Activation of TonEBP by calcium controls {beta}1,3-glucuronosyltransferase-I expression, a key regulator of glycosaminoglycan synthesis in cells of the intervertebral disc

Abstract

The goal of this investigation was to study the expression and regulation of beta1,3-Glucuronosyltransferase-I (GlcAT-I), a key enzyme regulating GAG synthesis in cells of the intervertebral disc. There was a robust expression of GlcAT-I in the nucleus pulposus in vivo. Treatment with the calcium ionophore ionomycin resulted in increased GlcAT-I expression, whereas GlcAT-I promoter constructs lacking TonE site or a mutant TonE were unresponsive to the ionophore. Experiments using TonEBP and DN-TonEBP constructs showed that TonEBP positively regulated GlcAT-I promoter activity. ChIP analysis confirmed binding of TonEBP to the promoter. We further validated the role of TonEBP in controlling GlcAT-I expression using mouse embryo fibroblasts from TonEBP null mice. GlcAT-I promoter activity in null cells was significantly lower than the wild type cells. In contrast to wild type cells, treatment with ionomycin failed to increase GlcAT-I promoter activity in null cells. We then investigated if calcineurin (Cn)-NFAT signaling played a regulatory role in GlcAT-I expression. Inhibition of Cn following ionomycin treatment did not block GlcAT-I and tauT, a TonEBP-responsive reporter activity. GlcAT-I promoter activity was suppressed by co-expression of Cn, NFAT2, NFAT3, and NFAT4. Moreover, following ionomycin treatment, fibroblasts from CnAalpha and CnAbeta null mice exhibited robust induction in GlcAT-I promoter activity compared with wild type cells. Results of these studies demonstrate that calcium regulates GlcAT-I expression in cells of the nucleus pulposus through a signaling network comprising both activator and suppressor molecules. The results suggest that by controlling both GAG and aggrecan synthesis, disc cells can autoregulate their osmotic environment and accommodate mechanical loading.

FIGURE 1.

Saggital and coronal sections of disc tissue from neonatal (A) and mature (C and E) rat spines stained with an antibody against GlcAT-I or stained with hematoxylin and eosin and Alcian blue (B, D, and F). Note that nucleus pulposus cells in the neonatal (A) as well as skeletally mature disc cells (C and E) express GlcAT-I protein; much of the staining is localized to the cytosol and plasma membrane (A and C; arrows). Furthermore, inner annulus fibrosus cells localized in amorphous Alcian blue-positive matrix (F; arrow) express GlcAT-I protein (E; arrow). Isotype and secondary antibody controls were negative (not shown). Magnification was ×20. G, Western blot analysis of GlcAT-I expression by nucleus pulposus (NP) and annulus fibrosus (AF) tissue and cultured cells. Note, the expression of the 43-kDa GlcAT-I band in tissue extracts. The native nucleus pulposus tissue and cells in culture expressed higher GlcAT-I protein levels than the annulus tissue and cells. H, real time reverse transcription-PCR analysis of GlcAT-I expression by cells treated with ionomycin (1 μm) and PMA (100 ng) (I + P) for 4–24 h. There was a time-dependent change in mRNA expression following treatment. At 4 h, GlcAT-I expression was suppressed; at 24 h, there was increased expression of the gene. I, immunofluorescent analysis of nucleus pulposus cells treated with ionomycin and PMA. Cells showed increased GlcAT-I expression 24 h after the treatment. J, Western blot of nucleus pulposus cells treated with ionomycin and PMA. Note the increased GlcAT-I expression 24 h after treatment.

FIGURE 2.

GlcAT-I promoter contains TonEBP and NFAT binding motifs. A, promoter organization of the human GlcAT-I gene. The transcription start site is marked as +1. The TonE site is shown as a flattened circle, Spls are indicted as ovals, and the NFAT binding motifs are shown as rectangles. B, DNA sequence of the promoter region of the human GlcAT-I gene. TonE (TTTCCA) and NFAT (TTTCC or GGAAA) consensus sequences are marked in boldface type and underlined. The arrows indicate the starting location of the primers used to generate promoter constructs. The transcription start site is marked as +1; ATG marks the translation start site. Spl sites are underlined and lie within first 100 bases. C, schematic diagram showing a map of successive PCR-generated 5′ deletion constructs of the human GlcAT-I promoter. The GlcAT-I-D construct consists of a 1,231-bp fragment containing 1,170 bp of the upstream GlcAT-I promoter sequence linked to 61 bp of exon 1 (i.e. –1170/+61), whereas GlcAT-I-M and GlcAT-I-P constructs contain a 335-bp fragment (–274/+61) and a 184-bp fragment (–123/+61), respectively. D, basal activities of GlcAT-I promoter constructs relative to full-length construct GlcAT-I-D in nucleus pulposus cells. Cells showed maximal luciferase activity for the GlcAT-I-D construct, whereas the shortest construct, GlcAT-I-P, showed minimal activity. Values shown are mean ± S.D. of three independent experiments. *, p < 0.05.

