help button home button Endocrine Society Molecular Endocrinology ENDO 08 Sessions Library
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2006-0558
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow NURSA Molecule Pages Link
Right arrow Reprints, Permissions and Rights
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schedlich, L. J.
Right arrow Articles by Baxter, R. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schedlich, L. J.
Right arrow Articles by Baxter, R. C.
Molecular Endocrinology 21 (10): 2378-2390
Copyright © 2007 by The Endocrine Society

Insulin-Like Growth Factor Binding Protein-5 Interacts with the Vitamin D Receptor and Modulates the Vitamin D Response in Osteoblasts

Lynette J. Schedlich, Anita Muthukaruppan, Michelle K. O’Han and Robert C. Baxter

Kolling Institute of Medical Research, University of Sydney, Royal North Shore Hospital, Sydney, New South Wales 2065, Australia

Address all correspondence and requests for reprints to: Dr. Lyn Schedlich, Kolling Institute of Medical Research, Royal North Shore Hospital, St. Leonards, New South Wales 2065, Australia. E-mail: lyns{at}med.usyd.edu.au.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The 1,25 dihydroxyvitamin D3 [1,25(OH)2D3]-induced differentiation of osteoblasts comprises the sequential induction of cell cycle arrest at G0/G1 and the expression of bone matrix proteins. Reports differ on the effects of IGF binding protein (IGFBP)-5 on bone cell growth and osteoblastic function. IGFBP-5 can be growth stimulatory or inhibitory and can enhance or impair osteoblast function. In previous studies, we have shown that IGFBP-5 localizes to the nucleus and interacts with the retinoid receptors. We now show that IGFBP-5 interacts with nuclear vitamin D receptor (VDR) and blocks retinoid X receptor (RXR):VDR heterodimerization. VDR and IGFBP-5 were shown to colocalize to the nuclei of MG-63 and U2-OS cells and coimmunoprecipitate in nuclear extracts from these cells. Induction of osteocalcin promoter activity and alkaline phosphatase activity by 1,25(OH)2D3 were significantly enhanced when IGFBP-5 was down-regulated in U2-OS cells. Moreover, we found IGFBP-5 increased basal alkaline phosphatase activity and collagen {alpha}1 type 1 expression, and that 1,25(OH)2D3 was unable to further induce the expression of these bone differentiation markers in MG-63 cells. Expression of IGFBP-5 inhibited MG-63 cell growth and caused cell cycle arrest at G0/G1 and G2/M. Furthermore, IGFBP-5 reduced the effects of 1,25(OH)2D3 in blocking cell cycle progression at G0/G1 and decreased the expression of cyclin D1. These results demonstrate that IGFBP-5 can interact with VDR to prevent RXR:VDR heterodimerization and suggest that IGFBP-5 may attenuate the 1,25(OH)2D3-induced expression of bone differentiation markers while having a modest effect on the 1,25(OH)2D3-mediated inhibition of cell cycle progression in bone cells.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
IGF BINDING PROTEINS (IGFBPs) play an integral role in modifying the effects of IGFs on a wide variety of cell types (1). Recent evidence suggests that a number of IGFBPs also have effects on cell growth that are independent of IGF signaling. In the case of IGFBP-3 and -5, these IGF-independent effects can be either growth inhibitory or stimulatory, depending on the cellular context. IGFBP-5 is the most abundantly expressed IGFBP in bone cells and is known to play an important role in bone cell function. The majority of studies have shown that the IGF-dependent and -independent effects of IGFBP-5 on bone cell growth are stimulatory (2, 3). The ability of IGFBP-5 to enhance IGF signaling is thought to relate to IGFBP-5 accumulating in bone through its binding to osteopontin and hydroxyapatite (4). This then serves to increase the local pool of IGFs and stimulate bone cell growth. However, IGFBP-5 has also been shown to stimulate bone cell growth in the absence of increased type I IGF receptor:IGF-I binding (3). These IGF-independent effects of IGFBP-5 were also demonstrated in IGF-I knockout mice, where the administration of IGFBP-5 caused an increase in bone formation (2). The mechanisms for these effects are not fully understood, but may be dependent upon coupling to other signaling systems.

The classic role of the vitamin D endocrine system is to stimulate calcium absorption in the intestine, thus maintaining normocalcemia and indirectly regulating bone mineralization (5). The actions of vitamin D are mediated through the vitamin D receptor (VDR), which acts as a ligand-activated transcription factor to regulate the expression of target genes. The VDR heterodimerizes with retinoid X receptor (RXR) (or less commonly forms VDR:VDR homodimers) and associates with the transcriptional complex on the promoter of target genes. Many tissues that do not have a role in bone and mineral homeostasis express the VDR and respond to 1,25 dihydroxyvitamin D3 [1,25-(OH)2D3], including both normal and neoplastic cells derived from bone, breast, colon, and prostate (6). 1,25-(OH)2D3 inhibits the growth and stimulates the differentiation of these cell types by regulating the expression of many growth factors and their receptors, cell-cycle intermediates, transcription factors, and genes involved in tissue-specific expression.

1,25-(OH)2D3 induces the differentiation of osteoblasts through the sequential induction of cell cycle arrest (7), the maturation of extracellular matrix and finally bone mineralization (8). The MG-63 cell line is considered to represent a population of undifferentiated human osteoblast-like cells and represents a good model for examining the initial phases of human bone cell differentiation. 1,25-(OH)2D3 or its synthetic analogs inhibit osteoblast proliferation by targeting various key regulators responsible for the G1/S transition, including the cyclin-dependent kinase (cdk) inhibitors p21CIP1/WAF1 and p27KIP1 (7, 9). Osteoblasts then respond to 1,25-(OH)2D3 by expressing the bone matrix proteins collagen type I, alkaline phosphatase (ALP) activity, osteocalcin, and osteopontin (10).

In previous studies, we have shown that IGFBP-3 interacts with RXR{alpha} and retinoic acid receptor (RAR)-{alpha} to modulate all-trans retinoic acid (atRA)-signaling in breast cancer cells (11). We now show that IGFBP-5 interacts with the VDR both in vitro and in living cells, and have examined the role of IGFBP-5 in the 1,25-(OH)2D3-signaling pathway in osteoblasts. Our results suggest that IGFBP-5 may alter the 1,25-(OH)2D3-induced expression of bone matrix proteins in osteoblast-like cells.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Characterization of the Interaction between IGFBP-5 and VDR
We have investigated the ability of IGFBP-5 to associate with RXR{alpha} and VDR using a glutathione S-transferase (GST) pull-down assay (Fig. 1AGo). Unfused GST, GST-RXR{alpha}, and GST-VDR proteins were immobilized on beads and incubated with IGFBP-5 (lanes 1–3, respectively), and the bound protein was identified by ligand blotting using 125I-IGF-I. In comparison to GST alone, GST-RXR{alpha} and GST-VDR both bound IGFBP-5 strongly. To confirm that the protein detected by ligand blotting was IGFBP-5, we conducted similar experiments where GST alone and GST-VDR were used to coprecipitate IGFBP-5 (Fig. 1AGo, lanes 4 and 5, respectively) and the bound protein was identified by Western immunoblot analysis using specific IGFBP-5 antiserum. The specificity of the interaction between VDR and members of the IGFBP family of proteins was investigated by incubating IGFBP-1, IGFBP-2 and IGFBP-5 with immobilized GST-VDR (Fig. 1BGo, lanes 2, 4 and 6, respectively). The bound proteins were detected by ligand blotting and compared with their respective standards (lanes 1, 3, and 5). Neither IGFBP-1 nor IGFBP-2 bound VDR, demonstrating the specificity of the interaction between VDR and IGFBP-5. As an additional negative control we investigated the ability of IGFBP-5 to bind an unrelated GST-fusion protein (Fig. 1CGo). For this, we used the steroid receptor coactivator 1 (SRC-1), which interacts with steroid hormone receptors to regulate transcriptional activation of target genes (12). The results showed that compared with binding to GST-RXR{alpha} (lane 3), GST-SRC-1 did not interact with IGFBP-5 (lane 2). We had previously shown that IGFBP-3 also bound VDR (our unpublished observation) and now sought to determine the relative affinities of IGFBP-3 and IGFBP-5 for VDR (Fig. 1DGo). An equimolar mixture of IGFBP-3 and -5 was incubated with decreasing amounts of GST-VDR (lanes 3–6) and bound IGFBP-3 and IGFBP-5 were visualized by ligand blotting and compared with IGFBP-3 (lane 1) and IGFBP-5 (lane 2) alone. As GST-VDR became limiting, IGFBP-3 binding appeared to decrease, whereas IGFBP-5 binding remained constant. Quantitation of these data (Fig. 1EGo) suggests that IGFBP-5 has the higher relative affinity for VDR than IGFBP-3.


