1,25-Dihydroxy Vitamin D 3 and Tri-iodothyronine Stimulate the Expression of a Protein Immunologically Related to Osteocalcin

Abstract

Osteocalcin (OC), a bone-specific protein, is a marker of late osteoblastic differentiation. Its expression is influenced by various growth factors and hormones. We investigated the effect of 1,25-dihydroxy vitamin D₃ (D₃) and tri-iodothyronine (T₃) on OC expression in osteoblast-like MC3T3-E1 cells. A heterologous OC green fluorescence protein (GFP) fusion vector was established and expressed to study possible effects on protein transport. Immunostaining of endogenous OC revealed a significant increase in the percentage of positive cells after D₃ and T₃ treatment. This was consistent for MC3T3-E1 cells as well as nonosteogenic NIH-3T3 and mammary carcinoma cells, but not for neuroblastoma cells. The perinuclear immunostaining corresponded to the NBD C₆ ceramide Golgi staining. Conversely, we found a strong induction of OC in MC3T3-E1 cells at the mRNA and protein levels only with T₃ and not with D₃. OC mRNA and protein expression was not detected in NIH fibroblasts. OC GFP transfection experiments indicate rapid transport and secretion of OC, because OC GFP was not found to be accumulated at intracellular compartments after hormone treatment. We conclude that the strong perinuclear immunostaining does not represent OC but a protein immunologically related to OC, as indicated by preabsorption experiments. The expression of this OC epitope-sharing protein is regulated by both D₃ and T₃ in the osteoblastic MC3T3-E1 and in nonosteogenic cells.

Keywords

osteocalcin immunohistochemistry green fluorescent protein tri-iodothyronine MC3T3-E1 cells

1,25-Dihydroxy vitamin D₃ (D₃) and tri-iodothyronine (T₃) exert their diverse effects on normal bone growth and development primarily through nuclear receptor-mediated processes (Haussler 1986; Evans 1988). In vitro studies suggest that both hormones play a role in promoting bone cell differentiation. This is indicated by increased bone-specific alkaline phosphatase activity, alterations in the synthesis of growth factors and matrix proteins, and inhibition of cell growth (Kurihara et al. 1986; Spiess et al. 1986; Kasono et al. 1988; Takeuchi et al. 1989; Kanatani et al. 1993; Ohishi et al. 1994; Varga et al. 1994; Klaushofer et al. 1995; Glantschnig et al. 1996; Luegmayr et al. 1996; Fratzl–Zelman et al. 1997).

Osteocalcin (OC) is a small (5.2–5.9 kD) protein that comprises 10–20% of the noncollagenous organic material of the extracellular bone matrix (Hauschka et al. 1975). To date, OC protein synthesis has been identified only in osteoblasts and odontoblasts (Price et al. 1976; Weinreb et al. 1990). However, low basal expression of OC mRNA was described in tissues other than bone (Fleet and Hock, 1994) and even in blood platelets (Thiede et al. 1994). More recently, ectopic secretion of osteocalcin by a myeloma cell line was reported and was found to be regulated by D₃ (Barille et al. 1996). The majority of OC secreted by osteoblasts is deposited into the bone matrix. Its most obvious property is the capacity for specific Gla (γ-car-boxyglutamic acid)-dependent binding of Ca²⁺. OC-deficient mice develop a phenotype characterized by higher bone mass without impaired bone resorption (Ducy et al. 1996). The level of OC in the blood circulation is considered to reflect primarily a spillover of osteoblast synthetic activity, and its concentration in serum has been increasingly used as an index of bone formation. However, the exact physiological functions of OC in bone metabolism are still not fully understood.

The effects of D₃ and T₃ on OC mRNA and protein expression have been studied in various chicken, mouse, rat, and human osteo-derived cell lines, indicating that the hormonal regulation is species- and differentiation-dependent (Pirskanen et al. 1991; Ohishi et al. 1994; Broess et al. 1995; Zhang et al. 1997; Varga et al. 1997). In this study we investigated the regulatory effects of D₃ and T₃ on OC synthesis and secretion in MC3T3-E1 cells. The intracellular transport processes were studied by use of a heterologous OC green fluorescence protein (GFP) fusion vector. Immunohistochemical localization and distribution of OC-positive cells were investigated using an antiserum against mouse OC.

