• Sox-4;
  • osteoblasts;
  • chondrocytes;
  • growth plate;
  • parathyroid hormone


  1. Top of page
  2. Abstract

Parathyroid hormone (PTH) and PTH-related protein (PTHrP) exert potent and diverse effects in cells of the osteoblastic and chondrocytic lineages. However, downstream mediators of these effects are characterized inadequately. We identified a complementary DNA (cDNA) clone encoding the 5′ end of the transcription factor Sox-4, using a subtracted cDNA library enriched in PTH-stimulated genes from the human osteoblast-like cell line OHS. The SOX-4 gene is a member of a gene family (SOX and SRY) comprising transcription factors that bind to DNA through their high mobility group (HMG)-type binding domain, and previous reports have implicated Sox proteins in various developmental processes. In situ hybridization of fetal and neonatal mouse hindlimbs showed that Sox-4 messenger RNA (mRNA) was expressed most intensely in the zone of mineralizing cartilage where chondrocytes undergo hypertrophy, and by embryonic day 17 (ED17), after the primary ossification center was formed, its expression was detected only in the region of hypertrophic chondrocytes. Sox-4 mRNA was detected in osteoblast-like cells of both human and rodent origin. In OHS cells, physiological concentrations (10−10–10−9 M) of human PTH 1-84 [hPTH(1-84)] and hPTH(1-34), but not hPTH(3-84), stimulated Sox-4 mRNA expression in a time-dependent manner, indicating involvement of the PTH/PTHrP receptor. Sox-4 transcripts also were detected in various nonosteoblastic human cell lines and tissues, in a pattern similar to that previously reported in mice. The presence of Sox-4 mRNA in hypertrophic chondrocytes within the mouse epiphyseal growth plate at sites that overlap or are adjacent to target cells for PTH and PTHrP, and its strong up-regulation via activated PTH/PTHrP receptors in OHS cells, makes it a promising candidate for mediating downstream effects of PTH and PTHrP in bone.


  1. Top of page
  2. Abstract

SOX-4 IS a member of a family of transcription factors consisting of the mammalian testis-determining factor SRY and the Sox proteins of which about 30 have been found so far. (1–3) Sox proteins are defined by the presence of a high mobility group (HMG) box of about 80 amino acids that also are found in the chromatin-associated proteins HMG-1 and HMG-2, suggesting that they represent architectural components of chromatin in addition to their function as transcription factors.(4)

Evidence that Sox proteins play important roles during embryonic development and cellular differentiation recently has been presented. Disruption of the Sox-4 gene in mice leads to circulatory failure in embryos homozygous for the mutation at embryonic day 14 (ED14). This is related to impaired development of the endocardial ridges, consistent with Sox-4 messenger RNA (mRNA) expression in the endocardial anlagen of wild-type embryos.(5) Furthermore, a role for Sox-4 in lymphocyte development is implicated from studies showing that the gene is expressed in cell lines representing early pre-B and T lymphocyte lineages(6) and that differentiation of B lymphocytes is arrested at the pro-B cell stage in SOX-4 (−/−) mice.(5) Recent studies have shown that Sox-4 mRNA is stimulated by the progesterone analogue progestin in breast cancer cells(7) and by prostaglandin (PG)A2 and Δ12-PGJ2 in Hep3B human hepatocellular carcinoma cells.(8) Several reports also suggest important roles of Sox proteins in chondrogenesis: Sox-4 is expressed in embryonic cartilage of the respiratory tract, the orthosympatic trunk, and the spinal cord(5); defects in the transcription factor Sox-9 lead to camptomelic dysplasia, which is characterized by skeletal malformation(9,10); and L-Sox-5, Sox-6, and Sox-9 cooperatively regulate chondrocyte-associated genes such as type II and type XI collagen.(11,12) Postnatally, although Sox-4 is expressed most abundantly in the heart, testis, ovary, and thymus,(6,13) gross histological examination of SOX-4 (−/−) embryos at ED14 did not reveal manifestations of any obvious abnormalities at these sites, except dysplasia of the semilunar valves.(5)

Parathyroid hormone (PTH) is produced in the parathyroid glands and is a main systemic regulator of calcium homeostasis and bone remodeling. Full-length PTH contains 84 amino acid residues, and although circulating forms of the hormone include distinct N-terminal, midregion, and C-terminal fragments, the latter forms predominate.(14) By comparison, the PTH-related protein (PTHrP) is produced locally in a wide variety of cell types and tissues, including the skeleton, and various proteolytic fragments of PTHrP also are known to exist. PTH and PTHrP share structural homology in the 34 amino acids N-terminal region, enabling the two proteins to bind and activate a common cloned G protein-coupled PTH/PTHrP receptor present on the surface of chondrocytes, osteoblasts, and renal epithelial cells(15,16) with equal potency. This then leads to stimulation of multiple effector systems, including adenylate cyclase and phospholipase C, and increases of intracellular calcium. (17–19) N- and C-terminal fragments of PTH and PTHrP also show distinct actions in bone-derived cells that are not mediated by the “classical” PTH/PTHrP receptor and thus are likely to be mediated by unique PTH and PTHrP receptors. (20–23) For example, a novel PTH receptor (C-PTH receptor) that is expressed by parathyroid and osteoblastic cells specifically binds C-terminal fragments of PTH but not PTHrP.(24,25) Although the biochemical properties and biological significance of these other receptors remain to be elucidated, the PTH/PTHrP receptor system has been shown to play crucial roles in embryonic skeletal development. Mice lacking functional copies of the respective genes show a similar phenotype, including severe defects in the epiphyseal growth plates associated with increased rate of chondrocyte differentiation.(26) The more severe skeletal abnormalities observed in mouse embryos, where the PTH/PTHrP receptor gene has been ablated as compared with those lacking one of the ligands (i.e., PTHrP (−/−) mutants), can be taken as evidence for a compensatory role of circulating PTH in the latter knockout model.(27)

The PTH and PTHrP signals in bone cells are transduced via activation of receptor-coupled secondary messenger systems that lead to transcriptional regulation of downstream target genes. We have used subtractive cloning to identify genes that are expressed differentially in response to PTH (or PTHrP) in the human osteoblast-like OHS osteosarcoma cell line. From a subtracted complementary DNA (cDNA) library, we have identified a clone encoding a part of the human Sox-4 gene. Our data showing for the first time that Sox-4 gene expression is stimulated by human PTH 1-84 [hPTH(1-84)]and PTH(1-34) in osteoblastic cells, implicate a member of the HMG-transcription factors as a novel component of the PTH signal transduction pathway in osteoblastic cells. Furthermore, expression of Sox-4 within the embryonic growth plate indicates that Sox-4 has an important function in bone formation.


