CTGF Mediates TGF-β–Induced Fibronectin Matrix Deposition... : Journal of the American Society of Nephrology (original) (raw)
Diabetic nephropathy (DN) is characterized by excessive deposition of extracellular matrix in the glomerular mesangium (1). Although hyperglycemia in diabetes is directly related to the severity of the disease (2), it is not entirely clear how high glucose concentration exerts its effects on mesangial cells. However many of the effects of high glucose are thought to be mediated through its induction of the growth factors, transforming growth factor-β (TGF-β) (3) and connective tissue growth factor (CTGF) (4–6). The level of both factors increase in the glomerulus in DN (3,6).
Connective tissue growth factor (CTGF) is a 36- to 38-kD cysteine-rich secreted protein (7). CTGF is encoded by an inducible immediate early gene and is one of six distinct members of the CCN family (8). The CTGF gene contains a novel TGF-β response element in its promoter region (9) and is thought to be a downstream mediator of the some of the effects of TGF-β (10). Several studies support this view (11–14), but CTGF is also induced by a variety of other factors such as thrombin and VEGF (15,16).
CTGF is thought to drive glomerulosclerosis and tubulointerstitial fibrosis in renal diseases (4). CTGF levels are elevated in different renal fibrotic disorders where excessive extracellular matrix formation is observed (11,17–19) and the growth factor induces matrix protein synthesis in mesangial cells in vitro (5,6,11,19).
One of the key extracellular matrix proteins secreted by human mesangial cells (HMC) is fibronectin. Levels of fibronectin are elevated in DN (20), and fibronectin secretion by HMC is increased when cultured under high glucose conditions (21–23). Fibronectin is assembled into an insoluble matrix outside the cell, and distinct regions of the molecule are required for matrix assembly (24). One of the most important regions in the fibronectin monomer for this is the cell-binding domain that incorporates the ninth and tenth type III repeats, which contain the synergy sites and RGD integrin-binding consensus respectively (24). The predominant integrin receptor that binds fibronectin and is intimately involved in matrix assembly is the α5β1 heterodimer. The importance of this receptor for fibronectin matrix assembly has been shown previously using monoclonal antibodies to block its function (25,26).
Levels of the α5β1 fibronectin receptor correlate directly with the degree of fibronectin matrix accumulation in renal disease (27,28), and it is known that the profibrotic factor TGF-β upregulates receptor expression in different cell types (29,30), including rat mesangial cells and glomeruli (27,31). Little is known about the effect of CTGF on α5β1 levels, but it has been shown to upregulate α5 integrin transcripts in NRK fibroblasts (14). However, in view of the increased levels of CTGF in the diabetic glomerulus and its stimulatory effect on fibronectin synthesis by mesangial cells (6), we hypothesized that the factor may play a key role in promoting excessive deposition of fibronectin around mesangial cells by upregulating the expression of their fibronectin receptors. We report experiments below showing that this is the case and that similar effects induced by TGF-β1 are in fact mediated by CTGF.
Materials and Methods
Materials
HMC were obtained from Biowhittaker UK Ltd, Wokingham, UK. Antibodies to total fibronectin and β-actin were from Sigma (Poole, UK). The antibody to cellular (EDA+) fibronectin (MAS 521) was from Harlan Sera-Lab, Loughborough, UK. The BIIG2 hybridoma developed by Dr. Caroline H. Damsky was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the University of Iowa, Department of Biological Sciences, Iowa City, IA. The 12G10 β1 antibody was a kind gift from Professor Martin Humphries, of the Wellcome Trust Center for Cell-Matrix Research, University of Manchester, UK. Recombinant CTGF, TGF-β, and phosphothioate antisense CTGF and control oligonucleotides were obtained as described previously (6). Phosphothioate antisense (TGG GCA GAC GAA CG) and control oligonucleotides (ACC GAC CGA CGT GT) directed to CTGF were designed and manufactured by Biognostik GmbH (Go[Combining Diaeresis]ttingen, Germany), who own the intellectual property rights to the sequences (6). Monoclonal antibodies MAB1969 and MAB 1999 were from Chemicon International, Ltd, Harrow, UK. All other reagents were purchased from Sigma except where stated otherwise.
Cell Culture Conditions
HMC were routinely cultured and maintained as described previously (32). Cells between passage 6 to 10 were used for experiments.
Measurement of Fibronectin Deposited as Insoluble Matrix by HMC in Culture
HMC were maintained in RPMI 1640 medium containing 4 mM D-glucose and 10% fetal bovine serum (FBS), previously depleted of fibronectin by passing FBS over gelatin-Sepharose (Amersham Pharmacia Biotech, Amersham, UK) (32). After serum deprivation for 24 h post log-phase cultures were treated with serum-free medium containing 4 mM D-glucose, L-[4,5-3H]Leucine (8.3 μCi · ml−1; Amersham Pharmacia Biotech), and supplemented with either rCTGF (80 ng/ml) or TGF-β1 (5 ng/ml) (2 ml/well). The cultures were incubated at 37°C for 48 h, after which the media were aspirated and stored at −20°C. The cell layers were extracted sequentially with detergent solutions to obtain a deoxycholate-insoluble matrix, as described in detail previously (32). 3H-Fibronectin was immunoprecipitated from all fractions before scintillation counting in a Beckman LS6000SC counter. Protein concentration in the cell-associated fraction was determined using the BCA-200 Protein Assay Kit (Pierce, Rockford, IL).
