Research ArticlePulmonary fibrosis

Caveolin-1–derived peptide limits development of pulmonary fibrosis

See allHide authors and affiliations

Science Translational Medicine  11 Dec 2019:
Vol. 11, Issue 522, eaat2848
DOI: 10.1126/scitranslmed.aat2848

Limiting lung disease

Fibrotic foci that form during idiopathic pulmonary fibrosis (IPF), a type of progressive and fatal interstitial lung disease, alter lung architecture, leading to cell death and loss of lung function. Marudamuthu et al. developed a caveolin-1–derived peptide, CSP7, that inhibited apoptosis of alveolar epithelial progenitor cells and activation of fibrotic lung fibroblasts. When delivered to three mouse models of lung fibrosis, CSP7 reduced extracellular matrix deposition, promoted epithelial cell survival, and improved lung function. CSP7 was also effective when administered ex vivo to lung tissue or cells isolated from people with end-stage IPF. These results support further development of CSP7 as a potential treatment.

Abstract

Idiopathic pulmonary fibrosis (IPF) is a fatal fibrotic lung disease with a median 5-year survival of ~20%. Current U.S. Food and Drug Administration–approved pharmacotherapies slow progression of IPF, providing hope that even more effective treatments can be developed. Alveolar epithelial progenitor type II cell (AEC) apoptosis and proliferation, and accumulation of activated myofibroblasts or fibrotic lung fibroblasts (fLfs) contribute to the progression of IPF. Full-length caveolin-1 scaffolding domain peptide (CSP; amino acids 82 to 101 of Cav1: DGIWKASFTTFTVTKYWFYR) inhibits AEC apoptosis and fLf activation and expansion and attenuates PF in bleomycin (BLM)–induced lung injury in mice. Like full-length CSP, a seven–amino acid deletion fragment of CSP, CSP7 (FTTFTVT), demonstrated antifibrotic effects in murine models of lung fibrosis. When CSP7 was administered during the fibrotic phase in three preclinical models [single-dose BLM, repeated-dose BLM, and adenovirus expressing constitutively active transforming growth factor–β1 (Ad-TGF-β1)–induced established PF], CSP7 reduced extracellular matrix (ECM) markers characteristic of PF, increased AEC survival, and improved lung function. CSP7 is amenable to both systemic (intraperitoneal) or direct lung delivery in a nebulized or dry powder form. Furthermore, CSP7 treatment of end-stage human IPF lung tissue explants attenuated ECM production and promoted AEC survival. Ames testing for mutagenicity and in vitro human peripheral blood lymphocyte and in vivo mouse micronucleus transformation assays indicated that CSP7 is not carcinogenic. Together, these findings support the further development of CSP7 as an antifibrotic treatment for patients with IPF or other interstitial lung diseases.

INTRODUCTION

The pathogenesis of interstitial lung diseases (ILDs), including idiopathic pulmonary fibrosis (IPF), is characterized by defective alveolar epithelial progenitor type II cell (AEC) renewal and aberrant proliferation and accumulation of activated myofibroblasts and fibrotic lung fibroblasts (fLfs), which produce an excess of extracellular matrix (ECM). IPF is the most common form of ILD and is characterized by spatial and temporal heterogeneity of usual interstitial pneumonitis lesions. IPF is a progressive and fatal lung disease, with a 5-year survival of only 20% after diagnosis (1). Epidemiological studies reveal that the incidence of IPF is estimated between 5 and 20 cases per 100,000 (2). Although the cause of IPF is unknown, environmental and occupational exposures, tobacco smoking, rheumatoid arthritis, comorbidities like gastroesophageal reflux disease, and genetic polymorphisms appear to be important contributing factors (3).

Currently, two U.S. Food and Drug Administration–approved compounds are marketed for the treatment of IPF in the United States. Pirfenidone, a compound that abrogates transforming growth factor β–1 (TGF-β1)–stimulated collagen synthesis and down-regulates proinflammatory cytokines, slowed disease progression over a 52-week treatment period (4). Nintedanib, a broad-spectrum tyrosine kinase inhibitor, also slowed the progression of disease in patients with IPF (5). These drugs reduce the rate of acute exacerbation of IPF (which often leads to death); however, neither treatment is known to be curative, and many patients experience side effects that reduce their quality of life.

Although the cause of IPF remains unclear, studies indicate that AEC apoptosis precedes activation of fLfs during formation of fibrotic foci (6). Activated fLfs secrete excessive amounts of ECM proteins, which obliterate functional alveolar units, decreasing gas exchange and thereby reducing lung function. Increased accumulation of ECM by fLfs also contributes to basement membrane disruption and AEC death. In addition, an increased influx of circulating fibrocytes has been observed in the lungs of patients with IPF (7). In both of these dysfunctional cell types in IPF, altered p53 signaling contributed to AEC apoptosis (8), fLf activation (9), and subsequent development of PF. In addition, we and others observed that the caveolin-1 (Cav1) scaffolding domain 20-mer peptide CSP inhibited p53 in AECs, which attenuated AEC apoptosis after lung injury and prevented development of PF (8, 1016).

Here, we demonstrate that the 7-mer deletion fragment of CSP, CSP7 (FTTFTVT), concurrently attenuates the expansion and profibrogenic activity of fLfs and AEC apoptosis in fibrotic lungs. CSP7 delivered systemically or locally reduced PF and inhibited AEC apoptosis in three mouse models of established PF. Human end-stage IPF lung tissues treated with CSP7 ex vivo inhibited profibrogenic protein expressions and suppressed p53 and caspase-3 activation in AECs. These observations support further development of CSP7 as a therapeutic for PF.

RESULTS

CSP and CSP7 exhibit antifibrotic effects in fLfs in vitro

Primary fLfs isolated from human lungs with end-stage IPF or from mice with established PF were treated with full-length CSP, CSP7, or a scrambled control peptide (CP). The conditioned medium was analyzed for soluble collagen 1α1 protein (COL1), and Western blotting was performed on cell lysates for α–smooth muscle actin (αSMA), platelet-derived growth factor–β (PDGFR-β), tenascin-C (TN-C), and p53. Treatment of human (Fig. 1A) fLfs (hfLfs) or mouse fLfs (mfLfs) (Fig. 1B) with CSP reduced COL1, PDGFR-β, TN-C, and αSMA, whereas naïve fLfs or fLfs exposed to CP still displayed elevated COL1 and αSMA. CSP-induced changes were associated with a parallel increase in the basal p53 expression, which was markedly reduced in untreated fLfs. To identify the minimal sequence capable of inducing antifibrotic responses, overlapping deletions of CSP were made, human fLfs were treated with the fragments, and activity was assessed. A 7–amino acid fragment of CSP, CSP7 (FTTFTVT), reduced COL1, TN-C, and αSMA in human fLfs (Fig. 1C). Treatment with CSP7 induced p53, indicating a link between increased p53 expression and the antifibrotic effects of these peptides. Further truncation or mutation of CSP7 abolished its ability to suppress αSMA in both hfLfs and mfLfs (Fig. 1D). Analyses of αSMA, COL1, fibronectin (FN), and TN-C mRNAs from hfLFs and mfLfs confirmed that CSP7 contained the minimal sequence required for the antifibrotic effects (Fig. 1E). Cellular metabolic activity measured by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay revealed that CSP7 significantly (P < 0.001) inhibited proliferation of hfLfs but did not affect the viability of human non-fLfs (hnLfs) (Fig. 1F).

