Formoterol attenuates increased oxidative stress and myosin protein loss in respiratory and limb muscles of cancer cachectic rats

Animal experiments

In vivo measurements in the animals

Food and water were administered ad libitum to the animals for the entire duration of the study. All the animals were maintained at a temperature of 22 ± 2 °C with a regular light-dark cycle (lights were on from 08:00 a.m. to 08:00 p.m.) and had free access to food and water. Body weight was determined in all animals on day 0 and prior to their sacrifice on day 7. Tumor weights were determined in all animals upon sacrifice. The percentage of body weight gain at the end of the period was calculated as follows: [(body weight on day 7 − tumor weight on day 7) − body weight on day 0]/ body weight on day 0 × 100 (Busquets et al., 2004; Busquets et al., 2011; Busquets et al., 2012; Fontes-Oliveira et al., 2013; Fontes-Oliveira et al., 2014; Toledo et al., 2011; Toledo et al., 2016).

Sacrifice and sample collection

On day 7 after tumor transplantation, the animals were weighed and anesthetized through an intraperitoneal injection of 3:1 ketamine/xylazine mixture (Imalgene^® 1000; Rhone Merieux, France and Rompun^®, Bayer AG, Leverkusen, Germany, respectively). In all animals, the pedal and blink reflexes were evaluated in order to verify total anesthesia depth. The diaphragm and gastrocnemius muscles were obtained from all the animals. In all samples, muscle specimens were immediately frozen in liquid nitrogen and subsequently stored at −80 °C. Frozen tissues were used to assess the expression of the target molecular markers (Busquets et al., 2004; Busquets et al., 2011; Busquets et al., 2012; Fontes-Oliveira et al., 2013; Fontes-Oliveira et al., 2014; Toledo et al., 2011; Toledo et al., 2016).

Muscle biology analyses

All the muscle biological experiments were performed in the laboratory at Hospital del Mar- IMIM-Universitat Pompeu Fabra (Barcelona).

Mitochondrial respiratory chain (MRC) complexes: enzyme activities in muscles

Statistical analysis

The normality of the study variables was verified using the Shapiro–Wilk test. Results are presented as mean (standard deviation). The comparisons between the different study groups were analyzed using the one-way analysis of variance (ANOVA), in which Tukey’s post hoc analysis was used to adjust for multiple comparisons. The sample size chosen was based on previous studies (Busquets et al., 2004; Busquets et al., 2011; Busquets et al., 2012; Chacon-Cabrera et al., 2014; Chacon-Cabrera et al., 2015; Chacon-Cabrera et al., 2016; Chacon-Cabrera, Gea & Barreiro, 2016; Fermoselle, Sanchez & Barreiro, 2011; Fermoselle et al., 2013; Fontes-Oliveira et al., 2013; Fontes-Oliveira et al., 2014; Marin-Corral et al., 2010; Salazar-Degracia et al., 2016; Toledo et al., 2011; Toledo et al., 2016) and on assumptions of 80% power to detect an improvement of more than 20% in measured outcomes at a level of significance of P ≤ 0.05. All statistical analyses were performed using the Statistical Package for the Social Sciences (Portable SPSS, PASW statistics 18.0 version for Windows; SPSS Inc., Chicago, IL, USA).

Physiological characteristics of the study animals

As shown in Table 1, at the end of the study period (day 7), cancer-cachexia rats exhibited a reduction in the variables final body weight and body weight gain compared to non-cachexia controls. In cancer-cachexia rats, formoterol treatment significantly improved body weight gain compared to the non-treated cachectic rats (Table 1). The weights of diaphragm and gastrocnemius muscles were significantly smaller in cancer-cachexia rats than in non-cachexia controls, and treatment with formoterol elicited a significant improvement in the weight of these muscles in the cachectic animals (Table 1).

