Loss of PTEN Expression by Dermal Fibroblasts Causes Skin Fibrosis

ORIGINAL ARTICLE

Loss of PTEN Expression by Dermal Fibroblasts Causes Skin Fibrosis Sunil K. Parapuram1,5, Xu Shi-wen2,5, Christopher Elliott1, Ian D. Welch3, Helen Jones2, Murray Baron4, Christopher P. Denton2, David J. Abraham2,6 and Andrew Leask1,6 Fibrosis represents a common pathway leading to organ failure and death in many diseases and has no effective therapy. Dysregulated repair and excessive tissue scarring provides a unifying mechanism for pathological fibrosis. The protein phosphatase and tensin homolog (PTEN) acts to dephosphorylate proteins, which promotes tissue repair and thus could be a key fibrogenic mediator. To test this hypothesis, we first showed that PTEN expression was reduced in skin fibroblasts from patients with the fibrotic autoimmune disease diffuse systemic sclerosis (dSSc). To evaluate whether this deficiency could be sufficient for fibrogenesis in vivo, we deleted PTEN in adult mouse fibroblasts. Compared with littermate control mice, loss of PTEN resulted in a 3-fold increase in dermal thickness due to excess deposition of collagen. PTEN-deleted fibroblasts showed elevated Akt phosphorylation and increased expression of connective tissue growth factor (CTGF/CCN2). Selective inhibition of the phosphatidylinositol 3-kinase/Akt pathway reduced the overexpression of collagen and CCN2 by PTEN-deficient fibroblasts. Overexpression of PTEN reduced the overexpression of type I collagen and CCN2 by dSSc fibroblasts. Thus, PTEN appears to be a potential in vivo master regulator of fibrogenesis; PTEN agonists may represent anti-fibrotic treatments. Journal of Investigative Dermatology (2011) 131, 1996–2003; doi:10.1038/jid.2011.156; published online 9 June 2011

INTRODUCTION Fibrosis is associated with several disorders, including systemic sclerosis (SSc), arthritis, nephropathy, liver cirrhosis, and cancer, and is characterized by excess deposition of collagen by activated connective tissue fibroblast cells and their contraction, resulting in destruction of normal tissue architecture/organ failure. There is no effective therapy for fibrotic diseases, in part, because the underlying mechanisms are poorly understood (Abraham et al., 2009). Different cytokines (e.g., transforming growth factor-b (TGFb), platelet1

Department of Dentistry, Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada; 2Centre for Rheumatology, University College London (Royal Free Campus), London, UK; 3 Department of Animal Care and Veterinary Services, Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada and 4Division of Rheumatology, McGill University and SMBD— Jewish General Hospital, Montreal, Quebec, Canada 5

These authors contributed equally to this work.

6

Co-senior authors

Correspondence: Andrew Leask, Division of Oral Biology, Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada N6A 5C1. E-mail: [email protected] Abbreviations: ALK, activin linked kinase; CTGF, connective tissue growth factor/CCN2; dcSSc, diffuse cutaneous systemic sclerosis; GAPDH, glyceraldehydes-3-phosphate dehydrogenase; NF, normal fibroblasts; PTEN, phosphatase and tensin homolog/PI3K phosphatidylinositol triphosphate; a-SMA, a-smooth muscle actin; SSc, systemic; TGFb, transforming growth factor b Received 5 November 2010; revised 2 March 2011; accepted 12 April 2011; published online 9 June 2011

1996 Journal of Investigative Dermatology (2011), Volume 131

derived growth factor, and endothelin-1) act in concert to promote fibrosis, and thus compounds targeting individual cytokines may not be effective (Leask, 2010). Moreover, different cytokines may be responsible for the fibrosis seen in individual patients, making a ‘‘one-size-fits-all’’ approach potentially problematic. Studies using fibroblasts from patients with the fibrotic autoimmune disease SSc have demonstrated that inhibition of individual cytokines results in partial alleviation of persistent fibrotic phenotype, but individual cytokines appear to be responsible for complementary, yet overlapping, aspects of the phenotype of these cells (Chen et al., 2006; Ishida et al., 2006; Shi-Wen et al., 2006; Leask, 2010, 2011). Indeed, clinical trials aimed at exploiting these strategies have been disappointing (Leask, 2011). An alternative anti-fibrotic strategy might be to attack proteins that are dysregulated within the actual target cells (i.e., the fibroblasts) responsible for the excessive collagen deposition characterizing scar tissue (Leask, 2011). A plausible model for fibrosis is that there is excessive and/ or persistent activation of a tissue repair program. One possible cause of this could be that proteins, which normally suppress this process, are defective in fibrosis. One such protein may be the phosphatase and tensin homolog (PTEN; Tsugawa et al., 2002). The main substrate of PTEN, a dual protein/lipid phosphatase, is phosphatidyl-inositol,3,4,5 triphosphate, which activates Akt. PI3K–Akt signaling has been implicated in collagen production and proliferation in lung fibroblasts (Lu et al., 2010). Akt has been shown to be elevated in activity in SSc skin fibroblasts and mediates the & 2011 The Society for Investigative Dermatology