Fig. 1 Suppression of profibrogenic responses in fLfs by CSP or its deletion peptides.

(A) Human non-fLfs (hnLfs) and hfLfs (hfLfs) isolated from the lung tissues of healthy donors and patients with IPF or (B) mouse non-fLfs (mnLfs) and mouse fLfs (mfLfs) from uninjured mice and mice with BLM-induced PF, respectively, were treated with PBS, CSP (10 μM), or CP for 48 hours. Conditioned media and cell lysates were immunoblotted. (C) A series of overlapping deletions were made in CSP, and hfLfs were treated with these peptides and assessed for p53 and profibrogenic proteins to identify the minimum amino acids required for the CSP effects. (D) A series of truncations/mutations were made in CSP7 and tested in hfLfs and mfLfs using αSMA expression to assess bioactivity. (E) Total RNA from hfLfs and mfLfs treated with CSP7 or truncated or mutated CSP7 were analyzed for COL1, αSMA, FN, and TN-C mRNA by qRT-PCR. The values are presented as bar graphs after normalization with β-actin transcript. (F) hnLfs and hfLfs treated with PBS, CSP7, or CP for 48 hours were subjected to MTT assay to assess proliferation. The experiments were repeated two (A to D) or three times (E and F). Data in (E) and (F) are represented as means ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 were obtained by one-way analysis of variance (ANOVA) with Turkey’s multiple comparison test.

Daily CSP7 mitigates intranasal single-dose BLM-induced PF

We previously showed that CSP protects against intranasal single-dose (1×) bleomycin (BLM) (2 U/kg)–induced PF (8, 10). Next, we investigated whether CSP7 provides similar protection when administered by daily intraperitoneal (1.5 mg/kg) injection from day 14 (d14) to d20 after 1× intranasal BLM (8 U/kg) injury. Analyses were performed 24 hours after the last treatment (d21). As shown in Fig. 2A, 14 days after BLM injury, a significant (P < 0.001) increase in total hydroxyproline (a moiety specific to all forms of collagen) was observed in wild-type (WT) C57BL/6 mice. Hydroxyproline remained elevated through d21. In addition, hematoxylin and eosin (H&E) and Masson’s trichrome histological analyses showed established PF at d14 and d21 after BLM injury (Fig. 2B). Intraperitoneal (IP) injection of CSP7 improved overall survival (Fig. 2C). Quantitative chest micro–computed tomography (CT) renditions (Fig. 2D) indicated that BLM exposure caused robust injury by d21, which was markedly reduced in CSP7-treated mice. Lung volumes were calculated from CT renditions at full inspiration, and compliance and resistance were measured by forced oscillation plethysmography Scientific Respiratory Equipment Inc. (SCIREQ) at 21 days post-BLM suggested restrictive lung remodeling. These changes were significantly (P < 0.01) improved after CSP7 treatment (Fig. 2, E to G). Likewise, reductions in lung weight (Fig. 2H), matrix proteins, PF (Fig. 2I), and total lung hydroxyproline content (Fig. 2J) were observed. CSP7 also suppressed the soluble collagen content of lung tissues (Fig. 2K).

Fig. 2 Intraperitoneal injection of CSP7 delivered daily limits 1× BLM-induced PF in mice.

Mice exposed to saline or BLM by intranasal instillation were euthanized 7 to 21 days later. (A) Total lung hydroxylproline (n = 4 to 5 per group). (B) H&E- and trichrome-stained lung sections. BLM-exposed mice received intraperitoneal injections with or without CSP7 or CP (1.5 mg/kg) daily for 7 days starting d14 after BLM injury. (C) Percent survival during 7-day treatment relative to number of mice (n = 19 per group) on d14 after BLM. (D) Representative micro-CT images on d21 after BLM. Lung volumes (E) were measured by quantitative-CT renditions (n = 4 to 6 per group), and lung compliance (F) and elastance (G) were measured using a flexiVent system (n = 3 per group). (H) Lung weights presented as a bar graph (n = 3 per group). (I) H&E- and trichrome-stained lung sections. Lung homogenates from mice treated with CSP7 from two sources were tested for (J) hydroxyproline (n = 10 per group) and (K) soluble collagen (n = 15 per group), (L) profibrogenic proteins, or (M) mRNA (n = 3 per group). Lung homogenates were tested for TGF-β and CTGF (N) protein, (O and P) mRNA (n = 3 per group), and (Q) MPO (n = 4 to 5 per group). Total (R) and differential (S) counts from BALF were determined (n = 3 per group). The experiments were repeated two to three times. Bars indicate means ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 were obtained by one-way ANOVA with Turkey’s multiple comparison and log-rank tests, respectively. Scale bars, 100 μm (B and I).

Immunoblotting of mouse lung homogenates showed a marked increase in profibrogenic proteins: COL1, FN, αSMA, and TN-C. CSP7 reversed this response (Fig. 2L). Consistent with changes in profibrogenic marker proteins, quantitative reverse transcription polymerase chain reaction (qRT-PCR) confirmed significant (P < 0.05, P < 0.01, or P < 0.001) inhibition of BLM-induced profibrogenic mRNAs after treatment with CSP7 (Fig. 2M). Furthermore, CSP7 markedly reduced the lung expression of profibrogenic cytokines such as TGF-β and connective tissue growth factor (CTGF) protein (Fig. 2N) and mRNAs (Fig. 2, O and P), which were otherwise markedly increased during BLM injury.

Analysis of lung homogenates and lung sections was performed to explore the mechanisms contributing to these effects. These analyses demonstrated that CSP7 reduced both TGF-β receptor I (TGF-βRI) and TGF-βRII expression and a member of the embryonic lethal abnormal vision (ELAV1)/Hu family of RNA binding proteins HuR, which was otherwise increased in BLM-induced mouse fibrotic lung tissues. These changes were associated with concurrent reduction in phosphorylation of SMAD2/3 in mice treated with BLM and exposed to CSP7 (fig. S1). Elevated pulmonary myeloperoxidase (MPO) activity (Fig. 2Q) and bronchoalveolar lavage fluid (BALF) inflammatory cell numbers were also observed, indicative of increased inflammation, after BLM exposure; CSP7 reduced total lung MPO activity and BALF inflammatory cells. Differential staining further revealed an increase in the proportion of lymphocytes and neutrophils, with a relative reduction in monocytes in the lungs of mice exposed to BLM (Fig. 2, R and S). These changes were also reversed by CSP7.

