CCN2 (connective tissue growth factor) promotes fibroblast adhesion to fibronectin

Abstract

In vivo, CCN2 (connective tissue growth factor) promotes angiogenesis, osteogenesis, tissue repair, and fibrosis, through largely unknown mechanisms. In vitro, CCN2 promotes cell adhesion in a variety of systems via integrins and heparin sulfate proteoglycans (HSPGs). However, the physiological relevance of CCN2-mediated cell adhesion is unknown. Here, we find that HSPGs and the mitogen-activated protein kinase kinase/extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase cascade are required for adult human dermal fibroblasts to adhere to CCN2. Endogenous CCN2 directly binds fibronectin and the fibronectin receptors integrins alpha4 beta1 and alpha5 and syndecan 4. Using Ccn2-/- mouse embryonic fibroblasts, we show that loss of endogenous CCN2 results in impaired spreading on fibronectin, delayed alpha-smooth muscle actin stress fiber formation, and reduced ERK and focal adhesion kinase phosphorylation. These results suggest that a physiological role of CCN2 is to potentiate the ability of fibroblasts to spread on fibronectin, which may be important in modulating fibroblast adhesion to the provisional matrix during tissue development and wound healing. These results are consistent with the notion that a principal function of CCN2 is to modulate receptor/ligand interactions in vivo.

Figure 1.

Adhesion of normal adult human dermal fibroblasts to CCN2 is HSPG dependent. (A) Normal adult human dermal fibroblasts (NADF) show little appreciable CCN2 expression. Primary normal adult dermal fibroblasts were cultured in DMEM/0.5% FBS for 24 h, detached with EDTA, and allowed to adhere to CCN2 that had been coated on a tissue culture plate. After 40 min, attached cells were lysed, and the resultant protein extracts were subjected to Western blot analysis with an anti-CCN2 antibody. NADF, normal adult dermal fibroblasts (20 μg/lane). rCCN2, 10 ng of recombinant CCN2 (10 ng/lane). (B) CCN2 promotes fibroblast adhesion in a dose-dependent manner. Wells of a 96-well plate were incubated overnight with BSA (10 μg/ml) or different concentrations of CCN2, as indicated. After a blocking step, primary adult dermal fibroblasts were allowed to adhere to CCN2 for 40 min, and cell adhesion was assessed as described in Materials and Methods (6 wells/data point; average [closed black dot] ± SEM [error bar] is shown). The values obtained from the BSA control experiment, which were not significantly above background, were subtracted from the experimental values. (C) Adhesion of fibroblasts to CCN2 is dependent on cell surface heparin-containing proteoglycans. Adhesion assays were repeated, but cells were incubated with or without heparinase (HEPase) or chondrotinase ABC (CHaseABC), or soluble chondrotinase A, B, or C, keratin sulfate, or heparin (50 μg/ml; 6 wells/data point; average [gray box] ± SEM [error bar] are shown).

Figure 2.

Fibronectin promotes adhesion of normal adult dermal fibroblasts in a dose-dependent manner. Wells of a 96-well plate were incubated overnight with different concentrations of fibronectin, as indicated. After a blocking step, primary adult dermal fibroblasts were allowed to adhere to fibronectin for 40 min, and cell adhesion was assessed as described in Materials and Methods (3 wells/data point; average [open diamond] ± SEM [error bar] are shown). The values obtained from a control adhesion experiment with BSA, which were not statistically significantly above background, were subtracted from the experimental values.

Figure 3.

Comparison of adult dermal fibroblast adhesion to CCN2 and fibronectin. Staining of cells adhered to CCN2 and fibronectin with phalloidin (ACTIN), anti-phospho-FAK (p-FAK), and anti-phospho-ERK (p-ERK) antibodies are shown. Fibroblasts were allowed to adhere to substrates, as described in Materials and Methods, for (A) 40 or (B) 120 min before fixation of cells in paraformaldehyde. (A) Fibroblasts adhered to fibronectin for 40 min showed extension of filopodia (arrow with f, actin), actin stress fibers, and punctuate phospho-FAK (arrow, p-FAK, FN) and phospho-ERK staining, consistent with the induction of focal adhesions. Conversely, fibroblasts adhered to CCN2 did not show extended filopodia, displayed disorganized actin staining, and did not show clustered, localized phospho-FAK staining (400×). Fibroblasts adhered to CCN2 and fibronectin showed punctuate foci of phospho-ERK staining (arrow, CCN2, FN, p-ERK), consistent with an induction of cellular signaling in fibroblasts adhered to both substrates. Conversely, cells adhered to BSA showed no phospho-ERK induction. Nuclei were detected by DAPI staining (blue).