Materials and Methods

Cell Culture

MC3T3-E1 mouse calvaria-derived osteoblast-like cells (Dr. Kumegawa; Sakado, Japan), NIH-3T3 mouse fibroblasts (ATCC; Rockville, MD), I34 mouse mammary carcinoma cells, and NB1 mouse neuroblastoma cells (Institute of Cancer Research and Tumor Biology; University of Vienna, Vienna, Austria) were cultured in αMEM supplemented with 4.5 g/liter glucose, 5% fetal calf serum (FCS), and 30 μg/ml gentamicin (regular medium). The cells were kept in humidified air under 5% CO₂ at 37C. They were subcultured twice a week using 0.001% pronase E and 0.02% EDTA in PBS. For the experiments, the cells were seeded at a density of 50,000/cm². Continuous treatment with T₃ (10⁻⁷ M; Sigma, Diesenhofen, Germany) or D₃ (10⁻⁸ M; Hoffmann LaRoche, Vienna, Austria) was started 24 hr after seeding. Medium was changed every third day.

Immunostaining of Endogenous OC

Cells were cultured on sterile glass coverslips placed in the culture dishes. At the respective time points (24 hr, and 3 and 8 days of culture) cells were rinsed with PBS, fixed with 4% paraformaldehyde (15 min), and permeabilized with methanol (5 min, -20C). After several rinses with PBS, cells were incubated with goat anti-mouse osteocalcin antiserum (RIA grade; BTI, Stoughton, MA; 1:100 in 3% BSA, 1 hr), followed by rabbit anti-goat IgG (Nordic Immunology; Til-burg, The Netherlands; 1:100 in 3% BSA, 45 min) and donkey anti-rabbit IgG conjugated with Cy3 (Accurate, West-bury, NY; 1:500 in 3% BSA, 45 min). According to the supplier, this OC antiserum is specific for the intact molecule, with the recognition site at the carboxyl terminal. The epitope is further along the molecule than the common Gla domain. In addition, for some experiments cell nuclei were counterstained with PicoGreen (Molecular Probes, Leiden, Netherlands; 1:1000 in A. dest., 45 min) after treatment with RNase A (Boehringer, Mannheim, Germany; 0.1 mg/ml in A. dest., 37C, 20 min). Coverslips were mounted on glass slides in Vectashield (Vector Laboratories; San Diego, CA) and examined using a confocal laser scanning microscope (Leica TCS^4D).

Percentages of immunopositive cells were determined by evaluating at least 300 cells for each time point and treatment. The specificity of the immune reaction was assessed by staining with nonimmune goat serum (BTI; 1:100) or by omitting the OC antiserum. Preabsorption of the OC antiserum with purified mouse OC (BTI) was performed with 5 μg, 500 ng, 50 ng, and 5 ng OC per ml of 1:100 diluted antiserum, overnight at 4C.

Golgi Staining with NBD C₆ Ceramide

MC3T3-E1 and NIH-3T3 cells grown on glass coverslips were incubated with NBD C₆ ceramide (Molecular Probes, Leiden, The Netherlands; 5 nM in 1% DF-BSA, 6C, 30 min), a vital stain for the Golgi apparatus. The cells were treated as described in Pagano et al. (1989), fixed, and examined as described above.

Western Immunoblotting

Denaturing SDS-PAGE of 20 μg protein extract/lane or 100 ng mouse OC standard (BTI) was performed on 12% or 15% gels, respectively, according to the method of Laemmli (1970), followed by semidry transfer to PVDF membranes (Immobilon-P; Millipore Intertech, Bedford, MA). The blots were probed with the goat anti-mouse OC antiserum (BTI) and were detected by a secondary antibody against goat IgG coupled to alkaline phosphatase (Sigma). For estimation of the molecular weight, a color marker (Amersham; Poole, UK) was used.

For determining the binding capacity of the OC antiserum to denatured OC, we applied 50 ng OC (BTI; diluted in 0.1% BSA) as well as 50 ng SDS- and/or heat-denatured OC on a nitrocellulose and/or PVDF membrane using a slot-blot filtration apparatus.

Radioimmunoassay

MC3T3-E1 and NIH-3T3 cells were seeded in 96-well culture plates and cultured as described above. At Days 3 and 4, respectively, the cell culture media were removed and kept at -20C until assayed by a competitive liquid-phase RIA (BTI). According to the supplier's description of the RIA, there is no crossreaction with bovine OC.

Northern Blotting

RNA electrophoresis, transfer, and hybridization were performed as described previously (Varga et al. 1997). As hybridization probes we used the mouse OC cDNA (Celeste et al. 1986) and the complete rat GAPDH. All probes were kindly provided by Dr. Busslinger (IMP; Vienna, Austria).

Construction of OC GFP Fusion Plasmid

The recombinant mouse OC-enhanced green fluorescence protein (GFP) fusion construct was obtained by fusing the mouse OC cDNA (Celeste et al. 1986) in a frame N-terminal to the GFP. For this a mouse OC cDNA restriction fragment (PstI/XmaI) of 210 bp, which encoded the 49-amino-acid leader peptide (prepro- and propeptide) and 16 of the 46 amino acids of the mature peptide, was ligated into a PstI/ XmaI cut pEGFP-N1 N-terminal fusion vector (Clontech; Palo Alto, CA) and was expressed under control of the immediate-early promoter of human cytomegalovirus. The correct fusion was confirmed by sequencing.

