IPF fibroblasts display enhanced proliferation on polymerized collagen matrices. (A) Proliferation assay of control (n = 8) and IPF (n = 6) fibroblasts on monomeric and polymerized collagen. *, P < 0.01 and 0.005 for the ratio of proliferation of IPF fibroblasts on polymerized versus monomeric collagen compared with control fibroblasts at days 3 and 6, respectively. (B) Quantification of DNA synthesis by BrdU incorporation in control (n = 4) and IPF (n = 5) fibroblasts on monomeric and polymerized collagen at day 3. *, P < 0.0001 for the percentage of BrdU-positive IPF and control fibroblasts at the indicated time. (C) Ratio of proliferation of wound (n = 3) and IPF (n = 6) fibroblasts on polymerized versus monomeric collagen at days 3 and 6 (*, P < 0.02 and 0.001, respectively). (D) Normal fibroblasts were exposed to TGF-β for 3–6 d or left untreated, and cell numbers were quantified (*, P < 0.05). (E) Proliferation assay of control (n = 4) and IPF (n = 5) fibroblasts on monomeric and polymerized collagen in defined media. Shown is the percent increase in cell number in response to the indicated growth factor treatment compared with cultures continued in the absence of growth factors at day 3. *, P < 0.05 or 0.001 for the proliferative response of IPF compared with control fibroblasts at the time in defined media containing insulin alone or insulin plus platelet-derived growth factor, respectively. Error bars represent SEM. Data are representative of three independent experiments.

PI3K–Akt–S6K1 signaling in control fibroblasts. (A) Western blot analysis of phospho-Akt and S6K1 in the absence or presence of the PI3K inhibitor wortmannin (WT) in response to control fibroblast attachment to monomeric collagen in the absence of serum. One representative example is shown. (B) Western blot analysis of phospho-Akt in control fibroblasts preincubated with the indicated integrin blocking antibody and plated on monomeric collagen. (C) Fibroblasts were plated on tissue culture plates coated with monomeric collagen and blocked by BSA. Cells were allowed to spread for 30 min. Cell areas of a random 150 cells were quantified by tracing the cell border using ImageJ software (available at http://rsb.info.nih.gov/ij/). (D) Western blot analysis of phospho-Akt and S6K1 in response to control fibroblast attachment to polymerized collagen in the absence of serum. (E) Western blot analysis of phospho-Akt and S6K1 in control fibroblasts cultured on monomeric or polymerized collagen in growth factor–replete media. One representative example is shown. (F) GD25, GD25 β1, and GD25 α2β1 fibroblasts were cultured on monomeric (left) or polymerized (right) collagen in the presence of serum, and cell numbers were quantified. (G) Proliferation assay of control fibroblasts expressing dominant-negative Akt (Ad-DN-Akt; left) or constitutively active p110 subunit of PI3K (Ad-PI3Kp110; right) and cultured on monomeric or polymerized collagen in growth factor–replete media, respectively. (top left) Western blot analyses of phospho- and total Akt. (bottom left) Shown is the percent change in cell growth. *, P < 0.01 and 0.001 for the percent change in fibroblast growth on monomeric collagen that was significantly less in control fibroblasts expressing dominant-negative Akt compared with cells expressing control vector at days 3 and 6, respectively. (top right) Western blot analyses of phospho- and total Akt. (bottom right) Shown is the percent change in cell growth. *, P < 0.01 and 0.0001 for the percent change in fibroblast growth on polymerized collagen that was significantly greater in control cells expressing constitutively active p110 subunit of PI3K compared with control vector at days 3 and 6, respectively. Error bars represent SEM. Data are representative of three independent experiments.

