Background
Aging is a risk factor for chronic obstructive pulmonary disease (COPD) [
1]. Recent evidence indicates that cellular senescence of various types of cells is accelerated in COPD patients, including alveolar type II cells, endothelial cells, fibroblasts, and peripheral blood lymphocytes [
2‐
5]. Cellular senescence is a state of essentially irreversible growth arrest that occurs either as a result of a large number of cell divisions (replicative senescence) or exposure to any of wide range of stimuli, including oncogene activation, oxidative stress, and DNA damage (premature senescence) [
6,
7]. Unlike apoptotic cells, senescent cells remain metabolically active and are capable of altering their microenvironment for as long as they persist [
6,
7]. Since senescent cells accumulate in vivo, they are presumed to contribute to the pathogenesis of age-related diseases, such as COPD and atherosclerosis, in at least two distinct ways, first inhibiting tissue repair, because they remain viable but are unable to divide and to repair tissue defects, and second, by acting as a source of chronic inflammation, because senescent cells have been shown to secrete pro-inflammatory mediators [
1,
6‐
10]. However, whether the senescence of airway epithelial cells contributes to the development of airway diseases is unknown.
Clara cells are the principal progenitors of the distal airway epithelium [
11‐
14]. Clara cells of mice and certain other species are rich in a cytochrome P450 enzyme (CYP2F2) and therefore are sensitive to the toxic effects of naphthalene (NA), which is metabolized to a toxic intermediate by the enzyme [
11‐
14]. Repair of the airway epithelium after NA injury is accomplished in several overlapping stages. In mice, the proliferative response peaks 1 to 2 days after NA injury and is followed by the differentiation phase, which is normally completed in 2 weeks [
13].
We hypothesized that senescence of airway epithelial cells impairs repair processes and exacerbates inflammation after an airway injury. To test this hypothesis, we utilized a well-established murine model of NA-induced Clara cell depletion. To induce airway epithelial cell senescence in this model, we intraperitoneally injected mice with the brominated thymidine analog 5-bromo-2'-deoxyuridine (BrdU) after NA injury. BrdU is incorporated into DNA during the S-phase of the cell cycle, and is commonly used to identify and track proliferating cells. However, emerging evidence indicates that BrdU imposes genotoxic stress that induces premature senescence and therefore limits cell's proliferative response to growth stimuli [
15‐
18]. In this study we demonstrated that administration of BrdU following repeated exposure to NA induced epithelial cell (Clara cell) senescence and p38 mitogen-activated protein kinase (MAPK)-dependent inflammation in the distal airway epithelium of mice. These findings suggest that airway epithelial cell senescence impairs repair processes and exacerbates inflammation after airway injury, and presumably contributes to pathological alterations in the airways of COPD patients.
Methods
Animal protocol
The animal protocol was reviewed and approved by the Animal Care, Use, and Ethics Committee of Tokyo Women's Medical University. Eight-week-old male C57/BL6J mice were intraperitoneally injected with NA (Kanto Chemical, Tokyo, Japan: 200 mg/kg body wt) or corn oil vehicle on day 0 alone (acute model), or on days 0, 7, and 14 (chronic model). Each NA injection was followed by intraperitoneal injection of BrdU (Sigma, St. Louis, MO: 200 mg/kg body wt) or 0.3% carboxymethycellulose, on 3 consecutive days (days 1-3, 8-10, and 15-17). This BrdU administration schedule was chosen because epithelial proliferation in mice is maximal 1 to 2 days after exposure to NA [
13]. The p38 mitogen-activated protein kinase (MAPK) inhibitor SB202190 (Enzo Life Sciences, Plymouth Meeting, PA) or 0.1% DMSO was administered by intraperitoneal injection 30 minutes before each BrdU injection. Animals were killed on days 1, 2, 3, 4, 11, or 28 by injecting an overdose of pentobarbital sodium [
19].
Human lung tissue samples
The protocol of the study conformed to the Declaration of Helsinki, and approval from the Tokyo Women's Medical University Institutional Review Board was obtained. Lung tissue blocks were obtained from COPD patients (
n = 14), asymptomatic smokers (
n = 7), and asymptomatic nonsmokers (
n = 8) during lung volume reduction surgery or pulmonary resection for localized lung cancer. The clinical information regarding these patients is shown in Table
1.
