Persistence of pulmonary tertiary lymphoid tissues and anti-nuclear antibodies following cessation of cigarette smoke exposure

Female BALB/c mice (6-8 weeks old) were purchased from Charles River Laboratories (Montreal, PQ, Canada). Female, 6-8 weeks old C57BL/6 and IL-1R1^-/- (C57BL/6 background) mice were purchased from The Jackson Laboratories (Bar Harbor, ME). All mice were kept in a 12-h light-dark cycle with food and water ad libidum. The McMaster University Animal Research Ethics Board approved all experiments described in this study.

Mice were exposed to cigarette smoke for 4 days, 8 weeks and 24 weeks using a whole-body exposure system (SIU48, PROMECH LAB AB, Vintrie, Sweden) as previously described in detail [13, 14]. Briefly, mice were exposed to the mainstream smoke of twelve 3R4F reference cigarettes (University of Kentucky, Lexington, USA) with filters removed. Mice were exposed 5 days per week, twice daily, for 50 minutes/exposure. Total particulate matter in the exposure chamber was measured weekly and ranged from 600 to 700 μg/L. We previously reported that cigarette smoke exposure is well tolerated and results in cotinine and carboxyhemoglobin levels comparable to human smokers [13]. Control animals were exposed to room air only.

Bronchoalveolar lavage (BAL), lung tissue, and blood were collected as previously described [13, 14]. Total cell counts in the BAL were determined using a haemocytometer. BAL cytospins were prepared for differential cell counts and stained with Hema 3 as per the manufacturer’s instructions (Biochemical Sciences Inc., Swedesboro, New Jersey, USA). At least 300 leukocytes were counted per cytospin. BAL fluid was stored at -80°C. Blood was collected by retro-orbital bleeding, serum was obtained and stored at -80°C. For histological assessment, the left lung was inflated with 10% neutral buffered formalin at a constant pressure of 30 cm of water, and then fixed in 10% neutral buffered formalin for 72 hours.

All clinical samples were collected with patients’ informed written consent and ethical approval from the Research Ethic Board at McMaster University. Sputum induction was performed as described by Pizzichini et al.[15]. Spirometry was performed according to standards of the American Thoracic Society.

Bronchoalveolar lavage levels of CXCL13 were measured by ELISA according to manufacturer’s specifications (R&D systems, Minneapolis, MN). Anti-nuclear antibodies were measured by ELISA. Nuclei were isolated from TC-1 cells (mouse lung epithelial cell line; grown according to ATCC guidelines) and A549 cells (human adenocarcinoma cell line; grown according to ATCC guidelines) according to manufacturer’s specifications (Nuclear extraction kit, Millipore, CA). NUNC 96-well plates (Nalge Nunc international, NY) were dry-coated with 1 μg of nuclear protein per well in 100 μl of water overnight at 37°C. Wells were washed three times (0.05% TWEEN-20 in PBS) prior to incubation with either mouse BALF (1:2 dilution), mouse serum (1:500 dilution) or human sputum supernatant (1:10 dilution) at room temperature (RT) for two hours. Plates were washed 3 times (0.05% TWEEN-20 in PBS) prior to incubation with the detection antibody for 1 hour at RT. Total mouse ANA were detected using rabbit anti-mouse conjugated to horseradish peroxidase (HRP) (1:1000; Invitrogen, NY). For isotype specific detection, wells were incubated with biotinylated goat anti-mouse IgM, IgA, or IgG (1/1000; Sigma Aldrich), washed, and incubated with streptavidin-HRP. Total human ANA were detected using anti-human immunoglobulin conjugated to HRP (1:1000; Invitrogen, NY). Wells were washed then incubated with chromogen TMB substrate (BioFX Laboratories, MD) and optical density was measured at 450 nm. The assay was validated using the serum and BALF of the autoimmune New Zealand Black (NZB) mouse as a positive control (Additional file 1: Figure S1).

Formalin-fixed, paraffin-embedded lungs were cut into 4 μm thick cross-sections and stained with hematoxylin and eosin. Four photos at a 16x magnification were taken for every lung. Whole lung cross-section area (Lung_area [pixel²/10¹⁰]) as well as the number of bronchus-associated TLT (TLT_number) were determined using the Image J Software and identified according to their distinctive morphology. Data are expressed as bronchus-associated TLTs per lung area (TLT_number/Lung_area).

