Thyroid Hormone Regulates Heparan Sulfate Proteoglycan Expression in the Growth Plate
http://www.100md.com
《内分泌学杂志》
Molecular Endocrinology Group, Division of Medicine and Medical Research Council Clinical Sciences Centre, Faculty of Medicine (J.H.D.B., R.S., G.R.W.), Imperial College London, Hammersmith Hospital, London W12 0NN, United Kingdom
Biomateriaux et Reparation Tissulaire Institut National de la Sante et de la Recherche Medicale U-443 (O.C.), Universite de Bordeaux 2, Zone Nord, 33076 Bordeaux, France
Laboratoire de Biologie Moleculaire et Cellulaire de l’Ecole Normale Superieure (J.S.), Unite Mixte de Recherche 5665 Centre National de la Recherche Scientifique, LA 913 Institut National de la Recherche Agronomique, 69364 Lyon Cedex, France
Abstract
Thyroid hormone is essential for normal skeletal development. Hypothyroidism is associated with growth arrest, failure of chondrocyte differentiation, and abnormal matrix synthesis. Thyroid hormone modulates the Indian hedgehog/PTHrP feedback loop and regulates fibroblast growth factor (FGF)/FGF receptor signaling. Because heparan sulfate (HS) proteoglycans (Prgs) (HSPGs) are absolutely required by these signaling pathways, we have investigated whether thyroid status affects HSPG expression within the growth plate. Tibial growth plate sections were obtained from 12-wk-old rats rendered euthyroid, thyrotoxic, or hypothyroid at 6 wk of age, 14-d-old congenitally hypothyroid Pax8-null mice, and TR/TR double-null mice lacking all thyroid hormone receptors. HS and chondroitin sulfate Prg expression was determined by immunohistochemistry using three monoclonal antibodies. There was increased HS staining in growth plates from hypothyroid animals predominantly within the extracellular matrix of reserve and proliferative zones. Cellular HS staining was also increased particularly in prehypertrophic chondrocytes. T3 regulation of HSPG core protein and HS synthetic and modification enzyme expression was studied in ATDC5 cells using semiquantitative RT-PCR. Thyroid hormone negatively regulated expression of the core protein Gpc6, the polymerase Ext1, and the modification enzyme Hs6st2. These studies demonstrate that the expression and distribution of growth plate Prgs are regulated by thyroid hormone, and the regulation of HSPG expression provides an important additional link between FGF and Indian hedgehog signaling and T3. These novel observations suggest that the cartilage matrix and especially HSPGs are critical mediators of the skeletal response to thyroid hormone.
Introduction
LONGITUDINAL SKELETAL GROWTH occurs by endochondral ossification, the sequential process of growth plate chondrocyte proliferation, matrix synthesis, cellular hypertrophy, matrix mineralization, vascular invasion, and apoptosis. The process is precisely regulated by systemic hormones and local paracrine factors including T3, fibroblast growth factors (FGFs), and the Indian hedgehog (Ihh)/PTHrP feedback loop (1, 2).
T3 is essential for skeletal development (3). Childhood hypothyroidism causes growth arrest, epiphyseal dysgenesis, and delayed skeletal development (4), whereas T4 replacement induces rapid catch-up growth. In contrast, untreated childhood thyrotoxicosis advances skeletal maturation and results in premature fusion of the epiphyses and short stature, with early fusion of the cranial sutures and craniosynostosis occurring in severe cases. In addition, the autosomal dominant syndrome of resistance to thyroid hormone, which results from dominant-negative mutations of the T3 receptor (TR) protein, is associated with skeletal abnormalities and short stature (5). We and others have shown that TRs are expressed in growth plate chondrocytes in vivo, that thyroid hormone is essential for the normal organization of proliferating and hypertrophic chondrocytes (6, 7, 8), and that T3 inhibits chondrocyte proliferation and promotes hypertrophic differentiation (9). Taken together, these observations indicate the developing skeleton is exquisitely sensitive to thyroid hormones.
T3 is essential for the normal deposition of growth plate cartilage matrix (6, 10). Using Alcian blue critical electrolyte staining techniques, we demonstrated that hypothyroid rat growth plates may contain increased sulfated proteoglycans (Prgs) with histochemical properties of heparan sulfate (HS) (6). Indeed, hypothyroidism (myxoedema) has long been associated with the accumulation of glycosaminoglycans (GAGs), both within the skin and in other tissues (11, 12). Although these GAGs predominantly consist of hyaluronic acid, increased levels of HS have also been reported (13, 14). Furthermore, recent evidence suggests that the HS core proteins glypican 2 and CD44, the hyaluronan receptor, are negatively regulated by thyroid hormone (15, 16, 17, 18).
Cartilage matrix and cell surface HS Prgs (HSPGs) are intimately involved in both the FGF and Ihh signaling pathways. HSPGs are required for functional FGF-FGF receptor (FGFR) binding (19) but also act as reservoirs for local ligand availability regulating diffusion, gradient formation, and degradation of FGFs. Furthermore, active hedgehog molecules require cell surface HSPGs for normal diffusion and gradient formation (20). Although HSPGs are quantitatively a minor component of cartilage Prgs, their critical importance to normal bone formation is demonstrated by mutations or targeted deletion of HSPG core proteins and synthetic enzymes that result in a wide variety of skeletal developmental disorders (21, 22, 23). At the cell surface, four syndecans, six glypicans, and CD44 may be substituted with HS and within the extracellular matrix perlecan and to a lesser extent aggrecan are substituted with HS (24). HSPGs are key and highly specific developmental regulators due to the structural diversity of their disaccharide polymer, which results from a highly complex series of modifications that includes deacetylation, sulfation, and epimerization (21, 25) (Fig. 1). The enzymes responsible are spatially and temporally regulated resulting in cell- and tissue-specific patterns of HS modification. These modifications generate specific binding sites for growth factors and morphogens, including FGF and Ihh. This specific binding to HSPGs regulates the degradation, sequestration, and diffusion of these growth factors and morphogens (26).
Mast cells, which synthesize and secrete matrix degrading enzymes, have also been implicated in the process of endochondral ossification, which requires degradation of cartilage matrix (27). Mast cells also synthesize heparin and HS and chondroitin sulfate (CS) Prgs. We previously demonstrated by histamine immunohistochemistry that hypothyroidism is associated with an increase in numbers of bone marrow mast cells (BMMCs), particularly in the subchondral region of the primary spongiosum immediately adjacent to the growth plate (28), leading to the possibility that T3 regulation of mast cell function contributes to the effects of thyroid hormone on the developing growth plate.
Based on the known biology and function of Ihh/PTHrP, FGF/FGFR, and HSPGs within the growth plate and our preliminary observations in hypothyroid rats, we hypothesized that T3 regulates HSPG synthesis and expression during bone development. To investigate this hypothesis further, we analyzed the expression of HSPGs using two distinct monoclonal antibodies to HS (29) in growth plate sections from euthyroid, hypothyroid, and thyrotoxic rats (6) and from Pax8–/– mice, which lack thyroid hormone due to thyroid gland agenesis but retain TRs, and TR0/0–/– mice, which lack TRs but have elevated T4, T3, and TSH. To further investigate the mechanism of HSPG regulation within the growth plate, we studied chondrogenesis in vitro in ATDC5 cells (30). This murine cell line expresses the same TR isoforms as primary growth plate chondrocytes, responds to T3 (9, 10), and undergoes a well-described program of chondrogenesis (31). In these studies, we used semiquantitative RT-PCR to investigate whether T3 regulates HS core protein, synthetic enzyme, or modification enzyme mRNA expression.
Materials and Methods
Animals
Male Sprague Dawley rats were rendered hypothyroid or thyrotoxic at 6 wk and were examined together with euthyroid controls at 12 wk as previously described (6). Rat studies were performed under license in compliance with the Animals (Scientific Procedures) Act 1986 and were approved by the Imperial College of Science, Technology, and Medicine Biological Services Unit (London, UK) ethical review process.
Pax8–/–, TR0/0–/–, and their wild-type (wt) littermate controls were bred and genotyped as previously described, and mice were examined at d 14 postnatal (32, 33). Mouse breeding and handling were carried out in a certified animal facility at Ecole Normale Superieure (Lyon, France) according to procedures approved by the local animal care and use committee.
Histology
Rat and mouse tibias were fixed for 24–48 h in 10% neutral buffered formalin and decalcified in 10% formic acid and 10% neutral buffered formalin at 20 C for 5–7 d. Sections (3 μm) were cut onto 3-aminopropyltriethoxysilane-coated slides (Sigma Chemical, Poole, Dorset, UK), deparaffinized in xylene, rehydrated, and stained with hematoxylin and eosin (H&E) or Alcian blue 8GX and van Gieson (6, 34).
Monoclonal antibodies against HSPGs and CSPGs
HS expression was determined using the biotin-conjugated monoclonal antibodies 10E4 (mouse IgM chain) and 3G10 (mouse IgG2b chain) (Seikagaku, Tokyo, Japan) (29). 10E4 (native HS) recognizes a specific tetrasaccharide epitope in HS (Fig. 1), the distribution of which may vary between tissues and between core proteins (35), whereas 3G10 (HS-stub) recognizes terminal desaturated glucuronate residues (-HS) that are revealed after heparitinase I digestion (Seikagaku) (Fig. 1). Thus, 10E4 (native HS) determines the abundance of HS with the specific tetrasaccharide epitope, whereas 3G10 (HS-stub) reflects the total number of HS side chains (36). CS expression was determined using the biotin-conjugated monoclonal antibody 3B3 (CS-stub), which recognizes the terminal unsaturated 6-sulfated disaccharide that remains after chondroitinase ABC (chondroitin 6-sulfate, dermatan sulfate, chondroitin 4-sulfate lyase) digestion (Seikagaku) (Fig. 1) and thus reflects the total number of CS side chains (37).
Heparitinase and chondroitinase pretreatment of slides
Sections stained using the 3G10 and 10E4 antibodies were predigested with chondroitinase ABC (50 mU/slide) and chondroitinase ACII (50 mU/slide) (Seikagaku) to optimize staining sensitivity (38). In 3G10 sections, heparitinase I (4 mU/slide) was also included to generate the HS-stub epitope. Sections stained using the 3B3 antibody were predigested with chondroitinase ABC (50 mU/slide) alone to generate the CS-stub epitope. Predigestion in each case was carried out in 0.2 ml of buffer [50 mM HEPES (pH 7.0), 0.1 M NaCl, 1 mM CaCl2, 50 μg/ml BSA (Invitrogen, Carlsbad, CA), supplemented with protease inhibitors, 2.5 μg/ml pepstatin A, 1 μg/ml aprotinin, 20 μg/ml leupeptin, and 1 mM phenylmethylsulfonylflouride] in a humidified chamber at 37 C for 3 h (36).
Immunohistochemistry
After predigestion, sections were washed in PBS, and endogenous peroxidase activity was removed by incubation with 0.3% hydrogen peroxide in methanol (30 min, room temperature). Sections were subsequently washed in PBS and preincubated in 0.2 ml 3% BSA in PBS (30 min, room temperature) and then incubated overnight in a humidified chamber at 4 C with 0.2 ml 3% BSA-containing antibody at the following dilutions [10E4 (1:100), 3G10 (1:100), 3B3(1:400)]. After washing in PBS, sections were stained with 0.05% 3,3'-diaminobenzidine tetrahydrochloride (Sigma) and 0.01% hydrogen peroxide in 50 mM Tris HCl buffer, pH 7.6, for 10 min. Sections were washed in running water and nuclei were stained with 5% hematoxylin. Two sections at two different levels of the tibial growth plate from each of four euthyroid, thyrotoxic, and hypothyroid rats and each of 2 Pax8–/–, TR0/0–/–, and wt mice were stained under optimal conditions for identical times and analyzed for each of the three monoclonal antibodies.
Immunohistochemistry controls
Rat and mouse tibial growth plate sections and embryo sections were also used as negative controls to further establish antibody specificity and the effects of enzymatic preincubation on staining. All appropriate combinations of heparitinase I, chondroitinase ABC, and chondroitinase AC II were used, and streptavidin-horseradish peroxidase was omitted to ensure the absence of endogenous peroxidase activity. Omission of the primary antibody or substitution of the antibody with biotin-conjugated anti-cd34 (Chemicon International, Temecula, CA), an antigen not expressed in the growth plate (39), were also used to confirm that staining with 3B3, 3G10, and 10E4 antibodies was specific (data not shown).
Quantitation of immunohistochemical staining
Intensity of immunohistochemical staining was analyzed using ImageJ version 1.24o software (http://rsb.info.nih.gov/ij/). Mean growth plate gray levels (range 1–256) were calculated for two sections at two different levels of the tibial growth plate from each of four euthyroid, thyrotoxic, and hypothyroid rats and each of two Pax8–/–, TR0/0–/–, and wt mice. Immunohistochemical staining of sections from all groups of animals was performed in parallel, and comparisons of staining intensities were made within a single experiment.
Cell culture
ATDC5 cells were maintained in DMEM/Ham’s F12 (1:1; Invitrogen) containing 5% fetal bovine serum (Bio-Whittaker, Wokingham, UK), 10 μg/ml transferrin (Sigma), and 3 x 10–8 mol/liter sodium selenite (Sigma) at 37 C in 5% CO2, as described (10, 30). Cells were harvested at d 0, resuspended in medium containing 5% charcoal-stripped serum (CSS) (40), and plated at 350,000 cells/75-cm2 flask. Differentiation medium, which contained 5% CSS with transferrin, sodium selenite, and 10 μg/ml bovine insulin (Sigma) was added on d 1. After 21 d, differentiation medium was replaced with mineralization medium, consisting of MEM (Invitrogen) plus 5% CSS, transferrin, sodium selenite, and insulin. Cells were then cultured in 3% CO2 at 37 C to facilitate calcification of chondrocyte nodules (10, 30). Cells were treated with or without 100 nM T3 throughout the culture period.
