The objective of this study was to investigate the effects of cigarette smoke exposures on skeletal muscle cell signaling involved in protein synthesis and breakdown, and to assess the reversibility of these effects following smoking cessation. This study emphasizes three key messages. First, the cigarette smoke exposure results in reduced skeletal muscle weight and increases in expression of pro-atrophy related genes. Second, intensity of smoke-induced cell signaling events is increased alongside the total cigarette exposure period. Third, cell signaling returns to control levels when cigarette smoke exposure is stopped for a prolonged period of time.
Mice were exposed to cigarette smoke using a well-characterized whole body smoke exposure system 
. We previously reported that cigarette smoke exposure is well tolerated and results in cotinine and COHb levels similar to what has been found in human smokers 
. Cigarette smoke exposure elicits an inflammatory response in the lungs that is characterized by increased mononuclear cells and neutrophils. Mechanistically, this inflammatory response is IL-1α/IL-1R1 dependent 
, but redundant of TNF-α (unpublished data). This model also exhibits similar levels of physical activity, as well as equivalent food and water intake compared to room-air conditions (Debigaré et al, unpublished data).
Despite a decreased body weight after 8 weeks, mice exposed to cigarette smoke exhibited a 5.4% body weight gain between week 8 and week 24. Comparatively, body weight gain was 3.8% in room air-exposed mice over the same period. Mice were not weighed at time zero. Nevertheless, mice were matched for age and gender, and female Balb/c mice of similar age purchased from the supplier (Charles Rivers) typically average 17–18 grams. Therefore, the reported body weight difference is probably the consequence of a failure to gain weight normally during the first weeks of cigarette smoke exposure. The fact that tibia length was similar in smoke-exposed mice and in controls after 24 weeks reveals that the overall skeletal growth was comparable in both conditions. Consequently, the discrepancies in body weight originate from modifications in body composition (i.e. fat versus fat-free mass). Unfortunately, our data do not allow distinction between fat and fat-free mass distribution. Body composition measurements are therefore needed to further document the impact of cigarette smoke exposures on the dynamic and compartmentalized nature of body weight loss in this model.
The effects of cigarette smoke exposures on muscle mass were different relative to the muscle group being assessed. Gastrocnemius, a predominantly glycolytic muscle, was significantly smaller in 8-week smoke exposed mice when compared to their counterparts, whereas only a tendency toward significance was found between both groups after 24 weeks (p
0.055). Conflicting reports exist concerning the effects of cigarette smoke exposure on gastrocnemius weight 
. This discrepancy is most probably related to the methodological divergences that exist in the smoke exposure protocols currently used in the literature; e.g. nose-only versus whole body exposure, number of daily exposures, number of days/week, total duration of the protocols, type of cigarettes, animal gender, etc. The fact that gastrocnemius mass was more severely impaired after an 8-week smoke exposure than after 24 weeks (10.8% versus 6.5% decreases) is intriguing. De Paepe reported a type IIa to type IIb fiber switch in gastrocnemius muscle of 18-week smoke-exposed mice 
. Although speculative, this fiber type transition from hybrid to glycolytic fibers could be the result of an adaptive period where muscle growth is altered, perhaps during the first 8 weeks of the protocol. Following this period, the adapted (more glycolytic) gastrocnemius muscle would then gain mass at a similar rate to non-smoking mice, as seen during the 8 to 24 weeks interval in our study. The soleus, a phasic and predominantly oxidative muscle, was significantly smaller in smoke-exposed mice after 24 weeks. This result confirms previous studies reporting a reduction in soleus weight after 18 weeks 
and 24 weeks 
of smoke exposure, although that latter study only reported a strong statistical tendency (p<0.06). It should be noted that nose-only smoke exposures failed to produce soleus muscle weight reduction in 3 and 6 months protocols 
. With our data, it is impossible to conclude whether the reduction in muscle weight represents atrophy (specific loss of mass and fibers) or failure to gain weight. However, both atrophy and failure to gain weight are likely responses to increased expression of pro-atrophy genes due to smoke exposure. Taken together, our results on muscle weight highlight a differential effect of cigarette smoke exposure on muscle groups, a variance that could be explained by myofibrillar composition or muscle function (postural versus mobility). Specific investigations are required to explore this notion.