Figure 1

MC3T3-E1 osteoblast-like cells stained with the OC antiserum, picogreen, and NBD C₆ ceramide. Left row, control cells; middle row, T₃-treated cells; right row, D₃-treated cells. (A–C) Day 3: immunostaining with OC antiserum (red) and DNA counterstaining with picogreen (green). (D–F) Day 8: immunostaining with OC antiserum. (G–I) Day 8: Golgi staining with NBD C₆ ceramide. n, nucleus; arrows indicate perinuclear immunostaining. Bar = 10 μm.

Transfection Experiments

MC3T3-E1 cells were seeded on coverslips placed in culture dishes and were grown overnight in regular medium. The cells were transfected with 1 μg OC GFP fusion vector and 5 μg of transfection reagent DOSPER (Boehringer) in 700 μl αMEM for 6 hr, and this was continued for a further 18 hr after addition of 2.5 ml regular medium. Then the medium was completely replaced by regular medium with or without T₃ (10⁻⁶ to 10⁻⁹ M) or D₃ (10⁻⁷ to 10⁻¹⁰ M). A temperature block (15C for 1 or 2 hr; cf. Kaether and Gerdes 1995) was performed to visualize the fusion protein along its secretory pathway. During this period the medium was buffered with 20 mM HEPES (pH 7.4). Release of the temperature block was started by replacing HEPES-buffered culture medium with 37C regular medium. After fixation, at least 10,000 cells/coverslip were investigated to determine the number of transport-arrested cells after the respective hormone treatments.

Figure 2

MC3T3-E1 osteoblast-like cells treated with D₃. (A–D) Day 8. (A) Positive immunostaining with OC antiserum. Note binucleate cell (^∗) and intracellular immunopositive vesicles. (B) Inhibition of immunostaining by preabsorption of the OC antiserum with purified OC. (C) Negative staining with nonimmune serum. (D) Golgi staining of four binucleate cells with NBD C₆ ceramide. (E–F) Day 1. (E) Positive immunostaining with OC antiserum. (F) Phase-contrast image of E. N, nucleus. Bar = 10 μm.

Statistical Analysis

ANOVA (Statview 4.0; post hoc test, Scheffe) was used for statistical analysis.

Results

Immunostaining of Endogenous OC

MC3T3-E1 Mouse Osteoblast-like Cells. After 3 days of culture, cells reached a subconfluent monolayer. In controls, only few cells exhibited positive immunostaining (Figure 1A). In contrast, the majority of T₃- and D₃-treated cells exhibited intense perinuclear staining (Figures 1B and 1C). Three days of treatment with T₃ and D₃ resulted in a highly significant increase in the percentage of positive cells (cf. Figure 4; Co 16 ± 7%, T₃ 69 ± 5%, D₃ 82 ± 3%).

Figure 3

Nonosteogenic cells after 3 days of culture, immunostained with the OC antiserum. (A–D) NIH-3T3 fibroblasts. (A) Control cells; (B) T₃-treated cells; (C) D₃-treated cells; (D) negative control by omitting the primary antiserum. (E) I34 mammary carcinoma cells, D₃-treated. 3-D reconstruction from eight optical sections taken at 0.9-μm intervals. (F) NB1 neuroblastoma cells, D₃-treated; 3-D reconstruction from three optical sections taken at 1-μm intervals. n, nucleus; arrows indicate perinuclear immunostaining. Bars = 10 μm.

After 8 days of culture the percentage of immunopositive cells remained constant compared with Day 3 (cf. Figure 4; Co 17 ± 6%, T₃ 78 ± 5%, and D₃ 85 ± 3%). However, an increase in the intensity of the immunostaining in a distinct juxtanuclear region could be observed (Figures 1D-1F), especially after treatment with D₃ (Figure 1F). The tubulo-vesicular structure of the staining appeared consistent with an accumulation of the antigen in the Golgi region. This is supported by comparison with the NBD C₆ ceramide Golgi staining (Figures 1G-1I). In general, the stained Golgi stacks appeared to become enlarged with culture time. However, enlarged Golgi complexes were more abundant in cultures treated with T₃ and D₃. In D₃-treated cultures, many large binucleate cells could be observed, showing extensive strong immunostaining mostly located between the two nuclei (Figure 2A). A similar localization was observed after NBD C₆ ceramide Golgi staining (Figure 2D). Such conspicuous binucleate figures were found exclusively in D₃-treated and, less commonly, in T₃-treated MC3T3-E1 cells.