SK Parapuram et al. Loss of PTEN Results in Fibrosis

ability of endothelin-1 to elevate myofibroblast activity in SSc fibroblasts (Shi-wen et al., 2004; Jun et al., 2005) and elevated Akt activity contributes to fibrogenic gene expression in SSc fibroblasts (Shi-wen et al., 2009). However, whether dysregulation of the PTEN/Akt pathway is a hallmark of SSc and is responsible for the persistent fibrotic phenotype of SSc fibroblasts is unclear. Moreover, whether loss of PTEN expression is sufficient for a fibrotic phenotype in vivo is not known. In this report, we investigate the contribution of PTEN to the sclerosis observed in diffuse cutaneous SSc (dcSSc) by examining the expression of PTEN in normal and dcSSc skin fibroblasts and the effect of loss of PTEN expression in vivo using a conditional knockout strategy (Zheng et al., 2002; Cai et al., 2008; Liu et al., 2008, 2009). Our results show valuable insights into the molecular mechanism underlying fibrosis and suggest that reduced PTEN expression by dSSc fibroblasts may be responsible for dermal fibrosis seen in SSc patients. RESULTS PTEN protein expression is reduced in dcSSc dermal fibroblasts

We subjected dcSSc dermal fibroblasts isolated from explant culture to western blot analyses using an anti-PTEN antibody. Dermal fibroblasts from six healthy individuals and six individuals with dcSSc (lesional) were used. PTEN protein expression was significantly reduced in dcSSc fibroblasts compared to normal, healthy fibroblasts (SScF vs. normal fibroblasts in five out of six samples; Figure 1). Similarly, phosphorylation of Akt was elevated in SScF (Figure 1). Loss of PTEN expression by fibroblasts is sufficient for fibrogenesis in vivo

PTEN GAPDH p-Akt Akt

Densitometry image units

NF1 NF2 NF3 NF4 NF5 NF6 SScF1 SScF2 SScF3 SScF4 SScF5 SScF6

As we observed reduced PTEN expression in dcSSc fibroblasts, we investigated whether loss of PTEN expression by fibroblasts resulted in fibrosis in vivo. We used a strategy well established in our laboratories to specifically delete target

10 8 6 4 2 0 8 6 4 2 0

PTEN

* * p-Akt

NF

SScF

Figure 1. Scleroderma fibroblasts show reduced PTEN expression and elevated Akt phosphorylation. Dermal fibroblasts from six normal individuals (NF) and six individuals with SSc (SScF) were cultured. Equal amounts of protein extracts were subjected to SDS–PAGE and western blot analysis with anti-protein phosphatase and tensin homolog (PTEN), anti-phospho-Akt, anti-Akt, and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibodies. Statistical analysis using a Student’s t-test shows that there is a statistical significance (Po0.05) between PTEN expression and phospho-Akt levels in SSc and control fibroblasts.

genes in fibroblasts of mice (Zheng et al., 2002; Cai et al., 2008; Liu et al., 2008, 2009). Briefly, mice with Pten gene flanked by loxP sites (Groszer et al., 2001) were crossed with mice expressing tamoxifen-dependent Cre recombinase under the control of fibroblast-specific collagen 1a2 promoter/enhancer (Zheng et al., 2002). Mice homozygous for loxP–Pten allele and hemizygous for tamoxifen-dependent collagen 1a2 promoter/enhancer-controlled Cre recombinase were generated. Administration of tamoxifen to these mice allowed conditional deletion of the Pten gene (K/K) in fibroblasts. Genetically identical littermates or littermates that did not have the Cre allele were administered corn oil (vehicle) or tamoxifen, respectively, and served as control mice (C/C). PCR analysis of DNA obtained from biopsies of the ear showed that Pten gene was deleted only in animals in which tamoxifen-activated Cre enzyme was present (Figure 2a, lanes 5 and 6). Deletion of the PTEN gene was further verified by immunohistochemical analysis of skin (Figure 2b) and by subjecting dermal fibroblasts isolated from these mice to indirect immunofluorescence analysis with an anti-PTEN antibody (Figure 2c). By 42 days post-cessation of tamoxifen injection, PTENdeficient mice, compared with their wild-type counterparts, showed a B1.5-fold increase in dermal thickness (Figure 3a), as revealed by hematoxylin and eosin staining, paralleled by elevated collagen production, as visualized by Trichrome stain (Figure 3b, Trichrome). Increase in collagen was further validated by assessing hydroxyproline levels (Figure 3b, hydroxyproline). Moreover, within the dermis, PTEN-deficient animals possessed elevated expression of the fibrogenic marker CCN2 (Leask et al., 2009) (Figure 3c). Also, fibrotic lesions are characterized by the presence of a-smooth muscle actin (a-SMA)-expressing myofibroblasts (Chen et al., 2005; Abraham et al., 2009); loss of PTEN resulted in an increase in a-SMA-positive cells (Figure 3d). Finally, fibrosis is a fibroproliferative condition (Abraham et al., 2009); loss of PTEN resulted in an increase in proliferating-cell nuclear antigen-positive cells (Figure 3e). Collectively, these data suggest that loss of PTEN expression by fibroblasts is sufficient to result in fibrogenesis. PTEN-deficient fibroblasts show overexpression of type I collagen and CCN2 in an Akt/PI3-kinase-dependent fashion

