CaMKII inhibition in type II pneumocytes protects from bleomycin-induced pulmonary fibrosis by preventing Ca²⁺-dependent apoptosis

Abstract

The calcium and calmodulin-dependent kinase II (CaMKII) translates increases in intracellular Ca²⁺ into downstream signaling events. Its function in pulmonary pathologies remains largely unknown. CaMKII is a well-known mediator of apoptosis and regulator of endoplasmic reticulum (ER) Ca²⁺. ER stress and apoptosis of type II pneumocytes lead to aberrant tissue repair and progressive collagen deposition in pulmonary fibrosis. Thus we hypothesized that CaMKII inhibition alleviates fibrosis in response to bleomycin by attenuating apoptosis and ER stress of type II pneumocytes. We first established that CaMKII was strongly expressed in the distal respiratory epithelium, in particular in surfactant protein-C-positive type II pneumocytes, and activated after bleomycin instillation. We generated a novel transgenic model of inducible expression of the CaMKII inhibitor peptide AC3-I limited to type II pneumocytes (Tg SPC-AC3-I). Tg SPC-AC3-I mice were protected from development of pulmonary fibrosis after bleomycin exposure compared with wild-type mice. CaMKII inhibition also provided protection from apoptosis in type II pneumocytes in vitro and in vivo. Moreover, intracellular Ca²⁺ levels and ER stress were increased by bleomycin and significantly blunted with CaMKII inhibition in vitro. These data demonstrate that CaMKII inhibition prevents type II pneumocyte apoptosis and development of pulmonary fibrosis in response to bleomycin. CaMKII inhibition may therefore be a promising approach to prevent or ameliorate the progression of pulmonary fibrosis.

pulmonary fibrosis occurs as an idiopathic disorder or as a result of inflammatory states or injuries, such as radiation or bleomycin administration. There are currently limited treatment strategies that prevent or slow pulmonary fibrosis progression (34), suggesting that an improved understanding of disease mechanisms will be necessary to identify new therapeutic targets. The current concept of its pathogenesis encompasses an initial alveolar epithelial injury that leads to abnormal wound healing, which is associated with the formation of patchy fibroblast-myofibroblast foci that ultimately evolve to fibrosis (41).

Numerous cell types have been implicated in fibrosis formation, including macrophages and respiratory epithelium. Several studies point to type II pneumocytes as critical mediators of fibrosis development (2, 16). Patients with pulmonary fibrosis display increased endoplasmic reticulum (ER) stress (19) and apoptosis of type II pneumocytes (1), which is thought to lead to increased collagen deposition. Consistent with observations in patients, bleomycin rapidly induces apoptosis of respiratory epithelium in murine models (12). Consistently, the number of type II pneumocytes is strongly decreased within 7 days after instillation of bleomycin (13, 50). In addition, selective genetic deletion of type II pneumocytes has proven sufficient to induce fibrosis formation (43). The potential mechanisms by which compromised type II pneumocyte function lead to fibrosis include impaired homeostatic ability of the lung to replace type I cells and the loss of important signals that suppress proliferation and collagen synthesis by lung fibroblasts and production of profibrotic factors such as TGF-β by residual injured type II epithelium (50).

The Ca²⁺ and calmodulin-dependent kinase II (CaMKII) is an attractive candidate signal for promoting diseases where elevated intracellular Ca²⁺ or oxidative stress contributes to disease initiation or progression (7, 38). Under resting conditions, CaMKII is held in an inactive state. Numerous signaling events increase levels of calcified calmodulin (Ca²⁺/CaM), which then binds to and thereby activates CaMKII (10). Once activated by Ca²⁺/CaM, CaMKII can remain active even in the absence of Ca²⁺/CaM binding by oxidation or autophosphorylation (8). Active CaMKII catalyzes the phosphorylation of proteins that increase inflammatory signaling (42), apoptosis (8, 52), and cell proliferation (23). While CaMKII function is well defined in cardiac myocytes and neurons, we recently found that excessive levels of active CaMKII induces mechanisms that drive key features of allergic asthma, suggesting that CaMKII could participate in other pulmonary disease processes.

Here we show that CaMKII is robustly expressed in type II pneumocytes and activated after bleomycin exposure. In a new transgenic mouse model of type II pneumocyte-targeted CaMKII inhibition, pulmonary fibrosis was significantly reduced compared with wild-type (WT) littermates after bleomycin exposure. Conversely, CaMKII inhibition abolished apoptosis in response to bleomycin in vitro and in vivo. As a potential mechanism, we ascertained that bleomycin administration increased intracellular Ca²⁺ concentrations and ER stress. CaMKII inhibition significantly reduced ER Ca²⁺ levels and ER stress. Our data define CaMKII as a mediator of cellular mechanisms that are crucial for the development of lung fibrosis.