Preabsorption of the OC antiserum with purified OC (5 μg) resulted in complete inhibition of immunostaining (Figure 2B) comparable to the negative staining when nonimmune goat serum was used instead of the OC antiserum (Figure 2C). The perinuclear immunostaining gradually reappeared in response to increasing dilutions of the OC protein used for the preabsorption experiment (500 ng, 50 ng, and 5 ng OC), indicating the specificity of the antiserum for OC.

Figure 4

Effect of D₃ and T₃ treatment on the percentage of immunopositive cells as determined by counting at least 300 cells per time point and treatment. The stimulatory effect of the hormones was most profound in the osteoblastic MC3T3-E1 cells. Mean ± SEM; ^∗∗∗ p < 0.001 (control vs T₃ and D₃).

To evaluate whether both hormones also stimulate the immunostaining at the single-cell level in actively growing MC3T3-E1 cells, cells were seeded and immediately treated with T₃ or D₃. Only 24 hr (Day 1) after seeding, many single cells without contact with neighboring cells showed positive perinuclear and vesicular staining in the cytoplasm (Figures 2E and 2F), predominately in hormone-treated but also in control cells.

Figure 5

Immunoblotting with the OC antiserum. (A) SDS-PAGE of MC3T3-E1 protein extracts (Lanes 1–4) and mouse OC protein (Lane 5). Three distinct bands of ∼ 14, 18, and 40 kD were detected in the MC3T3-E1 protein extracts: Lane 1, Day 4 controls; Lane 2, Day 4 T₃-treated; Lane 3, Day 8 controls; Lane 4, Day 8 T₃-treated. The OC antiserum did not react with 100 ng mouse OC after SDS-PAGE (Lane 5). (B) Slot-blot analysis of native OC (upper slot) and denatured OC (lower slot) revealed that the antibodies failed to bind to denatured OC.

Figure 6

OC mRNA expression in MC3T3-E1 (Lanes 1–3) and NIH 3T3 (Lanes 4–6) cells. Northern blots of 10 μg RNA isolated from MC3T3-E1 or NIH 3T3 cells treated for 3 days with T₃ (Lanes 2 and 5) or D₃ (Lanes 3 and 6) or without hormones (Lanes 1 and 4) are shown. Control hybridizations were performed using a rat GAPDH probe. The 18S ribosomal RNA marker is indicated.

NIH-3T3 Mouse Fibroblasts, I34 Mouse Mammary Carcinoma Cells, and NB1 Mouse Neuroblastoma Cells. Surprisingly, some nonosteoblastic cell lines also showed similar positive results when stained with the OC antiserum.

NIH-3T3 Cells. After 3 days of culture, many control as well as T₃- and D₃-treated cells showed perinuclear immunostaining as well as immunoreactive vesicles throughout the cytoplasm (Figures 3A-3C). Again, immunostained regions corresponded to NBD C₆ ceramide-positive Golgi elements (data not shown). However, in contrast to the results with MC3T3-E1 cells, a higher percentage of untreated NIH-3T3 cells were immunopositive (cf. Figure 4; Co 75 ± 5%, T₃ 88 ± 2%, D₃ 95 ± 1%).

I34 mammary carcinoma cells also showed a similar positive immunostaining pattern after 3 days of culture (Figure 3e), but the stimulatory effects of D₃ or T₃ did not reach significance (cf. Figure 4; Co 23 ± 5%, T₃ 32 ± 3%, D₃ 42 ± 12%).

Neither untreated nor hormone-treated NB1 neuroblastoma cells showed any immunohistochemical reaction with the OC antiserum (Figure 3F).

Table 1.

Comparison of the amount of OC in the cell culture media of mouse osteoblasts and mouse fibroblasts after treatment with D₃ or T₃ a

MC3T3-E1			NIH-3T3
Control	T₃	D₃	Control	T₃	D₃
0.22 ± 0.1	0.43 ± 0.3	0.18 ± 0.2	0.19 ± 0.1	0.19 ± 0.1	0.18 ± 0.1
	∗	n.s.	n.s.	n.s.	n.s.
	∗∗

Cells were cultivated for 3 days (NIH-3T3) or for 4 days (MC3T3-E1) with or without D₃ or T₃. The values are given as μg/ml (mean ± SEM).

∗

p<0.01 (T₃ vs Co);

∗∗

p<0.001 (T₃ vs D₃). n.s., not significant.