Daily intraperitoneal CSP7 alleviates TGF-β1–induced PF and improves lung function

To further test the therapeutic potential of CSP7 against PF induced by TGF-β1, adenovirus expressing constitutively active TGF-β1 (Ad-TGF-β1) [109 plaque-forming units (PFU)], a viral vector that constitutively expresses active TGF-β1, was intranasally instilled in WT mice. Controls received an empty viral vector (Ad-Ev). Fourteen days after receiving Ad-TGF-β1, mice received daily intraperitoneal injection of CSP7 or CP (1.5 mg/kg) for 2 weeks (d14 to d27) and were euthanized for analyses on d28. CSP7-treated mice showed improved overall survival (Fig. 3A). Increased opacifications were observed on micro-CT scans of mice 28 days after transduction with Ad-TGF-β1, and CSP7 reduced these opacifications (Fig. 3B). Consistent with the findings in 1× BLM–induced PF, a significant (P < 0.01) reduction in lung volume and compliance with an increase in resistance was observed 28 days after Ad-TGF-β1 transduction (Fig. 3, B to E). CSP7 protected against the Ad-TGF-β1–induced decline in lung function.

Fig. 3 Intraperitoneal injection of CSP7 delivered daily inhibits progression of TGF-β1–induced established PF.

Mice were intranasally instilled with Ad-TGF-β1 to induce PF. Control mice were exposed to Ad-Ev. After 14 days, Ad-TGF-β1–treated mice received intraperitoneal injections of CSP, CSP7, or CP (1.5 mg/kg). (A) The percent survival 14 to 28 days after-Ad-TGF-β1 presented as a line graph relative to total number mice (n = 13 per group) at the beginning of CSP7 treatment. (B) Representative micro-CT images at 28 days after Ad-TGF-β1. (C) Lung volumes were measured using quantitative-CT renditions (n = 3 to 7 per group). Lung (D) compliance and (E) elastance were obtained by the flexiVent system (n = 3 to 9 per group). (F) Lung weights presented as bar graph (n = 3 per group). (G) Lung sections stained with H&E and trichrome. Scale bars, 100 μm. Whole-lung homogenates analyzed for (H) total hydroxyproline and (I) soluble collagen content (n = 5 to 6 per group). (J) Total protein and (K) RNA (n = 3 per group) from lung homogenates tested for profibrogenic marker protein and mRNA, respectively. (L) Neutrophil accumulation measured by MPO assay in lung homogenates (n = 4 per group). Lung lavage (M) total and (N) differential numbers of leukocytes (n = 3 per group). The experiments were repeated two times. Bars represent means ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 were obtained by log-rank and one-way ANOVA with Turkey’s multiple comparison tests, respectively.

CSP7 also reduced the Ad-TGF-β1–induced increase in total lung weight (Fig. 3F). Both CSP and CSP7 reduced TGF-β1–mediated induction of collagen and other ECM proteins with relatively well-preserved lung architecture, indicating mitigation of PF (Fig. 3G), which was otherwise markedly increased in untreated mice or CP-treated mice. Likewise, a significant (P < 0.01) reduction in total lung hydroxyproline and soluble collagen was observed (Fig. 3, H and I). Analyses of whole-lung homogenates for COL1, αSMA, FN, and TN-C protein by Western blotting (Fig. 3J) or mRNA by qRT-PCR (Fig. 3K) demonstrated reductions in the expression of these markers in CSP- and CSP7-treated animals. We also observed a reduction in pulmonary MPO activity (Fig. 3L), inflammatory cells in the BALF, and decreased numbers of leukocytes and the proportion of lymphocytes and neutrophils in mice transduced with Ad-TGF-β1 treated with CSP7 (Fig. 3, M and N).

Daily intraperitoneal CSP7 attenuates repetitive-dose BLM–induced PF

Next, mice were exposed to eight intranasal doses of BLM (2 U/kg) to establish irreversible PF over 16 weeks. Two weeks after the final BLM dose, mice were either not treated or treated with intraperitoneal injection of CSP, CSP7, or CP (1.5 mg/ml) once daily for two consecutive weeks. Twenty weeks later, animals were euthanized after lung function analyses. Although the average body weight or survival between various groups was not significantly different during the last 2 weeks of CSP7 treatment (Fig. 4A), increased mortality was noted in all mice exposed to BLM due to development of PF by week 18. Micro-CT scanning revealed markedly attenuated PF in CSP7 animals compared to CP or untreated controls (Fig. 4B). A significant (P < 0.05) reduction in lung volume and function was also observed, which was reversed after CSP7 treatment (Fig. 4, C to E). CSP7 treatment also reduced total lung weight, which was otherwise significantly (P < 0.001) increased due to BLM-induced PF (Fig. 4F). H&E and trichrome staining of lung sections (Fig. 4G) showed reduced ECM after CSP or CSP7 treatment, indicating mitigation of PF. CSP and CSP7 also reduced total lung hydroxyproline and soluble collagen (Fig. 4, H and I). Western blotting of lung homogenates (Fig. 4J) or qRT-PCR (Fig. 4K) revealed that both CSP and CSP7 reduced COL1, FN, αSMA, and TN-C. CSP7 also down-regulated MPO activity in whole-lung homogenates (Fig. 4L) and decreased total inflammatory cells in the BALF (Fig. 4M). However, CSP7 treatment failed to significantly alter BLM-induced changes in the proportion of monocytes, lymphocytes, and neutrophils (Fig. 4N).

Fig. 4 PF induced by chronic repetitive-dose BLM was attenuated by CSP or CSP7 treatment in mice.

Mice were exposed to BLM (2 U/kg) once every 2 weeks for a total of 8× in 4 months by intranasal instillation. Two weeks after the final dose of BLM, BLM-treated mice (n = 10 per group) received intraperitoneal injections of CSP, CSP7, and CP (1.5 mg/kg) daily for 2 weeks. (A) The percent survival during 2 weeks of CSP7 treatment presented as a line graph. (B) Representative micro-CT images 2 weeks after peptide treatment. (C) Lung volumes were measured using quantitative-CT renditions (n = 3 to 4 per group). Lung (D) compliance and (E) elastance were obtained by the flexiVent (n = 3 to 5 per group). (F) Lung weights presented as bar graph (n = 3 per group). (G) Lung sections stained with H&E and trichrome to assess PF. Scale bars, 100 μm. The lung homogenates were tested for (H) hydroxyproline and (I) soluble collagen (n = 6 to 10 per group). The lung protein and RNA were analyzed for profibrogenic (J) proteins and (K) mRNAs (n = 3 per group). (L) Neutrophil accumulation was measured by MPO assay in lung homogenates (n = 4 per group). (M) Total and (N) differential counts of lung lavage leukocytes (n = 3 per group). The experiments were repeated twice. Bars depict means ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 were obtained by one-way ANOVA with Turkey’s multiple comparison test and log-rank tests, respectively.