Figure 1
View larger version (48K):
[in this window]
[in a new window]

  		Fig. 1. Characterization of the Interaction between IGFBP-5 and VDR

A, IGFBP-5 (0.5 µg) was incubated with GST (lane 1), GST-RXR{alpha} (lane 2) or GST-VDR (lane 3) immobilized on glutathione-Sepharose beads. The bound IGFBP-5 was detected by ligand blotting using 125I-IGF-I and compared with 0.5 µg IGFBP-5 standard. IGFBP-5 was coprecipitated with GST (lane 4) or GST-VDR (lane 5) and identified as IGFBP-5 by Western immunoblot analysis. B, IGFBP-1 (lane 2), IGFBP-2 (lane 4) and IGFBP-5 (lane 6) were incubated with GST-VDR and the bound IGFBPs were detected by ligand blotting using 125I-IGF-I and compared with their respective standards (lanes 1, 3, and 5). C, IGFBP-5 was incubated with GST-SRC-1 (lane 2) or GST-RXR{alpha} (lane 3) and the amount of bound IGFBP-5 was compared with IGFBP-5 standard (lane 1). D, Equimolar amounts of IGFBP-3 and IGFBP-5 were incubated with decreasing amounts of GST-VDR (lanes 3–6). E, The proportion of bound IGFBP-3 and IGFBP-5 was compared with IGFBP-3 (lane 1) and IGFBP-5 (lane 2) alone and quantitated as described in Materials and Methods. F, IGFBP-5 was incubated with GST (lane 1) or GST-VDR in the absence (lane 2) or the presence of 10 nM (lane 3) or 100 nM (lane 4) 1,25-(OH)2D3, and bound IGFBP-5 was detected by ligand blotting. IGFBP-5 was incubated with GST (lanes 5 and 9) or GST-VDR in the absence (lanes 6 and 10) or the presence of a 2.5-fold molar excess (lanes 7 and 11) or a 10-fold molar excess (lanes 8 and 12) of IGF-I, and bound IGFBP-5 was detected. IGF-I was either added to GST-VDR at the same time as IGFBP-5 (lanes 7 and 8) or preincubated with IGFBP-5 for 1 h before incubation with GST-VDR (lanes 11 and 12). BP, IGFBP.

 
Next we determined whether 1,25-(OH)2D3 or IGF-I could interfere with the interaction between IGFBP-5 and VDR (Fig. 1FGo). Immobilized GST-VDR was incubated with IGFBP-5 in the absence (lanes 2 and 6) or the presence of 10 nM (lane 3) or 100 nM (lane 4) 1,25-(OH)2D3 or a 2.5-fold (lane 7) and 10-fold (lane 8) molar excess of IGF-I and the amount of bound IGFBP-5 was determined after SDS-PAGE by ligand blotting using 125I-IGF-I. We found that IGF-I was able to compete with VDR for binding to IGFBP-5, reducing the VDR:IGFBP-5 interaction by 25% and 35% in the presence of a 2.5- and 10-fold molar excess, respectively. In contrast, the presence of 1,25-(OH)2D3 had no effect on the VDR:IGFBP-5 interaction. Because IGFBP-5 and IGF-I form a very slowly dissociating complex (13), we investigated whether preincubation of IGF-I with IGFBP-5 before its addition to GST-VDR would have a more profound effect on the VDR:IGFBP-5 interaction (Fig. 1FGo). The results showed that when IGFBP-5 was preincubated for 1 h with a 2.5-fold (lane 11) or 10-fold (lane 12) molar excess of IGF-I, the VDR:IGFBP-5 interaction (lane 10) was completely eliminated with binding equivalent to that seen for the unfused GST control (lane 9).

IGFBP-5 Prevents the Heterodimerization of RXR{alpha} with VDR
To determine whether IGFBP-5 could sequester RXR{alpha} and/or VDR and block the formation of RXR{alpha}:VDR heterodimers, immobilized GST-RXR{alpha} was preincubated with or without increasing amounts of IGFBP-5 and then His6-VDR was added to each reaction. The amount of His6-VDR that was able to bind RXR{alpha} was determined by Western immunoblotting (Fig. 2AGo). In the absence of IGFBP-5, VDR coprecipitated with RXR{alpha}, demonstrating that heterodimerization had occurred. However, in the presence of IGFBP-5 the amount of His6-VDR that bound to RXR{alpha} was significantly reduced. Quantitation of these data showed that a 2- and 4-fold molar excess of IGFBP-5 (with respect to His6-VDR) significantly reduced heterodimerization by 58% (P < 0.02) and by 76% of control (P < 0.01), respectively (Fig. 2BGo).


Figure 2
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 2. IGFBP-5 Inhibits RXR:VDR Heterodimerization

A, Immobilized GST-RXR{alpha} was preincubated without or with increasing amounts of IGFBP-5 (1-, 2-, and 4-fold molar excess with respect to His6-VDR). His6-VDR was then added to each reaction and bound His6-VDR was analyzed by Western immunoblotting and compared with His6-VDR run as standard. B, Data from two independent experiments was quantified as described in Materials and Methods. Values are expressed as means ± SD. *, P < 0.02; **, P < 0.01 vs. nil molar excess of IGFBP-5.

 
IGFBP-5 Localizes to the Nucleus in Human Osteoblast-Like Cells Where It Interacts with VDR
To determine whether the biochemical interaction described between IGFBP-5 and VDR in vitro might be relevant in living cells, we compared their subcellular distribution in MG-63 and U2-OS cells (Fig. 3Go). Cells expressing recombinant IGFBP-5 and VDR were stimulated with 1,25-(OH)2D3 for 4 h before fixation, and VDR and IGFBP-5 were detected using specific antibodies and fluorescent microscopy. The immunocytochemical study demonstrates that both VDR (A) and IGFBP-5 (B) localized to the nucleus in MG-63 (upper panels). IGFBP-5 (B) was present in the nuclei of all U2-OS cells (lower panels), with evidence of strong perinuclear labeling in a percentage of cells. This is consistent with the presence of IGFBP-5 in the secretory pathway. Only a proportion of the U2-OS cells contained nuclear VDR (A), which colocalized with IGFBP-5. Interestingly, there appeared to be a strong signal for VDR in the cytoplasm of U2-OS cells. The antibodies were specific for VDR and IGFBP-5 because no signal was detected when they were replaced with rabbit (D) or chicken (E) IgGs, respectively.


Figure 3
View larger version (41K):
[in this window]
[in a new window]

  		Fig. 3. IGFBP-5 Colocalizes with VDR to the Nuclei of MG-63 and U2-OS Cells

MG-63 (upper panel) and U2-OS cells (lower panel) expressing human recombinant IGFBP-5 and VDR were treated with 1,25-(OH)2D3 for 4 h and then fixed and permeabilized. The cells were incubated sequentially with a rabbit antihuman VDR antibody and antirabbit IgG conjugated with Alexa594 to detect VDR (A), and chicken antihuman IGFBP-5 antiserum and antichicken IgG conjugated with FITC to detect IGFBP-5 (B). Control experiments replaced the VDR and IGFBP-5-specific antibodies with purified rabbit IgG (D) or chicken IgY (E), respectively, followed by incubating with either antirabbit Alexa594 or antichicken FITC. The nuclei are identified by 4',6-diamidino-2-phenylindole (DAPI) staining (C and F). Representative images are shown from at least two independent experiments. Scale bar, 10 µm.

 
To confirm that IGFBP-5 interacts directly with VDR in the nucleus of human osteoblast-like cells, we performed a series of coimmunoprecipitation experiments using nuclear and cytoplasmic extracts. These extracts were prepared from MG-63 (Fig. 4Go, A and B) and U2-OS (Fig. 4Go, C and D) cells expressing recombinant human IGFBP-5, and the cells were treated with 1,25-(OH)2D3 for 4 h before harvesting. VDR was detected in the cytoplasmic and nuclear input samples (A and C; lanes 1 and 4, respectively), although it was in relatively low abundance in the nuclear extract derived from U2-OS cells. This is consistent with the observation that VDR is localized to only a proportion of US-OS cell nuclei (Fig. 3AGo). VDR was immunoprecipitated from nuclear extract (lanes 6), but not from cytoplasmic extracts (lanes 3) nor in the IgG controls (lanes 2 and 5). IGFBP-5 was present in the nuclear input samples (B and D; lanes 4) and coimmunoprecipitated with VDR in nuclear extracts (lanes 6). The identification of this IGFBP as IGFBP-5 was based on its apparent molecular weight (identical with the IGFBP-5 standard) and the fact that neither cell type expresses detectable levels of IGFBP-3.


Figure 4
View larger version (35K):
[in this window]
[in a new window]

  		Fig. 4. IGFBP-5 Coimmunoprecipitates with the VDR in Nuclear Extracts from Osteoblast-Like Cells

MG-63 (A and B) and U2-OS (C and D) cells expressing recombinant human IGFBP-5 were treated with 1,25-(OH)2D3 for 4 h before the extraction of nuclear and cytoplasmic fractions. The soluble cytoplasmic (lanes 2 and 3) and nuclear extracts (lanes 5 and 6) (100 µg) were subjected to immunoprecipitation (IP) using rat antihuman VDR antibodies (lanes 3 and 6) or purified rat IgG (lanes 2 and 5) as a control. The VDR and IGFBP-5 in the protein complexes recovered by immunoprecipitation were detected by Western immunoblot (IB) analysis using rabbit antihuman VDR antibodies (A and C) and ligand blotting for IGFBP-5 using 125I-IGF-II (B and D), respectively. Ten micrograms of the soluble cytoplasmic (lane 1) and nuclear (lane 4) extracts (10% input) were also analyzed to evaluate the relative abundance of the target proteins.