Western Immunoblot

The specificity of the OC antiserum was tested by immunoblotting under denaturing conditions. Three distinct bands of approximately 14, 18, and 40 kD were detected in MC3T3-E1 (Figure 5A) and in NIH-3T3 protein extracts (data not shown), none of them having the appropriate molecular weight of osteocalcin (5.8 kD). The OC antiserum did not react with 100 ng mouse OC standard after SDS-PAGE (Figure 5A) but immunoprecipitated native iodinated mouse OC (data not shown). Slot-blots with native mouse OC and denatured OC revealed that the OC antiserum failed to bind to denatured OC, whereas strong binding to native osteocalcin could be observed (Figure 5B).

Figure 7

Mouse OC green fluorescent protein (GFP) fusion protein construct. The mouse OC cDNA sequence (bold letters) was fused in frame with the GFP coding sequence. The coding sequence is indicated by the derived amino acid sequence above. Numbering of the amino acids is with respect to residue 1 being the amino terminus of the mature OC. Numbering of nucleotides is shown at right. Indicated within shaded boxes are the amino acid residues of the 49-residue OC leader peptide (prepro-, propeptide) and 16 N-terminal residues of the mature peptide. Eight N-terminal amino acid residues of the GFP are marked within an unshaded box.

Northern Blot and RIA

Expression of OC was studied at both the mRNA and the protein level in MC3T3-E1 and NIH-3T3 cells. No OC mRNA was detected in untreated or D₃-treated cultures. T₃ significantly increased OC mRNA levels in the osteoblastic but not in the fibroblastic cells (Figure 6).

Measurement of the OC protein by RIA in the cell culture media exactly reflected the mRNA data (Table 1).

Transfection Experiments with OC GFP

By recombinant DNA techniques, we generated an OC GFP fusion protein. This fusion protein consisted of the OC prepropeptide and propeptide with specific recognition sites for signal peptidases and maturases in the secretory transport process (Figure 7). The correct recombination of the OC GFP was confirmed by sequencing and its expression at the mRNA level was proved by Northern blotting. The expression of the construct was not influenced by D₃ or T₃ treatment. The “functional integrity” of the fusion protein at the secretory pathway was demonstrated by incubating transiently transfected MC3T3-E1 cells for 1 or 2 hr at 15C. This arrest of protein transport due to low temperature resulted in strong perinuclear staining (Figure 8A). Release of the transport block led to a vesicular staining pattern (Figures 8B and 8C) and the complete disappearance of fluorescent OC GFP from the cells after appropriate incubation at 37C. Transfection of the pEGFP-N1 vector into MC3T3-E1 cells without the prepropeptide of OC resulted in uniform expression of GFP in the cytoplasm and nuclei (Figure 8D).

Different doses of T₃ and D₃ were used to study possible inhibitory effects on OC GFP transport and secretion. None of the hormones altered transport of OC GFP in comparison to untreated controls; 99.8% of the investigated cells did not show any accumulation of OC GFP in the Golgi complex (Figures 8E and 8F).

Discussion

Osteocalcin expression is a characteristic of the mature osteoblast phenotype (Stein et al. 1990; Weinreb et al. 1990). In the mouse osteoblast-like cell line MC3T3-E1, OC expression is progressively upregulated over culture time (Choi et al. 1996). Both D₃ and T₃ influence osteoblastic differentiation and therefore also regulate OC expression (Mahonen et al. 1990; Pirskanen et al. 1991; Ohishi et al. 1994; Pockwinse et al. 1995; Van Leeuwen et al. 1996; Varga et al. 1997). We initiated this study to investigate immunohistochemically the effects of D₃ and T₃ on localization and distribution of OC in MC3T3-E1 cells at different levels of phenotypic differentiation.

Treatment with D₃ and T₃ resulted in a highly significant increase in the percentage of immunopositive MC3T3-E1 cells that showed strong perinuclear staining. A comparable localization of OC has been shown previously (Malaval et al. 1994; Bos et al. 1996; Varga et al. 1997) and appeared consistent with an accumulation of the antigen at the Golgi region. In this study we show that between Day 3 and Day 8 the percentage of OC-positive cells remained constant. These immunohistochemical results were contradictory to the mRNA and RIA data, which demonstrated strong induction of OC expression only with T₃ and not with D₃. While this work was in progress, Zhang et al. (1997) reported that expression of OC in the mouse, unlike in rat and human, is downregulated by D₃. This was also confirmed in MC3T3-E1 cells (Lian et al. 1997). These findings support our suggestion that in mouse osteoblasts T₃ acts as an inducer of OC expression (Varga et al. 1997).

At this point we speculated that the contradiction of apparent immunolocalization of intracellular OC in the absence of detectable mRNA or protein secretion after treatment with D₃ might be based on transport-inhibitory phenomena. To test this possibility, we initiated experiments with GFP (Prasher et al. 1992), which is a fluorescent genetic tag that can be used as an excellent marker for monitoring fusion proteins in the cytosol (Kaether and Gerdes 1995; Gerdes and Kaether 1996).

