CTGF Promotes Inflammatory Cell Infiltration of the Renal Interstitium by Activating NF-κB

Abstract

Connective tissue growth factor (CTGF) is an important profibrotic factor in kidney diseases. Blockade of endogenous CTGF ameliorates experimental renal damage and inhibits synthesis of extracellular matrix in cultured renal cells. CTGF regulates several cellular responses, including adhesion, migration, proliferation, and synthesis of proinflammatory factors. Here, we investigated whether CTGF participates in the inflammatory process in the kidney by evaluating the nuclear factor-kappa B (NF-κB) pathway, a key signaling system that controls inflammation and immune responses. Systemic administration of CTGF to mice for 24 h induced marked infiltration of inflammatory cells in the renal interstitium (T lymphocytes and monocytes/macrophages) and led to elevated renal NF-κB activity. Administration of CTGF increased renal expression of chemokines (MCP-1 and RANTES) and cytokines (INF-γ, IL-6, and IL-4) that recruit immune cells and promote inflammation. Treatment with a NF-κB inhibitor, parthenolide, inhibited CTGF-induced renal inflammatory responses, including the up-regulation of chemokines and cytokines. In cultured murine tubuloepithelial cells, CTGF rapidly activated the NF-κB pathway and the cascade of mitogen-activated protein kinases, demonstrating crosstalk between these signaling pathways. CTGF, via mitogen-activated protein kinase and NF-κB activation, increased proinflammatory gene expression. These data show that in addition to its profibrotic properties, CTGF contributes to the recruitment of inflammatory cells in the kidney by activating the NF-κB pathway.

Connective tissue growth factor (CTGF) is a member of the C-terminal cystein-rich proteins (CCN) family of early response genes. CTGF is a 38-kD cystein-rich secreted protein that is up-regulated in proliferative disorders or fibrotic lesions in several human diseases, including skin disorders, atherosclerosis, pulmonary fibrosis, and kidney diseases.^1,2 In human biopsies of different renal pathologies and in experimental models of kidney injury, renal CTGF overexpression was correlated with cellular proliferation and extracellular matrix (ECM) accumulation, both at glomerular and interstitial areas.^2–4 In the diabetic kidney, elevated CTGF expression co-localizes with sites of epithelial-to-mesenchymal transition (EMT) on the tubular epithelium.⁵ In cultured renal cells, recombinant CTGF significantly increases ECM production and induces transition of tubuloepithelial cells to myofibroblasts.^6–8 In experimental diabetic nephropathy in mice, the blockade on endogenous CTGF, by antisense oligonucleotides, has beneficial effects on renal damage progression.⁹ In cultured renal cells, CTGF blockade inhibits ECM accumulation and EMT caused by angiotensin II and transforming growth factor-β (TGF-β).^3,10 These data suggest that CTGF could be an important target for the treatment of renal fibrosis.

CTGF also induces other cellular responses. Depending on the cell type, CTGF regulates cell growth, proliferation, and apoptosis. CTGF is a downstream mediator of TGF-β-induced apoptosis of mesothelial cells,¹¹ but contributes to the survival of hepatic stellate cells.¹² CTGF may play a role as a secreted tumor suppressor protein¹³ or contribute to promote tumor cell growth and invasion.¹⁴ Some studies suggested that CTGF could also be involved in the inflammatory response. CTGF is a chemotactic factor for monocytes¹⁵ and regulates cellular adhesion and migration in mesangial cells.¹⁶ Moreover, in cultured mesangial cells, CTGF enhances the production of proinflammatory factors, including chemotactic molecules, and activates nuclear factor-kappa B (NF-κB).¹⁷ However, there is no data about the in vivo effect of CTGF on the renal inflammatory process.

The molecular mechanisms involved in CTGF signaling are far from being understood. CTGF interacts with tyrosine kinase receptors and integrins that activate multiple signaling systems including NF-κB and mitogen-activated protein kinase (MAPK) pathways.^12,17–19 Although the regulation of the inflammatory response in the kidney is a complex process, the activation of NF-κB plays a pivotal role. Experimental studies have shown that NF-κB blockade by different methods, including I-κB overexpression, NF-κB decoy oligonucleotides, NF-κB inhibitors (parthenolide among others), or indirectly by statins, glucocorticoids, and antioxidants, prevents renal damage.^20–23 Activation of renal NF-κB has been described in human kidney diseases, associated to proinflammatory factors overexpression.^24,25 We have now investigated whether CTGF could modulate the inflammatory response in the kidney and the mechanisms underlying this process, evaluating the involvement of the NF-κB signaling pathway.

RESULTS

Systemic Administration of CTGF Induces Interstitial Inflammatory Cell Infiltration in the Kidney

To investigate the in vivo effect of CTGF in the kidney, a model of acute administration of CTGF in mice was performed. C57Bl/6 mice were intraperitoneally injected with human recombinant CTGF (2.5 ng/g of body weight) and studied 24 h later. As control, a group of vehicle-injected (saline) mice was used. CTGF-injected mice did not present marked histologic changes in the kidney after 24 h (Figure 1A; Masson Trichrome staining), we just observed focal interstitial inflammatory cell infiltration when compared with the control group. CTGF-treated mice had a marked increase of CD68 positive macrophages and CD3 positive T lymphocytes in interstitial areas (Figure 1A) that were not found in control mice (Figure 1B shows staining quantification).