Western Blotting
Cell layers were washed three times with PBS and then solubilized in RIPA buffer (1% vol/vol NP-40; 0.5% wt/vol sodium deoxycholate; 0.1% wt/vol SDS). Protein samples were separated by reducing SDS-PAGE on 7.5% gels and then transferred to Immobilon-P PVDF membranes (Millipore, Bedford, UK). Nonspecific sites were blocked before incubation of the membrane with primary antibody (anti-fibronectin [1/2000]; anti-EDA+ fibronectin [1/400]; anti-β-actin [1/3000]) overnight at 4°C. Incubation with an appropriate secondary antibody for 1 h at room temperature was followed by band visualization with enhanced chemiluminescence (ECL, Amersham Pharmacia Biotech). SDS-PAGE molecular weight markers (Rainbow colored protein molecular weight markers, Amersham Pharmacia Biotech) were run to calibrate the gel.
Use of CTGF Antisense Oligonucleotides
CTGF phosphothioate antisense oligonucleotide or a CG matched control was added directly to cultures at a final concentration of 2 μM (according to manufacturer’s recommendation). Previous experiments have shown that this concentration leads to an appreciable decline in CTGF mRNA levels from HMC in culture (6).
FACS Analysis of Cell Surface Levels of α5β1 Integrin Receptor
HMC were seeded onto cell culture dishes (Corning Ltd, Corning, NY) in media containing 10% FBS previously depleted of fibronectin. Post log-phase cultures were serum-starved for 24 h before treatment with medium containing 4 mM D-glucose alone or media supplemented with either rCTGF (80 ng/ml), TGF-β (5 ng/ml), or rCTGF and TGF-β for different times (30 min to 72 h). Cells were then transferred to the wells of a 96-well conical-bottomed microtiter plate (Nunc International, Roskilde, Denmark). Cells were pelleted, resuspended in 100 μl of FACS buffer (HBSS containing 50 U · ml−1 penicillin, 50 μg · ml−1 streptomycin, 300 μg · ml−1 glutamine, and 2.5% FBS) containing a saturating concentration of mouse anti-human VLA-5 (α5β1) integrin monoclonal antibody (10 μg/ml; MAB1999 Chemicon International, Inc, Temecula, CA), and incubated on ice for 30 min. After washing, the cells were incubated with goat anti-mouse IgG FITC conjugate (1/50; Dako, Glostrup, Denmark) for 30 min on ice in the dark. Washed cells were then resuspended in a total volume of 500 μl of PBS containing 1% paraformaldehyde (PFA) before analysis on a FACScan machine (Becton Dickinson, Franklin Lakes, NJ). Cells were also incubated with an isotype-matched control antibody, mouse IgG2b,κ, (Sigma), or with secondary antibody alone. Data were analyzed using the CellQuest program. Relative fluorescence intensities were calculated by dividing the mean fluorescence intensity of test antibody by the mean fluorescence intensity of the appropriate isotype-matched control antibody or secondary antibody alone.
Additional experiments were undertaken, as above, for cells stimulated with either rCTGF or TGF-β1 for 48 h and probed with the β1 integrin antibody 12G10 (33), using a saturating concentration of 10 μg/ml.
RT-PCR Analyses
Total RNA was isolated using RNAzol B (AMS Biotechnology, Abingdon, UK) (34). Two micrograms of each RNA sample was converted into cDNA with Superscript II RNase H− reverse transcriptase and random hexamer primers (Life Technologies BRL, Paisley, Scotland, UK). Equal amounts of cDNA were subsequently amplified by PCR. Control amplifications were performed with GAPDH to confirm the use of equal amounts of RNA. The amount of reverse transcription reaction used for the amplification (0.5 μl) was selected as being nonsaturating for the PCR product of the genes under investigation after 27 cycles of amplification. The sequences of the primers used for the amplification of integrin alpha5 mRNA were: (sense) 5′-TGCATCAACCTTAGCTTCTGCCT -3′ and (antisense) 5′-ACCAGCAGGCGGCTCTGGTTCAC -3′ (35). The sequences of the primers used for the amplification of integrin beta1 mRNA were: (sense) 5′-GGAAAACGGCAAATTGTCAGAAGG-3′ and (antisense) 5′-TGGACCAGTGGGACACTCTGGATT-3′ (35). The sequences of the primers used for the amplification of GAPDH mRNA were: (sense) 5′-ACCACAGTCCATGCCATCAC-3′ and (antisense) 5′-TCCACCACCCTGTTGCTGTA-3′ (32). Amplification products were electrophoresed through 1.5% (wt/vol) agarose gels and visualized by ethidium bromide staining. Bands were scanned and quantitated by densitometry using NIH Image software.