To understand the mechanism by which loss of PTEN induces a fibrotic phenotype, we first assessed the phosphorylation status of Akt, a known target of PTEN (Carnero, 2010), and found a significant increase in the number of dermal fibroblasts expressing p-Akt in PTEN K/K mice compared with controls (Figure 4a). These data were further verified by western blot analysis of protein extracted from dermal fibroblasts isolated from PTEN K/K and C/C mice. PTEN-deficient fibroblasts showed elevated Akt phosphorylation and increased expression of type I collagen and CCN2 (Figure 4b). To assess whether the Akt/PI3 kinase pathway was responsible for this overexpression of type I collagen and CCN2, control and PTEN-deficient cells were treated overnight with wortmannin (100 nM) or LY294002 (10 mM). www.jidonline.org 1997

SK Parapuram et al. Loss of PTEN Results in Fibrosis

Both wortmannin and LY294002 reduced the overexpression of type I collagen and CCN2 by PTEN-deficient cells, indicating the involvement of the PI3K–Akt pathway

1

2

3

4

5

6 loxP–Pten wt Pten Pten deletion

Cre

C/C

K/K Epidermis

Dermis

C/C

K/K

Fibroblasts

Figure 2. Generation of mice bearing a fibroblast-specific deletion of PTEN. We generated mice homozygous for loxP–Pten allele and hemizygous for tamoxifen-dependent Cre expressed under the control of a fibroblast-specific collagen 1a2 promoter/enhancer. Administration of tamoxifen deleted the Pten gene only in mice in which tamoxifen-activated Cre enzyme was present. (a) Deletion of Pten gene was tested by PCR genotyping of DNA extracted from ear biopsies (lanes 1 and 2, corn-oil administration; lanes 3–6, tamoxifen administration. (b) Loss of Pten was further verified by indirect immunofluorescence analysis of skin tissue with an anti-protein phosphatase and tensin homolog (PTEN) antibody. Whereas dermal fibroblasts of control mice (C/C) showed immunostaining for PTEN protein, there was virtually no stain for PTEN in the fibroblasts of mice deleted for PTEN (K/K). Bar ¼ 100 mM. (c) Dermal fibroblasts isolated from PTEN C/C and K/K mice were also subjected to indirect immunofluorescence with anti-PTEN antibody (red). Cells were counterstained with 40 ,6-diamidino-2-phenylindole (blue) to detect nuclei.

in inducing the fibrotic phenotype in PTEN K/K mice (Figure 4c). Treatment of cells with the ALK5 inhibitor SB431542 (10 mM) did not reverse the overexpression of CCN2 or type I collagen in PTEN-deficient cells (Figure 4d). (It is interesting to note, however, that the ALK5 inhibitor reduced basal expression of type I collagen in wild-type cells, consistent with previous observations that ALK5 inhibition reduced basal expression of type I collagen in normal, human dermal fibroblasts (Chen et al., 2005)). PTEN overexpression results in the rescue of the type I collagen and CCN2 overexpression phenotype of dSSc fibroblasts

On the basis of these data, we assessed whether adenoviralbased overexpression of PTEN could reverse the persistent fibrotic phenotype of dSSc dermal fibroblasts, which overexpress type I collagen and CCN2 compared with healthy control fibroblasts (Shi-wen et al., 1997; Shi-wen et al., 2000). We found that, compared to cells transfected with control virus, cells transfected with adenovirus containing PTEN expression vector resulted in the overexpression of PTEN and the reduced expression type I collagen and CCN2 protein expression (Figure 5). Collectively, these results indicate that reduced PTEN expression is sufficient for fibrogenic responses in skin fibroblasts. DISCUSSION There is no treatment for fibrotic disease in general and in SSc in particular largely because the underlying deficiency causing fibrosis is unclear. Our results are consistent with previous findings that PTEN expression is downregulated in liver tissues in a rat model of hepatic fibrosis (Hao et al., 2009) and a recent report that showed first that Akt levels are heightened in a PTEN-dependent fashion in fibroblasts isolated from idiopathic pulmonary fibrosis patients and second that PTEN-deficient mice are more susceptible to bleomycin-induced lung fibrosis (Xia et al., 2008). However, whether loss of PTEN is sufficient for fibrogenesis and whether PTEN overexpression rescues the fibrotic phenotype of cells derived from SSc skin lesions has not, at least as far as we are aware, been shown until now. TGFb suppresses PTEN expression (Kato et al., 2009; Trojanowska, 2009; Yang et al., 2009) in normal cells. It is interesting that ALK5 inhibition did not affect the phenotype of the PTEN-deficient cells. This result is consistent with the previous observations that ALK5 inhibitors do not affect CCN2 overexpression in SSc fibroblasts (Chen et al., 2006). The data contained in our report therefore suggest that PTEN expression in SSc fibroblasts is reduced in a fashion

