SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Objective

Systemic sclerosis (SSc) is characterized by fibrosis of the skin and visceral organs, vascular dysfunction, and immunologic dysregulation. Platelet-derived growth factors (PDGFs) have been implicated in the development of fibrosis and dysregulation of vascular function. We investigated the effects of sunitinib and sorafenib, two tyrosine kinase inhibitors that interfere with PDGF signaling, in a mouse model of diffuse SSc.

Methods

SSc was induced in BALB/c mice by subcutaneous injections of HOCl daily for 6 weeks. Mice were randomized to treatment with sunitinib, sorafenib, or vehicle. The levels of native and phosphorylated PDGF receptor β (PDGFRβ) and vascular endothelial growth factor receptor (VEGFR) in the skin were assessed by Western blot and immunohistochemical analyses. Skin and lung fibrosis were evaluated by histologic and biochemical methods. Autoantibodies were detected by enzyme-linked immunosorbent assay, and spleen cell populations were analyzed by flow cytometry.

Results

Phosphorylation of PDGFRβ and VEGFR was higher in fibrotic skin from HOCl-injected mice with SSc than from PBS-injected mice. Injections of HOCl induced cutaneous and lung fibrosis, increased the proliferation rate of fibroblasts in areas of fibrotic skin, increased splenic B cell and T cell counts, and increased anti–DNA topoisomerase I autoantibody levels in BALB/c mice. All of these features were reduced by sunitinib but not by sorafenib. Sunitinib significantly reduced the phosphorylation of both PDGF and VEGF receptors.

Conclusion

Inhibition of the hyperactivated PDGF and VEGF pathways by sunitinib prevented the development of fibrosis in HOCl-induced murine SSc and may represent a new SSc treatment for testing in clinical trials.

Systemic sclerosis (SSc) is a connective tissue disorder characterized by fibrosis of the skin and visceral organs, vascular dysfunction, and immunologic dysregulation associated with autoantibodies (1). To date, the mechanisms that determine the clinical manifestations of the disease remain unclear (2–4).

Platelet-derived growth factors (PDGFs) are potent mitogens and chemoattractants for cells of mesenchymal and neuroectodermal origin (5). Members of the PDGF family play a major role during embryonic development and contribute to the maintenance of connective tissue in adults (6). Increased levels of PDGF and PDGF receptors (PDGFRs) have been found in skin and lung biopsy samples from patients with scleroderma (7–9). Moreover, sera from patients with SSc may contain autoantibodies directed toward PDGFRs (10). These antibodies can induce the production of reactive oxygen species that activate the MAP kinase/ERK-1/2 pathway and lead to fibroblast proliferation (11). Transforming growth factor β (TGFβ) and interleukin-1α (IL-1α) can also trigger PDGFR signaling through increasing either PDGFR levels or PDGF synthesis by fibroblasts (12). Thus, SSc skin and lung fibroblasts are more sensitive to mitogenic stimulation by PDGF than normal fibroblasts are. The functional significance of the activation of PDGF signaling in fibroblasts has not been fully evaluated in scleroderma, but it may contribute to their enhanced proliferative, migratory, and contractile potential both in vitro and in vivo (13)

Protein tyrosine kinase inhibitors (TKIs) are a new class of therapeutic agents that were initially developed for the treatment of cancers (14, 15). They can inhibit proliferative signals via the blockade of the tyrosine kinases Bcr-Abl or c-Kit. Sunitinib and sorafenib also block the tyrosine kinase activity of PDGFRs, as well as the VEGF/VEGFR pathway, which plays a role in the pathogenesis of SSc (16). The aim of this study was to investigate the effects of sunitinib and sorafenib in the HOCl-induced mouse model of SSc (17). We found that these TKIs prevent fibrosis development, immune activation, and endothelial dysfunction in SSc, and thus appear to be attractive therapeutic tools for this disease.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Animals, cells, and chemicals.

Six-week-old female BALB/c mice were purchased from Harlan and were given humane care according to our institutional guidelines. The project was approved by the Regional Ethics Committee on Animal Experimentation. All chemicals were obtained from Sigma, except for sunitinib, which was from Pfizer, and sorafenib, which was from Bayer Health Care.

Experimental procedure.

Induction of SSc.

Mice were randomly distributed into experimental and control groups (n = 14 per group). A total of 200 μl of substances that generate HOCl was injected intradermally into the back of the mice every day for 6 weeks, as previously described (17). Control groups received injections of 200 μl of sterilized phosphate buffered saline (PBS).

Treatment with TKIs.

Each mouse receiving subcutaneous injections was randomized to receive 6 weeks of oral treatment (by gavage) with sunitinib (50 mg/kg/day), sorafenib (50 mg/kg/day), or vehicle alone. The dosage of 50 mg/kg/day was chosen as being consistent with the report from the European Medicines Agency on sunitinib malate and sorafenib tosylate.

One week after the end of the injections and treatment, the animals were euthanized by cervical dislocation. Lungs were collected, and biopsies of the skin of the back were performed with a punch (6 mm in diameter). Samples for determination of collagen content were stored at –80°C. Samples for histopathologic analysis were fixed in 10% formalin.

Immunofluorescence analysis of skin sections.

Diseased skin was taken from each mouse in each treated and untreated group. Formalin-fixed paraffin-embedded skin sections were dewaxed, and an enzymatic antigen retrieval method was used to overcome antigen masking. Slides were washed for 1 hour at room temperature with sodium borohydrate and then blocked with mouse serum. Slides were stained overnight at 4°C with a 1:200 dilution of anti-PDGFRβ, anti–phosphorylated PDGFRβ, anti-VEGFR, or anti–phosphorylated VEGFR monoclonal antibodies or with isotype control (all from Santa Cruz Biotechnology). A secondary fluorescein isothiocyanate (FITC)–labeled antibody was then applied, and after washing in PBS, slides were examined using an Olympus microscope equipped with an epifluorescence system. Photographs were captured with an Olympus DP70 camera and analyzed with accompanying controller software as previously described (18).

Western blot experiments.