Figure 1.

CTGF causes tubulointerstitial inflammatory cell infiltration in the kidney, and NF-κB inhibition ameliorates CTGF induced renal injury. Animals were injected with 2.5 ng/g recombinant CTGF or saline and killed 24 h later. Some animals were pretreated 24 h before with 3.5 mg/g parthenolide (NF-κB inhibitor) or its vehicle (0.05% DMSO, control group) before CTGF administration. Paraffin-embedded kidney sections were studied. (A) Representative Masson trichrome and immunohistochemistry stainings. Inflammatory cell infiltration was evaluated using anti-CD68 (macrophages) and anti-CD3 (T lymphocytes) antibodies. Arrows indicate stained interstitial inflammatory cells. Magnification, ×400. (B) Immunohistochemistry staining quantification expressed as mean ± SEM of seven animals per group. *P < 0.05 versus control. #P < 0.05 versus CTGF.

Renal function was also evaluated. There were no statistically significant differences in urea (24 ± 4 mg/dl in controls versus 30 ± 2 mg/dl in CTGF-injected mice, P = 0.124) and creatinine plasma levels (0.09 ± 0.03 mg/dl versus 0.06 ± 0.01 mg/dl, P = 0.171) or in the urinary protein/creatinine ratio (3.84 ± 0.27 versus 5.33 ± 1, P = 0.356) between controls and CTGF-injected mice. This suggests that CTGF does not induce loss of renal function.

CTGF Activates the NF-κB Pathway in the Kidney

In nuclear protein extracts of kidneys from mice treated with CTGF, NF-κB DNA binding activity was increased compared with controls, as assessed by electrophoretic mobility shift assay (EMSA) (Figure 2, A and B). Moreover, CTGF-injected mice had increased nuclear p50 and p65 protein levels compared with control animals, as shown by western blot analysis (Figure 2, C and D). These data indicate that CTGF activates the renal NF-κB pathway in vivo.

Figure 2.

CTGF activates the NF-κB pathway in the kidney. NF-κB DNA binding activity was determined in 30 μg nuclear tissue protein extracts by EMSA. The specificity of the NF-κB complexes was determined by competition with an excess of 100-fold cold NF-κB oligonucleotide. The position of free probe is indicated. (A) A representative gel, and (B) the mean ± SEM of seven animals per group. Nuclear p65 and p50 protein levels were analyzed by western blot analysis in tissue nuclear protein extracts. (C) Representative gels of p50 and p65 subunits detected by western blot analysis. Lamin b was used as loading control. (D) Nuclear protein levels expressed as mean ± SEM of seven animals per group. *P < 0.05 versus control. #P < 0.05 versus CTGF.

Blockade of the NF-κB Pathway Diminishes CTGF-Induced Inflammatory Cell Infiltration in the Kidney

To test the in vivo contribution of NF-κB activation to CTGF responses, animals were pretreated with the NF-κB inhibitor parthenolide (3.5 mg/g of body weight) 24 h before CTGF administration. As a control, some animals were injected with 0.05% DMSO (parthenolide vehicle), showing no difference with saline-injected mice. Parthenolide is a sesquiterpene lactone that inhibits NF-κB activation by preventing the degradation of I-κBα.²⁶ This NF-κB inhibitor has previously shown beneficial effects in several experimental models of renal damage.^22,27 Treatment with parthenolide ameliorated CTGF-induced renal damage and markedly diminished the presence of infiltrating monocytes/macrophages and T cells in renal interstitium (Figure 1). Parthenolide significantly diminished renal NF-κB activation caused by CTGF (Figure 2). These data suggest that NF-κB is an important intracellular signaling pathway involved in CTGF-induced renal responses.

CTGF Increases the Renal Expression of Chemokines and Proinflammatory Cytokines Via NF-κB Activation

The local renal production of chemokines, some of them under NF-κB control, regulates inflammatory cell infiltration.²⁰ In CTGF-injected mice, the renal mRNA expression of the chemokines monocyte chemotactic protein-1 (MCP-1) and RANTES (regulated on activation normal T cell expressed and secreted) was significantly up-regulated compared with control animals (real-time PCR, Figure 3A). The expression of both chemokines at the protein level was confirmed by immunohistochemistry. MCP-1 and RANTES positive staining was observed in tubuloepithelial cells and interstitial infiltrating cells (Figure 3B). Treatment with the NF-κB inhibitor parthenolide restored MCP-1 and RANTES gene and protein expression to control levels (Figure 3, A and B), suggesting that NF-κB signaling participates in their regulation.

Figure 3.

In vivo CTGF increases renal expression of chemokines and proinflammatory cytokines through the NF-κB pathway. (A) MCP-1 and RANTES gene expression as mean ± SEM of seven animals per group analyzed by real-time PCR. (B). Immunostainings for MCP-1 and RANTES. A representative animal out of six studied in each group is shown. Magnification, ×400. (C). In renal extracts, cytokine levels were evaluated by ELISA and shown as mean ± SEM of seven animals per group. *P < 0.05 versus control. #P < 0.05 versus CTGF.

