The Th17/Treg Cytokine Imbalance in Chronic Obstructive Pulmonary Disease Exacerbation in an Animal Model of Cigarette Smoke Exposure and Lipopolysaccharide Challenge Association

We proposed an experimental model to verify the Th17/Treg cytokine imbalance in COPD exacerbation. Forty C57BL/6 mice were exposed to room air or cigarette smoke (CS) (12 ± 1 cigarettes, twice a day, 30 min/exposure and 5 days/week) and received saline (50 µl) or lipopolysaccharide (LPS) (1 mg/kg in 50 µl of saline) intratracheal instillations. We analyzed the mean linear intercept, epithelial thickness and inflammatory profiles of the bronchoalveolar lavage fluid and lungs. We evaluated macrophages, neutrophils, CD4+ and CD8+ T cells, Treg cells, and IL-10+ and IL-17+ cells, as well as STAT-3, STAT-5, phospho-STAT3 and phospho-STAT5 levels using immunohistochemistry and IL-17, IL-6, IL-10, INF-γ, CXCL1 and CXCL2 levels using ELISA. The study showed that CS exposure and LPS challenge increased the numbers of neutrophils, macrophages, and CD4+ and CD8+ T cells. Simultaneous exposure to CS/LPS intensified this response and lung parenchymal damage. The densities of Tregs and IL-17+ cells and levels of IL-17 and IL-6 were increased in both LPS groups, while IL-10 level was only increased in the Control/LPS group. The increased numbers of STAT-3, phospho-STAT3, STAT-5 and phospho-STAT5+ cells corroborated the increased numbers of IL-17+ and Treg cells. These findings point to simultaneous challenge with CS and LPS exacerbated the inflammatory response and induced diffuse structural changes in the alveolar parenchyma characterized by an increase in Th17 cytokine release. Although the Treg cell differentiation was observed, the lack of IL-10 expression and the decrease in the density of IL-10+ cells observed in the CS/LPS group suggest that a failure to release this cytokine plays a pivotal role in the exacerbated inflammatory response in this proposed model.

. Mice were placed on a heated animal bed (30 °C) until they showed active movements. They were also subjected to post-surgical recovery procedure are received tramadol chlorhydrate (5 mg/mL) via intramuscular injections (Fig. 1). Thus, we did not observe any considerable differences in animal's behavior. After the recovery period, animals were returned to the cage under normal conditions. The animals were active and did not display signs of post-operative pain.