Figure 3. Loss of PTEN expression in dermal fibroblasts results in increased skin thickness, collagen deposition, and CCN2 expression. Sections of paraffin embedded skin tissue from protein phosphatase and tensin homolog (PTEN) C/C and K/K mice 42 days post-cessation of tamoxifen injection were examined. (a) Hematoxylin and eosin (H and E) staining. Note increase in dermal thickness in PTEN-deficient mouse skin. (**Po0.001, N ¼ 4). (b) Trichrome staining. A representative of stained section is shown. Hydroxyproline analysis of skin further confirmed the increase of collagen in PTEN-deficient mice (*Po0.05, N ¼ 6). Indirect immunofluorescence analysis by (c) anti-CCN2 antibody was also performed. Note abundant CCN2-expressing cells in PTEN K/K mice as opposed to control mice (C/C; **Po0.001, N ¼ 4). (d) Anti-a-smooth muscle actin (SMA) antibody was also performed. Note abundant a-SMA-expressing cells in PTEN K/K mice as opposed to control mice (C/C; **Po0.001, N ¼ 4). (e) Anti-proliferating-cell nuclear antigen (PCNA) antibody was also performed. Note abundant PCNA-expressing cells in PTEN K/K mice as opposed to control mice (C/C; **Po0.001, N ¼ 4). Statistics were performed using the Student’s t-test. Cells were counterstained with 40 ,6-diamidino-2-phenylindole (blue) to detect nuclei. Bar ¼ 100 mM.

1998 Journal of Investigative Dermatology (2011), Volume 131

SK Parapuram et al. Loss of PTEN Results in Fibrosis

a

H and E C/C

K/K Dermal thickness

**

Distance (μM)

200 150 100 50 0

C/C

K/K

Collagen

b

C/C

K/K

Hydroxyproline (g per 100 g tissue)