Kidneys from CTGF-treated mice presented elevated levels of the T helper (Th) 1/2 cytokines IFN-γ, IL-6, and IL-4 compared with control mice, which were abolished by parthenolide pretreatment (Figure 3C). By contrast, IL-10 (main regulatory T cells cytokine secreted) was down-regulated by CTGF, but was not modified by NF-κB inhibition (Figure 3C). CTGF has no effect on IL-17 production, the main cytokine secreted by TH17 cells, showing that TH17 response was not activated by CTGF (Figure 3C).

CTGF Activates the NF-κB Pathway in Tubuloepithelial Cells

To further investigate the mechanism involved in the inflammatory effects of CTGF in the kidney, experiments were performed in cultured murine proximal tubuloepithelial cells (MCT cells), an important cell implicated in renal inflammation.²⁸

First, we evaluated whether CTGF could activate the NF-κB pathway in these cells. In MCTs, increased NF-κB DNA binding activity in response to CTGF was observed as soon as 15 min, peaked at 30 min, and returned to control levels at 2 h (Figure 4, A and B). CTGF induced a dose-dependent NF-κB activation, with a maximal response between 10 to 100 ng/ml (Figure 4A). CTGF also induced the phosphorylation and consequent degradation of the cytosolic inhibitory subunit I-κBα, observed at 30 min, the time of the maximal NF-κB activation (western blot analysis, Figure 4, C and D). By immunofluorescence, we have located NF-κB subunits. In control cells, a diffuse cytoplasmic fluorescence was seen with anti-p50 and anti-p65 antibodies. After 30 min with CTGF, an intense nuclear fluorescence was noted, showing a translocation of p50 and p65 NF-κB subunits into the nucleus (Figure 4E). By transient transfection of MCT cells with a NF-κB-luciferase reporter vector, we have found that CTGF significantly increased NF-κB promoter activity, similar to that observed with TNF-α (not shown), indicating NF-κB-dependent gene transcription.

Figure 4.

CTGF activates the NF-κB pathway in murine tubuloepithelial cells. Murine tubuloepithelial cells (MCT cell line) were growth-arrested for 24 h and then stimulated with 10 ng/ml CTGF. NF-κB DNA binding activity was determined in 10 μg nuclear extracts by EMSA. (A) The time-response curve as mean ± SEM of three experiments is shown. *P < 0.05 versus control. (B) A representative dose-response EMSA after 30 min of CTGF incubation, out of three performed. (C) Phosphorylated I-κBα versus I-κBα protein levels as mean ± SEM of three experiments of western blot analysis. (D) Phosphorylated I-κBα and I-κBα representative gels of western blot analysis of three experiments performed. (E) CTGF causes nuclear translocation of p50 and p65 NF-κB subunits in MCT. Cells were treated with 10 ng/ml CTGF for 30 min. Indirect immunofluorescence was done using p50 and p65 antibodies and FITC-labeled secondary antibodies (green staining). Nuclei were stained with propidium iodide (PI, in red). In the merge confocal microscopy image, the yellow color indicates nuclear localization of NF-κB subunits. A representative experiment out of three performed for each subunit is shown.

To further study NF-κB activation induced by CTGF, we evaluated this effect in human tubuloepithelial cell line (HK-2). In these cells, CTGF activates the NF-κB transcription factor as soon as 30 min, remaining elevated after 1 h; but a higher dose than in MCT cells is required (50 ng/ml) (Figure 5).

Figure 5.

CTGF activates the NF-κB pathway in HK-2 cells. Human tubuloepithelial cells (HK-2 cell line) were growth-arrested for 24 h and were pretreated with parthenolide (NF-κB inhibitor 10⁻⁶ M) for 1 h. Then they were stimulated with 50 ng/ml CTGF for 30 min and 1 h. NF-κB DNA-binding activity was determined in 10 μg nuclear extracts by EMSA. (A) A representative gel of NF-κB activation by EMSA. (B) NF-κB activation levels as mean ± SEM of three experiments. *P < 0.05 versus control. #P < 0.05 versus CTGF.

CTGF Activates the NF-κB Pathway Through MAPK Activation in Murine Tubuloepithelial Cells

To investigate the upstream mechanisms involved in NF-κB activation induced by CTGF, we have studied the MAPK cascade. First, we have evaluated whether CTGF could activate the MAPK signaling pathway in MCTs. By western blot analysis, we have observed that CTGF increased the phosphorylation of JNK1/2 at 10 min, ERK1/2 at 15 min, and p38 at 20 min (Figure 6, A and B), preceding the maximal NF-κB activation. The potential role of MAPKs in NF-κB activation was studied using specific inhibitors of ERK1/2 (PD98059), p38 (SB203580), and JNK (SP600125).⁸ Treatment with any of the inhibitors reduced NF-κB DNA binding activity (Figure 7, A and B) and p65 nuclear translocation (Figure 7C) caused by CTGF, showing the involvement of all three MAPKs, ERK1/2, p38, and JNK, in this process.