Three micrometer thick sections from formalin-fixed, paraffin-embedded lungs were stained by immunochemistry for B220, CD138, IgM, IgA and IgG protein expression. Briefly, lung sections were deparaffinized in xylene and rehydrated in ethanol:water. Endogenous peroxidases were blocked in 3% H₂O₂ in methanol. Citrate buffer antigen retrieval was performed (45 minutes). Sections were blocked with 1% swine serum in TBS 0.01% Tween 20. Sections were stained at 4°C overnight with either rat anti-mouse B220 (1:50; Abcam), rabbit anti-mouse CD138, biotinylated goat anti-mouse IgM, IgA or IgG. Antibodies were detected with a biotinylated goat anti-rat IgG (B220) or anti-rabbit IgG (CD138) (1:100; 1 h RT; BioLegend) followed by streptavidin-HRP (DACO) or directly by streptavidin-HRP (IgM, IgA, IgG). Staining was visualized using 3-amino-9-ethylcarbazole (AEC) substrate reagent. Sections were counterstained with hematoxylin. Negative staining controls are presented in Additional file 1: Figure S2.

Statistical analyses were performed using StatView Software. Two-group comparisons were made using an unpaired Student’s t-test with a significant threshold at 0.05. Experimental protocols with more than two groups were compared using a two-way ANOVA (significant threshold at 0.05) followed, if applicable, by a Bonnferoni post-hoc test.

A characteristic histopathological feature of advanced COPD is the emergence of pulmonary tertiary lymphoid tissue (TLT) [16]. Using a murine model, we investigated the time course of cigarette smoke-induced TLT formation and whether TLTs persist following smoking cessation. BALB/c mice were exposed to cigarette smoke for 4 days, 8 weeks, and 24 weeks. Control groups were exposed to room air only. No TLTs were observed after 4 days (data not shown) and 8 weeks of cigarette smoke exposure (Figure 1A). In contrast, TLTs were present in the lungs of mice exposed to cigarette smoke for 24 weeks (Figure 1B). No TLT formation was observed in age-matched room air-exposed animals, providing evidence that formation of these lymphoid structures was a consequence of cigarette smoke exposure rather than aging. Similarly, we observed TLT formation in C57BL/6 mice following 24, but not 8, weeks of cigarette smoke exposure (data not shown). Of note, TLT persisted in the lungs following smoking cessation for 60 and 120 days (Figure 1B). Quantification of TLT provided evidence of their expansion following smoking cessation (Figure 1C). These data demonstrate that cigarette smoke exposure leads to the formation of TLTs that persist following smoking cessation.

Cell types found in cigarette smoke-induced pulmonary TLTs include T cells, B cells, dendritic cells, and macrophages [16]. Using immunohistochemistry, we characterized the phenotype of B cells in TLTs of BALB/c mice exposed to cigarette smoke for 24 weeks. Cells within TLTs stained positive for B220, a commonly used B cell marker (Figure 2A and 2B). The predominant isotype expressed within TLTs was IgM (Figure 2C and 2D), while only few IgG⁺ (Figure 2F) and IgA⁺ (Figure 2H) cells were observed. TLT B cells displayed a distinct membrane-staining pattern for IgM (Figure 2D) and lacked expression of the plasma cell marker CD138⁺ (Figure 2J). Of note, IgM⁺ (Figure 2E), IgG⁺ (Figure 2G), and IgA⁺ (Figure 2I) cells, displaying a distinct cytoplasmic staining pattern were observed in areas immediately adjacent to TLTs, often nearby blood vessels. This staining pattern is indicative of plasma cells, as B cell that differentiate into antibody producing plasma cells loose surface expression of membrane bound immunoglobulin while greatly expanding their immunoglobulin synthesis machinery. This gives plasma cells a characteristic cytoplasmic immunoglobulin-staining pattern. This observation was supported by the presence of cells expressing the plasma cell marker CD138⁺ in similar locations (Figure 2K and 2L).

