help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology, doi:10.1210/en.2007-1556
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
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 Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hewitt, J.
Right arrow Articles by Nanes, M. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hewitt, J.
Right arrow Articles by Nanes, M. S.
Endocrinology Vol. 149, No. 7 3698-3707
Copyright © 2008 by The Endocrine Society

The Muscle Transcription Factor MyoD Promotes Osteoblast Differentiation by Stimulation of the Osterix Promoter

Jocelyn Hewitt1, Xiaghuai Lu1, Linda Gilbert and Mark S. Nanes

Division of Endocrinology, Metabolism, and Lipids, Department of Medicine Emory University School of Medicine, and Veterans Affairs Medical Center, Atlanta, Georgia 30033

Address all correspondence and requests for reprints to: Mark S. Nanes, M.D., Ph.D., Mail Code 111, 1670 Clairmont Road, Decatur, Georgia 30033. E-mail: mnanes{at}emory.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Transcription factors regulate tissue-specific differentiation of pluripotent mesenchyme to osteoblast (OB), myoblast (MB), and other lineages. Osterix (Osx) is an essential transcription factor for bone development because knockout results in lack of a mineralized skeleton. The proximal Osx promoter contains numerous binding sequences for MyoD and 14 repeats of a binding sequence for Myf5. These basic helix-loop-helix (bHLH) transcription factors have a critical role in MB differentiation and muscle development. We tested the hypothesis that bHLH transcription factors also support OB differentiation through regulation of Osx. Transfection of a MyoD expression vector into two primitive mesenchymal cell lines, C3H/10T1/2 and C2C12, stimulated a 1.2-kb Osx promoter-luciferase reporter 70-fold. Myf5 stimulated the Osx promoter 6-fold. Deletion analysis of the promoter revealed that one of three proximal bHLH sites is essential for MyoD activity. The Myf5 repeat conferred 60% of Myf5 activity with additional upstream sequence required for full activity. MyoD bound the active bHLH sequence and its 3'-flanking region, as shown by EMSA and chromatin immunoprecipitation assays. Real-time PCR revealed that primitive C2C12 and C3H/10T1/2 cells, pre-osteoblastic MC3T3 cells, and undifferentiated primary marrow stromal cells express the muscle transcription factors. C2C12 cells, which differentiate to MB spontaneously and form myotubules, were treated with bone morphogenetic protein 2 (BMP-2) to induce OB differentiation. BMP-2 stimulated expression of Osx and the differentiation marker alkaline phosphatase and blocked myotubule development. BMP-2 suppressed the muscle transcription factor myogenin, but expression of MyoD and Myf5 persisted. Silencing of MyoD inhibited BMP-2 stimulation of Osx and blocked the later appearance of bone alkaline phosphatase. MyoD support of Osx transcription contributes to early OB differentiation.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
BONE REQUIRES ADEQUATE numbers of osteoblasts (OB) for the formation of a mineralized matrix during skeletal development, renewal, and fracture repair. OB are derived from pluripotent precursors that have the capacity to differentiate toward myocyte, adipocyte, chondrocyte, or fibroblast phenotypes (1). The commitment of mesenchymal precursor cells along any one of these pathways is regulated by transcription factors that direct tissue-specific gene expression. Transcription factors organize target genes within the nuclear matrix to support a specific differentiation pathway. For OB, important transcription factors for phenotype commitment include the runt-related factor (RUNX2/Cbfa1/AML1/Pepb2{alpha}A) and osterix (Osx/SP7). In vitro cell studies and murine transgenic experiments support the essential roles of RUNX2 and Osx for OB differentiation (2, 3, 4, 5, 6, 7, 8). Deletion of either of these genes results in a cartilaginous skeleton devoid of OB and marrow, a lethal outcome. Differentiation toward myoblasts is regulated by members of the basic helix-loop-helix (bHLH) family including MyoD, Myf5, and myogenin (9). MyoD and Myf5 exhibit redundancy in the early selection of myoblasts but are essential, along with myogenin, for later formation of myotubes and mature muscle (10, 11, 12, 13). During differentiation of myoblast precursors, expression of muscle factors increase; however, if cells are treated with bone morphogenetic proteins (BMP-2 and BMP-7), expression of the muscle factors decreases, whereas expression of RUNX2 and Osx increases as cells acquire the OB phenotype (14, 15, 16, 17). Thus, transcription factors may stimulate expression of genes for one pathway while simultaneously suppressing expression of genes for another. However, wingless factors (Wnt), which simultaneously increase RUNX2, Osx, and bHLH factors, stimulate both OB and myoblast differentiation at an early stage (18, 19, 20, 21, 22, 23). Thus, the suppression of the bHLH factors is not a prerequisite for RUNX2 and Osx expression or for OB differentiation.

Osx is posttranscriptionally up-regulated by BMP-2 (24, 25). Less is known about the transcription factors directly regulating the Osx promoter. We, and others, previously reported the structure and regulatory sequences in the proximal Osx promoter (26, 27). The promoter contains two regions capable of independently supporting transcription of the gene, two transcription start sites, a potent TNF/MAPK suppressor region, and a weak RUNX2 enhancer. Our analysis of the Osx promoter unexpectedly revealed 14 overlapping repeats of a putative GC-rich Myf5 binding sequence and additional upstream sites containing canonical CAANTG E-boxes known to bind bHLH factors of the MyoD family. This finding led us to hypothesize that factors previously thought to regulate muscle development may also play a role in Osx regulation and, ultimately, OB differentiation. We tested this hypothesis by studying bHLH regulation of the Osx promoter and localizing their effects through mutational analysis and EMSA. We further investigated the role of MyoD and Myf5 on Osx expression and OB differentiation by silencing MyoD or Myf5 with small interfering RNA (siRNA). Our results revealed that bHLH transcription factors, long known to regulate muscle cell differentiation, also contribute to the regulation of Osx and the differentiation of OB.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Materials
MC3T3-E1 (clone 14) mouse pre-OB cells were provided by Dr. R. Franceschi, University of Michigan, (Ann Arbor, MI). C3H/10T1/2 primitive mesenchymal cells and C2C12 pre-myoblasts were obtained from the American Type Culture Collection (Manassas, VA). MEM-{alpha} was purchased from GIBCO/Invitrogen (Grand Island, NY) and fetal bovine serum (FBS) from HyClone (Logan, UT). Unless otherwise indicated, chemicals were purchased from Sigma Chemical Co. (St. Louis, MO). pcDNA3 expression vector for MyoD was kindly provided by Dr. L. Kedes, Institute for Genetic Medicine (Los Angeles, CA) and the pBabe-Myf5 vector was kindly provided by Dr. N. Imbalzano, University of Massachusetts Medical School (Worcester, MA). The siRNA to MyoD and Myf5 were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) and negative control siRNA from Ambion (Austin, TX). The –1269/+91 Osx promoter-luciferase reporter (Osx-luc) and the techniques to create deletion mutants have been described previously (26). Consensus binding analysis of the Osx promoter was done using Genomatix software (Ann Arbor, MI) with a stringency of 0.85 (28). The preparation of Osx promoter deletion constructs has been described previously (26).