Cell Adhesion Assays
Twenty-four–well flat-bottom plates (Corning Ltd) were coated overnight at 4°C with 200 μl of a 10 μg/ml solution of fibronectin (Sigma, Poole, UK). Wells were washed with PBS and blocked with 500 μl of 10 mg/ml denatured bovine serum albumin (BSA) for 1 h at 37°C. Cell adhesion assays were performed essentially as described by Kagami et al. (31). Briefly, growth-arrested HMC were treated with media containing 4 mM D-glucose and either CTGF (80 ng/ml), TGF-β (5 ng/ml), or CTGF and TGF-β for 48 h. Control cultures that were maintained in serum-free media containing 4 mM D-glucose were also included. Cells were detached with trypsin, washed, and pelleted. Cells were resuspended in RPMI containing 1 mM MnCl2 and 1 mM CaCl2 at a concentration 1 × 105 cells/ml. Cells were plated onto fibronectin-coated wells (300 μl of cell suspension/well) and placed at 37°C for 50 min. Adherent cells were fixed and stained with 0.1% crystal violet. Finally, extensively washed cells were solubilized with 1% Triton X-100, and the absorbance of the solubilized cell extracts was measured at 595 nm.
For antibody blocking experiments the α5 integrin antibody BIIG2 (36) was purified from BIIG2 hybridoma cultures to ensure that a sufficient amount of antibody was available for the adhesion experiments. Cell adhesion assays were carried out as described above, the only difference being that cell suspensions were incubated with 20 μg/ml purified BIIG2 antibody (30 min, room temperature) before seeding onto fibronectin-coated tissue-culture plastic. Control suspensions, where the cells were incubated under the same conditions in the absence of BIIG2 antibody were carried out.
Immunofluorescence
HMC (passage 6 to 10) were seeded onto coverslips in medium containing 4 mM D-glucose and FBS depleted of fibronectin. Once the monolayer had reached approximately 50 to 60% confluence, the cells were serum-starved overnight and were subsequently treated with serum-free media supplemented with 4 mM D-glucose and 5 ng/ml TGF-β1 or 80 ng/ml rCTGF. After 48 h, the cells were fixed and permeabilized. Nonspecific sites were then blocked. The primary antibodies (anti-human α5β1 [1/50] MAB1969, Chemicon; anti-human paxillin [1/100], Sigma) were incubated with the cells overnight at 4°C in a humidified chamber. Washed cells were incubated with the appropriate secondary antibody-conjugate (1/200) for 1 h at room temperature. In certain cases, incubation with the secondary antibody was followed by incubation with phalloidin-TRITC (50 μg · ml−1) for 1 h at room temperature. The coverslips were washed and mounted with Vectashield mounting medium (Vector Laboratories Inc, Burlingame, VT). The cells were inspected in a fluorescence microscope (Olympus Provis), and photographs were taken using 1600ASA Kodak color film. Control experiments, in which the primary antiserum was omitted, were also carried out.
For α5β1 integrin-blocking experiments, adherent cells were treated with either 10 μg/ml or 20 μg/ml α5β1 blocking antibody (MAB1969, Chemicon) in media containing 10% FBS depleted of fibronectin. After 48 h, the cells were processed for immunofluorescence as above, although they were not permeablized and were probed with an antibody to human fibronectin (1/500, Sigma).
Results
The Deposition of Deoxycholate-Insoluble Fibronectin Matrix by HMC in Culture Is Stimulated by TGF-β and CTGF
Post log-phase human mesangial cell cultures were treated with either 80 ng/ml CTGF or 5 ng/ml TGF-β for 48 h to examine their effects on fibronectin synthesis. 3H-Leucine was included in the culture media to label all newly synthesized protein. After treatment, the deoxycholate-insoluble extracellular matrix fraction, which was solubilized with 1% SDS and 5 mM DTT at 37°C for 30 min, was isolated from the cell layer. Immunoprecipitation of 3H-FN from the medium and all fractions of the cell layer was followed by scintillation counting to determine the effect of the two growth factors on fibronectin synthesis and insoluble fibronectin-matrix deposition. The results from two separate experiments are shown in Table 1. The addition of CTGF or TGF-β to HMC cultures appreciably increased total fibronectin synthesis over basal values, during the 48-h labeling period. There was no difference between the total amount of fibronectin synthesis in cultures treated with CTGF or TGF-β. Measurement of 3H-FN in the deoxycholate-insoluble fraction indicated that treatment with CTGF or TGF-β led to a greater proportion of total counts being deposited in this fraction than in control cultures. The counts in both the CTGF- and TGF-β-treated cultures were ≥60% higher than those observed in control cultures.
The effect of connective tissue growth factor (CTGF) and transforming growth factor-β (TGF-β) on fibronectin synthesis and deoxycholate-insoluble fibronectin matrix deposition by human mesangial cellsa
Western blot analyses confirmed that fibronectin levels were increased in cell layer extracts of cultures treated with either CTGF or TGF-β for 48 h compared with control cultures (Figure 1, A and B). The EDA+ splice variant of the molecule was also increased (Figure 1, A and B). Taken together, these data indicate that, in addition to upregulating total fibronectin synthesis by mesangial cells in culture, both CTGF and TGF-β specifically promote fibronectin matrix deposition by these cells.
Effect of connective tissue growth factor (CTGF) and transforming growth factor-β (TGF-β) on fibronectin present in mesangial cell layer extracts. (A) Treated cells (48 h) were harvested in RIPA buffer and protein samples were subjected to SDS-PAGE (7.5% acrylamide) and Western blotting with antibodies to (i) total fibronectin, (ii) EDA+ splice variant fibronectin, and (iii) beta-actin. Lane 1, 4 mM glucose control cultures; lane 2, 4 mM glucose + CTGF (80 ng/ml); lane 3, 4 mM glucose + TGF-β (5 ng/ml). Representative blots are shown. (B) The protein bands were quantified with a scanning densitometer. The results shown are the ratio of the integrated absorbance of the fibronectin or EDA+ fibronectin bands and the corresponding β-actin band and are the means ± SEM for three separate experiments, each with duplicate cultures (total n = 6).
