Proteasomal inhibition after injury prevents fibrosis by modulating TGF-β(1) signalling

Abstract

Background: The development of organ fibrosis after injury requires activation of transforming growth factor β(1) which regulates the transcription of profibrotic genes. The systemic administration of a proteasomal inhibitor has been reported to prevent the development of fibrosis in the liver, kidney and bone marrow. It is hypothesised that proteasomal inhibition would prevent lung and skin fibrosis after injury by inhibiting TGF-β(1)-mediated transcription.

Methods: Bortezomib, a small molecule proteasome inhibitor in widespread clinical use, was administered to mice beginning 7 days after the intratracheal or intradermal administration of bleomycin and lung and skin fibrosis was measured after 21 or 40 days, respectively. To examine the mechanism of this protection, bortezomib was administered to primary normal lung fibroblasts and primary lung and skin fibroblasts obtained from patients with idiopathic pulmonary fibrosis and scleroderma, respectively.

Results: Bortezomib promoted normal repair and prevented lung and skin fibrosis when administered beginning 7 days after the initiation of bleomycin. In primary human lung fibroblasts from normal individuals and patients with idiopathic pulmonary fibrosis and in skin fibroblasts from a patient with scleroderma, bortezomib inhibited TGF-β(1)-mediated target gene expression by inhibiting transcription induced by activated Smads. An increase in the abundance and activity of the nuclear hormone receptor PPARγ, a repressor of Smad-mediated transcription, contributed to this response.

Conclusions: Proteasomal inhibition prevents lung and skin fibrosis after injury in part by increasing the abundance and activity of PPARγ. Proteasomal inhibition may offer a novel therapeutic alternative in patients with dysregulated tissue repair and fibrosis.

Figure 1

Bortezomib abrogates pulmonary and skin fibrosis induced by bleomycin in mice. (A) Mice (C57BL/6, male) were treated with intratracheal bleomycin or saline (0.075 IU). Seven and 14 days later the mice were treated with bortezomib (120 μg/kg intraperitoneally) or saline. The mice were killed on day 21 and the lungs were examined by trichrome staining. (B) A separate group of mice (BALB/c) were treated with daily subcutaneous injections of bleomycin (20 μg) in the same region of the skin on their back and bortezomib (400 μg/kg intraperitoneally twice weekly) or saline was begun 7 days later. After 4 weeks, H&E-stained skin sections of the treated skin were obtained. The thickest regions of skin for each treatment are shown. (C) Total lung collagen was quantified by picrosirius red collagen precipitation and (D) skin thickness was measured microscopically (n≥8 (lung) and n=5 (skin) for each treatment arm). p Values for comparisons between groups are indicated in italics above the bars.

Figure 2

Bortezomib does not inhibit transforming growth factor β₁ (TGF-β₁)-induced Smad3 phosphorylation or nuclear translocation but inhibits Smad-mediated transcription. (A) Primary cultures of normal human lung fibroblasts were transfected with a plasmid containing SBE-luciferase and 24 h later treated with TGF-β₁ (5 ng/ml) with or without bortezomib. SBE-luciferase activity was measured 24 h later. Treatment with the ALK-5 inhibitor SB431542 (10 μM) is shown as a control. p<0.0001 for difference in dose determined by one-way ANOVA; Bonferroni-corrected p<0.05 for comparisons between TGF-β₁ alone and all other conditions. (B, C) Primary cultures of normal human lung fibroblasts were grown to 70% confluence and incubated with TGF-β₁ (5 ng/ml) with or without bortezomib (0.2 βM). The level of phosphorylated Smad3 was measured in (B) total cell lysates and (C) nuclear extracts using total Smad3 and RNA polymerase II (RNA Pol II) as loading controls, respectively. Phosphorylated-Smad3 expression was quantified using densitometry. p Values for comparisons are indicated in italics above the bars (N ≥3 for all measures).