Semiquantitative RT-PCR
cDNA was synthesized from total RNA (2.5 μg) using SuperScript II RT (Invitrogen), and 1 μl cDNA was used for PCR amplification of aggrecan (Agc1), perlecan (Hspg2), syndecans (Sdc1–4), glypicans (Gpc1–6), exostosins (Ext1, 2, and Extl2), N-deacetylase/N-sulfotransferases (Ndst1–4), glucuronyl C5-epimerase (Glce), HS 2-O-sulfotransferase (Hs2st), HS 6-O-sulfotransferases (6osts; Hs6st1–4), Fgfr3, and 18s rRNA, as described (9, 41). Primers were designed from GenBank sequences and crossed exonic boundaries. Nucleotide (nt) positions are given in the 5'–3' direction of the synthesized oligonucleotide: Agc1, forward primer Agc1F (nt 6190–6209), reverse primer Agc1R (nt 6540–6559) (GenBank accession no. NM_007424); Hspg2, forward primer Hspg2F (nt 11488–11506), Hspg2R (nt 11857–11874) (GenBank accession no. M77174); Sdc1, forward primer Sdc1F (nt 971–990), reverse primer Sdc1R (nt 1332–1351) (GenBank accession no. NM_011519); Sdc2, forward primer Sdc2F (nt 847–869), reverse primer Sdc2R (nt 1224–1246) (GenBank accession no. NM_008304); Sdc3, forward primer Sdc3F (nt 1216–1234), reverse primer Sdc3R (nt 1588–1607) (GenBank accession no. NM_011520); Sdc4, forward primer Sdc4F (nt 496–515), reverse primer Sdc4R (nt 858–877) (GenBank accession no. NM_011521); Gpc1, forward primer Gpc1F (nt 1438–1457), reverse primer Gpc1R (nt 1813–1833) (GenBank accession no. NM_016696); Gpc2, forward primer Gpc21F (nt 1346–1365), reverse primer Gpc2R (nt 1781–1800) (GenBank accession no. NM_172412); Gpc3, forward primer Gpc3F (nt 1454–1474), reverse primer Gpc3R (nt 1802–1821) (GenBank accession no. NM_016697); Gpc4, forward primer Gpc4F (nt 1552–1571), reverse primer Gpc4R (nt 1923–1944) (GenBank accession no. NM_008150); Gpc5, forward primer Gpc5F (nt 1657–1676), reverse primer Gpc5R (nt 2024–2043) (GenBank accession no. NM_175500); Gpc6, forward primer Gpc6F (nt 1607–1626), reverse primer Gpc6R (nt 1958–1976) (GenBank accession no. NM_011821); Cd44v3, forward primer Cd44v3F (nt 782–801), reverse primer Cd44v3R (nt 1103–1122) (GenBank accession no. NM_009851); Ext1, forward primer Ext1F (nt 2531–2549), reverse primer Ext1R (nt 2877–2896) (GenBank accession no. NM_010162); Ext2, forward primer Ext2F (nt 2027–2047), reverse primer Ext2R (nt 2407–2426) (GenBank accession no. NM_010163); Extl2, forward primer Extl2F (nt 576–595), reverse primer Extl2R (nt 941–960) (GenBank accession no. NM_021388); Ndst1, forward primer Ndst1F (nt 2504–2523), reverse primer Ndst1R (nt 2875–2894) (GenBank accession no. NM_008306); Ndst2, forward primer Ndst2F (nt 2789–2808), reverse primer Ndst2R (nt 3163–3182) (GenBank accession no. NM_010811); Ndst3, forward primer Ndst3F (nt 2587–2606), reverse primer Ndst3R (nt 2934–2953) (GenBank accession no. NM_031186); Ndst4, forward primer Ndst4F (nt 2212–2231), reverse primer Ndst4R (nt 2540–2559) (GenBank accession no. NM_022565); Glce, forward primer GlceF (nt 807–826), reverse primer GlceR (nt 1150–1169) (GenBank accession no. NM_033320); Hs2st, forward primer Hs2stF (nt 227–246), reverse primer Hs2stR (nt 599–618) (GenBank accession no. NM_0011828); Hs6st1, forward primer Hs6st1F (nt 558–576), reverse primer Hs6st1R (nt 921–940) (GenBank accession no. NM_015818); Hs6st2, forward primer Hs6st2F (nt 429–448), reverse primer Hs6st2R (nt 806–823) (GenBank accession no. NM_015819); Hs6st3, forward primer Hs6st3F (nt 771–790), reverse primer Hs6st3R (nt 1111–1129) (GenBank accession no. NM_015820); Fgfr3, forward primer Fgfr3F (nt 330–349), reverse primer Fgfr3R (nt 1377–1396) (GenBank accession no. NM_008010); and 18s rRNA, forward primer 18sF (nt 1577–1596), reverse primer 18sR (nt 1727–1708) (GenBank accession no. 00686). PCRs were performed with an initial denaturation step at 94 C for 2 min, cycles of 30 sec at 94 C, 30 sec at an annealing temperature ranging between 55 and 60 C depending on the primer pairs used, and 30 sec at 68 C, followed by a termination step at 68 C for 2 min. Semiquantitative RT-PCR was optimized to detect linear accumulation of each PCR product as described (41). A range of input RNA concentrations (0.625–5 μg) was tested over a range of 20–40 PCR cycles. The range of linear accumulation of each PCR product was determined by agarose gel electrophoresis, GelDoc 2000 digitization, and QuantityOne version 4.1.0 analysis (Bio-Rad, Hercules, CA). For each primer pair, assays were designed to detect PCR product accumulation in the middle of the linear range to facilitate their relative quantitation (data not shown). Thus, 1 μl cDNA prepared from 2.5 μg RNA was amplified using the optimized number of PCR cycles for each gene (18 cycles for 18s rRNA; 28 cycles for Sdc1, Gpc4, Ext2, Extl2, Glce, and Hs2st; 30 cycles for Hspg2, Sdc3, Sdc4, Gpc1, Gpc3, Gpc6, Ndst1, and Ndst2; 34 cycles for Agc1, Sdc2, Gpc2, and Hs6st2; 38 cycles for Ext1 and Ndst3; 40 cycles for Ndst4 and Hs6st3). Gpc5, Cd44v3, and Hs6st1 expression was not detected in ATDC5 cells.
Statistical analysis
Growth plate immunohistochemical staining intensity data were analyzed by one-way ANOVA and subsequent Tukey multiple comparison post hoc testing as appropriate using GraphPad Prism version 4.03 (GraphPad Software Inc., San Diego, CA). 18s normalized mRNA expression data were analyzed by Student’s t test using GraphPad Prism version 4.03.
Results
Tibial growth plate sections from thyroid manipulated rats were analyzed by histology and immunohistochemistry. Positive immunohistochemistry controls for HS and CS expression in cartilage (cochlea and ribs), skin, kidney, and peripheral nerves were performed on sagittal whole-animal sections from E17.5 and newborn mice (Fig. 2 and legend).
Histology of thyroid-manipulated rats
Growth plates from euthyroid and thyrotoxic rats were organized into reserve, proliferative, and hypertrophic zones (Fig. 3, A, B, I, and J). Bony trabeculae in the primary spongiosum were orientated in line with the columns of chondrocytes, indicating normal functional progression of endochondral ossification. In contrast, in growth plates from hypothyroid rats, chondrocyte columns were irregularly arranged, and hypertrophic chondrocytes were morphologically indistinct (Fig. 3, Q and R). Furthermore, we have previously shown, by in situ hybridization, that hypothyroid growth plates fail to express collagen X, indicating markedly impaired hypertrophic chondrocyte differentiation in thyroid hormone deficiency (6). This impaired hypertrophic differentiation was accompanied by patchy and reduced Alcian blue staining of the growth plate matrix and separation of the growth plate from the primary spongiosum by a layer of bone. Furthermore, the orientation of primary spongiosum bony trabeculae relative to proliferative chondrocytes columns was also disorganized indicating disruption of normal endochondral ossification in hypothyroidism.
Immunohistochemistry
CS expression with 3B3 antibody (CS-stub) in thyroid-manipulated rats.
In both thyrotoxic and euthyroid rat growth plates, the CS-stub antibody stained the growth plate matrix throughout the proliferative and hypertrophic zones, but there was a relative paucity of staining in the reserve zone (Fig. 3, C, D, K, and L). In contrast, in hypothyroid growth plates, the pattern of CS expression was reversed, with the highest levels of staining present in reserve and upper proliferative zone matrix and a paucity of staining in the lower region of the growth plate containing prehypertrophic chondrocytes adjacent to the primary spongiosum (Fig. 3, S and T). However, there was no significant difference in total growth plate staining intensity among thyrotoxic, euthyroid, and hypothyroid sections.
HS expression with 3G10 antibody (HS-stub) in thyroid-manipulated rats.
In both euthyroid and thyrotoxic rat sections, the HS-stub antibody stained the extracellular matrix in a predominantly pericellular distribution throughout all regions of the growth plate (Fig. 3, E, F, M, and N). In hypothyroid sections, the HS staining was more intense and evenly distributed throughout the matrix but was particularly concentrated in the reserve and proliferative zones (Fig. 3, U and V). In the euthyroid growth plates, the intensity of HS staining within the matrix appeared intermediate between the intensity observed in thyrotoxic and hypothyroid sections (Fig. 3, F, N, and V). The total growth plate staining intensity was increased in hypothyroid compared with thyrotoxic and euthyroid sections (P < 0.001 in both cases).
HS expression with 10E4 antibody (native HS) in thyroid-manipulated rats.
In thyrotoxic growth plates, the native HS staining was restricted to hypertrophic chondrocytes, and little staining of the matrix was evident (Fig. 3, G and H). In hypothyroid growth plates, the intensity of cellular HS staining was markedly increased and present throughout the growth plate with additional staining of the matrix evident in the reserve and upper proliferative zones (Fig. 3, W and X). Euthyroid sections revealed an intermediate level of HS staining, similar to that identified with the 3G10 antibody (Fig. 3, O and P). The total growth plate staining intensity was increased in hypothyroid compared with thyrotoxic and euthyroid sections (P < 0.01 and P < 0.001, respectively).
In summary, hypothyroidism resulted in increased expression of growth plate HS Prgs, particularly in regions containing reserve zone progenitor cells and proliferating chondrocytes. These data are in keeping with our previous documentation of increased Alcian blue critical electrolyte staining of HS in this region (6). Furthermore, we have also shown that TR expression is restricted to these regions of the growth plate in vivo (9), suggesting that thyroid hormone-regulated HS expression is mediated by TRs.
To investigate further, tibial growth plates from Pax8–/– and TR0/0–/– mice were also analyzed by histology and immunohistochemistry. Pax8–/– mice have congenital hypothyroidism due to thyroid agenesis. Circulating T4 and T3 are undetectable, and TSH is elevated 400-fold (33). TR0/0–/– mice lack all TR proteins and have gross resistance to thyroid hormone with elevated T4 (14x), T3 (13x), and TSH (>200x) levels (32). These mice represent a model of nuclear receptor deficiency rather than deficiency of ligand.
Histology in Pax8- and TR-null mice
H&E and Alcian blue van Gieson staining of wt growth plate sections from 2-wk-old animals revealed the presence of a secondary ossification center and organization of the growth plate into discrete zones of normal architecture (Fig. 4, A and B). In contrast, analysis of Pax8–/– and TR0/0TR–/– growth plates revealed markedly delayed endochondral ossification with a reduction in size of the skeletal elements, failure of formation of the secondary ossification center, and an epiphysis containing immature chondrocyte precursors (Fig. 4, I, J, Q, and R). There was no difference in the quality of Alcian blue staining of the growth plate matrix between wt and Pax8–/– or TR0/0TR–/– mutant mice.
Immunohistochemistry
CS expression with 3B3 antibody (CS-stub) in Pax8- and TR-null mice.
In wt mice CS expression was evident throughout the growth plate matrix, and staining appeared more intense in the hypertrophic zone (Fig. 4, C and D). In Pax8–/– growth plates, in contrast, this pattern was reversed with increased intensity of CS staining apparent in the reserve zone and upper proliferative zone (Fig. 4, K and L). These findings are similar to those seen in hypothyroid rat growth plates. In contrast, in TR0/0–/– growth plates, there was a reduced and patchy expression of CS distributed in an irregular pattern (Fig. 4, S and T), and total growth plate staining intensity was reduced compared with WT and Pax8–/– sections (P < 0.001 in both cases). Thus, CS expression in receptor-deficient mice is reduced and distinct from that observed in congenitally hypothyroid Pax8–/– mice in which TR expression is retained.
HS expression with 3G10 (HS-stub) antibody in Pax8- and TR-null mice
In wt growth plates, the HS stub antibody 3G10 stained matrix in a predominantly pericellular distribution (Fig. 4, E and F). In hypothyroid Pax8–/– sections, HS staining was more intense and evenly distributed but particularly concentrated around the proliferating chondrocytes (Fig. 4, M and N). These findings were similar to observations in hypothyroid rats. In contrast, HS expression was reduced and patchy in TR0/0–/– growth plates compared with wt and Pax8–/– mice (Fig. 4, U and V), and total growth plate staining intensity was reduced compared with Pax8–/– sections (P < 0.01).
HS expression with 10E4 native antibody in Pax8- and TR-null mice
In wt sections, 10E4 staining of native HS was seen in chondrocytes in the epiphyseal, reserve, and proliferative zones but was absent from the hypertrophic zone (Fig. 4, G and H). HS staining in Pax8–/– growth plates was increased compared with wt and staining was evident both in chondrocytes and the growth plate matrix (Fig. 4, O and P). Furthermore, the pattern of HS staining in Pax8–/– differed; a region of reduced staining intensity was evident in the upper proliferative zone, and a zone of increased intensity was seen at the junction between the proliferative and hypertrophic zones. The overall level of staining revealed by the 10E4 native antibody was reduced in TR0/0–/– sections compared with Pax8–/– mice, but the distribution of HS expression was similar, and the zone of increased staining intensity in prehypertrophic chondrocytes was particularly prominent (Fig. 4, W and X).