Modified CSP7 formulated to improve stability retains antifibrotic activity

Because intraperitoneal injection is not a feasible way to deliver drugs to humans, we developed a nebulized formulation of CSP7 using a vibrating mesh nebulizer to allow direct delivery to the lung. First, we assessed whether capping the CSP7 peptide would improve its antifibrotic effects in the BLM model. As shown in Fig. 5A, both capped and uncapped CSP peptides inhibited BLM-induced increments in the expression of COL1, FN, and TN-C. This finding was confirmed by measurement of total lung hydroxyproline content (Fig. 5B). Next, we tested whether mannitol, a sugar alcohol that is sometimes used as a drug stabilizer in nebulized formulations, affects the antifibrotic activity of CSP7. Treatment of hfLfs with CSP7 with or without mannitol for 48 hours reduced expression of COL1, FN, and TN-C (Fig. 5C). Similarly, CSP7 containing mannitol reduced the expression of profibrogenic proteins in mice with BLM-induced PF (Fig. 5D). To confirm stability, CSP7 was dissolved in either Dulbecco’s phosphate-buffered saline (DPBS) or normal saline (NS) at a concentration of 0.5 mg/ml and then nebulized via an Aeroneb Pro (vibrating mesh nebulizer) or EZ Breathe Atomizer (vibrating mesh nebulizer) as previously described (17). Percent recovery at each step (in red boxes) was calculated considering input as 100% (Fig. 5E). Mass spectroscopy of the nebulized fluid yielded a single peak, indicating structural stability (Fig. 5F). Nebulized CSP7 in DPBS suppressed COL1, FN, and TN-C expression. Baseline-reduced p53 expression in fLfs was also restored to that observed in nLfs (Fig. 5G). As shown in Fig. 5H, treatment of two human fLf lines with CSP7 resuspended in a stabilizer solution containing lactose monohydrate showed antifibrotic activity, whereas CP-suspended stabilizer solution or stabilizer solution alone was ineffective.

Fig. 5 Effect of CSP7 modification and nebulization on antifibrotic activity and peptide structure.

WT mice with 1× BLM-induced PF were treated with capped CSP7 (using acetyl group at the N termini) or uncapped CSP7 and CP daily for 7 days, starting d14 after BLM injury. The lung homogenates were immunoblotted for (A) profibrogenic proteins or analyzed for (B) total hydroxyproline content (n = 4 to 5 per group). (C) hfLfs treated with PBS, CSP, or CSP7 with (CSP7-M) or without (CSP7) mannitol for 48 hours were immunoblotted for profibrogenic proteins. (D) The lung homogenates from mice treated with CSP7 or CSP7-M as in (A) were immunoblotted for profibrogenic proteins. (E) Percent recovery of nebulized CSP7. (F) Mass spectroscopy of fluids collected after nebulization. (G) Immunoblots of lysates of hfLfs exposed to DPBS, CSP7, CP, or CSP7 nebulized using Aeroneb Pro (CSP7N) analyzed for protein expression. m/z, mass/charge ratio. (H) Immunoblots of lysates from two hfLf lines (#9 and #17) treated with DPBS containing lactose solution (PBS) or formulated with CSP7 or CP (50 μM) in DPBS containing lactose for 48 hours analyzed for profibrogenic proteins. The experiments were repeated two times. Bars show mean ± SD. ****P < 0.0001 were obtained by one-way ANOVA with Turkey’s multiple comparison test.

Nebulization of CSP7 inhibits intranasal 1× BLM-induced PF

Mice with BLM-induced established PF were treated with nebulized CSP7 for 2 hours (0.05 mg kg−1 day−1) daily for 7 days (d14 to d20) using a nose-only inhalation exposure chamber. Overall survival of mice exposed to BLM with PF improved significantly (P < 0.05) with CSP7 treatment (Fig. 6A), and there was a trend toward an improvement in body weight (Fig. 6B). As shown in Fig. 6C, lungs of mice exposed to BLM showed a significant (P < 0.001) increase in lung weight, which was reduced by nebulized CSP7. Total lung hydroxyproline (Fig. 6D) and soluble collagen (Fig. 6E) were reduced, as were expression of profibrogenic proteins (Fig. 6F) and mRNAs (Fig. 6G). Nebulization of CSP7 reduced BLM-induced increments in lung homogenate concentrations of TGF-β and CTGF protein (Fig. 6H) and their mRNAs (Fig. 6, I and J). Lung collagen and other matrix protein deposition were reduced (Fig. 6K) consistent with resolution of PF. Last, nebulized CSP7 significantly (P < 0.05) inhibited 1× BLM-induced pulmonary MPO activity (Fig. 6L).

Fig. 6 Nebulization of CSP7 delivered daily resolved single-dose BLM–induced PF.

WT mice were exposed to saline or BLM (8 U/kg). After 14 days, mice exposed to BLM were treated with/without placebo, formulated CSP7, or CP containing lactose monohydrate stabilizer for 2 hours daily for 7 days using a nose-only nebulization tower. (A) The percent survival from 14 to 21 days after BLM injury in each group (n = 20 per group) presented as a line graph relative to total number of mice at the beginning (as 100%). (B) The body weights of treated and untreated animals were measured and represented as a bar graph (n = 5 to 9 per group). All mice were euthanized 21 days after BLM exposure. (C) Lung weights presented as bar graph (n = 4 to 9 per group). Whole-lung homogenates were analyzed for (D) total hydroxyproline (n = 5 to 10 per group) and (E) soluble collagen (n = 5 to 7 per group). (F) Total lung proteins were immunoblotted for profibrogenic proteins. (G) Whole-lung RNA was tested for changes in profibrogenic mRNAs (n = 3 per group). Lung homogenates were tested for TGF-β and CTGF (H) protein and (I and J) mRNA (n = 4 per group). (K) H&E and trichrome staining of lung sections. Magnification, ×20. Scale bars, 100 μm. (L) MPO assay of lung homogenates for pulmonary neutrophil accumulation (n = 3 to 4 per group). The experiments were repeated twice. Bars represent mean ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 were obtained by log-rank and one-way ANOVA with Turkey’s multiple comparison tests, respectively.

