SMAD3 Functions as a Transcriptional Repressor of Acid-Sensing Ion Channel 3 (ASIC3) in Nucleus Pulposus Cells of the Intervertebral Disc

Abstract

The goal of this investigation was to study the regulation of acid-sensing ion channel (ASIC)3 expression by TGFβ in the nucleus pulposus cells of the intervertebral disc. Analysis of human nucleus pulposus tissue indicated decreased ASIC3 and elevated TGFβ expression in the degenerate state. In a parallel study, treatment of nucleus pulposus cells with TGFβ resulted in decreased expression of ASIC3 mRNA and protein. Suppression of ASIC3 promoter activity was evident when the nucleus pulposus cells were treated with TGFβ or co-transfected with the constitutively active ALK5 or a smad3 construct. On the other hand, co-transfection of dominant negative smad3 or smad7 restored ASIC3 promoter activity. We validated the role of smad3 in controlling ASIC3 expression using cells derived from smad3-null mice. ASIC3 promoter activity in the null cells was 2- to 3-fold higher than the wildtype cells. Moreover, expression of smad3 in null cells decreased ASIC3 promoter activity by almost 50%. Further studies using deletion constructs and trichostatin A treatment showed that the full-length smad3 was necessary, and the suppression involved recruitment of histone deacetylase to the promoter. To determine the mechanism, we evaluated the rat ASIC3 promoter sequence and noted the presence of two smad interacting CAGA box motifs. Gel-shift and supershift analysis indicated that smad3 protein was bound to this motif. Chromatin immunoprecipitation analysis confirmed that smad3 bound both the CAGA elements. Results of these studies clearly show that TGFβ is highly expressed in the degenerate disc and through smad3 serves as a negative regulator of ASIC3 expression.

INTRODUCTION

The intervertebral disc is a specialized tissue that permits motion between vertebrae and accommodates applied compressive forces. Within the disc, secretion of complex matrix macromolecules serves to maintain the high osmotic pressure of the nucleus pulposus. The hyperosmotic status of the nucleus pulposus and the restricted vascular supply imposes metabolic restraints on the cells.(1) As a result, these cells generate almost all of their metabolic energy through the glycolytic pathway.(2,3) One consequence of the reliance on anaerobic metabolism is that the disc pH is low.(4)

The mechanism by which tissues adapt to a low pH is tissue specific. In neural tissues, cells adjust to variations in extracellular pH by regulating the activities of acid-sensing ion channel (ASIC) proteins. These proteins are members of the amiloride-sensitive epithelial Na⁺ channel (ENaC)/degenerin family. Four distinct genes encode six ASIC isoforms.(5–10) These polypeptides form both homomeric and heteromeric functional membrane channels, most likely tetramers, each with distinct electrophysiological characteristics.(11,12) Recent studies by Jahr et al.(13) and Uchiyama et al.(14) showed that skeletal tissues also express members of ASIC family. ASIC3 and ASIC2b are the predominant expressed isoforms in the disc. Of these channel proteins, ASIC3 has been shown to be required for maintenance of a number of physiological functions and implicated in pain transduction associated with ischemic or inflamed tissue acidosis.(15–20) In general, these channel proteins respond to extracellular acidification by regulating transmembrane Na⁺, K⁺, or Ca²⁺ flux.

The Na⁺ transport activity during acute injury is regulated by TGFβ. For example, in lung epithelial cells, TGFβ downregulates αENac expression.(21) It is important to note that our knowledge of the repressive mechanisms by which TGFβ regulates cell function is limited. TGFβ inhibits telomerase activity by regulating the interaction of smad3 with the TERT promoter.(22) However, smad3 suppresses transcription of c-myc by interacting with transcriptional co-repressors, E2F4/5-p107.(23) There is also evidence that TGFβ blocks MyoD and MEF2 activity and represses myogenesis,(24,25) whereas, in osteoblasts and hepatocytes, TGFβ/smad3 recruits class I or class II histone deacetylase (HDAC) to Runx2 or HNF4α, respectively. As a result, in bone cells, there is repression of osteocalcin expression, whereas in liver cells, cholesterol 7α-hydroxylase expression is impaired.(26,27)

