Bone Morphogenetic Protein-7 Inhibits Constitutive and Interleukin-1β-Induced Monocyte Chemoattractant Protein-1 Expression in Human Mesangial Cells: Role for JNK/AP-1 Pathway ¹

Monocyte chemoattractant protein-1 (MCP-1)³ is an important mediator for monocyte/macrophage infiltration in acute and chronic inflammatory renal diseases (1, 2, 3, 4). A critical role for MCP-1 in the pathogenesis of renal inflammation was established by renal parenchymal cell culture studies as well as by blocking MCP-1 activity with neutralizing Abs in animal models of inflammatory renal disease (3, 4, 5, 6). Therefore, clinically applicable anti-MCP-1 agents have been sought that inhibit local MCP-1 production by renal cells, neutralize MCP-1 activity, or block receptors for MCP-1.

Bone morphogenetic proteins (BMPs) are a subgroup of the TGF-β superfamily, originally identified by their ability to induce the formation of endochondral bone at extraskeletal sites (7). BMPs, comprising over 15 members in humans, share the highly conserved seven cysteines in the C terminus, characteristic of members of the TGF-β superfamily (8). BMPs play an important role in the development of solid organs, including lung, heart, teeth, gut, skin, and kidney (9).

BMP-7, also known as osteogenic protein-1, is required for normal renal development and BMP-7 expression remains high in the adult kidney (10, 11). BMP-7 mRNA expression was demonstrated in the renal glomerular and tubule cells (12), and membrane-bound, specific, high affinity BMP receptors binding to BMP-7 were localized in the kidney, including mesangial cells (12, 13). However, little is known about the physiological and pathophysiological function of BMP-7 in the adult kidney.

BMP-7 has been reported to reduce histological and functional injury in a rat model of ischemic acute renal failure (14, 15). Recently, there were some reports demonstrating the renal antifibrotic effect of BMP-7 in rat models of unilateral ureteral obstruction (UUO) (16, 17) and diabetic nephropathy (18). In a rat model of UUO, BMP-7 effectively prevented interstitial inflammation and tubulointerstitial fibrosis, leading to preservation of renal function. In addition, BMP-7 treatment significantly reduced interstitial macrophage infiltration, raising the possibility of regulation of chemokine expression in renal cells by BMP-7. Gould et al. (19) demonstrated that BMP-7 repressed the basal and TNF-α-induced expression of proinflammatory cytokines IL-6 and IL-1β, and chemokines MCP-1 and IL-8 in human proximal tubule epithelial cells. However, the effect of BMP-7 on chemokine expression in renal glomerular cells, and the mechanisms involved in the regulation of chemokine expression have not been determined.

In this study, we examined the effects of BMP-7 on constitutive and IL-1β-induced MCP-1 expression in human mesangial cells and monocyte chemotactic activity released from IL-1β-stimulated mesangial cells. To explore the mechanism involved, we also investigated the effect of BMP-7 on IL-1β-induced transcription factors, NF-κB and AP-1, and c-Jun N-terminal kinase (JNK) activity.

Recombinant human BMP-7, IL-1β, MCP-1, and anti-MCP-1 Abs were purchased from R&D Systems (Minneapolis, MN). Anti-IκBα, anti-phosphorylated-JNK, anti-c-Jun, anti-c-Fos, anti-p65, anti-actin Abs, NF-κB consensus oligonucleotides, AP-1 consensus oligonucleotides, and AP-1 mutant oligonucleotides were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-pan JNK Ab was purchased from Cell Signaling Technology (Beverly, MA). The [³²P]dCTP and [γ-³²P]ATP were from NEN (Boston, MA). Curcumin and pyrrolidine dithiocarbamate (PDTC) were purchased from Sigma-Aldrich (Deisenhofen, Germany). All supplements for cell culture were purchased from Life Technologies (Gaithersburg, MD).

