Chronic obstructive pulmonary disease (COPD) is a growing public health issue associated with high morbidity and mortality worldwide (Rubi et al., 2010). COPD is characterized by small-airway disease and parenchymal destruction, resulting in airflow limitation that is not fully reversible (Papaioannou et al., 2010; Yamamoto et al., 2010; Decramer et al., 2015).

Although aerobic training has shown beneficial effects on dyspnea and exercise tolerance in patients with COPD (Amin et al., 2014; Jacome and Marques, 2014), its impact on cardiorespiratory impairment have shown conflicting results (Oliveira et al., 2014). Additionally, experimental studies have reported controversies regarding the effects of aerobic exercise in emphysema. The role of aerobic training has been evaluated in different emphysema models: induced by elastase (Kononov et al., 2001) or papain (Flo et al., 2006). In both models, the aerobic training caused alveolar damage with marked enlargement of distal air spaces. These results were attributed to the stretching of newly deposited elastin and collagen fibers, resulting in larger distortions compared to the normal tissue, thus contributing to emphysema progression (Suki et al., 2005). Otherwise, aerobic training has been shown to induce release of anti-inflammatory mediators (Petersen and Pedersen, 2005; Radom-Aizik et al., 2008; Moir et al., 2010) and reduce levels of growth factors associated with fibrogenesis (O'Callaghan and Williams, 2000; Czarkowska-Paczek et al., 2006; Radom-Aizik et al., 2008; Moir et al., 2010). These differing data may be attributed to: (1) differences in experimental models of emphysema, which do not closely mimic the pulmonary and cardiovascular changes observed in human emphysema; and (2) the timing of exercise initiation, i.e., before, during, or after establishment of emphysema.

We hypothesized that moderate aerobic training, initiated after establishment of emphysema and in animals with recognized cardiorespiratory impairment, would lead to beneficial effects on cardiorespiratory function, as well as modulate proteins implicated in emphysema progression. To answer this question, we investigated the therapeutic effects of aerobic training on lung mechanics, inflammation, remodeling and morphometry, biological markers associated with inflammation, and endothelial cell damage, and cardiac function in a model of elastase-induced emphysema.

This study was approved by the Ethics Committee of the Health Sciences Centre, Federal University of Rio de Janeiro (CEUA-CCS-IBCCF 019). All animals received humane care in compliance with the “Principles of Laboratory Animal Care” formulated by the National Society for Medical Research and the “Guide for the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences, USA.

After 4 weeks of emphysema induction, ELA compared to C group exhibited: (1) higher protein levels of IL-1β and TNF-α, and decreased of VEGF (Figure 2); (2) increased alveolar collapse, lung hyperinflation, mean linear intercept, and number of neutrophils (Table 1, Figure 3); (3) increased collagen and elastic fiber content (Figure 4) and number of apoptotic cells in lung tissue (Figure 5); (4) ultrastructural damage of epithelial and endothelial cells, elastolysis, and increased number of fibroblasts (Table 2, Figure 6); (5) decreased static lung elastance (Est,L; in vivo lung mechanics) and increased tissue elastance (Figure 7), with no changes in tissue resistance or hysterisivity (in vitro lung mechanics; data not shown); and 6) increased right ventricular diastolic area (RVA; Table 3, Figure 8). No alterations were observed in left ventricular area (data not shown).

Aerobic training did not affect any of the aforementioned parameters in the C groups. Regarding the ELA subgroups, trained (T) compared to sedentary (S) animals exhibited: (1) an increased number of neutrophils in lung tissue (Table 1, Figure 3); (2) reduced alveolar hyperinflation, mean linear intercept (Table 1), and number of apoptotic cells in lung tissue (Figure 5); (3) decreased collagen and elastic fiber content (Figure 4), elastolysis, and alveolar wall rupture (Table 2, Figure 6); (4) reduced levels of TNF-α (Figure 2); (5) increased VEGF level (Figure 2), Est,L and E (Figure 7). Moreover, aerobic training reduced RVA (Table 3, Figure 8). No differences in left ventricular area were found among groups (data not shown).