Proteins for pPDGFR/PDGFR analysis were extracted from purified primary skin fibroblasts; those for pVEGFR/VEGFR analysis were extracted from the skin. Proteins (30 μg per sample) were subjected to immunoprecipitation with PDGFRβ antibodies and VEGFR antibodies, respectively, using a protein G immunoprecipitation kit (Sigma-Aldrich). Samples were then subjected to 10% polyacrylamide gel electrophoresis, transferred onto nitrocellulose membranes, blocked for 2 hours with 5% dry milk in Tris buffered saline–Tween (TBST), and then incubated overnight at 4°C with anti-PDGFRβ antibody, anti–phosphorylated PDGFRβ antibody Tyr1021, anti-VEGFR (fetal liver kinase 1) antibody, or anti–phosphorylated VEGFR antibody Tyr996. The membranes were then washed and incubated for 1 hour at room temperature with a horseradish peroxidase–conjugated secondary antibody (all from Santa Cruz Biotechnology).

For α-smooth muscle actin (α-SMA) Western blots, primary skin fibroblasts were treated with several doses of sorafenib or sunitinib. Cell pellets were then thawed and mixed in radioimmunoprecipitation assay buffer and stored at –80°C. Proteins were then subjected to 10% polyacrylamide gel electrophoresis, transferred onto nitrocellulose membranes, blocked for 1 hour with 5% dry milk in TBST, and then incubated overnight at 4°C with an anti–α-SMA antibodies diluted 1:2,000 (clone 1A4; Sigma-Aldrich). As an internal control, we used β-actin.

Assessment of dermal thickness.

The thickness of the skin of the shaved back of each mouse was measured with calipers, and the results were expressed in millimeters. Measurements were performed every week and on the day of euthanization by the same operator (NK). Two sides on the back of each mouse were measured, and the mean was calculated and recorded.

Histopathologic analysis.

A 5-μm–thick tissue section was prepared from the midportion of paraffin-embedded lung and skin pieces and stained with hematoxylin and eosin. Slides were examined under standard brightfield microscopy (Olympus BX60 microscope) by a pathologist (FB) who was blinded to the group assignment of the animal.

Measurement of collagen content in skin and lung tissues.

Skin and lung pieces were diced using a sharp scalpel and were mixed with pepsin (1:10 weight ratio) and 0.5M acetic acid overnight at room temperature, with constant stirring. The collagen content assay was based on the quantitative dye-binding Sircol method (Biocolor) (19).

Isolation of skin fibroblasts and proliferation assays.

Skin fibroblasts were extracted as previously described. Primary fibroblasts (2 × 103/well) were seeded in 96-well plates and incubated for 48 hours with 150 μl of culture medium or a solution of sorafenib (0.75–100 mg/ml) or sunitinib (0.16–25 mg/ml). Cell proliferation was determined by pulsing the cells with 3H-thymidine (1 μCi/well) during the last 16 hours of culture. Results were expressed as absolute counts per minute or as the ratio of the absolute counts per minute with sorafenib or with sunitinib to the absolute counts per minute with medium alone.

Detection of serum antibodies.

Levels of anti–DNA topoisomerase I IgG (anti–topo I IgG) antibodies were detected using DNA topo I–coated enzyme-linked immunosorbent assay (ELISA) microplates (ImmunoVision). A 1:50 serum dilution was used for the determination of anti–topo I IgG antibodies.

Flow cytometric analysis of spleen cell subsets.

Cell suspensions from spleens were prepared after hypotonic lysis of erythrocytes. Cells were incubated with the appropriate labeled antibodies for 45 minutes at 4°C in PBS with 0.1% sodium azide and 5% normal rat serum. Cells were then analyzed with a FACSCanto flow cytometer (BD Biosciences). The antibodies used in this study were FITC-conjugated anti-CD11b, phycoerythrin (PE)–conjugated anti-B220, allophycocyanin (APC)–Cy7–conjugated anti-CD4, PE–Cy7–conjugated anti-CD8, PerCP–Cy–conjugated anti-CD3, and FITC-conjugated anti-B7.1 monoclonal antibodies (BD PharMingen).

ELISA determination of IL-6 and TGFβ production by B cells.

B cells were isolated from splenocytes with CD45R (B220) microbeads and MS columns according to the manufacturer's instructions (Miltenyi Biotec). B cell suspensions were then seeded (1 × 106 cells) in 96-well flat-bottomed plates and cultured for 48 hours in complete medium in the presence of 10 μg/ml of lipopolysaccharide (Sigma-Aldrich). Supernatants were collected, and IL-6 and TGFβ concentrations determined by ELISA (eBioscience). Results are expressed in nanograms per milliliter.

ELISA determination of soluble vascular cell adhesion molecule (sVCAM) in serum.

Levels of sVCAM in mouse serum were measured by ELISA. A 1:800 serum dilution and a mouse sVCAM/CD106 DuoSet kit were used according to the manufacturer's instructions (R&D Systems).

Statistical analysis.

All quantitative data were expressed as the mean ± SEM. Data were compared using a nonparametric Mann-Whitney test or Student's paired t-test. When the analysis included more than 2 groups, one-way analysis of variance was used. P values less than 0.05 were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

High levels of phosphorylated PDGFRβ in skin fibroblasts from mice with HOCl-induced fibrosis.

Higher amounts of phosphorylated PDGFRβ were found in fibrotic areas of skin from HOCl-injected mice than in skin from PBS-injected mice, as demonstrated by immunohistochemistry (Figures 1A and B) and by Western blotting using protein extracts from fibroblasts (Figure 1C). Treatment of mice with HOCl-induced SSc with the TKI sunitinib abrogated the phosphorylation of PDGFRβ in skin fibroblasts, whereas phosphorylated PDGFRs could still be detected in the skin of mice with SSc treated with sorafenib (Figures 1A–C). No significant difference in the expression of nonphosphorylated PDGFRs in mice exposed to HOCl versus control mice exposed to PBS was observed (Figure 1C).

thumbnail image

Figure 1. Effects of sunitinib and sorafenib on the levels of platelet-derived growth factor receptor (PDGFR) and phosphorylated PDGFR in the skin of mice exposed to HOCl. A, Detection of pPDGFR by immunofluorescence in skin sections from a representative mouse from each of the 4 experimental groups (n = 7 per group): phosphate buffered saline (PBS) injected, HOCl injected, HOCl injected and sorafenib treated, and HOCl injected and sunitinib treated. Original magnification × 40. B, Quantification of pPDGFR staining in skin sections, as determined using ImageJ software. Values are the mean ± SEM of 7 mice per group. ∗ = P < 0.05; ∗∗ = P < 0.01 by Mann-Whitney unpaired U test. NS = not significant. C, Levels of native and phosphorylated PDGFR, as determined by Western blotting of protein extracts from purified fibroblasts isolated from mouse skin. β-actin was used as an internal control. Levels of pPDGFR were increased in skin fibroblasts from HOCl-injected mice. Sunitinib down-regulated the phosphorylation of PDGFR. Results are from 2 representative mice (n = 7 per group).