Cell culture and transfection
C3H/10T1/2, MC3T3, and C2C12 cells were cultured in 1 ml MEM-{alpha} with 10% fetal bovine serum in 12-well plates. OB differentiation medium consisted of the addition of 50 µg/ml L-ascorbate and 5 mM β-glycerophosphate. The plates were incubated at 37 C with 5% CO2 under humidified conditions. Medium was changed every 48–72 h. Cells were transfected with promoter reporter constructs, expression vectors, or siRNA using Lipofectamine 2000 according to the manufacturer’s specification (Invitrogen, Carlsbad, CA).

PCR
Probes used for qualitative PCR and primers for real-time RT-PCR are shown in Table 1Go. Conditions for qualitative PCR were 94 C for 2 min, 95 C for 30 sec for 35 cycles, 60 C for 30 sec, 72 C for 1 min, and a final extension of 72 C, 10 min. Conditions for Real-time PCR conditions were 95 C for 30 sec, 95 C for 3 min, 95 C for 15 sec, and 60 C for 30 sec (40 times). Melting curves were measured at the end of each run using a starting temperature of 60 C with an increase of 0.5 C at each cycle to a final temperature of 99.5 C (80 times).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Primers used in qualitative and real-time RT-PCR

 
EMSA
32P-labeled double-stranded oligonucleotides (0.01 pmol) spanning 20 bp of the indicated sequences was added to 20 µl EMSA buffer [10 mM HEPES (pH 7.9), 1 mM EDTA, 1 mM dithiothreitol, 60 mM KCl, 7% glycerol, and 2 µg poly (deoxyinosine-deoxycytosine)]. For preparation of each 32P-labeled probe, the oligonucleotides were annealed, radiolabeled with T4 polynucleotide kinase and {gamma}-[32P]ATP, and separated from unreacted nucleotides by purification with QIAquick Nucleotide removal Kit (QIAGEN, Valencia, CA). The reaction mix included recombinant MyoD (1 µg, murine MyoD 1–318 full-length; Santa Cruz Biotechnology) or IgG control. After 30 min of incubation, the reaction mixture was loaded onto a standard 5% nondenatured polyacrylamide gel in 1x Tris glycine/EDTA buffer. After electrophoresis, the gel was dried and exposed to radiographic film using intensifying screens at –80 C.

Chromatin immunoprecipitation (ChIP) assay
C2C12 cells were grown to 50% confluence in 150-mm flasks for 24 h. Cross-linking was performed by adding formaldehyde directly to culture media to a final concentration of 1%. Cells were washed twice in ice-cold PBS containing protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, and 1 µg/ml pepstatin). Cells were pelleted by centrifugation at 2000 x g for 4 min at 4 C and resuspended in protease inhibitor containing sodium, dodecyl sulfate (SDS) lysis buffer [50 mM Tris-HCl, 10 mM EDTA, and 1% SDS (pH 8.1)]. Samples were sonicated on ice to lengths of 200-1000 bp and then centrifuged at 13,000 x g for 10 min to remove insoluble material. An equal amount of soluble chromatin from each sample was then diluted 10-fold in ChIP dilution buffer (16.7 mM Tris, 167 mM NaCl, 1.1% Triton X-100, and 0.01% SDS). Diluted soluble chromatin was precleaned by 20 µl salmon sperm DNA/protein A-agarose beads and mixed with 2 µg antibody (MyoD or Myf5) (Santa Cruz Biotechnology). Parallel preimmune control precipitation was performed by normal IgG. Antibodies and chromatin were allowed to mix and were rotated overnight at 4 C and precipitated with 60 µl salmon sperm DNA/protein A-agarose slurry for 2 h at 4 C with rotation to collect the antibody/chromatin complex. Beads were washed by rotation at 4 C with 1 ml buffer in the following order: low-salt immune complexes wash buffer [20 mM Tris-HCl (pH 8.1), 150 mM NaCl, 2 mM EDTA, 1%Triton X-100, 1% SDS], high-salt immune complex wash buffer [20 mM Tris-HCl (pH 8.1), 500 mM NaCl, 2 mM EDTA, 1%Triton X-100, 1% SDS], and LiCl immune complex wash buffer [10 mM Tris-HCl (pH 8.1), 0.25 mM LiCl, 1% deoxycholate, 1% Nonidet P-40, 0.1% SDS], followed by two washes with Tris-EDTA buffer [10 mM Tris-HCl, 1 mM EDTA (pH 8.1)]. Precipitated complexes were eluted twice for 15 min each at room temperature with 250 µl elution buffer (1% SDS, 0.1 M NaHCO3). Reversal of cross-linking was performed by heating the elution mixture at 65 C for 4 h in the presence of 20 µl 5 M NaCl. Proteins were digested with proteinase K by incubating for 1 h at 45 C in the presence of EDTA. DNA was then recovered by phenol-chloroform extraction and ethanol precipitation, followed by resuspension in nuclease-free water for PCR amplification. Immunoprecipitated DNA was analyzed by PCR amplification. For amplification of MyoD response element of Osx promoter, the primer set was 5'-ttgtc tgtgt tcatt catca-3' (forward,–829) and 5'-aaagg aacaa acccc aaacc-3' (reverse, –692). For amplification of Myf5 response element of Osx promoter, the primer set was 5'-gctgggaggg atctg ctcct ctctc-3' (forward, –780) and 5'-gtgta tgtca gtgtg ttcca gtctt-3' (reverse, –480).

