The ethics Committee of the School of Medicine of University of Sao Paulo approved all the experiments. Male BALB/c mice (8 weeks old) were purchased from the University of Sao Paulo (Sao Paulo, Brazil) and maintained as described in the Guide for the Care and Use of Laboratory Animals—NIH (20). Mice were divided into four groups: SAL (non-sensitized and non-exposed to cigarette smoke), CS (exposed to cigarette smoke), OVA (OVA sensitized and challenged), and O + CS (OVA sensitized, challenged, and exposed to cigarette smoke).

Animals from OVA and O + CS groups were sensitized with two intraperitoneal injections containing 20 µg/mL of OVA (Grade V, Sigma Chemical Co., St. Louis, MO, USA) adsorbed in 3 mg/mL of aluminum hydroxide (Pepsamar gel, Sanofi-Synthelabo S.A., Rio de Janeiro, Brazil) (21). Animals received the intraperitoneal injections of OVA in a total volume of 0.2 mL on days 0 and 14. Animals from SAL and CS groups received two intraperitoneal injection containing saline solution (NaCl 0.9%) and aluminum hydroxide on same days. On days 21, 23, 25, and 27, the animals were challenged via aerosol for 30 min. OVA sensitized groups were challenged with OVA solution at 1% and non-sensitized groups were challenged with saline solution. On day 28, the animals were studied (Figure 1A).

The CS and O + CS groups were placed inside a box divided in eight sections. The box was connected to a Venturi system that drove the cigarette smoke directly into the box, mimetizing a first-hand smoke system. Animals were exposed to seven commercial cigarettes (Derby Filter Red, 0.8 mg of nicotine, 10 mg of tar, and 10 mg of CO per cigarette, Souza Cruz, UDI, Brazil) without filters twice per day for 30 min (22) during 12 consecutive days from day 16 to 27 (Figure 1A). The concentration of CO inside the box was maintained between 350 and 400 ppm during entire time of exposition. The interval between each exposition was of 3 h. The aerosol challenge was performed 1 h after the first cigarette smoke exposition so it would not affect the antigen presentation.

To evaluate COHb levels in blood and to see if sensitization and challenge would interfere in these levels, mice (four per group) were euthanized right after last exposure to cigarette smoke (day 28 of the protocol) and 0.2 mL of blood was collected. Levels of COHb were obtained by blood gas analyzer (ABL800, Radiometer, CA, USA). The quantification of cotinine levels in blood was evaluated using a cotinine ELISA kit as manufacturer instructions (Abnova, CA, USA).

Twenty-four hours after the last challenge and cigarette smoke exposition, we evaluated airway responsiveness of eight to ten mice per group, in a whole-body plethysmograph system (Harvard Apparatus, Boston, MA, USA). After the baseline reading, anesthetized (thiopental 33 mg, 0.01 mL/g), tracheostomized, and ventilated mice were serially exposed to increasing concentrations of nebulized methacholine (Mch) at 3, 30, and 300 mg/mL for 1 min per dose. The values of elastance and resistance were captured after each dose of Mch at 30 s, 1, 2, and 3 min (21). The highest dose, regardless of the time, was chosen to determine maximal respiratory response.

Right after the airway responsiveness evaluation, 0.3 mL of blood from five to ten mice per group, was collected and diluted into 600 µL of saline solution. Then it was centrifuged at 3,000 rpm for 10 min at 4°C. The serum was maintained at −20°C until IgE analysis. ELISA kit was used to quantify serum and IgE levels according to manufacture instructions (R&D Systems, MN, USA). The levels of IL-4, IL-10, IL-13, and transforming growth factor (TGF)-β1 in the lung homogenate was measured using ELISA kit according to manufacture instructions (R&D Systems, MN, USA).

Left lungs of five to ten mice per group were fixed in 10% formalin and embedded in paraffin. Five-micrometer-thick sections were used to make histology slides. Lung histology slides were stained for periodic acid-Schiff-alcian blue (PAS-AB) in order to quantify the area of the bronchial epithelium area and mucus. The evaluation was made using an optical microscope with one of the oculars designed with a square with 50 lines and 100 dots and with a known area of 10,000 µm². The mucus and epithelium area was measured by counting the number of dots that reached the mucus or epithelium respectively. The area was calculated knowing that 100 dots are equivalent of 10,000 µm², and then the number of dots that reached mucus/epithelium would be equivalent of the area in square micrometer. In order to quantify collagen fibers, histology slides with lung slices were stained with picrossirius red. Using a microscope with camera, pictures of four airways per slide were taken and analyzed for the area in square micrometer of collagen fibers using Image Pro Plus software (Media Cybernetics Inc., MA, USA) (21, 23).