Download figure to PowerPoint

Better prevention of skin fibrosis in HOCl-injected mice by sunitinib than by sorafenib.

As previously observed, subcutaneous injections of HOCl in BALB/c mice increased dermal thickness and the concentration of acid- and pepsin-soluble type I collagen in the skin as compared with injections of PBS (P < 0.001 for dermal thickness and P = 0.003 for collagen concentration in the skin) (Figures 2A and C). Histopathologic analysis confirmed the presence of dermal fibrosis (Figure 2B).

thumbnail image

Figure 2. Effects of sunitinib and sorafenib on skin and lung collagen contents following subcutaneous injections of an HOCl-generating agent or phosphate buffered saline (PBS) in BALB/c mice (n = 7 per group). Skin and lungs were collected at the time of euthanization. A, Dermal thickness in the injected areas of skin from the 4 experimental groups of BALB/c mice. B, Representative sections of injected skin from BALB/c mice. Fibrosis was inhibited by sunitinib more than by sorafenib. Hematoxylin and eosin stained; original magnification × 20. C, Collagen content in skin samples obtained with a 6-mm punch biopsy, as measured by the quantitative dye-binding Sircol method. D, Representative sections of lung from BALB/c mice. Lung fibrosis was inhibited by sunitinib more than by sorafenib. Hematoxylin and eosin stained; original magnification × 20. E, Collagen content in the lung, as measured by the quantitative dye-binding Sircol method. Sunitinib inhibited both skin and lung fibrosis. Values in A, C, and E are the mean ± SEM of 7 mice per group. ∗ = P < 0.05 by Mann-Whitney unpaired U test. NS = not significant.

Download figure to PowerPoint

To evaluate whether the inhibition of tyrosine kinases affects the development of dermal fibrosis in this model of SSc, mice exposed to HOCl were simultaneously treated with sorafenib or sunitinib, two different TKIs that interfere with the PDGF pathway.

Sorafenib moderately reduced the dermal thickness (P = 0.001) and the accumulation of collagen (P = 0.529) induced by HOCl as compared with untreated mice exposed to HOCl (Figures 2A and C). These results were confirmed by histopathologic analysis of skin biopsy samples stained with hematoxylin and eosin (Figure 2B), which showed a moderate decrease in dermal thickness in HOCl-injected BALB/c mice treated with sorafenib.

In contrast, sunitinib significantly reduced dermal thickness and accumulation of collagen in the skin of HOCl-injected mice (P = 0.0003 and P = 0.039, respectively, versus untreated HOCl-injected mice) (Figures 2A and C). These results were confirmed by histopathologic analysis (Figure 2B).

Better prevention of lung fibrosis in HOCl-injected mice by sunitinib than by sorafenib.

In addition to skin fibrosis, HOCl-injected mice developed lung fibrosis, as shown by histopathologic analysis (Figure 2D) and by the higher concentration of type I collagen in the lungs of HOCl-injected mice than in PBS-injected mice (P < 0.001) (Figure 2E). Sorafenib did not abrogate the development of lung fibrosis induced by HOCl, as shown by histopathologic analysis and by the accumulation of type I collagen in lungs (P = 0.139 versus untreated HOCl-injected mice) (Figures 2D and E).

In contrast, sunitinib significantly reduced the concentration of type I collagen in the lungs as compared with untreated HOCl-injected mice (P = 0.012) (Figure 2E). These findings were confirmed by histopathologic analysis of lung biopsy sections stained with hematoxylin and eosin (Figure 2D), which showed a more marked decrease in lung fibrosis in BALB/c mice treated with sunitinib than in untreated HOCl-injected mice.

Analysis of bronchoalveolar lavage fluid from 4 mice showed a significant increase in the numbers of total leukocytes (P = 0.029), neutrophils (P = 0.028), monocytes (P = 0.036), and lymphocytes (P = 0.032) in HOCl-injected mice, which were decreased with sunitinib (P = 0.079, P = 0.88, and P = 0.048, respectively) (data not shown).

Normalization of the rate of dermal fibroblast proliferation by sunitinib and sorafenib treatment.

We next investigated whether treatment with various TKIs modified the growth of fibroblasts isolated from the fibrotic skin of mice with SSc. As previously reported, skin fibroblasts isolated from HOCl-injected mice displayed a higher rate of proliferation than did fibroblasts from mice injected with PBS (P = 0.007) (Figure 3A). The rate of proliferation of fibroblasts isolated from HOCl-injected mice treated with sorafenib or sunitinib was lower than that of fibroblasts isolated from HOCl-injected mice treated with PBS (P = 0.028 for sorafenib and P < 0.001 for sunitinib) (Figure 3A). Sunitinib was more efficient than sorafenib in reducing the rate of dermal fibroblast proliferation (P = 0.019) (Figure 3A).

thumbnail image

Figure 3. Proliferative properties of skin fibroblasts. A, Fibroblasts were isolated from injected areas of skin from the 4 experimental groups obtained at the time of euthanization, and cell proliferation was determined by measurement of 3H-thymidine (1 μCi/well) incorporation during the last 16 hours of culture. Fibroblasts from HOCl-injected mice treated with sunitinib showed a lower proliferation rate than those from HOCl-injected mice treated with vehicle alone. Values are the mean ± SEM. ∗ = P < 0.05 by Mann-Whitney unpaired U test. PBS = phosphate buffered saline. B–D, Fibroblasts from HOCl-injected mice were isolated and treated in vitro with various doses of sunitinib or sorafenib. Western blotting for the expression of α-smooth muscle actin (α-SMA) was performed in untreated cells and in cells treated with 2 different concentrations of sorafenib or sunitinib (B). β-actin was used as an internal control. The fibroblast proliferation rate was measured in 4 HOCl-injected mice after exposure to increasing concentrations of sunitinib (C) or sorafenib (D). Results in C and D are expressed as the ratio of proliferation with sunitinib or sorafenib, respectively, to the proliferation with medium alone.