 
IGFBP-5 Attenuates Ligand-Induced Transactivation of the VDR Response Element (VDRE) and ALP Activity in Osteoblasts
The ability of IGFBP-5 to modulate 1,25-(OH)2D3-induced transactivation of the osteocalcin promoter was investigated in U2-OS cells. U2-OS cells were transiently cotransfected with a luciferase reporter plasmid (expressing the firefly luciferase gene) containing the VDREs from the human osteocalcin promoter and pRL-TK vector (expressing the Renilla luciferase gene). The cells were subsequently transduced with adenovirus expressing recombinant human IGFBP-5 or vector control sequences and stimulated with 1,25-(OH)2D3 or vehicle for 24 h. Conditioned media were collected and cell lysates were harvested and assayed for luciferase activity. The level of secreted and nuclear IGFBP-5 was determined by RIA on the conditioned media and nuclear extracts, respectively. Secreted IGFBP-5 increased from 40 ng/ml in vector control cells to 450 ng/ml in cells expressing recombinant human IGFBP-5 (Fig. 5AGo). This IGFBP-5 concentration is comparable to that in human serum (14) and in media conditioned by some untransfected cell lines, e.g. T47D human breast cancer cells (Baxter, R. C., unpublished data). However, stimulation with 1,25-(OH)2D3 had no effect on the expression of IGFBP-5 in U2-OS cells. Nuclear IGFBP-5 levels were 0.5 ng/µg nuclear extract in vector control cells and increased to 3.5 ng IGFBP-5/µg nuclear extract in cells expressing recombinant IGFBP-5 (data not shown).


Figure 5
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 5. Endogenous IGFBP-5 Attenuates Ligand-Induced Transactivation of the VDRE

A and B, U2-OS cells were transiently cotransfected with a luciferase reporter plasmid containing the VDREs from the human osteocalcin promoter and the pRL-TK vector, and then transduced with adenoviral constructs expressing human IGFBP-5 or vector control sequences. Cells were treated with vehicle (white bars) or 1,25-(OH)2D3 (black bars) for 24 h, and conditioned media were assayed for IGFBP-5 expression (A) and lysates were assayed for luciferase activity (B). Each experimental condition was performed in triplicate and applied in three independent experiments. C and D, U2-OS cells were transiently transfected with the osteocalcin promoter:luciferase reporter construct and the pRL-TK vector and either an siRNA targeted to IGFBP-5 or a scrambled control siRNA. Cells were treated with vehicle (white bars) or 1,25-(OH)2D3 (black bars) for 24 h, and conditioned media were assayed for IGFBP-5 expression (C) and lysates were assayed for luciferase activity (D). E, Cell lysates were prepared from U2-OS cells expressing control siRNAs [treated with vehicle (open squares) or 1,25-(OH)2D3 (filled squares)] or siRNAs targeted to IGFBP-5 [treated with vehicle (open circle) or 1,25-(OH)2D3 (filled circle)] and assayed for ALP activity. Each experimental condition was performed in triplicate and applied in two independent experiments. Data from the time courses were analyzed by Repeated Measures ANOVA followed by Fisher’s protected least significant difference test using Statview 4.02 (Abacus Concepts, Inc., Berkeley, CA). Values are expressed as means ± SD. *, P < 0.01, for cells transfected with IGFBP-5 siRNA ± 1,25-(OH)2D3; **, P < 0.0001, for 1,25-(OH)2D3-treated cells where IGFBP-5 is recombinantly expressed or down-regulated compared with control cells.

 
We were unable to discern any effect of recombinantly expressed IGFBP-5 on osteocalcin promoter activity under basal non-1,25-(OH)2D3 stimulated conditions (Fig. 5BGo). As has been previously described osteocalcin promoter activity was enhanced (3-fold) in control cells treated with 1,25-(OH)2D3. However, in cells overexpressing IGFBP-5, the ability of 1,25-(OH)2D3 to induce promoter activity was lost (P < 0.0001). Similar experiments were carried out where the addition of increasing amounts of recombinant IGFBP-5 (100–1000 ng/ml) to the culture medium was substituted for the expression of endogenous IGFBP-5. We found that exogenously added IGFBP-5 had no effect on the 1,25-(OH)2D3-induction of the VDREs in the osteocalcin promoter (data not shown). These data demonstrate that endogenously expressed (but not exogenously added) IGFBP-5 inhibits the promoter’s response to 1,25-(OH)2D3.

To determine whether natural endogenously expressed IGFBP-5 was also inhibitory, we studied the effects of knocking down the expression of IGFBP-5 on the 1,25-(OH)2D3 response. U2-OS cells were transiently cotransfected with the osteocalcin promoter:luciferase construct, the pRL-TK vector and either a small interfering RNA (siRNA) targeted against IGFBP-5 or a scrambled control siRNA. The cells were treated with 1,25-(OH)2D3 or vehicle for 24 h. Using real-time RT-PCR, IGFBP-5 mRNA levels were shown to be reduced by 85% in cells transfected with the IGFBP-5 siRNA compared with control cells (data not shown). This translated into a reduction in secreted IGFBP-5 from 60 ng/ml in scrambled control cells to 10 ng/ml in cells transfected with the IGFBP-5 siRNA (Fig. 5CGo). Down-regulation of IGFBP-5 expression had no significant effect on the basal activity of the osteocalcin promoter (Fig. 5DGo). Osteocalcin promoter activity was induced 3-fold by 1,25-(OH)2D3 in scrambled siRNA control cells, consistent with the level of induction observed for vector virus control cells (Fig. 5BGo). However, in the cells where IGFBP-5 had been effectively knocked down, osteocalcin promoter activity was significantly enhanced to almost 6-fold (P < 0.0001) (Fig. 5DGo). Similar findings were observed for a second siRNA targeting the 3'-untranslated region of IGFBP-5 (data not shown). These results indicate clearly that in U2-OS cells, endogenous IGFBP-5 exerts a tonic inhibitory effect on 1,25-(OH)2D3 responsiveness, which can be relieved by IGFBP-5 down-regulation.

Although U2-OS cells express VDR and are capable of eliciting a 1,25-(OH)2D3 response on an isolated promoter-reporter construct (see above), they are not growth inhibited by 1,25-(OH)2D3 (our unpublished observations) nor is ALP activity, collagen {alpha}1 type 1 or osteocalcin expression induced by 1,25-(OH)2D3 (15). Therefore, we investigated whether depletion of IGFBP-5 from these cells would render them sensitive to 1,25-(OH)2D3. U2-OS cells were transfected with siRNAs against IGFBP-5 or control siRNAs before treatment with 1,25-(OH)2D3 for 72 h and assayed for ALP activity (Fig. 5EGo). As previously reported, 1,25-(OH)2D3 did not induce ALP activity in cell lysates derived from control cells, but we observed a significant increase in the induction of ALP activity in cells where IGFBP-5 expression had been down-regulated (P < 0.01). This suggests that the presence of IGFBP-5 in these cells may, at least in part, be responsible for their lack of responsiveness to 1,25-(OH)2D3.

Effect of IGFBP-5 on the Induction of Bone Differentiation Markers by 1,25-(OH)2D3
To further investigate the role of IGFBP-5 in modulating bone cell differentiation, we examined its effect on 1,25-(OH)2D3-mediated induction of the bone differentiation markers, ALP activity and collagen {alpha}1 type 1 expression in MG-63 cells. Because these cells do not express IGFBP-5 (our unpublished observation), we were unable to examine its effects in a knock-down system. Therefore, MG-63 cells expressing recombinant human IGFBP-5 or vector control sequences were stimulated with 1,25-(OH)2D3 or vehicle for 48 h and cell lysates were assayed for ALP activity (Fig. 6AGo). 1,25-(OH)2D3 and IGFBP-5 independently induced ALP activity by 1.8-fold (P < 0.0001) and 1.6-fold (P < 0.0001), respectively, compared with untreated vector control cells (Fig. 6BGo). Although 1,25-(OH)2D3 was still able to induce ALP activity in cells expressing IGFBP-5 by 1.2-fold (P = 0.002), this induction was significantly less than that seen in vector expression cells (P = 0.013) (Fig. 6CGo). Thus, IGFBP-5, while itself stimulating ALP activity, is able to effectively decrease ALP responsiveness to 1,25-(OH)2D3.


