Recently, several studies have reported that increased free radicals production may be associated with respiratory diseases, such as asthma, pulmonar fibrosis, chronic obstructive pulmonary disease (COPD), acute lung injury (ALI) and lung cancer (Valko et al., 2007; Ciencewicki et al., 2008; Park et al., 2009). In this line, the lung is continuously exposed to high levels of oxidants, which together with its large surface area and blood supply render it prone to injury mediated by free radicals (Rahman, 2002, 2012; Park et al., 2009).

The mechanical properties of lung tissue are important determinants of the overall mechanical behavior of the respiratory system, contributing significantly to elastic, resistive, and viscoelastic properties of the lung (Faffe and Zin, 2009; Suki et al., 2011). During disease states, mechanical properties can be altered considerably, which contributes importantly to a decline in lung function (Suki et al., 2003; Bates et al., 2007; Gladysheva et al., 2010). Several mechanisms have been suggested to explain the functional decline in lungs, such as thickened alveolar membranes (Rocco et al., 2001), increased cellularity (Casquilho et al., 2011; Oliveira et al., 2015), extracellular matrix remodeling (Rocco et al., 2001) and fibrosis (Sansores et al., 1966). In the context, prior studies report changes in lung resistance and elastance resulting from lung diseases in which oxidative stress plays a key role (Suki et al., 2003; Bezerra et al., 2011; Lima-Trajano et al., 2011).

Numerous compounds induce oxidative stress, as expressed by redox decomposition of hydroperoxides or hydrogen peroxide by metal ions, and thermal decomposition of free radical initiators, including hyponitrites, peroxides and azo compounds (Terao and Niki, 1986; Dooley et al., 1990; Yokozawa et al., 2000). In this study, 2,2′-azobis (2-amidinopropane) dihydrochloride (AAPH) was chosen as the free radical generating system to impair lung function. This water-soluble azo compound directly generates peroxyl radicals (Dooley et al., 1990; Krainev and Bigelow, 1996; Musialik et al., 2008) at physiological temperature and at a constant and reproducible rate (Fiorentini et al., 1999) without generating H₂O₂ (hydrogen pexoxide) as an intermediate, thus rendering its effect independent of tissue content of enzymatic antioxidants. The half-life of AAPH is about 175 h (37°C at neutral pH), allowing the rate of free radical generation to be essentially constant during the first several hours in solution (Landi et al., 1995; Peluso et al., 2002). Because of all these advantages, the AAPH has been used extensively over the years as a predictable and controllable source of free radicals, to mimic the oxidative stress and assess the free-radical-scavenging activity of drugs (Terao and Niki, 1986). Indeed, in in vitro experimental methods, AAPH induces hemolysis of human erythrocytes (Qasim and Mahmood, 2015), lipid peroxidation of rat thymocytes and liver (Landi et al., 1995; Cai et al., 2003) and inhibition of spontaneous motility and contractile responses of guinea pig ileum and rabbit jejunum (Peluso et al., 2002). Furthermore, AAPH has been used in in vivo model of oxidative stress. In this line, induction of colitis in rats by intrarectal administration of AAPH, causes erythema, edema and histologically mucosal inflammation (Tamai et al., 1992). Intraperitoneal administration of AAPH can induce lipid peroxidation in liver, kidney and heart of rats (Dooley et al., 1990). Recently, a rat model of acute pancreatitis induced by AAPH has been proven useful to investigate the early events of oxidative stress insult to the pancreas (Tukaj et al., 2012).

Although AAPH was used in vivo, to our knowledge there is no study in the literature concerning the pulmonary outcomes of the intranasal administration of AAPH. Hence, we aimed at evaluating whether the exposure to AAPH impairs lung functional, morphological and biochemical profiles.

ΔP1 was significantly higher in AAPH100 and AAPH200 rats than in CTRL and AAPH25 groups that did not differ among them. Additionally, AAPH200 was also higher than AAPH50 (Figure 1A). Accordingly, a higher resistive pressure (that reflects Newtonian resistence) suggest an increase in airway resistance to airflow. Est was larger in AAPH100 and AAPH200 animals than in the CTRL group (Figure 1D), indicating stiffer lungs. Intranasal instillation of AAPH did not alter ΔP2 and ΔE (Figures 1B,C, respectively), which means that lung tissue viscoelastic properties were not substantially altered. Bronchoconstriction index showed a significant increase in all AAPH groups compared to CTRL (Table 1), which supports the higher pressure used to overcome central airway resistance (ΔP1 seen in Figure 1).