In the intervertebral disc, nucleus pulposus cells respond to TGFβ; Peng et al.(28) showed that degenerate, painful human discs contained elevated levels of both TGFβ protein and its receptors. In a parallel study, Nerlich et al.(29) reported TGFβ1 synthetic activity in matrix remodeling areas of the degenerate human nucleus pulposus. These studies were further validated by Murakami et al.(30) and Sobajima et al.,(31) who observed elevated levels of TGFβ in early degenerative discs from old animals and in an animal model of disc injury, respectively. These findings lent considerable strength to the hypothesis that elevated TGFβ expression and secretion is linked to disc injury and repair. Based on this information, we asked the following question: does TGFβ regulate ASIC3 expression in cells of the nucleus pulposus? Results of this study clearly show that this growth factor inhibits ASIC3 expression. Suppression is mediated by smad3, which binds to the ASIC3 promoter through conserved CAGA box motifs. Based on these observations, it is probable that this pathway may provide a protective role for cells of the nucleus pulposus cells, during degeneration or injury.

MATERIALS AND METHODS

Reagents and plasmids

Wildtype (WT) and null Smad3 mouse embryonic fibroblasts (MEFs), Flag-Smad3C (aa 199–424, MH2 domain), Flag-Smad3NL (aa 1–211, MH1 domain and linker), and Flag-Smad3ΔC (aa 1–381, lacks the C-terminal 43 amino acids and functions as DN-Smad3) vectors were provided by Dr Rik Derynck, University of California at San Francisco. Plasmids were kindly provided by Dr Yan Chen, Indiana University School of Medicine (CA-ALK5)(32); Dr Adam Glick, The Pennsylvania State University (Flag-Smad3, Flag-Smad7); and Dr Joan Massague, Memorial Sloan Kettering Cancer Center (3TP-Lux reporter). ASIC3 reporter constructs have been described previously.(14) As an internal transfection control, vector pRL-TK (Promega) containing the Renilla reniformis luciferase gene was used. The amount of transfected plasmid, the pretransfection period after seeding, and the post-transfection period before harvesting have been optimized for rat nucleus pulposus cells using pSV β-galactosidase plasmid (Promega).(2) Well-characterized rat neuronal PC12 cells were used as controls in some experiments. These cells are known to be unresponsive to TGFβ treatment because of the lack of TGFβ type II receptors and hence do not show smad3 activation in response to TGFβ.(33)

Isolation of nucleus pulposus cells

Nucleus pulposus cells were isolated from the rat spine using a method reported earlier(3) and approved by the Institutional Animal Care Committee of Thomas Jefferson University. Briefly, male Wistar rats (250 g) were killed with CO₂, and the lumbar intervertebral discs were removed from the spinal column. These human tissues were collected as surgical waste during spinal surgical procedures. In line with Thomas Jefferson University's Institutional Review Board guidelines, informed consent for sample collection was obtained for each patient. Assessment of the disease state was performed using the modified Thompson grading. The gel-like nucleus pulposus was separated, using a dissecting microscope, and the nucleus pulposus tissue was treated with 0.1% collagenase and 10 U/ml hyaluronidase for 4–6 h. This procedure partially digested the tissue and thereby enhanced the subsequent release of cells trapped in the dense matrix. The partially digested tissue was maintained as an explant in DMEM and 10% FBS supplemented with antibiotics. Nucleus pulposus cells migrated out of the explant after 1 week. When confluent, the cells were lifted using a trypsin (0.25%) EDTA (1 mM) solution and subcultured in 10-cm dishes. These cells were treated with TGFβ (1–10 ng/ml).

Real-time RT-PCR analysis

At the end of TGFβ treatment, total RNA was extracted from nucleus pulposus cells using RNAeasy mini columns (Quiagen). Before elution from the column, RNA was treated with RNase-free DNase I. Total RNA (100 ng) was used as template for real-time PCR analysis. Reactions were set up in microcapillary tubes using 1 μl RNA with 9 μl of a LightCycler FastStart DNA Master SYBR Green I mix (Roche Diagnostics, Indianapolis, IN, USA) to which gene-specific forward and reverse PCR primers were added (ASIC3: NCBI NM_173135, Fwd: 5`-tggcaacggactggagattatgct-3`: 621–644 bp, Rev: 5`-tcatcctggctgtgaatctgcact-3`: 717–740 bp). Each set of samples included a template-free control. PCR reactions were performed in a LightCycler (Roche) according to the manufacturer's instructions. All the primers used were synthesized by Integrated DNA Technologies (Coralville, IA, USA).