  1. Top of page
  2. Abstract


Commercial premade blots were from Clontech (Palo Alto, CA, U.S.A.). Recombinant hPTH(1-84) and hPTH(3-84) were prepared in our laboratory as described.(28,29) hPTH(1-34) was purchased from Sigma (St. Louis, MO, U.S.A.); ExpressHyb and Marathon-Ready cDNA were from Clontech; the OHS, KPDXM, and KRIB cell lines were obtained from Dr. Ø. Bruland (The Norwegian Radium Hospital, Oslo, Norway). MNNG-HOS, SaOS-2, ROS 17/2.8, UMR-106, and MDA-MB-231 cell lines, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA were from the American Type Culture Collection (ATCC; Manassas, VA, U.S.A.); genomic DNA for mouse type X collagen was obtained from Dr. B. R. Olsen (Harvard Medical School, Boston, MA, U.S.A.); and the rat PTH/PTHrP receptor cDNA and C29 cell line were obtained from Dr. H. Jüppner (Massachusetts General Hospital, Boston, MA, U.S.A.).

Construction of cDNA libraries and directional tag polymerase chain reaction subtraction

Cytoplasmic RNA was isolated from OHS cells treated with 10−7 M hPTH(1-84) or vehicle for 24 h and enriched for poly(A+) RNA using oligo (dT)25 Dynabeads(DYNAL, Oslo, Norway). A target cDNA library was prepared from the PTH-treated material as described(30) except that the Sal I linker (Gibco BRL, Stafford, TX, U.S.A.), rather than EcoRI linker, was ligated to the double-stranded cDNA, and pSPORT-Sal I/Not I (Gibco BRL), rather than pT7T3D, was used as vector. The cDNA library was transformed into Epicurian coli SURE electroporation-competent cells according to vendor of the cells (Stratagene, La Jolla, CA, U.S.A.). The driver library was prepared as described(30) from poly(A)+ RNA from untreated OHS cells, except that pT7T3D-EcoRI/Not I/bacterial alkaline phosphatase (BAP) was used rather than the pGEM11Zf(−) vector. The subtractive hybridization was performed once using the previously described procedures.(30,31) Briefly, both libraries were linearized with Not I, followed by in vitro transcription with T7 RNA polymerase using the megascript kit (Ambion, Austin, TX, U.S.A.). Thirty micrograms of the sense strand OHS-purified complementary RNA (cRNA) were used directly in the subtractive hybridization procedure as driver, whereas the target OHS-PTH had to go through an additional [32P]-labeled antisense cDNA synthesis prepared as described in the First-Strand cDNA Synthesis Kit (Amersham Pharmacia Biotech; Uppsala, Sweden), using 3 μg of the sense cRNA, 0.75 μg of Not I-d(T)18 primer, and 150 μCi of α[32P]deoxycytosine triphosphate (dCTP) to allow later quantification of the target cDNA. The product was treated with alkali to hydrolyze the RNA template and purified before annealing for 24 h at 68°C in 15 μl of hybridization buffer(30) with the 30 μg of the driver sense cRNA (ratio 1:10). After chromatography on hydroxyapatite, the single-stranded target cDNA fraction corresponded to 3% of the input material, as judged by tracer quantitation. The sample was then purified and an aliquot (1/10) of the single-stranded material was used as template in a 30-cycle polymerase chain reaction (PCR) (94°C for 15 s, 60°C for 30 s, and 72°C for 1 minute) using primers corresponding to the tag sequences in the target (5′-AACTGGAAGAATTCGCGG-3′ and 5′-ACCGGTCCGGAATTCCCG-3′) and Taq polymerase (Perkin Elmer, Branchburg, NJ, U.S.A.). The amplification product was cleaved sequentially with Not I and EcoRI and then separated from primers and the flanking sequences with a Sephacryl S 400 spin column, before it was ligated into the pT7T3D EcoRI/Not I/BAP vector. An aliquot (1/10) of this subtracted library was electroporated into Epicurian coli cells. The subtracted library was named ΔOHS-PTH and represented 1.5 × 106 clones.

Differential screening of the subtracted cDNA library

The single-stranded OHS-PTH cDNA target library was used to prepare a cDNA probe. This subtracted cDNA (ΔOHS-PTH) was labeled using PCR (same conditions and primers as described previously for construction of the cDNA subtraction library, except for a reduced dCTP concentration and the use of α[32P]dCTP(30) and was used as probe for the subtractive library, ΔOHS-PTH probe. To prepare a representative probe for the target and driver libraries, 2 μg from the OHS-PTH sense cRNA and 2 μg from OHS sense cRNA, respectively, were used to synthesize the corresponding antisense cDNAs. These probes were prepared in the same way as the single-stranded cDNA that was used as target in the subtractive hybridization. An aliquot (1/100) each of the single-stranded cDNA (OHS-PTH and OHS) was labeled by PCR using the same conditions as for the subtracted probe. The OHS probe was synthesized by PCR using the same primers as for the subtracted probe. To synthesize the OHS probe (driver), the primers were 5′-ACTGGAAGAATTCGCGG-3′ and 5′-AGGCCAAGAATTCGGCACGA-3′. To screen the subtracted cDNA library, 960 single colonies were picked and grown up in microtiter plates and three identical nylon replica filters were prepared from each plate. These identical filters were hybridized to each of the three different probes, target probe (OHS-PTH), driver probe (OHS), and subtraction probe (ΔOHS-PTH), using equal amounts of32P radioactivity (>106 cpm/ml) for each filter. Colonies that gave strong signals with the subtractive probe (ΔOHS-PTH) and with the target probe (OHS-PTH) but no signals with the driver probe (OHS) were picked for further analysis. The inserts were sequenced, using the ThermoSequenase kit (Amersham, Arlington Heights, IL, U.S.A.). The inserts were then isolated from these clones and used as probes in Northern blot analysis.