Protein oxidation

Compared to non-cachexia controls, total reactive carbonyls, both HNE- and MDA- protein adducts, and total protein tyrosine nitration levels were increased only in the limb muscle of the cancer-cachexia rats (Figs. 1A–1D and Figs. S1–S4). Treatment with formoterol attenuated the rise in levels of total reactive carbonyls, HNE- and MDA-protein adducts, and protein tyrosine nitration in the gastrocnemius of the cancer-cachexia rats (Figs. 1A–1D and Figs. S1–S4). No significant differences were seen in any of these markers in respiratory or limb muscles between non-cachexia controls and non-cachexia control-formoterol animals (Figs. 1A–1D and Figs. S1–S4). Several glycolytic enzymes, albumin, actin and tropomyosin were shown to be carbonylated in gastrocnemius muscles of the study groups (Table 2 and Fig. S5).

Antioxidants

SOD1 levels were significantly higher in the limb muscle of cancer-cachexia rats than those detected in non-cachexia control animals (Fig. 2A and Fig. S6). SOD2 protein levels did not significantly differ between cancer-cachexia and the non-cachexia control rats (Fig. 2B and Fig. S7). Catalase protein content significantly increased in limb muscles of cancer-cachexia rats compared to non-cachexia controls (Fig. 2C and Fig. S8). In cancer-cachexia rats, treatment with formoterol did not significantly modify protein SOD1, SOD2, or catalase levels in either diaphragm or gastrocnemius muscles (Figs. 2A–2C and Figs. S6–S8). No significant differences were seen in any of these markers in respiratory or limb muscles between non-cachexia controls and non-cachexia control-formoterol animals (Figs. 2A–2C and Figs. S6–S8).

Muscle inflammation

The number of infiltrated inflammatory cells (leukocytes and macrophages) was significantly higher in diaphragm and gastrocnemius muscles of cancer-cachexia rats than in non-cachexia controls (Figs. 3A–3D). Treatment with formoterol of cachectic rats attenuated inflammatory cell counts in the study muscles, especially in the diaphragm (Figs. 3A–3D). No significant differences were seen in any of these markers in respiratory or limb muscles between non-cachexia controls and non-cachexia control-formoterol animals (Figs. 3A–3D).

Specific muscle proteins

In the diaphragm and gastrocnemius muscles of cancer-cachexia rats, protein content of MyHC-I and MyHC-II isoforms, and total MyHC were significantly lower than in muscles of the non-cachexia controls, whereas actin, creatine kinase and carbonic anhydrase-3 protein levels did not significantly differ among the experimental groups (Figs. 4A–4C, 5A–5C and Figs. S9–S13, respectively). Treatment of cancer-cachexia rats with formoterol significantly attenuated the decrease in MyHC-I and MyHC-II isoforms, and total MyHC protein content, while it did not modify the content of the other study muscle proteins (Figs. 4A–4C, 5A–5C and Figs. S9–S13, respectively). No significant differences were seen in any of these markers in respiratory or limb muscles between non-cachexia controls and non-cachexia control-formoterol animals (Figs. 4A–4C, 5A–5C and Figs. S9–S13, respectively).

MRC complexes: enzyme activities

In cancer-cachexia rats compared to non-cachexia controls, activity of citrate synthase (CS) did not significantly differ between cancer-cachexia rats and the controls animals in any study muscle (Fig. 6A). Activities of complex I, II and IV did not significantly differ in any of the study muscles among the different experimental groups (Figs. 6B–6D). Treatment with formoterol did not modify activity levels of CS or complex I, II or IV in either diaphragm or gastrocnemius muscles among the study groups (Figs. 6A–6D). No significant differences were seen in any of these markers in respiratory or limb muscles between non-cachexia controls and non-cachexia control-formoterol animals (Figs. 6A–6D).

Redox balance markers in respiratory and limb muscles

As previously demonstrated (Busquets et al., 2004; Busquets et al., 2011; Busquets et al., 2012; Marin-Corral et al., 2010; Toledo et al., 2011; Toledo et al., 2016), the Yoshida AH-130 ascites hepatoma is a suitable approach to study the underlying biology of muscle mass loss in cancer cachexia. The tumor growth induces a progressive loss of body weight and skeletal muscle proteins in the host. Our group has published extensively on the elucidation of the biological mechanisms involved in muscle wasting on the basis of this experimental model of oncologic cachexia (Busquets et al., 2004; Busquets et al., 2011; Busquets et al., 2012; Chacon-Cabrera et al., 2014; Chacon-Cabrera et al., 2015; Chacon-Cabrera et al., 2017; Fermoselle et al., 2013; Fontes-Oliveira et al., 2013; Fontes-Oliveira et al., 2014).