FIGURE 3.

Calcium regulates GlcAT-I promoter activity through TonEBP. A, GlcAT-I-D or GlcAT-I-P reporter activity measured following ionomycin and PMA (I + P) treatment. Treatment resulted in induction in GlcAT-I-D but not GlcAT-I-P reporter activity. B, effect of BAPTA-AM (10 μm) on GlcAT-I-D reporter activity. The calcium chelator completely blocks ionomycin-mediated activity but not basal activity of the GlcAT-I-D reporter in nucleus pulposus cells. C and D, effect of ionomycin treatment on cells transfected with GlcAT-I-D reporter construct along with DN-TonEBP or empty backbone FLAG-CMV2. Note that the expression of DN-TonEBP resulted in a complete suppression of ionomycin-mediated induction in GlcAT-I promoter activity. In addition, when TonEBP function was blocked, there was suppression of basal GlcAT-I-D activity. E, effect of TonEBP on GlcAT-I promoter activity. When pTonEBP was co-expressed, there was a dose-dependent increase in GlcAT-I reporter activity. F, GlcAT-I promoter activity of TonEBP/NFAT5 null and wild type cells. Null cells evidenced decreased basal activity of reporter compared with wild type cells. G, effect of co-expression of TonEBP and GlcAT-D in null cells. When co-expressed, TonEBP increased GlcAT-I-D reporter activity. Values shown are mean ± S.D. of three independent experiments performed in triplicate. *, p < 0.05.

FIGURE 4.

Ionomycin mediated induction of GlcAT-I promoter activity requires TonEBP binding to TonE. Effect of a 4-bp mutation introduced into the TonE motif of the GlcAT-I-D reporter plasmid. Nucleus pulposus cells were transfected with wild type GlcAT-I-D (W-GlcAT-I) or mutant GlcAT-I-D (M-GlcAT-I) reporter plasmids, and the induction of the luciferase activity was determined following ionomycin treatment. Treatment caused an induction of wild type reporter activity, whereas the mutant reporter failed to increase activity. Values shown are mean ± S.D. of three independent experiments. *, p < 0.05. B, interaction of TonEBP with the GlcAT-I promoter measured using a chromatin immunoprecipitation assay. COS7 cells were transfected with GlcAT-I-D along with either FLAG-TonEBP or FLAG-CMV2 empty vector. PCR amplification was performed using primer pairs that encompass TonE sequences of the GlcAT-I promoter. The use of anti-FLAG antibody resulted in generation of a PCR amplicon containing TonE only when FLAG-TonEBP was present. The addition of increasing amounts of FLAG-TonEBP vector evidenced enhanced binding to TonE. Pull-down using anti-FLAG antibody did not result in the formation of a PCR product when cells received empty FLAG-CMV2 vector.

FIGURE 5.

Effect of calcium ions on TonEBP expression. A, nucleus pulposus cells were treated with the calcium ionophore ionomycin (I; 1 μm), along with PMA (P; 100 ng), and TonEBP expression was measured. Ionomycin treatment resulted in significant increase in TonEBP mRNA expression. B and C, immunofluorescence and Western blot analysis of cells as treated in A. note the increase in TonEBP protein after ionomycin treatment. The addition of Cn inhibitors FK506 (10 ng/ml) and cyclosporine A (CsA; 1 μg/ml) did not suppress synthesis of TonEBP protein. When nucleus pulposus cells were cultured under hyperosmotic conditions (450 mosmol/kg), there was high induction in TonEBP, whereas TonEBP protein was undetectable in TonEBP null cells. D and E, reporter activity of nucleus pulposus cells transfected with wild type (D) or mutant (E) tauT reporter and treated with ionomycin and PMA with or without FK506 and cyclosporine. Note that treatment with ionomycin increased the activity of the WT but not the TonE-MT tauT reporter. FK506 and CsA did not inhibit ionomycin-mediated increase in tauT-WT promoter activity. As expected under hyperosmotic conditions, there is a robust induction in activity of tauT-WT but not tauT-MT reporter. F, induction of the activity of 3×NFAT reporter in NP cells following ionomycin and PMA treatment. The reporter is highly induced, whereas the addition of FK506 and cyclosporine completely blocks activation, indicating a requirement for Cn in this process. Values shown are mean ± S.D. of three independent experiments performed in triplicate. *, p < 0.05.