Immunofluorescence microscopy

Cells were plated in flat bottom 96-well plates (5000 cells/well) and treated with TGFβ (10 ng/ml) for 6 h or left untreated. After incubation, cells were fixed with 4% paraformaldehyde, permeabilized with 0.2% triton-X 100 in PBS for 10 min, blocked with PBS containing 5% FBS, and incubated with anti-ASIC3 (1:200; Alpha Diagnostics) or anti-smad3 (1:200; Aviva Systems Biology) antibodies at 4°C overnight. As a negative control, cells were reacted with isotype IgG under similar conditions. After washing, the cells were incubated with Alexa fluor-488–conjugated anti-mouse secondary antibody (Molecular Probes, St Louis, MO, USA), at a dilution of 1:50 for 1 h at room temperature. Cells were washed and imaged using a laser scanning confocal microscope (Olympus Fluoview).

Western blotting

Total cell lysates were resolved on 10% SDS-polyacrylamide gels. Proteins were transferred by electroblotting to nitrocellulose membranes (Bio-Rad). The membranes were blocked with 5% nonfat dry milk in TBST (50 mM Tris, pH 7.6, 150 mM NaCl, 0.1% Tween 20) and incubated overnight at 4°C in 3% nonfat dry milk in TBST with the antibodies against ASIC3 (1:500; Alamone Laboratories, Haifa, Israel), and tubulin (1:5000; Santa Cruz). Immunolabeling was detected using the ECL reagent (Amersham Biosciences).

Transfections and dual luciferase assay

Nucleus pulposus cells or WT and null Smad3 MEF were transferred to 24-well plates at a density of 5.0–7.5 × 10⁴ cells/well 1 day before transfection. LipofectAMINE 2000 (Invitrogen) or Transgater (America Pharma Source) was used as a transfection reagent. For each transfection, desired concentrations and combination of plasmids were premixed with the transfection reagent. In some experiments, 24 h after transfection, cells were either treated with TGFβ (10 ng/ml) with or without anti-TGFβ (2 μg/ml) or SB431542 (10 μM), a specific ALK5 inhibitor. The next day, the cells were harvested, and a Dual-Luciferase reporter assay system (Promega) was used for sequential measurements of firefly and Renilla luciferase activities. Quantification of luciferase activities and calculation of relative ratios were carried out using a luminometer (TD-20/20; Turner Designs, Sunnyvale, CA, USA). At least three independent transfections were performed, and all analyses were carried out in triplicate.

Electrophoretic mobility shift assays

Nuclear extracts were prepared using the CellLytic NuCLEAR extraction kit (Sigma-Aldrich, St Louis, MO, USA). Electromobility shift assays were performed as previously described.(34) Briefly, the binding reaction was carried out in 12.5 mM HEPES, pH 7.9, 50–100 mM NaCl, 5% glycerol, 0.5 mg/ml BSA, 1–2 μg poly-dIdC, 0.1 mM EDTA, and 0.1 mM DTT, using 50 fmol of biotin-end-labeled double-stranded oligonucleotide (top strand sequence: 5`-taactgcatgaccagacaagcatttgtact-3` [NCBI AF527175, 2791–2817 bp] for DistalCAGA-ASIC3 and 5`-cctacccccccagactgctctgctgcc-3` [NCBI AF527175, 391–420 bp] for ProximalCAGA-ASIC3 and 10 μg of nuclear protein. After incubation for 45 min at room temperature, extracts were loaded onto 5% acrylamide-0.5× Tris-borate-EDTA gels, electrophoresed at 130 V for 1 h, and transferred onto a positively charged nylon membrane (Hybond-N⁺; Pierce) in 0.5× Tris borate/EDTA at 100 V for 45 min. To perform supershift experiments, 1–2 μl (1 mg/ml) of anti-Smad3 antibody (Aviva Systems Biology) was added to the binding reaction for 25 min at room temperature, before the addition of the labeled DNA probe. After addition of the DNA probe, the reactants were incubated together for an additional 20 min at room temperature. Transferred DNAs were cross-linked to the membrane at 120 mJ/cm² and detected using horseradish peroxidase–conjugated streptavidin, according to the manufacturer's instructions (LightShift Chemiluminescent EMSA kit; Pierce).