Bronchoalveolar lavage fluid (BALF).
At the end of the protocol, mice were intraperitoneally anaesthetized with thiopental (70 mg/kg) and euthanized by transecting the abdominal aorta. BALF was obtained through the tracheal cannula by washing the airway lumen with 3 × 0.5 ml of sterile saline. BALF samples were centrifuged at 1000 rpm at 4 °C for 10 min and the cell pellet was resuspended in 0.2 mL of sterile saline for the Control groups and 1.5 mL of sterile saline for LPS groups. Total cell numbers were determined using a Neubauer hemocytometer counting chamber (Carl Roth, Karlsruhe, Germany). Differential cell counts were evaluated by a microscopic examination of BALF samples prepared on cytocentrifuge slides that were stained with Diff Quick (Medion Diagnostics, Dündingen, Switzerland) 35 .
Lung preparation. Lungs were removed en bloc and fixed with 4% formaldehyde infused through the trachea at constant pressure of 20 cmH 2 O for 24 h. Lungs were embedded in paraffin and cut into 5 µm coronal sections.
Morphometry. For conventional morphometry, we used an eyepiece with a coherent system of 50 lines and 100 points with a known area attached to the microscope ocular to perform the mean linear intercept (Lm) measurements, an indicator of the mean alveolar diameter 36 . Tissue samples were stained with hematoxylin and eosin (H&E). For each animal, images of 20 fields at a magnification of 200× were captured. Lm was obtained by counting the number of times that the lines of the reticulum intercepted the alveolar walls and calculated using the following equation: Lm = Ltotal/NI. We performed the Lm analysis in subpleural airspaces and peribronchial airspaces.
Epithelial thickness. Histological sections were stained with H&E to evaluate the epithelial thickness. We used the Panoramic Viewer 1.5 (3DHISTECH, Budapest, Hungary), an image analysis system. We quantified 5 airways from each animal at a magnification of 400×. Epithelial thickness was defined as the distance between the basement membrane and the luminal cell membrane, excluding the cilia 37 , and the length of basement membrane was determined. The epithelial thickness was expressed as a relationship between the epithelial thickness and the length of basement membrane.
Immunohistochemistry. Tissue sections were deparaffinized and hydrated. After blocking endogenous peroxidase activity, antigen retrieval was performed with a high-temperature citrate buffer (pH = 6.0). The primary antibodies used in this study were: a rat monoclonal antibody against Mac-2 (1:50000, Cedarlane, Ontario, CA), a rabbit polyclonal antibody against neutrophil elastase ( 34 and replaced with BSA for the incubation with tissue sections. Images of the lung tissues were captured from 15-20 random parenchymal fields for each lung sample at 400× magnification using the Panoramic Viewer 1.5 (3DHISTECH, Budapest, Hungary). We used a digital analysis system and specific software (Image-Pro Plus version 4.5 for Windows, Media Cybernetics, MD, USA) to determine the area of pulmonary parenchyma. Then, we quantified the density of positive cells for Mac-2, neutrophil Double immunohistochemical staining. We performed double immunostaining to examine the deficiency in IL-10 expression in Treg cells. Firstly, lung tissue sections were stained with a rabbit polyclonal against FOXP3 (1:300, Abcam, Cambridge, UK) using an immunoperoxidase procedure and DAB as chromogen. Afterwards, sections were incubated with a rat monoclonal antibody against IL-10 (1:100, Santa Cruz, CA, USA) using an equivalent protocol with an immunoalkaline phosphatase procedure and a red colored reaction product (PermaRed/ AP, Diagnostic BioSystems, Pleasanton, CA). Finally, Harris hematoxylin was used to counterstain tissue sections. Images of the lung tissues were captured with an AxioCam digital camera MRc5 using the software Zen from Carl Zeiss (München-Hallbergmoos, Germany). We captured images at magnifications of 400× and 1000×.
Cytokine analysis. Lungs were homogenized in a saline solution (0.9% NaCl), centrifuged, and the supernatants were stored at −80 °C until subsequent analyses. The levels of IL-17, interferon-gamma (IFN-γ), chemokine C-X-C motif ligand 1 (CXCL1) and CXCL2 were quantified using enzyme linked immunosorbent assays (ELISA). The kits were purchased from R&D System (Minneapolis, MN), and the levels of IL-6 and IL-10 were quantified using ELISA kits purchased from BioLegend (San Diego, CA). The ELISA was performed according to the manufacturers' instructions.
Statistical analysis. The statistical analysis was performed using SigmaStat statistical software (Systat Software, San Jose, CA). For immunohistochemistry and cytokine analyses, we logarithmically transformed the data. Data were analyzed using one-way ANOVA followed by Holm-Sidak post hoc analysis for data with a parametric distribution or Kruskal-Wallis test for data with a nonparametric distribution and Dunn's post hoc analysis. Results are presented as means ± SE and p value < 0.05 was considered statically significant.  Lung morphometry. Alveolar enlargement was observed in the subpleural airspaces of animals exposed to CS compared to Control groups. However, we only observed alveolar enlargement in the peribronchial airspaces in the CS/LPS group, but not the Control/SAL group ( Lung tissue inflammation. We observed increased densities of macrophages and neutrophils in the lung parenchyma of animals that were challenged with LPS compared to the Control/SAL and CS/SAL. We also observed an increase in the densities of these cells in the CS/LPS group compared to the Control/LPS group. Furthermore, we observed an increase in the CS/SAL group compared to the Control/SAL group (Fig. 4A-J, respectively).