Figure 6
View larger version (32K):
[in this window]
[in a new window]

  		Fig. 6. IGFBP-5 Induces Bone Differentiation Markers but Attenuates Their Induction by 1,25-(OH)2D3

A, Cell lysates were prepared from MG-63 cells expressing vector control sequences [treated with vehicle (open squares) or 1,25-(OH)2D3 (filled squares)] or recombinant human IGFBP-5 [treated with vehicle (open circle) or 1,25-(OH)2D3 (filled circle)] and assayed for ALP activity at 90 and 180 min. Each experimental condition was performed in triplicate and applied in two independent experiments. B, Each sample used to generate the data in panel A was expressed as the rate of ALP activity/mg protein for cells treated with vehicle (white bars) or 1,25-(OH)2D3 (black bars) in the presence or absence of IGFBP-5. C, Fold induction of ALP activity by 1,25-(OH)2D3 (from B) in vector control cells and IGFBP-5 expression cells. Values are expressed as means ± SD. *, P = 0.013; **, P = 0.002; ***, P < 0.0001. D, Control MG-63 cells (lanes 1 and 2) or cells expressing vector sequences (lanes 3 and 4) or IGFBP-5 (lanes 5 and 6) were treated with vehicle (lanes 1, 3, and 5) or 1,25-(OH)2D3 (lanes 2, 4, and 6) for 48 h and the expression of collagen {alpha}1 type 1 in cell lysates was analyzed by Western immunoblotting. {alpha}-Tubulin acted as a loading control. E, Data from two independent experiments (from D above) were quantified as described in Materials and Methods [vehicle- (white bars) and 1,25-(OH)2D3-treated samples (black bars)]. Values are expressed as means ± SD. *, P < 0.0001.

 
To assess the role of IGFBP-5 in the 1,25-(OH)2D3-induced expression of collagen {alpha}1 type 1, MG-63 cells were treated as described above and collagen {alpha}1 type 1 expression was detected using Western immunoblotting (Fig. 6DGo) and quantitated as described in Materials and Methods (Fig. 6EGo). As expected, 1,25-(OH)2D3 induced the expression of collagen {alpha}1 type 1 in wild-type MG-63 cells (lanes 1 and 2) and cells expressing vector control sequences (lanes 3 and 4) by 2.7-fold (P < 0.0001). Interestingly, IGFBP-5 also induced the expression of collagen {alpha}1 type 1 by 2.7-fold under basal conditions (lane 5) when compared with cells expressing vector sequences only (lane 3) (P < 0.0001). However, in cells expressing IGFBP-5, 1,25-(OH)2D3 was unable to further induce collagen {alpha}1 type 1 synthesis (lane 6), but rather we observed a small, nonsignificant decrease in collagen {alpha}1 type 1 expression under these experimental conditions. Thus, similar to ALP activity, the collagen {alpha}1 type 1 responsiveness to 1,25-(OH)2D3 stimulation is attenuated by IGFBP-5.

Expression of IGFBP-5 Has a Modest Effect on 1,25-(OH)2D3-Induced Cell Cycle Arrest in Osteoblast-Like Cells
1,25-(OH)2D3 can induce cellular differentiation in osteoblast-like cells by inducing cell cycle arrest at G1 and preventing entry into S phase (7). To determine whether expression of IGFBP-5 could relieve the 1,25-(OH)2D3-induced block in cell cycle progression, wild-type MG-63 cells or MG-63 cells expressing recombinant human IGFBP-5 or vector control sequences were treated with vehicle or 1,25-(OH)2D3 for 48 h. Cells were harvested to determine viable cell numbers, and the effects of treatment on cell cycle progression. There was no significant difference in any of the parameters tested between cells expressing vector control sequences and wild-type MG-63 cells (data not shown). Expression of IGFBP-5 caused a decrease in cell number compared with cells expressing vector control sequences (P < 0.0001) and treatment with 1,25-(OH)2D3 inhibited cell growth additively with the expression of IGFBP-5 (P < 0.005) (Fig. 7AGo).


Figure 7
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 7. Effect of IGFBP-5 Expression on Osteoblast Proliferation and Cell Cycle Progression

A, MG-63 cells expressing recombinant human IGFBP-5 or vector control sequences were treated with vehicle (white bars) or 1,25-(OH)2D3 (black bars) for 48 h and viable cell numbers determined. B, Cell cycle analysis was performed using flow cytometry. Cell cycle distribution in vector-expressing cells [without (white bars) or with (black bars)] 1,25-(OH)2D3) or IGFBP-5-expressing cells [without (light gray bars)] or with (dark gray bars) 1,25-(OH)2D3) was quantitated using Multicycle software. Each experimental condition was performed in triplicate and applied in two independent experiments. Values are expressed as means ± SD. *, P < 0.01; **, P < 0.005; ***, P < 0.0001. Vec, Vector; BP, IGFBP.

 
Cell cycle distribution was analyzed using flow cytometry and quantitated using Multicycle software (Fig. 7BGo). Expression of IGFBP-5 caused a significant increase in the percentage of cells in G0/G1 (from 54% in vector control cells to 63% in IGFBP-5 expressing cells: P < 0.005) and G2/M (from 16% in control cells to 22% in IGFBP-5 expressing cells: P < 0.005). The presence of IGFBP-5 also decreased the percentage of cells in S phase (from 29% in vector control cells to 17% in IGFBP-5 expressing cells: P < 0.0001). This suggests that IGFBP-5 is blocking progression at both stages of the cell cycle in MG-63 cells.

Vector-transduced cells treated with 1,25-(OH)2D3 displayed an increase in the percentage of cells in G0/G1 (from 54% in control cells to 76% in 1,25-(OH)2D3-treated cells: P < 0.0001) with a resultant decrease in the percentage of cells in S phase (from 29% in control cells to 10% in 1,25-(OH)2D3-treated cells: P < 0.0001). This is consistent with a block at the G0/G1 phase of the cell cycle as has been previously described for 1,25-(OH)2D3 (7). Although these trends were also seen in cells expressing IGFBP-5, there was a modest but reproducible decrease in the percentage of cells in G0/G1 when IGFBP-5 expressing cells were treated with 1,25-(OH)2D3 (from 76% in vector control cells to 71% in IGFBP-5 expressing cells: P < 0.01), suggesting that IGFBP-5 may be able to partially desensitize the cells to the effects of 1,25-(OH)2D3 at this point in the cell cycle.

Effect of IGFBP-5 on the Major Mediators of 1,25-(OH)2D3-Induced Cell Cycle Arrest
The cell cycle arrest induced by 1,25-(OH)2D3 has been studied in many cell types and appears to be mediated by induction of the cdk inhibitors p21CIP1/WAF1 and p27KIP1 and the inhibition of the regulatory subunit of cdk4, cyclin D1 (16, 17). Previous studies have shown that 1,25-(OH)2D3 does not increase protein levels of these cdk inhibitors in MG-63 cells (7), and we have confirmed these findings under our experimental conditions (data not shown). To determine whether IGFBP-5 has a role in modulating the expression of cyclin D1 under basal or 1,25-(OH)2D3-stimulated conditions, lysates were prepared from wild-type MG-63 cells (Fig. 8AGo, lanes 1 and 2) or MG-63 cells expressing vector control sequences (lanes 3 and 4) or recombinant IGFBP-5 (lanes 5 and 6) that had been treated with vehicle (lanes 1, 3, and 5) or 1,25-(OH)2D3 (lanes 2, 4, and 6). The expression of cyclin D1 was determined using Western immunoblot analysis (Fig. 8AGo) and quantitated as described in Materials and Methods (Fig. 8BGo). We found that 1,25-(OH)2D3 down-regulated the expression of cyclin D1 in control, vector and IGFBP-5 expressing cells to 4.7% (P < 0.0005), 6.6% (P < 0.001) and 18% (P < 0.01), respectively, compared with the corresponding vehicle-treated cells. These findings are consistent with the 1,25-(OH)2D3-induced block at G0/G1 observed by flow cytometry (Fig. 7BGo). Interestingly, there was a large basal inhibition of cyclin D1 expression by IGFBP-5 alone (P < 0.001), which is consistent with the observation that IGFBP-5 caused an increase in the percentage of cells in G0/G1 and a decrease in cells in S phase.


Figure 8
View larger version (35K):
[in this window]
[in a new window]

  		Fig. 8. IGFBP-5 Down-Regulates the Expression of Cyclin D1

A, Control MG-63 cells (lanes 1 and 2) or cells expressing vector sequences (lanes 3 and 4) or IGFBP-5 (lanes 5 and 6) were treated with vehicle (lanes 1, 3, and 5) or 1,25-(OH)2D3 (lanes 2, 4, and 6) for 48 h and the expression of cyclin D1 in cell lysates was analyzed by Western immunoblotting. {alpha}-Tubulin acted as a loading control. B, Data from two independent experiments were quantified as described in Materials and Methods (vehicle- [white bars] and 1,25-(OH)2D3-treated samples [black bars]). Values are expressed as means ± SD. *, P < 0.01; **, P < 0.001; ***, P < 0.0005; for 1,25-(OH)2D3-treated vs. the corresponding vehicle-treated cells.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
We and others (18, 19, 20) have shown that IGFBP-3 and IGFBP-5 localize to the nucleus in many cell types and interact with the retinoid receptors, RXR and RAR (11, 21). Recently we have demonstrated that IGFBP-3 can attenuate atRA-mediated signaling, prevent retinoid receptor heterodimerization and reduce the amount of ligand-activated receptor complexes (11). This current study is the first to demonstrate an interaction between IGFBP-5 and the VDR, and the disruption of RXR:VDR heterodimers by IGFBP-5. Furthermore, we present evidence that the endogenous expression of IGFBP-5 attenuates cell responsiveness to 1,25-(OH)2D3 in osteoblast-like cells.