Figure 8

Transport of OC GFP fusion protein along the secretory pathway in MC3T3-E1 cells. (A) OC GFP transport is arrested at the Golgi level after 2-hr incubation at 15C. Release of the 15C block resulted in vesicular transport of OC GFP after additional incubation at 37C for 10 min (B) and 30 min (C). (D) After transfection of pEGFP-N1 fusion vector, GFP is uniformly expressed in the cytoplasm and nuclei. (E) Treatment with 10⁻⁶ M T₃ did not result in accumulation of OC GFP at the Golgi compartment. (F) Phase-contrast image of E. Images (A–D) were constructed from four optical sections taken at 0.6-μm intervals. n, nucleus. Bars = 10 μm.

We established an OC GFP fusion vector and transiently transfected this fusion vector into MC3T3-E1 cells treated with or without D₃ and T₃. To demonstrate its functional integrity, the fusion protein was visualized along the secretory pathway by exposing the cells to low temperature (15C), which resulted in a simple and reversible arrest of intracellular protein transport at the Golgi level (Kaether and Gerdes 1995). In contrast to the arrest of transport caused by temperature block, treatment with D₃ and T₃ did not lead to accumulation of fluorescent OC GFP at the Golgi compartment. We therefore conclude that in control and in treated MC3T3-E1 cells OC GFP is rapidly transported and secreted into the culture medium, a process that also occurred after release of the temperature block.

The discrepancies between OC immunohistochemistry and the mRNA and RIA data, as well as the OC GFP transfection results found in MC3T3-E1 cells, suggest that the OC antiserum reacts with one or more immunogenic epitopes that are not solely specific to OC. Therefore, we tested it in nonosteogenic cells, i.e., NIH-3T3 mouse fibroblasts and I34 mouse mammary carcinoma cells. Unexpectedly, we found positive immunostaining identical to that of the osteoblastic cells. However, we did not detect OC mRNA or OC protein (RIA) in NIH-3T3 fibroblasts, a result that is in agreement with earlier reports showing the inability of skin fibroblasts to synthesize OC (Lian et al. 1989; Bradbeer et al. 1994). High OC expression is generally accepted to be present solely in osteoblasts of the mature phenotype. In our MC3T3-E1 cell cultures we find significant amounts of OC expression at the mRNA and protein level after treatment with T₃ only when cells had reached confluency (Varga et al. 1997). In contrast, in this study the percentage of immunopositive MC3T3-E1 cells remained constant from the early (single cells on Day 1) to the late (confluent cells on Day 8) cultures. This means that the number of positive cells was influenced only by the treatments but not by the culture time or the development of cell–cell or cell–matrix contacts.

Some evidence exists in the literature that OC anti-sera may recognize epitopes on proteins other than OC (Lian and Canalis 1986; Mahonen et al. 1990; Bradbeer et al. 1994). In agreement with these authors, we believe that the OC antiserum used in our study cross-reacts with a protein(s) that shares epitopes in common with OC, an assumption that is supported by the preabsorption experiments with purified OC. Moreover, by use of an OC GFP fusion vector we found that it is unlikely that large quantities of OC are accumulated at intracellular compartments. Instead, endogenous OC appears to be rapidly transported and secreted into the culture medium, as expected for a secretory protein that is mainly deposited into the bone matrix (Hauschka et al. 1975; Price et al. 1976). Therefore, our data indicate that the strong perinuclear OC immunostaining at least partially represents peptide epitopes related to but not identical to OC. The functional importance of such protein(s), which according to our data are influenced by D₃ and T₃ treatment, remains to be established.

Footnotes

Acknowledgement

We thank Prof Dr M.P.M. Erlee for critical reading of the manuscript.

References

Barille

Pellat–Deceunynck

Bataille

Amiot

(1996) Ectopic secretion of osteocalcin, the major non-collagenous bone protein, by the myeloma cell line NCI-H929. J Bone Miner Res 11:466–471

Bos

Van der Meer

Feyen

Herrmann–Erlee

(1996) Expression of the parathyroid hormone receptor and correlation with other osteoblastic parameters in fetal rat osteoblasts. Calcif Tissue Int 58:95–100

Bradbeer

Virdi

Serre

Beresford

Delmas

Reeve

Triffitt

(1994) A number of osteocalcin antisera recognize epitopes on proteins other than osteocalcin in cultured skin fibroblasts: implications for the identification of cells of the osteoblastic lineage in vitro. J Bone Miner Res 9:1221–1228