Download figure to PowerPoint

We then analyzed the in vitro effects of both TKIs on fibroblast proliferation. At low doses (0.16–0.75 mg/ml), sunitinib had no effect, whereas at 3.125 mg/ml, the proliferation rate was reduced by 75% (Figure 3C). In vitro treatment with sorafenib also had major effects on fibroblast proliferation, but effective doses were 4 times higher than the effective doses of sunitinib (Figure 3D). Indeed, the 50% inhibition concentration (IC50) of sunitinib was 2.75 mg/ml, whereas for sorafenib, the IC50 was 9.75 mg/ml (P = 0.029). These results were confirmed by Western blotting showing α-SMA expression in fibroblasts after in vitro exposure to different doses of sorafenib or sunitinib (Figure 3B).

Decreased serum levels of HOCl-induced anti–topo I autoantibodies by sunitinib and sorafenib treatment.

We next tested the effects of the two TKIs on the specific autoimmune response to DNA topo I that characterizes the diffuse cutaneous SSc phenotype. IgG antibodies directed toward DNA topo I were found, as usual, in the sera of mice exposed to HOCl for 6 weeks (P < 0.001 versus PBS-injected mice) (Figure 4C). As previously observed, no significant levels of other autoantibodies, such as anti–DNA IgG antibodies, anti–cardiolipin IgG antibodies, or rheumatoid factors, could be detected in the sera of HOCl-injected mice (data not shown).

thumbnail image

Figure 4. Effects of sunitinib and sorafenib on spleen cells, anti–DNA topoisomerase I (anti–topo I) autoantibody production, and interleukin-6 (IL-6) and transforming growth factor β (TGFβ) production. BALB/c mice were injected daily for 6 weeks with either an HOCl-generating agent or phosphate buffered saline (PBS) and treated with sunitinib, sorafenib, or vehicle alone. Spleens and sera were collected at the time of euthanization, and cells were enumerated by flow cytometry. A, Total numbers of B220+ cells in the spleen. B, Total numbers of CD4+ T cells in the spleen. C, Levels of anti–topo I antibodies (Abs), as determined by enzyme-linked immunosorbent assay in sera, expressed in arbitrary units (AU). D, Percentages of B7.1+ B cells in the spleen. E, Concentrations of IL-6 in supernatants from B cells stimulated for 48 hours with lipopolysaccharide (LPS). F, Concentrations of TGFβ in supernatants from B cells stimulated for 48 hours with LPS. Sunitinib treatment decreased splenic B cell and CD4+ T cell numbers, the concentration of anti–topo I antibodies, and the level of activation of B cells. Values are the mean ± SEM of 7 mice per group. ∗ = P < 0.05; ∗∗ = P < 0.01 by Mann-Whitney unpaired U test. NS = not significant.

Download figure to PowerPoint

Both sunitinib and sorafenib prevented the development of anti–topo I IgG antibodies in mice exposed to HOCl (P < 0.001 for comparison of each TKI versus untreated mice) (Figure 4C).

Decreased expansion of splenic B cells in HOCl-injected mice by sunitinib, but not sorafenib, treatment.

We next investigated the effects of the 2 TKIs on the different spleen cell populations. Daily subcutaneous exposure to HOCl for 6 weeks increased the numbers of splenic B220+ B cells and CD4+ T cells in the HOCl-injected mice as compared to PBS-injected mice (P < 0.001 for each comparison) (Figures 4A and B). Sunitinib prevented the increase in splenic B cell and CD4+ T cell numbers in mice exposed to HOCl (P = 0.002 and P = 0.004, respectively, versus untreated mice) (Figures 4A and B). Moreover, treatment with sunitinib reduced the expression of B7.1 on splenic B cells (P = 0.0089). Sorafenib tended to decrease B cell and CD4+ T cell numbers and B7.1 expression on B cells, but these results didn't achieve significance (P = 0.095, P = 0.48, and P = 0.21, respectively) (Figures 4A, B, and D).

Reduced production of IL-6 and TGFβ by B cells in HOCl-injected mice by sunitinib, but not sorafenib, treatment.

Since sunitinib and sorafenib exerted beneficial effects on autoantibody production and B cell activation, we tested their actions on the production of profibrotic cytokines by B cells. Both sorafenib and sunitinib decreased the production of IL-6 (P = 0.032 and P = 0.029, respectively) and TGFβ (P = 0.097 and P = 0.010, respectively) (Figures 4E and F).

Significant reduction of phosphorylated VEGFR concentrations in the skin of mice with HOCl-induced fibrosis by sunitinib treatment.

Since VEGF signaling has been found to be abnormal in SSc patients, we investigated the level of phosphorylation of VEGF receptors in the fibrotic skin of HOCl-injected mice and the effect of the two TKIs on this pathway. Higher levels of phosphorylated VEGFRs were detected in the diseased areas of skin from HOCl-injected mice than in skin from control mice (Figure 5). Sunitinib prevented the phosphorylation of this receptor, since no phosphorylated VEGFRs were found in the diseased areas of skin from HOCl-injected mice treated with sunitinib, as assessed by immunohistochemistry and Western blotting (Figure 5). In contrast, sorafenib had only a weak effect (Figure 5).

thumbnail image

Figure 5. Effects of sunitinib and sorafenib on the levels of phosphorylated vascular endothelial growth factor receptor (pVEGFR) in the skin of mice injected with HOCl. A, Immunohistochemistry of skin sections obtained from a representative mouse from each of the 4 experimental groups (n = 7 per group). Slides were stained with anti-pVEGFR monoclonal antibody. Original magnification × 40. PBS = phosphate buffered saline. B, Quantification of pVEGFR staining in skin sections, as determined using ImageJ software. Values are the mean ± SEM of 7 mice per group. ∗ = P < 0.05 by Mann-Whitney unpaired U test. NS = not significant. C, Levels of native and phosphorylated VEGFR, as determined by Western blotting of protein extracts from the skin. Sunitinib reduced the levels of pVEGFR in the skin of HOCl-injected mice. Results are from 2 representative mice (n = 7 per group).

