Th17/Treg imbalance in COPD progression: A temporal analysis using a CS-induced model

Background The imbalance between pro- and anti-inflammatory immune responses plays a pivotal role in chronic obstructive pulmonary disease (COPD) development and progression. To clarify the pathophysiological mechanisms of this disease, we performed a temporal analysis of immune response-mediated inflammatory progression in a cigarette smoke (CS)-induced mouse model with a focus on the balance between Th17 and Treg responses. Methods C57BL/6 mice were exposed to CS for 1, 3 or 6 months to induce COPD, and the control groups were maintained under filtered air conditions for the same time intervals. We then performed functional (respiratory mechanics) and structural (alveolar enlargement) analyses. We also quantified the NF-κB, TNF-α, CD4, CD8, CD20, IL-17, IL-6, FOXP3, IL-10, or TGF-β positive cells in peribronchovascular areas and assessed FOXP3 and IL-10 expression through double-label immunofluorescence. Additionally, we evaluated the gene expression of NF-κB and TNF in bronchiolar epithelial cells. Results Our CS-induced COPD model exhibited an increased proinflammatory immune response (increased expression of the NF-κB, TNF-α, CD4, CD8, CD20, IL-17, and IL-6 markers) with a concomitantly decreased anti-inflammatory immune response (FOXP3, IL-10, and TGF-β markers) compared with the control mice. These changes in the immune responses were associated with increased alveolar enlargement and impaired lung function starting on the first month and third month of CS exposure, respectively, compared with the control mice. Conclusion Our results showed that the microenvironmental stimuli produced by the release of cytokines during COPD progression lead to a Th17/Treg imbalance.


Introduction
Chronic obstructive pulmonary disease (COPD) is characterized by a progressive airflow limitation that is not fully reversible and is associated with a chronic inflammatory response in the lungs [1]. COPD is one of the main causes of morbidity worldwide and is estimated to become the third most common cause of death by 2030 [2].
Smoking has been confirmed to be the major risk factor for COPD development. However, only 15-20% of smokers develop this disease, which suggests that individual intrinsic factors are responsible for COPD progression [3][4][5].
Several studies have highlighted the importance of innate [6,7] and adaptive [8,9] immune responses in the pathophysiology of COPD. Moreover, the imbalance between proinflammatory and anti-inflammatory immune responses mediated by the different subsets of T helper (Th) cells, such as Th17 and regulatory T (Treg) cells, respectively, plays a pivotal role in the progression of this disease [10,11].
The Th17 response has strong proinflammatory abilities mediated by the release of interleukin (IL)-17 [12], which can facilitate the proliferation of T cells and the expression of various inflammatory mediators [13]. In contrast, Treg cells are responsible for the secretion of antiinflammatory cytokines, such as IL-10 and TGF-β, which promote the control of the inflammatory response in COPD [11,[14][15][16][17]. Additionally, the abnormal Treg response observed in COPD patients might lead to persistent inflammation and thus progression of the disease [17,18].
Despite the Th17/Treg imbalance and the worsening of alveolar enlargement and lung function observed in clinical and experimental COPD studies [10,11,15,16], the pathophysiological mechanisms at different time points can only be evaluated using animal models, and this information is crucial for obtaining an improved understanding of the changes between the pro-and anti-inflammatory immune responses involved in the development and progression of COPD.
In this study, we used a CS-induced animal model and performed a temporal analysis of the inflammatory progression mediated by the adaptive immune response with a focus on the Th17/Treg responses.

