Stefano Schiaffino
Gene regulation and cell signalling in skeletal muscle

Our studies focus on the signaling pathways that control muscle gene regulation, with particular reference to the mechanisms that regulate muscle fiber type specification and fiber size. These studies have important clinical implications. Targeting intracellular signaling pathways that control muscle growth and fiber type profile can be used as a therapeutic strategy for preventing muscle wasting and loss of muscle force during ageing and for treating neuromuscular diseases, such as muscular dystrophies, as well as other diseases, such as type II diabetes.

We have been involved for many years in the study of the different fiber types present in mammalian skeletal muscle (see Schiaffino & Reggiani, Physiol Rev 76:371, 1996). A major contribution has been the identification of a novel fiber type, called type 2X, characterized by a specific myosin heavy chain isoformencoded by a distinct gene (De Nardi et al, J Cell Biol, 123:823, 1993). To define the signaling mechanisms and transcription factors involved in fiber type specification, we took advantage of a simple procedure developed in our lab for in vivo transfection in regenerating skeletal muscle by intramuscular injection of plasmid DNA (Vitadello et al, Hum Gene Ther 5:11, 1994). Constitutively active or dominant negative mutants of various signal transducers were transfected in vivo to produce selective perturbations of signaling pathways. Using this approach, we have explored the mechanisms responsible for muscle atrophy and hypertrophy. We previously reported that a Ras double mutant that selectively stimulates the PI3K-PKB/Akt pathway is able to induce hypertrophy of regenerating muscle fibers (Murgia et al, Nature Cell Biol 2:142, 2000). Constitutively active mutants of PKB/Akt also induce muscle hypertrophy, while rapamycin, a specific inhibitor of mTOR, a downstream effector of PKB/Akt, blocks regenerating muscle growth (Pallafacchina et al, PNAS 99:9213, 2002).
More recently, we explored the role of another target of Akt, the FoxO transcription factors, in muscle atrophy. Akt inhibits transcriptional activation by FoxO factors, causing their displacement from the nucleus into the cytoplasm. We asked whether Akt is involved in the prevention of muscle atrophy and whether FoxO is mediating this effect. Muscle atrophy is due to increased protein degradation via the ubiquitin-proteasome pathway and is accompanied by the induction of muscle specific ubiquitin ligases, atrogin-1/MAFbx and MURF1, that are responsible for muscle atrophy due to inactivity or cachexia. In collaboration with A. Goldberg (Harvard Medical School), we have obtained direct evidence that FoxO is a major target of Akt in muscle cells and controls the induction of atrogin-1 (Sandri et al, Cell,117:399, 2004). Conditions leading to muscle atrophy, such as denervation and fasting, cause nuclear translocation of FoxO and activation of FoxO transcriptional activity. In contrast, PKB/Akt causes FoxO nuclear export and down-regulation of FoxO-dependent reporters. The role of FoxO in mediating the atrophic process was demonstrated by the finding that a constitutively active FoxO3 causes atrogin-1 expression and muscle atrophy (Figure 1), while siRNA-mediated inhibition of Foxo1-3 inhibits atrogin-1 promoter activity.



Figure 1: Constitutively active FoxO3 mutant induces muscle atrophy. Rat skeletal muscle were transfected with a tagged caFoxO3 and examined 15 days later. Sections stained for the tag show that muscle fibers expressing this mutant are much smaller in size compared to untransfected surrounding fibers.




Similar approaches were used to dissect the pathways involved in nerve activity-dependent muscle fiber type specification. We previously found that Ras-ERK signaling is involved in the activation of the "slow" muscle gene program induced by slow motor neurons in regenerating muscle (Murgia et al, Nature Cell Biol 2:142, 2000). The calcineurin pathway is also involved in this process, as shown by the finding that the "slow" muscle gene program is blocked by the calcineurin inhibitors cyclosporin A, FK506 and cain/cabin1 (Serrano et al, PNAS 98:13108, 2001; Schiaffino & Serrano, Trends Pharmacol Sci 23:569, 2002).
More recently, we explored the role of the transcription factor NFAT, a major target of calcineurin, in fiber type specification using multiple approaches: i) NFAT-GFP fusion proteins to visualize NFAT nuclear translocation (Figure 2), ii) NFAT-dependent reporters to monitor NFAT transcriptional activity, iii) specific peptide inhibitors to block calcineurin-mediated NFAT activation, and iv) constitutively active NFAT mutants to stimulate NFAT activity. These studies showed that NFAT is a nerve activity sensor in skeletal muscle and controls activity-dependent fiber type switching, specifically the induction and maintenance of the "slow" gene program (McCullagh et al, PNAS 101:10590).







Figure 2: Nuclear translocation of NFATc1-GFP fusion protein is induced in vivo by "slow-like" (20 Hz, right panel) but not "fast-like" (100 Hz, left panel) electrostimulation in mouse tibialis anterior muscle.


The objective of ongoing studies is to determine how the various transduction pathways are coupled to the electrical and mechanical signals that accompany nerve-dependent muscle activity and how specific transcription factors control complex programs of muscle gene regulation.
A second aim is to explore whether and how experimental manipulation of these signaling pathways can be used to treat pathological conditions, such as muscle atrophy, muscular dystrophies and type II diabetes.