Hydroxyproline content 7 6 5 4 3 2 1 0

*

C/C

K/K

CCN2

c

C/C

K/K

% Of CCN2-positive fibroblasts

CCN2 immunostaining 60

**

50 40 30 20 10 0 C/C

K/K

α-SMA C/C

K/K

% Of α-SMA-positive fibroblasts

d

70 60 50 40 30 20 10 0

α-SMA immunostaining

**

C/C

K/K

PCNA C/C

K/K PCNA immunostaining % Of PCNA-positive fibroblasts

e

90 80 70 60 50 40 30 20 10 0

**

C/C

K/K

www.jidonline.org 1999

SK Parapuram et al. Loss of PTEN Results in Fibrosis

p-Akt C/C

K/K

p-Akt immunostaining % Of p-Akt-positive fibroblasts

50

**

40 30 20 10 0

α-SMA

PTEN

GAPDH C / /C C C +Ly /C +W o K K/ /K K K/ +Ly K+ W o

K

CCN2

K/

C

/C

Col 1

C

p-Akt

Densitometry image units

C/C

*

*

*

*

*

*

α-SMA

Col 1

CCN2

C

C /C /C C +Ly /C +W o K/ K K/ K+ K/ Ly K+ W o

Akt

10 8 6 4 2 12 10 8 6 4 2 10 8 6 4 2 0

K/K

p-ERK

K5

K/

5

/C

+A

C C

K

CCN2

/C

K5 AL

K+

5

K K/

LK

C

+A /C C

K/

/C

GAPDH

AL

CCN2

LK

p38

Col 1

*

K+

Col 1 p-p38

12 10 8 6 4 2 2.0 1.6 1.2 0.8 0.4 0

K/

Densitometry image units

ERK

Figure 4. Overexpression of type I collagen and CCN2 by PTEN-deficient fibroblasts depends on PI3K/Akt pathway. The mechanism by which loss of protein phosphatase and tensin homolog (PTEN) induces a fibrotic phenotype was determined. Indirect immunofluorescence analysis indicates abundant (a) p-Akt-positive in PTEN-deficient (K/K) mice compared with control mice (C/C; **Po0.001). Mice were examined 42 days post-cessation of tamoxifen injection. Statistics were performed using the Student’s t-test. Bar ¼ 100 mM. (b) Western blot analysis of protein extracts from dermal fibroblasts isolated from PTEN K/K mice also showed increased expression of p-Akt compared with controls. Note that p-p38 and p-ERK were not affected by loss of PTEN. (c) Cells were treated for 24 hours with DMSO (d), wortmannin (100 nM, Wo; a concentration shown to be specific/selective for Akt/PI3 kinase in cultured cells; Arcaro and Wymann, 1993; Young et al., 1995; Shi-wen et al., 2004), and LY294002 (10 mM, LY; a concentration shown to be specific/selective for Akt/PI3 kinase in cultured cells; Sellers et al., 2000; Shi-wen et al., 2004) prior to protein extraction and western blot analysis. The increased expression of collagen I, a-smooth muscle actin (SMA), and CCN2 in PTEN K/K dermal fibroblasts was reduced in the presence of wortmannin or LY294002 (*, decreased expression in the presence of inhibitor compared to DMSO control Po0.05; analysis of variance). (d) Cells were treated as in c, except the ALK5 inhibitor SB431542 (10 mM; a concentration shown to be specific/selective for ALK 4/5/7 in cultured cells, Inman et al., 2002) was used (*, decreased expression in the presence of inhibitor compared with DMSO control, Po0.05; analysis of variance). ERK, extracellular signal-regulated kinase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PCNA, proliferating-cell nuclear antigen; PI3K, phosphatidylinositol 3-kinase. Please note that human dermal fibroblasts treated with a small interfering RNA directed toward PTEN resulted in increases in Col1a1 and CCN2 mRNA expression (see Supplementary Figure S1 online).

independent of canonical TGFb signaling. The data are therefore consistent with the notion that SSc fibroblasts possess a gene expression signature reminiscent of TGFb signaling (Sargent et al., 2010), but are activated in a fashion that is autonomous of canonical TGFb signaling but dependent on non-canonical TGFb signaling pathways (Trojanowska, 2009; Leask, 2011). One possible mechanism underlying the reduced PTEN expression in SSc fibroblasts could be due to reduced expression of peroxisome prolif2000 Journal of Investigative Dermatology (2011), Volume 131

erator-activated receptor-g; peroxisome proliferator-activated receptor-g is reduced in SSc fibroblasts; and peroxisome proliferator-activated receptor-g can regulate PTEN expression (Teresi and Waite, 2008; Wei et al., 2010). Our data collectively suggest that PTEN normally suppresses fibrogenesis in vivo and that loss of PTEN expression is sufficient for fibrogenesis in vivo. Our data also indicate that the reduction in PTEN expression in SSc fibroblasts contributes to the persistent fibrotic phenotype

SK Parapuram et al.

Densitometry image units

Loss of PTEN Results in Fibrosis

SScF PTEN GAPDH –

–

–

+

+

4 2 0 Control + PTEN

SScF

Col 1

CCN2

GAPDH Adenovirus + Pten

*

6

Densitometry image units

Normal F

+

8

– – – + + +

– – – + + +

40 Col 1 30 20 10 0 CCN2 20

*

NF SScF

*

10 0

–

+

–

+

Figure 5. Overexpression of PTEN results in reduced expression of type I collagen and CCN2 protein by scleroderma fibroblasts. Dermal fibroblasts from three normal individuals (NF) and three individuals with SSc (SScF) were transfected with control adenovirus or adenovirus encoding protein phosphatase and tensin homolog (PTEN). Cell extracts were then subjected to western blot analyses with anti-CCN2, anti-PTEN, anti-type I collagen, and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibodies. Statistical significance (*Po0.05) was determined by analysis of variance.

of these cells; PTEN agonists may be a therapeutic approach to fibrotic diseases such as SSc. Moreover, PTEN knockout mice could represent a new model for the fibrosis observed in SSc. MATERIALS AND METHODS Patients and cell culture Lesional fibroblasts were grown in Dulbecco’s modified Eagle’s medium 10% fetal bovine serum (Invitrogen, Burlington, ON, Canada) by explant culture from forearm skin of SSc patients with diffuse cutaneous involvement (Leroy et al., 1988). The group of SSc patients fulfilled the criteria of the American College of Rheumatology for the diagnosis of SSc. The patients were female, and age-, sex-, and site-matched fibroblasts from healthy individuals were also used. Cells were used between passages 2 and 5 (Shi-wen et al., 1997). All procedures were approved by the appropriate committee at the University College London under written, informed patient consent and adherence to the declaration of Helsinki Guidelines. Dermal fibroblasts were isolated from PTEN-deficient and wild-type control mice as previously described (Liu et al., 2008).