Dry powder inhalation of CSP7 protects against 1× intratracheal BLM-induced PF

To expand local delivery options of CSP7 to directly treat the damaged lung and to avoid solubility, stability, and degradation issues associated with the liquid formulation, CSP7 (acetate counterion) was micronized by air jet milling for dry powder inhalation (DPI). In vitro aerodynamic particle size distribution characterization was conducted using a next-generation impactor (NGI) as described earlier (18). The results indicated that the DP formulation of CSP7 is ideally suited for human deep lung deposition because 84.5% of all particles fell within respirable fractions of aerodynamic size less than 5 μm (Fig. 7A). To assess the efficacy of CSP7 against BLM-induced established PF, we exposed mice to 1× of BLM (2.8 U/kg) by orotracheal instillation. Consistent with intranasal exposure, orotracheal instillation of 1× of BLM caused the anticipated PF in mice. This was attenuated by DPI of air jet–milled CSP7 (0.05 mg kg−1 day−1) administered from d14 to d20 after BLM exposure. All mice survived the CSP7 DPI treatment. Consistent with systemic administration of CSP7 (1.5 mg/kg) by intraperitoneal injection, airway delivery by DPI resulted in a significant (P < 0.01) decrease in lung weight, which was otherwise increased in mice with BLM-induced PF (Fig. 7B). Daily DPI of micronized CSP7 resulted in a significant (P < 0.001) reduction in total lung hydroxyproline and soluble collagen content (Fig. 7, C and D). As shown in Fig. 7E, DPI of CSP7 caused a marked reduction in BLM-induced increase in profibrogenic marker proteins, TGF-β and CTGF. These findings were also corroborated by mRNA expression (Fig. 7F). Furthermore, CSP7 delivered locally by DPI or intraperitoneal injection reduced total lung MPO activity induced because of BLM injury (Fig. 7G). Trichrome- and H&E-stained lung sections revealed a marked reduction in BLM-induced collagen and other ECM deposits (Fig. 7H). The results of these studies show that airway delivery of CSP7 by DPI was as efficacious as intraperitoneal injection in reducing BLM-induced existing PF at a dose about 30 times less than that used for intraperitoneal injection.

Fig. 7 DPI of air jet–milled CSP7 mitigates single-dose BLM–induced PF in mice.

(A) The aerosol performance of micronized CSP7 DP was evaluated by NGI, showing the particle deposit in the NGI stages. CSP7 exhibited high fine particle fraction (FPF) with respirable aerodynamic size (repeated three times). MMAD, mass median aerodynamic diameter; GSD, geometric standard deviation; MOC, micro-orifice collector. WT mice (n = 10 per group) were exposed to either saline or BLM (2.8 U/kg) by intratracheal instillation. After 14 days, BLM-treated mice were left untreated (none) or exposed to air jet–milled CSP7 using a rotating-brush generator, with a consistent chamber concentration of 0.1 mg/liter for 15 min daily for 7 days. Mice exposed to unprocessed CSP7 (1.5 mg kg−1 day−1) by intraperitoneal injection for 7 days were used as positive control. All mice were euthanized 21 days after BLM exposure. (B) Lung weights were measured and presented as bar graph (n = 5 per group). Whole-lung homogenates from mice (n = 3 to 10 per group) were analyzed for (C) total hydroxyproline and (D) soluble collagen contents. Total lung homogenates were immunoblotted for profibrogenic marker and cytokine (TGF-β and CTGF) (E) proteins and (F) mRNAs (n = 3 per group). (G) Pulmonary neutrophil accumulation was measured by MPO assay of lung homogenates (n = 3 per group). (H) Lung sections were H&E- and trichrome-stained. Magnification, ×20. Scale bars, 100 μm. The experiments were repeated two times. Bars indicate mean ± SD. *P < 0.05, **P < 0.01, and ***P < 0.001 were obtained by one-way ANOVA with Turkey’s multiple comparison test.

CSP7 inhibits AEC p53 expression and apoptosis in mice with early lung injury or late-stage established PF and in human end-stage IPF lung explants treated ex vivo

We previously reported that intraperitoneal administration of CSP immediately after lung injury (d1 to d3) prevented the development of PF by inhibiting AEC apoptosis via inhibition of p53 (8, 10). We extended this work by next testing the effects of various fragments of CSP 24 hours after BLM exposure. AECs were isolated 2 days later and tested for activation of caspase-3. Peptide fragments harboring FTTFTVT protected AECs (Fig. 8A). The purity of the AEC preparation was >90% as confirmed by staining isolated AECs from uninjured mice for inclusion bodies and surfactant protein-C (SP-C) (Fig. 8B). CSP7 significantly (P < 0.001) reduced BLM-induced AEC apoptosis in vitro as measured by lactate dehydrogenase (LDH) cytotoxicity (Fig. 8C). CSP and CSP7 decreased BLM-induced p53 expression: The process was associated with p53 degradation due to inhibition of serine-15 phosphorylation and acetylation of lysine-379 (Fig. 8D). Furthermore, CSP and CSP7 suppressed the activation of caspase-3. Increased numbers of TUNEL (terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate nick end labeling)–positive cells were detected in the lungs of mice exposed to BLM (Fig. 8E). TUNEL-positive AECs were reduced in the lungs of mice treated with CSP7 but not CP for 7 days starting 14 days after BLM. Immunohistochemistry (IHC) staining for activated caspase-3 demonstrated that CSP7 treatment of mice with BLM-induced PF markedly reduced AEC apoptosis. Similarly, AECs of mice with Ad-TGF-β1–induced PF showed elevated total/acetylated and serine-15–phosphorylated p53, along with increased caspase-3 activation. These changes were reversed by treatment with CSP7 (Fig. 8F). Elevated numbers of TUNEL-positive cells were observed in lung sections from mice with Ad-TGF-β1–induced PF (Fig. 8G), including those treated with CP, indicating that they were undergoing apoptosis; however, lungs from mice treated with CSP7 showed a reduction in TUNEL-positive AECs. Inhibitory effects of CSP7 on AEC apoptosis were confirmed by analyses of lung sections for activated caspase-3. Consistent with the findings of the other two mouse models, analyses of isolated AECs from the lungs of mice exposed to 8× BLM for cleaved caspase-3, acetylated and serine-15–phosphorylated, and total p53 revealed that CSP7-mediated inhibition of AEC apoptosis involves suppression of AEC p53 (Fig. 8H). Lung sections showed minimal TUNEL-positive AECs (Fig. 8I). The beneficial effects of CSP7 were independently confirmed by antigen staining of the same lung sections for activated caspase-3, which suggests that CSP7 also promotes AEC survival to exert robust beneficial effects in fibrotic lungs.

Fig. 8 CSP7 inhibits p53 and apoptosis in AECs of fibrotic lungs.