Emerging evidence suggests that cigarette smoking elicits autoantibody responses in susceptible individuals [6‐8]. We next investigated whether formation of TLT and accumulation of plasma cells in the lungs of cigarette smoke-exposed mice was associated with autoantibody production. We chose to investigate the presence of broad-spectrum autoantibodies against nuclear antigens; a type of autoantibody widely associated with autoimmune disorders. Anti-nuclear antibodies (ANA) were assessed in the BALF (local) and the serum (systemic) of cigarette smoke and room air-exposed mice. While no increase in ANA was observed following 4 days of cigarette smoke exposure, exposure to cigarette smoke for 8 and 24 weeks was associated with significantly increased levels of ANA in the BALF (Figure 3A). No increase in BALF ANA levels was observed in room air-exposed mice. In mice exposed to cigarette smoke for 8 weeks, increased ANA levels returned to baseline following smoking cessation for 20 days. In contrast, ANA remained elevated for more than 120 days post-cessation in mice exposed to cigarette smoke for 24 weeks (Figure 3A). Further analysis of BALF ANA following 8 weeks of cigarette smoke exposure revealed that the main isotypes expressed were IgG and IgA (Figure 3B). After 24 weeks of smoke exposure, the predominant ANA isotype in the BALF was IgA (Figure 3C), an isotype adapted for mucosal secretion. Expression of ANA was specific to the lungs, as no increase in ANA was observed in the serum of cigarette smoke-exposed mice (Figure 3D). This suggests that following 8 weeks of cigarette smoke exposure, and in the absence of TLT, pulmonary ANA expression is transient and resolves following smoking cessation. In contrast, following chronic exposure to cigarette smoke, and in the presence of TLT, pulmonary ANA persist following smoking cessation.

Previous clinical studies focused on serum analysis of autoantibodies [7, 17, 18]. To investigate whether cigarette smoking was associated with accumulation of autoantibodies in the human lung, as observed in our animal model (Figure 3A), ANA were assessed in the sputum of COPD patients. ANA levels were significantly elevated in the sputum of GOLD stage 2/3/4 COPD patients compared to GOLD stage 0/1 (Figure 3E) (see Table 1 for patient descriptions). Moreover, steroid-based therapy had no effect on sputum ANA levels (see Table 2 for patient descriptions) (Figure 3E). To our knowledge, this is the first study to report that autoantibodies accumulate locally in the sputum of COPD patients and that they are refractory to steroid-based therapy.

In COPD patients, pulmonary inflammation may persist following smoking cessation. We therefore investigated whether the presence of TLT and the elevated ANA levels were associated with persistent inflammation following smoking cessation. BALB/c mice were exposed to cigarette smoke for 4 days, 8 weeks, and 24 weeks, and the BAL cellular profile was assessed 18 hours after the last smoke exposure and following smoking cessation. We observed significantly increased total cell numbers (TCN), mononuclear cells (MNC), and neutrophils following 4 days, 8 weeks, and 24 weeks of cigarette smoke exposure (Figure 4A-C). In mice exposed to cigarette smoke for 4 days (Figure 4A), increased numbers of total cells, MNC and neutrophils resolved within 3 days of smoking cessation. In mice exposed for 8 weeks (Figure 4B), TCN and MNC returned to baseline within 20 days, while the neutrophilia persisted. In mice exposed to cigarette smoke for 24 weeks (Figure 4C), TCN, MNC, and neutrophils declined but remained significantly elevated compared to room air controls following smoking cessation for 60 days. Following 120 days of smoking cessation, TCN and MNC returned to baseline levels, while neutrophilia resolved by >95%. These findings suggest that the cellular inflammatory profile in the BAL resolves following smoking cessation; although resolution of inflammation appears to be delayed following prolonged cigarette smoke exposure.

IL-1R1 signaling pathways play a critical role in cigarette smoke-induced neutrophilia, dendritic cell accumulation, and T cell activation [10, 12]. We therefore investigated the role of IL-1R1 signaling in the generation of TLT and ANA following chronic cigarette smoke exposure. Confirming our previous observations, no neutrophils were found in the BAL of IL-1R1^-/- mice following 24 weeks of cigarette smoke exposure (Figure 5B). Moreover, neither TLT nor elevated ANA were observed in the lungs of cigarette smoke-exposed IL-1R1-deficient mice (Figure 5A,C,D). CXCL13, a B cell-attracting chemokine that has recently been implicated in cigarette smoke-induced TLT formation [19], was significantly increased in cigarette smoke compared to room air-exposed wild type mice. In contrast, levels of CXCL13 were not increased in cigarette smoke-exposed IL-1R1^-/- mice (Figure 5E). These findings suggest that IL-1R1 signaling pathways play a central role in cigarette smoke-induced innate immune inflammatory processes as well as processes associated with adaptive immune responses such as the formation of tertiary lymphoid tissues and autoantibody production.