MATERIALS AND METHODS

Transgenic model of CaMKII inhibition in type II pneumocytes.

All animal care and housing requirements of the National Institutes of Health Committee on Care and Use of Laboratory Animals were followed. All protocols were reviewed and approved by the University of Iowa Animal Care and Use Committee. To test if CaMKII activity in type II pneumocytes affects pulmonary fibrosis in vivo, we interbred mice with tetracycline-induced (tetO) transgenic expression of a CaMKII inhibitory peptide (AC3-I) fused with eGFP (Tg tetO-AC3-I/eGFP, a kind gift of Dr. Daniel Winder, Vanderbilt University) with mice expressing the tetracycline reverse transcriptional activator (rtTA) under the control of the human surfactant, pulmonary-associated protein C [SFTPC, B6.Cg-Tg(SFTPC-rtTA)5Jaw/J, Jackson No. 06235, referred to as Tg SPC-rtTA] (30). In the double transgenic SFTPC-rtTA X tetO-AC3-I/eGFP mice (referred to as Tg SPC AC3-I mice), AC3-I/eGFP expression restricted to lung peripheral epithelial cells was induced by feeding a doxycycline-supplemented chow (625 mg/kg). AC3I transgene expression was verified by GFP fluorescence in 10-μm frozen lung sections. For all in vivo experiments, littermate Tg SPC-rtTA mice served as WT controls. Male and female mice were used in equal proportions.

Bleomycin treatment.

Intratracheal administration of bleomycin (Hospira, Lake Forest, IL) was performed in 8- to 12-wk-old male and female mice under isoflurane anesthesia. Bleomycin was instilled at a dose of 2.5 U/kg in normal saline at a final volume of 50 μl. The operator was blinded to genetic status of mice.

Histology.

On days 1, 3, or 14 after bleomycin instillation, left lungs were fixed with 4% paraformaldehyde, submerged in a vacuum overnight, and then processed by paraffin embedding as we previously described (14, 38). To assess collagen deposition, 5-μm tissue sections were cut and stained with Masson's trichrome. Images were acquired on an Olympus BX-61 light microscope.

Immunofluorescence.

Lung sections from WT mice at 24 h after saline or bleomycin instillation were probed for phosphorylated or total CaMKII using commercial anti-phospho-CaMKII (3361; Cell Signaling) and anti-total-CaMKII antibodies (LS-c100735; LS Bio). Some sections were colabeled with an anti-prosurfactant protein C antibody (AB3786; EMD Milipore). Alexa-Fluor secondary antibodies (Invitrogen, CA) were used to visualize staining. Nuclei were counterstained with TO-PRO-3 (Invitrogen). Sections were mounted with VectaShield mounting media (VectaShield). Images were acquired with a Zeiss 710 Confocal Microscope at ×40 or ×63 magnification. All images were taken at the same time and using the same imaging settings.

Terminal deoxynucleotidyl transferase dUTP-mediated nick-end labeling (TUNEL) staining was performed with the In Situ Cell Death Detection Kit, TMR Red Kit as recommended by the manufacturer (12156792910; Roche).

Hydroxyproline assay.

Lung tissue was homogenized, dried to a stable weight, and then acidified with 6 N HCl and hydrolyzed by heating at 120°C for 24 h. Hydroxyproline measurements were determined as described previously (14) and normalized to dry lung weight.

ELISA assays.

With the use of a sandwich ELISA, tumor necrosis factor-α (TNF-α; cat no. DY410), interleukin-1β (IL-1β; cat no. DY401), and macrophage inflammatory protein-2 (MIP-2; cat no. DY452; R&D Systems) were quantified in duplicate in lung homogenates as specified by the manufacturer.

Cell culture.

MLE-12 cells, a cell line derived from mouse alveolar epithelial cells that express surfactant protein C, were purchased from ATCC. Cells in passages 6–12 were used.

Cells were seeded onto collagen-coated glass coverslips or cell culture plates. MLE-12 cells were grown to 70% confluence and infected with adenovirus containing control or the CaMKII peptide inhibitor CaMKIIN for 48 h (22). Similarly to AC3-I, CaMKIIN is a CaMKII inhibitor peptide that inhibits all isoforms and splice variants of CaMKII with great sensitivity and specificity (3). MLE-12 cells were exposed to 50 mU/ml of bleomycin for indicated times.