Experimental groups
Male C57BL/6 mice (aged 6-8 weeks and weighing 20-25g) were randomly divided into groups exposed to CS for 1, 3 or 6 months, and the mice in the control groups were maintained

CS exposure protocol
The animals were exposed to CS as previously described by Toledo et al. [20]. The flow rate was set such that the carbon monoxide (CO) levels ranged from 250 to 350 parts per million (ppm). Approximately 12 ± 1 commercially filtered cigarettes were used per exposure (0.8 mg of nicotine, 10 mg of tar, and 10 mg of CO per cigarette), resulting in a total particulate matter concentration of 354.8 ± 50.3 μg/m 3 /day. The animals were maintained in the CS environment for 30 min for each exposure, and the exposures were repeated twice per day, 5 days per week for a period of 1, 3 or 6 months. The control groups were maintained under filtered air conditions.
Morphometry. Lung tissue sections were stained with hematoxylin and eosin (H&E) to evaluate the mean linear intercept (Lm), which serves as an indicator of the mean diameter of airspaces, as described previously [23] using an eyepiece with a known area attached to the microscope ocular lens [24]. For each animal, 20 nonoverlapping fields around the airways and in the distal lung parenchyma were assessed at 400× magnification, and the Lm values are presented in micrometers (μm).
Immunohistochemically stained lung tissue sections were scanned using a high-resolution digital scanner (Pannoramic SCAN, 3DHISTECH Ltd., Budapest, HU), and 5 to 10 peribronchovascular areas [25] at 200× magnification were analyzed for each animal. All the inflammatory mediators were measured with imaging analysis software (Pannoramic Viewer 1.15.4, 3DHISTECH Ltd., Budapest, HU) by counting the number of positive cells and dividing by the respective peribronchovascular total area calculated by the software (results are expressed as cells/μm 2 ).

Tissue preparation for laser capture microdissection (LCM)
For real-time RT-PCR analysis, the lungs were removed en bloc, inflated with diluted Tissue-Tek OCT (Sakura Finetek USA., Torrance, CA, USA) (50% vol/vol) in ribonuclease (RNase)free phosphate-buffered saline with 10% sucrose, immediately frozen in liquid nitrogen, and stored at -80˚C [26]. The lungs were sectioned (12 μm) with a cryostat (Leica CM3050 S, Leica Biosystems, Wetzlar, DEU) and stained with 0.5% Nissl (Cresyl violet acetate) (Sigma Chemical, St. Louis, MO, USA)/0.1 M sodium acetate buffer for laser capture microdissection (LCM) of the bronchiolar epithelial cells using a Zeiss PALM Microbeam Laser Microdissection System (Carl Zeiss Microscopy GmbH, Jena, DEU). A total of 20,000 to 40,000 bursts of the laser were used to collect a minimum of 5,000 cells from each animal.
RNA extraction and reverse transcription. Total RNA from the laser-captured bronchiolar epithelial cells was extracted using an RNAqueous-Micro Total RNA Isolation kit (Ambion, Applied Biosystems, Foster City, CA, USA) following the manufacturer's instructions. The quantity and quality of the RNA extract were assessed by microcapillary electrophoresis with a Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA) using an Agilent RNA 6000 Pico kit (Agilent Technologies, Santa Clara, CA, USA). Total RNA was reverse transcribed for complementary DNA synthesis (cDNA) using a High-Capacity RNA-to-cDNA kit (Applied Biosystems, Foster City, CA, USA) following the manufacturer's instructions.
Real-time RT-PCR. The samples from each animal were assayed in duplicate, and realtime RT-PCR was performed with a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The polymerase chain reactions were performed with 1 μL of target TaqMan Gene Expression (NF-κB-Mm00476361_m1, TNF-α-Mm00443258_m1), 10 μL of TaqMan Universal Master Mix II (Applied Biosystems, Foster City, CA, USA), 7 μL of DEPC-treated water (Ambion, Applied Biosystems, Foster City, CA, USA), and 2 μL of cDNA.
The relative expression levels of the target gene were quantified using the 2 -ΔΔCt approach [27] with the housekeeping β-actin (Mm01205647_g1) as the endogenous control gene.

Statistical analysis
The statistical analyses were performed using GraphPad Prism 5.0 (GraphPad, San Diego, CA, USA). The normality of the data distribution was verified with the Shapiro-Wilk test, and the data are presented as the means ± standard deviations (SDs). Firstly, we compared the Smoke groups and their respective Control groups at different time points using t-test or Mann-Whitney test, depending on the normality of the data.
Because we did not observe significant differences among the Control groups at different time points (Table 1), we performed a temporal analysis of the three Smoke groups. For these comparisons, we used one-way analysis of variance (ANOVA) followed by Tukey's post-test and Kruskal-Wallis test followed by Dunn's post-test for parametric and non-parametric data, respectively. The differences were considered significant at P < 0.05.