Figure 6.

CTGF activates ERK1/2, JNK1/2, and p38 pathways in MCT cells. MCT cells were serum-deprived for 24 h and then stimulated with 10 ng/ml CTGF. Increased phosphorylated protein levels are considered evidence of MAPK pathway activation. (A) A representative gel for ERK1/2, JNK1/2 and p38 phosphorylated, and total proteins, and (B) mean ± SEM of three independent experiments. *P < 0.05 versus control.

Figure 7.

MAPK pathway inhibition abolishes NF-κB activation caused by CTGF in MCT cells. Growth-arrested cells were pretreated with p38 inhibitor (SB203580; 10⁻⁵ mol/L), ERK1/2 inhibitor (PD98059; 10⁻⁵ mol/L), and JNK inhibitor (SP600125; 10⁻⁵ mol/L). Then cells were treated with 10 ng/ml CTGF for 30 min. (A) A representative EMSA out of three performed, and (B) mean ± SEM of three independent experiments. *P < 0.05 versus control. #P < 0.05 versus CTGF. (C) Representative confocal microscopy images of p65 immunofluorescence staining out of three independent experiments.

CTGF Increases the Production of Proinflammatory Mediators in Tubuloepithelial Cells

We have also investigated whether CTGF regulates some proinflammatory factors under NF-κB control in MCTs. By real-time PCR, we have found that in MCTs, CTGF up-regulated gene expression of the chemokine MCP-1, the intercellular adhesion molecule-1 (ICAM-1), and the cytokine IL-6. Parthenolide prevented the up-regulation of these proinflammatory mediators (Figure 8). Moreover, treatment with specific inhibitors of ERK1/2 (PD98059), p38 (SB203580), and JNK (SP600125) reduced MCP-1 gene expression to control levels. However, ICAM-1 gene overexpression was reduced by p38 and JNK inhibition, and only the inhibition of JNK diminished IL-6 overexpression induced by CTGF.

Figure 8.

CTGF, via MAPK and NF-κB pathways, increases proinflammatory factors in MCT. Cells were pretreated with parthenolide (NF-κB inhibitor 10⁻⁶ M), SB203580 (p38 inhibitor; 10⁻⁵ mol/L), PD98059 (ERK1/2 inhibitor; 10⁻⁵ mol/L), and SP600125 (JNK inhibitor; 10⁻⁵ mol/L) for 1 h and then stimulated with 10 ng/ml CTGF for 6 h. Shown are MCP-1 (A), ICAM-1 (B), and IL-6 (C) gene expression levels determined by real-time PCR and expressed as mean ± SEM of three experiments. *P < 0.05 versus control. #P < 0.05 versus CTGF.

We also observed that CTGF induces IL-6, MCP-1, and ICAM-1 overexpression after 6 h in HK-2 cells (Figure 9) through NF-κB pathway activation.

Figure 9.

CTGF increases proinflammatory factors in HK-2 cell line. Cells were pretreated with parthenolide (NF-κB inhibitor 10⁻⁶ M) for 1 h and then stimulated with 50 ng/ml CTGF for 6 h. Shown are MCP-1 (A), ICAM-1 (B), and IL-6 (C) gene expression levels determined by real-time PCR and expressed as mean ± SEM of five experiments. *P < 0.05 versus control. #P < 0.05 versus CTGF.

DISCUSSION

In this work, we have demonstrated that CTGF in vivo promotes renal inflammatory cell infiltration via NF-κB pathway activation and up-regulation of proinflammatory mediators.

Previous studies suggested that CTGF could participate in the regulation of the inflammatory process. CTGF is a chemotactic factor for cultured monocytes¹⁵ and participates in mesangial migration and adhesion.¹⁶ We have observed that the systemic administration of CTGF in mice leads to tubulointerstitial infiltration of macrophages and T cells. These data are the first evidence clearly showing that CTGF contributes to renal inflammation in vivo.

Among the intracellular signaling system involved in the regulation of inflammatory and immune responses, NF-κB has special interest. NF-κB is a transcription factor that regulates the gene expression of proinflammatory mediators, including chemokines and cytokines, that contribute to the recruitment of inflammatory cells in the renal tissue.^20,21 We have observed that CTGF activates the NF-κB pathway in the kidney, and this is associated with increased tubuloepithelial cell expression of chemokines and recruitment of interstitial macrophages and T cells. Moreover, the blockade of NF-κB activation by pretreatment with parthenolide inhibited chemokines overexpression and inflammatory cell infiltration induced by CTGF. These data show that NF-κB is an important signaling pathway involved in the inflammatory response elicited by CTGF in the kidney.