When mice were exposed to cigarette smoke for 8 weeks, pro-atrophic genes (i.e. Atrogin-1, MuRF1 and FoxO3) 
exhibited higher mRNA expression levels when compared to controls, despite the fact that an emphysema-like phenotype and airway remodeling are observed only after prolonged exposure, usually following 4–6 months of cigarette smoke exposure 
. The effects of cigarette smoke exposure on cell signaling were exacerbated in a 24-week protocol where, in addition to gene expression alterations, intra-muscular protein levels and phosphorylation status of members of the PI3K/Akt and ubiquitin-proteasome pathways were altered. We observed hypophosphorylation of Akt when mice were exposed to cigarette smoke for 24 weeks. Given the central role of Akt in the control of global protein synthesis 
, these results intuitively point toward an altered synthesis process. To clarify this assertion, we analyzed direct and indirect downstream targets of Akt (i.e. GSK-3β and p70S6K respectively) in the same group of mice and found no difference in the phosphorylation ratio of both proteins. However, the total form of GSK-β, an inhibitor of protein translation when unphosphorylated 
, was reduced in smoke-exposed mice. Because the phospho-GSK-3β/total GSK-3β ratio was unaltered in smoke-exposed animals, the net result of the decreased total GSK-3β is a reduction in both phosphorylated and unphosphorylated forms of this protein, suggesting that protein synthesis could be favored in the gastrocnemius (i.e less unphosphorylated GSK-3β). Since Akt and p70S6K phosphorylation status were either decreased or unaltered in smoke exposed mice, we believe that restoration of the protein synthesis process to a level comparable to non-smokers is unlikely with the sole reduction in total GSK-β. Given the fact that protein synthesis rates were not directly measured in this study, these assumptions are speculative and remain to be proven. After 24 weeks of exposure, Akt hypophosphorylation was associated with an altered control of the degradation process, as depicted by increased MuRF1, Atrogin-1, and FoxO3 mRNA levels, as well as MuRF1 and polyubiquitin protein content upregulation. Comparatively, at week 8, increased MuRF1, Atrogin-1 and FoxO3 mRNA levels were observed alongside decreased phospho-Akt/total Akt ratio and increased total Akt (both not significant). As the resulting situation is unclear, we cannot affirm that changes in Akt activity were related to atrophy gene expression at this time-point. Other transcription factors or signaling pathways are likely involved. The fact that gastrocnemius was able to gain weight at a normal rate between week 8 and 24 in smoke-exposed mice remains intriguing, considering that muscle protein degradation signals were higher in these mice at week 24. These results are compatible with a situation where the protein pool present in muscle cells are either more prone to abnormal folding during the synthesis process, are more sensitive to oxidation, or are vulnerable to damage. These protein alterations are known to activate the ubiquitin-proteasome system. Interestingly, our group 
and others 
have reported preservation of the phosphorylation status of GSK-3β and p70S6K along with higher expression levels of pro-atrophic markers in the vastus lateralis muscle of COPD patients when compared to matching healthy controls. This situation highlights the relevance of our chronic smoke-exposure mouse model in order to understand the skeletal muscle dysfunction occurring in the context of COPD.
Our results indicate that a pro-inflammatory state in muscle tissue was induced by chronic cigarette smoke exposure, at least at the mRNA level. Systemic and local inflammation is among the most frequently cited underlying mechanism that could contribute to the development of skeletal muscle dysfunction in COPD 
. In our hands, evidence of increased protein oxidation in smoke-exposed mice was concurrent with an inflammatory state after 24 weeks of exposure, which is in agreement with other reports 
. In support, it has been shown recently that cigarette smoke exposure in mice elicits changes in cellular redox status in lung tissue 
, as well as accumulation of reactive carbonyls, 4-hydroxy-2-nonenal and malondialdehyde protein adducts in respiratory and limb muscles 
. Since the buildup in oxidized products was seen late in our model, one could speculate that either exhaustion of the antioxidant defense or proteolytic pathways saturation (e.g. proteasome, autophagy) contributed to this situation. Further studies are needed to investigate the dynamic interactions between production, quenching, and degradation of oxidized products 
Perhaps the most striking result of this study is the demonstration of the reversible nature of the smoke-induced skeletal muscle cell signaling perturbations upon smoking cessation. All modifications in regards to protein quantity and phosphorylation status were improved by a smoking cessation period of 60 days. The same phenomenon was observed with mRNA expression levels of all tested genes except IL-1β, a cytokine that remained over-expressed in ex-smoking mice. IL-1β is a pro-inflammatory cytokine known to induce catabolism when incubated with fully differentiated skeletal muscle cells 
. The result obtained with this cytokine could be a direct consequence of the persistent pulmonary inflammation present after smoking cessation (unpublished data) or an inflammatory process taking place in muscle tissue, independently of TNF-α and IL-6. Nevertheless, these hypotheses were untested in this study and remain to be investigated. Interestingly, gastrocnemius and soleus muscle masses displayed different responses to smoking cessation. Gastrocnemius muscle was found to be 8.3% smaller in ex-smoking mice when compared to room air controls, a value very similar to the result obtained when smoke-exposed mice and room air controls at 8 (-10.8%) and 24 weeks (-6.5%) were compared. Cell signaling was similar in gastrocnemius between groups after smoking cessation. Combination of gastrocnemius muscle weight data and cell signaling results strongly suggests that smoking cessation was associated with a normal muscle protein turnover resulting in normalized muscle homeostasis rather than a muscle mass expansion that would translate into a similar muscle mass between controls and ex-smokers. Surprisingly, soleus muscle was able to normalize its weight after the cessation period. Intuitively, one would not expect 10 months old mice to display muscle growth. The reason why soleus was able to gain weight in the smoke-free period may reside in the metabolic profile of this muscle. Despite the fact that soleus exhibits a fiber shift from type IIa to type IIb in smoke-exposed mice 
, this muscle is still largely composed of oxidative fibers under smoking conditions. This oxidative-dominant phenotype perhaps makes this muscle prone to gain weight when the systemic environment normalizes under smoking cessation conditions. Since data on soleus cell signaling is lacking, we cannot speculate whether soleus cell signaling was in an anabolic or catabolic state during the 60-day smoking cessation period. Further study of the soleus is needed to understand the exact mechanisms.
In conclusion, cigarette smoke perturbs cellular signaling in limb muscle tissue in as few as after 8 weeks of smoke exposure. Disruption in cellular signaling persist and broaden as the exposure to cigarette smoke is prolonged, suggesting further alterations in muscle function. Interestingly, almost every muscle signaling modifications examined in this study were reversible following smoking cessation, opening a door for further studies aimed at understanding the relationship between inhaled cigarette smoke and systemic stresses.