In this study, mice were subjected to a protocol of multiple doses of elastase and, once identified the cardiorespiratory function impairment associated with emphysema (4 weeks after induction), aerobic training was then initiated and performed for 4 weeks. This is the first experimental study using an emphysema model with repetitive elastase instillations protocol to evaluate the therapeutic effect of moderate aerobic training. Exercise protocol resulted in decreased mean linear intercept, lung hyperinflation, number of apoptotic cells, collagen deposition and rupture of alveolar walls, levels of TNF-α in lung tissue, as well as increased VEGF levels, static lung elastance (in vivo), and tissue elastance (in vitro). Moreover, aerobic training reduced RVA. Therefore, in agreement with our hypothesis, in this translational study setting, moderate aerobic training produced effects desirable in an adjunctive therapeutic strategy for emphysema, modulating the pulmonary inflammatory and remodeling processes induced by emphysema and reducing pulmonary arterial hypertension.

Several experimental models have been used in an attempt to reproduce the characteristics of human emphysema. The most widely used are inhalation of cigarette smoke (Sarir et al., 2010; Chen et al., 2011; Barreiro et al., 2012; Toledo et al., 2012) and instillation of elastolytic enzymes (PPE, papain or human neutrophil elastase) (Cheng et al., 2009; Inoue et al., 2010; Mador et al., 2010; Zhang et al., 2010; Ishii et al., 2012). In this study, we used the model of multiple instillations of PPE previously described by Luthje et al. (2009) and modified by our group (Cruz et al., 2012; Padilha et al., 2015), which has been demonstrated to yield the lung and cardiovascular damage. Additionally, this model yields lengthy progressive inflammatory and remodeling processes, as well as alveolar destruction and worse lung function in the time course of the disease.

Aerobic training was performed on a motorized treadmill for 30 min, at a speed of 8–12 m.min⁻¹ and 5% grade, corresponding to ~65–70% of maximal oxygen uptake (VO_2max), as determined by previous studies (Lowder et al., 2005, 2006). The exercise protocol was applied for 4 weeks, and this mode of exercise allows better monitoring of intensity and duration as compared with swimming or voluntary running (Pastva et al., 2004).

Emphysema is characterized by alveolar extracellular matrix destruction, resulting in airspace enlargement and reduction in alveolar-capillary exchange area (Morris and Sheppard, 2006; Hantos et al., 2008). In this study, we observed a time dependence of lung mechanics and histology. At 4 weeks, the reduction in Est,L reflected the decrease in lung elastic recoil, which is consistent with the increase in mean linear intercept (Lopes et al., 2009, 2013; Anciaes et al., 2011). This could be explained by two factors: (1) the emphysema working values were in the non-steep part of the respiratory system PV curve due to tissue loss; (2) the decrease in lung elastance is in line with great collagen and elastic fibers depositions. It is well-known that the increase in mean linear intercept cannot fully explain the decrease in matrix breakdown, especially when the collagen content per alveolar wall is increased (Ito et al., 2005). To test this hypothesis, we confirmed that the increase in tissue elastance was associated with increased deposition of collagen and elastic fibers (Figures 4, 7). At 8 weeks, deposition of elastic and collagen fibers was even greater compared to 4 weeks, but tissue elastance was lower in ELA-S compared to C-S animals. These findings run counter to those observed in fibrotic lungs, in which total collagen content also increased, but was accompanied by an increase in stiffness (Dolhnikoff et al., 1999). We may infer that the remodeling process—more precisely, the internal structural organization of the collagen fibers in the alveolar wall—should be different in emphysematous and fibrotic lungs (Suki et al., 2005). Some reports have shown a disorganized collagen deposition in emphysema, which the mechanical forces can break the alveolar walls during breathing process (Suki et al., 2003; Martin-Mosquero et al., 2006), contributing to progressive damage. TNF-α also increased from ELA at 4 weeks to ELA-S at 8 weeks (Figure 2), depicting a time dependence of this emphysema model, similar to that described in an in vitro model (Demirjian et al., 2006). The raised levels of TNF-α can be associated to higher macrophage phenotype 1 (M1), which has pro-inflammatory features, instead to regenerative macrophage phenotype 2 (M2; Padilha et al., 2015). Furthermore, the increase in TNF-α levels along time was associated with the increased number of apoptotic cells (Figure 5), which shows a disease progression in this emphysema model.