Broess

Riva

Gerstenfeld

(1995) Inhibitory effects of 1,25(OH)₂ vitamin D₃ on collagen type I, osteopontin, and osteocalcin gene expression in chicken osteoblasts. J Cell Biochem 57:440–451

Celeste

Rosen

Buecker

Kriz

Wang

Wozney

(1986) Isolation of the human gene for bone gla protein utilizing mouse and rat cDNA clones. EMBO J 5:1885–1890

Choi

Lee

Song

Park

Kim

Sohn

Ryoo

(1996) Expression pattern of bone-related proteins during osteoblastic differentiation in MC3T3-E1 cells. J Cell Biochem 61:609–918

Ducy

Desbois

Boyce

Pinerot

Story

Dunstan

Smith

Bonadio

Goldstein

Gundberg

Bradley

Karsenty

(1996) Increased bone formation in osteocalcin-deficient mice. Nature 382:448–452

Evans

(1988) The steroid and thyroid hormone receptor super-family. Science 240:889–895

Fleet

Hock

(1994) Identification of osteocalcin mRNA in nonosteoid tissue of rats and humans by reverse transcription-polymerase chain reaction. J Bone Miner Res 9:1565–1573

10.

Fratzl–Zelman

Hörandner

Luegmayr

Varga

Ellinger

Erlee

MPM

Klaushofer

(1997) Effects of triiodothyronine on the morphology of cells and matrix, the localization of alkaline phosphatase, and the frequency of apoptosis in long-term cultures of MC3T3-E1 cells. Bone 20:225–236

11.

Gerdes

Kaether

(1996) Green fluorescent protein: applications in cell biology. FEBS Lett 389:44–47

12.

Glantschnig

Varga

Klaushofer

(1996) Thyroid hormone (T₃) and retinoic acid induce the synthesis of insulin-like growth factor binding protein 4 in mouse osteoblastic cells. Endocrinology 137:281–286

13.

Hauschka

Lian

Gallop

(1975) Direct identification of the calcium-binding amino acid gamma-carboxyglutamic in mineralized tissue. Proc Natl Acad Sci USA 72:3925–3929

14.

Haussler

(1986) Vitamin D receptors: nature and function. Annu Rev Nutr 6:527–562

15.

Kaether

Gerdes

(1995) Visualization of protein transport along the secretory pathway using green fluorescent protein. FEBS Lett 369:267–271

16.

Kanatani

Sugimoto

Fukase

Chihara

(1993) Effect of 1,25-dihydroxyvitamin D₃ on the proliferation of osteoblastic MC3T3-E1 cells by modulating the release of local regulators from monocytes. Biochem Biophys Res Commun 190:529–535

17.

Kasono

Sato

Han

Fujii

Tsushima

Shizume

(1988) Stimulation of alkaline phospatase activity by thyroid hormone in mouse osteoblast-like cells (MC3T3-E1): a possible mechanism of hyperalkaline phosphatasia in hyperthyroidism. Bone Miner 4:355–363

18.

Klaushofer

Varga

Glantschnig

Fratzl–Zelman

Czerwenka

Leis

Koller

Peterlik

(1995) The regulatory role of thyroid hormones in bone cell growth and differentiation. J Nutr 125:1996S–2003S

19.

Kurihara

Ishizuka

Kiyoki

Haketa

Ikeda

Kumegawa

(1986) Effects of 1,25-dihydroxyvitamin D₃ on osteoblastic MC3T3-E1 cells. Endocrinology 118:940–947

20.

Laemmli

(1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T₄. Nature 227:680–685

21.

Lian

Canalis

(1986) Evidence for 1,25(OH)₂D₃ stimulation of a high molecular weight protein immunologically related to osteocalcin. J Bone Miner Res 1: S262

22.

Lian

Shalhoub

Aslam

Frenkel

Green

Hamrah

Stein

(1997) Species-specific glucocorticoid and 1,25-dihy-droxyvitamin D responsiveness in mouse MC3T3-E1 osteoblasts: dexamethasone inhibits osteoblast differentiation and vitamin D down-regulates osteocalcin gene expression. Endocrinology 138: 2117–2127

23.

Lian

Stewart

Puchacz

Mackowiak

Shaloub

Collart

Zambetti

Stein

(1989) Structure of the rat osteocalcin gene and regulation of vitamin D-dependent expression. Proc Natl Acad Sci USA 686:1143–1147

24.

Luegmayr

Varga

Frank

Roschger

Klaushofer

(1996) Effects of triiodothyronine on morphology, growth behaviour, and the actin cytoskeleton in mouse osteoblastic cells (MC3T3-E1). Bone 18:591–599

25.

Mahonen

Pirskanen

Keinänen

Mäenpää

(1990) Effect of 1,25(OH)₂D₃ on its receptor mRNA levels and osteocalcin synthesis in human osteosarcoma cells. Biochim Biophys Acta 1048:30–37

26.