RNA analysis Cells were lysed in Trizol (Invitrogen). RNA was prepared, quantified spectrophotometrically, and subjected to real-time PCR analysis using Assays-on-Demand primers and One-Step Master Mix as previously described (Applied Biosystems, Foster City, CA; Pala et al., 2008). Relative expression was assessed, compared with glyceraldehyde-3-phosphate dehydrogenase mRNA, using the DD Ct method. Samples were run in triplicate; expression values represent averages (±SD) from different patients/mice.

Cell culture immunofluorescence and western analysis Cells were lysed in 2% SDS, proteins quantified (Pierce, Nepean, ON, Canada) and subjected to western blot analysis as previously described with anti-PTEN (1:500, Cell Signaling, Danvers, MA), anti-glyceraldehyde-3-phosphate dehydrogenase (1:5,000, Sigma,

St Louis, MO), anti-CCN2 (1:500, Abcam, Cambridge, MA), antip-Akt, anti-Akt, anti-p-p38, anti-p-38, anti-p-ERK, anti-ERK (all 1:500, Cell Signaling), anti-a-SMA (1:1,000, Sigma), and anti-type I collagen (1:1,000, Biodesign, Abingdon, UK) antibodies (Chen et al., 2005). Cells were incubated for 24 hours in the presence or absence of DMSO, wortmannin (100 nM; Arcaro and Wymann, 1993; Young et al., 1995; Shi-wen et al., 2004) or LY294002 (10 mM; Sellers et al., 2000; Shi-wen et al., 2004) or the ALK5 inhibitor SB431542 (10 mM; Inman et al., 2002; Sigma). Cells were stained with anti-PTEN antibody and with a Texas Red-conjugated secondary antibody (Jackson Laboratories, Bar Harbor, ME) prior to subjecting cells to immunofluorescence microscopy (Zeiss, North York, ON, Canada).

Animal studies Mice with exon 5 of Pten gene flanked by loxP sites were obtained from The Jackson Laboratory and mated with mice expressing tamoxifen-dependent Cre recombinase under the control of fibroblast-specific regulatory sequence from the pro-a2 (I) collagen gene (Zheng et al., 2002). Mice homozygous for loxP–Pten and hemizygous for Cre were administered tamoxifen (1 mg per mouse for 5 days) at 21–24 days of age to delete Pten gene, specifically in fibroblasts. Deletion of PTEN was tested by PCR genotyping. The deletion of floxed Pten allele was determined by forward primer 50 -ACTCAAGGCAGGGATGAGC-30 and two reverse primers 50 -AATCTAGGGCCTCTTGTGCC-30 and 50 -gcttgatatc gaattcctgcagc-30 (Lesche et al., 2002). Presence of Cre sequence was determined using primers 50 -ATCCGAAAAGAAAACGTT GA-30 and 50 -ATCCAGGTTACGGATATAGT-30 (Zheng et al., 2002). To delete PTEN, 3-week-old mice were given intraperitoneal injections of tamoxifen suspension (0.1 ml of 10 mg ml1 4-hydroxytamoxifen, Sigma) over 5 days. Deletion of PTEN was tested by PCR genotyping. Experiments were performed on littermate mice homozygous for the loxP–PTEN allele (Jackson Laboratories) and heterozygous for type I collagen Cre (Zheng et al., 2002) that were treated with either tamoxifen (‘‘knockout PTEN’’, K/K) or www.jidonline.org 2001

SK Parapuram et al. Loss of PTEN Results in Fibrosis

corn oil (‘‘control PTEN’’, C/C) alone. Mice were killed by CO2 euthanasia at time points indicated. Skin samples were collected for histology, immunohistochemistry, hydroxyproline assay, and for fibroblast cell culture. All animal protocols were given ethical approval by the University of Western Ontario’s Animal Care Committee.

SUPPLEMENTARY MATERIAL

Immunohistochemistry and assessment

Arcaro A, Wymann MP (1993) Wortmannin is a potent phosphatidylinositol 3-kinase inhibitor: the role of phosphatidylinositol 3,4,5-trisphosphate in neutrophil responses. Biochem J 296(Part 2):297–301

Skin tissue sections (0.5 mm) were cut using a microtome (Leica, Richmond Hill, ON, Canada) and collected on Superfrost Plus slides (Fisher Scientific, Ottawa, ON, Canada). Sections were then dewaxed in xylene and rehydrated by successive immersion in descending concentrations of alcohol. When indicated, tissue was stained with Harris’s hematoxylin and eosin. Skin thickness measurements were assessed using image analysis software (Northern Eclipse, Empix, Missassagua, ON, Canada). To assess the effects of PTEN deletion on collagen synthesis, Masson’s Trichrome collagen stain was used. Sections were also subjected for immunofluorescence staining. Tissue sections were blocked by incubation with 5% BSA, 0.1% Triton X-100 in with phosphate-buffered saline (overnight, 4 1C), washed with phosphate-buffered saline and then incubated with primary antibodies for 1 hour at room temperature under humidified conditions. Primary antibodies used were antiCCN2 (1:500 dilution, Abcam) and anti-p-Akt (1:50 dilution, Cell Signaling). After primary antibody incubation, sections were then washed with phosphate-buffered saline and incubated with appropriate fluorescent secondary antibodies (Invitrogen) 1 hour at 37 1C. Sections were washed with phosphate-buffered saline, mounted using 40 ,6-diamidino-2-phenylindole and photographed using a Zeiss fluorescence microscope and Northern Eclipse software (Empix). The percentage of positive cell staining was calculated by numbers of cells per mm2 and was detected using image analysis software (Northern Eclipse, Empix).