Table 1
Characteristics of the subjects
Male/females, n | 12/2 | 7/0 | 2/6 |
Age, years | 65.9 ± 2.2 | 60.9 ± 6.3 | 64.3 ± 3.8 |
Smoking, pack years | 80.0 ± 14.1†† | 50.7 ± 6.2† | 0 ± 0 |
FEV1, liters | 0.91 ± 0.11** | 2.35 ± 0.17 | 2.14 ± 0.12 |
FEV1/FVC, % | 34.0 ± 3.4** | 75.4 ± 2.9 | 75.0 ± 4.3 |
FEV1, % predicted | 35.5 ± 4.0** | 91.0 ± 6.4 | 101.2 ± 5.4 |
Tissue preparation
Lungs of mice were inflation fixed in situ for 5 minutes with 10% neutral buffered formalin (NBF) at 25 cm water pressure, removed, and immersion fixed in NBF for 24 hours. Formalin-fixed tissue was embedded in paraffin, and sectioned (3 μm). For frozen fixation, lungs were inflated by manual instillation of 50% optimal cutting temperature compound, quickly frozen, and sectioned (3 μm). The tissue blocks from human lungs were fixed in NBF, embedded in paraffin, and sectioned (3 μm).
Cell culture
NCI-H441 cells (the American Type Culture Collection, Rockville, MD), a Clara-cell-like human lung adenocarcinoma cell line, were cultured in RPMI 1640 supplemented with 10% FCS. Cells were exposed to BrdU by culturing for 10 days in the presence of BrdU (25, 50, or 100 μM), with a medium exchange on day 5; control cells were similarly cultured in the absence of BrdU. In some experiments, the p38 MAPK inhibitor SB202190 was added to a concentration 10 μM [
19]. For telomerase inhibition, cells were cultured for 28 days in the presence of MST-312 (2.5 μM: Calbiochem, Gibbstown, NJ), with passages every 7 days; control cells were similarly cultured in the absence of MST-312 [
20]. Cell numbers were counted manually or by Alamar
®blue assay (Invitrogen, Camarillo, CA). Population doubling (PD) at each passage was calculated by using the formula: PD = ln (number of cells recovered/number of cells inoculated)/ln2.
Epithelial repair assay
NCI-H441 cells were cultured on 30 mm-plates in RPMI 1640 supplemented with 10% FCS in the presence or absence of 25 μM BrdU for 10 days. Cell monolayers were then damaged mechanically by crossing three times with a 10-200 μl volume universal pipette tip (Corning, NY, USA) and epithelial repair after mechanical damage was monitored for 72 hours. (
See Additional file
1 for details.)
Enzyme-linked immunosorbent assay (ELISA)
The concentrations of cytokines/chemokines in the cell culture supernatants were measured by using ELISA kits (Biosource International, Camarillo, CA), and values were normalized to the number of cells.
Senescence-associated β-galactosidase (SA β-gal) staining
SA β-gal staining was performed as described previously [
21]. (
See Additional file
1 for details.)
Immunohistochemistry and immunofluorescence
The primary antibodies against Clara cell 10-kDa secretory protein (CC10), β-tubulin IV, Ki-67, BrdU, p16
INK4a (p16), p21
WAF1/CIP1 (p21), phospho(Thr180/Tyr182)-p38 MAPK, polyclonal anti-phospho(Ser/Thr)-ataxia teleangiectasia mutated kinase (ATM)/ataxia teleangiectasia and Rad3-related kinase (ATR) substrate, phospho(Ser139)-H2AX (γH2AX), CD45, and CD90.2 were used. For immunohistochemistry and immunocytochemistry, the primary antibodies were detected with a secondary antibody conjugated with a horseradish-peroxidase (HRP)-labeled polymer (Envison+
®, DAKO Japan, Tokyo, Japan; Histofine
® Simple Stain, Nichirei Biosciences, Tokyo Japan). Immunoreactants were detected with a diaminobenzidine substrate or a HistoGreen
® substrate (AbCys, Paris, France). (
See Additional file
1 for details.) For immunofluorescence staining, the primary antibodies were reacted with secondary anti-IgG antibodies conjugated with Alexa Fluor 350, Alexa Fluor 488, or Alexa Fluor 594 (Invitrogen, Carlsbad, CA). Images were acquired by using an Olympus BX60 microscope (Olympus Optical Co., Ltd., Tokyo, Japan) equipped with a digital camera, and processed with a computerized color image analysis software system (Win Roof Version 3.5; Mitani Corporation, Fukui, Japan) and Adobe Photoshop software (San Jose, CA). The numbers of γH2AX-foci in the cell nuclei of at least 50 cells were counted visually through an Olympus BX60 microscope equipped with a 100× objective as described previously [
22,
23].