(A) CSP deletion fragment sequences. Boxed region shows FTTFTVT. Mice received intraperitoneal injections of CSP deletion fragments or CP (1.5 mg/kg) 1 day after 1× BLM. AECs isolated 3 days later were immunoblotted. (B) Lithium carbonate and SP-C–immunofluorescence staining of uninjured AECs. DAPI, 4′,6-diamidino-2-phenylindole. (C) LDH assay of cultured medium from mouse naïve AECs or AECs treated with BLM (40 μg/ml) alone, BLM + CSP7, or BLM + CP in vitro (n = 3 to 6 per group). (D) Pooled AECs from mice (n = 4 per group) receiving intraperitoneal injections of CSP, CSP7, or CP (1.5 mg/kg) daily for 7 days starting d14 after BLM injury were tested for p53/apoptosis. (E) Lung sections were TUNEL-stained and IHC-stained for apoptosis [cleaved caspase-3 (Cl. cas3)]. (F) Pooled AECs from mice (n = 4 to 5 per group) treated intranasally with Ad-TGF-β1 before daily intraperitoneal injections of CSP7 or CP (1.5 mg/kg) for 2 weeks starting 14 days after transduction were tested for p53/apoptosis. (G) TUNEL and IHC of lung sections. (H) Pooled AECs from mice (n = 4 to 5 per group) exposed to BLM (2 U/kg) by intranasal instillation once every 2 weeks for a total of 8× in 4 months, treated with intraperitoneal injections of CSP7 or CP (1.5 mg/kg) daily for 2 weeks, starting 2 weeks after the final dose of BLM, were tested for p53 and apoptosis. (I) TUNEL and IHC of lung sections. Magnification, ×20. Scale bars, 100 μm. (J) Pooled human normal and IPF lung tissues (n = 4) untreated or treated with CSP7 or CP ex vivo were immunoblotted. (K) AECs isolated from tissues treated as in (J) were immunoblotted. (L) Lysates of AECs isolated from control and IPF tissues treated in vitro as in (J) were immunoblotted. The experiments were repeated twice. Bars represent means ± SD. ***P < 0.001 was obtained by one-way ANOVA with Turkey’s multiple comparison test.

We next treated human IPF lung tissues with CSP7 ex vivo and compared the responses with similarly treated control lung tissues from donors without PF. Baseline expression of COL1 and αSMA were elevated in IPF lung tissues (Fig. 8J), and treatment with CSP7 reduced profibrogenic marker expression. Expression remained elevated in naïve IPF lung tissues or IPF lung tissues exposed to CP for 72 hours (Fig. 8J). CSP7 had minimal, if any, effects on normal lung tissues. Isolated AECs from IPF lung tissues showed a marked increase in p53 and activated capase-3. However, AECs isolated from IPF lung tissues treated with CSP7 ex vivo showed suppression of both p53 and apoptosis (Fig. 8K). Treatment of control lung tissues with CSP7 failed to alter low-baseline p53 expression or apoptosis, suggesting selective targeting of injured AECs by CSP7. To confirm that CSP7 inhibits p53 expression and apoptosis in injured AECs of IPF lungs, we isolated AECs from IPF lungs and treated them with CSP7 in vitro. As shown in Fig. 8L, treatment with CSP7 markedly reduced p53 expression and activation of caspase-3; however, p53 and apoptosis remained elevated in untreated or CP-treated AECs from IPF lungs. Uninjured AECs isolated from non-IPF lung tissue failed to respond to CSP7, suggesting that CSP7 specifically targets injured AECs with dysregulated p53 expression.

CSP7 is not mutagenic nor clastogenic

To assess the safety of CSP7, we subjected CSP7 to various tests. The results of in vitro Ames testing did not produce any positive mutagenic response at CSP7 concentrations as high as 5000 μg per plate (table S1). Furthermore, micronucleus screening for chromosomal aberration in human peripheral blood lymphocytes treated with CSP7 (3.4 to 500 μg/ml) in vitro did not cause an increase in micronuclei at any concentration tested (table S2). Similarly, mice treated with CSP7 (≤12 mg/kg) showed no mortality, signs of lethargy, or piloerection in either sex. In addition, flow cytometry of peripheral blood samples of mice exposed to CSP7 (0, 3, 6, or 12 mg/kg) for 48 hours did not show any increase in micronucleated polychromatic erythrocytes (table S3), suggesting that CSP7 was not clastogenic at these doses.

DISCUSSION

We and others previously reported that CSP inhibits BLM-induced AEC apoptosis and prevents development of PF (8, 1016). CSP also inhibits AEC apoptosis in mice exposed to silica or cigarette smoke (11, 19, 20). Preventing development of PF by inhibiting early lung epithelial injury is a conceptually sound approach, whereby CSP or CSP7 iterations could be of potential benefit even in relatively advanced stages of disease. Myofibroblasts that derive from resident Lfs or circulating fibrocytes likewise contribute to the pathogenesis of IPF, and we recently found that plasminogen activator inhibitor-1 regulates expansion of these cells. Here, we found that CSP reduced COL1 and αSMA expression in primary fLfs isolated from fibrotic lungs of patients with IPF or mice with BLM-induced established PF. In addition, deletion studies of CSP revealed that a seven–amino acid fragment (CSP7), FTTFTVT, was the minimal sequence needed to maintain antifibrotic activity because further deletion or mutation of CSP7 resulted in loss of its ability to inhibit COL1 or αSMA expression.

This study explored the therapeutic potential of CSP7, which was initially tested in a mouse model of 1× BLM-induced PF (9). A comprehensive panel of imaging, physiological, and biochemical analyses confirmed the potent antifibrotic activity of CSP7. The ability of CSP7 to significantly (P < 0.001) inhibit hydroxyproline and soluble collagen contents in the lung tissues of mice exposed to a higher dose of BLM (8 U/kg) provides important insights about the potency of CSP7 to resolve established PF. We also demonstrated that CSP7 inhibited expression of cytokines such as TGF-β and CTGF, profibrotic marker proteins, and their mRNAs, observations relevant to the therapeutic mechanisms of PF mitigation. Together, our findings demonstrated the in vivo benefits of CSP7 against existing PF.

Transduction of Ad-TGF-β1 produces a fibrotic lung phonotype with extensive deposition of collagen and other matrix proteins and recruitment of myofibroblasts after a brief inflammatory phase in rodents (21, 22). To test the potential of CSP and CSP7 to resolve TGF-β1–induced established PF, we treated mice with CSP and CSP7 daily for 2 weeks, starting d14 after Ad-TGF-β1 transduction. Here, again, imaging, tissue, and biochemical analyses revealed benefits of CSP7 that were comparable to those of CSP against existing PF. Treatment with either CSP or CSP7 caused a significant (P < 0.01) inhibition of hydroxyproline and soluble collagen contents in lung tissues. Both CSP and CSP7 suppressed TGF-β1–induced expression of profibrotic marker proteins and mRNAs.

Although the preclinical 1× BLM and Ad-TGF-β1 mouse models of PF have played important roles in our understanding of the paradigms associated with fibrotic lung remodeling, they fail to recapitulate features of end-stage human IPF (2, 23). Recent literature suggests that repetitive exposure to BLM results in a molecular signature and pathological phenotype that is more similar to end-stage IPF, as inflammation is less pronounced and the rate of PF development is protracted (24). Therefore, we treated mice with established 8× BLM-induced PF with CSP or CSP7 daily for 2 weeks, starting 2 weeks after the last dose of BLM. Measurement of total hydroxyproline and collagen contents in the lungs revealed that treatment with either CSP or CSP7 inhibited established PF.