Download figure to PowerPoint

Prevention of increased serum levels of VCAM by sunitinib and sorafenib treatment.

Elevated serum levels of markers of endothelial cell damage such as sVCAM are observed in SSc. This was confirmed in mice exposed to HOCl as compared with those exposed to PBS (P = 0.037) (Figure 6). Sunitinib and sorafenib prevented the increase in sVCAM in mice exposed to HOCl (P = 0.015 for sunitinib versus vehicle alone and P = 0.010 for sorafenib versus vehicle alone) (Figure 6).

thumbnail image

Figure 6. Effects of sunitinib and sorafenib on serum concentrations of soluble vascular cell adhesion molecule (sVCAM). BALB/c mice were injected daily for 6 weeks with either an HOCl-generating agent or phosphate buffered saline (PBS) and treated with sunitinib, sorafenib, or vehicle alone. Sera were collected at the time of euthanization, and concentrations of sVCAM were determined by enzyme-linked immunosorbent assay. Soluble VCAM concentrations were lower in sera from HOCl-injected mice treated with sunitinib than in those from HOCl-injected mice treated with vehicle alone. Values are the mean ± SEM of 7 mice per group. ∗ = P < 0.05 by Mann-Whitney unpaired U test. NS = not significant.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

In the present study, we describe the hyperactivation of PDGFRβ and VEGFR in areas of fibrotic skin obtained from mice with HOCl-induced SSc. In addition, we show that TKIs targeting the PDGF and VEGF pathways can represent effective treatments in SSc by abrogating the fibrotic process, endothelial damage, and autoimmune activation that characterize the disease.

PDGFs and their receptors are physiologically involved in the embryogenesis of many organs and in the formation of blood vessels (6). They are also implicated in various diseases. Autocrine or paracrine activation of PDGF signaling pathways has been demonstrated in certain type of cancers and has been shown to affect tumor growth, angiogenesis, invasion, and metastasis. PDGFs and their receptors also play a role in vascular disorders, such as atherosclerosis and pulmonary hypertension, as well as in fibrotic diseases, including pulmonary fibrosis, liver cirrhosis, and cardiac fibrosis.

We first assessed the presence of high levels of PDGFRβ and its phosphorylation in the fibrotic skin of HOCl-injected mice. The elevated rate of phosphorylation, which reflects receptor hyperactivation, is consistent with the role played by PDGF in several fibrotic diseases. For example, high levels of phosphorylated PDGFRs have been reported in a histologic study of 2 patients with SSc (20). In other studies in which the phosphorylation levels were not investigated, elevated expression of PDGFs or PDGFRs in various tissues obtained from SSc patients have been reported. In contrast, PDGF is almost undetectable in healthy skin or lung (7, 8). Similarly, elevated levels of PDGF-A and PDGF-B have been found in bronchoalveolar lavage fluid from scleroderma patients (9). PDGFRs and their signaling pathway may be activated not only by the binding of PDGF, but also by the autoantibodies found in SSc patients (10). Another possible mechanism, which has been observed in SSc fibroblasts, is that TGFβ released by platelets or infiltrating mononuclear cells increases the expression of PDGFRs and enhances the mitogenic effect of PDGF-A (21). Taken together, those data strongly suggest a crucial role of the PDGF signaling pathway in the pathogenesis of SSc.

We therefore undertook an investigation of the potential inhibiting effects of 2 TKIs that interfere with the PDGF signaling pathway: sunitinib and sorafenib. Sunitinib was more effective than the same dosage of sorafenib in preventing skin and lung fibrosis in the HOCl-induced SSc model. The higher capacity of sunitinib over sorafenib to prevent fibrosis was associated with a stronger reduction of the phosphorylation rates of PDGFRβ and VEGFRs. Sorafenib can in fact reduce some features of autoimmunity and abrogate endothelial cell damage, but because of its low capacity to prevent fibrosis, the drug, when administered at a dosage of 50 mg/kg/day, is not able to prevent the development of disease; in contrast, sunitinib acts simultaneously on the vascular, immune, and fibrotic features of the disease.

Many differences between sorafenib and sunitinib can account for the superiority of the latter drug in our experiments. First, sorafenib has a shorter half-life (22), being 30–40 hours, whereas the half-life of sunitinib is 80–100 hours. Second, sunitinib is metabolized as the active metabolite SU012662, increasing the duration of the effects of the molecule. SU012662 is 2-fold less potent than sunitinib, but it inhibits PDGF and VEGF receptors and contributes to the pharmacologic activity of sunitinib. Third, sorafenib and sunitinib do not have identical activity on their target receptors (23). With regard to PDGFRβ, for example, the IC50 of sorafenib is 1129 nM, whereas it is 75 nM for sunitinib (i.e., >15-fold less). Moreover, the two TKIs do not have identical kinase selectivity, as they both are multikinase inhibitors. Sunitinib has a wider spectrum of inhibition than sorafenib and could inhibit additional receptors, possibly affecting the development of scleroderma in our model (23). Despite these differences, however, it is possible that sorafenib would have been more potent in SSc at higher doses, and further studies are needed to confirm the effects of sorafenib on fibrosis and vascular dysfunctions.

The reduction of dermal fibrosis induced by the profibrotic drug bleomycin has been observed in studies using other TKIs that target PDGFRs, such as imatinib, dasatinib, and nilotinib (24, 25). However, the effects on lung fibrosis have not yet been studied. Sorafenib, in contrast, has been shown to attenuate intrahepatic fibrogenesis, hydroxyproline accumulation, and collagen deposition in 2 rat models of liver fibrosis (26). Our study is the first to show clearcut effectiveness of sunitinib on the development of lung fibrosis in experimental SSc. A study of 5 SSc patients with lung fibrosis treated with a combination of imatinib and cyclophosphamide reported variable results among the patients (27). However, the small number of patients and the absence of a control group render the interpretation of these data difficult (27).