CS and HS expression in BMMCs in thyroid-manipulated rats
Examination of the primary spongiosum in euthyroid and thyroid manipulated rats revealed no mast cell staining with the anti-CS stub antibody 3B3 (Fig. 5, A, D, and G), whereas 3G10- and 10E4-positive mast cells were identified throughout the bone marrow cavity in all animals (data not shown). In the euthyroid and thyrotoxic primary spongiosum, very few HS-positive mast cells were evident adjacent to the growth plate (Fig. 5, B, C, E, and F), whereas in hypothyroid animals, large numbers of HS-positive mast cells were identified immediately close to the growth plate in the primary spongiosum (Fig. 5, H and I). In this region of the hypothyroid growth plate, hypertrophic chondrocyte differentiation is impaired, and growth plate vascularization is greatly reduced compared with euthyroid and thyrotoxic sections (6).
Semiquantitative RT-PCR from ATDC5 cell mRNA
The relative concentrations of mRNAs encoding HSPG core proteins (Agc1, Hspg2, Sdc1–4, Gpc1–6, and Cd44v3), HS synthetic enzymes (Extl2, Ext1, and Ext2) and HS modification enzymes (Ndst1–4, Glce, Hs2st, and Hs6st1–3) were determined in ATDC5 cells undergoing chondrogenesis in the presence and absence of T3. ATDC5 cells undergo a well-described reproducible program of chondrogenesis that is characterized by a proliferative phase between d 0 and 12, associated with the expression of collagen II and aggrecan, and a hypertrophic phase of cell differentiation between d 14 to 21, associated with the expression of collagen X. Finally, after d 21, mineralization of chondrogenic nodules occurs (30, 31). RT-PCR analysis was performed using RNA extracted at d 6 (early proliferative phase), 12 (prehypertrophic phase), 21 (hypertrophic phase), and 28 (mineralization phase), and Fgfr3 mRNA expression was examined as a control for a target gene that is induced by T3 (see legend to Fig. 6 and Bassett, J. H. D., and G. R. Williams, unpublished observations).
Expression of Gpc6, Ext1, and Hs6st2 mRNAs was negatively regulated by T3 (Fig. 6). No other core proteins or enzymes were regulated by T3 during ATDC5 cell differentiation in these experiments (data not shown). Gpc6 mRNA expression was inhibited by T3 after 6, 12, and 21 d of culture (P < 0.05) with the maximum change in expression of 4.6-fold observed after 12 d. Ext1 mRNA expression was inhibited by T3 after 6 and 12 d in culture (P < 0.05) with a maximum change in expression of 2.9-fold at 12 d. Hs6st2 mRNA expression was also inhibited by T3 after 6, 12, and 21 d in culture (P < 0.05) with a maximum change in expression of 4.2-fold.
Discussion
We have examined the expression of heparan and CS Prgs in the epiphyseal growth plates of rats pharmacologically manipulated for thyroid status, in mice lacking thyroid hormone but with intact nuclear receptors (Pax8–/–), and in mice lacking all thyroid hormone nuclear receptors but with intact thyroid hormones (TR0/0–/–). We have used chondroitin and HS monoclonal antibodies for the first time to examine abnormalities of growth plate endochondral ossification and have verified their specificity in multiple embryonic tissues. In addition, we have used a well-characterized in vitro model of chondrogenesis and semiquantitative RT-PCR to examine the role of thyroid hormone in the regulation of HSPG expression.
These studies show that CS and HSPG expression and distribution within the growth plate is modulated by changes in thyroid status and by TR occupancy. Furthermore, they categorically demonstrate that growth plate HSPG expression is up-regulated in hypothyroidism. The differing pattern of HS staining with HS-stub (3G10) and native HS (10E4) antibodies is in agreement with previous studies that indicate that 10E4 recognizes a specific HS subpopulation (29, 35, 36).
We have also demonstrated that thyroid hormone negatively regulates HSPG core proteins, HS synthesis, and modification enzymes’ mRNA expression, which correlates with the HSPG immunohistochemistry.
The Glypican 6 is a widely distributed core protein, and its expression in skeletal tissues has been described previously (42). Glypicans are associated with the plasma membrane by a glycosylphosphatidylinositol linkage, have a critical role in developmental morphogenesis, and have been implicated in the regulation of Hedgehog diffusion (43, 44).
The glycosyltransferases exostosin 1 (Ext1) and exostosin 2 (Ext2) form the hetero-oligomeric HS polymerase (Fig. 1). Inactivating mutations of EXT1/2 are associated with hereditary multiple exostosis, a condition characterized by multiple benign osteochondromas adjacent to the growth plates of long bones (22). A model has been proposed in which reduced EXT1 expression within the growth plate resulted in a local perturbation of Ihh diffusion, a reduction in PTHrP-negative feedback and enhanced chondrocyte differentiation (22). This model is entirely consistent with thyroid hormone’s negative regulation of Ext1 and its promotion of hypertrophic differentiation (9). Furthermore, although deletion of Ext1 results in early embryonic lethality in mice, adult heterozygote animals show reduced bone mineral density (45), a phenotype also associated with thyroid hormone excess (46).
Most of the interactions between HS and growth factors such as FGFs occur at short highly sulfated regions within HS chains (Fig. 1). The 6osts catalyze the transfer of sulfate to C6 in N-sulfoglucosamine, which is required in formation of the FGF2 binding site and for its subsequent functional signaling (47) (Fig. 1). Three 6osts have been identified in mice, with Hs6st2 having two functionally similar isoforms. These isoforms result from alternative splicing with the shorter transcript lacking exons 2 and 3 (48). In ATDC5 cells, we identified expression of only the shorter transcript both in the absence or presence of T3 (Fig. 6). The three 6osts show different tissue distribution and substrate specificity. 6ost1 is thought to act in regions of HS containing only iduronate, whereas 6ost2 acts in regions containing both iduronate and glucoronate residues, suggesting that it is 6ost2 that is required for 6-O-sulfotransfation of the FGF2 binding site (47, 48) (Fig. 1). Negative regulation of Hs6st2 mRNA expression by thyroid hormone suggests that T3 may have a role in regulating the generation of FGF binding sites within the growth plate matrix and thus regulating FGF sequestration diffusion and activity.
These data and the previous identification of glypican 2 and CD44 as negatively regulated T3 response genes suggest that thyroid hormone physiologically inhibits HSPG synthesis, regulates its spatial expression, and its subsequent modification by mechanisms that involve the thyroid hormone receptor. The physiological and pathological importance of these observations is demonstrated by the abnormalities of growth and skeletal development seen in thyroid disorders (3, 4, 5, 6, 34) and inherited disorders of HS and CS Prgs (21, 22, 49).
TRs are ligand-inducible transcription factors that are constitutively localized to T3 target cell nuclei that, in the absence of ligand, recruit corepressor protein leading to transcriptional repression of target genes. Thus, unliganded or Apo-TRs are potent repressors of T3 action and are postulated to play a key role in a developmental switch during amphibian metamorphosis and mammalian weaning (50). T3 binding results in dissociation of corepressors, recruitment of coactivators, and transcriptional activation. Despite this well-understood process, one microarray study of hepatic gene regulation showed that 75% of the T3 target genes identified were repressed in response to hormone (51). Although the molecular mechanisms of negative regulation of T3 target genes (which include TSH) have not been precisely characterized, it also involves interactions among TRs, corepressors, and coactivators (51, 52). In this context, the differences in Prg staining observed between congenitally hypothyroid Pax8–/– mice and TR-deficient TR0/0–/– mice are likely to reflect the importance of Apo-TRs in the regulation of HSPG expression during endochondral ossification.
In support of this view, the short stature and delayed endochondral ossification seen in Pax8–/– mice are more severe than that seen in TR0/0–/– mice. This has been interpreted to indicate an important developmental role for unliganded TRs. Indeed, rescue of the Pax8–/– growth phenotype by deletion of all TR isoforms in Pax8–/–TR0/0 double mutants indicates that TR, rather than TR, plays the major role in the skeleton (53). However, the persistence of a severe growth phenotype in Pax8–/–TR1–/– double mutants suggests that the non-T3 binding isoforms TR2 or TR2 may also have an important role in skeletal development (54). Taken together with our previous finding that TR is expressed at 10–12-fold higher levels than TR in bone (34), these studies suggest that developmental regulation of HSPG by thyroid hormones involves the TR proteins predominantly.
Nevertheless, it is important to consider the possibility that differences in expression of HSPGs that are evident between Pax8–/– and TR0/0–/– mice could merely reflect differences in the degree of developmental delay. This issue could potentially be resolved by analysis of older Pax8–/– and TR0/0–/– mice, but unfortunately, Pax8–/– mice do not survive beyond weaning unless they are treated with thyroid hormone, which would confound interpretation of the progression of skeletal growth (53).
The changes in HSPG expression identified in these studies are likely to influence the FGF and Ihh signaling pathways that are critical to normal growth plate development (26, 55, 56). HS binding protects FGFs from degradation and results in local sequestration of FGF ligands. Thus, for FGFs to exert long-range effects, local binding capacity must be saturated or FGFs released by HS degradation (57). Of the FGFs identified in the growth plate, only FGF18, expressed in the perichondrium, has been shown to be essential for chondrogenesis (58). Our demonstration of increased HS but reduced FGFR expression (Williams, G. R., unpublished observations) in hypothyroid growth plates suggests a coordinate regulation of FGF signaling by both ligand sequestration and a reduction in receptor expression. Alternatively, because HS is required for FGFR dimerization and functional ligand binding, increased cell surface HS expression in hypothyroid growth plates might enhance FGF/FGFR signaling in specific areas of the developing growth plate in vivo (57).
Cell surface HSPG expression is also essential for normal Hedgehog diffusion (20). In growth plates, Ihh is expressed in prehypertrophic chondrocytes yet stimulates PTHrP expression in the perichondrium (56, 59). Thus, the suggestion that increased chondrocyte HS expression in hypothyroid rats and Pax8–/– mice might facilitate Ihh diffusion between the prehypertrophic chondrocytes and perichondrium is consistent with our previously demonstration of increased PTHrP expression in hypothyroid growth plates (6). In summary, abnormal Ihh/PTHrP feedback loop activity and changes in FGF/FGFR signaling that result from increased HS expression may contribute to the block in chondrocyte differentiation observed in the hypothyroid growth plate (56, 60).
BMMCs express Prg1, which is substituted by heparin, HS, and CS and subsequently degraded by heparinase (61, 62). In BMMC, the CS is predominantly type E and therefore not recognized by the 3B3 antibody (37, 63). In contrast, in hypothyroid rat primary spongiosum, mast cells were numerous and strongly positive for both HS antibodies. Mast cells synthesize, store, and release matrix-degrading enzymes implicated in release of FGFs and vascular endothelial growth factor from extracellular matrix HSPGs (27, 64). In addition, mast cell heparin and heparinase may themselves modulate FGF and vascular endothelial growth factor signaling (65). FGFs induce mast cell chemomigration; therefore, the degradation of hypothyroid growth plate matrix HSPGs could release additional FGFs leading to enhanced mast cell migration. The increased numbers and redistribution of mast cells adjacent to the chondro-osseous junction in hypothyroidism suggests that mast cell play a role in mediating thyroid hormone effects in bone.
In summary, we have shown that hypothyroidism results in increased HSPG expression using immunohistochemistry methods and have correlated these data with results from RT-PCR experiments showing changes in expression of mRNAs encoding HSPG core proteins, HS synthesis enzymes, and HS modification enzymes. These observations open a new field of study that provides a sound basis for the further investigation of the role of T3 in the regulation of bone matrix synthesis, Prg core proteins, and HS synthesis and modification enzymes during skeletal development.
Footnotes
This work was supported by a Medical Research Council Career Establishment Grant and by an Arthritis Research Campaign Project Grant (to G.R.W.) and by a Medical Research Council Career Clinician Scientist Fellowship (to J.H.D.B.).
First Published Online October 13, 2005
Abbreviations: BMMC, Bone marrow mast cell; CS, chondroitin sulfate; CSS, charcoal-stripped serum; FGF, fibroblast growth factor; FGFR, FGF receptor; GAG, glycosaminoglycan; GlcA, glucuronic acid; GlcNAc, N-acetyl-D-glucosamine; H&E, hematoxylin and eosin; HS, heparan sulfate; HSPG, HS proteoglycan; Ihh, Indian hedgehog; nt, nucleotide; 6ost, 6-O-sulfotransferase; Prg, proteoglycan; TR, T3 receptor; wt, wild type.
Accepted for publication September 23, 2005.