We have further investigated the mechanisms involved in NF-κB activation by CTGF in cell culture studies. In renal tubuloepithelial cells, CTGF induced a transient increase in NF-κB DNA binding activity. NF-κB proteins are associated with I-κB proteins and retained in the cytoplasm where they are inactive. Activation of the classical NF-κB pathway consists of the dissociation of the inhibitory subunit I-κBα from the complex and its subsequent degradation. Then the active NF-κB complex is translocated to the nucleus where it activates the transcription of target genes.^20,21 CTGF elicited nuclear translocation of p65 and p50 subunits and decreased cytosolic I-κBα levels, indicative of I-κBα degradation, with a time-course consistent with the DNA binding assay. CTGF also increased NF-κB-dependent gene transcription and up-regulated the expression of NF-κB-controlled genes, such as MCP-1, IL-6, and ICAM-1, involved in the inflammatory response. These data show that CTGF activates the classical NF-κB, linked to inflammation, and are consistent with the observed in vivo effects in the kidney. In addition, previously reported NF-κB activation in cultured hepatic, mesangial, and pancreatic cells support our findings.^12,17,29 Moreover, we have demonstrated that CTGF increases NF-κB activation and induces MCP-1, IL-6, and ICAM-1 overexpression in HK-2 cell line, corroborating that this CTGF effect is not cell type-specific response.

Chemokines are important mediators in the recruitment of specific subpopulations of inflammatory cells into renal compartments. MCP-1/CCL2 has a key role in monocyte/macrophage recruitment in animal models of renal injury and in renal biopsies from patients with type 1 and 2 diabetes.³⁰ Experimental blockade of MCP-1 has demonstrated the involvement of this chemokine in the pathogenesis of progressive glomerular and tubulointerstitial lesions in different animal models of renal damage. T lymphocyte recruitment is influenced by up-regulation of RANTES/CCL5.³¹ Systemic administration of CTGF in mice induces the renal production of chemotactic mediators, such as MCP-1 and RANTES, mainly located in tubuloepithelial cells. During kidney damage, tubuloepithelial cells are activated and acquire a proinflammatory phenotype, contributing to inflammatory cell recruitment. Moreover, the direct contact between activated Th2 cells and tubuloepithelial cells can amplify the local inflammatory response in the kidney, through the production of inflammatory mediators.

As we have noted, CTGF could regulate the Th1 or Th2 responses. Th1 cells secrete one characteristic set of cytokines, namely IFN-γ, IL-2, IL-12, IL-18, and CD40L, and evoke cell-mediated immunity and phagocyte-dependent inflammation. Th2 cells, which secrete another set of cytokines, such as IL-4, IL-5, IL-6, IL-9, IL-10, and IL-13, evoke strong antibody responses (including those of the IgE class) and eosinophil accumulation, but inhibit several functions of phagocytic cells (phagocyte-independent inflammation).³² IFN-γ participates in renal damage through the regulation of immune cell migration contributing to the accumulation of inflammatory cells, mainly macrophages and leukocytes, in the injured kidney.^33,34 IL-6 is expressed by tubuloepithelial cells in human acute and chronic renal disease.^35,36 In renal cells, IL-6 and IL-4 induce the production of cytokines, chemokines, ECM proteins, and regulate cell growth.^37,38 IL-4 amplifies the response of other cytokines. In tubuloepithelial cells, IL-4 increases CD40-induced RANTES production.³⁹ CTGF-injected mice presented elevated renal production of the cytokines IFN-γ (Th1) and IL-4 and IL-6 (Th2), that were down-regulated by NF-κB inhibition. IL-10 inhibits the production of proinflammatory cytokines, chemokines, and growth factors controlling tissue damage and exerts beneficial effects in experimental models of renal damage.^40,41 In mice, renal IL-10 was down-regulated by CTGF, indicating that CTGF not only increases proinflammatory mediators but also down-regulates anti-inflammatory cytokines, such as IL-10.

Recents studies have described that some T cells could differentiate in IL-17 secreting cells, inducing TH17 response.⁴² The presence of TGF-β and IL-6 gives origin to murine TH17 cells, however human TH17 cells appear to originate in response to IL-23 and IL-1β.⁴³ CTGF acute administration did not increase IL-17 production, showing that TH17 responses is not activated in the kidney of these animals.

Inflammatory molecules increase NF-κB activity through multiple signaling pathways depending on cell type. Previous studies have demonstrated the relation between MAPK and NF-κB pathways.⁴⁴ CTGF activates MAPK cascade in several cell types.^{16,18,19,45,46} In MCTs, CTGF induced a rapid activation of ERK1/2, p38, and JNK1/2. The specific blockade of the activation of either of the three MAPKs, ERK1/2, p38, or JNK, inhibited NF-κB DNA binding activity and abolished p65 nuclear translocation caused by CTGF, suggesting that in MCT cells, all MAPKs participate in the activation of NF-κB. However, depending on the cell type, the specific cellular responses elicited by CTGF are differentially modulated by MAPKs. In condrocytes, cell growth is mediated by JNK,¹⁹ proliferation via ERK1/2, and differentiation by p38.⁴² In mesangial cells, ERK1/2 mediates CTGF-induced fibrosis, migration, cytoskeleton reorganization, and chemokines production.¹⁶ In human tubuloepithelial cells, CTGF induces the expression of integrin-linked kinase via ERK1/2 activation.⁴⁶ We have found that in MCTs cells, the proinflammatory mediators, MCP-1, IL-6, and ICAM-1 induced by CTGF are controlled by NF-κB pathway, but different MAPKs are involved in their regulation. In this sense, MCP-1 overexpression is regulated by all three MAPKs, as described for other factors involved in renal inflammation, such as angiotensin II.⁴⁷ We observed that JNK activation by CTGF also regulates IL-6 overexpression, as observed in human tubuloepithelial cells in response to IL-1β⁴⁸. In vivo p38 inhibition reduced renal MCP-1 levels and the severity of tissue injury but did not suppress renal levels of IL-6. By contrast, in cultured mesangial cells, p38 blockade down-regulated the production of MCP-1 and IL-6 by cytokines.⁴⁹ These data indicate that JNK/NF-κB plays a key role in IL-6 regulation in these cells. Previous studies in human tubuloepithelial and vascular cells have shown that selective inhibitors of p38 (SB203580), JNK (SP600125), and ERK (PD98059) suppressed ICAM-1 expression by cytokines.^50,51 In endothelial cells, p38-mediated activation of NF-κB and AP-1, resulted in increased expression of ICAM-1.⁵² In response to CTGF stimulation in murine tubuloepithelial cells, ICAM-1 is up-regulated by both p38 and JNK activation. Our data suggest that multiple signaling pathways are activated during CTGF-induced damage in tubuloepithelial cells, showing a complex mechanism that gives to CTGF a proinflammatory cytokine behavior.