Cloning of the 5′ untranslated region by rapid amplification of cDNA ends

For 5′ rapid amplification of cDNA ends (RACE), the AP1 and AP2 anchoring primers in the Marathon-Ready cDNA kit (Clontech) were used, in addition to the following downstream primers: internal (5′AGCAAACTGCAGCGCGGTGAGAGAG) and external (5′TCGTTTGACGTCGCGCCACTCTCTC). Marathon-Ready cDNA from the MOLT4 human T lymphocyte cell line was used as template as these cells contained large amounts of the transcript. PCR was performed with four cycles of annealing at 68°C and then four cycles at 65°C and finally 30 cycles at 63°C. Extension time was 60 s at 70°C, and denaturation time was 15 s at 94°C for all cycles. The GeneAmp XL PCR kit (Perkin Elmer, Foster City, CA, U.S.A.) was used according to the manufacturer's instructions, and the product was cloned into pCRII-TOPO (Invitrogen, Carlsbad, CA, U.S.A.) and sequenced using Thermo Sequenase (Amersham). These sequence data are available from GenBank/EMBL/DDBJ under the accession number AF124147.

Tissue preparation and in situ hybridization

Animals were maintained in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. Pregnant Swiss-Webster mice were killed by CO2, and hindlimbs of fetuses and neonatal pups were removed and dipped into freshly prepared 4% paraformaldehyde (PFA)-phosphate-buffered saline (PBS; pH 7.4) for at least 1 h. Bones were dehydrated and embedded in paraffin wax. Bone sections of 5 μm were prepared by standard histological procedures and mounted on Superfrost/Plus slides (Fisher, Pittsburgh, PA, U.S.A.).

Antisense or sense [35S]-labeled RNAs were synthesized from linearized plasmids containing the genomic DNA for mouse type X collagen DNA (nucleotides 903-2215; no. X65121), the 350-base pair (bp) Sox-4 human cDNA, or the rat PTH/PTHrP receptor cDNA (nucleotides 258-1671; no. M77184) using the Gemini Transcription kit (Promega, Madison, WI, U.S.A.). The specificity of in situ hybridization was indicated by the absence of hybridization signal when adjacent tissue sections were subjected to identical conditions with radiolabeled sense RNA probes (data not shown).

In situ hybridization was performed as described.(32) In brief, sections were dewaxed with xylene, dehydrated with increasing concentrations of ethanol, and postfixed with 4% PFA-PBS for 15 minutes. After washing with PBS and another treatment with 4% PFA-PBS for 10 minutes, sections then were washed sequentially with PBS, incubated with 0.2N HCl (10 minutes), again washed with PBS, acetylated with 0.25% acetic anhydride in the presence of triethanolamine (0.1 M; 10 minutes), dehydrated with increasing concentrations of ethanol, and air-dried. Hybridizations with [35S]-labeled cRNAs (5 × 107 cpm/ml) were performed in a humidified chamber in a solution containing 50% formamide, 10% dextran sulfate, 1× Denhardt's solution (0.02% Ficoll 400, 0.02% polyvinylpyrrolidone, and 0.02% bovine serum albumin [BSA]), 600 mM NaCl, 10 mM Tris-HCl, 1 mM EDTA, 50 mM dithiothreitol (DTT), 0.25% sodium dodecyl sulfate (SDS), and 200 μg/ml transfer RNA (18 h; 55°C). After hybridization, sections were washed briefly with 5× SSC at 50°C, with 50% formamide in 2× SSC at 50°C for 30 minutes, and then with 10 mM Tris-HCl (pH 7.6), 500 mM NaCl, and 1 mM EDTA (TNE) at 37°C for 30 minutes. Sections were treated with 10 μg/ml RNase A in TNE (37°C; 30 minutes). After being washed with TNE, sections were incubated once with 2× SSC (50°C; 20 minutes) and twice with 0.2× SSC (50°C; 20 minutes). Sections were dehydrated with increasing concentrations of ethanol and air-dried. Slides were then placed on X-ray films (hyperfilm β-max; Amersham) and film autoradiographs were obtained after overnight exposure. Slides were dipped into NTB-2 (Eastman Kodak, New Haven, CT, U.S.A.) and stored at 4°C for the times estimated from the intensity of expression on X-ray film (1-7 days). After development, sections were counterstained with hematoxylin and eosin and mounted.

Northern blot analysis

Total RNA was prepared using the TRIZOL reagent (Life Technologies, Rockville, MD, U.S.A.). Poly(A+) RNA was isolated from total RNA preparations using Dynabeads according to the manufacturer's instructions. Northern blots were prepared as described previously(33) using Hybond N+ filters (Amersham). The filters were prehybridized, hybridized with a radiolabeled Sox-4 cDNA fragment, washed, and exposed to X-ray films according to the manufacturer's recommendation. Quantification of the Northern blot signals on X-ray film was performed by scanning densitometry using an XRS 3sc scanner and the BioImage System (Millipore Corp., Ann Arbor, MI, U.S.A.). Values for Sox-4 mRNA signals were normalized to those of the GAPDH or β-actin to correct for variations in loading and are presented as fold stimulation in PTH-treated cultures relative to untreated control.

Cell culture

Normal osteoblasts (NOs) were prepared from bone biopsy specimens of patients undergoing hip surgery and cultured as described.(34) OHS, KPDXM, and SaOS-2 cells were cultured in RPMI-1640 medium with 10% fetal calf serum (FCS), L-glutamine (0.3 mg/ml), penicillin (5 × 105 IU/liter), and streptomycin (50 mg/liter). The MNNG-HOS, SaOS-2, ROS 17/2.8, UMR-106, and MDA-MB-231 cell lines were cultured according to instructions from ATCC and subcultured weekly. The medium was changed every 3-4 days, and the cells always received fresh medium containing 5% FCS 24 h before the experiments were started. C29 cells were cultured as described.(35) Full-length PTH or truncated PTH peptides of concentrations or vehicle (acetic acid, 1.7 × 10−5 M final concentration) were added to the culture medium of cells that had just reached confluence, and the cells were harvested at the given time points, as described in the figure legends.