Many growth factors with classical roles as extracellular signaling molecules have been localized to the cell nucleus where they may modulate a variety of cellular responses (22). In many cases, these small bioactive peptides are actively imported into the nucleus by larger, nuclear localization signal-containing, proteins for which they have a high affinity. Such is the case for IGF-I, which has been shown to be cotransported into the nucleus by IGFBP-3 in opossum kidney cells (19) and human breast cancer cells (23). Given the high structural homology between IGFBP-3 and -5, together with their shared ability to localize to the cell nucleus and bind RXR, RAR, and VDR, it seems plausible that IGFBP-5 may also be capable of cotransporting IGF-I to the nucleus. The inability of IGFBP-5 to interact with VDR when bound to IGF-I suggests that if IGFBP-5 enters the nucleus complexed to IGF-I, it may be unable to regulate the 1,25-(OH)2D3 response in these cells.

Previous work has shown that IGFBP-3 interacts directly with RXR{alpha} and enhances the transactivation caused by an RXR-specific ligand (21). In contrast, IGFBP-3 binding to RAR{alpha} inhibits the transactivation of the retinoic acid response element by atRA and modulates the atRA-responsiveness in human breast cancer cells (11, 21). We have shown that IGFBP-5 is also capable of interacting with members of the nuclear receptor superfamily, including the VDR, and modulating the 1,25-(OH)2D3 response. Together, these studies suggest that IGFBP-3 and -5 may have an important role in atRA- and 1,25-(OH)2D3 signaling. For the interaction between the IGFBPs and nuclear receptors to be physiologically relevant, it must be capable of occurring within intact cells. This has been shown for IGFBP-3 and RXR{alpha}, where they colocalize to the nuclei of LAPC-4 prostate cancer cells and coprecipitate in HeLa cell nuclear extracts (21). In the present study, we have shown that IGFBP-5 and VDR colocalize to the nuclei of human osteoblast-like cells. Strong nuclear staining for VDR was seen in all MG-63 cells and in a proportion of the U2-OS cells. Interestingly, we observed marked extranuclear staining for VDR in U2-OS cells, suggesting that these cells may possess a nonnuclear VDR signaling pathway. IGFBP-5 was present in the nuclei of MG-63 and U2-OS cells and was coimmunoprecipitated with VDR from nuclear extracts derived from these cells.

The impact of IGFBP-5 on the ability of 1,25-(OH)2D3 to transactivate the human osteocalcin promoter was examined in U2-OS cells. We were unable to discern any difference in basal promoter activity as a result of either increasing or decreasing IGFBP-5 expression, although a previous study has shown that recombinant IGFBP-5 can increase osteocalcin production in MG-63 cells (24). This may be related to posttranscriptional event(s) that control the expression of a naturally occurring gene that are not seen when the same promoter is driving a reporter gene. However, the induction of osteocalcin promoter activity by 1,25-(OH)2D3 was significantly reduced when the reporter constructs were coexpressed with IGFBP-5. The suggestion that IGFBP-5 could attenuate 1,25-(OH)2D3-induced promoter activity was strengthened by the observation that targeted knock-down of endogenous cellular IGFBP-5 enhanced the 1,25-(OH)2D3-response. This clearly demonstrates that endogenous IGFBP-5 is able to exert an inhibitory effect on the cellular responsiveness to 1,25-(OH)2D3. Interestingly, IGFBP-3 (which has high homology to IGFBP-5) has been shown to block 1,25-(OH)2D3-induced transcriptional activity in LAPC-4 cells and the expression of CD11b on HL-60 promyelocytic leukemia cells (25). We have shown that IGFBP-3 also binds the VDR and this interaction may be responsible for the observed effects in these cell types.

There are conflicting reports on the effect of IGFBP-5 on osteoblastic function, with the majority of studies showing that IGFBP-5 stimulates the expression of bone differentiation markers (24, 26). However, transgenic mice overexpressing IGFBP-5 have impaired osteoblast function (27). We have examined the effects of IGFBP-5 on the 1,25-(OH)2D3-induced expression of bone differentiation markers in U2-OS and MG-63 cells. U2-OS cells express very low levels of ALP and its activity is not induced by 1,25-(OH)2D3 (15). When IGFBP-5 was down-regulated in U2-OS cells, we observed a significant increase in ALP activity that was not observed in the presence of endogenously expressed IGFBP-5. In complementary studies carried out in MG-63 cells, we found that although IGFBP-5 and 1,25-(OH)2D3 independently induced ALP activity, the expression of IGFBP-5 prevented any further induction by 1,25-(OH)2D3. Together, these findings suggest that IGFBP-5 can attenuate the 1,25-(OH)2D3 response. Interestingly, a previous study found no difference in IGFBP-5-induced ALP activity in MG-63 cells treated with vehicle or 1,25-(OH)2D3 (24). These differing findings may relate to the source of IGFBP-5 used in these studies; whereas the previous work used exogenous IGFBP-5, we have used endogenously expressed IGFBP-5. In support of this, we were unable to discern any difference in the 1,25-(OH)2D3-induced activity of the osteocalcin promoter when exogenous IGFBP-5 was added to the culture medium. The mechanism underlying the stimulatory effect of IGFBP-5 alone is unclear. Zhao et al. (28) have recently shown that amino-terminal IGFBP-5 residues that are involved in its transactivation activity are masked by the central and carboxy-terminal domains. We speculate that endogenously expressed IGFBP-5 may undergo a conformational change, perhaps induced by phosphorylation, that may allow the unmasking of the critical residues, thus allowing interaction with transcriptional regulators, perhaps including VDR.

Previous studies have shown that 1,25-(OH)2D3 (29) and IGFBP-5 (30) independently induce the expression of collagen {alpha}1 type 1 and that IGFBP-5 decreases the level of serum carboxyl-terminal cross-linked telopeptide of type I collagen in mice (24). We have shown that 1,25-(OH)2D3 and IGFBP-5 independently induce the expression of collagen {alpha}1 type 1 in MG-63 cells, and that IGFBP-5 completely abrogates any further 1,25-(OH)2D3 induction of collagen {alpha}1 type 1. These findings suggest that although endogenous IGFBP-5 is itself stimulatory, when expressed in cells with an activated 1,25-(OH)2D3-signaling pathway, IGFBP-5 has the effect of inhibiting responsiveness to 1,25-(OH)2D3.

There are also differing reports on the role of IGFBP-5 on bone cell growth. Many studies have shown that IGFBP-5 stimulates osteoblast proliferation (2, 3); however, IGFBP-5 has also been shown to inhibit osteoblast cell growth (31) and to induce apoptosis (32). In this current study, we found that IGFBP-5 had an antiproliferative effect toward MG-63 cells. The explanation for these observed discrepancies remains unclear but may be related to differences in the degree of differentiation of the osteoblast models used in these studies. It may also be related to the mode of presentation of IGFBP-5 to these cells. When IGFBP-5 is exogenously added to cells it stimulates osteoblast cell growth (2, 3); however, when IGFBP-5 is endogenously expressed by osteoblast-like cells it is growth inhibitory (31). Previous studies have shown that endogenously expressed and exogenously added IGFBP-3 and IGFBP-5 have differing effects on cell proliferation and apoptosis in a number of cell systems. For example, expression of human recombinant IGFBP-3 in T47D human breast cancer cells inhibited cell proliferation in early passages after transfection but stimulated proliferation in later passage cells (33). In contrast, the addition of exogenous IGFBP-3 to T47D cells expressing vector control sequences had no effect on cell growth (34). Expression of recombinant IGFBP-5 by MDA-MB-231 and Hs578T human breast cancer cells caused cell cycle arrest and the induction of apoptosis. Neither of these effects were observed when these cells were exposed to exogenous IGFBP-5 (35). Furthermore, in U2-OS cells, both the knockdown of endogenous IGFBP-5 expression and the addition of exogenous IGFBP-5 caused an increase in the apoptotic index (32).

The decrease in osteoblast cell proliferation accompanying IGFBP-5 expression caused an increase in the percentage of cells in G0/G1 and G2/M of the cell cycle and a decrease in the percentage of cells in S phase compared with control cells. Previous work from our laboratory has shown that IGFBP-5 blocks cell cycle progression at G2/M in a breast cancer cell line, and this was associated with the induction of apoptosis (35). Interestingly, IGFBP-5 reduced the expression of cyclin D1, the regulatory subunit of cdk4. Because transition from G1 to S phase requires activation of the cyclin D/cdk complex, the reduction in cyclin D1 expression seen in IGFBP-5 expressing cells may be responsible for the accumulation of these cells in G0/G1 compared with vector expressing cells. The reason why the marked decrease in cyclin D1 expression is accompanied by a relatively small increase in the percentage of cells in G0/G1 is unclear, but may be related to differences in the accuracy of the two techniques used (flow cytometry being more quantitative than Western immunoblot analysis) or changes in other components of the cell cycle machinery.