Apoptosis assays.

Cell death was assayed by TUNEL staining. Twenty-four hours after exposure to 50 mU/ml bleomycin, MLE-12 cells were fixed in 4% paraformaldehyde and processed as recommended by the manufacturer (In Situ Cell Death Detection Kit; Roche). The nuclei were counterstained with DAPI. Images were taken at ×20 and the number of apoptotic to total number of nuclei was determined in five visual fields per condition. Three independent experiments were performed. Additional assays were performed in MLE-12 cells treated with 100 nM thapsigargin for 24 h using the annexin-based apoptosis assay for the MUSE cell analyzer (Milipore).

Cytoplasmic Ca²⁺ measurements.

MLE-12 cells were infected with adenovirus as indicated. The cells were loaded with fluo-4 AM (Life Technologies) in normal Hanks' buffered saline solution with Ca²⁺ and Mg²⁺ at room temperature for 20 min. Baseline signal recordings were performed for 5 min. Then, bleomycin (50 mU/ml) was added and fluorescence at 520 nm recorded every 30 or 60 s for the next 30 min. Additional measurements were performed with thapsigargin (100 nM), histamine (10 μM), and thapsigargin (100 nM) or in Ca²⁺-free solution. Measurements in cells treated with Hanks' buffered saline solution alone served as negative controls. The negative control data were subtracted from the measurements with agonists. The results are displayed as fold change.

Moreover, cytoplasmic Ca²⁺ responses were assessed by confocal microscopy in Rhod-2-loaded cells (1 μmol/l; Life Technologies) after 10 min of bleomycin treatment, using with a Zeiss LSM 510 confocal microscope. Images were taken at ×20 and five visual fields were analyzed using ImageJ.

Immunoblotting.

Fifty-microgram aliquots of whole tissue lysate from MLE-12 cells were resolved by SDS-polyacrylamide electrophoresis and transferred to polyvinylidene difluoride membrane. After blocking with 5% nonfat milk, the membranes were incubated with primary antibodies for p-CaMKII (3361S; Cell Signaling) CaMKII (#LS-c100735; LS Bio), transcription factor C/EBP homologous protein (CHOP; ABC 955; EMD Millipore,), the molecular chaperone BiP (3138; Cell Signaling), and GAPDH (5174; Cell Signaling) followed by horseradish peroxidase-conjugated secondary antibody. The proteins were visualized with the ECL chemiluminescence system (Amersham). Densitometry was performed using National Institutes of Health ImageJ software.

RT-PCR.

Total RNA was isolated from mouse lungs per manufacturer's recommendation (Qiagen) followed by digestion with proteinase K and DNase I to eliminate genomic DNA contamination. cDNA was transcribed from 1 μg RNA using Superscript III enzyme (Invitrogen) and random nanomer primers, followed by PCR amplification and visualization by agarose gel electrophoresis. The primers for AC3-I are as follows: AC3-I forward: 5′-GCA AGG CAG TCA ACT GCC TCC TGG-3′ and AC3-I reverse: 5′-ATG GTG AGC AAG GGC GAG GAG CTG-3′.

Statistical analysis.

All values in the text and figures are presented as means ± SE. Statistical significance of multiple treatments was determined using GraphPad Prism Software Version 6 by Student's t-test or ANOVA followed by Sidak or Tukey multiple comparison test when appropriate. P < 0.05 was considered significant.

RESULTS

CaMKII is expressed in type II pneumocytes and activated in response to bleomycin.

We first established that CaMKII is expressed in the lung parenchyma by immunofluorescence. In addition to robust expression in respiratory endothelium as we described in the past (38), CaMKII protein was present in alveolar epithelium (Fig. 1A). Its presence in type II pneumocytes was confirmed through colocalization with the prosurfactant protein C (Fig. 1B).

Fig. 1.

calcium and calmodulin-dependent kinase II (CaMKII) expression in distal respiratory epithelium. A: immunofluorescence for CaMKII (green) in murine lung sections. Nuclei, TO-PRO-3 (blue). No primary antibody serves as negative control (×20). B: prosurfactant protein C and CaMKII colocalize in distal respiratory epithelium. Left and middle: prosurfactant protein C (red) or CaMKII (green), TO-PRO-3 (blue), and bright field image. Right: merged image of prosurfactant protein C, CaMKII, and TO-PRO-3. Arrows denote cells with positive staining for CaMKII and prosurfactant protein C (×63). C: immunofluorescence for active phosphorylated CaMKII (p-CaMKII: green; nuclei TO-PRO-3: blue) in murine lung sections at baseline and 24 h after bleomycin instillation (×63).