FIGURE 6.

Ionomycin mediates GlcAT-I promoter activation, although TonEBP is independent of calcineurin. Nucleus pulposus cells (A), TonEBP/NFAT5 wild type (B), and TonEBP/NFAT5 null (C) MEFs were transfected with GlcAT-I-D reporter, and activity was measured following treatment with ionomycin with or without FK506 and cyclosporine A. Unlike TonEBP/NFAT5, null MEFs GlcAT-I-D reporter activity was induced in both nucleus pulposus and NFAT5 wild type cells. Calcineurin inhibitors did not suppress ionomycin-induced or basal activity of GlcAT-I reporter in any of the cell types. D, effect of ionomycin treatment on GlcAT-I-P reporter activity. The reporter was nonresponsive to ionomycin treatment in both TonEBP/NFAT5 wild type and null MEFs. Values shown are mean ± S.D. of three independent experiments performed in triplicate. *, p < 0.05.

FIGURE 7.

Calcineurin suppresses GlcAT-I promoter function. A, the GlcAT-I reporter activity following cotransfection with calcineurin A/B constructs or empty vector pBJ5 in nucleus pulposus cells. Co-expression of calcineurin subunits suppressed GlcAT-I reporter activity in transfected cells. B–F, reporter activity of medullary fibroblasts derived from CnA WT or CnAα or CnAβ null (–/–) mice transfected with GlcAT-I or tauT (WT or MT) reporter and treated with ionomycin and PMA with or without FK506 and CsA. Ionomycin did not increase GlcAT-I reporter activity in wild type cells (B), but a robust induction was observed in both the CnAα or CnAβ null cells (C). A similar high induction in the activity of tauT-WT reporter was observed in CnAα or CnAβ null cells; a relatively small inductive effect was seen in wild type cells, which remained constant after the addition of calcineurin inhibitors FK506 and CsA. Neither wild type (D) nor the null cells (E and F) displayed induction in tauT-MT reporter activity. Values shown are mean ± S.D. of three independent experiments performed in triplicate. *, p < 0.05.

FIGURE 8.

GlcAT-I promoter activity suppressed by NFAT signaling in nucleus pulposus cells. Effect of NFAT vectors (NFAT1, -2, -3, and -4) and/or CnA and CnB on GlcAT-I-D reporter activity of nucleus pulposus cells. A, NFAT1 alone did not affect GlcAT-I reporter activity. When calcineurin was added alone or together with NFAT1, there was a suppression in reporter activity. CA-NFAT2 (B), NFAT3 (C), or NFAT4 (D) alone or when added with CnA/B significantly suppressed the GlcAT-I-D reporter activity in nucleus pulposus cells. E, nucleus pulposus cells were transfected with 3×NFAT luciferase construct with or without NFAT1, CA-NFAT2, NFAT3, or NFAT4, and reporter activity was measured. Co-expression of NFAT1 to -4 resulted in a significant increase in activity of 3×NFAT reporter plasmid, indicating functionality of expressed proteins. Values shown are mean ± S.D., of three independent experiments. *, p < 0.05; ns, nonsignificant. F, activation and nuclear localization of NFAT2/3/4 following ionomycin treatment of nucleus pulposus cells.

FIGURE 9.

TonEBP and Spl are both required for calcium-mediated regulation of GlcAT-I promoter activation in nucleus pulposus cells. Cells were transfected with GlcAT-I-D (A) or GlcAT-I-P (B) promoter constructs and treated with ionomycin with or without WP631 (50–100 nm), an Spl inhibitor. Note that suppression of Spl function resulted in complete loss of ionomycin-mediated induction of GlcAT-I reporter. B, ionomycin failed to induce activity of GlcAT-I-P construct lacking the TonE motif in nucleus pulposus cells; WP631 has no effect on the activity of this reporter. Values shown are mean ± S.D. of three independent experiments. *, p < 0.05; ns, nonsignificant.