Chromatin immunoprecipitation assay

Chromatin immunoprecipitation (ChIP) assays were performed as described elsewhere.(35) The DNA–protein complexes in cells were cross-linked by adding formaldehyde and sheared to size between 200 and 1000 bp in lysis buffer containing protease inhibitors. Lysate was incubated with anti-smad3 antibody (Aviva Systems) or preimmune rabbit IgG, overnight at 4°C. The immune complexes were washed and eluted, and cross-links were reversed at 65°C for 6 h in elution buffer; the RNA was removed by RNase treatment. Co-precipitated DNA was extracted with phenol–chloroform and reprecipitated. PCR analysis was performed using the following primer sequences in ASIC3 promoter (NCBI AF527175): distal site, fwd: 5`-aggtcactctcccactgagaatcca-3` (301–325 bp) and rev: 5`-gtcctctgcaccaatcactcctgaa-3` (449–473 bp); proximal site, fwd: 5`-ctctagtctctgacaacagtgactc-3` (2720–2744 bp), and rev: 5`-actgagctagagctgcggcagagga-3` (2846–2870 bp). To further confirm the identity of the PCR products, they were cloned into a pCR2.1 vector and sequenced.

Statistical analysis

All measurements were performed in triplicate; data are presented as mean ± SD. Differences between groups were analyzed by Student's t-test; *p< 0.05.

RESULTS

Nucleus pulposus cells were maintained in culture for 24 h and treated with an antibody against ASIC3. Figure 1A shows that ASIC3 is expressed by the nucleus pulposus cells in culture. We next examined the effect of TGFβ on expression of ASIC3. When the nucleus pulposus cells are treated with TGFβ, there is suppression in ASIC3 mRNA expression (Fig. 1B). Moreover, TGFβ treatment results in decreased expression of ASIC3 protein (Figs. 1C and 1D). At a concentration of 10 ng/ml, there is suppression of both the 60- and 90-kDa isoform of the ASIC3 protein (Fig. 1C). In a parallel study, we analyzed human degenerate tissue by RT-PCR. Figure 1E shows that degenerate nucleus pulposus evidence decreased expression of ASIC3 mRNA compared with a normal control tissue (grade 1). In contrast, expression of TGFβ mRNA was highly elevated in degenerate tissue.

FIG. 1.

Effect of TGFβ on ASIC-3 expression in nucleus pulposus cells. (A) Immunofluorescent detection of ASIC3 in nucleus pulposus cells. Cells were treated with an antibody to ASIC3, and cell nuclei were stained with propidium iodide (PI). NP cells showed strong expression of ASIC3. Magnification, ×20. (B and C) Cells were treated with TGFβ (1–10 ng/ml) for 24–72 h, and ASIC3 expression was analyzed. (B) Real-time RT-PCR analysis of ASIC3 indicates decreased ASIC3 mRNA expression after treatment with TGFβ. (C) Western blots analysis showed a dose-dependent reduction in ASIC3 protein expression in TGFβ-treated NP cells. (D) Multiple Western blots of ASIC3 quantified by densitometric analysis. Note, NP cells exhibited a suppression of ASIC3 levels when TGFβ was used at a concentration of 10 ng/ml. (E) RT-PCR analysis of normal (N) and Thompson graded (grade 3, G3; and grade 4, G4) degenerate human nucleus pulposus tissue from lumbar discs. Note the high level of expression of ASIC3 in normal tissue that is significantly decreased in degenerate tissue. Degenerate tissue evidenced a higher expression of TGFβ mRNA than normal control tissue. Quantitative data represents mean ± SD of three independent experiments (n = 3); *p < 0.05; ns, nonsignificant.