Next, we investigated whether CaMKII is activated in pneumocytes after bleomycin instillation. At baseline, low levels of active phosphorylated CaMKII were detected (Fig. 1C). After exposure to bleomycin for 24 h, labeling for active phosphorylated CaMKII was strongly increased, suggesting that CaMKII is activated after bleomycin treatment through Ca²⁺/calmodulin-dependent CaMKII autophosphorylation. No differences in CaMKII protein levels were seen after bleomycin treatment (data not shown).

Transgenic model of CaMKII inhibition in type II pneumocytes.

To assess whether CaMKII in type II pneumocytes contributes to apoptosis and bleomycin-induced fibrosis, we developed a transgenic model that expresses the CaMKII inhibitor peptide AC3-I in type II pneumocytes when doxycycline is supplemented in the chow (Fig. 2A). As in most tissues outside the nervous system, CaMKII isoforms-γ and -δ are expressed in the lung (47). AC3-I is a peptide inhibitor that serves as a pseudosubstrate and inhibits all isoforms of CaMKII with >100-fold selectivity relative to PKC, PKA, and CaMKIV (30). Similar models of CaMKII inhibition by expression of AC3-I as a fusion protein with eGFP have been validated in other tissues and cell types (38, 51). We ascertained that the transgene RNA is only expressed in lungs of double transgenic Tg SPC-AC3-I mice compared with littermate controls after treatment with doxycycline-containing chow (Fig. 2B). Accordingly, in WT littermate mice, no significant eGFP fluorescence was detected in TG SPC-rtTA or in TG AC3-I/eGFP littermate control mice after administration of doxycycline-containing chow (Fig. 2C). In contrast, eGFP-positive cells were seen throughout the alveoli in double transgenic mice Tg SPC-AC3-I mice (Fig. 2C).

Fig. 2.

Transgenic model of CaMKII inhibition in type II pneumocytes. A: schematic overview of AC3-I expression in SPC-rtTA X AC3-I/eGFP mice (Tg SPC AC3-I). B: RT-PCR for AC3-I reveals induction of transgene expression in Tg SPC AC3-I mice only after administration of doxycycline-containing chow for 7 days. C: immunofluorescence for eGFP (green) and nuclei (DAPI) demonstrates AC3-I/eGFP expression only in Tg SPC AC3-I mice but not in Tg SPC-rtTA or Tg AC3-I/eGFP control mice after administration of Doxycycline-containing chow for 7 days (×10).

Pulmonary fibrosis is decreased after CaMKII inhibition in type II pneumocytes.

Because apoptosis of type II pneumocytes ultimately leads to lung fibrosis (43) and CaMKII controls apoptosis in other cell types and tissues, we assessed whether mice with CaMKII inhibition in type II pneumocytes are protected from bleomycin-induced fibrosis. We performed Masson's trichrome staining of lung sections from WT and Tg SPC-AC3-I mice at baseline and on day 14 after bleomycin exposure (Fig. 3A). In WT mice, bleomycin induced dense collagen deposition and pronounced destruction of lung architecture, whereas Tg SPC-AC3-I mice had essentially normal lung parenchyma.

Fig. 3.

CaMKII inhibition in type II pneumocytes abrogates bleomycin-induced pulmonary fibrosis. A: representative Masson's trichrome staining in lung sections from wild-type (WT) and Tg SPC-AC3-I mice at baseline and on day 14 after bleomycin administration (×20). B: hydroxyproline assay in lungs in untreated littermate WT (Tg SPC-rtTA) and Tg SPC-AC3-I mice on day 14 after bleomycin administration. (n = 3–8 per group). C: terminal deoxynucleotidyl transferase dUTP-mediated nick-end labeling (TUNEL) staining (red), immunofluorescence for prosurfactant protein C (green), and nuclear counterstain (DAPI: blue) in lungs on day 3 after exposure to bleomycin or saline (cntrl). Arrows indicate apoptotic type II pneumocytes. D: quantification of apoptotic type II pneumocytes (positive for TUNEL and prosurfactant protein C staining) (n = 3 visual fields, 3 samples per group). *P < 0.05.