References
Stevens DA, Williams GR 1999 Hormone regulation of chondrocyte differentiation and endochondral bone formation. Mol Cell Endocrinol 151:195–204
Robson H, Siebler T, Shalet SM, Williams GR 2002 Interactions between GH, IGF-I, glucocorticoids, and thyroid hormones during skeletal growth. Pediatr Res 52:137–147
Bassett JH, Williams GR 2003 The molecular actions of thyroid hormone in bone. Trends Endocrinol Metab 14:356–364
Boersma B, Otten BJ, Stoelinga GB, Wit JM 1996 Catch-up growth after prolonged hypothyroidism. Eur J Pediatr 155:362–367
Weiss RE, Refetoff S 1996 Effect of thyroid hormone on growth. Lessons from the syndrome of resistance to thyroid hormone. Endocrinol Metab Clin North Am 25:719–730
Stevens DA, Hasserjian RP, Robson H, Siebler T, Shalet SM, Williams GR 2000 Thyroid hormones regulate hypertrophic chondrocyte differentiation and expression of parathyroid hormone-related peptide and its receptor during endochondral bone formation. J Bone Miner Res 15:2431–2442
Lewinson D, Harel Z, Shenzer P, Silbermann M, Hochberg Z 1989 Effect of thyroid hormone and growth hormone on recovery from hypothyroidism of epiphyseal growth plate cartilage and its adjacent bone. Endocrinology 124:937–945
Freitas FR, Capelo LP, O’Shea PJ, Jorgetti V, Moriscot AS, Scanlan TS, Williams GR, Zorn TM, Gouveia CH 2005 The thyroid hormone receptor -specific agonist GC-1 selectively affects the bone development of hypothyroid rats. J Bone Miner Res 20:294–304
Robson H, Siebler T, Stevens DA, Shalet SM, Williams GR 2000 Thyroid hormone acts directly on growth plate chondrocytes to promote hypertrophic differentiation and inhibit clonal expansion and cell proliferation. Endocrinology 141:3887–3897
Siebler T, Robson H, Shalet SM, Williams GR 2002 Dexamethasone inhibits and thyroid hormone promotes differentiation of mouse chondrogenic ATDC5 cells. Bone 31:457–464
Ord W 1878 On myxoedema, term proposed to be applied to an essential condition in the "cretinoid" affection occasionally observed in middle-aged women. Med Chir Trans Lond 61:57–78
Safer JD, Fraser LM, Ray S, Holick MF 2001 Topical triiodothyronine stimulates epidermal proliferation, dermal thickening, and hair growth in mice and rats. Thyroid 11:717–724
Parving HH, Helin G, Garbarsch C, Johansen AA, Jensen BA, Helin P, Lund P, Lyngsoe J 1982 Acid glycosaminoglycans in myxoedema. Clin Endocrinol (Oxf) 16:207–210
Shishiba Y, Yanagishita M, Hascall VC 1988 Effect of thyroid hormone deficiency on proteoglycan synthesis by human skin fibroblast cultures. Connect Tissue Res 17:119–135
Calloni GW, Alvarez-Silva M, Vituri C, Trentin AG 2001 Thyroid hormone deficiency alters extracellular matrix protein expression in rat brain. Brain Res Dev Brain Res 126:121–124
Kim SW, Ho SC, Hong SJ, Kim KM, So EC, Christoffolete M, Harney JW 2005 A novel mechanism of thyroid hormone-dependent negative regulation by thyroid hormone receptor, nuclear receptor corepressor (NCoR), and GAGA-binding factor on the rat CD44 promoter. J Biol Chem 280:14545–14555
Kerr JM, Gordon DF, Woodmansee WW, Sarapura VD, Ridgway EC, Wood WM 2005 Growth arrest of thyrotropic tumors by thyroid hormone is correlated with novel changes in Wnt-10A. Mol Cell Endocrinol 238:57–67
Greenfield B, Wang WC, Marquardt H, Piepkorn M, Wolff EA, Aruffo A, Bennett KL 1999 Characterization of the heparan sulfate and chondroitin sulfate assembly sites in CD44. J Biol Chem 274:2511–2517
Schlessinger J, Plotnikov AN, Ibrahimi OA, Eliseenkova AV, Yeh BK, Yayon A, Linhardt RJ, Mohammadi M 2000 Crystal structure of a ternary FGF-FGFR-heparin complex reveals a dual role for heparin in FGFR binding and dimerization. Mol Cell 6:743–750
Jeong J, McMahon AP 2002 Cholesterol modification of Hedgehog family proteins. J Clin Invest 110:591–596
Forsberg E, Kjellen L 2001 Heparan sulfate: lessons from knockout mice. J Clin Invest 108:175–180
Duncan G, McCormick C, Tufaro F 2001 The link between heparan sulfate and hereditary bone disease: finding a function for the EXT family of putative tumor suppressor proteins. J Clin Invest 108:511–516
Li JP, Gong F, Hagner-McWhirter A, Forsberg E, Abrink M, Kisilevsky R, Zhang X, Lindahl U 2003 Targeted disruption of a murine glucuronyl C5-epimerase gene results in heparan sulfate lacking L-iduronic acid and in neonatal lethality. J Biol Chem 278:28363–28366
Govindraj P, West L, Koob TJ, Neame P, Doege K, Hassell JR 2002 Isolation and identification of the major heparan sulfate proteoglycans in the developing bovine rib growth plate. J Biol Chem 277:19461–19469
Gallagher JT 2001 Heparan sulfate: growth control with a restricted sequence menu. J Clin Invest 108:357–361
Bernfield M, Gotte M, Park PW, Reizes O, Fitzgerald ML, Lincecum J, Zako M 1999 Functions of cell surface heparan sulfate proteoglycans. Annu Rev Biochem 68:729–777
Fang KC, Wolters PJ, Steinhoff M, Bidgol A, Blount JL, Caughey GH 1999 Mast cell expression of gelatinases A and B is regulated by kit ligand and TGF-. J Immunol 162:5528–5535
Siebler T, Robson H, Bromley M, Stevens DA, Shalet SM, Williams GR 2002 Thyroid status affects number and localization of thyroid hormone receptor expressing mast cells in bone marrow. Bone 30:259–266
David G, Bai XM, Van der Schueren B, Cassiman JJ, Van den Berghe H 1992 Developmental changes in heparan sulfate expression: in situ detection with mAbs. J Cell Biol 119:961–975
Shukunami C, Shigeno C, Atsumi T, Ishizeki K, Suzuki F, Hiraki Y 1996 Chondrogenic differentiation of clonal mouse embryonic cell line ATDC5 in vitro: differentiation-dependent gene expression of parathyroid hormone (PTH)/PTH-related peptide receptor. J Cell Biol 133:457–468
Shukunami C, Ishizeki K, Atsumi T, Ohta Y, Suzuki F, Hiraki Y 1997 Cellular hypertrophy and calcification of embryonal carcinoma-derived chondrogenic cell line ATDC5 in vitro. J Bone Miner Res 12:1174–1188
Gauthier K, Plateroti M, Harvey CB, Williams GR, Weiss RE, Refetoff S, Willott JF, Sundin V, Roux JP, Malaval L, Hara M, Samarut J, Chassande O 2001 Genetic analysis reveals different functions for the products of the thyroid hormone receptor locus. Mol Cell Biol 21:4748–4760
Mansouri A, Chowdhury K, Gruss P 1998 Follicular cells of the thyroid gland require Pax8 gene function. Nat Genet 19:87–90
O’Shea PJ, Harvey CB, Suzuki H, Kaneshige M, Kaneshige K, Cheng SY, Williams GR 2003 A thyrotoxic skeletal phenotype of advanced bone formation in mice with resistance to thyroid hormone. Mol Endocrinol 17:1410–1424
van den Born J, Salmivirta K, Henttinen T, Ostman N, Ishimaru T, Miyaura S, Yoshida K, Salmivirta M 2005 Novel heparan sulfate structures revealed by monoclonal antibodies. J Biol Chem 280:20516–20523
Nackaerts K, Verbeken E, Deneffe G, Vanderschueren B, Demedts M, David G 1997 Heparan sulfate proteoglycan expression in human lung-cancer cells. Int J Cancer 74:335–345
Bayliss MT, Osborne D, Woodhouse S, Davidson C 1999 Sulfation of chondroitin sulfate in human articular cartilage. The effect of age, topographical position, and zone of cartilage on tissue composition. J Biol Chem 274:15892–15900
SundarRaj N, Fite D, Ledbetter S, Chakravarti S, Hassell JR 1995 Perlecan is a component of cartilage matrix and promotes chondrocyte attachment. J Cell Sci 108:2663–2672
Krause DS, Ito T, Fackler MJ, Smith OM, Collector MI, Sharkis SJ, May WS 1994 Characterization of murine CD34, a marker for hematopoietic progenitor and stem cells. Blood 84:691–701
Samuels HH, Stanley F, Casanova J 1979 Depletion of L-3,5,3'-triiodothyronine and L-thyroxine in euthyroid calf serum for use in cell culture studies of the action of thyroid hormone. Endocrinology 105:80–85
Williams GR 2000 Cloning and characterization of two novel thyroid hormone receptor isoforms. Mol Cell Biol 20:8329–8342
Veugelers M, De Cat B, Ceulemans H, Bruystens AM, Coomans C, Durr J, Vermeesch J, Marynen P, David G 1999 Glypican-6, a new member of the glypican family of cell surface heparan sulfate proteoglycans. J Biol Chem 274:26968–27743
Han C, Belenkaya TY, Wang B, Lin X 2004 Drosophila glypicans control the cell-to-cell movement of Hedgehog by a dynamin-independent process. Development 131:601–611
Filmus J, Selleck SB 2001 Glypicans: proteoglycans with a surprise. J Clin Invest 108:497–501
Hilton MJ, Gutierrez L, Martinez DA, Wells DE 2005 EXT1 regulates chondrocyte proliferation and differentiation during endochondral bone development. Bone 36:379–386
Murphy E, Williams GR 2004 The thyroid and the skeleton. Clin Endocrinol (Oxf) 61:285–298
Habuchi H, Tanaka M, Habuchi O, Yoshida K, Suzuki H, Ban K, Kimata K 2000 The occurrence of three isoforms of heparan sulfate 6-O-sulfotransferase having different specificities for hexuronic acid adjacent to the targeted N-sulfoglucosamine. J Biol Chem 275:2859–2868
Habuchi H, Miyake G, Nogami K, Kuroiwa A, Matsuda Y, Kusche-Gullberg M, Habuchi O, Tanaka M, Kimata K 2003 Biosynthesis of heparan sulphate with diverse structures and functions: two alternatively spliced forms of human heparan sulphate 6-O-sulphotransferase-2 having different expression patterns and properties. Biochem J 371:131–142
Thiele H, Sakano M, Kitagawa H, Sugahara K, Rajab A, Hohne W, Ritter H, Leschik G, Nurnberg P, Mundlos S 2004 Loss of chondroitin 6-O-sulfotransferase-1 function results in severe human chondrodysplasia with progressive spinal involvement. Proc Natl Acad Sci USA 101:10155–10160
Chassande O 2003 Do unliganded thyroid hormone receptors have physiological functions J Mol Endocrinol 31:9–20
Feng X, Jiang Y, Meltzer P, Yen PM 2000 Thyroid hormone regulation of hepatic genes in vivo detected by complementary DNA microarray. Mol Endocrinol 14:947–955
Lazar MA 2003 Thyroid hormone action: a binding contract. J Clin Invest 112:497–499
Flamant F, Poguet AL, Plateroti M, Chassande O, Gauthier K, Streichenberger N, Mansouri A, Samarut J 2002 Congenital hypothyroid Pax8(–/–) mutant mice can be rescued by inactivating the TR gene. Mol Endocrinol 16:24–32
Mittag J, Friedrichsen S, Heuer H, Polsfuss S, Visser TJ, Bauer K 2005 Athyroid Pax8–/– mice cannot be rescued by the inactivation of thyroid hormone receptor 1. Endocrinology 146:3179–3184
Ornitz DM, Marie PJ 2002 FGF signaling pathways in endochondral and intramembranous bone development and human genetic disease. Genes Dev 16:1446–1465
Chung UI, Schipani E, McMahon AP, Kronenberg HM 2001 Indian hedgehog couples chondrogenesis to osteogenesis in endochondral bone development. J Clin Invest 107:295–304
Ornitz DM, Itoh N 2001 Fibroblast growth factors. Genome Biol 2:REVIEWS3005
Liu Z, Xu J, Colvin JS, Ornitz DM 2002 Coordination of chondrogenesis and osteogenesis by fibroblast growth factor 18. Genes Dev 16:859–869
Karp SJ, Schipani E, St-Jacques B, Hunzelman J, Kronenberg H, McMahon AP 2000 Indian hedgehog coordinates endochondral bone growth and morphogenesis via parathyroid hormone related-protein-dependent and -independent pathways. Development 127:543–548
Naski MC, Colvin JS, Coffin JD, Ornitz DM 1998 Repression of hedgehog signaling and BMP4 expression in growth plate cartilage by fibroblast growth factor receptor 3. Development 125:4977–4988
Humphries DE, Nicodemus CF, Schiller V, Stevens RL 1992 The human serglycin gene. Nucleotide sequence and methylation pattern in human promyelocytic leukemia HL-60 cells and T-lymphoblast Molt-4 cells. J Biol Chem 267:13558–13563
Gong F, Jemth P, Escobar Galvis ML, Vlodavsky I, Horner A, Lindahl U, Li JP 2003 Processing of macromolecular heparin by heparanase. J Biol Chem 278:35152–35158
Razin E, Stevens RL, Akiyama F, Schmid K, Austen KF 1982 Culture from mouse bone marrow of a subclass of mast cells possessing a distinct chondroitin sulfate proteoglycan with glycosaminoglycans rich in N-acetylgalactosamine-4,6-disulfate. J Biol Chem 257:7229–7236
Walgenbach KJ, Gorospe JR, Gratas C, Brunagel G, Hoffman EP, Shestak KC 2002 A potential role for mast cells in the of bFGF from normal myocytes during angiogenesis in vivo. J Invest Surg 15:153–162
Kato M, Wang H, Kainulainen V, Fitzgerald ML, Ledbetter S, Ornitz DM, Bernfield M 1998 Physiological degradation converts the soluble syndecan-1 ectodomain from an inhibitor to a potent activator of FGF-2. Nat Med 4:691–697
Fine JD, Couchman JR 1988 Chondroitin-6-sulfate-containing proteoglycan: a new component of human skin dermoepidermal junction. J Invest Dermatol 90:283–288
O’Shea PJ, Bassett JH, Sriskantharajah S, Ying H, Cheng SY, Williams GR 28 July 2005 Contrasting skeletal phenotypes in mice with an identical mutation targeted to thyroid hormone receptor 1 or . Mol Endocrinol 10.1210/me.2005-0224(J. H. D. Bassett, R. Swinhoe, O. Chassan)
Biomateriaux et Reparation Tissulaire Institut National de la Sante et de la Recherche Medicale U-443 (O.C.), Universite de Bordeaux 2, Zone Nord, 33076 Bordeaux, France
Laboratoire de Biologie Moleculaire et Cellulaire de l’Ecole Normale Superieure (J.S.), Unite Mixte de Recherche 5665 Centre National de la Recherche Scientifique, LA 913 Institut National de la Recherche Agronomique, 69364 Lyon Cedex, France
Abstract
Thyroid hormone is essential for normal skeletal development. Hypothyroidism is associated with growth arrest, failure of chondrocyte differentiation, and abnormal matrix synthesis. Thyroid hormone modulates the Indian hedgehog/PTHrP feedback loop and regulates fibroblast growth factor (FGF)/FGF receptor signaling. Because heparan sulfate (HS) proteoglycans (Prgs) (HSPGs) are absolutely required by these signaling pathways, we have investigated whether thyroid status affects HSPG expression within the growth plate. Tibial growth plate sections were obtained from 12-wk-old rats rendered euthyroid, thyrotoxic, or hypothyroid at 6 wk of age, 14-d-old congenitally hypothyroid Pax8-null mice, and TR/TR double-null mice lacking all thyroid hormone receptors. HS and chondroitin sulfate Prg expression was determined by immunohistochemistry using three monoclonal antibodies. There was increased HS staining in growth plates from hypothyroid animals predominantly within the extracellular matrix of reserve and proliferative zones. Cellular HS staining was also increased particularly in prehypertrophic chondrocytes. T3 regulation of HSPG core protein and HS synthetic and modification enzyme expression was studied in ATDC5 cells using semiquantitative RT-PCR. Thyroid hormone negatively regulated expression of the core protein Gpc6, the polymerase Ext1, and the modification enzyme Hs6st2. These studies demonstrate that the expression and distribution of growth plate Prgs are regulated by thyroid hormone, and the regulation of HSPG expression provides an important additional link between FGF and Indian hedgehog signaling and T3. These novel observations suggest that the cartilage matrix and especially HSPGs are critical mediators of the skeletal response to thyroid hormone.