TGF-β and CTGF Increase the Cell-Surface Expression of the Fibronectin Receptor, α5β1 Integrin, by HMC in Culture
The α5β1 integrin receptor is essential for fibronectin matrix assembly (24) and is expressed by many cell types, including mesangial cells (37). TGF-β and CTGF were both shown to increase fibronectin matrix deposition by mesangial cells; we therefore examined whether this was related to differences in the level of cell-surface α5β1 integrin receptor. Examination of the distribution of the α5β1 integrin receptor in HMC by immunofluorescence revealed that the receptor was widely distributed on the cell in a characteristic fibrillar adhesion pattern, which appeared to be concentrated toward the center of the cell (Figure 2A). No differences in the amount or intensity of staining were detected using this technique between control cells and cells treated with either CTGF or TGF-β for up to 48 h (data not shown). The distribution of the α5β1 integrin receptor was compared with that observed of paxillin, a component of focal adhesions (Figure 2B). As expected, paxillin staining was observed at the ends of the actin filaments, unlike α5β1 integrin, the distribution of which would be expected to be closely associated with fibronectin matrix assembly sites in cells actively depositing an insoluble fibronectin matrix (38).
The distribution of the α5β1 integrin receptor in human mesangial cells (HMC). (A) The distribution of the α5β1 integrin receptor in HMC was analyzed by immunofluorescence as described in the Materials and Methods. It was compared to the distribution of (B) paxillin in the same cell-type, which is localized at the end of the phalloidin stained actin filaments at the sites of focal adhesions. A representative result is shown. Two further experiments gave similar results. Magnification ×100.
FACS analysis was used to quantify cell-surface α5β1 integrin levels. These were high on HMC maintained in 4 mM glucose medium. Moreover, exposure to either TGF-β or CTGF for periods of up to 8 h had no effect on the level of α5β1 expressed at the cell surface (Figure 3A). However, longer treatments with TGF-β or CTGF led to an increase in cell-surface α5β1, with TGF-β having the greatest effect (Figure 3A). Levels of α5β1 plateaued after 48 h treatment with either growth factor, with TGF-β inducing approximately a 65% increase and CTGF approximately a 25% increase over basal levels. These levels were maintained after treatment for 72 h. A typical FACS profile for cells stimulated for 48 h is shown in Figure 3B. Control experiments showed that neither an isotype-matched control antibody, mouse IgG2b,κ, nor the FITC-labeled secondary antibody bound HMC to any significant degree (Figure 3B). To determine whether changes in cell-surface levels of α5β1 were associated with changes in the subunit mRNA levels, total RNA was extracted from cells treated with either TGF-β or CTGF for 48 h. Semiquantitative RT-PCR indicated that CTGF and TGF-β treatment brought about significant increases in the level of mRNA for each subunit, when compared with control cultures (Figure 3, C and D). The extent of the increases was similar for both growth factors.
Effect of CTGF and TGF-β on α5β1 integrin expression by HMC. (A) Cell cultures were treated with either CTGF (80 ng/ml; filled triangles) or TGF-β (5 ng/ml; filled boxes). At the time points indicated, cells were labeled with an antibody to the α5β1 integrin receptor, followed by a FITC-labeled secondary antibody, prior to analysis by FACS, as described in the Materials and Methods. Triplicate cultures were treated in all experiments, which were repeated three times. The results shown are the mean ± SD for all experiments (total n = 9). (B) Representative FACS profile of cells treated with either CTGF (dashed line) or TGF-β (dotted line) for 48 h, as compared with the profile for control cultures (thick black line). The thin black line shows the profile observed for an isotype-matched control. (C) Effect of CTGF and TGF-β on the level of α5 and β1 mRNA levels in HMC. Cells were treated either with control medium (lane 1), CTGF (lane 2), or TGF-β (lane 3) for 48 h, after which total RNA was extracted and 2 μg reverse-transcribed. The product was amplified by PCR using primers for human (a) α5, (b) β1, and (c) GAPDH as a control. The cDNA products were analyzed on a 1.5% agarose gel and visualized with ethidium bromide. Two other experiments gave the same result. (D) The cDNA bands were quantified with a scanning densitometer. The results shown are the ratio of the integrated absorbance of the α5 or β1 bands and the corresponding GAPDH band and are the means ± SEM for three separate RT-PCR analyses. The Mann-Whitney U test was used to assess statistical significance. * P ≤ 0.05 (total n = 9).