Western analysis
Whole-cell extract was prepared, proteins separated by electrophoresis, and the MyoD band detected using a specific anti-MyoD antibody (Santa Cruz Biotechnology), as previously described (29).

Statistics
Statistical evaluation was done using GraphPad Software Prism version 3.03 (San Diego, CA). Differences between multiple groups were evaluated by ANOVA with multiple comparisons done by the method of Tukey. Comparison of multiple groups to a single control was done by the method of Dunnett. P < 0.05 was used to reject the null hypothesis. Experiments were done in triplicate and repeated.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The Osx promoter contains bHLH enhancers
Consensus binding analysis for bHLH homologous enhancers revealed 14 repeats of a Myf5 binding sequence (–669/–589) just upstream of the Osx2 promoter. Figure 1Go shows the relative location of the Osx1 promoter and a previously described TNF/MAPK suppressor region for reference (26). In addition, canonical CANNTG bHLH sites were identified. One of these is 120 bp upstream of a previously reported RUNX2 binding site (–849/–838) and two additional bHLH repeats are located further upstream (–1087/–1040). A less homologous myogenin consensus was also identified (–1000/–985) (not shown).


Figure 1
View larger version (16K):
[in this window]
[in a new window]

  		FIG. 1. Structure of the Osx proximal promoter showing two independently functioning transcription start sites (Osx1 and Osx2), putative bHLH enhancers, Myf5 repeat, and the previously described RUNX2 and TNF/MAPK regulatory elements. The sequence of the Myf5 repeat is shown below the promoter, and a consensus bHLH sequence is displayed above.

 
To assess the transcriptional response of the Osx promoter to MyoD or Myf5, we transfected the primitive pluripotent cell line, C3H/10T1/2, with a –1269/+91 Osx promoter-luciferase construct (Osx-luc) and vectors constitutively expressing either MyoD, Myf5, or empty vector control under the regulation of the cytomegalovirus promoter. The effect of bHLH factors on Osx promoter activity is shown in Fig. 2Go, A and B. Figure 2AGo shows a dose-dependent stimulation of the Osx promoter by MyoD that reaches 70-fold between 1 and 2 µg of MyoD plasmid. Myf5 expression (1 µg plasmid) also increased Osx promoter activity 6-fold (Fig. 2BGo).


Figure 2
View larger version (13K):
[in this window]
[in a new window]

  		FIG. 2. MyoD and Myf5 stimulate Osx promoter transcriptional activity. A, Dose response of MyoD expression plasmid stimulation of the Osx promoter. MyoD or control expression plasmid was transfected into C3H/10T1/2 cells with the –1269/+91 Osx-luc reporter. Results show relative stimulation of MyoD/control as arbitrary luciferase units corrected for transfection efficiency. B, Myf5 stimulates Osx promoter activity. Myf5 plasmid (1 µg) was transfected into C3H/10T1/2 cells with the Osx-luc reporter. Results show arbitrary luciferase units corrected for transfection efficiency. *, P < 0.05, Myf5 vs. empty vector control.

 
To determine whether the sequences identified by consensus analysis were conferring stimulation by MyoD or Myf5, the effect of promoter deletions on transcriptional activation was evaluated. Figure 3Go shows a schematic representation of the Osx promoter and deletion constructs used in the analysis. Reference labels are included to identify the two previously described regions of independent promoter activity, Osx1 and Osx2, the Myf5 repeat, and the putative MyoD sites. To the right of each construct is shown the effect of MyoD expression on promoter activity as a percentage of maximal promoter stimulation, where the –1269/+91 construct is arbitrarily assigned 100% (Fig. 3AGo). Stimulation by MyoD was conserved and slightly increased with deletion of the Osx1 promoter (Fig. 3BGo; –1269/–270). Further deletions showed that MyoD activity required the consensus MyoD sites but not the Myf5 repeat (Fig. 3Go, C and D). Inclusion of sequence 3' to the proximal bHLH site and the Runx2 site (Fig. 3EGo; –770/+91) resulted in maintenance of 40% of the MyoD activity, indicating a partial contribution of the sequence 3' to the bHLH element to this activity. Inclusion of all three bHLH sites retained almost 100% stimulation (Fig. 3FGo; –1100/+91). A MyoD response was not observed in a heterologous viral promoter containing three copies of the Myf5 sequence, confirming the lack of contribution of the Myf5 repeat to the MyoD response (Fig. 3GGo). Selective mutation of the bHLH regions was done to further determine their relative contribution to transcriptional activation by MyoD within the context of the full –1269/+91 reporter (Fig. 4Go). Mutation of the more proximal bHLH core sequence (–849/–838) completely eliminated the MyoD response, whereas deletion of the more distal (–999/989) region eliminated 40% of the activity.


Figure 3
View larger version (19K):
[in this window]
[in a new window]

  		FIG. 3. Localization of the MyoD response element. A map of the proximal Osx promoter is shown at the top with each deletion mutant shown below (A–G). The effect of 1 µg MyoD expression plasmid on Osx promoter activity is shown as percent maximal stimulation (% Max Stim) of the –1269/+91 Osx-luc reporter (70-fold over a pSV40 basic-luciferase control). Dotted vertical lines encompass the region required for MyoD responsiveness.

 

Figure 4
View larger version (27K):
[in this window]
[in a new window]

  		FIG. 4. Relative contribution of the bHLH homologous regions for MyoD responsiveness in the Osx promoter. A, Map of the control (wild-type) region of the modified Osx-luc reporter showing the three bHLH regions, 1, 2, and 3 (top), deletion of bHLH sites 1 and 2 within the otherwise intact –1269/+91 Osx promoter (middle), and deletion of bHLH site 3 (bottom). B, Effect of deletion of bHLH sites 1 + 2 or site 3 on the response to MyoD. MyoD (1 µg) expression plasmid or a control plasmid was transfected into C3H/10T1/2 cells with the indicated reporter plasmid. Results are mean ±SEM. *, P < 0.05, MyoD vs. empty control expression plasmid.