Introduction
LONGITUDINAL SKELETAL GROWTH occurs by endochondral ossification, the sequential process of growth plate chondrocyte proliferation, matrix synthesis, cellular hypertrophy, matrix mineralization, vascular invasion, and apoptosis. The process is precisely regulated by systemic hormones and local paracrine factors including T3, fibroblast growth factors (FGFs), and the Indian hedgehog (Ihh)/PTHrP feedback loop (1, 2).
T3 is essential for skeletal development (3). Childhood hypothyroidism causes growth arrest, epiphyseal dysgenesis, and delayed skeletal development (4), whereas T4 replacement induces rapid catch-up growth. In contrast, untreated childhood thyrotoxicosis advances skeletal maturation and results in premature fusion of the epiphyses and short stature, with early fusion of the cranial sutures and craniosynostosis occurring in severe cases. In addition, the autosomal dominant syndrome of resistance to thyroid hormone, which results from dominant-negative mutations of the T3 receptor (TR) protein, is associated with skeletal abnormalities and short stature (5). We and others have shown that TRs are expressed in growth plate chondrocytes in vivo, that thyroid hormone is essential for the normal organization of proliferating and hypertrophic chondrocytes (6, 7, 8), and that T3 inhibits chondrocyte proliferation and promotes hypertrophic differentiation (9). Taken together, these observations indicate the developing skeleton is exquisitely sensitive to thyroid hormones.
T3 is essential for the normal deposition of growth plate cartilage matrix (6, 10). Using Alcian blue critical electrolyte staining techniques, we demonstrated that hypothyroid rat growth plates may contain increased sulfated proteoglycans (Prgs) with histochemical properties of heparan sulfate (HS) (6). Indeed, hypothyroidism (myxoedema) has long been associated with the accumulation of glycosaminoglycans (GAGs), both within the skin and in other tissues (11, 12). Although these GAGs predominantly consist of hyaluronic acid, increased levels of HS have also been reported (13, 14). Furthermore, recent evidence suggests that the HS core proteins glypican 2 and CD44, the hyaluronan receptor, are negatively regulated by thyroid hormone (15, 16, 17, 18).
Cartilage matrix and cell surface HS Prgs (HSPGs) are intimately involved in both the FGF and Ihh signaling pathways. HSPGs are required for functional FGF-FGF receptor (FGFR) binding (19) but also act as reservoirs for local ligand availability regulating diffusion, gradient formation, and degradation of FGFs. Furthermore, active hedgehog molecules require cell surface HSPGs for normal diffusion and gradient formation (20). Although HSPGs are quantitatively a minor component of cartilage Prgs, their critical importance to normal bone formation is demonstrated by mutations or targeted deletion of HSPG core proteins and synthetic enzymes that result in a wide variety of skeletal developmental disorders (21, 22, 23). At the cell surface, four syndecans, six glypicans, and CD44 may be substituted with HS and within the extracellular matrix perlecan and to a lesser extent aggrecan are substituted with HS (24). HSPGs are key and highly specific developmental regulators due to the structural diversity of their disaccharide polymer, which results from a highly complex series of modifications that includes deacetylation, sulfation, and epimerization (21, 25) (Fig. 1). The enzymes responsible are spatially and temporally regulated resulting in cell- and tissue-specific patterns of HS modification. These modifications generate specific binding sites for growth factors and morphogens, including FGF and Ihh. This specific binding to HSPGs regulates the degradation, sequestration, and diffusion of these growth factors and morphogens (26).
Mast cells, which synthesize and secrete matrix degrading enzymes, have also been implicated in the process of endochondral ossification, which requires degradation of cartilage matrix (27). Mast cells also synthesize heparin and HS and chondroitin sulfate (CS) Prgs. We previously demonstrated by histamine immunohistochemistry that hypothyroidism is associated with an increase in numbers of bone marrow mast cells (BMMCs), particularly in the subchondral region of the primary spongiosum immediately adjacent to the growth plate (28), leading to the possibility that T3 regulation of mast cell function contributes to the effects of thyroid hormone on the developing growth plate.
Based on the known biology and function of Ihh/PTHrP, FGF/FGFR, and HSPGs within the growth plate and our preliminary observations in hypothyroid rats, we hypothesized that T3 regulates HSPG synthesis and expression during bone development. To investigate this hypothesis further, we analyzed the expression of HSPGs using two distinct monoclonal antibodies to HS (29) in growth plate sections from euthyroid, hypothyroid, and thyrotoxic rats (6) and from Pax8–/– mice, which lack thyroid hormone due to thyroid gland agenesis but retain TRs, and TR0/0–/– mice, which lack TRs but have elevated T4, T3, and TSH. To further investigate the mechanism of HSPG regulation within the growth plate, we studied chondrogenesis in vitro in ATDC5 cells (30). This murine cell line expresses the same TR isoforms as primary growth plate chondrocytes, responds to T3 (9, 10), and undergoes a well-described program of chondrogenesis (31). In these studies, we used semiquantitative RT-PCR to investigate whether T3 regulates HS core protein, synthetic enzyme, or modification enzyme mRNA expression.
Materials and Methods
Animals
Male Sprague Dawley rats were rendered hypothyroid or thyrotoxic at 6 wk and were examined together with euthyroid controls at 12 wk as previously described (6). Rat studies were performed under license in compliance with the Animals (Scientific Procedures) Act 1986 and were approved by the Imperial College of Science, Technology, and Medicine Biological Services Unit (London, UK) ethical review process.
Pax8–/–, TR0/0–/–, and their wild-type (wt) littermate controls were bred and genotyped as previously described, and mice were examined at d 14 postnatal (32, 33). Mouse breeding and handling were carried out in a certified animal facility at Ecole Normale Superieure (Lyon, France) according to procedures approved by the local animal care and use committee.
Histology
Rat and mouse tibias were fixed for 24–48 h in 10% neutral buffered formalin and decalcified in 10% formic acid and 10% neutral buffered formalin at 20 C for 5–7 d. Sections (3 μm) were cut onto 3-aminopropyltriethoxysilane-coated slides (Sigma Chemical, Poole, Dorset, UK), deparaffinized in xylene, rehydrated, and stained with hematoxylin and eosin (H&E) or Alcian blue 8GX and van Gieson (6, 34).
Monoclonal antibodies against HSPGs and CSPGs
HS expression was determined using the biotin-conjugated monoclonal antibodies 10E4 (mouse IgM chain) and 3G10 (mouse IgG2b chain) (Seikagaku, Tokyo, Japan) (29). 10E4 (native HS) recognizes a specific tetrasaccharide epitope in HS (Fig. 1), the distribution of which may vary between tissues and between core proteins (35), whereas 3G10 (HS-stub) recognizes terminal desaturated glucuronate residues (-HS) that are revealed after heparitinase I digestion (Seikagaku) (Fig. 1). Thus, 10E4 (native HS) determines the abundance of HS with the specific tetrasaccharide epitope, whereas 3G10 (HS-stub) reflects the total number of HS side chains (36). CS expression was determined using the biotin-conjugated monoclonal antibody 3B3 (CS-stub), which recognizes the terminal unsaturated 6-sulfated disaccharide that remains after chondroitinase ABC (chondroitin 6-sulfate, dermatan sulfate, chondroitin 4-sulfate lyase) digestion (Seikagaku) (Fig. 1) and thus reflects the total number of CS side chains (37).
Heparitinase and chondroitinase pretreatment of slides
Sections stained using the 3G10 and 10E4 antibodies were predigested with chondroitinase ABC (50 mU/slide) and chondroitinase ACII (50 mU/slide) (Seikagaku) to optimize staining sensitivity (38). In 3G10 sections, heparitinase I (4 mU/slide) was also included to generate the HS-stub epitope. Sections stained using the 3B3 antibody were predigested with chondroitinase ABC (50 mU/slide) alone to generate the CS-stub epitope. Predigestion in each case was carried out in 0.2 ml of buffer [50 mM HEPES (pH 7.0), 0.1 M NaCl, 1 mM CaCl2, 50 μg/ml BSA (Invitrogen, Carlsbad, CA), supplemented with protease inhibitors, 2.5 μg/ml pepstatin A, 1 μg/ml aprotinin, 20 μg/ml leupeptin, and 1 mM phenylmethylsulfonylflouride] in a humidified chamber at 37 C for 3 h (36).
Immunohistochemistry
After predigestion, sections were washed in PBS, and endogenous peroxidase activity was removed by incubation with 0.3% hydrogen peroxide in methanol (30 min, room temperature). Sections were subsequently washed in PBS and preincubated in 0.2 ml 3% BSA in PBS (30 min, room temperature) and then incubated overnight in a humidified chamber at 4 C with 0.2 ml 3% BSA-containing antibody at the following dilutions [10E4 (1:100), 3G10 (1:100), 3B3(1:400)]. After washing in PBS, sections were stained with 0.05% 3,3'-diaminobenzidine tetrahydrochloride (Sigma) and 0.01% hydrogen peroxide in 50 mM Tris HCl buffer, pH 7.6, for 10 min. Sections were washed in running water and nuclei were stained with 5% hematoxylin. Two sections at two different levels of the tibial growth plate from each of four euthyroid, thyrotoxic, and hypothyroid rats and each of 2 Pax8–/–, TR0/0–/–, and wt mice were stained under optimal conditions for identical times and analyzed for each of the three monoclonal antibodies.
Immunohistochemistry controls
Rat and mouse tibial growth plate sections and embryo sections were also used as negative controls to further establish antibody specificity and the effects of enzymatic preincubation on staining. All appropriate combinations of heparitinase I, chondroitinase ABC, and chondroitinase AC II were used, and streptavidin-horseradish peroxidase was omitted to ensure the absence of endogenous peroxidase activity. Omission of the primary antibody or substitution of the antibody with biotin-conjugated anti-cd34 (Chemicon International, Temecula, CA), an antigen not expressed in the growth plate (39), were also used to confirm that staining with 3B3, 3G10, and 10E4 antibodies was specific (data not shown).
Quantitation of immunohistochemical staining
Intensity of immunohistochemical staining was analyzed using ImageJ version 1.24o software (http://rsb.info.nih.gov/ij/). Mean growth plate gray levels (range 1–256) were calculated for two sections at two different levels of the tibial growth plate from each of four euthyroid, thyrotoxic, and hypothyroid rats and each of two Pax8–/–, TR0/0–/–, and wt mice. Immunohistochemical staining of sections from all groups of animals was performed in parallel, and comparisons of staining intensities were made within a single experiment.
Cell culture
ATDC5 cells were maintained in DMEM/Ham’s F12 (1:1; Invitrogen) containing 5% fetal bovine serum (Bio-Whittaker, Wokingham, UK), 10 μg/ml transferrin (Sigma), and 3 x 10–8 mol/liter sodium selenite (Sigma) at 37 C in 5% CO2, as described (10, 30). Cells were harvested at d 0, resuspended in medium containing 5% charcoal-stripped serum (CSS) (40), and plated at 350,000 cells/75-cm2 flask. Differentiation medium, which contained 5% CSS with transferrin, sodium selenite, and 10 μg/ml bovine insulin (Sigma) was added on d 1. After 21 d, differentiation medium was replaced with mineralization medium, consisting of MEM (Invitrogen) plus 5% CSS, transferrin, sodium selenite, and insulin. Cells were then cultured in 3% CO2 at 37 C to facilitate calcification of chondrocyte nodules (10, 30). Cells were treated with or without 100 nM T3 throughout the culture period.