The skin and lungs of patients with SSc contain myofibroblasts, a cell population that produces high amounts of collagen, expresses α-SMA, and displays an excessive rate of proliferation (28–30). The same observation has been made in the mouse model of HOCl-induced SSc (18). The reduction of fibroblast proliferation by sunitinib and sorafenib in fibrotic areas of the skin of mice exposed to HOCl is consistent with the inhibitory role of other TKIs: imatinib has been shown to inhibit the proliferation of normal and sclerodermatous human fibroblasts in the presence or absence of PDGF or TGFβ in vitro (8,20). Moreover, treatment with dasatinib and nilotinib was shown to strongly decrease the number of myofibroblasts in bleomycin-induced dermal fibrosis (25).

In addition to the inhibition of fibroblast proliferation and myofibroblast differentiation, we suggest that 2 other mechanisms can also lead to the reduction of skin fibrosis by sunitinib and, to a lesser extent, sorafenib in our SSc model. First, by inhibiting the phosphorylation of PDGFRs, sunitinib and sorafenib may directly decrease collagen production by fibroblasts. This phenomenon has been observed in 2 models of liver fibrosis in which sorafenib treatment directly reduced collagen production by hepatic stellate cells in vitro (26). Dasatinib and nilotinib, 2 other TKIs that interfere with the PDGF pathway, have been shown to decrease collagen synthesis in dermal fibroblasts from bleomycin-treated mice or from scleroderma patients (25). Alternatively, sunitinib and sorafenib can indirectly mediate a reduction in collagen deposition by decreasing the expression of tissue inhibitor of metalloproteinases 1 (TIMP-1), thereby potentially enhancing extracellular matrix degradation. Such a phenomenon, which is caused by an alteration in the balance between matrix metalloproteinase 13 and TIMP-1, has been observed in hepatic stellate cells treated with sorafenib and in dermal fibroblasts exposed to dasatinib or nilotinib (26).

As in patients with diffuse cutaneous SSc, the mice with HOCl-induced SSc develop anti–topo I autoantibodies, and increased numbers of B cells and CD4+ T cells are found in the spleens of these mice. Both sunitinib and sorafenib prevent the development of these autoantibodies as well as the B cell production of IL-6 and TGFβ. However, only sunitinib significantly limits the expansion of B cells and CD4+ T cells in the spleen and B cell activation through down-regulation of B7.1. To our knowledge, the effects of sunitinib on the immune system have not previously been reported.

Imatinib may induce a reversible dose-dependent lymphopenia and hypogammaglobulinemia, inhibit the development of dendritic cells derived from human CD34+ progenitor cells, and inhibit the expansion of memory cytotoxic T lymphocytes without affecting primary T cell or B cell responses (31, 32). These effects of imatinib may be linked to the inhibition of the Abl protein tyrosine kinase, but the exact mechanism remains unclear. Sunitinib only slightly interferes with Abl, but it strongly inhibits the VEGF receptor. VEGFs play a key role in activating naive T cells (33). VEGFR-1, which is expressed in monocyte/macrophages at both the messenger RNA and protein levels, is involved in the migration of these cells and in the initiation and perpetuation of chronic inflammation (34, 35). Mice lacking VEGFR-1 do not develop type II collagen arthritis because they display a decreased inflammatory response of monocyte/macrophages and decreased hematopoietic proliferation (35).

Based on these findings, we hypothesize that the effects of sunitinib on the immune cells observed in our SSc model may be dependent on the inhibition of the phosphorylation of VEGFR. Moreover, sunitinib strongly inhibits B7.1 expression on B cells. The effects of B7.1 downstream signaling on B cell/T cell cooperation may also explain why sunitinib is more effective than sorafenib in preventing disease progression. Indeed, experiments using SCID mice have clearly demonstrated that the immune system is required for the systemic spreading of the disease in the HOCl model of SSc (17).

In addition to fibroblast hyperproliferation and collagen hyperproduction, SSc is characterized by vascular abnormalities. High levels of soluble VCAM, a marker of damage to the endothelium, are found in the sera of SSc patients (36) and in sera from mice with HOCl-induced SSc. Sunitinib and, to a lesser extent, sorafenib prevent the elevation of VCAM in the sera of these mice. Little is known about the events that initiate vascular injury, prevent its repair, and lead to loss of angiogenesis (37). Early stages of SSc are characterized by an exaggerated angiogenic response, which is later followed by fibrosis (38, 39). One of the predominant growth factors associated with vascular endothelial proliferation, migration, and survival is VEGF (40). VEGF binds to 2 major receptors, VEGFR-1 and VEGFR-2. The VEGF-A/VEGFR-2 signaling pathway promotes the proliferation, survival, adhesion, and migration of endothelial cells. Several groups of investigators have reported the up-regulation of VEGF-A and its receptors VEGFRs in sclerodermatous skin (41–43), consistent with the data reported herein.

Taken together, all of these studies indicate a dysregulation of the VEGF/VEGFR axis in SSc. This dysregulation varies with the stage of the disease but is undisputable. A chronic and uncontrolled activation of VEGF-A/VEGFR-2 signaling may play a role in the microangiopathy of SSc (41). In addition, a prolonged exposure to VEGF and uncontrolled activation of VEGFRs can induce the formation of irregularly abnormal vessels with reduced blood flow in vitro (44). Both sunitinib and sorafenib block the kinase activity associated with VEGFRs. The beneficial effects following the use of these 2 drugs on the markers of endothelium damage emphasize the above hypothesis that excessive activation of the VEGFR signaling pathway may trigger the endothelial dysfunction in SSc.

In conclusion, the findings of this study emphasize the pivotal role of the PDGF/PDGFRβ signaling pathway in SSc by showing a steady phosphorylated state of PDGFRβ in the fibrotic skin of a mouse model of SSc. In addition, the results of this work strengthen the hypothesis that the uncontrolled activation of the VEGF/VEGFR pathway could be deleterious in this disease and could be directly involved in the microangiopathy of SSc. Our findings also suggest that some TKIs that target both PDGFRs and VEGFRs could be a new and attractive therapeutic tool for use in SSc by acting on the fibrotic, immune, and vascular components of the disease.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Batteux had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Kavian, Servettaz, Weill, Batteux.