  1. Top of page
  2. Abstract

Cloning and sequencing of a highly conserved region within the 5′ untranslated region of Sox-4

The osteogenic human OHS cell line exhibits characteristics of a relatively well-differentiated osteoblastic phenotype, including high expression of alkaline phosphatase and PTH-stimulated adenylyl cyclase activities, (36–38) and thus represents a valuable model system to investigate downstream effectors of PTH and PTHrP in bone. Immediate and early response genes such as c-fos are activated within minutes in osteoblast-like cells on PTH stimulation without the need of foregoing protein synthesis.(39,40) Secondary target genes are then transcriptionally activated, and in order to enrich for their mRNAs, OHS cells were stimulated with PTH over a period of 24 h. Subtractive cloning was used to identify the target genes for PTH signaling. A cDNA clone of 350 bp isolated from the subtracted library (see Introduction) showed 86% identity to the 5′ untranslated region of mouse Sox-4 but no homology to the published human sequence(13) (Fig. 1). To map the 5′ end of the Sox-4 transcript, 5′ RACE was performed (see Materials and Methods). We identified a putative transcription start site at position −767 relative to the translational start codon, in agreement with the 5′ end of a number of expressed sequence tags (EST) clones with identical or very similar sequences (e.g., the clones with GenBank accession numbers W25179, W16493, and AA534016).

thumbnail image

Figure FIG. 1. Schematic drawing displaying percent identity of the 350-bp cDNA clone isolated from the subtraction library within the published mouse and human cDNA sequences.

Download figure to PowerPoint

Expression of Sox-4 mRNA in fetal and newborn mouse hindlimbs

The spatial and temporal expression of Sox-4 mRNA in developing mouse tibias from ED15.5 throughout postnatal day 1 (P1) was studied using in situ hybridization. By ED15.5, Sox-4 mRNA was most highly expressed in the zone of mineralizing cartilage where chondrocytes characteristically express type X collagen (Kaechong Lee, personal communication, 1999), undergo hypertrophy, and the future bone marrow cavity is formed (Figs. 2A–2D and 2G). Furthermore, although expression of the PTH/PTHrP receptor is greatest in the zones flanking the central portion of the cartilage core, lower levels of the transcript are expressed throughout the central portion, overlapping with the Sox-4 expressing domain ( Figs. 2C–2F). At ED17.0, the primary ossification center has been formed in the tibias, giving rise to areas consisting of epiphyseal cartilage at the ends of the developing bones that are separated by a central marrow cavity (Figs. 2H and 2I). At this stage, expression of Sox-4 mRNA is evident in the zone of hypertrophic chondrocytes, although the levels of Sox-4 transcripts are lower than at ED15.5. By ED18.5, although mRNA for Sox-4 colocalizes with that of type X collagen and thereby identifies the Sox-4 expressing cells as hypertrophic chondrocytes, its expression is adjacent to or partly overlapping with the zone of PTH/PTHrP receptor mRNA expressing cells (Figs. 2J–2O). Postnatally (P1), expression of Sox-4 is reduced compared with earlier time points (Figs. 2P–2S).

thumbnail image

Figure FIG. 2. Sox-4 mRNA expression in developing endochondral bone from ED15.5 to P1. In situ hybridization analysis of tissue sections from mouse hindlimbs using [35S]-labeled probes. (A, C, E, G, H, J, L, N, P, and R) Sections stained with hematoxylin and eosin and (B, D, F, I, K, M, O, Q, and S) corresponding dark field views are shown. G is an enlargement of the center of C. The blue-stained areas of G indicate regions of calcified matrix. Hybridizations to the human Sox-4 antisense probe are shown in B, D, I, K, and Q; hybridizations to the antisense PTH/PTHrP receptor probe are shown in F and M; hybridization to the antisense type X collagen is shown in O; and hybridization to the human Sox-4 sense probe is shown S. The cartilage core is composed of chondrocytes flanked by several layers of cells defining the periosteum/perichondrium. Sox-4 mRNA expression is detected in the cartilage core from ED15.5 and onward. (A-G) Expression is most intense at ED15.5, (I) diminished at ED17.0 and (K) further reduced at ED18.5, and (Q) strongly reduced at P1. Sox-4 hybridization partly overlaps the most intense part of that of the PTH/PTHrP receptor at ED15.5 as well as at ED18.5 and it overlaps with that of type X collagen at ED18.5. Bars = 60 μm.

Download figure to PowerPoint

Regulation of Sox-4 mRNA by PTH in osteoblast-like cells

Dose and time response studies were carried out in OHS cell cultures using the full-length PTH(1-84) form. Northern analysis revealed a single Sox-4 mRNA species of about 5 kilobases (kb), as previously reported in humans and mice.(6,13) The Sox-4 mRNA levels increased dose dependently from 10−10 M PTH(1-84), and maximum response (∼10-fold stimulation) was observed at a dose of 10−9 M (Fig. 3A). Sox-4 mRNA levels peaked after 12-24 h of treatment with PTH(1-84) (10−7 M; Fig. 3B), and elevated levels relative to the untreated controls were observed also at 48 h and 5 days (not shown). Similar effects of PTH on Sox-4 mRNA expression were found in the rat UMR-106 osteosarcoma cell line (not shown).

thumbnail image

Figure FIG. 3. Regulation of Sox-4 mRNA expression by PTH in OHS cells. (A-C) Northern blots hybridized to the Sox-4 cDNA and GAPDH cDNA probe as control. Each lane was loaded with 20 μg of total RNA. The results of experiments A, B, and C are given as fold regulation (PTH-treated/control), representing densitometric scanning values corrected for loading variations (see Materials and Methods section). Average values of 2-10 independent experiments with SE of mean are shown in the bottom row. Unless otherwise stated, the concentration of all forms of PTH was 10−7 M. (A) A representative blot showing the effects of increasing concentrations of PTH(1-84) on the levels of Sox-4 mRNA in OHS cells. The cells were treated with vehicle (control) or 10−11–10−7 M PTH(1-84) for 24 h as indicated. (B) Time-course of effect of PTH on Sox-4 mRNA expression in human OHS cells. Total RNA was isolated after the indicated time points. A representative Northern blot is displayed. (C) A typical Northern blot showing the effects of stimulation of various forms of PTH on expression of the Sox-4 transcript levels.

Download figure to PowerPoint

To examine whether Sox-4 mRNA was regulated by PTH also in NOs, calvarial osteoblasts (released from calvaria by collagenase digestion) were grown in the presence of 10% fetal bovine serum supplemented with β-glycerophosphate and vitamin C to induce osteogenic differentiation.(41,42) Northern analysis of total RNA from prolonged cultures of these cells showed that levels of Sox-4 mRNA increased about 2-fold in response to treatment with 2 × 10−9 M hPTH(1-84) for 16 h (not shown). Our data show that PTH regulation of Sox-4 mRNA takes place also in normal primary osteoblasts and not only in clonal osteoblast-like cells.