Figure 2 depicts photomicrographs of lung parenchyma showing alveolar collapse, thickened septa, cellular infiltration, and hepatized areas after instillation of AAPH. Additionally, each panel insert shows thickening of the airway epithelium in AAPH animals. However, AAPH100 and AAPH200 groups displayed a more important lung injury than the other AAPH groups. These findings support the presence of an inflammatory process, which probably is correlated with the increase in resistive and elastic lung mechanical parameters (ΔP1 and Est, Figures 1A,D).

The area occupied by collagen fibers was significantly higher in AAPH100 and AAPH200 groups, whereas the area of elastic fibers was significantly higher than CTRL only in AAPH200 group (Table 1; Figure 3). Therefore, alterations in extracellular matrix composition, as expressed by collagen and elastic fiber contents indicate pulmonary remodeling, which is in line with Figure 2.

The number of PMN and MN cells were significantly higher in AAPH200 rats than in CTRL group (Table 1). Regarding PMN influx, it should be pointed out that the significance level between CTRL and AAPH100 was 0.07, and, thus, the results should be cautiously interpreted.

AAPH induced inflammatory changes, as expressed by increased release of pro-inflammatory cytokines IL-1β, IL-6 e TNF-α. IL-1β was significantly higher in AAPH200 group than in CTRL animals (Figure 4A). IL-6 presented larger values in AAPH200 rats than in CTRL and AAPH25 groups (Figure 4B). AAPH200 group also presented higher TNF-α levels than CTRL, AAPH25 and AAPH50 rats (Figure 4C). In all instances, a trend to progressively higher values can be seen between CTRL and AAPH200 rats (Figure 4).

SOD activity was higher in AAPH25 rats than in the remaining groups and AAPH200 rats presented lower values than the other groups (Figure 5A). CAT activity was significantly greater in AAPH200 group than in the remaining groups (Figure 5B).

Levels of carbonyl groups were largern in AAPH200 rats than in the other groups, AAPH50 and AAPH100 were higher than CTRL and AAPH rats (Figure 6B). Levels of MDA were larger in AAPH200 group than in the remaining groups (Figure 6A).

The aim of the present study was to investigate whether AAPH is able to generate functional, histological, inflammatory, and biochemical impairments in the lung. For such purpose, lung mechanics and histology, as well as redox and inflammatory state were used as markers. Our findings demonstrated that AAPH exposure caused significant alterations in vivo in mechanics, together with histologic and inflammatory alterations and redox imbalance in rat lungs.

The literature shows that reactive oxygen species (ROS) can be harmful or beneficial in biological systems depending on the environment (Phaniendra et al., 2015). At high concentrations, ROS can mediate damage to cell structures, including lipids, proteins, and nucleic acids (Rahman, 2007; Valko et al., 2007). Remarkably, it is known that oxidative stress plays a central role in the pathogenesis of various lung disorders such as asthma, lung cancer, pulmonary fibrosis, COPD and ALI in human beings (Valko et al., 2007; Park et al., 2009). Hence, the redox environment is of paramount importance in triggering respiratory disease.

Oxidative stress can be addressed by a variety of agents, including hydrogen peroxide, xanthine oxidase and organic hydroperoxides (Miki et al., 1987; Musialik et al., 2008), but AAPH offers some specific advantages. AAPH generates free radicals by its unimolecular thermal decomposition, which quickly react with oxygen, producing peroxyl radicals (Dooley et al., 1990; Landi et al., 1995; Hanlon and Seybert, 1997; Yokozawa et al., 2000; Zhu et al., 2005). In their turn, peroxyl radicals, which are high-energy species and display biological diversity in their actions, induce lipid peroxidation. The determination of lipid peroxidation is the most frequently used approach to reveal involvement of peroxyl radicals in human disease (Gutteridge, 1995; Cadenas and Sies, 1998). As there is no report in the literature on the use of AAPH intranasally, we conducted pilot experiments and chose the timeline and doses used in this study accordingly.