To further study the regulation of ASIC3, we measured ASIC3 promoter activity in the presence of exogenous TGFβ; for this study, we used a 2.8-kb rat ASIC3 promoter luciferase construct (pASIC3D) and a shorter promoter fragment corresponding to the proximal promoter domain (pASIC3P; Fig. 2A). Treatment of nucleus pulposus cells with TGFβ results in a significant suppression of basal activity of the full-length and proximal promoter construct (Fig. 2B). On the other hand, when PC12 cells are treated with TGFβ, no change in the activity of the ASIC3 promoter is observed (Fig. 2C). In a parallel study, we measured the effect of TGFβ on ASIC3 promoter activity in the presence of an antibody against TGF, or a specific inhibitor of ALK5, SB431542. Figure 2D shows that, when the anti-TGFβ antibody is present, TGFβ is unable to suppress ASIC3 promoter activity. Similarly, the suppressive action of TGFβ on ASIC3 promoter activity is blocked when the nucleus pulposus cells are treated with SB431542 (Fig. 2D). We monitored activation of 3TP-Lux, a TGFβ-responsive reporter in both nucleus pulposus and PC12 cells (Fig. 2E). After TGFβ treatment, nucleus pulposus cells exhibit a 2.5-fold induction in 3TP-Lux activity, whereas, as expected, PC12 cells exhibit a minimal change in activity based on the fact that they do not express TGFβ type II receptors. In a separate experiment, forced expression of constitutively active (CA)-ALK5, a type I TGF-βR in nucleus pulposus cells, also resulted in a 40% reduction in basal activity of the ASIC3 promoter construct (Fig. 2F).

FIG. 2.

Effect of TGFβ in ASIC3 promoter activity. (A) Cartoon showing map of successive PCR generated 5` deletion constructs of the rat ASIC3 promoter. The ASIC3 promoter contains three distinct domains: proximal, middle, and a distal activating domain. The ASIC3-D construct is a 2925-bp fragment containing 2831 bp of the upstream ASIC3 promoter sequence linked to 94 bp of exon 1 (i.e., −2831 to +94), whereas the ASIC3-P constructs contains a 1065-bp (−971 to + 94) fragment. (B and C) NP or PC12 cells were transfected with ASIC-3 reporter constructs along with pRL-TK vector. Cells were treated with TGFβ (10 ng/ml) for 24 h, and luciferase reporter activity was measured. NP cells showed significant suppression in activities of both ASIC-3 constructs (B). PC12 cells did not show a TGFβ-mediated decrease in pASIC-3D activity (D) NP cells transfected with ASIC3 reporter were treated with anti-TGFβ antibody (10 μg/ml) or SB431542 (10 μM) along with TGFβ3. Both anti-TGFβ antibody and SB431542 reversed the suppressive effects of TGFβ on ASIC3 promoter activity. (E) TGFβ reactivity of NP and PC12 cells assessed by measuring activation of the 3TP-Lux reporter. Note, only NP cells showed induction in 3TP-Lux activity after TGFβ treatment. (F) NP cells were transfected with an ASIC3 reporter along with constitutively active type I TGFßR (ALK5) construct or empty vector pcDNA3. Constitutive ALK5 signaling suppressed ASIC3 promoter activity in NP cells. Values shown are mean ± SD of three independent experiments performed in triplicate. *p < 0.05.

Nucleus pulposus cells under basal conditions expressed smad3 protein and, when treated with TGFβ, showed activation and nuclear accumulation of this protein (Fig. 3A). We determined whether TGFβ modulates ASIC3 promoter function through smad3 signaling. Nucleus pulposus and PC12 cells were transfected with a full-length WT smad3 expression plasmid. Figure 3B shows that forced expression of smad3 results in 60% suppression of ASIC3 basal promoter activity in nucleus pulposus cells. A similar level of ASIC3 promoter activity suppression is evident in PC12 cells (Fig. 3C). By measuring activation of the 3TP-Lux reporter, we assessed the transcriptional activity of the transfected smad3 construct. When co-transfected into both nucleus pulposus and PC12 cells, there is increased 3TP-Lux reporter activity (Fig. 3D). Moreover, when nucleus pulposus cells are transfected with the dominant negative (DN)-smad3 construct, the TGFβ-mediated inhibition of ASIC3 promoter activity is reversed (Fig. 3E). Similarly, transfection with a WT-smad7 expression vector results in maintenance of the ASIC3 promoter activity after TGFβ treatment (Fig. 3F); no change in promoter activity is observed when smad7 was co-transfected alone (data not shown).

FIG. 3.