 
To determine whether the proximal bHLH region bound MyoD, EMSA was done. Recombinant MyoD was incubated with five 32P-labeled probes spanning the active bHLH site. Figure 5AGo shows the location of the probes. Strong binding of MyoD to probe 2 was observed (Fig. 5BGo). Probe 2 was the sequence that included the proximal bHLH binding site. These results were concordant with the mutational studies identifying the proximal bHLH site as conferring the majority of MyoD transcriptional activation. A small amount of binding was also observed using probe 3, a sequence that included the 3'-flanking region of the core bHLH site and partial MyoD activity (–770/+91 reporter, Fig. 3EGo).


Figure 5
View larger version (53K):
[in this window]
[in a new window]

  		FIG. 5. MyoD binds to the active bHLH region. A map of the Osx promoter is shown at the top for reference. 32P-labeled oligonucleotide probes 1–5 were synthesized corresponding to the indicated regions surrounding the bHLH site found to be active in Fig. 4Go. B, EMSA of probe binding to recombinant MyoD. The highest affinity binding is seen in probe 2, which spanned the region suspected to confer MyoD activity.

 
The effect of Osx promoter deletions on the response to Myf5 was then examined (Fig. 6Go). The –1269/+91 construct response to Myf5 was arbitrarily assigned as 100% of stimulation (Fig. 6AGo). When the entire region upstream of the Osx1 promoter was removed, Myf5 stimulation was mostly lost (Fig. 6BGo); however, a –665/+91 reporter construct, which included the entire Myf5 repeat, maintained only 38% of the response to Myf5 (Fig. 6CGo). Retention of the Myf5 repeat and the Runx2 site but no bHLH sites (Fig. 6DGo) or of the Myf5 repeat, the Runx2 site, and all three bHLH sites (–1100/+91; Fig. 6EGo), resulted in 46 and 52% activity, respectively. Thus, additional sequences upstream of –1100 are necessary for the full Myf5 response. This revealed that the Myf5 repeat did not fully account for the enhancer effect of Myf5. Furthermore, EMSA revealed only weak binding of recombinant Myf5 to the Myf5 sequence of the Osx promoter (not shown). Enhancer activity of the Myf5 sequence was confirmed by inserting three copies in a heterologous SV40 viral promoter (Fig. 6FGo). The response to Myf5 was consistently 10-fold less potent than the response to MyoD, despite the presence of 14 repeats of the Myf5 enhancer. ChIP assay confirmed in vivo binding of MyoD and Myf5 to their respective transcriptionally active regions (Fig. 7Go).


Figure 6
View larger version (18K):
[in this window]
[in a new window]

  		FIG. 6. Localization of the Myf5 response element on the Osx promoter. A map of the proximal Osx promoter is shown at the top with each deletion mutant shown below (A–F). The effect of 1 µg Myf5 expression plasmid on Osx promoter activity is shown as percent maximal stimulation (% Max Stim) of the –1269/+91 Osx-luc reporter (6-fold over a pSV40 basic luciferase reporter). Dotted vertical lines encompass a portion of the Myf5 repeat used to construct the SV40 heterologous promoter in F.

 

Figure 7
View larger version (34K):
[in this window]
[in a new window]

  		FIG. 7. ChIP assay showing Myf5 (primer set A) and MyoD (primer set B) occupancy of the regions encompassing the previously identified transcriptionally responsive elements of the Osx promoter. Myf5 or MyoD bands are indicated by the arrows. An input signal is shown in lane 1.

 
MyoD family transcription factors are expressed in differentiating pre-OB
To determine whether muscle transcription factors are expressed in primitive mesenchymal and pre-osteoblastic cells, MC3T3 cells (committed pre-OB) and C3H/10T1/2 cells (primitive pluripotent mesenchymal cells) were grown under conditions that support OB differentiation. Figure 7Go shows the results of qualitative PCR for Myf5, MyoD, myogenin, and Id1 mRNA on d 2 and 5 of culture. Myf5, myogenin, and Id1 were expressed in these cell lines on day 2 (Fig. 8AGo), and MyoD was expressed on d 2 and 5. Osx was also expressed abundantly by d 2 (Fig. 8AGo). In primary bone marrow stromal cell cultures that had been committed to OB differentiation, we did not observe MyoD or Myf5 expression. However, a PCR signal for MyoD and Myf5 was detectable in freshly isolated primary murine marrow stromal cells before plating (Myf5, seven of nine samples; MyoD, six of nine samples).


Figure 8
View larger version (48K):
[in this window]
[in a new window]

  		FIG. 8. MyoD family transcription factors are expressed in pre-OB cells. A, Qualitative RT-PCR for Myf5, myogenin, and Id1 mRNA on d 2 of culture in MC3T3 pre-OB cells and duplicate cultures of C3H/10T1/2 cells. Osx expression is shown on the right. B, Expression of MyoD on d 2 and 5 in MC3T3 and C3H/10T1/2 cells as measured by qualitative RT-PCR. GAPDH expression is shown on the right.