Comparisons between the Smoke groups and their respective Control groups
Respiratory mechanics and lung morphometry. Although there were no significant differences in the Raw parameter between the Smoke and Control groups at the different time points (Fig 1A; 1 st month, P = 0.6784, t-test; 3 rd month, P = 0.7879, Mann-Whitney test; 6 th month, P = 0.6485, Mann-Whitney test), decreases in the Gtis value were observed in the Smoke groups after 3 and 6 months of exposure compared with the respective Control groups One month of CS exposure increased the Lm in the peribronchial area ( Fig 1D; 1 st month, P = 0.0002, Mann-Whitney test; 3 rd month, P = 0.0002, t-test; 6 th month, P = 0.0001, t-test) and the distal parenchyma ( Fig 1E; 1 st month, P = 0.0002, Mann-Whitney test; 3 rd month, P = 0.0008, t-test; 6 th month, P = 0.0004, t-test), suggesting alveolar enlargement, and these observations were also observed after 3 and 6 months of exposure.
An increase in IL-17 + cells was detected in the Smoke group compared with the Control group only after 6 months of exposure (Fig 7; 1 st month, P = 0.8458, t-test; 3 rd month, P = 0.0931, Mann-Whitney test; 6 th month, P = 0.0048, t-test). The numbers of IL-6 + cells showed an increase after 3 months of CS exposure, and this increase was also observed after 6 months (Fig 8; 1 st month, P = 0.2014, t-test; 3 rd month, P = 0.0065, t-test; 6 th month, P = 0.0108, t-test).

Comparisons among Smoke groups at different time points
Compared with the first month, decreases in Gtis and Htis were observed after 3 months of exposure and onward (Fig 12A; Raw, P = 0.3273, One-way ANOVA and Tukey's post-test; Gtis, P = 0.0014, One-way ANOVA and Tukey's post-test; Htis, P = 0.0004, One-way ANOVA and Tukey's post-test). Additionally, the Lm values were increased after 6 months in both the peribronchial area and distal parenchyma compared with the same areas after one month ( Fig  12B; Peribronchial area, P = 0.0019, Kruskal-Wallis test and Dunn's post-test; Distal parenchyma, P = 0.0202, Kruskal-Wallis test and Dunn's post-test), which suggested progressive functional and structural changes.
The analysis of the immune response revealed increases in CD4 + and CD8 + T cells after 3 months compared with after one month, and increases in these cells were still observed after 6 months. The numbers of CD20 + B cells were increased after 6 months compared with the numbers observed after 1 and 3 months (Fig 12C; CD4 + T cells, P = 0.0003, One-way ANOVA and Tukey's post-test; CD8 + T cells, P < 0.0001, Kruskal-Wallis test and Dunn's post-test; CD20 + B cells, P < 0.0001, One-way ANOVA and Tukey's post-test).

Immunofluorescence analysis
The results from immunofluorescence analysis was consistent with the immunohistochemical findings. The merged images revealed that the FOXP3 + cells (green) colocalized with the IL-10 + cells (red) and nuclei (blue) in all the control groups. The analysis of the Smoke groups after 1 month revealed only a few IL-10 + cells (red), whereas exposure to CS for 3 and 6 months resulted in weak signals for both FOXP3 + and IL-10 + cells (Fig 13).