Several studies have demonstrated that inhibition of endogenous CTGF is beneficial for the treatment of fibrotic diseases. In nephrectomized TGF-β1 transgenic mice, CTGF inhibition, using antisense oligodeoxynucleotides, diminished renal interstitial fibrogenesis.⁵³ In a model of diabetic nephropathy in mice, treatment with CTGF antisense oligonucleotides reduced renal CTGF, ECM, and proteinase inhibitor mRNA expression.⁹ In a model of liver fibrosis, intraportal vein CTGF small interfering RNA (siRNA) injection reduced CTGF, type I and III collagen, and laminin mRNA levels, attenuating liver fibrosis by reducing ECM accumulation.⁵⁴ In human diabetic nephropathy, plasma CTGF contributes to predict the progression to end-stage renal disease in patients with type 1 diabetic nephropathy, suggesting that CTGF could be considered as a risk marker of diabetic renal and vascular disease.^55–57 Studies in patients with IgA nephropathy showed that expression of CTGF and TGF-β mRNA in tubuloepithelial cells were correlated with the degree of tubulointerstitial damage.⁵⁸ Moreover, studies in patients with chronic cardiac damage suggest that plasma CTGF levels can be used as a marker of cardiac dysfunction and myocardial fibrosis.⁵⁹

The data presented in this work exhibit that CTGF, in addition to its well-known profibrotic effects, promote renal interstitial inflammation increasing renal resident cell activation, inflammatory cell infiltration, and cytokines and chemokines overexpression. Our data show for the first time that CTGF participates in the inflammatory process in vivo and support the idea that CTGF could be an interesting therapeutic target to reduce renal damage.

CONCISE METHODS

Antibodies and Reagents

We purchased culture reagents from Life Technologies BRL (Paisley, Scotland, UK), and recombinant human CTGF (endotoxin level is <0.1 ng/μg CTGF) from MBL International Corporation (Woburn, MA). We obtained recombinant TNF-α from Peprotech (London, UK), and NF-κB consensus oligonucleotide from Promega (Madison, WI).

We used the following inhibitors: PD98059 (ERK1/2 inhibitor), SB203580 (p38 MAPK inhibitor), and SP600125, (JNK-1,-2,-3 inhibitor) from Stressgen Bioreagents Corp. (Victoria, B.C., Canada); and parthenolide (NF-κB inhibitor) from Calbiochem (EMD Chemical, San Diego, CA).

Cell Cultured Studies

MCT cells are a well-characterized line of murine proximal tubular epithelial cells.⁶⁰ Cells were counted, seeded, and grown in RPMI 1640 medium (Life Technologies, Grand Island, NY), supplemented with 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM glutamine in the presence of 10% heat-inactivated fetal bovine serum (FBS). We then cultured the cells at 37°C in 5% CO₂ atmosphere. At confluence, cells were made quiescent for 24 h, and then different studies were performed.

We grew HK-2 cells in RPMI 1640 medium with 10% of heat-inactivated FBS, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM glutamine, 5 μg/ml ITS (Insulin Transferrin Selenium) (Sigma, St. Louis, MO) and 36 ng/ml hydrocortisone (Sigma) at 37°C in 5% CO₂ atmosphere. At confluence, cells were serum-deprived for 24 h, and then experiments were done.

In Vivo Studies

We performed studies in adult male C57Bl/6 mice (9 to 12 wk old, 20 g; Harlan Interfauna Ibérica, S.A., Barcelona, Spain). All of the procedures on animals were performed according to the International and Institutional Animal Research Committee guidelines.