1,25-(OH)2D3 suppresses cell growth by inhibiting cell cycle progression and inducing apoptosis in many cell types (36), and its major target is the G1 phase of the cell cycle. We have shown that 1,25-(OH)2D3 blocks cell cycle progression at G0/G1 in MG-63 cells and that the expression of IGFBP-5 partially desensitized the cells to these effects. The expression of cyclin D1 was down-regulated by 1,25-(OH)2D3 in MG-63 cells. Although this effect was less marked in the presence of IGFBP-5, it is unlikely to be relevant given the significant decrease in cyclin D1 expression caused by IGFBP-5 alone. Previous studies have shown that MG-63 cells do not express IGFs (37); however, these cells are very sensitive to their proliferative and antiapoptotic effects (our unpublished observations). Although our studies were carried out under serum-free conditions, it is possible that residual IGFs derived from the growth media were present. Therefore, an alternative explanation is that IGFBP-5 acts in an IGF-dependent manner to inhibit cell growth and cell cycle progression by sequestering IGFs. In this scenario, 1,25-(OH)2D3 and IGFs would be acting independently of each other. Thus, the inhibitory effects of 1,25-(OH)2D3 would appear to be less marked in the presence of IGFBP-5 because it had the effect of attenuating IGF-induced cell growth.

Many studies have shown that IGFBP-5 has an important role in bone cell function. The differences frequently observed in its effects on osteoblast proliferation and differentiation may be related to the site of action of IGFBP-5. If IGFBP-5 is acting primarily in the extracellular matrix or at the cell surface (as would be seen with exogenous IGFBP-5), then IGFBP-5 may act in an IGF-dependent manner. However, if the response relies on an intracellular effect of IGFBP-5 that could only be initiated by IGFBP-5 being routed through its synthesis pathway (as would be seen when IGFBP-5 was endogenously expressed), then the effects of IGFBP-5 may be IGF independent. This would suggest important differences in the paracrine and autocrine effects of IGFBP-5 and possibly other IGFBPs as well. We have shown that IGFBP-5 colocalizes with VDR to the cell nucleus where they physically interact, and prevents the formation of active RXR:VDR heterodimers. Consistent with this, endogenous IGFBP-5 exerts a tonic inhibitory effect on cellular responsiveness to 1,25-(OH)2D3. Together with the observation that IGFBP-5 and 1,25-(OH)2D3 independently induce the expression of bone differentiation markers, and that IGFBP-5 effectively desensitizes cells to any further induction by 1,25-(OH)2D3, our results suggest an important role for IGFBP-5 in modulating the 1,25-(OH)2D3-response in osteoblast-like cells.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Materials
Recombinant human IGFBP-3 and -5 were produced in 911 human retinoblastoma cells. IGFBP-1 was purified from human amniotic fluid, recombinant human IGFBP-2 was provided by Sandoz (Basel, Switzerland), and IGF-I was provided by Genentech (South San Francisco, CA). Antiserum against IGFBP-5 was prepared in this laboratory. Plasmids expressing human RXR{alpha}, human VDR, and SRC-1 fused to GST were kindly provided by Thorsten Heinzel (Georg Speyer Institute, Germany). Human VDR cDNA was amplified by PCR from GST-VDR and inserted into pRSET (Invitrogen) to generate His6-VDR. Plasmids expressing the osteocalcin promoter-luciferase reporter construct and the VDR expression construct were kindly provided by Gary Leong (Institute for Molecular Biosciences, University of Queensland, Queensland, Australia) and the IGFBP-5 expression construct was kindly provided by Dr. Sue Firth (Kolling Institute of Medical Research, Sydney, Australia). 1,25-(OH)2D3 was purchased from Calbiochem (La Jolla, CA), propidium iodide was from Molecular Probes (Eugene, OR) and ribonuclease and p-nitrophenyl phosphate were from Sigma (St. Louis, MO).

GST Pull-Down Assay
GST, GST-VDR, GST-RXR{alpha}, and GST-SRC-1 were expressed in Escherichia coli and captured from cell lysates using glutathione-Sepharose beads (Amersham Biosciences, Piscataway, NJ). The immobilized proteins were incubated with the indicated IGFBP (0.5 µg) as previously described (11) and the bound IGFBPs were separated on 10% SDS-PAGE and detected by Western ligand blotting using 125I-IGF-I or Western immunoblotting using chicken antihuman IGFBP-5 antiserum, antichicken horseradish peroxidase-conjugated secondary antibodies (Amersham Biosciences), and SuperSignal West Dura (Pierce, Rockford, IL). Imaging was carried out using a Fuji-Film LAS-3000 imaging system (Tokyo, Japan). Where IGFBP-5 was preincubated with IGF-I, the reaction took place in 50 mM sodium phosphate (pH 6.8) and 0.1% BSA at 22 C for 1 h.

In another series of experiments, GST-RXR{alpha} (3 µg) was immobilized on glutathione beads and preincubated without or with IGFBP-5 at 1-, 2-, or 4-fold molar excess with respect to His6-VDR for 1 h at 22 C with gentle rotation. His6-VDR was expressed in Escherichia coli and purified from cell lysates on a Ni-NTA agarose column (QIAGEN). His6-VDR (4.5 µg) was added to each reaction and incubated as above for a further 15 min. Bound His6-VDR was analyzed on reducing SDS-PAGE and Western immunoblotting using a monoclonal anti-polyhistidine antibody (Sigma). Imaging was carried out using a Fuji-Film LAS-3000 imaging system, with quantitation performed using the Image Gauge 3.11 software.

Cell Culture
The human osteosarcoma cell lines U2-OS and MG-63 (American Type Culture Collection, Manassas, VA) were maintained in McCoys 5A medium and RPMI medium (Invitrogen, Carlsbad, CA), respectively, supplemented with 25 mM NaHCO3 and 10% fetal calf serum. Adenovirus expressing recombinant human IGFBP-5 or vector control sequences were prepared as previously described (35). The number of live viral particles was quantitated using the Adeno-X Rapid Titer Kit (BD Biosciences, San Jose, CA). Cells were transduced with 1.4 x 109 infective particles/ml serum-free medium for 6 h. After removal of the virus, the cells were cultured in serum-free medium overnight and then stimulated with 1,25-(OH)2D3 or vehicle for 48 h. Secreted and intracellular levels of IGFBP-5 were analyzed by RIA on conditioned media and nuclear and cytoplasmic extracts (prepared using NE-PER nuclear and cytoplasmic extraction reagents; Pierce).

Indirect Immunocytochemistry
MG-63 and U2-OS cells were transiently cotransfected with 1µg each of plasmids expressing human VDR (pCMV-FLAG:VDR) and human IGFBP-5 (pcDNA-IGFBP-5) using the Nucleofector II (Amaxa, Cologne, Germany) according to the manufacturer’s instructions. The cells were grown on coverslips in 12-well plates seeded at a density of 1.4 x 105 cells/well. They were treated with 1,25-(OH)2D3 (10–8 M) for 4 h before fixation and permeabilization with 4% paraformaldehyde and 0.3% Triton X-100 at 22 C for 15 min. After overnight incubation at 4 C in blocking buffer (2% BSA in PBS), cells were incubated with a 1:50 dilution rabbit antihuman VDR antibody (Santa Cruz Biotechnology, Santa Cruz, CA) or purified rabbit IgG (Pierce) diluted in blocking buffer at 22 C for 2 h. Cells were then washed and incubated with a 1:200 dilution of goat antirabbit IgG conjugated with Alexa594 (Molecular Probes) at 22 C for 1 h. After another series of washes, the cells were incubated with 1:50 dilution chicken antihuman IGFBP-5 antiserum or purified chicken IgY (Institute of Medical and Veterinary Science, Adelaide, South Australia, Australia) at 22 C for 2 h. Cells were then washed and incubated with 1:200 dilution rabbit antichicken IgY conjugated with fluorescein isothiocyanate (FITC) (Pierce) at 22 C for 1 h. After a final series of washes, the cells were incubated with 300 nM 4',6-diamidino-2-phenylindole, rinsed and mounted in an antifade medium, and examined using an IX70 inverted fluorescent microscope (Olympus, Australia). The images were collected using a SPOT-RT camera (Diagnostic Instruments, Sterling Heights, MI) under identical and nonsaturating conditions.

Immunoprecipitations
MG-63 and U2-OS cells expressing adenoviral-derived human IGFBP-5 were stimulated with 1,25-(OH)2D3 (10–8 M) 4 h before the extraction of nuclear and cytoplasmic fractions using NE-PER reagents (Pierce). The nuclear and cytoplasmic extracts (100 µg) were precleared by incubation in IPP buffer [20 mM HEPES (pH 7.5), 20% glycerol, 0.1% Nonidet P-40 and protease inhibitors (Complete protease inhibitor cocktail; Roche Diagnostics, Indianapolis, IN)] and 25 µl protein G plus/protein A-agarose (Calbiochem) for 1 h at 4 C with gentle rotation. The precleared lysates were collected and VDR was immunoprecipitated by incubating with 2 µg rat antihuman VDR antibody (Calbiochem) or purified rat IgG (Pierce) for 20 h at 4 C with gentle rotation. Protein G plus/protein A-agarose (25 µl) was added, and the incubation continued for a further 1 h at 4 C to allow capture of the antibody-protein complexes by the agarose beads. The immunoprecipitated protein complexes were washed in buffer containing 20 mM HEPES (pH 7.5), 75 mM KCl, 2.5 mM MgCl2, 0.1% Nonidet P-40, and protease inhibitors, recovered from the agarose beads and separated on SDS-PAGE. IGFBP-5 was detected under nonreducing conditions by ligand blotting using 125I-IGF-II, and the VDR was detected under reducing conditions by Western immunoblotting using a rabbit antihuman VDR antibody (Santa Cruz Biotechnology).