The present investigation provides evidence on the posttranslational modifications induced by ROS on proteins in both respiratory and limb muscles and the response to treatment with the beta₂ agonist formoterol in rats bearing the Yoshida ascites hepatoma. As previously demonstrated in several models of cancer-induced cachexia (Fermoselle, Sanchez & Barreiro, 2011; Marin-Corral et al., 2010; Salazar-Degracia et al., 2016), increased levels of protein oxidation and nitration have also been shown in the gastrocnemius of rats bearing the Yoshida ascites hepatoma in the current study. Importantly, the beta₂ agonist formoterol attenuated the rise in protein oxidation and nitration levels detected in the limb muscle of the cancer cachectic rats. In fact, increased levels of cAMP were shown to inhibit ROS production and ROS-mediated effects probably via protein kinase A and mitogen-activated protein kinases (MAPK) pathways in tissues such as the vasculature of spontaneously hypertensive rats (Gusan & Anand-Srivastava, 2013).

The enzyme isoforms of SOD catalyze the conversion of two superoxide anions into hydrogen peroxide and molecular oxygen. In the current study, levels of cytosolic SOD1, but not those of SOD2, were significantly greater in the limb muscle of the cancer-cachexia rats than in control animals. These findings are somehow consistent with those previously reported, in which SOD2 levels were not modified in the limb muscles of tumor-bearing rats (Marin-Corral et al., 2010). Nonetheless, they are counter to what shown in mice with lung cancer-induced cachexia in which protein levels of SOD2 were significantly decreased in their respiratory and limb muscles, while levels of SOD1 were not modified by cachexia in the same animals (Chacon-Cabrera et al., 2017). Catalase is a ubiquitous heme protein that is responsible for the detoxification of H₂O₂ in tissues along with other enzymes such as glutathione peroxidases and peroxiredoxins. Importantly, the gastrocnemius muscle of the cachectic rats exhibited a significant rise in protein catalase content compared to control animals. A previous investigation also demonstrated a rise in protein catalase content in the fast-twitch fiber muscles of tumor-bearing rats (Marin-Corral et al., 2010). Collectively, these findings suggest that increased protein levels of SOD and catalase were induced probably to offset the deleterious effects of ROS in limb muscles of the cancer-cachexia rats. Interestingly, treatment of the cachectic rats with formoterol attenuated the rise in antioxidants seen in the tumor-bearing rats. It is likely that as cAMP inhibits ROS production (Ryall & Lynch, 2008), upregulation of antioxidants was not required in this model of cancer cachexia.

Levels of reactive carbonyls, MDA- and HNE-protein adducts, and those of protein tyrosine nitration were not modified in the diaphragm of cancer cachectic rats. These findings suggest that lower levels of oxidants were probably produced in the main respiratory muscle compared to the limb muscle in the cancer cachectic rats. Respiratory muscles need to overcome elastic and resistive loads and they are exposed to voluntary and involuntary control. Moreover, the diaphragm needs to contract rhythmically to generate the required forces for ventilation throughout the whole existence of the individuals, whereas the limb muscles are only subject to involuntary control and are not essential to life. The fibers in the diaphragm are unique in that this muscle must continually contract in animals and humans (Zuo et al., 2014; Zuo et al., 2015). On this basis, the differences seen in oxidative stress profile between the gastrocnemius and the diaphragm may be the result of differences in function and activity between these two muscles (Zuo et al., 2014; Zuo et al., 2015). As most of the investigations conducted so far have focused on the study of either respiratory or limb muscles, the current results are of interest. Moreover, in a model of lung carcinogenesis in which cachexia was also induced in mice, oxidative stress markers were also only increased in the gastrocnemius muscle but not in the diaphragm (Salazar-Degracia et al., 2016).