The PI3K–Akt–S6K1 pathway is pathologically regulated in IPF fibroblasts. (A) Western blot analyses of phospho-Akt and S6K1 in response to IPF fibroblast attachment to monomeric collagen in the absence of serum. One representative example is shown. (B) Western blot analysis of phospho- and total Akt in response to treatment of control and IPF fibroblasts with TS2/16 β1 integrin activating antibody. (C) Control (n = 3) or IPF (n = 3) fibroblasts were stained with anti–β1 integrin antibodies and analyzed by FACS. Anti–β1 integrin antibodies used were against ligand-induced binding sites (9EG7), activating sites (TS2/16), or total β1 integrin regardless of activation state (P5D2). (D) Control and IPF fibroblasts were plated on monomeric collagen. Shown is a Western blot analysis of phosphorylated FAK (tyr397). Data are representative of three independent experiments from one control and one IPF cell line. (E) PI3K activity in control and IPF fibroblasts plated on tissue culture dishes and treated with TS2/16 β1 integrin activating antibody. Data are representative of three independent experiments from one IPF cell line. (F) Western blot analysis of phospho-Akt and S6K1 in response to IPF fibroblast attachment to polymerized collagen. One representative example is shown. (G) Western blot analysis of phospho-Akt in IPF fibroblasts preincubated with the indicated integrin blocking antibody and plated on polymerized collagen. Data are representative of three independent experiments from one representative IPF cell line. (H) Western blot analysis of phospho-Akt and S6K1 in IPF fibroblasts cultured on monomeric or polymerized collagen in growth factor–replete media. Data are representative of three independent experiments from five IFP cell lines. (I) Proliferation assay of IPF fibroblasts expressing dominant-negative Akt (left) or treated with 100 nM rapamycin (right) and cultured on polymerized collagen in growth factor–replete media. Shown is the percent change in cell growth. *, P < 0.01 and 0.0001 for the percent change in cell growth on polymerized collagen that was significantly less in IPF fibroblasts expressing dominant-negative Akt at days 3 and 6, respectively; or P < 0.002 or 0.001 for the percent change in cell growth on polymerized collagen that was significantly less in IPF fibroblasts treated with rapamycin at days 3 or 6, respectively. Error bars represent SEM. Data are representative of three independent experiments from one IPF cell line. (J) IPF fibroblasts were preincubated with the indicated integrin blocking antibody and plated on polymerized collagen in growth factor–replete media in the presence of antibody. BrdU staining as a measure of DNA synthesis was assessed at 24 h. Error bars represent SEM. Data are representative of three independent experiments from three IPF cell lines.

Regulation of PTEN is altered in IPF fibroblasts. (A and B) Western blot analysis of total PTEN in control (A) or IPF (B) fibroblasts cultured on monomeric or polymerized collagen in growth factor–replete media. One representative sample out of five control and five IPF cell lines examined is shown. (C and D) PTEN phosphatase assay showing the percent change in PTEN activity in control (n = 3; C, left) or IFP (n = 3; D, left) fibroblasts cultured on polymerized or monomeric collagen in growth factor–replete media compared with cells cultured under proliferation-prohibitive conditions. *, P < 0.001 for PTEN activity in control fibroblasts on polymerized collagen that was significantly higher at day 3 compared with monomeric collagen; P < 0.01 for PTEN activity in IPF fibroblasts on polymerized collagen that was significantly less on day 3 compared with monomeric collagen. Shown is PI3K activity in control (C, right) and IPF (D, right) fibroblasts cultured on monomeric or polymerized collagen in growth factor–replete media for 3 d. *, P < 0.005 for PI3K activity in control fibroblasts cultured on polymerized collagen that was significantly less at day 3 compared with monomeric collagen; or P < 0.05 for PI3K activity in IPF fibroblasts cultured on polymerized collagen that was significantly higher at day 3 compared with monomeric collagen. Error bars represent SEM. (E) PTEN protein levels in membrane and cytosolic fractions of cell lysates were measured by Western blot analysis. Data are representative of three independent experiments.