Semiquantitative RT-PCR
cDNA was synthesized from total RNA (2.5 μg) using SuperScript II RT (Invitrogen), and 1 μl cDNA was used for PCR amplification of aggrecan (Agc1), perlecan (Hspg2), syndecans (Sdc1–4), glypicans (Gpc1–6), exostosins (Ext1, 2, and Extl2), N-deacetylase/N-sulfotransferases (Ndst1–4), glucuronyl C5-epimerase (Glce), HS 2-O-sulfotransferase (Hs2st), HS 6-O-sulfotransferases (6osts; Hs6st1–4), Fgfr3, and 18s rRNA, as described (9, 41). Primers were designed from GenBank sequences and crossed exonic boundaries. Nucleotide (nt) positions are given in the 5'–3' direction of the synthesized oligonucleotide: Agc1, forward primer Agc1F (nt 6190–6209), reverse primer Agc1R (nt 6540–6559) (GenBank accession no. NM_007424); Hspg2, forward primer Hspg2F (nt 11488–11506), Hspg2R (nt 11857–11874) (GenBank accession no. M77174); Sdc1, forward primer Sdc1F (nt 971–990), reverse primer Sdc1R (nt 1332–1351) (GenBank accession no. NM_011519); Sdc2, forward primer Sdc2F (nt 847–869), reverse primer Sdc2R (nt 1224–1246) (GenBank accession no. NM_008304); Sdc3, forward primer Sdc3F (nt 1216–1234), reverse primer Sdc3R (nt 1588–1607) (GenBank accession no. NM_011520); Sdc4, forward primer Sdc4F (nt 496–515), reverse primer Sdc4R (nt 858–877) (GenBank accession no. NM_011521); Gpc1, forward primer Gpc1F (nt 1438–1457), reverse primer Gpc1R (nt 1813–1833) (GenBank accession no. NM_016696); Gpc2, forward primer Gpc21F (nt 1346–1365), reverse primer Gpc2R (nt 1781–1800) (GenBank accession no. NM_172412); Gpc3, forward primer Gpc3F (nt 1454–1474), reverse primer Gpc3R (nt 1802–1821) (GenBank accession no. NM_016697); Gpc4, forward primer Gpc4F (nt 1552–1571), reverse primer Gpc4R (nt 1923–1944) (GenBank accession no. NM_008150); Gpc5, forward primer Gpc5F (nt 1657–1676), reverse primer Gpc5R (nt 2024–2043) (GenBank accession no. NM_175500); Gpc6, forward primer Gpc6F (nt 1607–1626), reverse primer Gpc6R (nt 1958–1976) (GenBank accession no. NM_011821); Cd44v3, forward primer Cd44v3F (nt 782–801), reverse primer Cd44v3R (nt 1103–1122) (GenBank accession no. NM_009851); Ext1, forward primer Ext1F (nt 2531–2549), reverse primer Ext1R (nt 2877–2896) (GenBank accession no. NM_010162); Ext2, forward primer Ext2F (nt 2027–2047), reverse primer Ext2R (nt 2407–2426) (GenBank accession no. NM_010163); Extl2, forward primer Extl2F (nt 576–595), reverse primer Extl2R (nt 941–960) (GenBank accession no. NM_021388); Ndst1, forward primer Ndst1F (nt 2504–2523), reverse primer Ndst1R (nt 2875–2894) (GenBank accession no. NM_008306); Ndst2, forward primer Ndst2F (nt 2789–2808), reverse primer Ndst2R (nt 3163–3182) (GenBank accession no. NM_010811); Ndst3, forward primer Ndst3F (nt 2587–2606), reverse primer Ndst3R (nt 2934–2953) (GenBank accession no. NM_031186); Ndst4, forward primer Ndst4F (nt 2212–2231), reverse primer Ndst4R (nt 2540–2559) (GenBank accession no. NM_022565); Glce, forward primer GlceF (nt 807–826), reverse primer GlceR (nt 1150–1169) (GenBank accession no. NM_033320); Hs2st, forward primer Hs2stF (nt 227–246), reverse primer Hs2stR (nt 599–618) (GenBank accession no. NM_0011828); Hs6st1, forward primer Hs6st1F (nt 558–576), reverse primer Hs6st1R (nt 921–940) (GenBank accession no. NM_015818); Hs6st2, forward primer Hs6st2F (nt 429–448), reverse primer Hs6st2R (nt 806–823) (GenBank accession no. NM_015819); Hs6st3, forward primer Hs6st3F (nt 771–790), reverse primer Hs6st3R (nt 1111–1129) (GenBank accession no. NM_015820); Fgfr3, forward primer Fgfr3F (nt 330–349), reverse primer Fgfr3R (nt 1377–1396) (GenBank accession no. NM_008010); and 18s rRNA, forward primer 18sF (nt 1577–1596), reverse primer 18sR (nt 1727–1708) (GenBank accession no. 00686). PCRs were performed with an initial denaturation step at 94 C for 2 min, cycles of 30 sec at 94 C, 30 sec at an annealing temperature ranging between 55 and 60 C depending on the primer pairs used, and 30 sec at 68 C, followed by a termination step at 68 C for 2 min. Semiquantitative RT-PCR was optimized to detect linear accumulation of each PCR product as described (41). A range of input RNA concentrations (0.625–5 μg) was tested over a range of 20–40 PCR cycles. The range of linear accumulation of each PCR product was determined by agarose gel electrophoresis, GelDoc 2000 digitization, and QuantityOne version 4.1.0 analysis (Bio-Rad, Hercules, CA). For each primer pair, assays were designed to detect PCR product accumulation in the middle of the linear range to facilitate their relative quantitation (data not shown). Thus, 1 μl cDNA prepared from 2.5 μg RNA was amplified using the optimized number of PCR cycles for each gene (18 cycles for 18s rRNA; 28 cycles for Sdc1, Gpc4, Ext2, Extl2, Glce, and Hs2st; 30 cycles for Hspg2, Sdc3, Sdc4, Gpc1, Gpc3, Gpc6, Ndst1, and Ndst2; 34 cycles for Agc1, Sdc2, Gpc2, and Hs6st2; 38 cycles for Ext1 and Ndst3; 40 cycles for Ndst4 and Hs6st3). Gpc5, Cd44v3, and Hs6st1 expression was not detected in ATDC5 cells.
Statistical analysis
Growth plate immunohistochemical staining intensity data were analyzed by one-way ANOVA and subsequent Tukey multiple comparison post hoc testing as appropriate using GraphPad Prism version 4.03 (GraphPad Software Inc., San Diego, CA). 18s normalized mRNA expression data were analyzed by Student’s t test using GraphPad Prism version 4.03.
Results
Tibial growth plate sections from thyroid manipulated rats were analyzed by histology and immunohistochemistry. Positive immunohistochemistry controls for HS and CS expression in cartilage (cochlea and ribs), skin, kidney, and peripheral nerves were performed on sagittal whole-animal sections from E17.5 and newborn mice (Fig. 2 and legend).
Histology of thyroid-manipulated rats
Growth plates from euthyroid and thyrotoxic rats were organized into reserve, proliferative, and hypertrophic zones (Fig. 3, A, B, I, and J). Bony trabeculae in the primary spongiosum were orientated in line with the columns of chondrocytes, indicating normal functional progression of endochondral ossification. In contrast, in growth plates from hypothyroid rats, chondrocyte columns were irregularly arranged, and hypertrophic chondrocytes were morphologically indistinct (Fig. 3, Q and R). Furthermore, we have previously shown, by in situ hybridization, that hypothyroid growth plates fail to express collagen X, indicating markedly impaired hypertrophic chondrocyte differentiation in thyroid hormone deficiency (6). This impaired hypertrophic differentiation was accompanied by patchy and reduced Alcian blue staining of the growth plate matrix and separation of the growth plate from the primary spongiosum by a layer of bone. Furthermore, the orientation of primary spongiosum bony trabeculae relative to proliferative chondrocytes columns was also disorganized indicating disruption of normal endochondral ossification in hypothyroidism.
Immunohistochemistry
CS expression with 3B3 antibody (CS-stub) in thyroid-manipulated rats.
In both thyrotoxic and euthyroid rat growth plates, the CS-stub antibody stained the growth plate matrix throughout the proliferative and hypertrophic zones, but there was a relative paucity of staining in the reserve zone (Fig. 3, C, D, K, and L). In contrast, in hypothyroid growth plates, the pattern of CS expression was reversed, with the highest levels of staining present in reserve and upper proliferative zone matrix and a paucity of staining in the lower region of the growth plate containing prehypertrophic chondrocytes adjacent to the primary spongiosum (Fig. 3, S and T). However, there was no significant difference in total growth plate staining intensity among thyrotoxic, euthyroid, and hypothyroid sections.
HS expression with 3G10 antibody (HS-stub) in thyroid-manipulated rats.
In both euthyroid and thyrotoxic rat sections, the HS-stub antibody stained the extracellular matrix in a predominantly pericellular distribution throughout all regions of the growth plate (Fig. 3, E, F, M, and N). In hypothyroid sections, the HS staining was more intense and evenly distributed throughout the matrix but was particularly concentrated in the reserve and proliferative zones (Fig. 3, U and V). In the euthyroid growth plates, the intensity of HS staining within the matrix appeared intermediate between the intensity observed in thyrotoxic and hypothyroid sections (Fig. 3, F, N, and V). The total growth plate staining intensity was increased in hypothyroid compared with thyrotoxic and euthyroid sections (P < 0.001 in both cases).
HS expression with 10E4 antibody (native HS) in thyroid-manipulated rats.
In thyrotoxic growth plates, the native HS staining was restricted to hypertrophic chondrocytes, and little staining of the matrix was evident (Fig. 3, G and H). In hypothyroid growth plates, the intensity of cellular HS staining was markedly increased and present throughout the growth plate with additional staining of the matrix evident in the reserve and upper proliferative zones (Fig. 3, W and X). Euthyroid sections revealed an intermediate level of HS staining, similar to that identified with the 3G10 antibody (Fig. 3, O and P). The total growth plate staining intensity was increased in hypothyroid compared with thyrotoxic and euthyroid sections (P < 0.01 and P < 0.001, respectively).
In summary, hypothyroidism resulted in increased expression of growth plate HS Prgs, particularly in regions containing reserve zone progenitor cells and proliferating chondrocytes. These data are in keeping with our previous documentation of increased Alcian blue critical electrolyte staining of HS in this region (6). Furthermore, we have also shown that TR expression is restricted to these regions of the growth plate in vivo (9), suggesting that thyroid hormone-regulated HS expression is mediated by TRs.
To investigate further, tibial growth plates from Pax8–/– and TR0/0–/– mice were also analyzed by histology and immunohistochemistry. Pax8–/– mice have congenital hypothyroidism due to thyroid agenesis. Circulating T4 and T3 are undetectable, and TSH is elevated 400-fold (33). TR0/0–/– mice lack all TR proteins and have gross resistance to thyroid hormone with elevated T4 (14x), T3 (13x), and TSH (>200x) levels (32). These mice represent a model of nuclear receptor deficiency rather than deficiency of ligand.
Histology in Pax8- and TR-null mice
H&E and Alcian blue van Gieson staining of wt growth plate sections from 2-wk-old animals revealed the presence of a secondary ossification center and organization of the growth plate into discrete zones of normal architecture (Fig. 4, A and B). In contrast, analysis of Pax8–/– and TR0/0TR–/– growth plates revealed markedly delayed endochondral ossification with a reduction in size of the skeletal elements, failure of formation of the secondary ossification center, and an epiphysis containing immature chondrocyte precursors (Fig. 4, I, J, Q, and R). There was no difference in the quality of Alcian blue staining of the growth plate matrix between wt and Pax8–/– or TR0/0TR–/– mutant mice.
Immunohistochemistry
CS expression with 3B3 antibody (CS-stub) in Pax8- and TR-null mice.
In wt mice CS expression was evident throughout the growth plate matrix, and staining appeared more intense in the hypertrophic zone (Fig. 4, C and D). In Pax8–/– growth plates, in contrast, this pattern was reversed with increased intensity of CS staining apparent in the reserve zone and upper proliferative zone (Fig. 4, K and L). These findings are similar to those seen in hypothyroid rat growth plates. In contrast, in TR0/0–/– growth plates, there was a reduced and patchy expression of CS distributed in an irregular pattern (Fig. 4, S and T), and total growth plate staining intensity was reduced compared with WT and Pax8–/– sections (P < 0.001 in both cases). Thus, CS expression in receptor-deficient mice is reduced and distinct from that observed in congenitally hypothyroid Pax8–/– mice in which TR expression is retained.
HS expression with 3G10 (HS-stub) antibody in Pax8- and TR-null mice
In wt growth plates, the HS stub antibody 3G10 stained matrix in a predominantly pericellular distribution (Fig. 4, E and F). In hypothyroid Pax8–/– sections, HS staining was more intense and evenly distributed but particularly concentrated around the proliferating chondrocytes (Fig. 4, M and N). These findings were similar to observations in hypothyroid rats. In contrast, HS expression was reduced and patchy in TR0/0–/– growth plates compared with wt and Pax8–/– mice (Fig. 4, U and V), and total growth plate staining intensity was reduced compared with Pax8–/– sections (P < 0.01).
HS expression with 10E4 native antibody in Pax8- and TR-null mice
In wt sections, 10E4 staining of native HS was seen in chondrocytes in the epiphyseal, reserve, and proliferative zones but was absent from the hypertrophic zone (Fig. 4, G and H). HS staining in Pax8–/– growth plates was increased compared with wt and staining was evident both in chondrocytes and the growth plate matrix (Fig. 4, O and P). Furthermore, the pattern of HS staining in Pax8–/– differed; a region of reduced staining intensity was evident in the upper proliferative zone, and a zone of increased intensity was seen at the junction between the proliferative and hypertrophic zones. The overall level of staining revealed by the 10E4 native antibody was reduced in TR0/0–/– sections compared with Pax8–/– mice, but the distribution of HS expression was similar, and the zone of increased staining intensity in prehypertrophic chondrocytes was particularly prominent (Fig. 4, W and X).