Malaval

Modrowski

Gupta

Aubin

(1994) Cellular expression of bone-related proteins during in vitro osteogenesis in rat bone marrow stromal cell cultures. J Cell Physiol 158:555–572

27.

Ohishi

Ishida

Nagata

Yamauchi

Tsurumi

Nishikawa

Wakano

(1994) Thyroid hormone suppresses the differentiation of osteoprogenitor cells to osteoblasts, but enhances functional activities of mature osteoblasts in cultured rat calvaria cells. J Cell Physiol 161:544–552

28.

Pagano

Sepanski

Martin

(1989) Molecular trapping of a fluorescent ceramide analogue at the Golgi apparatus of fixed cells: interaction with endogenous lipids provides a trans-Golgi marker for both light and electron microscopy. J Cell Biol 109:2067–2079

29.

Pirskanen

Mahonen

Maenpaa

(1991) Modulation of 1,25(OH)₂D₃-induced osteocalcin synthesis in human osteosarcoma cells by other steroidal hormones. Mol Cell Endocrinol 7:149–159

30.

Pockwinse

Stein

Lian

Stein

(1995) Developmental stage-specific cellular responses to vitamin D and glucocorticoids during differentiation of the osteoblast phenotype: interrelationship of morphology and gene expression by in situ hybridization. Exp Cell Res 216:244–260

31.

Prasher

Eckenrode

Ward

Prendergast

Cormier

(1992) Primary structure of the Aequorea victoria green-fluorescent protein. Gene 111:229–233

32.

Price

Otuska

Poser

Kristaposis

Raman

(1976) Characterization of a gamma-carboxyglutamic containing protein from bone. Proc Natl Acad Sci USA 73:1447–1451

33.

Spiess

Price

Deftos

Manolagas

(1986) Phenotype-associated changes in the effects of 1,25-Dihydroxyvitamin D₃ on alkaline phosphatase and bone Gla-protein of rat osteoblastic cells. Endocrinology 118:1340–1346

34.

Stein

Lian

Owen

(1990) Relationship of cell growth to the regulation of tissue-specific gene expression during osteoblast differentiation. FASEB J 4:3111–3123

35.

Takeuchi

Matsumoto

Ogata

Shishiba

(1989) 1,25-Dihy-droxyvitamin D₃ inhibits synthesis and enhances degradation of proteoglycans in osteoblastic cells. J Biol Chem 264:18407–18413

36.

Thiede

Smock

Petersen

Grasser

Thompson

Nishimoto

(1994) Presence of messenger ribonucleic acid encoding osteocalcin, a marker of bone turnover, in bone marrow megakaryocytes and peripheral blood platelets. Endocrinology 135:929–937

37.

Van Leeuwen

Birkenhager

Van den Bemd

Pols

(1996) Evidence for coordinated regulation of osteoblast function by 1,25-dihydrovitamin D₃ and parathyroid hormone. Biochim Biophys Acta 1312:54–62

38.

Varga

Rumpler

Klaushofer

(1994) Thyroid hormones increase insulin-like growth factor mRNA levels in the clonal osteoblastic cell line MC3T3-E1. FEBS Lett 345:67–70

39.

Varga

Rumpler

Luegmayr

Fratzl–Zelman

Glantschnig

Klaushofer

(1997) Triiodothyronine a regulator of osteoblastic differentiation: depression of histone-H4, attenuation of c-fos/c-jun and induction of osteocalcin expression. Calcif Tiss Int 61:404–411

40.

Weinreb

Shinar

Rodan

(1990) Different pattern of alkaline phosphatase, osteopontin and osteocalcin expression in developing rat bone visualized by in situ hybridization. J Bone Miner Res 5:831–842

41.

Zhang

Ducy

Karsenty

(1997) 1,25-Dihydroxyvitamin D₃ inhibits osteocalcin expression in mouse through an indirect mechanism. J Biol Chem 272:110–116

1,25-Dihydroxy Vitamin D 3 and Tri-iodothyronine Stimulate the Expression of a Protein Immunologically Related to Osteocalcin

Abstract

Keywords

Materials and Methods

Cell Culture

Immunostaining of Endogenous OC

Golgi Staining with NBD C6 Ceramide

Western Immunoblotting

Radioimmunoassay

Northern Blotting

Construction of OC GFP Fusion Plasmid

Transfection Experiments

Statistical Analysis

Results

Immunostaining of Endogenous OC

Western Immunoblot

Northern Blot and RIA

Transfection Experiments with OC GFP

Discussion

Footnotes

Acknowledgement

References

Golgi Staining with NBD C₆ Ceramide