Respiratory mechanical parameters were measured by the end-inflation occlusion method that allows the identification of elastic, resistive, and viscoelastic and/or inhomogeneous lung mechanical components (Bates et al., 1985, 1988). Our results showed that Est was significantly higher in AAPH100 and AAPH200 than in CTRL rats, indicating stiffer lungs and greater inspiratory impedance (Figure 1). Possibly, the increased elastance is partly secondary to the presence of inflammation of the lung parenchyma, characterized by edema, influx of inflammatory cells, hepatized areas and septal thickening (Figure 2). Indeed, in this context Table 1 shows that there is a significant increase in the amount of polymorpho- and mononuclear cells in AAPH200 group. There is also a trend in the same direction in AAPH100 rats.

In line with our results, evidence suggest that inflammatory processes can potentially impair the synthesis and/or storage of pulmonary surfactant, elevating alveolar surface tension and thus producing alveolar collapse with consequent increase in Est (Jain et al., 1992; Liu et al., 1996). In this context, type II alveolar epithelial cells produce and secrete surfactant proteins to maintain morphological organization, biophysical, and immune functions, and biochemical composition in lung surface film (Lu et al., 2010). The methionine residues in these proteins can be oxidized and inactived by ROS, which diminishes the ability of the surfactant to lower lung surface tension (Chao et al., 1997; Manzanares et al., 2007). Decomposition of AAPH can oxidize amino acid residues in proteins, including methionine, tyrosine, and tryptophan (Chao et al., 1997; Ji et al., 2009). Pro-inflammatory cytokines, such as TNF-α, IL-1β, and IL-6, are involved in both asthma and COPD and may play a role in amplifying inflammation and thus determining disease severity (Barnes, 2008). Indeed, we observed influx of inflammatory cells, alveolar collapse (Figure 2) and increase in pro-inflammatory citokines IL-1β, TNF-α, and IL-6 (Figure 4) in treated rats. Thus, it would be reasonable to hypothesize that lung exposure to AAPH increased Est owing to inflammation and alveolar collapse resulting from the impairment in the synthesis and/or storage of pulmonary surfactant.

Our results also disclossed that higher doses of AAPH significantly increased lung resistive pressure (ΔP1), suggesting that the airways were also affected. This is evidenced by progressive thickening of the airway wall with increasing dose of AAPH (Figure 2), and the bronchoconstriction index disclosed that all AAPH groups similarly displayed reduced lumina (Table 1). Our data corroborate previous studies in organ baths, which evaluate the contractile effect of H₂O₂ on rat (Szarek and Schmidt, 1990) and cat (Bauer et al., 1997) airway smooth muscle. Moreover, respiratory diseases are related to increased free radical production in airway smooth muscle, which alters the excitation-contraction coupling, increasing baseline smooth muscle tonus (Hulsmann et al., 1994; Sadeghi-Hashjin et al., 1996; Cortijo et al., 1999; Wiegman et al., 2015).

The amount of elastic and collagen fibers was increased in groups treated with high doses of AAPH (Table 1; Figure 3). A complex mechanism modulate the transcription of molecules that destroy the lung extracellular matrix proteins and produces activation/inhibition of diverse cells of the lung tissue, which altogether regulate matrix remodeling (Rocco et al., 2001). These modifications may alter lung geometry, and, hence, its elastic component (James and Wenzel, 2007). Rocco et al. (2001) demonstrated increased collagen and elastic components in a model of ALI, yielding a process of fibroelastosis that started early in the evolution of the lesion. Therefore, our data are consistent with the literature, since ROS may contribute to the development of pulmonary fibrosis owing to their effects on the production of cytokines and growth factors, such as TGF-β (Park et al., 2009). Additionally, the increase in ROS can stimulate fibroblast migration to injured areas, where they will secrete collagen and other matrix proteins, contributing to the stiffening of lung parenchyma (Kirkham and Barnes, 2013). In fact, oxidative stress plays an important role in the remodeling process, interfering with the architecture and, consequently, with lung compliance, which is observed in patients with asthma, COPD, ALI, and pulmonary fibrosis (MacNee, 2001; Kluchová et al., 2007). Therefore, our finding of increased pro-inflammatory cytokines levels (Figure 4) may explain the augmented deposition of collagen and elastic fibers (Table 1; Figure 3), as well as the increased Est (Figure 1).