We also found that both uncapped and N-terminal capping of CSP7 using acetyl group showed robust antifibrotic properties. Similarly, excipients such as lactose or mannitol in the nebulization formulation did not affect the antifibrotic effects of CSP7. Mice with BLM-induced PF that were treated with CSP7- and saline-treated control mice tend to gain more body weight than BLM controls. CSP and CSP7, when therapeutically administrated, inhibited AEC apoptosis by blocking p53 induction through suppression of serine-phosphorylated and acetylated p53. Our results show that CSP7 treatment of mice with existing PF not only affects apoptosis of AECs but also inhibits matrix deposition by activated fLfs, which likewise contribute to the fibrotic response. In addition, our observations demonstrate that CSP7 is involved in the suppression of p53 in injured AECs. These studies show that concurrent targeting of both apoptotic AECs and activated fLfs by CSP or CSP7 can effectively resolve existing PF.

In the lung epithelium after BLM injury, an increase in endogenous Cav1 expression appears to induce not only AEC apoptosis but also growth arrest and senescence (19, 20, 2527). These effects potentiate alveolar epithelial injury and prevent proper re-epithelialization due to loss of AECs renewal and differentiation, which stimulates fibrotic remodeling (8, 10, 12, 19). We previously found that Cav1 expression is lost in fLfs (9), whereas TGF-β/TGF-βR–mediated phosphorylation of SMAD2/3 is increased, leading to increased profibrogenic responses. Our data now reveal that CSP7 treatment of mice with PF reduced TGF-βRI and TGF-βRII expression and downstream SMAD2/3 signaling in the lung. These results suggest that CSP7 may promote the internalization and degradation of TGF-βR to cause the observed antifibrotic responses. In addition, CSP7 reduced TGF-β and CTGF protein and mRNA expression. Furthermore, accumulation of leukocytes and AEC apoptosis in lung tissues were reduced. Consistent with induction of TGF-β, we found increased expression of HuR, an mRNA-binding protein that augments TGF-β mRNA stabilization (28), in mice with BLM-induced PF. Increased TGF-β expression can induce CTGF (29), TGF-βR (30), and PDGFR-β expression, all of which are overexpressed in IPF fibroblasts. These were all reduced after treatment with CSP7, suggesting that it acts via multiple targets to exert robust antifibrotic effects.

Because intraperitoneal injection is not a tenable means of drug delivery in humans, a nebulized form of CSP7 was developed for direct lung delivery, and efficacy was assessed in mice. Nebulization of CSP7 in liquid formulation significantly reduced total lung weight (P < 0.05), total hydroxyproline (P < 0.05), and soluble collagen (P < 0.001) content in whole-lung tissues, which are otherwise increased in mice with BLM-induced PF. CSP7 also markedly reduced ECM deposition in the lung tissues by histologic analyses. Given that DP formulations retain structural integrity with shipping and storage and can be locally delivered to the airway, a micronized form of CSP7 was developed and tested. Micronized CSP7 significantly reduced total lung hydroxyproline (P < 0.001) and soluble collagen (P < 0.0001) contents in mice with BLM-induced existing PF. These changes were associated with marked reduction in BLM-induced expression of profibrogenic markers and collagen and other matrix protein deposits in the lung homogenates. Suppression of profibrogenic proteins in IPF lung tissues treated with CSP7 ex vivo occurred with concurrent inhibition of p53 expression and apoptosis in AECs. There were no detectable effects in normal lung tissues. These observations underscore the potential of CSP7 to resolve fibrotic lung diseases including IPF. Preliminary characterization of micronized CSP7 using an NGI indicated that the formulation was likewise capable of promoting effective delivery to the human lung parenchyma.

A limitation of our study is that none of the mouse models fully recapitulate all of the cardinal manifestations of human IPF, such as fibrotic foci, a protracted and progressive disease course, and temporal heterogeneity (31). Therefore, three different mouse models were used to authenticate the therapeutic potential of CSP7. In addition, we evaluated the beneficial effects of CSP7 on human and mouse Lfs and AECs treated in vitro, and human IPF lung explants treated ex vivo, to relate to human IPF. We acknowledge that CSP/CSP7 may mitigate PF via acting on multiple cell types and attenuation of both inflammatory and fibrotic pathways in vivo. For example, CSP7 suppressed the influx of inflammatory cells into the lungs, which might be due to inhibited leucocyte recruitment (32). However, the relevance of inflammation to pathogenesis of IPF is contested (33). We also recognize that in vitro or ex vivo response of isolated AECs, Lfs, or IPF tissues treated with CSP7 could differ from their responses in live animals. Mouse models are tightly controlled, which facilitates the detection of efficacy in responses to drug treatment. Therefore, the efficacy of CSP7 treatment in patients with IPF given diverse genetics, effects of comorbidities, and other factors could be more challenging. Nonetheless, our findings provide a strong predicate for further exploration of the complex mechanisms that underlie the antifibrotic effects of CSP7.

In summary, data generated from multiple mouse models demonstrate that CSP7 effectively reduces existing PF in a range of well-accepted preclinical models. This candidate intervention targets multiple cellular signaling pathways. Our findings demonstrate that CSP7 enhances AEC viability via a mechanism of action involving attenuation of p53 while concurrently slowing the expansion of activated fLfs by increasing expression of p53. These effects limit the progression of PF. The intervention appears to be well tolerated and offers a candidate for the treatment of fibrosing lung diseases such as IPF or other ILDs.

MATERIALS AND METHODS

Study design

Previous studies from our laboratory (811, 19) and others (13, 32, 34, 35) have implicated Cav1 in modulating pulmonary, dermal, heart, and kidney fibrosis and CSP inhibiting development of fibrosis in multiple organs. However, whether CSP7 is an effective therapeutic for fibrotic lung diseases is unknown. Therefore, our primary objective was to perform a rigorous evaluation of the efficacy of CSP7 administered systemically or locally in multiple (1× and 8× BLM and Ad-TGF-β1) mouse models of PF. The human relevance of the beneficial effects of CSP7 was tested by analyzing p53, profibrogenic, and proapoptotic marker proteins in lung fibroblasts, AECs, and lung tissue explants from patients with IPF and control donors treated ex vivo. A minimum of 8 to 10 mice per group of both sexes were used in each in vivo experiment to meet the power analyses. Most in vitro and in vivo experiments were performed at least twice. Moreover, three distinct lung injury models, three distinct formulations, and two routes of administration of CSP7 were assessed. All data points were included. All mice were weighed before the beginning of the experiments. Mice with approximately equal body weights were divided into two groups. One group was exposed to vehicles, and the second group was exposed to BLM or Ad-TGF-β1 to induce PF. To mitigate bias because of disparities in lung injury before initiating CSP7 or other intervention, the mice were redistributed by percentage of total weight lost because of lung injury such that the average weight loss of each treatment group was within 1%. End points were chosen on the basis of our expertise and literature regarding preclinical evaluation of antifibrotic therapeutics (36). Treatment outcome was determined by analyzing the expression of profibrogenic marker proteins and their mRNAs in total lung homogenates, matrix protein deposition, micro-CT scans, pulmonary function, and overall survival in the lung fibrosis model. Peptides from three sources were validated and delivered by different routes in different injury models to extend the scope of reproducibility. Furthermore, peptides of scrambled sequence, deletion, and mutant peptides with or without overlapping CSP7 sequences were used to establish the specificity of CSP7 effects. Experiments were performed by multiple members, and labeling of vehicle, test, and CPs was concealed from the laboratory personnel by giving different serial numbers each time to avoid bias. Laboratory personnel were blinded of sample and tissue labeling during analyses by providing serial numbers rather than group names. In vivo micronucleus assays were done using mice and were performed by BioReliance (Rockville, MD). Additional details are provided in the Supplementary Materials.