Human mesangial cells were cultured as described previously (20). Mesangial cells were grown in DMEM supplemented with 17% heat-inactivated FBS, 10 mM HEPES, 0.01% nonessential amino acid, 13 mg/ml insulin, and an antibiotic/antifungal solution. For the experiments, cells were used between passages three and eight.

After a 24-h starvation, confluent cells were preincubated with or without various concentration of BMP-7 (50–200 ng/ml) for 30 min, and then were stimulated with IL-1β (250 pg/ml) for 24 h. The conditioned medium was collected, centrifuged at 10,000 × g (4°C) for 3 min and stored at −70°C until assayed. For the time course, confluent cells incubated in the serum and insulin-free medium were preincubated with or without BMP-7 (200 ng/ml) for 30 min, and then IL-1β (250 pg/ml) was added for the indicated time points. For MCP-1 ELISA, mouse anti-human MCP-1 Ab (2 μg/ml) and biotinylated goat anti-human MCP-1 Ab (1 μg/ml) were used. The TMB microwell peroxidase substrate system (Kirkegaard & Perry Laboratories, Gaithersburg, MD) was used as an enzyme substrate and the reaction was followed by measurement of OD₄₅₀.

Mesangial cell viability was assessed using a lactate dehydrogenase (LDH) cytotoxicity detection kit (Takara Biomedicals, Kyoto, Japan) that measures LDH release into the culture medium. The manufacturer’s protocols were followed.

Serum-starved cells were preincubated with or without BMP-7 for 30 min and were incubated with IL-1β (250 pg/ml) for 24 h. Monocyte chemotactic activity was determined in modified Boyden chambers by using freshly prepared human PBMCs as previously described (21). Conditioned medium obtained from mesangial cells was diluted 1/4 to 1/9 and assayed for monocyte chemotactic activity to determine the optimal dilution for maximal stimulation of monocyte migration. Chemotactic activity is expressed as the mean number of migrated monocytes per field in 10 high power fields (×1,000 magnification).

Serum-starved confluent cells were preincubated with or without BMP-7 for 30 min, and then were stimulated with IL-1β (250 pg/ml) for the indicated time points. RNA was isolated by one-step guanidium-thiocyanate-phenol-chloroform extraction. RNA quantification was determined by absorbance at 260 nm. Total RNA was fractionated by electrophoresis on a 1% agarose-formaldehyde gel, blotted by capillary transfer on a nylon membrane (Gene Screen; NEN), and cross-linked by UV irradiation (UV Strata-linker 1800; Stratagene, La Jolla, CA). The membrane was probed with baboon MCP-1 cDNA labeled by random priming using a commercial kit (Amersham, Buckinghamshire, U.K.) and [³²P]dCTP. The blot was prehybridized at 42°C for 2 h in 50% formamide, 0.1% SDS, 2× Denhardt’s solution, 5× standard SSPE, and 0.1 mg/ml salmon sperm DNA. Probe (10⁶ cpm/ml) was added to the prehybridization solution, and the blot was hybridized overnight at 42°C. The blot was then washed twice for 10 min each at 55°C in 2× standard saline citrate and 0.1% SDS, and autoradiography was performed with Fuji x-ray film (Tokyo, Japan) and an intensifying screen at −70°C. The MCP-1 probe was then removed by boiling, and the same blot was rehybridized to a cDNA probe encoding for the GAPDH. The intensity of the blots was quantified by densitometric analysis.