 
MyoD contributes to Osx expression and phenotype commitment of mesenchymal precursors
A C2C12 cell model was used to test whether endogenous MyoD regulates expression of Osx and OB differentiation. C2C12 cells are pluripotent myoblast precursors that spontaneously form organized myotubules in culture. Treatment of C2C12 cells with BMP-2 abolishes myotubule formation and stimulates the expression of OB phenotypic genes. We first determined the effect of BMP-2 treatment on the temporal expression of bHLH proteins and second, the effect of silencing MyoD on BMP-2-stimulated expression of Osx and the OB phenotypic gene alkaline phosphatase (AlkPhos). Figure 9Go shows the effect of BMP-2 treatment on the expression of the muscle factors and the bone-related factors. In the absence of BMP-2 treatment, C2C12 cells express increasing amounts of MyoD, Myf5, and myogenin, peaking on d 7, followed by a decline at d 9 (Fig. 9Go, A–C). No Osx or AlkPhos is detected in C2C12 cells in the absence of BMP treatment (Fig. 9Go, D and E). Treatment with BMP-2 reduced but did not eliminate the expression of MyoD and Myf5 and suppressed myogenin expression (Fig. 9Go, F–H). BMP-2 treatment also caused an early peak of Osx mRNA on d 2 with continued expression through d 9 and a peak of AlkPhos on d 7 (Fig. 9Go, I and J). Myotubule formation was observed uniformly throughout the cultures after d 5 in the absence of BMP-2, but no myotubules were observed in BMP-2-treated cultures (not shown), as reported previously (30). These results showed that MyoD and Myf5 expression persist in BMP-2-treated C2C12 cells as they acquire the OB phenotype.


Figure 9
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 9. Expression of muscle and bone transcription factors in C2C12 cells cultured with or without BMP-2. A–E, Expression of MyoD, Myf5, myogenin, Osx, and AlkPhos as measured by quantitative real-time RT-PCR without BMP-2 treatment; F and G, expression of the same mRNA with BMP-2 treatment. Induction of the OB phenotype (Osx and AlkPhos expression) by BMP-2 is associated with continued expression of MyoD and Myf5. Results are mean ± SEM. *, P < 0.05 vs. d-2 value.

 
We then examined the effect of silencing MyoD on Osx expression and OB differentiation using a siRNA. C2C12 cells were plated on d 0 and treated with or without BMP-2 beginning d 2. The siRNA targeting MyoD or a random sequence control siRNA was transfected on d 1, before BMP-2 treatment. Total RNA was collected on d 5 and 9 for analysis of MyoD, Osx, and AlkPhos expression. Figure 10Go shows that the siRNA targeting MyoD was specific. A 90% suppression of MyoD expression was observed in cells treated with the MyoD siRNA (Fig. 10AGo). The suppression of MyoD was maintained for 48 h, after which MyoD levels began to rise. Myf5 and myogenin were not significantly diminished 24 h after the siRNA treatment (Fig. 10Go, B and C). Western analysis revealed inhibition of MyoD protein by the siMyoD (Fig. 10DGo). The effect of silencing MyoD on Osx and AlkPhos expression is shown in Fig. 11Go. C2C12 cells were treated with BMP-2 with or without siMyoD or siControl on d 2 of culture, and the mRNA was measured on d 5 and 9. Silencing of MyoD caused a 50% or greater reduction in Osx mRNA expression on d 5 and 9, suggesting that endogenous MyoD was required for maximal Osx expression (Fig. 11AGo). MyoD silencing also prevented the expected rise in AlkPhos by d 9 (Fig. 11BGo).


Figure 10
View larger version (14K):
[in this window]
[in a new window]

  		FIG. 10. Specificity of siMyoD. A–C, C2C12 cells were transfected with 30 nM siMyoD or si control on d 2 of culture, and the expression of MyoD (A), Myf5 (B), and myogenin (C) was measured 24 h later. D, Western analysis of MyoD in cell extract from C2C12 cells treated with control Si, C, or Si-MyoD, Si. The middle lane is 30 nM Si and the right lane 60 nM Si. Results are mean ± SEM. *, P < 0.05 vs. si control.

 

Figure 11
View larger version (14K):
[in this window]
[in a new window]

  		FIG. 11. Silencing of MyoD inhibits BMP-2-stimulated differentiation toward the OB phenotype. C2C12 cells were treated with BMP-2 and either siMyoD (Si) or si control (C) on d 2 of culture. mRNA levels of Osx and AlkPhos were measured on d 5 and 9. siMyoD suppressed Osx expression and blocked the expected rise in AlkPhos on d 9. Results are mean ± SEM. *, P < 0.05 vs. si control on that day.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Our results suggest that MyoD, and possibly Myf5, stimulate the transcription of Osx and OB differentiation. MyoD is the most potent stimulator of Osx transcription that we have tested, exceeding by at least 10-fold the response to BMP-2 or homolog of Drosophila muscle segment 2 or the reported 2-fold response to RUNX2 (27). Consensus analysis identified several bHLH binding sites, a GC-rich Myf5 consensus repeat, and a homologous myogenin binding site upstream of the first Osx transcriptional start site. The regions conferring stimulation by MyoD and Myf5 were identified by selective deletions and mutations. For MyoD, the key responsive region spanned three bHLH homologous sequences 1100 bp upstream of the transcriptional start site. One of these bHLH sequences bound recombinant MyoD with high affinity. Discrete mutations of the bases critical for MyoD binding at this site or a broader deletion spanning this sequence abolished the stimulatory effect of MyoD. The MyoD binding site is located within a regulatory hot spot bound by numerous transcription factors including the RUNX2 organizer, homeobox proteins, and a potent TNF/MAPK suppressor. Our data also revealed some MyoD enhancer activity of DNA 3' to the critical bHLH site. This site bound MyoD with low affinity and conferred some transcriptional response to MyoD. However, the proximal bHLH site was absolutely required for MyoD action in the context of the –1269/+91 promoter. The role of RUNX2 in MyoD action has not been determined. Overexpression of Runx2 alone has been shown to increase Osx promoter activity 2-fold while potently suppressing MyoD expression (27, 31). RUNX2 functions as an organizer of the osteocalcin promoter, where it permits histone acetylation and a transcriptional response to the vitamin D receptor (32, 33). More data are needed to determine whether RUNX2 acts cooperatively with MyoD for histone remodeling and transcriptional activation.

Transfection of cells with a Myf5 expression plasmid stimulated the Osx promoter less potently. Fourteen repeats of a GC-rich Myf5 consensus binding site partly confers this response, although full activity requires additional upstream sequence. The authenticity of the Myf5 enhancer was confirmed by showing that Myf5 transcriptional activation could be transferred to a heterologous viral promoter. We did not observe an additive effect on Osx expression with transfection of both MyoD and Myf5 (not shown). Additional work will be needed to determine whether these transcription factors function as redundant stimuli for the Osx promoter in vivo. ChIP assay did confirm that the identified promoter regions for MyoD and Myf5 were occupied in MC3T3 cells, supporting regulation of the promoter by the muscle transcription factors.