Hydroxyproline assay Hydroxyproline assay was performed as a marker of collagen synthesis in wound tissues as previously described (Samuel, 2009). Tissues were homogenized in saline, hydrolyzed with 2 N NaOH for 30 minutes at 120 1C. Briefly, skin biopsies were freeze-dried overnight, dry-weight noted, followed by submersion in 6 M HCl before being hydrolyzed at 110 1C for 18–24 hours. The samples were dessicated in the presence of NaOH overnight and then reconstituted in 0.1 M HCl. Hydroxyproline levels were assessed by modification of the Neumann and Logan’s reaction using Chloramine T and Ehrlich’s reagent (Sigma) using a hydroxyproline standard curve and measuring at 558 nm. Values were expressed as mg of hydroxyproline per mg of protein. CONFLICT OF INTEREST The authors state no conflict of interest.

ACKNOWLEDGMENTS This work was supported by grants from the Canadian Foundation for Innovation, the Canadian Institutes of Health Research (to AL and MB), and the Arthritis Research Campaign and the Reynaud’s and Scleroderma Association (to DJA and CPD). AL is a New Investigator of the Arthritis Society (Scleroderma Society of Ontario), the recipient of an Early Researcher Award, and a member of the Canadian Scleroderma Research Group New Emerging Team. SKP was supported by the Canadian Scleroderma Research Group.

2002 Journal of Investigative Dermatology (2011), Volume 131

Supplementary material is linked to the online version of the paper at http:// www.nature.com/jid

REFERENCES Abraham DJ, Krieg T, Distler J et al. (2009) Overview of pathogenesis of systemic sclerosis. Rheumatology (Oxford) 48(Suppl 3):iii3–7

Cai CL, Martin JC, Sun Y et al. (2008) A myocardial lineage derives from Tbx18 epicardial cells. Nature 454:104–8 Carnero A (2010) The PKB/AKT pathway in cancer. Curr Pharm Des 16:34–44 Chen Y, Shi-wen X, van Beek JKennedy L et al. (2005) Matrix contraction by dermal fibroblasts requires TGFbeta/ALK5, heparan sulfate containing proteoglycans and MEK/ERK: insights into pathological scarring in chronic fibrotic disease. Am J Pathol 167:1699–711 Chen Y, Shi-wen X, Eastwood M et al. (2006) Contribution of activin receptorlike kinase 5 (transforming growth factor beta receptor type I) signaling to the fibrotic phenotype of scleroderma fibroblasts. Arthritis Rheum 54:1309–16 Groszer M, Erickson R, Scripture-Adams DD et al. (2001) Negative regulation of neural stem/progenitor cell proliferation by the Pten tumor suppressor gene in vivo. Science 294:2186–9 Hao LS, Zhang XL, An JY et al. (2009) PTEN expression is down-regulated in liver tissues of rats with hepatic fibrosis induced by biliary stenosis. APMIS 117:681–91 Inman GJ, Nicola´s FJ, Callahan JF et al. (2002) SB-431542 is a potent and specific inhibitor of transforming growth factor-beta superfamily type I activin receptor-like kinase (ALK) receptors ALK4, ALK5, and ALK7. Mol Pharmacol 62:65–74 Ishida W, Mori Y, Lakos G et al. (2006) Intracellular TGF-beta receptor blockade abrogates Smad-dependent fibroblast activation in vitro and in vivo. J Invest Dermatol 126:1733–44 Jun JB, Kuechle M, Min J et al. (2005) Scleroderma fibroblasts demonstrate enhanced activation of Akt (protein kinase B) in situ. J Invest Dermatol 124:298–303 Kato M, Putta S, Wang M et al. (2009) TGF-beta activates Akt kinase through a microRNA-dependent amplifying circuit targeting PTEN. Nat Cell Biol 11:881–9 Leask A (2010) Towards an anti-fibrotic therapy for scleroderma: targeting myofibroblast differentiation and recruitment. Fibrogenesis Tissue Repair 3:8 Leask A (2011) Possible strategies in antifibrotic drug intervention in scleroderma. J Cell Commun Signal 5:125–9 Leask A, Parapuram SK, Shi-Wen X et al. (2009) Connective tissue growth factor (CTGF, CCN2) gene regulation: a potent clinical bio-marker of fibroproliferative disease? J Cell Commun Signal 3:89–94 LeRoy EC, Black C, Fleischmajer R et al. (1988) Scleroderma (systemic sclerosis): classification, subsets and pathogenesis. J Rheumatol 15:202–5 Lesche R, Groszer M, Gao J et al. (2002) Cre/loxP-mediated inactivation of the murine Pten tumor suppressor gene. Genesis 32:148–9 Liu S, Kapoor M, Leask A (2009) Rac1 expression by fibroblasts is required for tissue repair in vivo. Am J Pathol 174:1847–56 Liu S, Kapoor M, Shi-wen X et al. (2008) Investigation of the role of Rac1 in a bleomycin-induced scleroderma model using fibroblast-specific Rac1 knockout mice. Arthritis Rheum 58:2189–95 Lu Y, Azad N, Wang L et al. (2010) Phosphatidylinositol-3-kinase/akt regulates bleomycin-induced fibroblast proliferation and collagen production. Am J Respir Cell Mol Biol 42:432–41 Pala D, Kapoor M, Woods A et al. (2008) Focal adhesion kinase/Src suppresses early chondrogenesis: central role of CCN2. J Biol Chem 283:9239–47 Samuel CS (2009) Determination of collagen content, concentration and subtypes in kidney tissue from: methods in molecular biology. In: Hewitson TD, Becker GJ (eds), Kidney Research, vol. 466. Humana Press: Totowa, NJ