Acquisition of data. Kavian, Marut, Nicco, Chéreau.

Analysis and interpretation of data. Kavian, Servettaz, Marut, Nicco, Chéreau, Batteux.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

The authors are indebted to Ms Agnes Colle for typing the manuscript.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES
  • 1
    Gabrielli A, Avvedimento EV, Krieg T. Scleroderma. N Engl J Med 2009; 360: 19892003.
  • 2
    Mukerjee D, St George D, Coleiro B, Knight C, Denton CP, Davar J, et al. Prevalence and outcome in systemic sclerosis associated pulmonary arterial hypertension: application of a registry approach. Ann Rheum Dis 2003; 62: 108893.
  • 3
    Matucci-Cerinic M, Denton CP, Furst DE, Mayes MD, Hsu VM, Carpentier P, et al. Bosentan treatment of digital ulcers related to systemic sclerosis: results from the RAPIDS-2 randomised, double-blind, placebo-controlled trial. Ann Rheum Dis 2011; 70: 328.
  • 4
    Launay D, Sitbon O, Le Pavec J, Savale L, Tcherakian C, Yaici A, et al. Long-term outcome of systemic sclerosis-associated pulmonary arterial hypertension treated with bosentan as first-line monotherapy followed or not by the addition of prostanoids or sildenafil. Rheumatology (Oxford) 2010; 49: 490500.
  • 5
    Fredriksson L, Li H, Eriksson U. The PDGF family: four gene products form five dimeric isoforms. Cytokine Growth Factor Rev 2004; 15: 197204.
  • 6
    Andrae J, Gallini R, Betsholtz C. Role of platelet-derived growth factors in physiology and medicine. Genes Dev 2008; 22: 1276312.
  • 7
    Gay S, Jones JR, Huang GQ, Gay R. Immunohistologic demonstration of platelet-derived growth factor (PDGF) and sis-oncogene expression in scleroderma. J Invest Dermatol 1989; 92: 3013.
  • 8
    Soria A, Cario-Andre M, Lepreux S, Rezvani HR, Pasquet JM, Pain C, et al. The effect of imatinib (Glivec) on scleroderma and normal dermal fibroblasts: a preclinical study. Dermatology 2008; 216: 10917.
  • 9
    Ludwicka A, Ohba T, Trojanowska M, Yamakage A, Strange C, Smith EA, et al. Elevated levels of platelet derived growth factor and transforming growth factor-β1 in bronchoalveolar lavage fluid from patients with scleroderma. J Rheumatol 1995; 22: 187683.
  • 10
    Svegliati Baroni S, Santillo M, Bevilacqua F, Luchetti M, Spadoni T, Mancini M, et al. Stimulatory autoantibodies to the PDGF receptor in systemic sclerosis. N Engl J Med 2006; 354: 266776.
  • 11
    Svegliati S, Cancello R, Sambo P, Luchetti M, Paroncini P, Orlandini G, et al. Platelet-derived growth factor and reactive oxygen species (ROS) regulate Ras protein levels in primary human fibroblasts via ERK1/2. J Biol Chem 2005; 280: 3647482.
  • 12
    Trojanowska M. Role of PDGF in fibrotic diseases and systemic sclerosis. Rheumatology (Oxford) 2008; 47: 24.
  • 13
    Tingstrom A, Heldin CH, Rubin K. Regulation of fibroblast-mediated collagen gel contraction by platelet-derived growth factor, interleukin-1α and transforming growth factor-β1. J Cell Sci 1992; 102: 31522.
  • 14
    Virgili A, Koptyra M, Dasgupta Y, Glodkowska-Mrowka E, Stoklosa T, Nacheva EP, et al. Imatinib sensitivity in BCR-ABL1-positive chronic myeloid leukemia cells is regulated by the remaining normal ABL1 allele. Cancer Res 2011; 71: 53816.
  • 15
    Mahmood ST, Agresta S, Vigil C, Zhao X, Han G, D'Amato G, et al. Phase II study of sunitinib malate, a multitargeted tyrosine kinase inhibitor in patients with relapsed or refractory soft tissue sarcomas: focus on 3 prevalent histologies: leiomyosarcoma, liposarcoma, and malignant fibrous histiocytoma. Int J Cancer 2011; 129: 19639.
  • 16
    Pendergrass SA, Hayes E, Farina G, Lemaire R, Farber HW, Whitfield ML, et al. Limited systemic sclerosis patients with pulmonary arterial hypertension show biomarkers of inflammation and vascular injury. PLoS One 2010; 5: e12106.
  • 17
    Servettaz A, Goulvestre C, Kavian N, Nicco C, Guilpain P, Chereau C, et al. Selective oxidation of DNA topoisomerase I induced systemic sclerosis in the mouse. J Immunol 2009; 182: 585564.
  • 18
    Kavian N, Servettaz A, Mongaret C, Wang A, Nicco C, Chereau C, et al. Targeting ADAM-17/Notch signaling abrogates the development of systemic sclerosis in a murine model. Arthritis Rheum 2010; 62: 347787.
  • 19
    Burdick MD, Murray LA, Keane MP, Xue YY, Zisman DA, Belperio JA, et al. CXCL11 attenuates bleomycin-induced pulmonary fibrosis via inhibition of vascular remodeling. Am J Respir Crit Care Med 2005; 171: 2618.
  • 20
    Chung L, Fiorentino DF, BenBarak MJ, Adler AS, Mariano MM, Paniagua RT, et al. Molecular framework for response to imatinib mesylate in systemic sclerosis. Arthritis Rheum 2009; 60: 58491.
  • 21
    Yamakage A, Kikuchi K, Smith EA, LeRoy EC, Trojanowska M. Selective upregulation of platelet-derived growth factor α receptors by transforming growth factor β in scleroderma fibroblasts. J Exp Med 1992; 175: 122734.
  • 22
    Clark JW, Eder JP, Ryan D, Lathia C, Lenz HJ. Safety and pharmacokinetics of the dual action Raf kinase and vascular endothelial growth factor receptor inhibitor, BAY 43-9006, in patients with advanced, refractory solid tumors. Clin Cancer Res 2005; 11: 547280.