Next, we performed hydroxyproline assays in samples from Tg SPC-AC3-I mice and WT littermate at baseline and on day 14 after bleomycin instillation. Hydroxyproline levels were similar in Tg SPC-AC3-I mice and WT littermates at baseline (Fig. 3B). In contrast, the concentration of hydroxyproline was significantly increased in WT mice on day 14 after treatment, whereas collagen production was almost near the control concentration in Tg SPC-AC3-I mice. Since early apoptosis of type II pneumocytes initiates the disease process that leads to fibrosis, we tested whether CaMKII inhibition protects from apoptosis in vivo. In WT mice, increased apoptosis of type II pneumocytes was detected, whereas, as predicted, type II pneumocytes in Tg SPC-AC3-I mice were protected (Fig. 3, C and D). These data strongly suggest that CaMKII has a key role in the development of a fibrotic phenotype, possibly by decreasing bleomycin-induced apoptosis of type II pneumocytes.

CaMKII inhibition in type II pneumocytes decreases the inflammatory response to bleomycin.

Because bleomycin-induced lung injury is, in part, mediated by a proinflammatory state, we tested whether CaMKII inhibition in type II pneumocytes is sufficient to block the inflammatory response of lung tissue to bleomycin injury. For this purpose, we performed ELISA assays in whole lung lysates from Tg SPC-AC3-I mice and WT littermates on days 3 or 14 after instillation of bleomycin or normal saline. While increases in cytokines were present in bleomycin- compared with saline-treated control samples on day 3, no differences between the genotypes were detected. In contrast, on day 14, the overall concentrations of MIP-2 (17), IL-1β (32), and TNF-α (29, 31) were greatly increased in bleomycin-treated lungs compared with day 3 and a reduction, especially in TNF-α and MIP-2, was seen in Tg SPC AC3-I mice compared with WT mice (Fig. 4, A–C).

Fig. 4.

CaMKII inhibition in type II pneumocytes decreases the inflammatory response to bleomycin. ELISA assays for TNF-α (A), IL-1β (B), and macrophage inflammatory protein-2 (MIP-2; C) in homogenates of lungs from WT or TG SPC AC3-I mice on days 3 and 14 after treatment with bleomycin or normal saline (cntrl) [n = 7 (WT cntrl), 4 (Tg SPC AC3-I cntrl), 13 (WT Bleo), and 8 (Tg SPC AC3-I Bleo)]. * P < 0.05.

CaMKII inhibition blocks bleomycin-induced apoptosis and ER stress in MLE-12 cells.

To understand the mechanisms by which CaMKII inhibition in type II pneumocytes reduces pulmonary fibrosis, we conducted in vitro experiments in MLE-12 cells, an immortalized murine mouse epithelial cell line with type II characteristics. First, we tested whether bleomycin induces CaMKII activity by performing immunoblots for activated CaMKII. Similarly to our findings in vivo (Fig. 1C), the level of p-CaMKII was increased at 1 and 4 h (Fig. 5, A and B). Since CaMKII autophosphorylation depends on binding of calcified calmodulin, these findings suggest that the intracellular Ca²⁺ levels are increased after bleomycin application. Next, we examined whether CaMKII inhibition protects from bleomycin-induced apoptosis. For this purpose, we performed TUNEL staining in cells that were infected with control adenovirus or an adenovirus that expresses the potent and specific CaMKII inhibitor CaMKIIN (22). In untreated MLE-12 cells, very few TUNEL-positive cells were detected regardless of the adenovirus used (Fig. 5, C and D). Incubation with bleomycin for 24 h significantly increased the number of TUNEL-positive cells infected with control adenovirus. In contrast, significantly fewer apoptotic cells were detected after infection with CaMKIIN (Fig. 5, C and D). ER stress is believed to mediate type II pneumocyte apoptosis and thereby predispose to the development of lung fibrosis after treatment with bleomycin (45). To assess whether CaMKIIN protected from ER stress, we performed immunoblots for the ER stress markers CHOP in bleomycin-treated MLE-12 cells. While the protein levels of CHOP and BiP were strongly increased with bleomycin treatment, the response was abolished with CaMKII inhibition (Fig. 5, E–G).

Fig. 5.

CaMKII inhibition prevents apoptosis and endoplasmic reticulum (ER) stress in type II pneumocytes in vitro. A: immunoblots for CaMKII, active phosphorylated p-CaMKII in MLE-12 cells after bleomycin treatment as indicated. NT, no treatment. B: quantification of 3 independent experiments. The relative increase in signal intensity compared with untreated conditions was determined by densitometry and is indicated as fold increase. C: representative TUNEL staining in MLE-12 cells that were transduced with control adenovirus or adenovirus expressing CaMKIIN for 48 h before bleomycin or sham treatment for additional 24 h. D: quantification of TUNEL-positive cells in 3 independent experiments; 5 visual fields per conditions were analyzed. E: representative immunoblot for CHOP and BiP in MLE-12 cells after bleomycin treatment as indicated. MLE-12 cells were infected with control adenovirus (C) or adenovirus expressing CaMKIIN (N) for 48 h before bleomycin was added. Quantification of CHOP (F) and BiP immunoblots (G), adjusted to GAPDH. Data are the average of 5 (CHOP) and 4 (BiP) independent experiments. *P < 0.05.