The requirement for MyoD in muscle development is well established, but less is known regarding a possible role in skeletal development. MyoD is expressed in the developing mouse embryo by embryonic d 10.5, simultaneously with the onset of skeletal formation, and is therefore temporally available to regulate the selection of both myoblasts and early OB (34). We found that MyoD and Myf5 are expressed in mesenchymal and pre-osteoblastic cell lines. In addition, both MyoD and Myf5 are expressed in freshly isolated bone marrow stromal cells. Although some reports indicate that primary cultures or C2C12 cells lose detectable MyoD expression after stimulation with BMP, we found that MyoD and Myf5 remained detectable and that levels continued to rise as the cells acquired the OB phenotype (14, 15, 31). Our results showed that BMP-2 potently suppressed myogenin, which is required for myotubule formation as MyoD expression was retained. These results for BMP treatment are consistent with those of Katagiri et al. (15, 30) and Gersbach et al. (31). Using E-box oligonucleotide probes, Kazhdan et al. (35) showed that binding of bHLH factors could be demonstrated throughout differentiation in a variety of osteogenic precursors, although the specific bHLH proteins were not identified. More recently, a laser-capture RNA technique readily detected MyoD mRNA in individual mature OB from adult bone specimens as recognized by concurrent expression of AlkPhos and osteocalcin in these cells (36). Thus, MyoD is present in the precursors as they commit to OB differentiation and may be expressed at lower levels after the onset of differentiation.

We found that silencing of MyoD completely suppressed BMP-2-stimulated AlkPhos expression in C2C12 cells. The siRNA to MyoD also suppressed Osx 50% or more. The siRNA to MyoD was applied on d 2 of C2C12 culture and most active for the first 24–48 h after treatment. Commitment to the OB phenotype occurs within the first 24–48 h of these cultures and requires the expression of Osx and RUNX2. Later in the culture, Osx and RUNX2 may not be as important. Thus, the 90% suppression of MyoD by the siRNA during the first few days of culture may have been sufficient to block commitment to the OB pathway and the later expression of AlkPhos on d 9. Alternatively, other MyoD-regulated genes may be required for OB differentiation. Komaki et al. (37) previously showed that fetal calvaria precursors from MyoD–/– mice, which have a grossly normal skeleton, fail to differentiate to OB after treatment with BMP-7, unlike MyoD+/+ cells, which respond to BMP-7 with expression of Osx, RUNX2, and AlkPhos. Furthermore, forced expression of MyoD in MyoD–/– cells restored the response to BMP-7 and expression of OB phenotypic genes. Thus, our results support the previous work of Komaki et al. (37), in which MyoD was completely absent, and suggest that MyoD regulation of the Osx promoter may explain, in part, the requirement for MyoD in OB differentiation.

In vivo models of bHLH transcription factor action yield clues on their potential role in bone. For myogenesis, MyoD and Myf5 stimulate myoblast formation from mesenchyme, after which myogenin promotes the formation of myotubules (13). Deletion of MyoD delays myoblast differentiation without disturbing the final muscle phenotype (12). Deletion of Myf5 or Myf6 (which induces Myf5) also delays but does not abolish early myotome formation and reduces muscle mass somewhat, but most muscle formation is ultimately accomplished (10, 11). However, double knockouts of MyoD and Myf6 reveal a severe muscle deficiency, suggesting a functional redundancy of the bHLH factors (13). Redundancy of bHLH factors could explain the lack of a severe skeletal phenotype in MyoD knockout mice. The skeletons of MyoD or Myf5 single gene knockout mice have not been studied in detail to determine whether there is a reduced bone formation rate, low bone density, or subtle defect in patterning. Recently, abnormal endochondral bone formation was described in mice with combined deletion of Myf5 and MyoD (38, 39). These Myf5–/– MyoD–/– mice had fusion of cervical vertebrae, shortened long bones, clavicular hypoplasia, cleft palate and sternum, and other findings. The authors attributed the abnormal skeletal development to a lack of musculature-dependent loading forces in utero rather than a deficiency of these transcription factors that could contribute to the commitment of precursor cells to OB. The Osx promoter is a potential target of the redundant activity of MyoD and Myf5 during development.

The muscle-bone relationship is complex and includes a variety of shared signals during development as well as a reciprocal postnatal communication that influences the structure and function of both organ systems. During early development, wingless factors (Wnt), which are required for OB differentiation, stimulate expression of MyoD while simultaneously promoting the differentiation of myoblasts (40). Somewhat later, BMPs 2, 4, and 7 regulate muscle and bone in opposite directions through smad activation (41, 42). The regulation of phenotype commitment and development of the mature OB or muscle phenotype cannot be ascribed to the expression of a single transcription factor, rather, a combinatorial code involving sets of factors determine gene expression. Such a code may provide a greater selectivity, specificity, and demarcation between adjacent tissue types in development. In adults, muscle also signals bone by providing physical forces that conform skeletal structure to function and also through soluble signals, including myostatin (GDF8), a muscle secretory protein that inhibits hypertrophy of muscle while increasing bone formation and bone density (43). Our results provide one example of the mechanism through which cross-talk between skeletal and muscle tissues can occur at the level of a shared transcription factor.


   Footnotes
 
This work was supported by an Endocrine Fellows Foundation Grant (to J.H.) and a Veterans Affairs Merit Review Grant (to M.S.N.).

Disclosure Statement: The authors have nothing to disclose.

First Published Online March 27, 2008

1 J.H. and X.L. contributed equally to the work. Back

Abbreviations: AlkPhos, Alkaline phosphatase; bHLH, basic helix-loop-helix; BMP, bone morphogenetic protein; ChIP, chromatin immunoprecipitation; OB, osteoblast; Osx, osterix; RUNX2, runt-related factor; SDS, sodium dodecyl sulfate; siRNA, small interfering RNA.