Figure 3

Bortezomib inhibits transforming growth factor β₁ (TGF-β₁)-induced gene expression in primary normal human lung fibroblasts. (A) Early passage primary cultures of normal human lung fibroblasts were grown to 70% confluence, serum-starved for 24 h and incubated with TGF-β₁ (5 ng/ml) with or without bortezomib (200 nM). After 24 h, α-smooth muscle actin (α-SMA) mRNA and protein were measured using RT-qPCR and immunohistochemistry, respectively. (B) Identically treated cells were harvested for measurement of connective tissue growth factor (CTGF) mRNA and protein expression using RT-qPCR and immunoblotting, respectively. (C) Normal human lung fibroblasts were treated with bortezomib (200 nM) for 24 h, cultured in media without bortezomib for 5 days, treated with TGF-β₁ (5 ng/ml) in the absence of serum and 24 h later RNA was harvested for measurement of mRNA encoding the TGF-β₁ target genes CTGF and plasminogen activator-1 (PAI-1). The number of days after the single dose of bortezomib is indicated (N ≥3 for all conditions). p Values are indicated in italics above the bars.

Figure 4

Bortezomib increases PPARg abundance and activity in normal human lung fibroblasts. (A) Primary normal human lung fibroblasts were cultured in the presence of cycloheximide (10 μg/ml) with and without bortezomib (200 nM) and the protein abundance of PPARγ was assessed by immunoblotting at the indicated times. (B) The same experiment to that described in (A) was performed in cells treated with transforming growth factor β₁ (TGF-β₁, 5 ng/ml). (C) Primary normal human lung fibroblasts were infected with a plasmid containing PPARg response element (PPRE)-luciferase. Twenty-four hours after transfection the cells were treated with TGF-β₁ (5 ng/ml) with or without bortezomib and PPRE-luciferase activity was measured 24 h later. (D) Primary normal human lung fibroblasts were treated with TGF-β₁ (5 ng/ml) with or without bortezomib and the mRNA level of the PPARg target gene FABP4 was measured 24 h later by RT-qPCR. (E) Bortezomib (120 μg/kg intraperitoneally) was administered to mice and 24 h later lung homogenates were immunoblotted using an antibody against PPARγ. p Values are indicated in italics above the bars (N ≥3 for all measures).

Figure 5

Bortezomib modulates fibrotic gene expression via PPARγ. (A) Primary normal human lung fibroblasts were incubated with transforming growth factor β₁ (TGF-β_1, 5 ng/ml) with or without low-dose bortezomib (10 nM) or rosiglitazone (10 μM) and 24 h later connective tissue growth factor (CTGF) mRNA expression was measured using RT-qPCR. (B) Primary normal human lung fibroblasts were stably transfected with a shRNA against PPARγ or a control lentivurus and PPARγ levels were measured by immunoblotting (densitometry shown to the right of the blot). These cells were treated with TGF-β₁ (5 ng/ml) with or without bortezomib (200 nM) and CTGF mRNA was measured 24 h later (RT-qPCR). p Values are indicated in italics above the bars. In addition, p=0.026 for comparison between control transfected cells treated with TGF-β₁ or TGF-β₁ and bortezomib and p=0.12 for comparison between PPARγ and control shRNA cells treated with TGF-β₁ alone.

Figure 6

Bortezomib inhibits transforming growth factor β₁ (TGF-β₁) autocrine expression. (A) Primary normal human lung fibroblasts were incubated with TGF-β₁ (5 ng/ml) with or without bortezomib (200 nM) and 24 h later TGF-β₁ mRNA expression was measured using RT-qPCR. (B) Intratracheal bleomycin (0.075 IU) or saline was instilled into mice and intraperitoneal bortezomib (120 μg/kg) or saline was administered 7 days later. Bronchoalveolar lavage was performed 14 days after bleomycin treatment for measurement of the level of active TGF-β₁ by ELISA. p Values are indicated in italics above the bars.

Figure 7

Bortezomib inhibits profibrotic gene expression in biological samples obtained from patients with lung fibrosis or scleroderma. (A) Primary cultures of normal human lung fibroblasts transfected 24 h earlier with a plasmid containing SBE-luciferase were incubated with bronchoalveolar lavage fluid obtained from 10 patients with lung fibrosis or eight patients without evidence of lung fibrosis with or without bortezomib (200 nM). SBE-luciferase activity was measured 24 h later. (B) Primary cultures of skin fibroblasts obtained from a region of normal or fibrotic skin from a patient with scleroderma and (C) primary lung fibroblasts cultured from three patients (Pts 1, 2 and 3) with lung fibrosis were treated with transforming growth factor b1 (TGF-β₁, 5 ng/ml) with or without bortezomib or bortezomib alone and the levels of connective tissue growth factor (CTGF) and plasminogen activator-1 (PAI-1) mRNA were measured using RT-qPCR. p Values are indicated in italics above the bars.