CaMKII controls intracellular Ca²⁺ release and ER Ca²⁺ content in MLE-12 cells.

Since CaMKII promotes apoptosis, in part through regulation of ER Ca²⁺ uptake (36), we next assessed whether bleomycin increased intracellular Ca²⁺ levels in MLE-12 cells by fluo-4 fluorescence. Bleomycin induced a robust increase in intracellular Ca²⁺ that persisted over 30 min (Fig. 6, A and B). The response was blunted when the experiment was performed under nominally Ca²⁺-free conditions (Fig. 6, B and C), suggesting that an influx of Ca²⁺ from the extracellular space contributes to the bleomycin-dependent increase in cytoplasmic Ca²⁺. The sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA) inhibitor thapsigargin induces apoptosis in many cell types through depletion of ER Ca²⁺ stores (28). Therefore, as a positive control, we tested the response of MLE-12 cells to thapsigargin treatment. As expected, thapsigargin induced MLE-12 cell apoptosis (Fig. 6D). The addition of histamine to thapsigargin is predicted to maximize the Ca²⁺ release from the ER, and as expected, it further increased intracellular Ca²⁺ levels. In direct comparison, the cytoplasmic Ca²⁺ levels were insignificantly greater after bleomycin treatment compared with thapsigargin (Fig. 6, E and F). This finding implies that the increase in intracellular Ca²⁺ after bleomycin treatment is sufficient to induce apoptosis. Lastly, we tested the effect of CaMKII inhibition on cytoplasmic Ca²⁺ levels. CaMKII inhibition significantly reduced the bleomycin-induced increase in Ca²⁺ (Fig. 6, G and H). These findings were confirmed in separate experiments by confocal microscopy with the fluorescent dye Rhod-2 (Fig. 6I). Our data demonstrate that bleomycin increases cytoplasmic Ca²⁺ to a level comparable to thapsigargin and that CaMKII inhibition blocks the Ca²⁺ response to bleomycin. Since thapsigargin induces apoptosis through ER Ca²⁺ release and ER stress (37), it is very feasible that the reduced ER Ca²⁺ release under CaMKII inhibition diminishes the ER stress response and blunts apoptosis.

Fig. 6.

CaMKII regulates ER Ca²⁺ release in response to bleomycin. A: intracellular Ca²⁺ levels by fluo-4 fluorescence after bleomycin treatment for 10 min in MLE-12 cells. B: cytoplasmic Ca²⁺ levels were determined in MLE-12 cells during treatment with 50 mU/ml bleomycin in Hanks' buffered saline solution with and without Ca²⁺. Negative control data were subtracted from the measurements with agonists. A representative experiment is shown. C: area under the curve (AUC) was determined in 4 independent experiments and plotted as fold increase. D: apoptosis in MLE-12 cells after treatment with 100 nM thapsigargin for 24 h. E: cytoplasmic Ca²⁺ levels in MLE-12 cells during treatment with 50 mU/ml bleomycin, 100 nM thapsigargin, and 100 nM thapsigargin with 10 μM histamine. A representative experiment is shown. F: AUC was determined for each condition in 4 independent experiments and plotted as fold increase. G: cytoplasmic Ca²⁺ levels in MLE-12 cells that were infected with adenovirus expressing the CaMKII inhibitor peptide CaMKIIN or control for 48 h during treatment with 50 mU/ml bleomycin in Hanks' buffered saline solution with Ca²⁺. A representative experiment is shown. H: AUC was determined in 5 independent experiments and plotted as fold increase. I: Rhod-2 fluorescence in MLE-12 cells infected with adenovirus expressing the CaMKII inhibitor peptide CaMKIIN or control after bleomycin treatment for 10 min. *P < 0.05.