The TGF-β-Stimulated Increase in Mesangial Cell-Surface α5β1 Integrin Levels Is Partially Mediated by CTGF
We have previously shown that in HMC in culture TGF-β upregulates CTGF, and that the increase in fibronectin synthesis brought about by the addition of TGF-β is mediated through the induction of CTGF (6). To determine whether the ability of TGF-β to stimulate an increase in cell-surface α5β1 integrin in HMC was dependent on CTGF activity, cells were stimulated with TGF-β (5 ng/ml) for 48 h in the presence of 2 μM phosphothioate antisense CTGF oligonucleotide. This oligonucleotide has been used previously and has been shown to downregulate CTGF mRNA levels in HMC in culture (6). Cell-surface α5β1 integrin levels were then determined by FACS. Figure 4 shows that the CTGF antisense significantly reduced TGF-β–stimulated cell-surface α5β1 integrin levels. However, levels of α5β1 remained approximately 25% higher in the presence of antisense CTGF than in the control cultures. The control antisense oligonucleotide did not have any significant effect on TGF-β-stimulated α5β1 integrin levels. Interestingly, despite the marked inhibitory effect of antisense CTGF on TGF-β-stimulated α5β1 levels, exogenous rCTGF did not mimic exactly the effect of exogenous TGF-β. It is possible that the exogenous rCTGF is not as active as CTGF produced by the cultures, and it may be more restricted in access to receptors than the latter through binding to matrix heparan sulfate proteoglycans. The apparent increase in α5β1 levels in cultures treated with TGF-β and CTGF together over those treated with TGF-β alone was not statistically significant.
The effect of CTGF antisense oligonucleotide on the TGF-β–induced stimulation of cell surface α5β1 expression by HMC. Cells were exposed to different treatments for 48 h, after which cell-surface α5β1 integrin receptor levels were analyzed by FACS. The results are compared with the relative fluorescent intensity in untreated control cultures and represent the means ± SD for three separate experiments with quadruplicate cultures in each experiment for each condition. The Mann-Whitney U test was used to assess statistical significance. * P ≤ 0.05 (total n = 12).
CTGF and TGF-β Stimulate Mesangial Cell Adhesion to Fibronectin
To assess whether the growth-factor–stimulated increase in cell-surface α5β1 integrin levels affected HMC adhesion to the major α5β1 substrate, cells were seeded onto fibronectin-coated plates after stimulation with either CTGF or TGF-β for 48 h. After 50 min, adherent cells were stained with crystal violet. The results are shown in Figure 5. Pretreatment with either CTGF or TGF-β significantly increased the number of cells that adhered to the fibronectin-coated wells over this time, with there being no significant difference between the effect of either factor. Antisense CTGF had the effect of significantly reducing cell adhesion to fibronectin in TGF-β stimulated cells, although once more levels were not reduced to those in control cultures.
Effect of CTGF and TGF-β on mesangial cell adhesion to fibronectin. HMC were treated with either CTGF (80 ng/ml), TGF-β (5 ng/ml), both CTGF and TGF-β, or TGF-β and antisense CTGF for 48 h before seeding (3 × 104 cells) onto fibronectin-coated wells (10 μg/ml) for 50 min. Adherent cells were stained with crystal violet, and the level of staining (Abs 595 nm) was compared with untreated cells. Results represent the mean ± SD for a representative experiment, with four cell-cultures for each condition (n = 4). Each culture was seeded onto eight wells for the adhesion assay. A value of 1 was assigned to untreated cells. The level of staining in BSA-coated control wells was subtracted from all results before analysis. The Mann-Whitney U test was used to assess statistical significance. * P ≤ 0.05. The experiment was repeated three times.
Mesangial cells express an array of integrin receptors that are able to bind fibronectin (37). Thus, to assess the importance of α5β1 integrin in mediating mesangial cell adhesion to fibronectin, we treated cells with BIIG2 (20 μg/ml), an antibody to the α5 integrin receptor that blocks cell attachment to fibronectin (36), before seeding onto fibronectin. Table 2 shows that incubation with BIIG2 significantly reduced the number of adherent cells in all cell cultures to about 30% of that seen in control cultures. Control cultures were defined as those that had been previously subjected to the same treatments, but not exposed to BIIG2. The level of inhibition did not depend on the treatment to which the cells had been previously exposed. Taken together, these data indicate that both TGF-β and CTGF increase the adhesive properties of mesangial cells to fibronectin, through a mechanism partly dependent on the α5β1 integrin receptor. The TGF-β-induced effect is mediated, at least in part, via CTGF.
The effect of the α5 integrin-blocking antibody BIIG2 on mesangial cell adhesion to fibronectina
TGF-β and CTGF Increase the Number of β1 Integrin Receptors in the Ligand-Occupied State
The 12G10 monoclonal antibody is a β1 integrin monoclonal antibody that reacts preferentially with the ligand-occupied form of β1-integrin containing heterodimers (33). FACS experiments were undertaken to determine whether the level of 12G10 binding to HMC treated with CTGF or TGF-β was altered when compared with untreated control cultures. The results are shown in Figure 6. Treatment of HMC with CTGF (80 ng/ml) or TGF-β (5 ng/ml) for 48 h led to a shift in the FACS profile for both growth factors when compared with the profile for untreated cells (Figure 6A). The effects of TGF-β and CTGF on 12G10 binding were similar. CTGF and TGF-β increased binding of 12G10 to HMC by 19% and 26%, respectively (Figure 6B). These changes were statistically significant and indicated that both CTGF and TGF-β increased the level of ligand-occupied β1 integrin on the mesangial cell surface. Control experiments, where 12G10 was omitted, showed that the secondary antibody did not bind mesangial cells.
The effect of CTGF and TGF-β on the level of ligand-occupied β1 integrin. (A) FACS profile of cells treated with either CTGF (80 ng/ml; dashed line) or TGF-β (5 ng/ml; dotted line) for 48 h compared with the profile for control cultures (thick black line). The thin black line shows the profile observed for a secondary-antibody only control. FACS analysis was performed following detection of cell-surface ligand-occupied β1 integrin with the 12G10 monoclonal antibody, as described in the Materials and Methods. (B) The results are compared with the relative fluorescent intensity observed in untreated controls and represent the average value obtained for four separate cultures ± SD. The Mann-Whitney U test was used to assess statistical significance. * P ≤ 0.05 (n = 4). A representative result is shown. Two further experiments each with quadruplicate cultures for each condition gave similar results.
Blocking Antibodies to the α5β1 Integrin Inhibit Fibronectin Matrix Deposition by HMC In Vitro
A further set of experiments sought to confirm the importance of the α5β1 integrin receptor in fibronectin matrix deposition by mesangial cells in vitro. Newly adherent cells were fed with media supplemented with FBS previously depleted of fibronectin and containing antibody blocking α5β1 integrin (Chemicon, MAB 1969). After 48 h, the fibronectin matrix was visualized by immunofluorescence. This showed that addition of the blocking antibody to adherent cells had no effect on the number of cells that remained attached to the coverslip (data not shown). The fibronectin matrix deposited around the cells (which were non-permeablized for analysis) was visible in the form of characteristic fibrils (Figure 7A). However, the addition of the blocking antibody led to appreciably lower amounts of fibronectin matrix, which also appeared more randomly organized and less fibrillar in structure (Figure 7, B and C). Similar observations were also made with cells that were stimulated with CTGF, in that the presence of α5β1 integrin blocking antibody also brought about a decrease in the amount of fibronectin matrix deposited by the cells (data not shown).
The effect of α5β1 integrin-blocking antibodies on the fibronectin matrix deposited around human mesangial cells in culture. Fibronectin matrix was visualized by immunofluorescence around (A) untreated cells and cells exposed to blocking antibodies (MAB 1969) at concentrations of 10 μg/ml (B) and (C) 20 μg/ml. A representative experiment is shown. Two other experiments gave similar results (n = 3). Magnification × 50.
Discussion
Levels of the pro-fibrotic growth factor TGF-β and its downstream mediator CTGF are increased in the kidney in diabetic nephropathy (3,6,17). Accumulation of extracellular matrix in the glomerular mesangium is one of the most prominent features of the disorder (1). One of the main matrix proteins deposited is fibronectin, which associates to form an insoluble matrix in a well-defined manner (24). The α5β1 integrin receptor is key to fibronectin-matrix formation, providing an important point of contact between the cell and extracellular fibronectin fibrils.
Previous studies have shown that expression of α5β1 integrin parallels mesangial content of both fibronectin and TGF-β protein in renal disease (27,28,39). TGF-β has also been shown to directly upregulate α5β1 expression by cultured rat mesangial cells (31). However, very little is known about the relationship between the levels of CTGF and α5β1 expression. One study showed upregulation of α5 transcripts in NRK fibroblasts (14), and CTGF levels have been shown to correlate very well with α5 levels in the ileum from patients with inflammatory bowel disease, Crohn disease, and ulcerative colitis (40). Using an HMC culture system, we sought to establish whether the ability of TGF-β to stimulate α5β1 expression in these cells is mediated through CTGF.
In addition to upregulating total fibronectin synthesis, our results showed that TGF-β and CTGF preferentially stimulate the deposition of insoluble fibronectin matrix by HMC. Western blot analyses also revealed that synthesis of the EDA+ splice variant of fibronectin is upregulated by these factors. The inclusion of the EDA segment is thought to enhance the cell-adhesive activity of fibronectin by potentiating the interaction of fibronectin with the fibronectin receptor, the α5β1 integrin (41). These observations led us to investigate whether the level of α5β1 integrin is also altered, as this is an important factor for fibronectin matrix synthesis (24–26).
Both CTGF and TGF-β significantly upregulated α5 and β1 at the message level, as determined by RT-PCR. TGF-β stimulated cell-surface expression of the α5β1 integrin in a manner very similar to that previously observed in other cell types (29,30), including rat mesangial cells (31). Treatment of the cells with CTGF also induced an increase in cell-surface α5β1 integrin levels, although the stimulation was less pronounced than with TGF-β. However, treatment of cells with TGF-β and a phosphothioate antisense oligonucleotide to CTGF specifically abrogated much of the effect of TGF-β, suggesting that the effect of TGF-β is mediated at least in part through the induction of CTGF. The fact that antisense CTGF does not completely abolish TGF-β-induced cell-surface α5β1 expression suggests that TGF-β can act on this through more than one mechanism.
Treatment of HMC with TGF-β increased their ability to adhere to fibronectin, in a manner very similar to that observed with rat mesangial cells (31). Treatment with CTGF brought about similar effects, despite the fact that cell-surface levels of α5β1 were much lower in these cultures. This presumably relates to CTGF treatment inducing a similar increase in ligand-occupied integrin receptor on the cell-surface as treatment with TGF-β, as observed in experiments using the 12G10 antibody for detection. However, the use of blocking antibodies to the α5 integrin subunit suggested that, although α5β1 is the predominant receptor expressed by HMC that mediates adhesion to fibronectin, it is unlikely to be the only receptor for this.