Ethical approval. This study was performed under a protocol approved by the Ethics in Research
The groups that received LPS challenge showed increased densities of CD4 + and CD8 + T cells compared to the Control/SAL group (Figs 4K-O and 5A-E, respectively). We also identified an increase in the CS/SAL group compared to the Control/SAL group. In addition, we observed an increase in the density of CD4 + cells in the CS/ LPS group compared to the CS/SAL and Control/LPS groups (Fig. 4K-O).
The LPS challenge (Control/LPS and CS/LPS) increased the density of FOXP3 + cells compared to Control/ SAL group (Fig. 6A-E). The density of STAT3 + cells was increased in the lung parenchyma of animals challenged with LPS compared to the Control/SAL and CS/SAL groups. We also observed an increase in the CS/LPS group compared to the Control/LPS group. Furthermore, we observed an increase in the CS/SAL group compared to Control/SAL group (Fig. 5F-J). Moreover, the LPS challenge (Control/LPS and CS/LPS) increased the density In addition, we detected an increase in the density of phospho-STAT5 + cells in the lung parenchyma of animals challenged with LPS compared to the Control/SAL and CS/SAL groups, and we observed an increase in the CS/ LPS group compared to Control/LPS group (Fig. 6K-O). The density of IL-10 + cells was increased in the lung parenchyma of animals challenged with LPS compared to the Control/SAL group. We also observed an increase in the Control/LPS group compared to the CS/SAL and CS/ LPS groups (Fig. 7A-E). Moreover, the LPS challenge (Control/LPS and CS/LPS) increased the density of IL-17 + cells compared to the Control/SAL group (Fig. 7F-J).
Regarding the ratio of normalized phosphoSTAT3/STAT3, there were no significant differences among the experimental groups. However, the ratio of phosphoSTAT5/STAT5 was increased in the CS/LPS compared to the Control/SAL and CS/SAL groups (Fig. 8A,B, respectively).

Measurement of cytokine levels in lung homogenates. The LPS challenge significantly increased IL-17
and IL-6 levels compared to the Control/SAL group (Fig. 9A,B, respectively). In Control/LPS and CS/LPS groups, we also detected increased levels of the CXCL1 and CXCL2 chemokines compared to the Control/SAL and CS/SAL groups. In addition, an increase in CXCL1 levels was detected in the CS/SAL group compared to the Control/SAL group and an increase in CXCL2 levels in the CS/LPS group compared to the Control/LPS group (Fig. 9C,D, respectively). The IL-10 levels were increased in the Control/LPS group compared to the Control/SAL and CS/SAL groups (Fig. 9E). We did not observe significant differences in INF-γ levels among the groups (Fig. 9F).
Double Staining for Treg and IL-10. The analysis of Treg/IL-10-positive cells revealed an increase in the Control/LPS group compared to the CS/LPS group (Fig. 10A,B, respectively).