DISCUSSION

Apoptosis of type II pneumocytes plays an important role in the development of pulmonary fibrosis, but specific molecular mediators that regulate this process have not been identified. Here, we demonstrate that CaMKII is robustly expressed in type II pneumocytes and activated by bleomycin. Bleomycin-induced pulmonary fibrosis is abrogated in a novel transgenic model of CaMKII inhibition in type II pneumocytes. Consistently, hydroxyproline levels and inflammatory cytokines are significantly lower in lungs of mice with CaMKII inhibition. After bleomycin instillation, apoptosis of type II pneumocytes is greatly reduced in mice with CaMKII inhibition. In vitro experiments in MLE-12 cells revealed that bleomycin-induced apoptosis and ER stress are suppressed with CaMKII inhibition. Numerous apoptotic pathways including ER stress are induced by a rise in intracellular Ca²⁺ levels (28). Bleomycin increased intracellular Ca²⁺ levels in part through Ca²⁺ release from the ER. CaMKII inhibition significantly reduced the ER Ca²⁺ loading and release in MLE-12 cells. Thus we propose that bleomycin induces apoptosis and ER stress of type II pneumocytes at least in part through increasing cytoplasmic Ca²⁺ levels and that CaMKII controls apoptosis through regulating Ca²⁺ levels. Since apoptosis of type II pneumocytes impairs the homeostatic ability of the lung to replace type I cells, CaMKII inhibition in type II pneumocytes is sufficient to prevent pulmonary fibrosis. Thus CaMKII inhibition may be a promising novel approach to prevent or halt the progression of pulmonary fibrosis.

While intensively studied as a decoder of Ca²⁺-dependent signals in neurons and cardiac myocytes, CaMKII expression and function in the lung have remained largely unknown. A single report recently demonstrated that CaMKII is expressed in nonsmall cell lung cancer in vivo (21) where it promotes proliferation. In vitro, CaMKII function was studied in an adenocarcinoma human alveolar basal cell line (21) and respiratory smooth muscle (39) in the context of proliferation and inflammation. Moreover, CaMKII regulates surfactant secretion in rat alveolar type II cells by modulating lamellar body calcium (4). We recently developed a model of CaMKII inhibition in Clara cells and reported that CaMKII promotes mucus production, expression of inflammatory cytokines, eosinophil invasion, and airway hyperreactivity in vivo and in vitro models of allergic asthma (38). In this project, we concentrated on CaMKII expression and function in alveolar epithelium where we detected robust expression in surfactant C-positive type II pneumocytes. Our data suggest that CaMKII is an important yet widely unrecognized regulator of pulmonary pathophysiology.

Type II pneumocytes play various roles in alveolar fluid balance, coagulation/fibrinolysis, and host defense. Type II pneumocytes proliferate, differentiate into type I cells, and remove apoptotic type II pneumocytes cells by phagocytosis, thus contributing to epithelial repair. Deletion of type II pneumocytes is sufficient to induce pulmonary fibrosis (43). The bleomycin model is widely used to understand the various cellular and molecular pathways that contribute to pulmonary fibrosis, including apoptosis of type II pneumocytes (50). However, the mechanism by which bleomycin induces apoptosis is controversial since conflicting data providing evidence for activation of the intrinsic (48) and extrinsic (27) pathways have been reported. Oxidative stress-dependent mechanisms are widely accepted as upstream activators of bleomycin-induced apoptosis (48), although a few studies have directly implicated Ca²⁺ (18, 26). Interestingly, a recent study demonstrated that the administration of the Ca²⁺ channel blocker nifedipine significantly decreases extracellular matrix deposition after bleomycin exposure (26). Nifedipine administration had no effect on total inflammatory cell counts, indicating that its protective effect on lung fibrosis was due to the profibrotic rather than the inflammatory response to bleomycin (26). While initially believed to be a specific inhibitor of the L-type Ca²⁺ channel, nifedipine also blocks store-operated Ca²⁺ entry (SOCE) (5). Our data demonstrate that the Ca²⁺ response to bleomycin is dependent on influx of Ca²⁺ from the extracellular space. Ca²⁺ entry in type II pneumocytes is mainly dependent on transient receptor potential (TRP) channels (9) and SOCE via its molecular components stromal interaction molecule 1 (STIM1) and Orai1 (6a). In addition, the L-Type Ca²⁺ channel has been identified in type II pneumocyte-derived cell lines (6). Several lines of evidence link these channels to CaMKII. CaMKII was recently reported to promote TRP channel activity in endothelial cells (44). Conversely, knockdown of STIM1/Orai1 inhibits CaMKII activation. Moreover, CaMKII is known to phosphorylate the β-subunit of the L-Type Ca²⁺ channel, leading to increased opening probability of the channel (11, 33). Thus we speculate that CaMKII regulates Ca²⁺ channels in type II pneumocytes and that decreased influx of Ca²⁺ from the extracellular space drives the reduction in ER Ca²⁺ content under CaMKII inhibition.