Smad3 regulation of ASIC3 promoter activity. (A) Nucleus pulposus cells were treated with TGFβ for 6 h and probed with anti-Smad3 antibody. Note nuclear accumulation of smad3 protein in treated cells (arrows). (B and C) ASIC3 reporter construct along with full-length Smad3 construct, or empty backbone pcDNA3.1, was transfected into (B) NP cells and (C) PC12 cells, and luciferase activity was measured. Expression of smad3 resulted in significant suppression of ASIC3 promoter activity in both cell types. (D) Transcriptional activity of transfected Smad3 was measured by determining 3TP-Lux reporter activity. Note, smad3 induced 3TP-Lux activation in both NP and PC12 cells. (E) NP cells were transfected with ASIC3 reporter along with DN-Smad3 construct or empty backbone vector pRK5F and treated with TGFβ. Note, expression of DN-Smad3 rescued cells from the suppressive effect of TGFβ on ASIC3 promoter activity. (F) NP cells were transfected with full-length Smad7 construct or empty backbone vector pcDNA3 along with ASIC3 reporter, and luciferase activity was measured after TGFβ treatment. Smad7 expression preserved ASIC3 promoter activity in the presence of TGFβ. Values shown are mean ± SD of three independent experiments performed in triplicate. *p < 0.05.

The role of TGFβ and smad3 signaling in the regulation of ASIC3 promoter activity was further examined using cells derived from smad3-null (^−/−) mice. Figure 4A shows that the basal ASIC3 promoter activity in null cells was 2- to 3-fold higher than in the wildtype cells. Moreover, forced expression of smad3 expression vector in smad3-null cells results in a significant decrease in basal ASIC3 promoter activity (Fig. 4B). Using domain deletion expression plasmids, we studied whether suppression was conferred by a specific domain of the smad3 protein. Figure 5A shows that, when either the MH1 or MH2 domain of smad3 is absent, there is no suppression of ASIC3 promoter activity; suppression is evident only when a full-length WT-Smad3 construct is used. We determined whether recruitment of HDAC to the ASIC3 promoter mediates smad3 suppression. Figure 5B shows that when nucleus pulposus cells are treated with trichostatin A (TSA), a potent inhibitor of HDAC, smad3-mediated suppression of ASIC3 promoter is reversed.

FIG. 4.

Effect of Smad3 deletion on ASIC-3 promoter activity. (A) Smad3 wildtype (WT: ^+/+) and null (^−/−) MEF were transfected with ASIC-3 reporter constructs along with pRL-TK control vector, and luciferase reporter activity was measured. Null cells showed a significant increase in basal ASIC-3 reporter activity compared with WT cells. (B) Smad3-null cells were transfected with ASIC-3 reporter construct along with full-length Smad3 construct or empty backbone vector pCDNA3.1, and luciferase activity was measured. Forced expression of Smad3 in null cells resulted in a significant suppression in ASIC-3 promoter activity. Values shown are mean ± SD of three independent experiments. *p < 0.05.

FIG. 5.

Wildtype Smad3 is required for regulation of ASIC3 promoter activity. ASIC3 reporter construct along with wildtype (WT)-Smad3 construct or truncated Smad3-encoding plasmids; Smad3NL (MH1 domain and linker) or Smad3C (MH2 domain); or empty vector was transfected into NP cells, and luciferase activity was measured. Expression of WT-Smad3 resulted in suppression, whereas plasmids encoding truncated Smad3 protein did not suppress ASIC3 promoter activity. (B) NP cells were transfected with ASIC3 reporter along with smad3 construct, or empty vector, and treated with trichostatin A (TSA; 100 nM) or vehicle (DMSO) for 24 h. Treatment of TSA restored expression of ASIC3 in the presence of smad3. Values shown are mean ± SD of three independent experiments performed in triplicate. *p < 0.05.