CS and HS expression in BMMCs in thyroid-manipulated rats
Examination of the primary spongiosum in euthyroid and thyroid manipulated rats revealed no mast cell staining with the anti-CS stub antibody 3B3 (Fig. 5, A, D, and G), whereas 3G10- and 10E4-positive mast cells were identified throughout the bone marrow cavity in all animals (data not shown). In the euthyroid and thyrotoxic primary spongiosum, very few HS-positive mast cells were evident adjacent to the growth plate (Fig. 5, B, C, E, and F), whereas in hypothyroid animals, large numbers of HS-positive mast cells were identified immediately close to the growth plate in the primary spongiosum (Fig. 5, H and I). In this region of the hypothyroid growth plate, hypertrophic chondrocyte differentiation is impaired, and growth plate vascularization is greatly reduced compared with euthyroid and thyrotoxic sections (6).
Semiquantitative RT-PCR from ATDC5 cell mRNA
The relative concentrations of mRNAs encoding HSPG core proteins (Agc1, Hspg2, Sdc1–4, Gpc1–6, and Cd44v3), HS synthetic enzymes (Extl2, Ext1, and Ext2) and HS modification enzymes (Ndst1–4, Glce, Hs2st, and Hs6st1–3) were determined in ATDC5 cells undergoing chondrogenesis in the presence and absence of T3. ATDC5 cells undergo a well-described reproducible program of chondrogenesis that is characterized by a proliferative phase between d 0 and 12, associated with the expression of collagen II and aggrecan, and a hypertrophic phase of cell differentiation between d 14 to 21, associated with the expression of collagen X. Finally, after d 21, mineralization of chondrogenic nodules occurs (30, 31). RT-PCR analysis was performed using RNA extracted at d 6 (early proliferative phase), 12 (prehypertrophic phase), 21 (hypertrophic phase), and 28 (mineralization phase), and Fgfr3 mRNA expression was examined as a control for a target gene that is induced by T3 (see legend to Fig. 6 and Bassett, J. H. D., and G. R. Williams, unpublished observations).
Expression of Gpc6, Ext1, and Hs6st2 mRNAs was negatively regulated by T3 (Fig. 6). No other core proteins or enzymes were regulated by T3 during ATDC5 cell differentiation in these experiments (data not shown). Gpc6 mRNA expression was inhibited by T3 after 6, 12, and 21 d of culture (P < 0.05) with the maximum change in expression of 4.6-fold observed after 12 d. Ext1 mRNA expression was inhibited by T3 after 6 and 12 d in culture (P < 0.05) with a maximum change in expression of 2.9-fold at 12 d. Hs6st2 mRNA expression was also inhibited by T3 after 6, 12, and 21 d in culture (P < 0.05) with a maximum change in expression of 4.2-fold.
Discussion
We have examined the expression of heparan and CS Prgs in the epiphyseal growth plates of rats pharmacologically manipulated for thyroid status, in mice lacking thyroid hormone but with intact nuclear receptors (Pax8–/–), and in mice lacking all thyroid hormone nuclear receptors but with intact thyroid hormones (TR0/0–/–). We have used chondroitin and HS monoclonal antibodies for the first time to examine abnormalities of growth plate endochondral ossification and have verified their specificity in multiple embryonic tissues. In addition, we have used a well-characterized in vitro model of chondrogenesis and semiquantitative RT-PCR to examine the role of thyroid hormone in the regulation of HSPG expression.
These studies show that CS and HSPG expression and distribution within the growth plate is modulated by changes in thyroid status and by TR occupancy. Furthermore, they categorically demonstrate that growth plate HSPG expression is up-regulated in hypothyroidism. The differing pattern of HS staining with HS-stub (3G10) and native HS (10E4) antibodies is in agreement with previous studies that indicate that 10E4 recognizes a specific HS subpopulation (29, 35, 36).
We have also demonstrated that thyroid hormone negatively regulates HSPG core proteins, HS synthesis, and modification enzymes’ mRNA expression, which correlates with the HSPG immunohistochemistry.
The Glypican 6 is a widely distributed core protein, and its expression in skeletal tissues has been described previously (42). Glypicans are associated with the plasma membrane by a glycosylphosphatidylinositol linkage, have a critical role in developmental morphogenesis, and have been implicated in the regulation of Hedgehog diffusion (43, 44).
The glycosyltransferases exostosin 1 (Ext1) and exostosin 2 (Ext2) form the hetero-oligomeric HS polymerase (Fig. 1). Inactivating mutations of EXT1/2 are associated with hereditary multiple exostosis, a condition characterized by multiple benign osteochondromas adjacent to the growth plates of long bones (22). A model has been proposed in which reduced EXT1 expression within the growth plate resulted in a local perturbation of Ihh diffusion, a reduction in PTHrP-negative feedback and enhanced chondrocyte differentiation (22). This model is entirely consistent with thyroid hormone’s negative regulation of Ext1 and its promotion of hypertrophic differentiation (9). Furthermore, although deletion of Ext1 results in early embryonic lethality in mice, adult heterozygote animals show reduced bone mineral density (45), a phenotype also associated with thyroid hormone excess (46).
Most of the interactions between HS and growth factors such as FGFs occur at short highly sulfated regions within HS chains (Fig. 1). The 6osts catalyze the transfer of sulfate to C6 in N-sulfoglucosamine, which is required in formation of the FGF2 binding site and for its subsequent functional signaling (47) (Fig. 1). Three 6osts have been identified in mice, with Hs6st2 having two functionally similar isoforms. These isoforms result from alternative splicing with the shorter transcript lacking exons 2 and 3 (48). In ATDC5 cells, we identified expression of only the shorter transcript both in the absence or presence of T3 (Fig. 6). The three 6osts show different tissue distribution and substrate specificity. 6ost1 is thought to act in regions of HS containing only iduronate, whereas 6ost2 acts in regions containing both iduronate and glucoronate residues, suggesting that it is 6ost2 that is required for 6-O-sulfotransfation of the FGF2 binding site (47, 48) (Fig. 1). Negative regulation of Hs6st2 mRNA expression by thyroid hormone suggests that T3 may have a role in regulating the generation of FGF binding sites within the growth plate matrix and thus regulating FGF sequestration diffusion and activity.
These data and the previous identification of glypican 2 and CD44 as negatively regulated T3 response genes suggest that thyroid hormone physiologically inhibits HSPG synthesis, regulates its spatial expression, and its subsequent modification by mechanisms that involve the thyroid hormone receptor. The physiological and pathological importance of these observations is demonstrated by the abnormalities of growth and skeletal development seen in thyroid disorders (3, 4, 5, 6, 34) and inherited disorders of HS and CS Prgs (21, 22, 49).
TRs are ligand-inducible transcription factors that are constitutively localized to T3 target cell nuclei that, in the absence of ligand, recruit corepressor protein leading to transcriptional repression of target genes. Thus, unliganded or Apo-TRs are potent repressors of T3 action and are postulated to play a key role in a developmental switch during amphibian metamorphosis and mammalian weaning (50). T3 binding results in dissociation of corepressors, recruitment of coactivators, and transcriptional activation. Despite this well-understood process, one microarray study of hepatic gene regulation showed that 75% of the T3 target genes identified were repressed in response to hormone (51). Although the molecular mechanisms of negative regulation of T3 target genes (which include TSH) have not been precisely characterized, it also involves interactions among TRs, corepressors, and coactivators (51, 52). In this context, the differences in Prg staining observed between congenitally hypothyroid Pax8–/– mice and TR-deficient TR0/0–/– mice are likely to reflect the importance of Apo-TRs in the regulation of HSPG expression during endochondral ossification.
In support of this view, the short stature and delayed endochondral ossification seen in Pax8–/– mice are more severe than that seen in TR0/0–/– mice. This has been interpreted to indicate an important developmental role for unliganded TRs. Indeed, rescue of the Pax8–/– growth phenotype by deletion of all TR isoforms in Pax8–/–TR0/0 double mutants indicates that TR, rather than TR, plays the major role in the skeleton (53). However, the persistence of a severe growth phenotype in Pax8–/–TR1–/– double mutants suggests that the non-T3 binding isoforms TR2 or TR2 may also have an important role in skeletal development (54). Taken together with our previous finding that TR is expressed at 10–12-fold higher levels than TR in bone (34), these studies suggest that developmental regulation of HSPG by thyroid hormones involves the TR proteins predominantly.
Nevertheless, it is important to consider the possibility that differences in expression of HSPGs that are evident between Pax8–/– and TR0/0–/– mice could merely reflect differences in the degree of developmental delay. This issue could potentially be resolved by analysis of older Pax8–/– and TR0/0–/– mice, but unfortunately, Pax8–/– mice do not survive beyond weaning unless they are treated with thyroid hormone, which would confound interpretation of the progression of skeletal growth (53).
The changes in HSPG expression identified in these studies are likely to influence the FGF and Ihh signaling pathways that are critical to normal growth plate development (26, 55, 56). HS binding protects FGFs from degradation and results in local sequestration of FGF ligands. Thus, for FGFs to exert long-range effects, local binding capacity must be saturated or FGFs released by HS degradation (57). Of the FGFs identified in the growth plate, only FGF18, expressed in the perichondrium, has been shown to be essential for chondrogenesis (58). Our demonstration of increased HS but reduced FGFR expression (Williams, G. R., unpublished observations) in hypothyroid growth plates suggests a coordinate regulation of FGF signaling by both ligand sequestration and a reduction in receptor expression. Alternatively, because HS is required for FGFR dimerization and functional ligand binding, increased cell surface HS expression in hypothyroid growth plates might enhance FGF/FGFR signaling in specific areas of the developing growth plate in vivo (57).
Cell surface HSPG expression is also essential for normal Hedgehog diffusion (20). In growth plates, Ihh is expressed in prehypertrophic chondrocytes yet stimulates PTHrP expression in the perichondrium (56, 59). Thus, the suggestion that increased chondrocyte HS expression in hypothyroid rats and Pax8–/– mice might facilitate Ihh diffusion between the prehypertrophic chondrocytes and perichondrium is consistent with our previously demonstration of increased PTHrP expression in hypothyroid growth plates (6). In summary, abnormal Ihh/PTHrP feedback loop activity and changes in FGF/FGFR signaling that result from increased HS expression may contribute to the block in chondrocyte differentiation observed in the hypothyroid growth plate (56, 60).
BMMCs express Prg1, which is substituted by heparin, HS, and CS and subsequently degraded by heparinase (61, 62). In BMMC, the CS is predominantly type E and therefore not recognized by the 3B3 antibody (37, 63). In contrast, in hypothyroid rat primary spongiosum, mast cells were numerous and strongly positive for both HS antibodies. Mast cells synthesize, store, and release matrix-degrading enzymes implicated in release of FGFs and vascular endothelial growth factor from extracellular matrix HSPGs (27, 64). In addition, mast cell heparin and heparinase may themselves modulate FGF and vascular endothelial growth factor signaling (65). FGFs induce mast cell chemomigration; therefore, the degradation of hypothyroid growth plate matrix HSPGs could release additional FGFs leading to enhanced mast cell migration. The increased numbers and redistribution of mast cells adjacent to the chondro-osseous junction in hypothyroidism suggests that mast cell play a role in mediating thyroid hormone effects in bone.
In summary, we have shown that hypothyroidism results in increased HSPG expression using immunohistochemistry methods and have correlated these data with results from RT-PCR experiments showing changes in expression of mRNAs encoding HSPG core proteins, HS synthesis enzymes, and HS modification enzymes. These observations open a new field of study that provides a sound basis for the further investigation of the role of T3 in the regulation of bone matrix synthesis, Prg core proteins, and HS synthesis and modification enzymes during skeletal development.
Footnotes
This work was supported by a Medical Research Council Career Establishment Grant and by an Arthritis Research Campaign Project Grant (to G.R.W.) and by a Medical Research Council Career Clinician Scientist Fellowship (to J.H.D.B.).
First Published Online October 13, 2005
Abbreviations: BMMC, Bone marrow mast cell; CS, chondroitin sulfate; CSS, charcoal-stripped serum; FGF, fibroblast growth factor; FGFR, FGF receptor; GAG, glycosaminoglycan; GlcA, glucuronic acid; GlcNAc, N-acetyl-D-glucosamine; H&E, hematoxylin and eosin; HS, heparan sulfate; HSPG, HS proteoglycan; Ihh, Indian hedgehog; nt, nucleotide; 6ost, 6-O-sulfotransferase; Prg, proteoglycan; TR, T3 receptor; wt, wild type.
Accepted for publication September 23, 2005.