Received November 13, 2007.

Accepted for publication March 14, 2008.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Komori T 2006 Regulation of osteoblast differentiation by transcription factors. J Cell Biochem 99:1233–1239[CrossRef][Medline]
  2. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G 1997 Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89:747–754[CrossRef][Medline]
  3. Komori T, Yagi H, Nomura S, Yamaguchi A, Sasaki K, Deguchi K, Shimizu Y, Bronson RT, Gao YH, Inada M, Sato M, Okamoto R, Kitamura Y, Yoshiki S, Kishimoto T 1997 Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89:755–764[CrossRef][Medline]
  4. Karsenty G 2001 Transcriptional control of osteoblast differentiation. Endocrinology 142:2731–2733[Abstract/Free Full Text]
  5. Li YL, Xiao ZS 2007 Advances in Runx2 regulation and its isoforms. Med Hypotheses 68:169–175[CrossRef][Medline]
  6. Lian JB, Javed A, Zaidi SK, Lengner C, Montecino M, van Wijnen AJ, Stein JL, Stein GS 2004 Regulatory controls for osteoblast growth and differentiation: role of Runx/Cbfa/AML factors. Critical Rev Eukaryot Gene Expr 14:1–41[CrossRef]
  7. Nakashima K, Zhou X, Kunkel G, Zhang Z, Deng JM, Behringer RR, de Crombrugghe B 2002 The novel zinc finger-containing transcription factor osterix is required for osteoblast differentiation and bone formation. Cell 108:17–29[CrossRef][Medline]
  8. Tai G, Polak JM, Bishop AE, Christodoulou I, Buttery LD 2004 Differentiation of osteoblasts from murine embryonic stem cells by overexpression of the transcriptional factor osterix. Tissue Eng 10:1456–1466[Medline]
  9. Christ B, Brand-Saberi B 2002 Limb muscle development. Int J Dev Biol 46:905–914[Medline]
  10. Braun T, Arnold HH 1995 Inactivation of Myf-6 and Myf-5 genes in mice leads to alterations in skeletal muscle development. EMBO J 14:1176–1186[Medline]
  11. Braun T, Rudnicki MA, Arnold HH, Jaenisch R 1992 Targeted inactivation of the muscle regulatory gene Myf-5 results in abnormal rib development and perinatal death. Cell 71:369–382[CrossRef][Medline]
  12. Kablar B, Krastel K, Ying C, Asakura A, Tapscott SJ, Rudnicki MA 1997 MyoD and Myf-5 differentially regulate the development of limb versus trunk skeletal muscle. Development 124:4729–4738[Abstract]
  13. Rawls A, Valdez MR, Zhang W, Richardson J, Klein WH, Olson EN 1998 Overlapping functions of the myogenic bHLH genes MRF4 and MyoD revealed in double mutant mice. Development 125:2349–2358[Abstract]
  14. Murray SS, Murray EJ, Glackin CA, Urist MR 1993 Bone morphogenetic protein inhibits differentiation and affects expression of helix-loop-helix regulatory molecules in myoblastic cells. J Cell Biochem 53:51–60[CrossRef][Medline]
  15. Katagiri T, Yamaguchi A, Komaki M, Abe E, Takahashi N, Ikeda T, Rosen V, Wozney JM, Fujisawa-Sehara A, Suda T 1994 Bone morphogenetic protein-2 converts the differentiation pathway of C2C12 myoblasts into the osteoblast lineage. J Cell Biol 127:1755–1766[Abstract/Free Full Text]
  16. Gitelman SE, Kirk M, Ye JQ, Filvaroff EH, Kahn AJ, Derynck R 1995 Vgr-1/BMP-6 induces osteoblastic differentiation of pluripotential mesenchymal cells. Cell Growth Differ 6:827–836[Abstract]
  17. Yamaguchi A 1995 Regulation of differentiation pathway of skeletal mesenchymal cells in cell lines by transforming growth factor-β superfamily. Semin Cell Biol 6:165–173[Medline]
  18. Bain G, Muller T, Wang X, Papkoff J 2003 Activated β-catenin induces osteoblast differentiation of C3H10T1/2 cells and participates in BMP2 mediated signal transduction. Biochem Biophys Res Commun 301:84–91[CrossRef][Medline]
  19. Bennett CN, Longo KA, Wright WS, Suva LJ, Lane TF, Hankenson KD, MacDougald OA 2005 Regulation of osteoblastogenesis and bone mass by Wnt10b. Proc Natl Acad Sci USA 102:3324–3329[Abstract/Free Full Text]
  20. Gong Y, Slee RB, Fukai N, Rawadi G, Roman-Roman S, Reginato AM, Wang H, Cundy T, Glorieux FH, Lev D, Zacharin M, Oexle K, Marcelino J, Suwairi W, Heeger S, Sabatakos G, Apte S, Adkins WN, Allgrove J, Arslan-Kirchner M, Batch JA, Beighton P, Black GC, Boles RG, Boon LM, Borrone C, Brunner HG, Carle GF, Dallapiccola B, De Paepe A, Floege B, Halfhide ML, Hall B, Hennekam RC, Hirose T, Jans A, Juppner H, Kim CA, Keppler-Noreuil K, Kohlschuetter A, LaCombe D, Lambert M, Lemyre E, Letteboer T, Peltonen L, Ramesar RS, Romanengo M, Somer H, Steichen-Gersdorf E, Steinmann B, Sullivan B, Superti-Furga A, Swoboda W, van den Boogaard MJ, Van Hul W, Vikkula M, Votruba M, Zabel B, Garcia T, Baron R, Olsen BR, Warman ML 2001 LDL receptor-related protein 5 (LRP5) affects bone accrual and eye development. Cell 107:513–523[CrossRef][Medline]
  21. Mbalaviele G, Sheikh S, Stains JP, Salazar VS, Cheng SL, Chen D, Civitelli R 2005 β-Catenin and BMP-2 synergize to promote osteoblast differentiation and new bone formation. J Cell Biochem 94:403–418[CrossRef][Medline]
  22. Rodda SJ, McMahon AP 2006 Distinct roles for Hedgehog and canonical Wnt signaling in specification, differentiation and maintenance of osteoblast progenitors. Development 133:3231–3244[Abstract/Free Full Text]