Immunoblot analysis
Cell lysates were fractionated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane. The membrane was probed with primary antibodies against phospho(Thr180/Tyr182)-p38 MAPK, p38 MAPK, NF-κB p65, phospho-NF-κB p65 (Ser536), phospho(Ser139)-H2AX (γH2AX, Cell Signaling), p21, or actin (
See Additional file
1 for details.)
Cell cycle analysis
The DNA content of cells was analyzed by flow cytometry [
24].
Morphometric analysis in murine distal airways
Morphometric analysis was performed in the distal bronchiolar airway region. Since cell type representation varies with anatomical location, the analysis was limited to the final 200-μm basement membrane (BM) that ended in a well-defined bronchoalveolar duct junction [
25]. The distal bronchiolar airway epithelium was defined as the cells located between the basal lamina and the airway lumen, and the peribronchiolar interstitium was defined as the cells located between the basal lamina of the distal bronchiolar airway epithelium and an adjacent blood vessel, alveolus, or bronchiole. Ten distal bronchiolar airways were randomly selected on each slide and examined under a microscope at ×400 magnification.
Epithelial injury was quantified on hematoxylin-eosin-stained slides by counting the number of necrotic bronchial epithelial cells that had exfoliated into the airway lumen and dividing the number by the total length of the BM. Clara cells were identified by immunohistochemistry for CC10, and the number of CC10-positive cells in the epithelium was divided by the total length of the BM. Epithelial cell proliferation was quantified by dividing the number of Ki-67-labeled nuclei in the CC10-positive cells by the total number of CC10-positive cells, or the number of Ki-67-labeled nuclei in the CC10-negative epithelial cells by the total number of CC10-negative epithelial cells. Epithelial cell senescence was quantified by counting the number of p21-labeled nuclei in CC10-positive cells or the number of SA β-gal-positive cells that co-express CC10 and dividing the number by the total number of CC10-positive cells. DNA damage response was quantified by dividing the number of phospho-ATM/ATR substrate-labeled nuclei in the CC10-positive cells by the total number of CC10-positive cells, or by counting the number of γH2AX foci in CC10-positive cells. Activation of p38 MAPK was quantified by dividing the number of phospho-p38 MAPK-labeled nuclei in the CC10-positive cells by the total number of CC10-positive cells. Airway inflammation was evaluated by counting the number of CD45-positive cells (pan-leukocytes) and the number of CD90.2-positive cells (T-cells) in the peribronchiolar interstitium and dividing their numbers by the total length of the BM.
Morphometric analysis of human bronchiolar airways
Human lung tissue sections were triple immunofluorescence stained for CC10, p16, and phospho-p38 MAPK, and five microscopic fields of tissue from each patient containing a region of distal bronchiolar airway epithelium were examined under an epifluorescence microscope at ×400 magnification. The number of CC10-positive cells that stained positive for p16 was divided by the total number of CC10-positive cells, the number of CC10-positive cells that stained positive for phospho-p38 MAPK was divided by the total number of CC10-positive cells, and the number of CC10-positive cells that stained positive for both phospho-p38 MAPK and p16 was divided by the total number of CC10-positive cells. The number of CC10-positive cells that stained positive for both phospho-p38 MAPK and p16 was divided by the total number of CC10-positive cells that stained positive for p16 (p38 MAPK index for senescent Clara cells), and the number of CC10-positive cells that were positive for phospho-p38 MAPK but negative for p16 was divided by the total number of CC10-positive cells that were negative for p16 (p38 MAPK index for presenescent Clara cells).
Statistical analysis
Data are expressed as means ± SEM. Statistical analyses were performed by using the Excel X software program with the add-in software Statcel 2 (OMS, Tokyo, Japan). Data obtained from two groups were compared by using Student's t-test. Comparisons among three or more groups were made by analysis of variance (ANOVA), and any significant differences were further examined by the Tukey-Kramer comparisons post hoc test. Data were tested for correlations by the Spearman rank correlation test. A p value of < 0.05 was considered significant.