References
Stevens DA, Williams GR 1999 Hormone regulation of chondrocyte differentiation and endochondral bone formation. Mol Cell Endocrinol 151:195–204
Robson H, Siebler T, Shalet SM, Williams GR 2002 Interactions between GH, IGF-I, glucocorticoids, and thyroid hormones during skeletal growth. Pediatr Res 52:137–147
Bassett JH, Williams GR 2003 The molecular actions of thyroid hormone in bone. Trends Endocrinol Metab 14:356–364
Boersma B, Otten BJ, Stoelinga GB, Wit JM 1996 Catch-up growth after prolonged hypothyroidism. Eur J Pediatr 155:362–367
Weiss RE, Refetoff S 1996 Effect of thyroid hormone on growth. Lessons from the syndrome of resistance to thyroid hormone. Endocrinol Metab Clin North Am 25:719–730
Stevens DA, Hasserjian RP, Robson H, Siebler T, Shalet SM, Williams GR 2000 Thyroid hormones regulate hypertrophic chondrocyte differentiation and expression of parathyroid hormone-related peptide and its receptor during endochondral bone formation. J Bone Miner Res 15:2431–2442
Lewinson D, Harel Z, Shenzer P, Silbermann M, Hochberg Z 1989 Effect of thyroid hormone and growth hormone on recovery from hypothyroidism of epiphyseal growth plate cartilage and its adjacent bone. Endocrinology 124:937–945
Freitas FR, Capelo LP, O’Shea PJ, Jorgetti V, Moriscot AS, Scanlan TS, Williams GR, Zorn TM, Gouveia CH 2005 The thyroid hormone receptor -specific agonist GC-1 selectively affects the bone development of hypothyroid rats. J Bone Miner Res 20:294–304
Robson H, Siebler T, Stevens DA, Shalet SM, Williams GR 2000 Thyroid hormone acts directly on growth plate chondrocytes to promote hypertrophic differentiation and inhibit clonal expansion and cell proliferation. Endocrinology 141:3887–3897
Siebler T, Robson H, Shalet SM, Williams GR 2002 Dexamethasone inhibits and thyroid hormone promotes differentiation of mouse chondrogenic ATDC5 cells. Bone 31:457–464
Ord W 1878 On myxoedema, term proposed to be applied to an essential condition in the "cretinoid" affection occasionally observed in middle-aged women. Med Chir Trans Lond 61:57–78
Safer JD, Fraser LM, Ray S, Holick MF 2001 Topical triiodothyronine stimulates epidermal proliferation, dermal thickening, and hair growth in mice and rats. Thyroid 11:717–724
Parving HH, Helin G, Garbarsch C, Johansen AA, Jensen BA, Helin P, Lund P, Lyngsoe J 1982 Acid glycosaminoglycans in myxoedema. Clin Endocrinol (Oxf) 16:207–210
Shishiba Y, Yanagishita M, Hascall VC 1988 Effect of thyroid hormone deficiency on proteoglycan synthesis by human skin fibroblast cultures. Connect Tissue Res 17:119–135
Calloni GW, Alvarez-Silva M, Vituri C, Trentin AG 2001 Thyroid hormone deficiency alters extracellular matrix protein expression in rat brain. Brain Res Dev Brain Res 126:121–124
Kim SW, Ho SC, Hong SJ, Kim KM, So EC, Christoffolete M, Harney JW 2005 A novel mechanism of thyroid hormone-dependent negative regulation by thyroid hormone receptor, nuclear receptor corepressor (NCoR), and GAGA-binding factor on the rat CD44 promoter. J Biol Chem 280:14545–14555
Kerr JM, Gordon DF, Woodmansee WW, Sarapura VD, Ridgway EC, Wood WM 2005 Growth arrest of thyrotropic tumors by thyroid hormone is correlated with novel changes in Wnt-10A. Mol Cell Endocrinol 238:57–67
Greenfield B, Wang WC, Marquardt H, Piepkorn M, Wolff EA, Aruffo A, Bennett KL 1999 Characterization of the heparan sulfate and chondroitin sulfate assembly sites in CD44. J Biol Chem 274:2511–2517
Schlessinger J, Plotnikov AN, Ibrahimi OA, Eliseenkova AV, Yeh BK, Yayon A, Linhardt RJ, Mohammadi M 2000 Crystal structure of a ternary FGF-FGFR-heparin complex reveals a dual role for heparin in FGFR binding and dimerization. Mol Cell 6:743–750
Jeong J, McMahon AP 2002 Cholesterol modification of Hedgehog family proteins. J Clin Invest 110:591–596
Forsberg E, Kjellen L 2001 Heparan sulfate: lessons from knockout mice. J Clin Invest 108:175–180
Duncan G, McCormick C, Tufaro F 2001 The link between heparan sulfate and hereditary bone disease: finding a function for the EXT family of putative tumor suppressor proteins. J Clin Invest 108:511–516
Li JP, Gong F, Hagner-McWhirter A, Forsberg E, Abrink M, Kisilevsky R, Zhang X, Lindahl U 2003 Targeted disruption of a murine glucuronyl C5-epimerase gene results in heparan sulfate lacking L-iduronic acid and in neonatal lethality. J Biol Chem 278:28363–28366
Govindraj P, West L, Koob TJ, Neame P, Doege K, Hassell JR 2002 Isolation and identification of the major heparan sulfate proteoglycans in the developing bovine rib growth plate. J Biol Chem 277:19461–19469
Gallagher JT 2001 Heparan sulfate: growth control with a restricted sequence menu. J Clin Invest 108:357–361
Bernfield M, Gotte M, Park PW, Reizes O, Fitzgerald ML, Lincecum J, Zako M 1999 Functions of cell surface heparan sulfate proteoglycans. Annu Rev Biochem 68:729–777
Fang KC, Wolters PJ, Steinhoff M, Bidgol A, Blount JL, Caughey GH 1999 Mast cell expression of gelatinases A and B is regulated by kit ligand and TGF-. J Immunol 162:5528–5535
Siebler T, Robson H, Bromley M, Stevens DA, Shalet SM, Williams GR 2002 Thyroid status affects number and localization of thyroid hormone receptor expressing mast cells in bone marrow. Bone 30:259–266
David G, Bai XM, Van der Schueren B, Cassiman JJ, Van den Berghe H 1992 Developmental changes in heparan sulfate expression: in situ detection with mAbs. J Cell Biol 119:961–975
Shukunami C, Shigeno C, Atsumi T, Ishizeki K, Suzuki F, Hiraki Y 1996 Chondrogenic differentiation of clonal mouse embryonic cell line ATDC5 in vitro: differentiation-dependent gene expression of parathyroid hormone (PTH)/PTH-related peptide receptor. J Cell Biol 133:457–468
Shukunami C, Ishizeki K, Atsumi T, Ohta Y, Suzuki F, Hiraki Y 1997 Cellular hypertrophy and calcification of embryonal carcinoma-derived chondrogenic cell line ATDC5 in vitro. J Bone Miner Res 12:1174–1188
Gauthier K, Plateroti M, Harvey CB, Williams GR, Weiss RE, Refetoff S, Willott JF, Sundin V, Roux JP, Malaval L, Hara M, Samarut J, Chassande O 2001 Genetic analysis reveals different functions for the products of the thyroid hormone receptor locus. Mol Cell Biol 21:4748–4760
Mansouri A, Chowdhury K, Gruss P 1998 Follicular cells of the thyroid gland require Pax8 gene function. Nat Genet 19:87–90
O’Shea PJ, Harvey CB, Suzuki H, Kaneshige M, Kaneshige K, Cheng SY, Williams GR 2003 A thyrotoxic skeletal phenotype of advanced bone formation in mice with resistance to thyroid hormone. Mol Endocrinol 17:1410–1424
van den Born J, Salmivirta K, Henttinen T, Ostman N, Ishimaru T, Miyaura S, Yoshida K, Salmivirta M 2005 Novel heparan sulfate structures revealed by monoclonal antibodies. J Biol Chem 280:20516–20523
Nackaerts K, Verbeken E, Deneffe G, Vanderschueren B, Demedts M, David G 1997 Heparan sulfate proteoglycan expression in human lung-cancer cells. Int J Cancer 74:335–345
Bayliss MT, Osborne D, Woodhouse S, Davidson C 1999 Sulfation of chondroitin sulfate in human articular cartilage. The effect of age, topographical position, and zone of cartilage on tissue composition. J Biol Chem 274:15892–15900
SundarRaj N, Fite D, Ledbetter S, Chakravarti S, Hassell JR 1995 Perlecan is a component of cartilage matrix and promotes chondrocyte attachment. J Cell Sci 108:2663–2672
Krause DS, Ito T, Fackler MJ, Smith OM, Collector MI, Sharkis SJ, May WS 1994 Characterization of murine CD34, a marker for hematopoietic progenitor and stem cells. Blood 84:691–701
Samuels HH, Stanley F, Casanova J 1979 Depletion of L-3,5,3'-triiodothyronine and L-thyroxine in euthyroid calf serum for use in cell culture studies of the action of thyroid hormone. Endocrinology 105:80–85
Williams GR 2000 Cloning and characterization of two novel thyroid hormone receptor isoforms. Mol Cell Biol 20:8329–8342
Veugelers M, De Cat B, Ceulemans H, Bruystens AM, Coomans C, Durr J, Vermeesch J, Marynen P, David G 1999 Glypican-6, a new member of the glypican family of cell surface heparan sulfate proteoglycans. J Biol Chem 274:26968–27743
Han C, Belenkaya TY, Wang B, Lin X 2004 Drosophila glypicans control the cell-to-cell movement of Hedgehog by a dynamin-independent process. Development 131:601–611
Filmus J, Selleck SB 2001 Glypicans: proteoglycans with a surprise. J Clin Invest 108:497–501
Hilton MJ, Gutierrez L, Martinez DA, Wells DE 2005 EXT1 regulates chondrocyte proliferation and differentiation during endochondral bone development. Bone 36:379–386
Murphy E, Williams GR 2004 The thyroid and the skeleton. Clin Endocrinol (Oxf) 61:285–298
Habuchi H, Tanaka M, Habuchi O, Yoshida K, Suzuki H, Ban K, Kimata K 2000 The occurrence of three isoforms of heparan sulfate 6-O-sulfotransferase having different specificities for hexuronic acid adjacent to the targeted N-sulfoglucosamine. J Biol Chem 275:2859–2868
Habuchi H, Miyake G, Nogami K, Kuroiwa A, Matsuda Y, Kusche-Gullberg M, Habuchi O, Tanaka M, Kimata K 2003 Biosynthesis of heparan sulphate with diverse structures and functions: two alternatively spliced forms of human heparan sulphate 6-O-sulphotransferase-2 having different expression patterns and properties. Biochem J 371:131–142
Thiele H, Sakano M, Kitagawa H, Sugahara K, Rajab A, Hohne W, Ritter H, Leschik G, Nurnberg P, Mundlos S 2004 Loss of chondroitin 6-O-sulfotransferase-1 function results in severe human chondrodysplasia with progressive spinal involvement. Proc Natl Acad Sci USA 101:10155–10160
Chassande O 2003 Do unliganded thyroid hormone receptors have physiological functions J Mol Endocrinol 31:9–20
Feng X, Jiang Y, Meltzer P, Yen PM 2000 Thyroid hormone regulation of hepatic genes in vivo detected by complementary DNA microarray. Mol Endocrinol 14:947–955
Lazar MA 2003 Thyroid hormone action: a binding contract. J Clin Invest 112:497–499
Flamant F, Poguet AL, Plateroti M, Chassande O, Gauthier K, Streichenberger N, Mansouri A, Samarut J 2002 Congenital hypothyroid Pax8(–/–) mutant mice can be rescued by inactivating the TR gene. Mol Endocrinol 16:24–32
Mittag J, Friedrichsen S, Heuer H, Polsfuss S, Visser TJ, Bauer K 2005 Athyroid Pax8–/– mice cannot be rescued by the inactivation of thyroid hormone receptor 1. Endocrinology 146:3179–3184
Ornitz DM, Marie PJ 2002 FGF signaling pathways in endochondral and intramembranous bone development and human genetic disease. Genes Dev 16:1446–1465
Chung UI, Schipani E, McMahon AP, Kronenberg HM 2001 Indian hedgehog couples chondrogenesis to osteogenesis in endochondral bone development. J Clin Invest 107:295–304
Ornitz DM, Itoh N 2001 Fibroblast growth factors. Genome Biol 2:REVIEWS3005
Liu Z, Xu J, Colvin JS, Ornitz DM 2002 Coordination of chondrogenesis and osteogenesis by fibroblast growth factor 18. Genes Dev 16:859–869
Karp SJ, Schipani E, St-Jacques B, Hunzelman J, Kronenberg H, McMahon AP 2000 Indian hedgehog coordinates endochondral bone growth and morphogenesis via parathyroid hormone related-protein-dependent and -independent pathways. Development 127:543–548
Naski MC, Colvin JS, Coffin JD, Ornitz DM 1998 Repression of hedgehog signaling and BMP4 expression in growth plate cartilage by fibroblast growth factor receptor 3. Development 125:4977–4988
Humphries DE, Nicodemus CF, Schiller V, Stevens RL 1992 The human serglycin gene. Nucleotide sequence and methylation pattern in human promyelocytic leukemia HL-60 cells and T-lymphoblast Molt-4 cells. J Biol Chem 267:13558–13563
Gong F, Jemth P, Escobar Galvis ML, Vlodavsky I, Horner A, Lindahl U, Li JP 2003 Processing of macromolecular heparin by heparanase. J Biol Chem 278:35152–35158
Razin E, Stevens RL, Akiyama F, Schmid K, Austen KF 1982 Culture from mouse bone marrow of a subclass of mast cells possessing a distinct chondroitin sulfate proteoglycan with glycosaminoglycans rich in N-acetylgalactosamine-4,6-disulfate. J Biol Chem 257:7229–7236
Walgenbach KJ, Gorospe JR, Gratas C, Brunagel G, Hoffman EP, Shestak KC 2002 A potential role for mast cells in the of bFGF from normal myocytes during angiogenesis in vivo. J Invest Surg 15:153–162
Kato M, Wang H, Kainulainen V, Fitzgerald ML, Ledbetter S, Ornitz DM, Bernfield M 1998 Physiological degradation converts the soluble syndecan-1 ectodomain from an inhibitor to a potent activator of FGF-2. Nat Med 4:691–697
Fine JD, Couchman JR 1988 Chondroitin-6-sulfate-containing proteoglycan: a new component of human skin dermoepidermal junction. J Invest Dermatol 90:283–288
O’Shea PJ, Bassett JH, Sriskantharajah S, Ying H, Cheng SY, Williams GR 28 July 2005 Contrasting skeletal phenotypes in mice with an identical mutation targeted to thyroid hormone receptor 1 or . Mol Endocrinol 10.1210/me.2005-0224(J. H. D. Bassett, R. Swinhoe, O. Chassan)