Moreover, species and experimental models may also contribute to the lack of a rise in oxidative stress levels observed in the respiratory muscle of the cachectic rats, since increased protein oxidation has been reported in the diaphragm of lung cancer tumor-bearing mice (Chacon-Cabrera et al., 2017). Additionally, levels of SOD2 were also shown to be reduced in the diaphragm of mice with lung cancer-induced cachexia that were studied for one month (Chacon-Cabrera et al., 2017). A differential metabolic rate of muscle in each model may account for divergences in the expression of oxidative stress markers and antioxidants between types of rodents (mice and rats). Furthermore, the duration and severity of the muscle wasting process in each type of cancer-cachexia model may also partly explain the differences observed in the expression of redox balance markers in the target muscles (Chacon-Cabrera et al., 2017; Fermoselle, Sanchez & Barreiro, 2011; Marin-Corral et al., 2010).

Proteins susceptible to be modified by oxidative stress and intramuscular inflammation cells

In the study, the content of proteins that have been shown to be carbonylated and affected by muscle mass loss was detected in both respiratory and limb muscles of all groups of mice. Importantly, the content of MyHC (slow- and fast-twitch isoforms) was significantly lower in both respiratory and limb muscles of the cancer cachectic rats and treatment with formoterol attenuated such a decrease. These findings are very consistent with previous reports in which protein levels of contractile myosin were also shown to be significantly reduced in the respiratory (Marin-Corral et al., 2009; Ottenheijm et al., 2005; Ottenheijm et al., 2006) and limb muscles of patients with advanced COPD (Fermoselle et al., 2012; Puig-Vilanova et al., 2014c) and lung cancer cachexia (Puig-Vilanova et al., 2014c) and in both diaphragm and gastrocnemius muscles of cachectic mice with emphysema and/or lung cancer (Chacon-Cabrera et al., 2017; Fermoselle, Sanchez & Barreiro, 2011; Salazar-Degracia et al., 2016).

One of the deleterious effects of oxidative stress in tissues is that involving an increased susceptibility of proteins to be degraded by the proteolytic system. In this regard, we previously demonstrated that levels of carbonylation of MyHC protein were significantly greater in diaphragm (Marin-Corral et al., 2009; Salazar-Degracia et al., 2016) and limb muscles (Fermoselle et al., 2012; Puig-Vilanova et al., 2014c) of patients with advanced COPD than those detected in the controls. As the content of this contractile protein was significantly lower in those muscles, we concluded that oxidative stress may be a trigger for enhanced proteolysis of specific muscle proteins in muscle wasting associated with chronic conditions (Fermoselle et al., 2012; Marin-Corral et al., 2009; Puig-Vilanova et al., 2014c), which could also precede proteolysis in animal models of cancer-induced cachexia (Chacon-Cabrera et al., 2014; Chacon-Cabrera et al., 2017), emphysema (Fermoselle, Sanchez & Barreiro, 2011; Salazar-Degracia et al., 2016), and disuse muscle atrophy (Chacon-Cabrera et al., 2016). Indeed, the proteolytic release of myofilaments is a characteristic feature of muscle wasting conditions (Powers, Kavazis & McClung, 2007).

Levels of contractile actin and the functional proteins creatine kinase and carbonic anhydrase-3 did not significantly differ in any study muscle between cachectic mice and control rats. These findings suggest that despite the previously reported oxidation of contractile actin which may enhance its proteolytic degradation (Powell, Gurzenda & Wahezi, 2001), in our model of cancer-induced cachexia, the content of this protein was not reduced. Levels of creatine kinase, but not carbonic anhydrase-3, were shown to be decreased in diaphragm (Marin-Corral et al., 2009; Ribera et al., 2003) and vastus lateralis (Barreiro et al., 2005; Puig-Vilanova et al., 2014c) of patients with severe COPD and lung cancer cachexia. Differences in the experimental models, humans versus rodents, in the type of conditions, chronic versus rather “acute”, and in the fiber composition of the muscles (Yamashita & Yoshioka, 1991) may account for the lack of effects on creatine kinase protein levels in the cachectic muscles in the current investigation.