An intact N terminus of PTH is required for its stimulatory effect on Sox-4 mRNA expression in OHS cells as shown by the equipotent ability of PTH(1-84) and PTH(1-34) to increase levels of Sox-4, whereas the N-terminally truncated form PTH(3-84) had no effect. PTH(3-84) is able to bind to the receptor but unable to stimulate cyclic monophosphate (cAMP) production.(28) Furthermore, Sox-4 mRNA expression in OHS cells increased 3-fold when the OHS cells were treated with the cAMP analog 8-Br-cAMP, but not with phorbol-12-myristate-13-acetate (TPA; not shown). These results indicate that increased Sox-4 expression on PTH stimulation is mediated by the PTH/PTHrP receptor via increases in cAMP levels.

Sox-4 mRNA expression in human tissues and cell lines

Northern blot analysis was performed in order to define the pattern of Sox-4 mRNA expression in various cell lines. We found that Sox-4 mRNA is abundantly expressed in human MNNG-HOS, KRIB, OHS, SaOS-2, KPDXM, and in rat ROS 17/2.8 osteosarcoma cell lines (Fig. 4A), while lower levels of the transcript were detected in human osteoblast cultures derived from normal trabecular bone. The osteoblastic nature of the osteosarcoma cell lines was verified by positive hybridization signals using cDNA probes for the following osteoblastic markers: type I collagen, osteonectin, PTH/PTHrP receptor, and alkaline phosphatase (data not shown). No or very little expression of Sox-4 mRNA was found in the breast cancer cell line MDA-MB-231 (Fig. 4A).

thumbnail image

Figure FIG. 4. Expression of Sox-4 mRNA in various cell lines. (A) Two micrograms of poly(A+) RNA from the indicated cell lines was loaded in each well. The filter was hybridized with the Sox-4 probe and reprobed with the β-actin probe as control. OHS + PTH indicates stimulation of OHS cells with 10−7 M PTH for 24 h. NO, normal osteoblasts; KPDXM, SaOS-2, MNNG-HOS, and OHS, human osteosarcoma cell lines; KRIB, v-Ki-ras-oncogene-transformed human osteosarcoma cell line; ROS17/2.8, rat osteosarcoma cells; MDA-MB-231, human breast cancer cell line; C29 LLC-PK1 porcine renal epithelial cells stably transformed with the human PTH/PTHrP receptor. (B) The Sox-4 probe was hybridized to a commercial human cancer cell line (MTN) blot (Clontech) containing 2 μg of poly(A+) RNA in each lane. MOLT4, lymphoblastic leukemia cell line; A549, lung carcinoma cell line; SW480, colorectal adenocarcinoma cell line; HL60, HeLa, and S3, promyelocytic leukemia cell lines; Raji, Burkitt's lymphoma cell line; G363, melanoma cell line. Numbers indicate size (kb) of RNA standard.

Download figure to PowerPoint

Among other nonosteoblastic cell lines, high levels of Sox-4 mRNA expression were found in the lymphoblastic leukemia cell line MOLT4, moderate expression was found in the lung carcinoma cell line A549, and low expression was found in the colorectal adenocarcinoma cell line SW480 as revealed by hybridization of the Sox-4 probe to a commercial human cancer cell-line multiple tissues Northern blot (MTN; Clontech). No expression was detected in the promyelocytic leukemia cell line HL60, HeLa, and S3; in the Burkitt's lymphoma cell line Raji; or in the melanoma cell line G363 (Fig. 4B). These results are in agreement with previous reports showing Sox-4 expression in early B and T cell lineages.(6)

The expression pattern of Sox-4 mRNA in human tissues was analyzed by hybridization to a human RNA Master Blot (Clontech; Fig. 5). Sox-4 mRNA levels were found to be most abundant in fetal brain, but high levels also were detected in fetal heart, kidney, liver, spleen, and lung. With the exception of thymus, expression was higher in fetal tissues than in adult tissues. In adult tissue, although Sox-4 mRNA was most abundantly expressed in the thymus and ovary, signals also were detected in the frontal lobe, uterus, testis, prostate, stomach, lung trachea, placenta, small intestine, kidney, mammary gland, and appendix. Thus, the expression pattern is reminiscent of what has been found in mice.(6,5)

thumbnail image

Figure FIG. 5. Expression of Sox-4 mRNA in human tissues. The RNA Master Blot (Clontech) was hybridized to the Sox-4 probe. Each dot contains 100-500 ng of poly(A+) RNA normalized to compensate for unequal transcriptional activity between different tissues.

Download figure to PowerPoint


  1. Top of page
  2. Abstract

Using subtractive cloning, we have identified the HMG transcription factor Sox-4 as a novel putative regulatory component of the PTH signal transduction pathway in osteoblastic cells. In situ hybridization showed that Sox-4 transcripts are predominantly localized to the zone of hypertrophic chondrocytes in the growth plates of developing fetal mouse hindlimbs. We found that the Sox-4 gene is expressed by human and rat osteoblast-like clonal cell lines and primary osteoblastic cell cultures, and that the levels of Sox-4 mRNA in human OHS and rat UMR-106 osteosarcoma cell lines are time- and dose-dependently increased in response to treatment with full-length PTH(1-84) for 24 h.

DNA sequencing revealed high homology of the isolated clone to the 5′ untranslated region of mouse Sox-4. Mapping of the Sox-4 mRNA by 5′ RACE identified the end at position −767 relative to the translational start codon. Northern analysis further showed the presence of a single transcript of about 5 kb in various cell lines and tissues. Thus, our data indicate that the Sox-4 cDNA clone represents a hitherto unknown 5′ end of the human Sox-4 transcript. Interestingly, the interspecies conservation is nearly as high for this newly discovered 5′ region as for the HMG box and the serine-rich regions (94% and 91% sequence identity, respectively, whereas the identity between the 5′ ends is 86%). By comparison, the DNA sequence of the translated coding region of mouse and Sox-4 shows only 64% identity between the HMG box and the serine-rich region (Fig. 1). Because the Sox-4 protein is encoded by a single exon in mice(43) and is caused by stop codons downstream of the conserved sequence that are in frame with the protein coding sequence, this region is not likely to be part of a translated region. We therefore propose that this region has an important regulatory function.

Sox-4 mRNA is predominantly expressed in the growth plate of hypertrophic chondrocytes of fetal mouse hindlimbs as detected using in situ hybridization. Furthermore, Sox-4 transcripts are most abundant at ED15.5 and gradually decrease to undetectable levels at P1. Its expression overlaps the same area as that of the central and less intense part of the PTH/PTHrP receptor distribution. At ED 18.5 the common area of expression of Sox-4 mRNAs and the PTH/PTHrP receptor appears to be reduced to overlapping zones near the distal ends. We have found Sox-4 mRNA expression in cultured human articular chondrocytes (not shown), indicating a function for Sox-4 also in chondrocytes outside the embryonic growth plate.

Other members of the Sox family and various collagens are candidates for being involved with the regulative functions of Sox-4. L-Sox-5, Sox-6, and Sox-9 are all expressed in proliferating chondrocytes,(11) whereas Sox-4 mRNA is located in hypertrophic chondrocytes (Fig. 2). Sox-4 is able to bind to the same DNA binding site as other Sox proteins. Hence, it would be interesting to study whether Sox-4 acts as an inducer for type X collagen expression, which is used as a marker for hypertrophic chondrocytes, and participates in regulation of expression of other collagens. The type X collagen promoter has a Sox-9 responsive region that probably also will bind Sox-4,(44) and type II and type XI collagen as well as the cartilage-derived retinoic acid-sensitive protein (CD-RAP) can all be regulated by Sox-9 in chondrocytes through their HMG type protein binding site.(12, 45, 46)

Sox-4 mRNA was not detected in the osteoblasts of the perichondrium or the bone collar by in situ hybridization, perhaps because of the scattered distribution of osteoblasts in this tissue and/or lower levels of expression existing at various differentiation stages.

The effects of PTH on Sox-4 mRNA expression in osteoblastic cells are coupled via specific receptors. The dose-dependent increase of Sox-4 mRNA in OHS cells in response to PTH parallels the PTH-stimulated cAMP production in human OHS cells and LLC-PK1 cells stably transfected with the cloned human PTH/PTHrP receptor, with effective concentrations with half maximal effect (EC50) values of 2 × 10−9 M (unpublished results) and 5 × 10−10 M, respectively.(28,47) These results, together with the observation that PTH(3-84) was unable to stimulate cAMP production, make the classical PTH/PTHrP receptor a strong candidate for mediating the stimulatory effect of PTH on Sox-4 mRNA expression. Activated PTH/PTHrP receptors are capable of stimulating multiple intracellular effector systems, including adenylate cyclase and phospholipase C. (17–19) Stimulation of Sox-4 mRNA expression by the cAMP mimic 8-Br-cAMP but not TPA indicates that the effect of PTH is mediated through adenylate cyclase. The functional significance of Sox-4 in osteoblasts is unknown. PTH stimulation of human bone cells after several weeks of culture gave no significant Sox-4 mRNA stimulation. This may be caused by phenotypic alterations during in vitro conditions.

In addition to PTH, PTHrP, and their receptor(s), the transforming growth factor (TGF) superfamily including the bone morphogenetic proteins (BMPs) with receptors and Indian hedgehog are important players in the regulation of chondrogenesis and bone formation.(48) Recently, it was found that BMP-6 is negatively regulated by PTHrP, resulting in accelerated maturation of cultured chicken sternal chondrocytes.(49,50) Being downstream in the signaling pathway of PTH and PTHrP and being expressed in the same cells as BMP-6, Sox-4 is a candidate transcription factor for taking part in mediation of these effects.

We found Sox-4 mRNA to have similar expression pattern in human tissues and cell lines as has been found in the mouse, indicating similar roles in the two species.

High expression of Sox-4 mRNA in osteosarcoma cells and hypertrophic chondrocytes may suggest a role for Sox-4 in maturation of osteoblasts and chondrocytes, analogous to what has been found for B cell and T cell development, where lack of Sox-4 blocks further differentiation at the pro-B cell stage(5) and restricts thymocyte differentiation.(51) Interestingly, Sox-4 is highly expressed in the thymus, in which PTHrP and its receptor also is expressed.(52)

Being expressed within the embryonic growth plate and regulated via the common PTH/PTHrP receptor in osteoblast-like cells, Sox-4 appears to be an important regulator of bone development and can be added to the list of transcription factors regulated by PTH and PTHrP.


  1. Top of page
  2. Abstract

We thank Zeremariam Yohannes and Åse Karine Fjellheim for technical help, Sigurd Ørstavik for computational assistance, and Øyvind S. Bruland and Harald Jüppner for cell lines. This work was supported by The Norwegian Osteoporosis Society, The Norwegian Cancer Society, The Norwegian Research Council, Anders Jahre's Foundation for Promotion of Science, Rachel and Otto Bruuns legate, The Novo Nordisk Foundation, The Norwegian Society for Endocrinology, and Bristol-Myers Squibb. S.R. was financed by The Norwegian Foundation for Health and Rehabilitation through the Norwegian Women's Health Association.


