br The mechanisms underlying muscle depletion
The mechanisms underlying muscle depletion have been extensively studied in the past, acceleration of protein turnover rates being the main reason for muscle mass depletion . Specifically, myofibrillar protein catabolism occurs, although reduced protein synthesis might play a role as well . As for protein degradation, two main proteolytic systems operate in the skeletal muscle: the ubiquitin–proteasome-dependent pathway (UPS) and macroautophagy (hereafter referred to as autophagy). UPS activation in cancer cachexia has been reported both in preclinical rodent models and in humans with some differences according to tumor stage and type ( and references within). However, pharmacological UPS inhibition (using bortezomib) is unable to prevent muscle wasting in two distinct experimental models of
Autophagy and mitochondria in cancer cachexia 2675
cancer cachexia , and no clinical trial has attempted to directly interfere with proteasomal degradation in cachectic patients.
Also, excessive autophagic degradation has been proposed to play a role in the onset of muscle depletion in cancer cachexia, based on observations in both rodents  and cachectic cancer patients [6,7]. Autophagy modulation is a double-edged sword, since basal activation is required to remove damaged proteins and organelles maintaining muscle quality, while enhanced autophagy flux favors muscle loss. As a consequence, both excessive and defective au-tophagy worsen muscle function and impair muscle mass , suggesting the importance of a finely tuned balance between protein degradation and synthesis. Even in cancer cachexia, an impaired autophagy flux, likely due to overload, has been proposed to occur, while the pharmacological reactivation rescues mus-cle PD 98059 and counteracts cachexia . Those data are in conflict with others showing that the promotion of autophagy worsens muscle wasting. Specifically, the overexpression of the autophagy-regulating TP53INP2/DOR protein triggers muscle wasting in diabetic (streptozotocin-induced) mice . TP53INP2 activates basal autophagy in skeletal muscle directly interacting with LC3 and sustaining the degradation of ubiquitylated proteins. In this regard, muscle-specific TP53INP2 gain and loss of function have been shown to result in atrophy and hypertrophy, respectively .
When debating the role of autophagy in the regulation of muscle mass, a further consideration should be given to the cargo that is selectively degraded and might be different in healthy as compared to diseased conditions. Indeed, the degradation of damaged proteins and dysfunctional organelles is a premise for muscle quality, while the uncontrolled targeting of myofibrillar proteins or mitochondria negatively impact muscle mass with-out any positive effect. Such dysregulation might be the consequence of specific signals depending on inflammation and tumor-related factors, such as the increased expression of ubiquitin-ligases  and mitophagy triggers , both sufficient to cause atrophy.
So far, only indirect evidences on the beneficial or detrimental effect of autophagic degradation in experimental cancer cachexia are available. The present work is aimed at overcoming such limita-tions by directly modulating (blocking or activating) autophagy in order to measure the resulting effects on muscle mass and quality.
Beclin-1 deficiency does not prevent muscle wasting in tumor-bearing mice
To understand whether muscle wasting in cancer cachexia could be prevented by blocking stress-induced (beclin-1-regulated) autophagy, beclin-1 was knocked down in the tibialis anterior (TA) muscle via electroporation of a specific shRNA, leading to a 50% reduction of the protein levels (Fig. 1a). As expected, in C26-bearing mice TA mass was reduced (39%) when electroporated with scramble sequence, while it was moderately pro-tected against loss (24%) in condition of beclin-1 deficiency. Such a difference could not be appreci-ated in non-electroporated muscles such as the gastrocnemius (Fig. 1b).
However, beclin-1 knockdown did not result in improved TA fiber cross-sectional area (CSA; Fig. 1c) and induced p62 accumulation in the muscle of both control and tumor-bearing (TB) animals, in the latter going beyond the already increased expression (Fig.1d). Morphological data were con-firmed biochemically (Fig. 1e), showing that beclin-1 loss-of-function prompted an abnormal p62 increase and beclin-1-independent LC3B accumulation, likely due to a reduced autophagy flux, as previously suggested . Similar results were obtained in a different experimental setting, by blocking autopha-gy through electroporation of a vector harboring a non-phosphorylatable BCL2 AAA knock-in mutant (see Ref.  for details). In this experiment, TA mass was increased by BCL2 AAA expression in control mice (151 ± 12 versus 178 ± 20 mg % i.b.w., p = 0.018), whereas no effect was observed in TB mice (112 ± 27 versus 107 ± 31 mg % i.b.w.). Fiber CSA analysis showed no differences upon BCL2