Discussion
CS exposure and subsequent LPS challenge induced an inflammatory process in the BALF and lung parenchyma similar to the process observed in patients with COPD presenting an exacerbation 8,38,39 . Moreover, this dual CS/ LPS challenge increased the epithelial thickness and resulted in diffuse alveolar enlargement in the peribronchial and and subpleural airspaces, reflecting a noticeable injury to the parenchymal architecture. LPS alone increased the densities of both IL-17 + cells and Treg cells, but did not increase IL-10 levels in the CS/LPS group. LPS alone also decreased the number of IL-10 + cells in this group compared to the Control/LPS group. Therefore, we suggest that decreased IL-10 production plays a pivotal role in the inflammatory exacerbation in this proposed model.
We observed an inflammatory process in the lung parenchyma of CS groups that was mainly characterized by increased numbers of macrophages, neutrophils, CD4 + and CD8 + T cells, and the subsequent LPS challenge intensified this response. The BALF analysis only revealed statistically significant increases in the numbers of macrophages and total cells in LPS groups. Since the BALF analysis include airways other than the parenchymal areas, an intensified inflammatory response that mainly occurred in parenchymal areas was observed in this animal model. Alveolar macrophages are known to recruit other cell types to the site of inflammation by releasing chemokines 40 ; for example, CXCL1 (chemokine homologous to IL-8) and CXCL2 41 that are chemoattractants for neutrophils 42 . These findings are in consistent with the higher expression of CXCL1 and CXCL2 in lung homogenates observed in our study.
While CXCL1 expression was increased by exposure to CS or LPS challenge, the levels of CXCL2 were increased only in LPS groups, and the CS/LPS challenge exacerbated this response. Moreover, the administration of both stimuli further increased the neutrophil density in parenchyma, which reinforces the importance of innate immune response in this experimental model 16 .
In patients with COPD, the exacerbation due to bacterial colonization is mainly mediated by IL-8 released by neutrophils 43 , and both neutrophils and IL-8 production are associated with an increase in sputum production and worsening airway obstruction [44][45][46] .
Regarding T cell subtypes, we observed increased numbers of CD8 + and CD4 + T cells in both LPS and CS groups, consistent with previous studies using animal models of COPD 31,32,47 ; however, we reported a greater increase inin the number of CD4 + T cells in animals exposed to CS/LPS. The number of CD8 + T cells is increased in the respiratory tract and in the parenchyma of smokers with COPD 48 , and the activation of these cells might contribute to COPD progression 49 .
We analyzed the density of positive cells for STAT3 and phospho-STAT3 and STAT5 and phospho-STAT5 to evaluate the balance between Th17 and Treg differentiation.  The increase in the number of STAT3 + cells is consistent with the increased IL-6 levels in both groups challenged with LPS, since IL-6 is recognized to switch Treg cells to a Th17 response following chronic infection 52 .  On the other hand, the density of phosphoSTAT3 + cells was increased in both groups that received LPS, and the administration of CS did not exacerbate this response. Moreover, the numbers of phosphoSTAT3 + cells/ STAT3 + cells were decreased in the CS/LPS group, which was inconsistent with the increased IL-6 and IL-17 levels. Yew-Booth 53 and colleagues have previously described an increase in STAT3 levels, but not phospoSTAT3 levels, in patients with COPD compared to smokers using imunohistochemsitry, whereas when the authors analyzed the levels of these cytokines using western blotting, they detected increased levels of both STAT3 and phos-phoSTAT3. The authors suggested that the antibody used in the study was more suitable for western blotting than for imunhohistochemistry. Although we did not use the same antibody described by Yew-Booth and colleagues 53 , our results are consistent with the previous findings described above, reinforcing the hypothesis that the immunohistochemistry might not be the best technique to evaluate STAT-3 phosphorylation.
Interestingly, we also observed an increase in the density of STAT5 + and phosho-STAT5 + cells in the LPS groups, and the higher values for phosho-STAT5 + cells were observed in the CS/LPS group, corroborating the increased density of Treg cellsand the increased numbers of phosphoSTAT5 + cells/STAT5 + cells.
Nonetheless, the increase in Treg cells does not reflect an increase in IL-10 release 54 . In the present study, although we observed an increase in the number of Treg cells in both groups that received the LPS instillation, only the Control/LPS group showed concomitantly increased IL-10 levels and IL-10 + cells density compared to the other experimental groups.
The increased numbers of other cell types, such as macrophages and neutrophils, in CS/LPS groups that are mediated by the release of this anti-inflammatory interleukin 55 are insufficient to induce an increase in IL-10 levels.
In the majority of infections, the IL-10 is an essential regulator to control the inflammatory response, and has a more important role than any other cytokine 55 . In viral infections, the IL-10 produced by Treg cells is closely related to the maintenance of the immunopathological balance [56][57][58][59] . Nevertheless, the inhibition of IL-10 signaling may exacerbate a pro-inflammatory response that clears pathogens but damages the lung tissue 60 . Jin and colleagues 61 previously reported similar results in serum from patients with COPD collected during an exacerbation to our findings. The authors showed an increase in IL-17 levels compared to healthy nonsmokers and patients with stable COPD, with a concomitant increase in the number of Treg/CD4 + cells. Therefore, although the Tregs are upregulated during acute exacerbations, their generation and differentiation were not sufficient, suggesting that both pro-inflammatory and anti-inflammatory reactions are enhanced, with pro-inflammatory and anti-inflammatory reactions are enhanced, with pro-inflammatory reactions predominating during acute exacerbations in patients with COPD. Additionally, the authors showed normal Treg/IL-17 numbers; however, the Treg cells were not sufficient to suppress the exacerbated inflammatory process.In the present study, we did not evaluate how Treg function impaired IL-10 release. However, we showed the role of the TH17/Treg cytokine imbalance, highlighting the importance of a lack of IL-10 release in this COPD exacerbation model.

Conclusions
We showed that the Th17/Treg cytokine imbalance lead the inflammatory process exacerbation as well as the diffuse structural changes in lungs in this COPD exacerbation model.

Data Availability
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.