Analysis of the rat ASIC3 promoter indicates that there is conserved CAGA box elements at −2429 and −30 bp from the transcription start site (Fig. 6A). An electromobility gel shift assay, using an oligonucleotide probe that contained the wildtype CAGA motif at −30 bp, was performed to examine its functional interaction with smad3 protein. Figure 6B shows that there is a functional binding of the smad3 protein to the CAGA probe. Moreover, the level of binding is enhanced when cells are treated with TGFβ. The specificity of this binding reaction was confirmed by including the anti-smad3 antibody in the binding reaction. In the presence of the antibody, a supershift is observed (Fig. 6B). We used the ChIP assay to further validate that there is a functional interaction of the smad3 protein with the CAGA motifs of the rat ASIC3 promoter (Fig. 6C). Smad3 interaction with both the proximal (ASIC3Prox-CAGA) and distal (ASIC3Dist-CAGA) CAGA motifs of genomic DNA of the nucleus pulposus cells is evident under basal conditions and when cells are treated with TGFβ. Moreover, binding of smad3 to the proximal CAGA motif is enhanced after TGFβ treatment (Fig. 6C). As expected, when an isotype antibody is used in place of the anti-smad3 antibody, PCR analysis, using primer pairs, spanning both CAGA motifs, indicates that amplicons are not formed.

FIG. 6.

Interaction of Smad3 with ASIC-3 promoter. (A) DNA sequence of the promoter region of the rat ASIC3 gene. The CAGA consensus sequence is marked in bold and underlined. The ASIC3 promoter contains two conserved CAGA motifs at −30 and −2429 bp from the transcription start site. (B) Electromobility shift assay to examine functional binding of smad3 to CAGA motif in the rat ASIC3 gene promoter. An oligonucleotide probe containing the CAGA motif (−30 bp) in the rat ASIC3 promoter was incubated with nuclear extracts from nucleus pulposus cells treated with TGFβ and binding was detected using chemiluminescence. Note, there was induction of binding 6 h after TGFβ treatment. Specificity of binding was confirmed by inclusion of a probe containing mutation in the CAGA site (MT probe) or anti-smad3 antibody in the binding reaction. The binding signal is significantly diminished when a mutant probe was used, whereas a supershift of the band was seen in presence of anti-smad3 antibody. (C) ChIP assay was used to examine the interaction of Smad3 with the ASIC3 promoter. PCR amplification was performed using primers pairs that encompass both distal (Dist)-CAGA and proximal (Prox)-CAGA sequences of the ASIC3 promoter. Note, use of smad3 antibody resulted in generation of PCR amplicons containing both distal HRE and proximal HRE. Detectable binding was observed under basal conditions, which was significantly increased after TGFβ treatment. Use of isotype antibody control did not result in the formation of a PCR product.

DISCUSSION

The experiments described in this study confirmed that ASIC3, a pH-sensitive sodium channel, was expressed by nucleus pulposus cells in the hydrodynamically stressed, hyperosmolar microenvironment of the intervertebral disc. We showed for the first time that ASIC3 expression was negatively regulated by TGFβ, a pleiotropic cytokine. TGFβ suppression was mediated by smad3, which binds to the ASIC3 promoter through conserved CAGA box motifs. From a disease viewpoint, TGF is highly expressed by nucleus pulposus cells, and the levels of this cytokine are raised during disc degeneration. Accordingly, an increase in TGFβ levels, resulting from inflammatory or degenerative stimuli, would be expected to downregulate ASIC channel expression and protect cells of the nucleus pulposus from changes in Na⁺ flux linked to proteoglycan catabolism.(28–31)

The effect of TGFβ on ASIC3 was evaluated using RT-PCR and Western blot analysis. Because TGFβ caused suppression of ASIC3 mRNA levels and inhibition of both ASIC3 isoforms, it was inferred that regulation was mediated at the transcriptional level. Transient transfection assays using promoter constructs showed that TGFβ suppressed ASIC3 promoter activity. The specificity of TGFβ-mediated suppression of ASIC3 expression was confirmed using a pharmacological inhibitor, an anti-TGFβ antibody, and transfections with a constitutively active form of type I TGFβ receptor (CA-ALK5). For these studies, PC12 cells served as a negative control; these cells lack TGFβ type II receptors and hence do not respond to TGFβ.(33) Whereas results of these studies confirmed that TGFβ suppressed ASIC3 expression in nucleus pulposus cells, it is probable that these interactions are not confined to cells of the intervertebral disc. TGFβ is a major regulator of cell activity, whereas ASIC3 is functional in a variety of different cell types, especially those whose membrane function is dependent on maintenance of steep proton gradients. Protons directly gate depolarization of ASIC3 channels and promote Na⁺ entry into cells. Our results suggest that, if TGFβ signaling is mediated through ALK5, there is likely be a downregulation of ASIC3 and reduced ability to respond to changes in H⁺ and Na⁺ concentrations. In this way, in the hyperosmotic environment of the disc, cells of the nucleus pulposus would be protected from fluxes in Na⁺ brought about by degradation of the extracellular matrix.