Data from various other cell types have identified CaMKII as a regulator of cell proliferation and migration. Moreover, CaMKII has been reported to promote apoptosis (8, 36, 52). Salas et al. (36) demonstrated that CaMKII controls apoptosis through regulation of phosphorylation of ER proteins. In particular, CaMKII phosphorylates the SERCA inhibitor phospholamban at Thr-17 and thereby reverses the inhibition of SERCA (24). The de-inhibition of SERCA facilitates ER Ca²⁺ loading. Apoptosis is then induced upon ER Ca²⁺ release in response to proapoptotic stimuli such as ischemia-reperfusion (36). Several mechanisms for Ca²⁺ regulation of apoptosis have been reported in the last several decades (25). For example, cytochrome c is released from mitochondria after an initial insult and binds to and activates the IP₃ receptor. As a result, Ca²⁺ is released from the ER. The rise in cytoplasmic Ca²⁺ induces a mass exodus of cytochrome c from mitochondria, which then elicit a robust apoptotic response (25). Since we detected decreased ER Ca²⁺ loading in MLE-12 cells with CaMKII inhibition and were able to induce apoptosis with thapsigargin through ER Ca²⁺ depletion, we propose that bleomycin induces apoptosis in part through ER Ca²⁺ release and ER stress by a mechanism that includes CaMKII. These data are in line with other reports that induction of ER stress in type II pneumocytes predisposes to enhanced pulmonary fibrosis after bleomycin, at least in part by increased alveolar epithelial cell apoptosis (20, 45). Of note, exaggerated ER stress responses in type II pneumocytes have been reported in patients with idiopathic pulmonary fibrosis (19). Moreover, studies in a variety of cell types have provided evidence that CaMKII induces ER stress (46).

Apart from its role in type II pneumocytes and ER Ca²⁺ loading, CaMKII may contribute to pulmonary fibrosis through other mechanisms, for example, via impaired secretion of surfactant (4). In addition, CaMKII was recently identified in mitochondria in cardiac myocytes where it regulates ischemia-induced apoptosis (15). Thus CaMKII may selectively control mitochondria-dependent apoptotic pathways. While we concentrated on the role of type II pneumocytes in pulmonary fibrosis, numerous other cell types contribute to its pathogenesis, including macrophages. Recent studies have demonstrated that Ca²⁺-dependent ER stress in macrophages contributes to pulmonary fibrosis in the chrysotile asbestos model (35). Since CaMKII is expressed in macrophages (40), we cannot rule out that CaMKII also regulates Ca²⁺-dependent ER stress leading to pulmonary fibrosis in this cell type. Regardless, we found that selective inhibition of CaMKII in type II pneumocytes was sufficient to protect against collagen deposition and the development of pulmonary fibrosis. It is possible that global CaMKII inhibition may provide a greater protection against fibrosis than selective blockade in type II pneumocytes. Future studies that examine the therapeutic benefit of CaMKII inhibition in the lung following injury are warranted.

GRANTS

This work was supported by a grant from the Spear Family Pulmonary Research Fund (to O. A. Jaffer), National Institutes of Health Grants R01-HL-108932 (to I. M. Grumbach) and 2R01-ES-015981-06 (to A. B. Carter) and a fellowship from the Iowa Center for Research By Undergraduates (to C. J. Winters). This work was also supported by Department of Veterans Affairs, Office of Research and Development, Biomedical Laboratory Research and Development Grant 1 BX001135-03 (to A. B. Carter).

DISCLAIMERS

The content of this manuscript is solely the responsibility of the authors and do not necessarily represent the views of the granting agencies.

DISCLOSURES

No conflicts of interest, financial or otherwise are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: C.J.W., O.M.K., S.C.S., O.A.J., A.B.C., and I.M.G. conception and design of research; C.J.W., O.M.K., S.M., C.A., J.D.P., and O.A.J. performed experiments; C.J.W., O.M.K., S.M., C.A., S.C.S., J.D.P., O.A.J., A.B.C., and I.M.G. analyzed data; C.J.W., O.M.K., S.M., S.C.S., O.A.J., A.B.C., and I.M.G. interpreted results of experiments; C.J.W., O.M.K., S.C.S., and I.M.G. prepared figures; C.J.W., O.M.K., S.M., C.A., O.A.J., A.B.C., and I.M.G. edited and revised manuscript; C.J.W., O.M.K., S.M., C.A., S.C.S., O.A.J., A.B.C., and I.M.G. approved final version of manuscript; I.M.G. drafted manuscript.