Mice (eight per group) were euthanized and bronchoalveolar lavage (BAL) fluid was performed. Lungs were washed with 1.5 mL of saline solution, then the volume was centrifuged at 1,500 rpm for 10 min at 4°C. The pellet of cells was resuspended with 1 mL of saline and total number of cell was counted using a hemocytometer. After BAL, lungs were perfused with 5 mL of saline. Then, lungs were harvested and cut in small pieces with scissors, incubated in solution containing collagenase (Sigma-Aldrich, 0.7 mg/mL) and DNase1 (Sigma-Aldrich, 30 µg/mL) for 30 min at 37°C. After incubation, suspension cells were filtered with 40 µm strainer and the enzymes were blocked with PBS added of serum at least three times of the initial volume. The cell suspension was centrifuged at 1,500 rpm for 10 min and resuspended in 1 mL of saline for total cell counting. Lymph nodes were harvested and homogenized using saline solution and a 40 µm strainer. The volume was centrifuged and the pellet with cells was resuspended with 1 mL of saline for total cell counting. Cells were stained for cell surface and intracellular markers. The phenotype of the cells was evaluated using the following antibodies: anti-B220-PE-CY7 (RA3-6B2), anti-FOXP3-V450 (MF23), anti-LY6C and G-APC-CY7 (RB6-8C5), anti-MHC II-PE or V500 (M5/114.15.2 and M5115.15.2, respectively), anti-SIGLEC-F-PE-CF594 (E50-2440), anti-CD3-PERCP (145-2C11), anti-CD4-V500 (RM4-5), anti-CD8α-APC-Cy7 (53-6.7), anti-CD11b-PE-CY7 or BV605 (M1/70), anti-CD11c-FITC (HL3), anti-CD24-PERCP-CY5.5 (M1/69), anti-CD25-PE-CY7 (PC61), anti-CD69-FITC (H1.2F3), anti-CD86-Alexa Fluor 700 (GL1), anti-ICOSL-PE (hk5.3), and PDL2-APC (Ty25), all of which were purchased from BD Bioscience, NJ, USA. F4/80-eFluor-450 (BM8) and PDCA1-eFluor-450 (eBio927) were purchased from eBioscience, SD, USA. A total of 1 × 10⁵ live events was acquired and analyzed with flow cytometry (LSR Fortessa, BD, San Jose, CA, USA) and FlowJo 10.0.6 (Tree Star, Ashland, OR, USA). Fluorescence minus one controls were performed for all antibody panels. The analysis strategy is provided in Figures S1–S7 in Supplementary Material.

Statistical differences between experimental groups were detected by analysis of variance (two-way ANOVA) followed by the Holm–Sidak post hoc test for multiple comparisons (SigmaStat 2.03, SPSS, Chicago, IL, USA). p-Values <0.05 were considered significant.

We analyzed COHb and cotinine levels, and the levels of both substances were comparable to those observed in a heavy smoker (Figures 1B,C) (24, 25).

We next analyzed whether the OVA protocol was able to stimulate an inflammatory process similar to human asthma. As we found in this model, OVA sensitization and aerosol challenge causes pulmonary allergic inflammation that is similar asthma in humans and characterized by higher lung remodeling (Figure 1H) and an increase in IgE levels (Figure 1G). We also observed hyperresponsiveness in the OVA group with increased maximal response of resistance (Figure 1D). However, there was no difference of resistance or elastance among different doses of methacholine among all groups (Figures 1E,F). We still observed that the combination of cigarette smoke exposure in this model of asthma caused higher deposition of collagen fibers over the airway when compared to CS alone and OVA group and promoted hyperplasia/hypertrophy of bronchial epithelial cells (Figures 1H–J).

We investigated the Th2 adaptive immune response, which has the characteristic cytokines of the allergic inflammatory response of the model we used. For that, we quantified IL-4, which increased in the OVA group but decreased significantly in the O + CS group (Figure 2A). To clarify the possible mechanisms involved in the increased responsiveness and remodeling observed before, we analyzed IL-13, which showed an increased level in the OVA group and interestingly decreased in the O + CS group (Figure 2B). We also analyzed IL-10 and TGF-β, which are present in inflammation as regulatory cytokines. The OVA group presented higher levels of IL-10 and TGF-β than the SAL group. Both of these cytokines decreased when OVA was combined with CS (Figures 2C,D) leading to an inflammatory profile. No significant differences were found in IFN-γ levels among the groups (data not shown).

We observed, in the OVA group, an increased number of eosinophils (Figure 3A). When OVA allergy is combined with CS exposure, the number of eosinophils in BAL decreased compared to OVA group (Figure 3A), whereas the number of neutrophils significantly increased in lung tissue (Figure 3B), indicating a switch in asthma phenotype inflammation.

Macrophages are the most abundant immune cell present in the lung, and these cells have important functions in lung homeostasis such as responding to infection and inflammation (26). We observed both types, M1 and M2, of macrophages in the BAL, and we found that cigarette smoke alone was capable of increasing both of them. Although OVA alone only increased the M1 macrophages, interestingly, in the O + CS group, we found a decrease in M1 macrophages but a substantial increase in M2 macrophages relative to OVA (Figures 4A,B).

To identify what was responsible for the decrease in cytokine levels, we analyzed the number of lymphocytes and their activation. OVA sensitization increased both CD4⁺ and CD8⁺ T cells in the lung tissue. Exposure to OVA and CS significantly intensified this OVA-induced increase in both CD4⁺ and CD8⁺ T cells in addition to activating CD8⁺ T cells (Figures 5A–C). T reg cells increased with OVA sensitization, and a drastic decrease was observed when sensitization was combined with CS exposure (Figure 5D).

Dendritic cells are the most professional antigen-presenting cells, and they are capable of influencing or determining the immune response to any assault (27). CS exposition trends to increase the number of CD11b⁺ DCs in the lung, but with no significance. OVA sensitization combined with CS exposition decreases significantly CD11b⁺ DCs compared to CS alone (Figure 6A); no difference among groups was observed on activation analysis of CD11b⁺ DCs (data not shown). The number of pDCs increased in the lungs of the OVA group and was significantly reduced in the O + CS group (Figure 6B). Activation determined by CD86⁺, ICOSL⁺, and PDL2⁺ costimulatory molecules expression showed decreased in O + CS group (Figure 6C). We then analyzed lymph node DCs, and we observed an increase in the number of CD8α⁺ DCs in the OVA group. Interestingly, there was a strong and significantly increase in the number of CD8α⁺ DCs in the lymph nodes of the O + CS group (Figure 7).