Confluent cells were starved for 24 h and preincubated with or without various concentration of BMP-7 (50–200 ng/ml) for 30 min, and then were stimulated with IL-1β (250 pg/ml) for 30 min. Nuclear extracts were prepared from the treated cells that were pelleted (∼500 × g, 10 min, 4°C) and resuspended in hypotonic buffer (10 mM HEPES, pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM DTT, 0.5 mM PMSF, complete protease inhibit mixture tablets) and incubated for 10 min on ice. Cells were lyzed by the addition of Nonidet P-40 (final concentration of 0.1%). After centrifugation, the cytoplasmic extract was collected to measure IκBα protein. The nuclear pellet was resuspended in extraction buffer (20 mM HEPES, pH 7.9, 0.4 mM KCl, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 1 mM PMSF, complete protease inhibit mixture tablets) and incubated for 10 min on ice, and the tubes were centrifuged at 14,000 × g (4°C) for 10 min. The supernatants were stored at −70°C until used. The protein content in nuclear extracts was measured by the Bradford method (Bio-Rad, Hercules, CA). The oligonucleotide containing the NF-κB or AP-1 consensus binding sequence was radiolabeled with [γ-³²P]ATP by T4 polynucleotide kinase (Roche, Mannheim, Germany) for 30 min at room temperature. Unincorporated label was removed with a Chroma spin STE-10 column (Clontech Laboratories, Palo Alto, CA). The binding reaction was performed for 30 min at room temperature in a total volume of 20 μl containing binding buffer (50 mM Tris, pH 7.5, 250 mM NaCl, 2.5 mM EDTA, 20% glycerol, 2.5 mM DTT, and 5 mM MgCl₂), 5 μg of nuclear protein extracts, 1 μl of ³²P-labeled oligonucleotide, and 50 μg/ml poly(dI-dC) (Roche). For competition assay, an unlabeled AP-1 consensus oligonucleotide or AP-1 mutant oligonucleotide was added in 50-fold excess. For supershift analysis, the nuclear extracts were preincubated with the indicated Abs for 30 min at room temperature before binding reaction. The DNA-protein complexes were fractionated by electrophoresis over 6% polyacrylamide gels in 1× Tris-boric acid-EDTA buffer. The gels were dried and exposed to x-ray film for autoradiography.

Confluent cells were starved for 24 h and preincubated with or without various concentration of BMP-7 (50–200 ng/ml) for 30 min, and then were stimulated with IL-1β (250 pg/ml) for 30 min. The treated cells were lyzed for 10 min on ice in lysis buffer (50 mM Tris, pH 7.5, 40 mM NaCl, 1% Triton X-100, 2 mM EDTA, 1 μg/ml leupeptin, 2 mM DTT, and 1 mM PMSF). Lysates were cleared by centrifugation at 14,000 × g (4°C) for 10 min. Total protein was quantified by the Bradford assay. Equal amounts of lysates were fractionated by 10% SDS-PAGE and electrotransferred to Bio-Blot nitrocellulose membranes (Bio-Rad). The membranes were blocked with TBS (pH 7.6)/5% nonfat dry milk/0.05% Tween 20, and were blotted with the indicated Abs at 4°C overnight. The membranes were incubated with 1/1000 diluted HRP-conjugated secondary Ab (Amersham) at room temperature for 1 h and visualized by an ECL kit (Amersham). The membranes were stripped in solution (62.5 mM Tris-HCl, pH 7.5, 20 mM DTT, and 1% SDS) for 5 min at 50°C with agitation and reprobed with anti-pan JNK and anti-actin Abs.

Data were expressed as the mean ± SEM. Statistical comparisons between multiple groups were performed by ANOVA and Bonferroni’s method was applied to control for multiple testing. Statistical comparisons between multiple time points were performed by repeated measure ANOVA, with a minimum value of p < 0.05 considered to represent statistical significance.

MCP-1 protein was measured in the supernatants of mesangial cells stimulated with IL-1β (250 pg/ml) ± BMP-7 (200 ng/ml). MCP-1 was constitutively produced by mesangial cells and was significantly stimulated by IL-1β from 8 to 48 h (p < 0.01). Preincubation of cells with BMP-7 significantly inhibited constitutive (p < 0.05) and IL-1β-induced MCP-1 production from 8 (29%) to 48 h (95%) (p < 0.05; Fig. 1,A). We then measured MCP-1 protein in the supernatants of cells cultured for 24 h in the presence of IL-1β (250 pg/ml) with or without various concentration of BMP-7. BMP-7 inhibited constitutive and IL-1β-induced MCP-1 production in a concentration-dependent manner (Fig. 1 B). BMP-7, at any concentration, did not elevate LDH release above the control values, indicating that BMP-7 was not cytotoxic to cells (data not shown).