Figure 4.

Adhesion of adult dermal fibroblasts to fibronectin results in ERK phosphorylation. Normal adult dermal fibroblasts were cultured and detached as described in Materials and Methods and in the legend for Figure 1. Fibroblasts were allowed to adhere to six-well plates that had been coated with CCN2 (10 μg/ml) as described above. Protein extracts were prepared after adhesion at the times indicated. The resultant protein extracts (20 μg/lane) were subjected to SDS-PAGE and Western blot analysis with anti-ERK and anti-phospho-ERK antibodies. Densitometry was performed on the resultant autoradiograms, and ratio of phosphorylated/non-phosphorylated ERK is shown.

Figure 5.

Adhesion of adult dermal fibroblasts to CCN2, but not fibronectin, requires the MEK/ERK pathway. Fibroblasts were allowed to adhere for 40 min onto fibronectin (FN), CCN2, or BSA. Before performing the adhesion assay, suspended fibroblasts were preincubated, as indicated, in the presence or absence of soluble heparin (50 μg/ml) or the MEK1/2 inhibitor U0126 (10 μM). After a washing step, adherent cells were quantified as described in Materials and Methods. Assays were performed in 96-well plates, each condition, in triplicate, two independent times (N = 6). Average optical density values ± SEM are shown.

Figure 6.

CCN2 binds fibronectin and the fibronectin receptors integrin α4, integrin α5, integrin β1, and syndecan-4. (A) ¹²⁵I-CCN2 binds fibronectin. Wells of a 96 well plate were coated overnight at 4°C with 50 μg/ml fibronectin or 1% BSA in PBS or PBS and blocked for 1 h at room temperature with 1% BSA in PBS. ¹²⁵I-CCN2 (500,000 cpm/well) was added for 1 h, at room temperature. Wells were washed three times with 1% BSA in PBS and bound ¹²⁵I-CCN2 was measured with a scintillation counter. Values represent average ± SEM (N = 6). (B) Protein extracts (20 μg) from Ccn2+/+ and Ccn2-/- mouse embryonic fibroblasts (MEF) were subjected to Western blot analysis with an anti-CCN2 antibody as described in Materials and Methods. (C) Membrane preparations from Ccn2+/+ and Ccn2-/- mouse embryonic fibroblasts were immunoprecipated using protein A-Sepharose with rabbit anti-CCN2 antibody or rabbit IgG. Immunoprecipitates were subjected to Western blot analysis with anti-CCN2, anti-syndecan 4, anti-fibronectin, anti-endoglin, and anti-integrin α4, α5, and β1 antibodies, as indicated.

Figure 7.

Comparison of the adhesive abilities of Ccn2+/+ and Ccn2-/- mouse embryonic fibroblasts on CCN2. Ccn2+/+ and Ccn2-/- mouse embryonic fibroblasts were cultured in DMEM, 0.5% FBS for 24 h; detached in EDTA; and allowed to adhere to CCN2 that had been adsorbed to 96-well plates as described in Materials and Methods and the legend to Figure 1. Cell adhesion after 40 min was determined (N = 6; average ± SEM is shown) as described in Materials and Methods and in the legend to Figure 1B.

Figure 8.