Transcriptional Reporter Assays
U2-OS cells were transiently cotransfected with 1.35 µg of a luciferase reporter plasmid (expressing firefly luciferase) containing the VDREs from the human osteocalcin promoter and 0.15 µg of internal control reporter plasmid (pRL-TK) (Promega, Madison, WI) using the Nucleofector II (Amaxa) system. The pRL-TK vector expresses the Renilla luciferase gene and was used to control for transfection efficiency. The cells were seeded in 12-well plates at 2 x 105 cells/well and allowed to recover overnight before infection with adenovirus expressing recombinant human IGFBP-5 or vector control sequences. In another set of experiments, U2-OS cells were transfected with the osteocalcin promoter:luciferase construct and pRL-TK with either an siRNA targeted against IGFBP-5 or a control siRNA. The siRNA duplexes were designed and chemically synthesized by QIAGEN. The sequence of the IGFBP-5 siRNA corresponded to bases 648–666 from the open reading frame of human IGFBP-5 mRNA (5'-GCCCAAUUGUGACCGCAAA-dTdT-3'). A scrambled siRNA was used as a negative control (5'-CUGUGAAAACACUACGGUC-dTdT-3'). After treatment with 1,25-(OH)2D3 (10–7 M) or vehicle for 24 h, conditioned media were collected for IGFBP-5 RIA and cell lysates were prepared and assayed for firefly and Renilla luciferase activity using the Dual-Glo Luciferase Assay System (Promega). Luminescence was quantified with a TD-20/20 Luminometer (Turner Designs, Sunnyvale, CA).

Alkaline Phosphatase Assay
U2-OS and MG-63 cells were seeded in six-well plates at a density of 8 x 104 cells/well and transfected with siRNAs against IGFBP-5 or infected with adenovirus expressing recombinant human IGFBP-5 and treated with 1,25-(OH)2D3 (10–7 M) or vehicle. The cells were washed with PBS and scraped into cold TXM-buffer [10 mM Tris-HCl (pH 7.4), 1 mM MgCl2, 20 µM ZnCl2 and 0.1% Triton X-100] (15). After sonication, the cell debris was removed by centrifugation, and the lysates were assayed for ALP activity with p-nitrophenyl phosphate as substrate.

Western Blot Analysis
MG-63 cells were seeded in T12.5 cm2 flasks at 2.5 x 105 cells/flask and infected with adenovirus expressing recombinant human IGFBP-5 or vector control sequences and treated with 1,25-(OH)2D3 (10–7 M) or vehicle. The cells were harvested in 60 mM Tris-HCl (pH 7.5), 1% sodium dodecyl sulfate, 7.5% glycerol, and 5% ß-mercaptoethanol, sonicated and cell debris removed by centrifugation. Whole-cell lysates (10 µg/lane) were separated on SDS-PAGE before transfer to nitrocellulose and immunoblotted using antibodies against p21CIP1/WAF1 or collagen {alpha}1 type 1 (Santa Cruz Biotechnology), and p27KIP1 or cyclin D1 (Cell Signaling Technology, Beverly, MA). Immunoreactive proteins were detected using antimouse, antirabbit or antigoat horseradish peroxidase-conjugated secondary antibodies (Amersham Biosciences) and SuperSignal West Dura (Pierce). The membranes were stripped with Restore Western Blot Stripping Buffer (Pierce) and immunoblotted using a mouse monoclonal antibody against {alpha}-tubulin (Sigma). Imaging was carried out using a Fuji-Film LAS-3000 imaging system under nonsaturating conditions, with quantitation performed using the Image Gauge 3.11 software.

Cell Proliferation Assay and Cell Cycle Analysis
MG-63 cells were seeded in six-well plates at 8 x 104 cells/well and infected with adenovirus expressing recombinant human IGFBP-5 or vector control sequences and treated with 1,25-(OH)2D3 (10–7 M) or vehicle. The cells were harvested and cell numbers determined using a Coulter Particle Counter (Beckman Coulter, Fullerton, CA). For cell cycle analysis, the cells were washed with PBS and fixed in 70% ice-cold EtOH for 24 h. The cells were collected by centrifugation, resuspended in buffer [45 mM sodium phosphate, 2.5 mM citric acid, 0.1% Triton X-100 (pH 7.8)] and incubated at 22 C for 30 min. The buffer was removed by centrifugation, and the cell pellet was resuspended in 0.1% Triton X-100/PBS. Propidium iodide (50 µg/ml) and freshly prepared ribonuclease (1 mg/ml) were added and the cells were incubated at 22 C for 30 min. Cell cycle analysis was performed using a FACScan (Becton Dickinson, Franklin Lakes, NJ), and the data were analyzed using ModFit software (ModFitLT version 1.00; Becton Dickinson).

Statistical Analysis
Data were analyzed by analysis of variance followed by Fisher’s protected least significant difference test using Statview 4.02 (Abacus Concepts, Inc., Berkeley, CA).


   ACKNOWLEDGMENTS
 
We thank Prof. Thorsten Heinzel (Georg-Speyer-Haus Institute, Frankfurt am Main, Germany) and Dr. Gary Leong (Institute for Molecular Biosciences, University of Queensland, Queensland, Australia) for providing the nuclear receptor expression constructs and the osteocalcin promoter-luciferase reporter construct. We also thank Dr. Sue Firth (Kolling Institute of Medical Research, Sydney, Australia) for providing adenovirus expressing recombinant human IGFBP-5, the purified protein and the IGFBP-5 expression construct.


   FOOTNOTES
 
This work was supported by National Health and Medical Research Council, Australia, Grant 302171.

Present address for A.M.: Faculty of Medical and Health Sciences, University of Auckland, Private Bag 92019, Auckland, New Zealand.

Disclosure Statement: The authors have nothing to disclose.

First Published Online June 26, 2007

Abbreviations: ALP, Alkaline phosphatase; atRA, all-trans retinoic acid; cdk, cyclin-dependent kinase; FITC, fluorescein isothiocyanate; GST, glutathione S-transferase; IGFBP, IGF binding protein; 1,25(OH)2D3, 1,25 dihydroxyvitamin D3; RAR, retinoic acid receptor; RXR, retinoid X receptor; siRNA, small interfering RNA; SRC-1, steroid receptor coactivator 1; VDR, vitamin D receptor; VDRE, VDR response element.