The direct importance of the α5β1 integrin in fibronectin matrix assembly by HMC in culture was demonstrated by the fact that blocking antibodies to the receptor appreciably reduced matrix deposition by these cells. Soluble integrin receptors have been generated that have been shown to be functional (42,43). Specific targeting of the α5β1 integrin receptor could prove successful in the control of excessive fibronectin matrix production in DN, although targets further upstream in the disease cascade may prove to be better points of intervention. One such attractive target would be CTGF. The data presented in this study provide further evidence that this growth factor is a downstream mediator of many of the profibrotic actions of TGF-β.
We are grateful for support from Diabetes UK (BSW) and the Medical Research Council, and we would like to thank Dr. Helen Yarwood for her assistance with the FACS analysis and Professor Martin Humphries for helpful discussions.
1. Trevisan R, Barnes D, Viberti G: Pathogenesis of nephropathy. In: Textbook of Diabetes, Vol. 2, 2nd edition, edited by Pickup J, Williams G, London, Blackwell Scientific Publ, 1997, pp 52.1–52.21
2. Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New Engl J Med 329: 977–986, 1993
3. Chen S, Hong SW, Iglesias-de la Cruz MC, Isono M, Casaretto A, Ziyadeh FN: The key role of the transforming growth factor-beta system in the pathogenesis of diabetic nephropathy. Ren Fail 23: 471–481, 2001
4. Gupta S, Clarkson MR, Duggan J, Brady HR: Connective tissue growth factor: Potential role in glomerulosclerosis and tubulointerstitial fibrosis. Kidney Int 58: 1389–1399, 2000
5. Murphy M, Godson C, Cannon S, Kato S, Mackenzie HS, Martin F, Brady HR: Suppression subtractive hybridisation identifies high glucose levels as a stimulus for expression of connective tissue growth factor and other genes in human mesangial cells. J Biol Chem 274: 5830–5834, 1999
6. Abdel-Wahab N, Yevdokimova N, Weston BS, Roberts T, Jun Li X, Brinkman H, Mason RM: Role of connective tissue growth factor in the pathogenesis of diabetic nephropathy. Biochem J 359: 77–87, 2001
7. Bradham DM, Igarashi A, Potter RL, Grotendorst GR: Connective tissue growth factor: a cysteine-rich mitogen secreted by human vascular endothelial cells is related to the SRC-induced immediate early gene product CEF-10. J Cell Biol 114: 1285–1294, 1991
8. Bork P: The modular architecture of a new family of growth regulators related to connective tissue growth factor. FEBS Lett 327: 125–130, 1993
9. Grotendorst GR, Okochi H, Hayashi N: A novel transforming growth factor beta response element controls the expression of the connective tissue growth factor gene. Cell Growth Differ 7: 469–480, 1996
10. Grotendorst GR: Connective tissue growth factor: a mediator of TGF-β action on fibroblasts. Cytokine Growth Factor Rev 8: 171–179, 1997
11. Riser BL, Cortes P: Connective tissue growth factor and its regulation: A new element in diabetic glomerulosclerosis. Ren Fail 23: 459–470, 2001
12. Igarashi A, Okochi H, Bradham DM, Grotendorst GR: Regulation of connective tissue growth factor gene expression in human skin fibroblasts and during wound repair. Mol Biol Cell 4: 637–645, 1993
13. Kothapalli D, Frazier KS, Welply A, Segarini PR, Grotendorst GR: Transforming growth factor beta induces anchorage-independent growth of NRK fibroblasts via a connective tissue growth factor-dependent signaling pathway. Cell Growth Differ 8: 61–68, 1997
14. Frazier K, Williams S, Kothapalli D, Klapper H, Grotendorst GR: Stimulation of fibroblast cell growth, matrix production, and granulation tissue formation by connective tissue growth factor. J Invest Dermatol 107: 404–411, 1996
15. Pendurthi UR, Allen KE, Ezban M, Roa LV: Factor VIIa and thrombin induce the expression of Cyr61 and connective tissue growth factor, extracellular matrix signaling proteins that could act as possible downstream mediators in factor VIIa tissue factor-induced signal transduction. J Biol Chem 275: 14632–14641, 2000
16. Suzama K, Naruse K, Suzama I, Takahara N, Ueki K, Aiello LP, King GL: Vascular endothelial growth factor induces the expression of connective tissue growth factor via KDR.Flt 1:and phosphatidylinositol 3-kinase-akt-dependent pathways in retinal vascular cells. J Biol Chem 275: 40725–40731, 2000
17. Ito Y, Aten J, Bende RJ, Oemar BS, Rabelink TJ, Weening JJ, Goldschmeding R: Expression of connective tissue growth factor in human renal fibrosis. Kidney Int 53: 853–861, 1998
18. Wang S, Denichilo M, Brubaker C, Hirschberg R: Connective tissue growth factor in tubulointerstitial injury of diabetic nephropathy. Kidney Int 60: 96–105, 2001