Discussion
The results of the present study demonstrated that BrdU-induced senescence of airway epithelial cells impairs epithelial regeneration and stimulates p38 MAPK-dependent inflammation after NA-induced Clara cell depletion in mice. To our knowledge, this is the first evidence indicating that epithelial cell senescence contributes to incomplete repair and excessive inflammation in the airways of mice. The results of the study also showed for the first time that Clara cell senescence is accelerated in COPD patients and is accompanied by p38 MAPK activation, suggesting that epithelial cell senescence may contribute to the excessive inflammation in the airways of COPD patients.
We used BrdU as an inducer of premature senescence to model airway epithelial senescence in mice and using BrdU offered several advantages in the present study. First, induction of senescence by exposure to BrdU has well been established as a model of premature senescence in various types of cells [
16‐
19]. Second, since NA selectively injures Clara cells, using NA in combination with BrdU facilitated selective induction of senescence of the airway epithelial cells, and allowed only proliferative epithelial cells to incorporate BrdU into their DNA during the cell division that commenced to restore the NA-depleted pool of Clara cells. This is supported by our findings that while BrdU induced senescence in an in vitro culture of proliferating NCI-H441 cells, BrdU itself did not induce senescence of quiescent airway epithelial cells in mice that had not been exposed to NA. We therefore think that the senescent CC10-positive cells found in the mice exposed to NA and BrdU were mostly derived from Clara cells, which are the major progenitors of cells in the distal airways, but may have included a subpopulation of Clara cells, such as vCE cells or BASCs, that function as progenitors capable of renewing NA-injured airway epithelium [
26]. Third, immunostaining for Ki-67 (proliferation marker) and SA β-gal (senescence marker) in combination with BrdU immunostaining made it possible to track the fate of the epithelial cells that had incorporated BrdU into their DNA. In fact, we found that the epithelial cells that had incorporated BrdU into their DNA became senescent and no longer proliferated. However, a limitation of our study stems from the fact that the BrdU taken up by the cells is phosphorylated to deoxynucleotide monophosphate by the salvage pathway enzyme thymidine kinase, whose levels may differ from cell to cell [
28], and thus the repeated BrdU injection of mice may have selected for a subset of cells that had a lower level of the salvage enzyme and were no longer able to incorporate BrdU into their DNA. Such selection may have biased the results of our study. Another limitation of our study is the fact that we used BrdU, not cigarette smoke, to induce cell senescence, which may make it uncertain to translate the results of animal experiments to human COPD.
However, our murine model of Clara cell senescence provided clear evidence that senescence impairs regenerative response to airway injury. This finding is not surprising because senescent cells no longer proliferate in response to growth stimulation [
6,
7]. The impaired regenerative response in the present study was not due to a direct cytotoxic effect of BrdU, because BrdU did not cause any discernible epithelial damage, and it did not exacerbate the NA-induced epithelial damage in the airway of the mice (Figure
1). By contrast, BrdU imposed genotoxic stress, as demonstrated by the phosphorylation of ATM/ATR substrates and γH2AX (Figure
3), which triggers the DNA damage signaling pathway that causes p21-dependent cell cycle arrest, and eventually an irreversible senescence arrest [
6,
7,
29].
Recent evidence suggests that airway epithelial cells, including Clara cells, play a pro-inflammatory role in the immune response through secretion of pro-inflammatory cytokines [
30,
31]. In the present study we found that Clara cell senescence was accompanied by exacerbation of airway inflammation that was at least in part attributable to increased pro-inflammatory cytokine secretion by senescent epithelial cells (Clara cells). These findings corroborate those of previous studies showing that other senescence inducers, including oncogene activation, DNA damage, and telomere shortening, stimulate pro-inflammatory cytokine secretion by cultured fibroblasts and endothelial cells, a phenomenon termed the "senescence-associated secretory phenotype (SASP)" [
10,
32‐
38]. Our study also showed that senescent-associated inflammation occurs in vitro as well as in vivo, and identified p38 MAPK activation as a positive regulator of the senescence-associated inflammation. P38 MAPK activation is a crucial step in the synthesis of several pro-inflammatory cytokines and recent evidence indicates a critical role of the p38 MAPK pathway in proinflammatory cytokine production by cells that have undergone oncogene- and environmental stress-induced senescence [
39,
40]. Similar to the findings in our own study, a previous study showed that inhibition of p38 MAPK by SB202190 reduced expression of IL-8 by fibroblasts after oncogene-induced senescence [
33]. Other potential regulators of senescence-associated inflammation include the transcription factors NF-κB and C/EBPβ [
10,
41]. Although no significant NF-κB activation in the BrdU-induced senescent NCI-H441 cells was detected in this study, in a previous study we found that NF-κB was activated in response to telomerase-inhibitor-induced senescence of alveolar type II-like A549 cells [
42]. Since telomerase has been shown to locate to mitochondria, where it decreases ROS production, inhibition of telomerase may have increased the formation of ROS, and that may in turn have activated NF-κB [
43]. Thus, the mechanism of senescence-associated inflammation may differ according to the cell types and senescence inducer. Our findings also suggest that the pathways that regulate the senescence-associated inflammation may be distinct from the pathways that regulate the senescence growth arrest, because the p38 MAPK inhibitor SB202190 substantially diminished senescence-associated inflammation (Figures
6A and
8C) but did not inhibit BrdU-induced growth arrest, p21 expression, or the increased SA β-gal activity (Figures
6C,
8D and
8E).
