscle weakness. Cardiomyopathy is a "classical" feature in MFM and may precede, coincide or follow the skeletal muscle weakness. Cardiomyopathy includes arrhythmogenic type (with atrio-ventricular blocks, supraventricular and ventricular ectopic beats and tachycardia), hypertrophic, dilated or restrictive features (reviewed in Ferrer & Olive, 2008; Goldfarb & Dalakas, 2009; Schroder & Schoser, 2009; Selcen, 2011).
Serum creatine kinase (CK) is variable, sometimes elevated. Differences depending on the causal gene have been reported (Schroder & Schoser, 2009):
DES, CRYAB and MYOT: normal up to 5 fold (reported maximal 15 fold for MYOT) FLNC: normal up to 10 fold ZASP: normal up to 6 fold BAG3: 3 up to 15 fold
Histopathology
Muscle histology is essential and reveals characteristic features:
- Amorphous, hyaline or granular materials stained by trichrome.
- hyaline structures intensely positive for Congo red staining.
- significant decrease of oxidative enzyme activity in many abnormal fibers regions.
- small vacuoles in a variable number of fibers.
Immunohistochemistry performed on frozen sections (paraformaledehyde-fixed (4%)) from biopsies show abnormal ectopic expression of desmin, aB-crystallin, myotilin and dystrophin (Claeys et al., 2009; Schroder & Schoser, 2009; Selcen, 2011).
Electromyography
Electromyography (EMG) should be performed to confirm the histopathological data. EMG reveals abnormal electrical irritability often with myotonic discharges. The motor unit potentials are mostly myopathic, sometimes in combination with neurogenic features (Schroder & Schoser, 2009).
Electron microscopy
Electron microscopy of muscles showing progressive myofibrillar degeneration reveal abnormalities starting from the Z-disc area: streaming, defects in stacking, accumulation of granulo-filamentous dense materials, sarcomere disintegration, dislocated membranous materials, autophagic vacuoles, abnormal accumulation or location of mitochondria. Ultrastructural observation may allow to differentiate between granulofilamentous accumulation, "sandwich" formation, filamentous bundles, floccular thin filaments, tubular filamentous accumulations, and early apoptotic changes, to allow to direct diagnostic efforts toward the type of gene causing the MFM (Claeys et al., 2008).
Clinical perspectives and therapeutic considerations
There is currently no specific treatment for myofibrillar myopathies, nor clinical trial investigations (to the best of our knowledge).
MFM share protein aggregation with other aggregate-prone diseases, such as Alzheimer’s, Parkinson’s and Huntington’s diseases. Accumulation of toxic P-amyloid oligomers, impairment of the ubiquitin proteasome system, alteration of the efficiency of the autophagic process have also been reported in studies about these diseases (Aguzzi & O’Connor, 2010). Several treatments have been set up and are currently tested for clinical trials, and may possibly be studied in the case of myofibrillar myopathies. However, caution should be taken because reducing the formation of aggregates may not directly allow to the cure of the disease (Sanbe et al., 2005).
Curcumin, a polyphenol naturally occurring in plant products, has been shown to inhibit a-synuclein aggregates formation involved in Parkinson’s disease (Pandey et al., 2008).
Another way to reduce aggregates formation is to activate HSP, which may favour refolding or degradation of mutant proteins, or folding of the WT proteins expressed in the normal allele. Non-steroidal anti-inflammatory drugs, prostaglandins (Amici et al., 1992), Celastrol (Westerheide et al., 2004) or Geranylgeranylacetone (Sanbe et al., 2009) have been shown to activate the heat shock response. However, it is probably not judicious to induce aB-crystallin if this gene is mutated.
Activators of the autophagic pathway of degradation have been reported and could constitute a way to clear aggregates from the cardiomyocytes. Among many compounds actually tested, one can cite starvation, rapamycin, wortmanin, trehalose, etc (Sarkar & Rubinsztein, 2008).
Antioxydant agents, such as vitamin C, N-acetyl cysteine, ROS trapping agents like phenyl-N-tert-butylnitrone may also help to reduce aggregates formation, or help to their degradation (Squier, 2001).
Conclusion
There has been considerable improvement in the understanding of myofibrillar myopathies since they were first reported in 1978 (Fardeau et al., 1978). New avenues of discoveries have just opened with concepts of protein quality control, involving chaperone molecules, the ubiquitine proteasome system, and macroautophagy. Moreover, the sarcomere is not only seen now as a passive force generator, but also as a mechanosensor which signals to the nucleus and to the whole muscular fiber, and activates specific programs of renewal of sarcomeric proteins, in case of damages. These new findings will help to integrate more components of the muscle physiology, and to understand how they are modified when MFM-causing genes are mutated. However, only 20 % of MFM have found a "culprit" gene, so it may be necessary to analyze other Z-disc-specific structural, quality control or signaling genes to find new candidates responsible for MFMs.