Mice received 3.5 mg/g body weight parthenolide (NF-κB inhibitor) or its vehicle (0.05% DMSO) 24 h before 2.5 ng/g body weight CTGF or saline by intraperitoneal administration. We randomized treatment and control groups (n = 7) as follows: (i) parthenolide followed by CTGF; (ii) vehicle (DMSO) followed by CTGF; (iii) parthenolide followed by saline; (iv) vehicle (DMSO) followed by saline; and (v) saline followed by CTGF. We calculated the dose of CTGF according to in vitro experiments. The dose of parthenolide was established based on previous experience in the lab.^22,27

We perfused the kidneys in situ with cold saline before removal. One kidney from each mouse was fixed in buffered formalin, embedded in paraffin, and used for immunohistochemistry. The other kidney was snap-frozen in liquid nitrogen for gene and protein studies. We collected blood samples to analyze urea and creatinine plasma levels. We also collected urine to evaluate proteinuria and creatinine.

Gene Expression Studies

For mRNA isolation, we pulverized frozen kidney pieces in a metallic chamber and dissolved them in 1 ml Trizol (Invitrogen, San Diego, CA). We also isolated total RNA from cultured cells with Trizol. We synthesized cDNA utilizing the High Capacity cDNA Archive kit (Applied Biosystems, Foster City, CA) using 2 μg total RNA primed with random hexamer primers following the manufacturer's instructions.

We performed real-time RT-PCR using fluorogenic TaqMan MGB probes and primers designed by Assay-on-Demand gene expression products (Applied Biosystems): mouse MCP-1 Mm00441242_m1, mouse RANTES Mm01302428_m1, mouse IL-6 Mm00446190_m1, mouse ICAM_1 Mm00516023_m1, human IL-6 Hs_00174131_m1, human MCP-1 Hs_00234140_m1, and human ICAM-1 Hs_00164932_m1. We normalized data with murine glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and 18s eukaryotic ribosomal RNA expression (assay IDs: Mm99999915_g1 and Hs99999901_m1, respectively), and we calculated the mRNA copy numbers for each sample by the instrument software using cycle threshold (Ct) value (“arithmetic fit point analysis for the lightcycler”). We expressed results in copy numbers, calculated relative to unstimulated cells or control mice, after normalization against GAPDH and 18s, as described previously.²²

EMSA: NF-κB Activation

For protein extraction from tissues, we pulverized frozen kidney pieces and resuspended them in a cold extraction buffer [20 mmol/L HEPES-NaOH, pH 7.6, 20% (vol/vol) glycerol, 0.35 mol/L NaCl, 5 mmol/L MgCl₂, 0.1 mmol/L EDTA), 1 mmol/L dithiothreitol (DTT), 0.5 mmol/L phenylmethylsulfonyl fluoride (PMSF)]. The homogenate was shaken for 30 min and then centrifuged at 40,000 × g for 30 min at 4°C. For nuclear protein extraction from cultured cells, they were homogenized using extraction buffer (10 mmol/L HEPES, pH 7.8, 15 mmol/L KCl, 2 mmol/L MgCl₂, 0.1 mmol/L EDTA, 1 mmol/L DTT, 1 mmol/L PMSF). We separated nuclei and cytosolic fractions by centrifugation at 1000 × g for 10 min. We then resuspended the nuclei in extraction buffer to a final concentration of 0.39 mol/L KCl, followed by centrifugation at 100,000 × g for 30 min. We dialyzed the supernatants overnight against a binding buffer containing 20 mmol/L HEPES-NaOH, pH 7.6, 20% (vol/vol) glycerol, 0.1 mmol/L NaCl, 5 mmol/L MgCl₂, 0.1 mmol/L EDTA, 1 mmol/L DTT, and 0.5 mmol/L PMSF. We then centrifuged the dialysates at 10,000 × g for 15 min at 4°C and collected the supernatants. We quantified protein concentration by the BCA method (Pierce, Rockford, IL).

We evaluated NF-κB activity by binding 30 μg tissue nuclear extracts or 10 μg cell nuclear extracts²² to a NF-κB consensus oligonucleotide (5′-AGTTGAGGGGACTTTCCCAGGC-3′) ³²P-end-labeled by incubation for 10 min at 37°C with 10 U T4 polynucleotide kinase (Promega). The reaction was stopped by the addition of EDTA to a final concentration of 0.05 M. We equilibrated nuclear protein for 10 min in a binding buffer containing 4% glycerol, 1 mmol/L MgCl₂, 0.5 mmol/L EDTA, 0.5 mmol/L DTT, 50 mmol/L NaCl, 10 mmol/L Tris-HCl, pH 7.5, and 50 μl/ml poly(dI-dC) (Pharmacia, Uppsala, Sweden). We used HeLa cell nuclear extract as a known positive control because it contains NF-κB (not shown). To assess the specificity of the reaction, we performed the following controls: negative assay without cellular extracts; and competition assays with a 100-fold excess of unlabeled NF-κB, mutant NF-κB, and nonspecific oligonucleotides. For competition assays, we added the unlabeled probe 20 min before the addition of the labeled probe. We added 0.35 pmol labeled probe to the reaction, followed by incubation for 20 min at room temperature. The reaction was stopped by adding gel loading buffer (250 mM Tris-HCl, 0.2% bromophenol blue, 0.2% xylene cyanol, and 40% glycerol) and run on a nondenaturing, 4% acrylamide gel at 150 V at room temperature in 0.25% Tris-borate EDTA. We then dried the gel and exposed it to x-ray film.