It has been demonstrated that, in the emphysematous lung, alveolar walls can break at physiological stretch corresponding to normal breathing, suggesting that newly synthetized collagen could be weaker (Suki et al., 2005). Although we did not measure the dynamics of breathing in mice during aerobic training, we may infer that alveolar stretch might increase, which could be an important mechanical force acting toward alveolar wall breakdown (Mishima et al., 1999). However, we observed that, after moderate aerobic training, mean linear intercept, collagen, and elastic fiber content were reduced compared to those of sedentary animals. This was followed by an increase in Est,L and E. We may hypothesize that aerobic training does not generate a lung stretch strong enough to affect collagen and elastic fibers so as to stimulate matrix turnover. In contrast, it positively modulated the remodeling process, leading to elastic recoil restoration. In a previous study, Flo et al. demonstrated that exercise training starting 2 days after intratracheal papain infusion increased the severity of alveolar damage, worsening emphysema (Flo et al., 2006). The dissociation between the previous study and our results could be explained by: (1) the different agents (papain vs. elastase) used to induce emphysema, (2) the degree of exercise (6 days/week for 9 weeks vs. 3 days/week for 4 weeks), and (3) the interaction between acute papain exposure and intense exercise sessions. Taking into account the limitations related to extrapolation of animal experimentation findings to human conditions, it should be noted that rehabilitation starts when cardiorespiratory impairment is already present in patients with emphysema. In the present study, we observed signs of pulmonary hypertension after 4 and 8 weeks, characterized by increased RVA (Koskenvuo et al., 2010). Moderate aerobic exercise minimized the cardiovascular alterations observed in the ELA-S group (Figure 8). In line with these results, VEGF levels increased in response to moderate exercise training (Figure 2). This increase in VEGF level is consistent with attenuation of air space enlargement (Takahashi et al., 2013) and exercise-induced angiogenesis, which may contribute to increased vascular cross-section area and reduced right ventricular overload (Gustafsson and Kraus, 2001; da Silva et al., 2012).

Regular and moderate exercise improves the immune response by reducing production of inflammatory cytokines (Lee et al., 2010; Gholamnezhad et al., 2014). Moderate exercise training in mice exposed to cigarette smoke minimized the expression of monocyte chemoattractant protein-1 (MCP-1) by inflammatory cells in the lung parenchyma (Toledo et al., 2012). Furthermore, previous studies have shown that aerobic training in allergic lung inflammation (Vieira et al., 2007) and sepsis (Chen et al., 2007) decreased lung inflammation. In this line, we observed that aerobic training reduced levels of IL-1β and TNF-α in lung tissue, suggesting a reduction in local inflammatory response (Mooren et al., 2012; Weng et al., 2013). Not only lung inflammation, but also the percentage of lung cell apoptosis reduced after aerobic training. Aerobic exercise was associated with a delay in neutrophil apoptosis, indicating alert and activation of the nonadaptive immune system (Mooren et al., 2012). The consequence of this decrease in apoptosis rate would be an increase of peripheral neutrophil numbers, which was indeed observed in the present study (Table 1). During and after exercise, catecholamines and cortisol are released, and these substances may facilitate the transition of neutrophils from endothelial walls and bone marrow (Korchak et al., 1998). Although, ELA-T showed an increase in tissue neutrophils percentage, no further structural alterations were observed in this group due to their action, as elastolysis, which by the way, decreased. Furthermore, it should be noted that delays in apoptosis are observed in various inflammatory conditions (Maianski et al., 2004), and it may be related to neutrophils staying in lung tissue.

Regular and moderate therapeutic exercise in elastase-induced emphysema modulated lung inflammation and remodeling, and reduced pulmonary arterial hypertension. This study provides evidence that moderate aerobic exercise had beneficial effects on lung morphofunctional parameters and inflammatory mediators, as well as on cardiovascular function, in an animal model of emphysema.

Conceived and designed the experiments: IH, ML, GP, NR, PS, DX, PR; Performed experiments: IH, ML, GP, PM, RM, MA, NR, DX; Analyzed data: IH, ML, GP, PM, RM, MA, MM, NR, PS, DX, PR; Interpreted results of research: IH, ML, GP, PM, RM, MA, MM, NR, PS, DX, PR; Drafted, edited, critically revised paper: IH, ML, PS, PR; All authors read and approved final version of manuscript.

This study was supported by the Brazilian Council for Scientific and Technological Development (CNPq), the Rio de Janeiro State. Research Foundation (FAPERJ), the Department of Science and Technology (DECIT)/Brazilian Ministry of Health, and the Coordination for the Improvement of Higher Education Personnel (CAPES).