In this study, the number of inflammatory cell counts was in general low in muscles all of the groups. Nonetheless, a significant rise in the number of both inflammatory cell types was detected in the diaphragm and gastrocnemius of the cachectic rats and treatment with formoterol attenuated such an increase. These findings are consistent with previous reports in which inflammatory cell counts were also shown to be greater in the respiratory and limb muscles of mice with lung cancer cachexia (Chacon-Cabrera et al., 2014; Chacon-Cabrera et al., 2017). Moreover, in patients with severe COPD, the number of inflammatory cells was also greater in the vastus lateralis muscle than in the healthy controls (Barreiro et al., 2010; Barreiro et al., 2011). The respiratory muscles, however, did not show any significant differences in inflammatory cell counts between patients and the controls (Barreiro et al., 2011). Another interesting observation was that the range of the magnitude of inflammatory cell numbers was higher in the diaphragm than in the limb muscle of the cachectic rats. It is likely that the greater respiratory loads imposed by muscle wasting induced a larger increase in inflammatory cells in the respiratory muscle than the limb muscle of the cachectic rats. In fact, similar trends were also reported in cancer cachectic rodents in previous studies (Chacon-Cabrera et al., 2014; Chacon-Cabrera et al., 2017; Salazar-Degracia et al., 2016).

MRC activity in respiratory and limb muscles

Interestingly, the magnitude of the measurements of the MRC activity complexes was in general greater in the respiratory than in the limb muscle of the study mice of all groups. These findings were similar to those previously reported in muscles of cancer cachectic mice (Fermoselle et al., 2013). A novelty in the study was the analysis of the activity of complexes I, II, and IV in the diaphragm and gastrocnemius muscles of rats with cancer-induced cachexia. No significant differences were seen between cachectic and control conditions in the study muscles for any of the analyzed complexes, and formoterol did not induce any significant effects. It is likely that the lack of effects on the MRC complexes in the cachectic muscles analyzed in the investigation may account for the absence of modifications resulting from the treatment with the beta₂ agonist formoterol.

Importantly, these findings are somehow counter to what reported in previous investigations. For instance, in tumor-bearing mice, the activity of MRC complexes I, II, and IV were significantly reduced in both respiratory and limb muscles and treatment with several agents reverted such a decrease to different degrees in each muscle (Fermoselle et al., 2013). Altered expression of proteins involved in mitochondrial biogenesis and fusion were also detected as signs of initiation of cachexia in skeletal muscles through interleukin (IL)-6 regulation in ApcMin/+ mice (White et al., 2012). In the peripheral muscles of cancer cachectic rats, mitochondrial content was also reduced (Fontes-Oliveira et al., 2013; Fontes-Oliveira et al., 2014), while mitochondrial uncoupling was demonstrated in cachectic mice bearing the Lewis lung carcinoma (Tzika et al., 2013). Interestingly, reduced ATP synthesis and mitochondrial disruption events were also detected in limb muscles of cancer-induced cachectic rats (Fontes-Oliveira et al., 2013) and in mice with lung cancer-induced cachexia (Constantinou et al., 2011). Again differences in the metabolic rates of muscles in each experimental model may account for differences observed in mitochondrial structural and biochemical alterations between types of rodents (mice and rats). Furthermore, the duration and severity of the muscle wasting process in each type of cancer-cachexia model may also partly explain the observed differences.

Study limitations

We acknowledge that the lack of functional data is a limitation in this study. Nonetheless, a first step in this field of investigation was to explore whether oxidative stress levels may be attenuated by treatment with a beta₂ agonist in both respiratory and limb muscles of rats with cancer cachexia. As shown in former investigations conducted on mice exposed to conditions such as hypoxia and TNF-alpha overexpression (Zuo et al., 2013; Zuo et al., 2014; Zuo et al., 2015), impaired muscle contractility can also be anticipated in the current model. In fact, previous investigations from our group have also revealed that in the same cancer cachexia model, muscle strength and physical activity decreased (Busquets et al., 2011; Toledo et al., 2011), while treatment with formoterol improved those parameters in the rats (Busquets et al., 2011).