19. Riser BL, Denichilo M, Cortes P, Baker C, Grondin JM, Yee J, Narins RG: Regulation of connective tissue growth factor activity in cultured rat mesangial cells and its expression in experimental diabetic glomerulosclerosis. J Am Soc Nephrol 11: 25–38, 2000
20. Ikeda S, Makino H, Haramoto T, Shikata K, Kumagai I, Ota Z: Changes in glomerular extracellular matrices components in diabetic nephropathy. J Diabet Complications 5: 186–188, 1991
21. Abdel Wahab N, Harper K, Mason R: Expression of extracellular matrix molecules in human mesangial cells in response to prolonged hyperglycaemia. Biochem J 316: 985–992, 1996
22. Oh JH, Ha H, Yu MR, Lee HB: Sequential effects of high glucose on mesangial cell transforming growth factor-beta 1 and fibronectin synthesis. Kidney Int 54: 1872–1878, 1998
23. Ayo SH, Radnik RA, Glass WF2nd, Garoni JA, Rampt ER, Appling DR, Kreisberg JI: Increased extracellular-matrix synthesis in mesangial cells grown in high-glucose medium. Am J Physiol 260: F185–F191, 1991
24. Scharzbauer JE, Sechler J: Fibronectin fibrillogenesis: a paradigm for extracellular matrix assembly. Curr Opin Cell Biol 11: 622–627, 1999
25. Fogerty FJ, Akiyama SK, Yamada KM, Mosher DF: Inhibition of binding of fibronectin to matrix assembly sites by anti-integrin (α5β1) antibodies. J Cell Biol 111: 699–708, 1990
26. Akiyama SK, Yamada SS, Chen WT, Yamada KM: Analysis of fibronectin receptor function with monoclonal antibodies: Roles in cell adhesion, migration, matrix assembly and cytoskeletal organization. J Cell Biol 109: 863–875, 1989
27. Kagami S, Border WA, Ruoslahti E, Noble NA: Coordinated expression of beta 1 integrins and transforming growth factor-beta-induced matrix proteins in glomerulonephritis. Lab Invest 69: 68–76, 1993
28. Kuhara T, Kagami S, Kuroda Y: Expression of beta-1 integrins on activated mesangial cells in human glomerulonephritis. J Am Soc Nephrol 8: 1679–1687, 1997
29. Cai T, Lei QY, Wang LY, Zha XL: TGF- β1 modulated the expression of α5β1 integrin and integrin-mediated signalling in human hepatocarcinoma cells: Biochem Biophys Res Comm 274: 519–525, 2000
30. Collo G, Pepper MS: Endothelial cell integrin alpha5beta1 expression is modulated by cytokines and during migration in vitro. J Cell Sci 112: 569–578, 1999
31. Kagami S, Kuhara T, Yasutomo K, Okada K, Lo[Combining Diaeresis]ster K, Reutter W, Kuroda Y: Transforming growth factor-β (TGF- β) stimulates the expression of β1 integrins and adhesion by rat mesangial cells. Exp Cell Res 229: 1–6, 1996
32. Weston BS, Abdel-Wahab N, Roberts T, Mason RM: Bacitracin inhibits fibronectin matrix assembly by mesangial cells in high glucose. Kidney Int 60: 1756–1765, 2001
33. Mould AP, Garratt AN, Askari JA, Akiyama SK, Humphries MJ: Identification of a novel anti-integrin monoclonal antibody that recognises a ligand-induced binding site epitope on the beta 1 subunit. FEBS Lett 363: 118–122, 1995
34. Chomczynski P, Sacchi N: Single step methods of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: 156–159, 1987
35. Nista A, Leonetti C, Bernardini G, Mattioni M, Santoni A: Functional role of alpha4beta1 and alpha5beta1 integrin fibronectin receptors expressed on adriamycin-resistant MCF-7 human mammary carcinoma cells. Int J Cancer 72: 133–141, 1997
36. Damsky CH, Fitzgerald M, Fisher SJ: Distribution patterns of extracellular matrix components and β1 integrin receptors are intricately modulated during differentiation of human cryotrophoblast to an invasive phenotype in vivo. J Clin Invest 89: 210–222, 1992
37. Gauer S, Yao J, Schoecklmann HO, Sterzel RB: Adhesion molecules in the glomerular mesangium. Kidney Int 51: 1447–1453, 1997
38. Geiger B, Bershadsky A, Pankov R, Yamada KM: Transmembrane extracellular matrix–cytoskeleton crosstalk. Nature Reviews Mol Cell Biol 2: 793–805, 2001
39. Border WA: Transforming growth factor-beta and the pathogenesis of glomerular diseases. Curr Opin Nephrol Hypertens 3: 54–58, 1994
40. Dammeier J, Brauchle M, Falk W, Grotendorst GR, Werner S: Connective tissue growth factor: A novel regulator of mucosal repair and fibrosis in inflammatory bowel disease? Int J Biochem Cell Biol 30: 909–922, 1998
41. Manabe R, Oh-e N, Maeda T, Fukuda T, Sekiguchi K: Modulation of the cell-adhesive activity of fibronectin by the alternatively spliced EDA segment. J Cell Biol 139: 295–307, 1997
42. Stephens PE, Ortlepp S, Perkins VC, Robinson MK, Kirby H: Expression of a soluble functional form of the integrin alpha4beta1 in mammalian cells. Cell Adhes Commun 7: 377–390, 2000
43. Kainoh M, Tanaka T: Production of soluble integrin alpha2beta1 heterodimer complex functionally active in vitro and in vivo. Biochem Biophys Res Commun 290: 305–310, 2002
Copyright © 2003 The Authors. Published by Wolters Kluwer Health, Inc. All rights reserved.