In order to assess the redox activity, we analyzed the levels of the antioxidant enzymes SOD and CAT in lung homogenate. SOD activity was significantly augmented in AAPH25 (lowest dose) and diminished in AAPH200 (highest dose). Conversely, CAT activity was significantly increased only in AAPH200 (Figure 5). Indeed, a low level of oxidative stress can stimulate an increase in antioxidant enzyme activity, while a high level can induce protein damage and decreased antioxidant enzyme activity. In this case, a reduced enzyme activity may occur via direct oxidative damage of the enzyme molecules, or via oxidative stress-altered enzymes gene expression, or both (Dreher and Junod, 1996; Comhair et al., 2005; Park et al., 2009). In line with our CAT and SOD activities results in AAPH200 group, Escobar et al. (1996) reported that SOD was readily inactivated by AAPH in in vitro assay. Kwon et al. (2000) showed that exposure of SOD to AAPH led to the protein fragmentation, associated with enzymatic inactivation and generation of protein carbonyl (Kwon et al., 2000), which is in line with our protein damage data (Figure 6B). Increased CAT activity in AAPH200 group was probably caused in part due to increased cellular infiltration (Figure 2; Table 1). Consistent with this argument, the main cellular sources of ROS in the lung include neutrophils, eosinophils and alveolar macrophages (Ciencewicki et al., 2008). Increasing cellular infiltration releases ROS, which leads to augmented hydrogen peroxide content, the substrate of CAT. Taken these data together, it is reasonable to infer that the increase in CAT activity may be a compensatory mechanism for the overproduction of free radicals.

In order to assess the oxidative damage induced by AAPH, we measured lipid peroxidation by means of the end product MDA, and protein oxidation by the concentration of carbonyl groups (Figure 6; Valko et al., 2007; Kirkham and Barnes, 2013). Our data show that exposure to AAPH significantly increased lipid peroxidation only in AAPH200. On the other hand, levels of carbonyl groups were significantly higher in AAPH50, AAPH100, and AAPH200. Moreover, MDA can inactivate SOD by oxidizing the histidine residues located in the active site of the enzyme (Koh et al., 1999). Thus, these findings suggest that AAPH-derived free radicals cause oxidative damage by disturbing the antioxidant defense systems (Figure 6). Interestingly, increased levels of carbonyls and MDA were also detected in the lungs of patients with COPD (Kluchová et al., 2007). Additionally, COPD show elevated levels of ROS and carbonyls that may be associated with increased inflammation and airway remodeling (Kirkham and Barnes, 2013).

Our study presents limitations: (1) we did not measure cellularity in the BALF; (2) the induction of prostaglandins involved in the permeability of blood vessel was not determined.

In conclusion, exposure to AAPH caused significant inflammatory alterations and redox imbalance accompanied by altered lung mechanics and histology. Furthermore, we disclosed that exposure to AAPH, especially using the dose of 200 mg/kg, may represent a useful in vivo tool to trigger lung lesions.

Substantial contributions to the conception or design of the work: MM, GC, JL, SV, and WZ. Substantial contributions to the acquisition, analysis, or interpretation of data for the work: MM, GC, SV, AA, NC, and WZ. Drafting the work or revising it critically for important intellectual content: MM, GC, JL, SV, NC, and WZ. Final approval of the version to be published: MM, GC, JL, SV, NC, AA, and WZ. Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved: MM, GC, JL, SV, NC, AA, and WZ.

This work was supported by the “Fundação Cearense de Apoio ao desenvolvimento Científico e Tecnológico-FUNCAP” (grant 10430447-2); Conselho Nacional de Desenvolvimento Científico e Tecnológico e Inovação-CNPq (grants 300531/2012-5, 470495/2012-0); Fundação Carlos Chagas Filho de Apoio à Pesquisa do Estado do Rio de Janeiro-FAPERJ (grants E-26/112.092/2012, E-26/201.450/2014), Brazil. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.