Knockdown of PTEN enables control fibroblasts to elude the antiproliferative effects of polymerized collagen. Control fibroblasts were transfected with PTEN or control siRNA and cultured on polymerized collagen in serum. (A) Western blot analysis of total PTEN levels. (B) Western blot analysis showing PTEN protein expression in response to increasing concentrations of PTEN siRNA. (C) Proliferation assay of control fibroblasts on polymerized collagen in serum-replete media. After 6 d, there was a significant increase in proliferation on polymerized collagen in cells treated with PTEN siRNA compared with control siRNA (*, P < 0.04). Error bars represent SEM. Data are representative of three independent experiments.

Ectopic expression of PTEN in IPF fibroblasts reverses the abnormal proliferative phenotype. (A) Western blot analysis of total PTEN and phospho-Akt and S6K1 in IPF fibroblasts expressing wild-type PTEN (Ad-wtPTEN) or empty vector (Ad-GFP) and cultured on polymerized collagen in growth factor–replete media. (B) eIF4F activity (RLUC luminescence) in IPF fibroblasts expressing wild-type PTEN or control vector and cultured for 3 d on polymerized collagen. *, P < 0.0001 for significantly less eIF4F activity in IPF fibroblasts expressing wild-type PTEN compared with control. (C) Quantification of DNA synthesis by BrdU incorporation in IPF fibroblasts expressing wild-type PTEN or empty vector and cultured on polymerized collagen in growth factor–replete media for 3 d. Shown is the percentage of BrdU-positive IPF fibroblasts. *, P < 0.0001 for the percentage of BrdU-positive IPF fibroblasts expressing wild-type PTEN that was significantly less compared with control. Error bars represent SEM. Data are representative of three independent experiments from one IPF cell line.

PTEN deficiency leads to a more durable fibroproliferative response after tissue injury. (A) Shown is the surface area of granulation tissue in PTEN wild-type (+/+; n = 3) and haploinsufficient (+/−; n = 3) mice. There was a significant delay in the resolution of granulation tissue in PTEN +/− compared with +/+ mice (P < 0.0001). Error bars represent SEM. (B) Hematoxylin and eosin–stained sections of granulation tissue at day 18 in PTEN +/− (top left) and +/+ (top right) mice. Note the increased cellularity of granulation tissue in PTEN +/− mice. Ki67 staining of day 18 granulation tissue from PTEN +/− (middle left) and +/+ (middle right). Note the abundant Ki67-positive cells (nuclei brown) in the granulation tissue of PTEN +/− mice. Insets show positive Ki67 staining in the newly formed epithelium overlying the wound ulcer. (bottom right) Abundant α–smooth muscle actin staining of the granulation tissue of PTEN +/− mice. (bottom left) Double staining for Ki67 (nuclei brown) and α–smooth muscle actin. Bar, 50 μm.

PTEN deficiency exaggerates fibrosis after lung injury. PTEN +/− or +/+ mice were administered bleomycin intratracheally. (A) 21 d later, lungs were harvested and total lung collagen was determined by Sircol assay (n = 6 mice per group). (B) Hematoxylin and eosin (H&E)–stained sections of day 21 lung tissue in PTEN +/− (top left) and +/+ (top right) mice. Trichrome staining of day 21 lung tissue from PTEN +/− (left, middle and bottom) and +/+ (right, middle and bottom) mice. Results are representative of three mice per group. Bars: (top and middle) 200 μm; (bottom) 50 μm.

Akt is activated in fibrotic foci in IPF lung tissue. Immunohistochemistry, evaluating phospho-Akt and total PTEN expression, was performed on lung tissue obtained from patients with histologically normal lungs (n = 5) and IPF (n = 5). (A and B) Normal lung tissue. Note the presence of phospho-Akt in epithelial and endothelial cells (A). No phospho-Akt expression is apparent in mesenchymal cells in the vessel wall or in normal lung parenchyma in A. Arrows denote PTEN expression in spindle-shaped cells within normal lung parenchyma (B). (C and D) IPF lung tissue. Note the phospho-Akt expression in spindle-shaped cells within the fibrotic foci of IPF (C). Cells within fibrotic foci also displayed PTEN expression (D). (E) As a control, shown is fibrotic lung tissue stained with irrelevant primary antibody (anti-p53). Bars, 50 μm.