  • 23
    Kumar R, Crouthamel MC, Rominger DH, Gontarek RR, Tummino PJ, Levin RA, et al. Myelosuppression and kinase selectivity of multikinase angiogenesis inhibitors. Br J Cancer 2009; 101: 171723.
  • 24
    Akhmetshina A, Venalis P, Dees C, Busch N, Zwerina J, Schett G, et al. Treatment with imatinib prevents fibrosis in different preclinical models of systemic sclerosis and induces regression of established fibrosis. Arthritis Rheum 2009; 60: 21924.
  • 25
    Akhmetshina A, Dees C, Pileckyte M, Maurer B, Axmann R, Jungel A, et al. Dual inhibition of c-abl and PDGF receptor signaling by dasatinib and nilotinib for the treatment of dermal fibrosis. FASEB J 2008; 22: 221422.
  • 26
    Wang Y, Gao J, Zhang D, Zhang J, Ma J, Jiang H. New insights into the antifibrotic effects of sorafenib on hepatic stellate cells and liver fibrosis. J Hepatol 2010; 53: 13244.
  • 27
    Sabnani I, Zucker MJ, Rosenstein ED, Baran DA, Arroyo LH, Tsang P, et al. A novel therapeutic approach to the treatment of scleroderma-associated pulmonary complications: safety and efficacy of combination therapy with imatinib and cyclophosphamide. Rheumatology (Oxford) 2009; 48: 4952.
  • 28
    Rajkumar VS, Howell K, Csiszar K, Denton CP, Black CM, Abraham DJ. Shared expression of phenotypic markers in systemic sclerosis indicates a convergence of pericytes and fibroblasts to a myofibroblast lineage in fibrosis. Arthritis Res Ther 2005; 7: R111323.
  • 29
    Kirk TZ, Mark ME, Chua CC, Chua BH, Mayes MD. Myofibroblasts from scleroderma skin synthesize elevated levels of collagen and tissue inhibitor of metalloproteinase (TIMP-1) with two forms of TIMP-1. J Biol Chem 1995; 270: 34238.
  • 30
    Yamamoto T, Nishioka K. Animal model of sclerotic skin. V: increased expression of a smooth muscle actin in fibroblastic cells in bleomycin-induced scleroderma. Clin Immunol 2002; 102: 7783.
  • 31
    Appel S, Boehmler AM, Grunebach F, Muller MR, Rupf A, Weck MM, et al. Imatinib mesylate affects the development and function of dendritic cells generated from CD34+ peripheral blood progenitor cells. Blood 2004; 103: 53844.
  • 32
    Mumprecht S, Matter M, Pavelic V, Ochsenbein AF. Imatinib mesylate selectively impairs expansion of memory cytotoxic T cells without affecting the control of primary viral infections. Blood 2006; 108: 340613.
  • 33
    Kim YS, Hong SW, Choi JP, Shin TS, Moon HG, Choi EJ, et al. Vascular endothelial growth factor is a key mediator in the development of T cell priming and its polarization to type 1 and type 17 T helper cells in the airways. J Immunol 2009; 183: 511320.
  • 34
    Barleon B, Sozzani S, Zhou D, Weich HA, Mantovani A, Marme D. Migration of human monocytes in response to vascular endothelial growth factor (VEGF) is mediated via the VEGF receptor flt-1. Blood 1996; 87: 333643.
  • 35
    Murakami M, Iwai S, Hiratsuka S, Yamauchi M, Nakamura K, Iwakura Y, et al. Signaling of vascular endothelial growth factor receptor-1 tyrosine kinase promotes rheumatoid arthritis through activation of monocytes/macrophages. Blood 2006; 108: 184956.
  • 36
    Kuryliszyn-Moskal A, Klimiuk PA, Sierakowski S. Soluble adhesion molecules (sVCAM-1, sE-selectin), vascular endothelial growth factor (VEGF) and endothelin-1 in patients with systemic sclerosis: relationship to organ systemic involvement. Clin Rheumatol 2005; 24: 1116.
  • 37
    Konttinen YT, Mackiewicz Z, Ruuttila P, Ceponis A, Sukura A, Povilenaite D, et al. Vascular damage and lack of angiogenesis in systemic sclerosis skin. Clin Rheumatol 2003; 22: 196202.
  • 38
    Distler O, Del Rosso A, Giacomelli R, Cipriani P, Conforti ML, Guiducci S, et al. Angiogenic and angiostatic factors in systemic sclerosis: increased levels of vascular endothelial growth factor are a feature of the earliest disease stages and are associated with the absence of fingertip ulcers. Arthritis Res 2002; 4: R11.
  • 39
    Guiducci S, Distler O, Distler JH, Matucci-Cerinic M. Mechanisms of vascular damage in SSc—implications for vascular treatment strategies. Rheumatology (Oxford) 2008; 47: v1820.
  • 40
    Ferrara N. The role of VEGF in the regulation of physiological and pathological angiogenesis. EXS 2005; 94: 20931.
  • 41
    Jinnin M, Makino T, Kajihara I, Honda N, Makino K, Ogata A, et al. Serum levels of soluble vascular endothelial growth factor receptor-2 in patients with systemic sclerosis. Br J Dermatol 2010; 162: 7518.
  • 42
    Distler O, Distler JH, Scheid A, Acker T, Hirth A, Rethage J, et al. Uncontrolled expression of vascular endothelial growth factor and its receptors leads to insufficient skin angiogenesis in patients with systemic sclerosis. Circ Res 2004; 95: 10916.
  • 43
    Davies CD, Jeziorska M, Freemont AJ, Herrick AL. The differential expression of VEGF, VEGFR-2, and GLUT-1 proteins in disease subtypes of systemic sclerosis. Hum Pathol 2006; 37: 1907.
  • 44
    Dor Y, Djonov V, Abramovitch R, Itin A, Fishman GI, Carmeliet P, et al. Conditional switching of VEGF provides new insights into adult neovascularization and pro-angiogenic therapy. EMBO J 2002; 21: 193947.