  1. Top of page
  2. Abstract
  • 1
    Sinclair AH, Berta P, Palmer MS, Hawkins JR, Griffiths BL, Smith MJ, Foster JW, Frischauf AM, Lovell-Badge R, Goodfellow PN 1990 A gene from the human sex-determining region encodes a protein with homology to a conserved DNA-binding motif. Nature 346:240244.
  • 2
    Gubbay J, Collignon J, Koopman P, Capel B, Economou A, Munsterberg A, Vivian N, Goodfellow P, Lovell-Badge R 1990 A gene mapping to the sex-determining region of the mouse Y chromosome is a member of a novel family of embryonically expressed genes. Nature 346:245250.
  • 3
    Cremazy F, Soullier S, Berta P, Jay P 1998 Further complexity of the human SOX gene family revealed by the combined use of highly degenerate primers and nested PCR. FEBS Lett 438:311314.
  • 4
    Giese K, Cox J, Grosschedl R 1992 The HMG domain of lymphoid enhancer factor 1 bends DNA and facilitates assembly of functional nucleoprotein structures. Cell 69:185195.
  • 5
    Schilham MW, Oosterwegel MA, Moerer P, Ya J, de Boer PA, van de Wetering S, Verbeek S, Lamers WH, Kruisbeek AM, Cumano A, Clevers H 1996 Defects in cardiac outflow tract formation and pro-B-lymphocyte expansion in mice lacking Sox-4. Nature 380:711714.
  • 6
    van de Wetering M, Oosterwegel M, van Norren K, Clevers H 1993 Sox-4, an Sry-like HMG box protein, is a transcriptional activator in lymphocytes. EMBO J 12:38473854.
  • 7
    Graham JD, Hunt SM, Tran N, Clarke CL 1999 Regulation of the expression and activity by progestins of a member of the SOX gene family of transcriptional modulators. J Mol Endocrinol 22:295304.
  • 8
    Ahn SG, Cho GH, Jeong SY, Rhim H, Choi JY, Kim IK 1999 Identification of cDNAs for Sox-4, an HMG-Box protein, and a novel human homolog of yeast splicing factor SSF-1 differentially regulated during apoptosis induced by prostaglandin A2/Delta12-PGJ2 in Hep3B cells. Biochem Biophys Res Commun 260:216221.
  • 9
    Wagner T, Wirth J, Meyer J, Zabel B, Held M, Zimmer J, Pasantes J, Bricarelli FD, Keutel J, Hustert E 1994 Autosomal sex reversal and camptomelic dysplasia are caused by mutations in and around the SRY-related gene SOX9. Cell 79:11111120.
  • 10
    Foster JW, Dominguez-Steglich MA, Guioli S, Kowk G, Weller PA, Stevanovic M, Weissenbach J, Mansour S, Young ID, Goodfellow PN 1994 Campomelic dysplasia and autosomal sex reversal caused by mutations in an SRY-related gene. Nature 372:525530.
  • 11
    Lefebvre V, Li P, de Crombrugghe B 1998 A new long form of Sox5 (L-Sox5), Sox6 and Sox9 are coexpressed in chondrogenesis and cooperatively activate the type II collagen gene. EMBO J 17:57185733.
  • 12
    Bridgewater LC, Lefebvre V, de Crombrugghe B 1998 Chondrocyte-specific enhancer elements in the Col1a2 gene resemble the Col2a1 tissue-specific enhancer. J Biol Chem 273:1499815006.
  • 13
    Farr CJ, Easty DJ, Ragoussis J, Collignon J, Lovell-Badge R, Goodfellow PN 1993 Characterization and mapping of the human SOX4 gene. Mamm Genome 4:577584.
  • 14
    Goltzman D, Henderson B, Loveridge N 1980 Cytochemical bioassay of parathyroid hormone: Characteristics of the assay and analysis of circulating hormonal forms. J Clin Invest 65:13091317.
  • 15
    Schipani E, Karga H, Karaplis AC, Potts JTJ, Kronenberg HM, Segre GV, Abou-Samra AB, Jüppner H 1993 Identical complementary deoxyribonucleic acids encode a human renal and bone parathyroid hormone (PTH)/PTH-related peptide receptor. Endocrinology 132:21572165.
  • 16
    Abou-Samra AB, Jüppner H, Force T, Freeman MW, Kong XF, Schipani E, Urena P, Richards J, Bonventre JV, Potts JTJ 1992 Expression cloning of a common receptor for parathyroid hormone and parathyroid hormone-related peptide from rat osteoblast-like cells: A single receptor stimulates intracellular accumulation of both cAMP and inositol trisphosphates and increases intracellular free calcium. Proc Natl Acad Sci USA 89:27322736.
  • 17
    Partridge NC, Alcorn D, Michelangeli VP, Kemp BE, Ryan GB, Martin TJ 1981 Functional properties of hormonally responsive cultured normal and malignant rat osteoblastic cells. Endocrinology 108:213219.
  • 18
    Partridge NC, Kemp BE, Veroni MC, Martin TJ 1981 Activation of adenosine 3′,5′-monophosphate-dependent protein kinase in normal and malignant bone cells by parathyroid hormone, prostaglandin E2, and prostacyclin. Endocrinology 108:220225.
  • 19
    Civitelli R, Reid IR, Westbrook S, Avioli LV, Hruska KA 1988 PTH elevates inositol polyphosphates and diacylglycerol in a rat osteoblast-like cell line. Am J Physiol 255:E660E667.
  • 20
    De Miguel F, Martinez-Fernandez P, Guillen C, Valin A, Rodrigo A, Martinez ME, Esbrit P 1999 Parathyroid hormone-related protein (107-139) stimulates interleukin-6 expression in human osteoblastic cells. J Am Soc Nephrol 10:796803.
  • 21
    Nasu M, Sugimoto T, Kaji H, Kano J, Chihara K 1998 Carboxyl-terminal parathyroid hormone fragments stimulate type-1 procollagen and insulin-like growth factor-binding protein-5 mRNA expression in osteoblastic UMR-106 cells. Endocr J 45:229234.
  • 22
    Valin A, Garcia-Ocana A, De Miguel F, Sarasa JL, Esbrit P 1997 Antiproliferative effect of the C-terminal fragments of parathyroid hormone-related protein, PTHrP-(107-111) and (107-139), on osteoblastic osteosarcoma cells. J Cell Physiol 170:209215.
  • 23
    Martinez ME, Garcia-Ocana A, Sanchez M, Medina S, del Campo T, Valin A, Sanchez-Cabezudo MJ, Esbrit P 1997 C-terminal parathyroid hormone-related protein inhibits proliferation and differentiation of human osteoblast-like cells. J Bone Miner Res 12:778785.
  • 24
    Inomata N, Akiyama M, Kubota N, Jüppner H 1995 Characterization of a novel parathyroid hormone (PTH) receptor with specificity for the carboxyl-terminal region of PTH-(1-84). Endocrinology 136:47324740.
  • 25
    Takasu H, Baba H, Inomata N, Uchiyama Y, Kubota N, Kumaki K, Matsumoto A, Nakajima K, Kimura T, Sakakibara S, Fujita T, Chihara K, Nagai I 1996 The 69-84 amino acid region of the parathyroid hormone molecule is essential for the interaction of the hormone with the binding sites with carboxyl-terminal specificity. Endocrinology 137:55375543.