Confluent cultures of mesangial cells secreted monocyte chemotactic activity into the medium and IL-1β significantly stimulated monocyte chemotactic activity released from mesangial cells (17.0 ± 0.8 monocytes/high power field vs 30.8 ± 1.0 monocytes/high power field, p < 0.01). BMP-7 significantly inhibited IL-1β-induced monocyte chemotactic activity in a concentration-dependent manner (p < 0.01) and 200 ng/ml BMP-7 completely abolished IL-1β-induced monocyte chemotactic activity. Random migration in response to nonconditioned medium was minimal (Fig. 2).

Cells were preincubated with or without BMP-7 (200 ng/ml) for 30 min and then were incubated with IL-1β (250 pg/ml). Cells were harvested and total RNA was isolated at 4, 8, 16, and 24 h. Northern blot analysis revealed that IL-1β significantly stimulated MCP-1 mRNA expression from 4 to 24 h and BMP-7 inhibited IL-1β-induced MCP-1 mRNA expression in a time-dependent manner from 4 to 24 h. The peak inhibition was seen at 16 h. BMP-7 also inhibited constitutive expression of MCP-1 mRNA (Fig. 3,A). We then preincubated the cells with or without various concentrations of BMP-7 for 30 min and incubated with IL-1β (250 pg/ml) for 16 h. IL-1β increased MCP-1 mRNA expression 3.4-fold. At a dose from 50 to 200 ng/ml, BMP-7 significantly inhibited IL-1β-induced MCP-1 mRNA expression in a concentration-dependent manner, with complete inhibition at 100 and 200 ng/ml (p < 0.01). Furthermore, BMP-7, from 50 to 200 ng/ml, significantly inhibited constitutive MCP-1 mRNA expression (p < 0.01; Fig. 3 B).

The 5′-flanking region of the MCP-1 gene contains multiple AP-1 and NF-κB binding sites (22), suggesting a potential role for both transcription factors in the regulation of MCP-1 expression. IL-1β-induced activation of NF-κB activity leading to increased MCP-1 expression in mesangial cells has been described in various experiments (23, 24). However, it has not been clearly determined whether IL-1β stimulates AP-1 activity and whether IL-1β-induced AP-1 activity triggers MCP-1 expression in mesangial cells. Therefore, we examined AP-1 activity in mesangial cells stimulated with IL-1β by EMSA using AP-1 consensus oligonucleotides. AP-1 activity was significantly increased by IL-1β from 15 min and was maintained up to 240 min (Fig. 4,A). Supershift analysis indicated that the AP-1 complexes induced by IL-1β contained c-Jun/c-Fos proteins, as this complex was remarkably supershifted by anti-c-Jun Abs and anti-c-Fos Abs. Anti-p65 Ab was used as a negative control. The specificity of the AP-1 binding activity was tested using both an unlabeled consensus AP-1 oligonucleotide (cold probe) and a mutant AP-1 oligonucleotide, which has two base changes within the AP-1 binding motif (CA to TG), at the same molar quantity (50-fold excess) for competition. The specific band induced by IL-1β was deleted by preincubation of nuclear extracts with excessive amounts of unlabeled AP-1 oligonucleotides, but not by preincubation of nuclear extracts with mutant AP-1 oligonucleotides (Fig. 4 B).