To define the mechanism by which TGFβ influences the activity of ASIC3 promoter, smad3 gain and loss of function experiments were performed. Overexpression of the full-length smad3 construct in nucleus pulposus and PC12 cells, in the absence of exogenously added TGFβ, suppressed ASIC3 promoter activity. Smad3 suppressed ASIC3 reporter activity in nucleus pulposus and the TGFβ unresponsive PC12 cells. These results directly support the hypothesis that smad3 transduced the effect of TGFβ by repressing ASIC3 activity. Further support for this hypothesis was from studies performed using DN-smad3 and smad7 plasmids.(36) One direct test of the hypothesis that ASIC3 activity was influenced by smad3 was performed using smad3-null MEFs. We noted that, in the absence of smad3, basal ASIC3 promoter activity was raised. Finally, we noted that, in these null cells, overexpression of smad3 caused a decrease in ASIC3 promoter activity. Based on these gain and loss of function experiments, it is evident that smad3 can mimic the effects of TGFβ action on ASIC3 promoter function. Results of these studies lend direct support to the notion that smad3 is a functional repressor of ASIC3 promoter activity, and as such, its regulation may control ASIC3 repression in nucleus pulposus cells in vivo.

A clue to the mechanism by which smad3 regulated ASIC3 expression was forthcoming from promoter analysis. We found that the ASIC3 promoter contained two conserved CAGA motifs: a proximal motif at −30 bp and a distal motif at −2429 bp. Constructs encompassing both motifs indicated that they were equally active. The gel shift and ChIP assays further confirmed the active nature of these motifs and their individual contributions to the regulation of ASIC3 promoter activity. Noteworthy, results of the ChIP assay suggested increased binding of smad3 to the proximal motif. This may have important implications in regulating transcription from a CCAAT box corresponding to an inverted nonconsensus box at −62 bp from the transcription start site.(37) Interestingly, it was shown that in most TATA-less promoters such as ASIC3 with an inverted CCAAT box, the functional CCAAT site is present approximately at −63 ± 29 relative to the transcription start site.(38) In line with observations concerning smad3 functional repression of Runx2 HNF4b activity and c-Myc promoter function, TSA treatment provided further insight into regulatory mechanism. These studies suggest that, at the CAGA sites of the ASIC3 promoter, smad3 may recruit HDAC, a known transcriptional repressor.(23,26,27) Accordingly, smad3 would act to suppress the activity of transcription factors that regulate ASIC3 transcription through the CCAAT box.

It should be acknowledged that, whereas the goal of this study was to determine how a critical pleiotropic factor, TGFβ, modulated the expression of ASIC3, the detailed functional role of this proton-activated Na⁺ channel in disc cells is not understood. Our previous studies showed that activities of amiloride-sensitive Na⁺ channels, including ASIC3, are required for maintenance of disc cell survival in a hypertonic and acidic environment.(14) Results of this study build on this finding and suggest a new role for ASIC3 during intervertebral disc degeneration: protection against changes in ion concentrations. In development of the disease state, there is decreased water binding caused by loss of sulfated glycosaminoglycans.(39) When this occurs, the bound counter ion, mainly Na⁺, would be expected to accumulate within the disc, influencing the osmotic properties of the nucleus pulposus tissue and the ion activity product of the fluid phase of the extracellular matrix. Under such circumstances, TGFβ expression would be activated by the disc cells, and by regulating Na⁺/H⁺ flux, ASIC3 would serve to maintain cell viability and matrix production.(28–31) In this way, possibly ionotrophic degradatory events mediated by the change in extracellular ion concentrations would be obviated and cell function preserved. From this perspective, nucleus pulposus cells are prepared to survive environmental changes mediated by the inflammatory insult. Further studies using an animal model of disc degeneration, or degenerative human tissue samples, will be needed to test this hypothesis.