  • 26
    Lanske B, Karaplis AC, Lee K, Luz A, Vortkamp A, Pirro A, Karperien M, Defize LHK, Ho C, Mulligan RC, Abou-Samra AB, Juppner H, Segre GV, Kronenberg HM 1996 PTH/PTHrP receptor in early development and Indian hedgehog-regulated bone growth. Science 273:663666.
  • 27
    Lanske B, Divieti P, Kovacs CS, Pirro A, Landis WJ, Krane SM, Bringhurst FR, Kronenberg HM 1998 204 The parathyroid hormone (PTH)/PTH-related peptide receptor mediates actions of both ligands in murine bone. Endocrinology 139:51945.
  • 28
    Olstad OK, Reppe S, Løseth OP, Jemtland R, Gautvik KM 1997 Binding and cyclic AMP stimulation by N-terminally deleted human PTHs (3-84 and 4-84) in a homologous ligand receptor system. J Bone Miner Res 12:13481357.
  • 29
    Gabrielsen OS, Reppe S, Sæther O, Blingsmo OR, Sletten K, Gordeladze JO, Høgset A, Gautvik VT, Alestrøm P, Øyen TB, Gautvik KM 1990 Efficient secretion of human parathyroid hormone by saccharomyces cerevisiae. Gene 90:255262.
  • 30
    Usui H, Falk JD, Dopazo A, de Lecea L, Erlander MG, Sutcliffe JG 1994 Isolation of clones of rat striatum-specific mRNAs by directional tag PCR subtraction. J Neurosci 14:49154926.
  • 31
    Gautvik KM, de Lecea L, Gautvik VT, Danielson PE, Tranque P, Dopazo A, Bloom FE, Sutcliffe JG 1996 Overview of the most prevalent hypothalamus-specific mRNAs, as identified by directional tag PCR subtraction. Proc Natl Acad Sci USA 93:87338738.
  • 32
    Jemtland R, Lee K, Segre GV 1998 Heterogeneity among cells that express osteoclast-associated genes in developing bone. Endocrinology 139:340349.
  • 33
    Rian E, Jemtland R, Olstad OK, Gordeladze JO, Gautvik KM 1993 Synthesis of human parathyroid-hormone-related protein(1-141) in Saccharomyces cerevisiae. A correct amino-terminal processing vital for the hormone's biological activity is obtained by an ubiquitin fusion protein approach. Eur J Biochem 213:641648.
  • 34
    Marie PJ, de Vernejoul MC, Connes D, Hott M 1991 Decreased DNA synthesis by cultured osteoblastic cells in eugonadal osteoporotic men with defective bone formation. J Clin Invest 88:11671172.
  • 35
    Bringhurst FR, Jüppner H, Guo J, Urena P, Potts JTJ, Kronenberg HM, Abou-Samra AB, Segre GV 1993 Cloned, stably expressed parathyroid hormone (PTH)/PTH-related peptide receptors activate multiple messenger signals and biological responses in LLC-PK1 kidney cells. Endocrinology 132:20902098.
  • 36
    Fodstad Ø, Brøgger A, Bruland Ø, Solheim OP, Nesland JM, Pihl A 1986 Characteristics of a cell line established from a patient with multiple osteosarcoma, appearing 13 years after treatment for bilateral retinoblastoma. Int J Cancer 38:3340.
  • 37
    Kjønniksen I, Winderen M, Bruland Ø, Fodstad Ø 1994 Validity and usefulness of human tumor models established by intratibial cell inoculation in nude rats. Cancer Res 54:17151719.
  • 38
    Gordeladze JO, Lund HW, Jablonski G, Bruland ØS 1996 Diverse expression of G-proteins in human sarcoma cell lines with different osteogenic potential: Evidence for the involvement of Gi2 in cell proliferation. J Cell Biochem 60:95106.
  • 39
    Clohisy JC, Scott DK, Brakenhoff KD, Quinn CO, Partridge NC 1992 Parathyroid hormone induces c-fos and c-jun messenger RNA in rat osteoblastic cells. Mol Endocrinol 6:18341842.
  • 40
    Evans DB, Hipskind RA, Bilbe G 1996 Analysis of signaling pathways used by parathyroid hormone to activate the c-fos gene in human SaOS2 osteoblast-like cells. J Bone Miner Res 11:10661074.
  • 41
    Owen TA, Aronow M, Shalhoub V, Barone LM, Wilming L, Tassinari MS, Kennedy MB, Pockwinse S, Lian JB, Stein GS 1990 Progressive development of the rat osteoblast phenotype in vitro: Reciprocal relationships in expression of genes associated with osteoblast proliferation and differentiation during formation of the bone extracellular matrix. J Cell Physiol 143:420430.
  • 42
    Divieti P, Lanske B, Kronenberg HM, Bringhurst FR 1998 Conditionally immortalized murine osteoblasts lacking the type 1 PTH/PTHrP receptor. J Bone Miner Res 13:18351845.
  • 43
    Schilham MW, van Eijk M, van de Wetering M, Clevers HC 1993 The murine Sox-4 protein is encoded on a single exon. Nucleic Acids Res 21:2009.
  • 44
    Harada S, Nakayama K, Oshima H, Rodan GA 1998 Activation of SOX-9 response element by constitutively active type I BMP receptors via a SMADs dependent pathway. Bone 23(Suppl):T154. (abstract)
  • 45
    Lefebvre V, Huang W, Harley VR, Goodfellow PN, de Crombrugghe B 1997 SOX9 is a potent activator of the chondrocyte-specific enhancer of the pro alpha1(II) collagen gene. Mol Cell Biol 17:23362346.
  • 46
    Xie WF, Zhang X, Sakano S, Lefebvre V, Sandell LJ 1999 Trans-activation of the mouse cartilage-derived retinoic acid-sensitive protein gene by Sox9. J Bone Miner Res 14:757763.
  • 47
    Portale AA, Lonergan ET, Tanney DM, Halloran BP 1997 Aging alters calcium regulation of serum concentration of parathyroid hormone in healthy men. Am J Physiol 272:E139E146.
  • 48
    Volk SW, Leboy PS 1999 Regulating the regulators of chondrocyte hypertrophy. J Bone Miner Res 14:483486.
  • 49
    Grimsrud CD, Romano PR, D'Souza M, Puzas JE, Reynolds PR, Rosier RN, O'Keefe RJ 1999 BMP-6 is an autocrine stimulator of chondrocyte differentiation. J Bone Miner Res 14:475482.
  • 50
    Solloway MJ, Dudley AT, Bikoff EK, Lyons KM, Hogan BL, Robertson EJ 1998 Mice lacking Bmp6 function. Dev Genet 22:321339.
  • 51
    Schilham MW, Moerer P, Cumano A, Clevers HC 1997 Sox-4 facilitates thymocyte differentiation. Eur J Immunol 27:12921295.
  • 52
    Funk JL, Jones GV, Botham CA, Morgan G, Wooding P, Kendall MD 1999 Expression of parathyroid hormone-related protein and the parathyroid hormone parathyroid hormone-related protein receptor in rat thymic epithelial cells. J Anat 194:255264.