We examined whether IL-1β-induced activation of AP-1 is responsible for increased MCP-1 production. Cells were preincubated with the c-Jun/AP-1 inhibitor, curcumin, and/or the NF-κB inhibitor, PDTC, for 1 h and then were stimulated by IL-1β for 24 h. MCP-1 protein in the conditioned medium was measured by ELISA. IL-1β-induced MCP-1 production was inhibited by curcumin in a concentration-dependent manner and was completely abolished at a dose of 25 μM. It was also inhibited by a high dose of PDTC. In addition, PDTC inhibition of IL-1β-induced MCP-1 production was much more enhanced by the combination with curcumin, suggesting that AP-1 activation was involved in the regulation of MCP-1 expression (Fig. 4,C). The curcumin and PDTC were not cytotoxic at any concentration we tested, because LDH release was not elevated over control (Fig. 4 D).

To examine the effect of BMP-7 on IL-1β-induced AP-1 activity, cells were preincubated with or without various concentration of BMP-7 for 30 min and then incubated with IL-1β for 30 min. EMSA revealed that BMP-7 inhibited IL-1β-induced AP-1 activity in a concentration-dependent manner. IL-1β-induced AP-1 activity was almost completely abolished by 200 ng/ml BMP-7 (Fig. 5).

We also examined the effect of BMP-7 on IL-1β-induced NF-κB activity. Cells were preincubated with or without various concentration of BMP-7 for 30 min and then incubated with IL-1β for 30 min, because IL-1β-induced NF-κB activity in mesangial cells was peak at 30 min in our previous study. Surprisingly, BMP-7 did not inhibit NF-κB activity at a dose from 50 to 200 ng/ml in five independent experiments. We examined NF-κB inhibitory protein IκBα in the cytoplasmic extracts, which were isolated from the same cells as cells in which we isolated nuclear extracts to measure NF-κB activity. Western blot analysis represented that IκBα was significantly degraded by IL-1β, but was not affected by BMP-7 (Fig. 6). These data indicate that BMP7 does not regulate the degradation of IκBα as well as NF-κB activity.

Because the activation of JNK has been shown to be linked to AP-1 activity, particularly to the phosphorylation of c-Jun (25, 26), we examined the phosphorylation of JNK in the cells stimulated with IL-1β by Western blot analysis using specific anti-phospho-JNK Ab, which can recognize the phosphorylation at Thr¹⁸³ and Thr¹⁸⁵ of JNKs. IL-1β phosphorylated both 46 and 54 kDa JNKs in mesangial cells from 10 to 60 min and the peak phosphorylation was seen at 30 min (Fig. 7 A). IL-1β-induced phosphorylation of JNK, particularly of 54 kDa JNK, was inhibited by BMP-7 in a concentration-dependent manner. IL-1β stimulated the phosphorylation of 54 kDa JNK by 22-fold and 200 ng/ml BMP-7 inhibited IL-1β-induced phosphorylation of 54 kDa JNK by 75%.

In the present study, we demonstrated that BMP-7 inhibited constitutive and IL-1β-induced MCP-1 expression in human mesangial cells, partly through the suppression of JNK activity and AP-1 binding activity. Because glomerular macrophage infiltration plays a direct role in the progression to glomerulosclerosis (27, 28) and is mediated by MCP-1 (1, 2, 3, 4), BMP-7 may have therapeutic potential in the treatment of inflammatory glomerular diseases.

Recently, the therapeutic effect of BMP-7 in renal disease has been demonstrated in a rat model of UUO. In that animal study, BMP-7 treatment effectively prevented infiltration of the interstitium by macrophages, attenuated interstitial fibrosis, and preserved renal function (16). The finding that BMP-7 inhibits TNF-α-induced chemokine expression, including MCP-1, in the proximal tubule epithelial cells (19) could partly explain the anti-inflammatory effect of BMP-7 in UUO. However, little is known about whether BMP-7 regulates MCP- 1 expression in mesangial cells and how BMP-7 regulates MCP-1 expression.