Testing of CSP or CSP7 intervention in mouse models of BLM-induced and Ad-TGF-β1–induced established PF

WT mice were exposed to saline or BLM by intranasal or intratracheal instillation as described earlier (10, 12). Fourteen days after 1× BLM treatment, mice exposed to BLM received intraperitoneal injections daily [200 μl of vehicle or CSP or CSP7 or CP (1.5 mg/kg)] for 7 days. For local delivery, mice exposed to BLM were treated with formulated CSP7 in lactose monohydrate/Hank’s balanced salt solution using a nose-only inhalation system for 2 hours per day for 1 week. For DPI, mice exposed to BLM were exposed to air jet–milled CSP7 using a rotating brush generator for 15 min daily for 7 days. Twenty-one days after BLM injury, mice were evaluated for changes in PF. For the Ad-TGF-β1 model, WT mice were exposed to Ad-TGF-β1 (109 PFU) in saline via the nose as described earlier (25) to induce PF. Control WT mice were exposed to empty viral vector (Ad-Ev) in saline. Fourteen days later, mice treated with Ad-TGF-β1 received intraperitoneal injections with either vehicle or CSP or CSP7 (1.5 mg/kg) daily for the next 14 days. All mice were tested for changes in PF 28 days after Ad-TGF-β1 transduction. WT mice were exposed to BLM (2 U/kg) once every 2 weeks for 4 months by intranasal instillation under anesthesia to induce PF (24). Two weeks after the last dose of BLM exposure, mice received intraperitoneal injections daily with CSP or CSP7 (1.5 mg/kg) for the next 2 weeks. At the end of 2 weeks of peptide therapy, the mice were euthanized, and their lungs were analyzed for PF.

Treatment of human lung tissues with CSP7

Lung tissues from control subjects and patients with IPF, and AECs and fLfs isolated from these tissues, were treated with or without CSP7 for 72 hours ex vivo or in vitro. Lung homogenates and AEC and fLf lysates were immunoblotted.

Analyses of lung hydroxyproline, soluble collagen, and profibrogenic markers

These analyses were performed as previously described (12). Whole-lung homogenates, conditioned medium, and cell lysates were tested to assess COL1, FN, α-SMA, and TN-C protein and mRNAs. H&E and trichrome staining and IHC analyses of lung sections were performed as we previously reported (8, 9).

MPO activities and BALF cell count determinations

MPO activities in the lung tissues were quantified using a colorimetric assay. For BALF, mouse lungs were lavaged thrice with 0.8 ml of saline and pooled. Leukocytes were counted with a hemocytometer. Cytospin preparations of BAL cells were stained using a Kwik-Diff staining kit for differential counts.

Aerodynamic particle size distribution test

The aerodynamic characterization of the micronized CSP7 DP was characterized by NGI test. Briefly, DP was filled into size three hydroxypropyl methycellulose (HPMC) capsules (Capsugel) and aerosolized using an RS01 monodose DP inhaler (high resistance). Aerosols were produced over 4 s at an air flow rate of 60 liter/min to enable the DPI to achieve an inhalation volume of 4 liters and 4-kPa pressure drop across the device. NGI collection surfaces were coated with 2.5% (v/v) Tween 20 in methanol. One capsule was actuated from the DPI device for each run, and each sample was run in triplicate. Data were analyzed using the Copley Inhaler Testing Data Analysis Software (Copley Scientific, Nottingham UK).

Statistical analysis

GraphPad Prism software was used for data analysis. Statistical significance between multiple groups was compared using one-way ANOVA with Turkey’s multiple comparison test. Survival analysis was calculated using the log-rank (Mantel-Cox) test. P < 0.05 was considered significant. Primary data are reported in data file S1.

SUPPLEMENARY MATERIALS

stm.sciencemag.org/cgi/content/full/11/522/eaat2848/DC1

Materials and Methods

Fig. S1. Effect of CSP7 on TGF-β signaling and HuR expression in mouse lung with 1× BLM-induced PF.

Table S1. In vitro Ames test.

Table S2. In vitro micronucleus assay.

Table S3. In vivo micronucleus screening in reticulocytes for clastogenic activity.

Data file S1. Primary data.

Reference (37)

REFERENCES AND NOTES

Acknowledgments: We thank B. Starcher for editing the manuscript and assisting in early animal experiments. Funding: This work was supported, in part, by NIH grants R01HL133067-01 and R21ES025815 and Flight Attendant Medical Research Institute Clinical Innovator Awards CIA150063 and CIA082380 to S.S. Author contributions: A.S.M., Y.P.B., V.G., M.R.N., S.K.S., L.F., N.T., A.B.W., B.M., Y.Z., E.M., V.R., and S.S. performed the experiments. S.S. conceived and designed the experiments and wrote the manuscript. A.B.W., N.M., S.B., and G.J.C. provided unique reagents and human tissue samples. B.M. and R.O.W. performed in vitro characterization and formulation studies, and S.I. reviewed data and edited the manuscript. E.M. performed DP formulation at the Drug Dynamics Institute at the University of Austin under R.O.W.’s supervision. Competing interests: S.S. and S.I. have equity interest in Lung Therapeutics Inc., a University of Texas startup biotechnology firm that is commercializing CSP7 for the treatment of IPF. S.I. is an equity holder, founder, and board member of Lung Therapeutics Inc. and serves as Chief Scientific Officer of the company, which is in the process of commercializing CSP7 for treatment of IPF. S.S. is an equity holder and consultant for the company. S.S. and S.I. have patents (patent no. 8697840 B2 entitled Peptide inhibition of lung epithelial apoptosis and pulmonary fibrosis and patent no. 9,630,990 entitled Inhibition of pulmonary fibrosis with chemical and peptide inhibitors) issued for the use of the Cav1 scaffolding domain peptide and its fragments for the treatment of lung injury and PF. R.O.W., S.I., and S.S. are inventors in a patent application no. WO 2016/138413 entitled “Polypeptide therapeutics and uses thereof”. A.B.W. is an inventor in a U.S. provisional patent application no. 62/729,010 entitled “Dry powder formulation of caveolin-1 peptides and methods of use thereof”. R.O.W.’s research is sponsored by Lung Therapeutics Inc. Data and materials availability: All data associated with this study are present in the paper or the Supplementary Materials.
View Abstract

Stay Connected to Science Translational Medicine

Navigate This Article