The increased pro-inflammatory cytokine secretion by senescent epithelial cells (Clara cells) may not be the sole mechanism responsible for the exacerbated airway inflammation in our murine model of epithelial cell senescence. Previous studies have shown that CC10, the major Clara cell secretory protein (CCSP), exerts anti-inflammatory effects and can attenuate airway inflammation through inactivation of secretory phospholipase A2 or regulation of macrophage behavior [
44,
45]. Thus, the reduced CC-10 levels in the airway fluid resulting from ineffective restoration of Clara cells due to senescence growth arrest may also contribute to the mechanism of the increased airway inflammation.
Pro-inflammatory cytokine secretion is one of the complex features of the senescence-associated secretory phenotype, which include disruption of normal tissue structure, promotion of endothelial cell invasion, and stimulation of tumor cell growth and invasion [
10,
37,
46]. Why do senescent cells mount a pro-inflammatory cytokine response? Recent evidence suggests at least two important roles of senescence-associated pro-inflammatory cytokine secretion [
10,
37,
46]. First, pro-inflammatory cytokines such as IL-6 and IL-8 act in an autocrine feedback loop to reinforce the senescence growth arrest and thereby reduce the risk of oncogenic transformation in a cell-autonomous manner [
33,
46]. Second, the pro-inflammatory cytokines mobilize innate immune cells, such as natural killer cells, that clear senescent cells [
47,
48]. These roles suggest that senescence-associated inflammation is important, especially early after senescence induction, to ensure efficient growth arrest and eventually to stimulate the immune system to clear senescent cells [
10]. However, senescent cells accumulate in the tissues with age and in the affected tissues of patients with age-related diseases such as atherosclerosis and COPD, probably because either immune clearance is less efficient and/or the rate at which senescent cells are produced outpaces the rate of clearance [
2,
6,
9,
10]. Consequently, the deleterious effects of cellular senescence, i.e., impaired tissue restoration and chronic inflammation, may become apparent with time and contribute to the pathogenesis of age-related diseases.
If that is true, does cellular senescence contribute to the onset and progression of COPD? Our findings show accelerated senescence of Clara cells in the airways of COPD patients, and they extend the findings in previous studies, including our own previous study, demonstrating that various types of cells, including alveolar type II cells, endothelial cells, fibroblasts, and peripheral blood lymphocytes, senesced more rapidly in COPD patients than in control subjects [
2‐
5]. In the present study we also demonstrated an increase in the phosphorylated form of p38 MAPK in the Clara cells of COPD patients, corroborating a previous study showing increased numbers of phospho-p38 MAPK-positive macrophages and phospho-p38 MAPK-positive alveolar cells in the lungs of COPD patients [
49]. Importantly, we found that p38 MAPK is preferentially activated by senescent Clara cells rather than by presenescent cells, indicating a correlation between p38 MAPK activation and senescence at the cellular level in vivo. There is evidence that p38 MAPK activation plays a role in recruiting CD8 T lymphocytes into the lungs of COPD patients, and a p38 MAPK inhibitor has been shown to be effective in suppressing inflammation in a model of smoking-induced COPD in mice [
49,
50]. In light of all of this evidence, senescence-associated p38 MAPK activation in Clara cells appears to contribute to the onset and progression of airway inflammation in COPD.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
FZ carried out the animal studies, the cell culture studies, and the human lung tissue studies, and drafted the manuscript. SO carried out the human lung tissue studies. NA participated in the design of the study. KA conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.