SK Parapuram et al. Loss of PTEN Results in Fibrosis

Sargent JL, Milano A, Bhattacharyya S et al. (2010) A TGFbeta-responsive gene signature is associated with a subset of diffuse scleroderma with increased disease severity. J Invest Dermatol 130:694–705

Trojanowska M (2009) Noncanonical transforming growth factor beta signaling in scleroderma fibrosis. Curr Opin Rheumatol 21: 623–9

Sellers LA, Alderton F, Carruthers AM et al. (2000) Receptor isoforms mediate opposing proliferative effects through gbetagamma-activated p38 or Akt pathways. Mol Cell Biol 20:5974–85

Tsugawa K, Jones MK, Sugimachi K et al. (2002) Biological role of phosphatase PTEN in cancer and tissue injury healing. Front Biosci 7:e245–51

Shi-Wen X, Chen Y, Denton CP et al. (2004) Endothelin-1 promotes myofibroblast induction through the ETA receptor via a rac/phosphoinositide 3-kinase/Akt-dependent pathway and is essential for the enhanced contractile phenotype of fibrotic fibroblasts. Mol Biol Cell 15:2707–19

Wei J, Ghosh AK, Sargent JL et al. (2010) PPARg downregulation by TGFb in fibroblast and impaired expression and function in systemic sclerosis: a novel mechanism for progressive fibrogenesis. PLoS One 5:e13778

Shi-wen X, Denton CP, McWhirter A et al. (1997) Scleroderma lung fibroblasts exhibit elevated and dysregulated collagen type I biosynthesis. Arthritis Rheum 40:1237–44 Shi-wen X, Liu S, Eastwood M et al. (2009) Rac inhibition reverses the phenotype of fibrotic fibroblasts. PLoS One 4:e7438 Shi-Wen X, Renzoni EA, Kennedy L et al. (2006) Endogenous endothelin-1 signaling contributes to type I collagen and CCN2 overexpression in fibrotic fibroblasts. Matrix Biol 26:625–32 Shi-wen X, Pennington D, Holmes A et al. (2000) Autocrine overexpression of CTGF maintains fibrosis: RDA analysis of fibrosis in systemic sclerosis. Exp Cell Res 259:213–24 Teresi RE, Waite KA (2008) PPARgamma, PTEN, and the fight against cancer. PPAR Res 2008:932632

Xia H, Diebold D, Nho R et al. (2008) Pathological integrin signaling enhances proliferation of primary lung fibroblasts from patients with idiopathic pulmonary fibrosis. J Exp Med 205:1659–72 Yang Y, Zhou F, Fang Z et al. (2009) Post-transcriptional and posttranslational regulation of PTEN by transforming growth factor-beta1. J Cell Biochem 106:1102–12 Young AT, Dahl J, Hausdorff SF et al. (1995) Phosphatidylinositol 3-kinase binding to polyoma virus middle tumor antigen mediates elevation of glucose transport by increasing translocation of the GLUT1 transporter. Proc Natl Acad Sci USA 92:11613–7 Zheng B, Zhang Z, Black CM et al. (2002) Ligand-dependent genetic recombination in fibroblasts: a potentially powerful technique for investigating gene function in fibrosis. Am J Pathol 160: 1609–17

www.jidonline.org 2003