  23. Ridgeway AG, Petropoulos H, Wilton S, Skerjanc IS 2000 Wnt signaling regulates the function of MyoD and myogenin. J Biol Chem 275:32398–32405[Abstract/Free Full Text]
  24. Lee MH, Kwon TG, Park HS, Wozney JM, Ryoo HM 2003 BMP-2-induced Osterix expression is mediated by Dlx5 but is independent of Runx2. Biochem Biophys Res Commun 309:689–694[CrossRef][Medline]
  25. Celil AB, Campbell PG 2005 BMP-2 and insulin-like growth factor-I mediate osterix (Osx) expression in human mesenchymal stem cells via the MAPK and protein kinase D signaling pathways. J Biol Chem 280:31353–31359[Abstract/Free Full Text]
  26. Lu X, Gilbert L, He X, Rubin J, Nanes MS 2006 Transcriptional regulation of the osterix (Osx, Sp7) promoter by tumor necrosis factor identifies disparate effects of mitogen-activated protein kinase and NF{kappa}B pathways. J Biol Chem 281:6297–6306[Abstract/Free Full Text]
  27. Nishio Y, Dong Y, Paris M, O'Keefe RJ, Schwarz EM, Drissi H 2006 Runx2-mediated regulation of the zinc finger Osterix/Sp7 gene. Gene 372:62–70[CrossRef][Medline]
  28. Cartharius K, Frech K, Grote K, Klocke B, Haltmeier M, Klingenhoff A, Frisch M, Bayerlein M, Werner T 2005 MatInspector and beyond: promoter analysis based on transcription factor binding sites. Bioinformatics 21:2933–2942[Abstract/Free Full Text]
  29. Gilbert L, He X, Farmer P, Rubin J, Drissi H, van Wijnen AJ, Lian JB, Stein GS, Nanes MS 2002 Expression of the osteoblast differentiation factor RUNX2 (Cbfa1/AML3/Pebp2{alpha}A) is inhibited by tumor necrosis factor-{alpha}. J Biol Chem 277:2695–2701[Abstract/Free Full Text]
  30. Katagiri T, Akiyama S, Namiki M, Komaki M, Yamaguchi A, Rosen V, Wozney JM, Fujisawa-Sehara A, Suda T 1997 Bone morphogenetic protein-2 inhibits terminal differentiation of myogenic cells by suppressing the transcriptional activity of MyoD and myogenin. Exp Cell Res 230:342–351[CrossRef][Medline]
  31. Gersbach CA, Byers BA, Pavlath GK, Garcia AJ 2004 Runx2/Cbfa1 stimulates transdifferentiation of primary skeletal myoblasts into a mineralizing osteoblastic phenotype. Exp Cell Res 300:406–417[CrossRef][Medline]
  32. Shen J, Hovhannisyan H, Lian JB, Montecino MA, Stein GS, Stein JL, Van Wijnen AJ 2003 Transcriptional induction of the osteocalcin gene during osteoblast differentiation involves acetylation of histones h3 and h4. Mol Endocrinol 17:743–756[Abstract/Free Full Text]
  33. Shen J, Montecino M, Lian JB, Stein GS, Van Wijnen AJ, Stein JL 2002 Histone acetylation in vivo at the osteocalcin locus is functionally linked to vitamin D-dependent, bone tissue-specific transcription. J Biol Chem 277:20284–20292[Abstract/Free Full Text]
  34. Patapoutian A, Yoon JK, Miner JH, Wang S, Stark K, Wold B 1995 Disruption of the mouse MRF4 gene identifies multiple waves of myogenesis in the myotome. Development 121:3347–3358[Abstract]
  35. Kazhdan I, Rickard D, Leboy PS 1997 HLH transcription factor activity in osteogenic cells. J Cell Biochem 65:1–10[CrossRef][Medline]
  36. Benoyahu D, Akavia UD, Socher R, Shur I 2005 Gene expression in skeletal tissues: application of laser capture microdissection. J Microsc 220:1–8[CrossRef][Medline]
  37. Komaki M, Asakura A, Rudnicki MA, Sodek J, Cheifetz S 2004 MyoD enhances BMP7-induced osteogenic differentiation of myogenic cell cultures. J Cell Sci 117:1457–1468[Abstract/Free Full Text]
  38. Rot-Nikcevic I, Downing KJ, Hall BK, Kablar B 2007 Development of the mouse mandibles and clavicles in the absence of skeletal myogenesis. Histol Histopathol 22:51–60[Medline]
  39. Rot-Nikcevic I, Reddy T, Downing KJ, Belliveau AC, Hallgrimsson B, Hall BK, Kablar B 2006 Myf5–/– :MyoD–/– amyogenic fetuses reveal the importance of early contraction and static loading by striated muscle in mouse skeletogenesis. Dev Genes Evol 216:1–9[CrossRef][Medline]
  40. Schmidt M, Tanaka M, Munsterberg A 2000 Expression of β-catenin in the developing chick myotome is regulated by myogenic signals. Development 127:4105–4113[Abstract]
  41. Amthor H, Christ B, Patel K 1999 A molecular mechanism enabling continuous embryonic muscle growth: a balance between proliferation and differentiation. Development 126:1041–1053[Abstract]
  42. Amthor H, Christ B, Weil M, Patel K 1998 The importance of timing differentiation during limb muscle development. Curr Biol 8:642–652[CrossRef][Medline]
  43. Hamrick MW, Shi X, Zhang W, Pennington C, Thakore H, Haque M, Kang B, Isales CM, Fulzele S, Wenger KH 2007 Loss of myostatin (GDF8) function increases osteogenic differentiation of bone marrow-derived mesenchymal stem cells but the osteogenic effect is ablated with unloading. Bone 40:1544–1553[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
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 Request Copyright Permission
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hewitt, J.
Right arrow Articles by Nanes, M. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hewitt, J.
Right arrow Articles by Nanes, M. S.

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