Mesangial cells synthesize IL-1 and have IL-1 receptors. IL-1 induces a variety of biochemical and functional responses in mesangial cells, including chemokine production (29). Most of the biological effects of IL-1 take place in cells following nuclear translocation of NF-κB and AP-1, two nuclear factors common to many IL-1-induced genes. Following IL-1 stimulation, IκB is phosphorylated and rapidly degraded within the proteosome and then translocation of NF-κB to the nucleus is observed. Similar to NF-κB, AP-1 sites are present in the promoter regions of many IL-1-inducible genes. IL-1 increases nuclear binding of c-Jun and c-Fos, the two components of AP-1 in T lymphocytes and cultured hepatocytes (30).

The role of transcription factors NF-κB and AP-1 in the regulation of MCP-1 expression is both in a stimulus-specific and a tissue-specific manner. In mesangial cells, IL-1β-induced MCP-1 expression is mediated by NF-κB (31, 32), proteasome inhibitor-induced MCP-1 expression is mediated by the JNK/AP-1 pathway (33), and angiotensin III-induced MCP-1 expression is mediated by both NF-κB and AP-1 (34). The role of AP-1 activation in the up-regulation of MCP-1 expression has been described in various cells. Activation of AP-1 is required for the induction of MCP-1 by TGF-β in osteoblastic cells (35), and by IL-1β and shear stress in vascular endothelial cells (36, 37). However, it has not been clearly determined whether IL-1β-induced MCP-1 expression is mediated by AP-1 in mesangial cells. In our study, IL-1β stimulated AP-1 activity, both c-Jun and c-Fos components, from 15 to 240 min. Curcumin is known not only to inhibit the DNA binding of c-Jun/AP-1 transcription factor but also to down-regulate c-jun gene transcription (38). Therefore, by using this inhibitor, we explored the role of c-Jun/AP-1 in IL-1β-induced MCP-1 production. When mesangial cells were preincubated with curcumin, IL-1β-induced MCP-1 production was inhibited in a concentration-dependent manner. In addition, a NF-κB inhibitor, PDTC, also blocked IL-1β-induced MCP-1 production. However, dual inhibition by both curcumin and PDTC inhibited IL-1β-induced MCP-1 production much more compared with inhibition by a single inhibitor. These results indicate that AP-1 activation is involved in IL-1β-induced MCP-1 expression. Curcumin is also known to inhibit cytokine-induced NF-κB activation by blocking a signal leading to IκB kinase activity in human myelomonoblastic leukemia cells and intestinal epithelial cells (39, 40). However, our results showing that single inhibition of MCP-1 production by PDTC at high concentration did not reach the level of dual inhibition suggested the role of AP-1 in the regulation of MCP-1 expression. Therefore, blocking AP-1 activity is likely to be linked to the intracellular mechanism of BMP-7 inhibition of MCP-1 expression.

c-Jun contains a docking site for JNKs which phosphorylate Serines 63 and 73 thereby enhancing its transcriptional potential and stability (41). Activation of JNK results in greatly enhanced c-Jun transcriptional activity and induction of the AP-1 target gene (41). Most cells express two JNK isoforms, 45 and 54 kDa in size and termed JNK1 and JNK2, which are highly similar in their modes of regulation (42, 43). JNK is known to be activated by IL-1β in mesangial cells (42, 43). IL-1β-induced JNK activation and proteasome inhibitor-induced activation of the JNK/AP-1 pathway, which triggered MCP-1 expression in mesangial cells, have been demonstrated (33, 44). In our study, IL-1β-induced phosphorylation of JNK was inhibited by BMP-7, suggesting that blocking JNK activation contributed to the inhibition of c-jun activation, AP-1 activation, and finally MCP-1 expression. However, the mechanism of JNK inhibition by BMP-7 remains to be determined.

In summary, our data provide strong evidence that BMP-7 inhibits constitutive and IL-1β-induced MCP-1 expression in human mesangial cells partly by inhibiting JNK activity and subsequent AP-1 activity. This study also provides a rationale to investigate the effect of BMP-7 as a therapeutic in the inflammatory renal diseases.