Ccn2-/- mouse embryonic fibroblasts show impaired spreading on fibronectin. Ccn2+/+ and Ccn2-/- mouse embryonic fibroblasts (MEFs) were allowed to adhere to fibronectin for 40 min as described in Materials and Methods. In addition, Ccn2-/- fibroblasts were preincubated with rCCN2 for 1 h before adhesion to fibronectin. Cells were fixed in paraformaldehyde and subjected to indirect immunofluorescence analysis with anti-α-SMA antibody and detected with an appropriate FITC-conjugated antibody as described in Materials and Methods and in the legend to Figure 1C. Forty minutes postadhesion, Ccn2+/+ MEFs adhered to fibronectin show extension of filopodia and α-SMA localization to the cell periphery (arrow with f, Ccn2+/+) and display a few stress fibers (arrow with a, Ccn2+/+). Conversely, Ccn2-/- MEFs do not display filopodia, α-SMA localization to the cell periphery, or stress fibers (Ccn2-/-). Ccn2-/- fibroblasts that were preincubated with rCCN2 for 1 h before adhesion to fibronectin showed extension of filopodia and α-SMA localization to the cell periphery (arrow with s, Ccn2-/-) and display a few stress fibers (arrow with a, Ccn2-/- + rCCN2). Nuclei were detected by DAPI staining (blue).

Figure 9.

Ccn2-/- mouse embryonic fibroblasts show delayed induction of α-SMA stress fibers. Ccn2+/+ and Ccn2-/- mouse embryonic fibroblasts (MEFs) were allowed to adhere to fibronectin for 60 and 120 min (min on FN) as described in Materials and Methods. Sixty minutes postadhesion, Ccn2+/+ MEFs adhered to fibronectin show stress fiber formation (Ccn+/+, 60). Conversely, 60 minutes postadhesion, Ccn2-/- MEFs adhered to fibronectin show extension of filopodia and α-SMA localization to the cell periphery not display abundant stress fiber formation (arrow, Ccn2-/-, 60). Two hours postadhesion, Ccn2+/+ and Ccn2+/+ MEFs show similar phenotypes, as visualized by extension of filopodia and formation of α-SMA stress fiber networks. (Ccn2+/+, and Ccn2-/-, 120). Nuclei were detected by DAPI staining (blue).

Figure 10.

CCN2-/- mouse embryonic fibroblasts show impaired phosphorylation of FAK and ERK. Ccn2+/+ and Ccn2-/- mouse embryonic fibroblasts were allowed to adhere to fibronectin (FN), as described in Materials and Methods, for up to 120 min (min on FN). (A and B) Indirect immunofluorescence analysis. Cells were fixed as described in Materials and Methods and in the legend to Figure 7 and were stained with anti-phospho-ERK (A) and anti-phospho-FAK (B) antibodies. Whereas Ccn2+/+ MEFs showed abundant phospho-ERK (arrow, A, Ccn2+/+ 40) and phospho-FAK (arrow, Ccn2+/+ 120) staining, Ccn2-/- MEFs showed markedly reduced phospho-ERK and phospho-FAK staining (A and B, Ccn2-/-). Nuclei were detected by DAPI staining (blue). (C) Western blot analysis. Experiments were performed as described above, except cell layers were harvested at different time points postadhesion (min on FN), and equal amounts of protein were subjected to Western blot analysis with anti-phospho-FAK and anti-FAK antibodies or anti-phospho-ERK and ERK antibodies, as indicated. Densitomery values showing ratio between phosphorylated epitopes and total protein, are indicated.

Figure 11.

Soluble heparin and U0126 impede the ability of Ccn2+/+ mouse embryonic fibroblasts to spread on fibronectin. Experiments were performed essentially as described in the legend to Figure 9. Ccn2+/+ mouse embryonic fibroblasts were allowed to adhere to fibronectin for up to 120 min (min on FN) as described in Materials and Methods, in the presence of soluble heparin (50 μg/ml) or U0126 (10 μM). Cells were fixed after 45 min and subjected to indirect immunofluorescence analysis with an anti-α-SMA antibody and an appropriate FITC-conjugated secondary antibody. Nuclei were detected by DAPI staining (blue).

Figure 12.

Schematic diagram summarizing CCN2-promoted fibroblast adhesion. CCN2 supports cell adhesion by binding to fibronectin, cell surface proteoglycans, and integrins; potentiates the phosphorylation of FAK and ERK; and thus enhances focal adhesion formation, cell spreading, and myofibroblast formation.