Discussion
Our results show that the microenvironmental stimuli produced by pro-and anti-inflammatory cytokines during COPD progression lead to a Th17/Treg imbalance. To our knowledge, this study constitutes the first evaluation of Th17 and Treg cells and some of their specific cytokines at different time points during COPD development.
An increase in IL-17 + cells and a decrease in Treg cells were detected after six months compared with after one month, and at this time period we also observed higher Lm values in the peribronchial area and the distal parenchyma, which suggested a progression of alveolar enlargement over time. This progression of alveolar enlargement and inflammatory processes at different time points agreed with the respiratory mechanics findings.
We observed a transitory increase in tissue elastance after one month of CS exposure, and this increase was likely due to the presence of tissue edema and inflammatory infiltrates in the lung parenchyma [28]. However, from three months to the end of the exposure protocol, we detected a worsening of lung function, characterized by decreases in tissue elastance and tissue resistance. This finding corroborates previous studies that highlighted the pivotal role of a Th17/Treg imbalance in lung function impairment [11,16,29].
Decreases in IL-10 + cells and TGF-β + cells occurred before the reduction in Treg cells, suggesting that despite the relative abundance of Treg cells after one month of CS exposure, these
Previous clinical findings have shown the importance of Treg cells and IL-10 production in obstruction progression in smokers [11,13,14,29]. Sales et al. [14] reported decreases in Treg cells and IL-10 + cells and a consequent increase in IL-17 + cells in the airways of obstructed smokers compared with those of healthy smokers and control subjects [14]. Additionally, Wang et al. [11] reported decreases in the frequency of Tregs and the levels of IL-10 in the peripheral blood and serum, respectively, in patients with moderate and severe COPD compared with heavy smokers and healthy controls [11].
Through an experimental comparison of CS-exposed animals with air-exposed controls, Wang et al. [15] found initial increases in Treg cells and IL-10 expression in the first month and decreases in the sixth month [15]. Duan et al. [16] also showed reductions in Treg cells and IL-10 expression with a concomitant increase in IL-17 in the third month of CS exposure, and these changes were intensified in the sixth month [16].
The balance between Th17 and Treg cells is essential for immune homeostasis [19,29]. In our study, the Th17 response was observed in later stages after an increase in IL-6 + cells, and this finding reinforces the relevance of this cytokine for the differentiation of Th17 cells and the inhibition of Treg differentiation [11,29].
The potential inflammatory property of Th17 cells is mainly mediated by IL-17, which has been associated with the development of inflammatory lung diseases, such as COPD [31-33].
Our analyses of CD4 + T cells, CD8 + T cells and B cells revealed increases in the numbers of these cells, as previously observed in COPD patients [8,9,14,17,38] and animal models [15,16,39,40]. Specifically, increases in CD4 + T cells and CD8 + T cells were observed after three months of CS exposure and continued to be observed after six months. In contrast, the B cell numbers were increased only at the end of the protocol.
Increases in B cells have been observed in the small [8] and large [41] airways, and were found to be associated with COPD progression. Moreover, these cell types organize into lymphoid follicles around the small airways and in the lung parenchyma of COPD patients [8,39]. Although we observed increased numbers of B cells in the lung parenchyma, we did not detect lymphoid follicles in this CS-induced model at the various time points.
Considering the importance of the innate immune response in COPD development and the role of the epithelium in this response [6,42], we evaluated the NF-kB and TNF-α expression levels in peribronchovascular areas and the respiratory epithelium isolated by laser microdissection. We obtained similar results for both compartments: an increase in NF-κB + cells was observed after one month of CS exposure and onward, and increases in TNF-α + cells and Th17/Treg imbalance in COPD progression the gene expression of both inflammatory mediators were detected after three months and onward.
The comparison of our results with previous studies using human lung samples [10,14] revealed that the main difference is the lung compartments in which we detected the inflammatory signal: in humans, the inflammatory infiltrate has mainly been observed in the airways between the respiratory epithelium and adventitia [8,14], whereas in this animal model, we observed the inflammatory infiltrates mostly in peribronchovascular areas between the airways and adjacent blood vessels. This difference likely occurs due to the specialized capillaries in this area, which allow the rapid release of leukocytes and thereby results in edema and inflammatory infiltration [43].
Although the pathophysiological features of COPD show many differences between humans and animal models, it is important to note that we observed the same inflammatory profile, with increases in T and B cells and a reduction in Treg cells, in both organisms.