Protein Studies

We quantified protein levels by western blot analysis. Cytosolic extracts from cultured cells were obtained by homogenization and centrifugation as described previously. ²² Then, samples were separated by SDS-PAGE, transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA), blocked in PBS containing 0.1% Tween-20, 7.5% dry skimmed milk for 1 h at room temperature, and incubated in the same buffer with different primary antibodies overnight at 4°C. We used the following antibodies: phosphorylated I-κBα, I-κBα, lamin b, phospho-ERK1/2, ERK1/2, phospho-p38, p38, JNK1/2 (Santa Cruz Biotechnology, Santa Cruz, CA); and phospho-JNK1/2 (Stressgen Bioreagents Corp). After washing, we incubated the membranes with peroxidase-conjugated secondary antibody and developed them using an ECL chemiluminiscence kit (Amersham Pharmacia Biotech, Piscataway, NJ). We quantified proteins in all samples using the BCA method, and we loaded 50 μg protein in each lane. The quality of proteins and efficacy of protein transfer were evaluated by Red Ponceau staining (not shown). To evaluate equal loading, we stained the membranes with anti-α-tubulin antibody (Sigma), and we used the total MAPK protein levels in MAPK phosphorylation studies. We scanned the autoradiographs using the GS-800 Calibrated Densitometer (Quantity One; Bio-Rad, Madrid, Spain).

We assayed IL-10, IL-6, IL-4, and IFN-γ protein levels by an ELISA kit from BD Bioscience (Erembodegem, Belgium) and IL-17 from eBioscience (San Diego, CA). We analyzed cytokine levels in tissue total protein extracts in lysis buffer (50 mmol/L Tris-HCl, 150 mol/L NaCl, 2 mmol/L EDTA, 2 mmol/L EGTA, 0.2% Triton X-100, 0.3% IGEPAL, 10 μl/ml proteinase inhibitors cocktail, 10 μl/ml PMSF, and 10 μl/ml orthovanadate), following the manufacturer's instructions. We expressed data as n-fold increase over the mean of control mice levels.

Immunofluorescence Staining of NF-κB Subunits

To determine which NF-κB subunits were activated by CTGF, we seeded cells in 24-well plates over glass coverslips. After a 24-h serum starvation step, we pretreated the cells with inhibitors and stimulated with CTGF, fixed in Merckofix (Merck, Whitehouse Station, NJ), treated with 0.1% Triton-X100, blocked with 4% BSA in 1× PBS, and incubated with primary antibodies against p50 and p65 subunits overnight. After washing, we incubated the cells with FITC secondary antibody, and we stained the nuclei with 1 μg/ml propidium iodide. The absence of primary antibody was the negative control. We mounted samples in Mowiol 40 to 88 (Sigma) and examined them by a laser scanning confocal microscope (Leica Microsystems, Wetzlar, Germany).⁸

Transfection, DNA Constructs, and Promoter Studies

We transiently transfected MCT cells cultured in 6-well plates with FuGENE (Roche Molecular Biochemicals) and 1 μg NF-κB/luc promoter and 1 μg TK-renilla as internal control. After a 24-h serum starvation step, we stimulated cells for 24 h and then evaluated the luciferase/renilla activity with Promega kits.⁶¹

Renal Histology and Immunohistochemistry

We stained the paraffin-embedded kidney sections using standard histology procedures, including Masson Trichrome. Immunostaining was carried out in 4-μm-thick tissue sections that were deparafinized through xylene and hydrated through graded ethanol (100%, 96%, 90%, and 70%) and distilled water. Endogenous peroxidase was blocked. We incubated the tissue sections with trypsin for 30 min at 37°C and then for 1 h at room temperature with 4% BSA and serum in 1× PBS to eliminate nonspecific protein binding sites. We incubated primary antibodies to detect MCP-1 (Becton Dickinson, Franklin Lakes, NJ), RANTES (Chemicon International, Millipore), CD3, and CD68 (Serotec, Oxford, UK) overnight at 4°C. After washing, we treated the slides with the corresponding anti-IgG biotinylated-conjugated secondary antibody (Amersham Bioscience, Buckinghamshire, England) followed by the avidin-biotin-peroxidase complex (Dako, Dako Diagnósticos S.A, Barcelona, Spain), and 3,3′-diaminobenzidine as chromogen. We counterstained the sections with Carazzi's hematoxylin. We performed negative controls by incubation with a nonspecific Ig of the same isotype as the primary antibody and without primary antibody. Quantification of CD68-stained and CD3-stained cells was made by determining the total number of positive cells in 20 randomly chosen fields (×400) using Image-Pro Plus software (Media Cybernetics, Bethesda, MD). We expressed data as positive stained area versus total analyzed area. We examined samples from each animal in a blind manner.

Statistical Analysis

We expressed results as mean ± SEM fold change over control. We used one-way ANOVA to compare gene and/or protein expression levels between groups. When statistical significance was found, we used the Bonferroni post hoc comparison test to identify differences between groups. We considered differences significant at P < 0.05. Statistical analyses were performed using the SPSS statistical software, version 11.0 (SPSS Inc., Chicago, IL).

DISCLOSURES

None.