Received for publication December 29, 2006. Accepted for publication June 22, 2007.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Firth SM, Baxter RC 2002 Cellular actions of the insulin-like growth factor binding proteins. Endocr Rev 23:824–854[Abstract/Free Full Text]
  2. Miyakoshi N, Richman C, Kasukawa Y, Linkhart TA, Baylink DJ, Mohan S 2001 Evidence that IGF-binding protein-5 functions as a growth factor. J Clin Invest 107:73–81[CrossRef][Medline]
  3. Mohan S, Nakao Y, Honda Y, Landale E, Leser U, Dony C, Lang K, Baylink DJ 1995 Studies on the mechanisms by which insulin-like growth factor (IGF) binding protein-4 (IGFBP-4) and IGFBP-5 modulate IGF actions in bone cells. J Biol Chem 270:20424–20431[Abstract/Free Full Text]
  4. Campbell PG, Andress DL 1997 Insulin-like growth factor (IGF)-binding protein-5-(201–218) region regulates hydroxyapatite and IGF-I binding. Am J Physiol Endocrinol Metab 36:E1005–E1013
  5. van Driel M, Pols HA, van Leeuwen JP 2004 Osteoblast differentiation and control by vitamin D and vitamin D metabolites. Curr Pharm Des 10:2535–2555[CrossRef][Medline]
  6. Gurlek A, Pittelkow MR, Kumar R 2002 Modulation of growth factor/cytokine synthesis and signaling by 1{alpha},25-dihydroxyvitamin D(3): implications in cell growth and differentiation. Endocr Rev 23:763–786[Abstract/Free Full Text]
  7. Ryhanen S, Jaaskelainen T, Mahonen A, Maenpaa PH 2003 Inhibition of MG-63 cell cycle progression by synthetic vitamin D3 analogs mediated by p27, Cdk2, cyclin E, and the retinoblastoma protein. Biochem Pharmacol 66:495–504[CrossRef][Medline]
  8. Koshihara Y, Hoshi K, Ishibashi H, Shiraki M 1996 Vitamin K2 promotes 1{alpha},25(OH)2 vitamin D3-induced mineralization in human periosteal osteoblasts. Calcif Tissue Int 59:466–473[Medline]
  9. Matsumoto T, Sowa Y, Ohtani-Fujita N, Tamake T, Takenaka T, Kuribayashi K, Sakai T 1998 p53-independent induction of WAF1/Cip1 is correlated with osteoblastic differentiation by vitamin D3. Cancer Lett 129:61–68[CrossRef][Medline]
  10. De Blasio A, Messina C, Santulli A, Mangano V, Di Leonardo E, D’Anneo A, Tesoriere G, Vento R 2005 Differentiative pathway activated by 3-aminobenzamide, an inhibitor of PARP, in human osteosarcoma MG-63 cells. FEBS Lett 579:615–620[CrossRef][Medline]
  11. Schedlich LJ, O’Han MK, Leong GM, Baxter RC 2004 Insulin-like growth factor binding protein-3 prevents retinoid receptor heterodimerization: implications for retinoic acid-sensitivity in human breast cancer cells. Biochem Biophys Res Commun 314:83–88[CrossRef][Medline]
  12. McKenna NJ, Xu JM, Nawaz Z, Tsai SY, Tsai MJ, O’Malley BW 1999 Nuclear receptor coactivators: multiple enzymes, multiple complexes, multiple functions. J Steroid Biochem Mol Biol 69:3–12[CrossRef][Medline]
  13. Wong MS, Fong CC, Yang M 1999 Biosensor measurement of the interaction kinetics between insulin-like growth factors and their binding proteins. Biochim Biophys Acta 1432:293–301[CrossRef][Medline]
  14. Rajaram S, Baylink DJ, Mohan S 1997 Insulin-like growth factor-binding proteins in serum and other biological fluids: regulation and functions. Endocr Rev 18:801–831[Abstract/Free Full Text]
  15. Mahonen A, Pirskanen A, Keinanen R, Maenpaa PH 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[Medline]
  16. Jensen SS, Madsen MW, Lukas J, Binderup L, Bartek J 2001 Inhibitory effects of 1{alpha},25-dihydroxyvitamin D(3) on the G(1)-S phase-controlling machinery. Mol Endocrinol 15:1370–1380[Abstract/Free Full Text]
  17. Verlinden L, Verstuyf A, Convents R, Marcelis S, Van Camp M, Bouillon R 1998 Action of 1,25(OH)2D3 on the cell cycle genes, cyclin D1, p21 and p27 in MCF-7 cells. Mol Cell Endocrinol 142:57–65[CrossRef][Medline]
  18. Schedlich LJ, Young TF, Firth SM, Baxter RC 1998 Insulin-like growth factor-binding protein (IGFBP)-3 and IGFBP-5 share a common nuclear transport pathway in T47D human breast carcinoma cells. J Biol Chem 273:18347–18352[Abstract/Free Full Text]
  19. Li WL, Fawcett J, Widmer HR, Fielder PJ, Rabkin R, Keller GA 1997 Nuclear transport of insulin-like growth factor-I and insulin-like growth factor binding protein-3 in opossum kidney cells. Endocrinology 138:1763–1766[Abstract/Free Full Text]
  20. Jaques G, Noll K, Wegmann B, Witten S, Kogan E, Radulescu RT, Havemann K 1997 Nuclear localization of insulin-like growth factor binding protein 3 in a lung cancer cell line. Endocrinology 138:1767–1770[Abstract/Free Full Text]
  21. Liu B, Lee HY, Weinzimer SA, Powell DR, Clifford JL, Kurie JM, Cohen P 2000 Direct functional interactions between insulin-like growth factor-binding protein-3 and retinoid X receptor-{alpha} regulate transcriptional signaling and apoptosis. J Biol Chem 275:33607–33613[Abstract/Free Full Text]
  22. Peralta Soler A, Alemany J, Smith RM, de Pablo F, Jarett L 1990 The state of differentiation of embryonic chicken lens cells determines insulin-like growth factor I internalization. Endocrinology 127:595–603[Abstract/Free Full Text]
  23. Schedlich LJ, Graham LD 2002 Role of insulin-like growth factor binding protein-3 in breast cancer cell growth. Microsc Res Tech 59:12–22[CrossRef][Medline]
  24. Richman C, Baylink DJ, Lang K, Dony C, Mohan S 1999 Recombinant human insulin-like growth factor-binding protein-5 stimulates bone formation parameters in vitro and in vivo. Endocrinology. 140:4699–4705
  25. Ikezoe T, Tanosaki S, Krug U, Liu B, Cohen P, Taguchi H, Koeffler HP 2004 Insulin-like growth factor binding protein-3 antagonizes the effects of retinoids in myeloid leukemia cells. Blood 104:237–242[Abstract/Free Full Text]
  26. Andress DL 2001 IGF-binding protein-5 stimulates osteoblast activity and bone accretion in ovariectomized mice. Am J Physiol Endocrinol Metab 281:E283–E288
  27. Devlin RD, Du Z, Buccilli V, Jorgetti V, Canalis E 2002 Transgenic mice overexpressing insulin-like growth factor binding protein-5 display transiently decreased osteoblastic function and osteopenia. Endocrinology 143:3955–3962[Abstract/Free Full Text]
  28. Zhao Y, Yin P, Bach LA, Duan C 2006 Several acidic amino acids in the N-domain of insulin-like growth factor-binding protein-5 are important for its transactivation activity. J Biol Chem 281:14184–14191[Abstract/Free Full Text]
  29. Bonewald LF, Kester MB, Schwartz Z, Swain LD, Khare A, Johnson TL, Leach RJ, Boyan BD 1992 Effects of combining transforming growth factor ß and 1,25-dihydroxyvitamin D3 on differentiation of a human osteosarcoma (MG-63). J Biol Chem 267:8943–8949[Abstract/Free Full Text]
  30. Pilewski JM, Liu L, Henry AC, Knauer AV, Feghali-Bostwick CA 2005 Insulin-like growth factor binding proteins 3 and 5 are overexpressed in idiopathic pulmonary fibrosis and contribute to extracellular matrix deposition. Am J Pathol 166:399–407[Abstract/Free Full Text]
  31. Schneider MR, Zhou R, Hoeflich A, Krebs O, Schmidt J, Mohan S, Wolf E, Lahm H 2001 Insulin-like growth factor-binding protein-5 inhibits growth and induces differentiation of mouse osteosarcoma cells. Biochem Biophys Res Commun 288:435–442[CrossRef][Medline]
  32. Yin P, Xu Q, Duan C 2004 Paradoxical actions of endogenous and exogenous insulin-like growth factor-binding protein-5 revealed by RNA interference analysis. J Biol Chem 279:32660–32666[Abstract/Free Full Text]
  33. Firth SM, Fanayan S, Benn D, Baxter RC 1998 Development of resistance to insulin-like growth factor binding protein-3 in transfected T47D breast cancer cells. Biochem Biophys Res Commun 246:325–329[CrossRef][Medline]
  34. Fanayan S, Firth SM, Butt AJ, Baxter RC 2000 Growth inhibition by insulin-like growth factor-binding protein-3 in T47D breast cancer cells requires transforming growth factor-ß (TGF-ß) and the type II TGF-ß receptor. J Biol Chem 275:39146–39151[Abstract/Free Full Text]
  35. Butt AJ, Dickson KA, McDougall F, Baxter RC 2003 Insulin-like growth factor-binding protein-5 inhibits the growth of human breast cancer cells in vitro and in vivo. J Biol Chem 278:29676–29685[Abstract/Free Full Text]
  36. Banerjee P, Chatterjee M 2003 Antiproliferative role of vitamin D and its analogs—a brief overview. Mol Cell Biochem 253:247–254[CrossRef][Medline]
  37. Okazaki R, Conover CA, Harris SA, Spelsberg TC, Riggs BL 1995 Normal human osteoblast-like cells consistently express genes for insulin-like growth factors I and II but transformed human osteoblast cell lines do not. J Bone Miner Res 10:788–795[Medline]

NURSA Molecule Pages Link:

Nuclear Receptors:   VDR  |  RXRα
Ligands:   Calcitriol



This article has been cited by other articles:


Home page
 Am. J. Physiol. Gastrointest. Liver Physiol.Home page
R. S. Flynn, S. Mahavadi, K. S. Murthy, J. M. Kellum, and J. F. Kuemmerle
Insulin-like growth factor-binding protein-5 stimulates growth of human intestinal muscle cells by activation of G{alpha}i3
Am J Physiol Gastrointest Liver Physiol, December 1, 2009; 297(6): G1232 - G1238.
[Abstract] [Full Text] [PDF]

Home page
 FASEB J.Home page
G. Tripathi, D. A. M. Salih, A. C. Drozd, R. A. Cosgrove, L. J. Cobb, and J. M. Pell
IGF-independent effects of insulin-like growth factor binding protein-5 (Igfbp5) in vivo
FASEB J, August 1, 2009; 23(8): 2616 - 2626.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Endocrinol.Home page
A. Mukherjee and P. Rotwein
Insulin-Like Growth Factor-Binding Protein-5 Inhibits Osteoblast Differentiation and Skeletal Growth by Blocking Insulin-Like Growth Factor Actions
Mol. Endocrinol., May 1, 2008; 22(5): 1238 - 1250.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow NURSA Molecule Pages Link
Right arrow Reprints, Permissions and Rights
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Schedlich, L. J.
Right arrow Articles by Baxter, R. C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Schedlich, L. J.
Right arrow Articles by Baxter, R. C.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals