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.2005-0438
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
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 zur Nieden, N. I.
Right arrow Articles by Rancourt, D. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by zur Nieden, N. I.
Right arrow Articles by Rancourt, D. E.
Molecular Endocrinology 21 (3): 674-685
Copyright © 2007 by The Endocrine Society

Gene Profiling on Mixed Embryonic Stem Cell Populations Reveals a Biphasic Role for ß-Catenin in Osteogenic Differentiation

Nicole I. zur Nieden, Feodor D. Price, Lesley A. Davis, Rebecca E. Everitt and Derrick E. Rancourt

University of Calgary, Faculty of Medicine, Institute of Maternal and Child Health, Calgary, Alberta, Canada T2N 4N1

Address all correspondence and requests for reprints to: Nicole I. zur Nieden, Fraunhofer Institute for Cell Therapy and Immunology, Deutscher Platz 5e, 04103 Leipzig, Germany. E-mail: nicole.zurnieden{at}izi.fraunhofer.de.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The differentiation of embryonic stem cells (ESCs) into osteoblasts is enhanced to 60% when exposed to vitamin D3 (VD3) but leaves a remainder of one half of the cell population unidentified. To increase differentiation outcome, the known osteoinducers retinoic acid (RA) and bone morphogenetic protein-2 (BMP-2) were evaluated. Initial studies using RA and BMP-2 during early osteogenesis in addition to VD3 increased osteogenic yield in the case of RA, but surprisingly decreased osteogenesis when BMP-2 was administered together with VD3 or RA. This paper describes a comprehensive microarray study examining the gene expression profile of differentiating osteoblasts in these mixed ESC populations. In addition to five other families of signaling molecules (insulin growth factors, prostaglandin, follistatin, TGFß2, and Wnt molecules), we identified an endogenous expression pattern for BMPs and RA that differed from our previous exogenous administration of these molecules. By mimicking the change in expression of the RA and BMP-2 families with exogenous supplementation at the correct time, it was then possible to increase the number of ESC-derived osteoblasts to 90%. This effect was mediated through alteration in ß-catenin (CatnB) expression levels and nuclear CatnB activity, both of which are modulated by VD3, RA, and BMP-2. Our results suggest that blockage of CatnB activity by VD3 and RA is opposed by induction of CatnB activity through BMP-2 when administered together. Hence, osteoinduction, in vitro, is an intricate process involving both temporal and quantitative changes in gene expression and CatnB activity.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
TO GROW NEW bone and restore bone strength and function lost in injury or disease, such as osteoporosis and osteoarthritis, we first must know what controls bone differentiation, growth, and destruction. A new model system for evaluating bone development has recently emerged using mouse embryonic stem cells (ESCs). Murine ESCs are pluripotent cells, derived from the inner cell mass, a component of the blastocyst. They can be maintained in permanent culture without loosing their pluripotent characteristics by the addition of leukemia inhibitory factor. Upon withdrawal of leukemia inhibitory factor, ESCs form a three-dimensional embryo-like structure [embryoid body (EB)] and undergo stochastic differentiation into cells of all three germ layers. However, ESCs can be prompted to differentiate into particular lineages by supplementation with specific inducing agents or growth factors. In the last 5 yr, protocols have been established that allow differentiation of the two skeletal cell types: osteoblasts and chondrocytes.

We and others have shown that an alkaline environment, which is provided by adding ß-glycerophosphate and ascorbic acid, is necessary to force the cells to secrete a mineralized matrix (1, 2). Vitamin D3 (VD3) acts as an additional trigger for ESCs to activate the alkaline phosphatase (ALP) enzyme and to express bone marker genes, including Cbfa1, collagen type I (Col I), osteocalcin, and bone sialoprotein (BSP), in a temporally coordinated fashion (2). A proliferation stage (phase I) is succeeded by differentiation (phase II), mineralization (phase III), and maturation (phase IV), each stage being characterized by expression of certain markers. Buttery et al. (3) reported that supplementation with dexamethasone at d 7–14 of EB differentiation induced expression of osteoblast marker genes and mineralization. Phillips et al. (1) showed that retinoic acid (RA) is important for the commitment of ESCs into osteoblasts, demonstrating enhanced expression of several osteoblast marker genes, including osteocalcin, ALP, and osteopontin by differentiation d 15. Bone morphogenetic protein-2 (BMP-2) is used in the ESC model to stimulate chondrogenic differentiation (4, 5) but has also been described by Phillips et al. (1) to be a downstream target of statins that can enhance osteogenesis in ESCs through regulation of ALP and osteocalcin.

Cell therapy in a transplantation setting requires the cell population to be introduced into a recipient to be extremely pure. ESCs have been shown to form teratomas when transplanted into a syngeneic host (6). Here, teratoma formation was not prevented by subjecting ESCs to in vitro differentiation into neurons before injection. Although the established lineage selection protocol for the generation of ESC-derived neurons has been extensively characterized and moreover yields a very high percentage of neurons, the slightest impurity may cause teratoma formation. To date, VD3-induced osteogenesis yields only 40–60% osteoblasts as estimated by morphology. Ultimately, it will no longer be sufficient to describe the capability of ESCs to differentiate into a certain lineage, but rather define conditions that result in the highest possible yield or devise strategies to purify the desired cell type. Unfortunately, no exclusive markers exist for the osteoblast lineage to allow for the purification of a progenitor population. Thus, we have investigated other osteoinducers like RA and BMP-2 that together with VD3 could enhance osteogenic outcome. In the course of this study, we have undertaken a large microarray study of ESC-derived osteogenesis using EBs at different time points that contain mixed cell populations. As a result of this study, we could show that timing in supplementation is critical to the point that incorrect timing could induce the opposite effect. Furthermore, all three osteoinducers seem to regulate the activity of ß-catenin (CatnB) in the nucleus, a molecule known to be involved in various cellular processes of normal and cancerous cells, such as proliferation and differentiation.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Effect of RA and BMP-2 on VD3-Induced Osteogenesis Early in Differentiation
RA and BMP-2 have been described by Phillips et al. (1) as osteogenic enhancers in ESCs. We were interested to see whether they would act similarly in the cultures that were triggered into osteogenesis by VD3 and possibly would enhance the osteogenic response. EBs were treated from d 3 onward (phase II) with RA or BMP-2 and starting with d 5 (phase III) with or without VD3 on two kinds of substrates, which could potentially increase the osteoblast yield further. Figure 1Go shows the morphology of these cultures treated with various inducer combinations when adherent to Primaria-coated plasticware or Col I-coated plasticware, respectively (Fig. 1AGo). Mineralized osteoblasts appear as black dots and clusters in phase contrast and have been shown to correspond to osteocalcin-immunopositive and tetracycline-incorporating cells (2). Incubation of the cultures with 0.5 M EDTA for 20 min decreased this black stain (Fig. 1AGo), further suggesting the presence of minerals in the matrix. In general, a higher degree of mineralization was achieved by culturing the EBs on the collagen substrate. VD3 as well as RA and BMP-2 alone were sufficient to direct differentiation into mineralized cells in the following order: RA greater than VD3 greater than BMP-2, sorted by the magnitude of the response.


Figure 1
View larger version (46K):
[in this window]
[in a new window]

  		Fig. 1. Treatment of EBs with RA and BMP-2 Early in Differentiation

A, Morphology of mineralized EBs on d 30 of culture. Black cells correspond to mineralized osteocalcin-immunopositive and tetracycline-responding osteoblasts (2 ). Top panel, VD3-induced osteogenic EBs were incubated with 0.5 M EDTA, a demineralizing agent. Black cells completely disappear after 20 min. Middle and bottom panels, EBs were supplemented with various combinations of VD3 (5 x 10–8 M), BMP-2 (10 ng/µl), and RA (1 x 10–7 M) as indicated with the addition of ß-glycerophosphate and ascorbic acid (GPAA) on Primaria- or Col I-coated plasticware. Although all three compounds induced mineralization in the EBs by themselves, combining VD3 and RA enhanced osteogenesis. Conversely, every combination that contained BMP-2 diminished the osteogenic response. Bar, 500 µm. B and C, VD3, BMP-2, and RA were given as indicated with or without the addition of ß-glycerophosphate and ascorbic acid (GPAA) on Primaria (black bars)- or Col I (white bars)-coated plasticware. The expression of three marker genes for bone tissue was quantified by real-time PCR. Expression levels were standardized against GAPDH and normalized to corresponding nontreated, spontaneously differentiated controls (B). Determination of Ca2+ content in the EBs showed that, similar to the expression of bone markers, RA increased bone formation and BMP-2 decreased mineralization when given from d 3 onward (C). Addition of GPAA was found to be essential for expression of marker genes and mineralization. Increased osteogenesis was discovered when EBs were grown on Col I. All values represent means of three independent experiments ± SD. *, P < 0.1; **, P < 0.01; ***, P < 0.001.

 
Similar conclusions can be drawn from studies investigating several bone markers, such as expression analysis of the characteristic marker genes osteocalcin, BSP, and Cbfa1, and evaluation of the degree of mineralization equal to the calcium content of the matrix (Fig. 1Go, B and C). However, in the combinatorial treatments, RA and VD3 enhanced bone marker gene expression and calcium content compared with single treatment, i.e. osteocalcin expression was greater in RA plus BMP than in BMP-2 alone, VD3 plus BMP was greater than in BMP-2 alone or RA plus VD3 was greater than either one alone. However, BMP-2 surprisingly decreased osteogenic marker expression in combination with the other two supplements (RA plus BMP-2 < RA; RA plus VD3 plus BMP-2 < RA plus VD3).

We subsequently determined that this effect was concentration dependent: higher concentrations of BMP-2 (up to 100 ng/ml) decreased osteocalcin and BSP expression in combination with VD3 and RA (supplemental Fig. 1, published as supplemental data on The Endocrine Society’s Journals Online web site at http://mend.endojournals.org). In contrast, RA increased expression of these two exclusive bone markers in a concentration-dependent manner. The initial VD3 concentration of 5 x 10–8 M seems to be critical, because the cells treated with 1 x 10–7 M VD3 had decreased osteoblast marker expression.

Expression Analysis Using Microarray Profiling
The down-regulation of osteogenic markers through BMP-2 was contradictory to its described role in the literature. Accordingly, we have reconciled that high-fidelity microarray analysis may be a useful approach to identify changes in gene expression during important stages of differentiation, such as in response to growth factors.

Total RNA derived from d-4-, -5-, -6-, -8-, -16-, -25-, and -30-old embryoid bodies and their VD3-treated counterparts (d 6–30) was hybridized to the Affymetrix M430A genomic chip. Only probes that showed an absent or present signal in at least one of the time points over all three biological replicates were considered for further analysis. Probes identified in a pairwise comparison that were 2-fold or higher differentially regulated in all three replicates were taken into the final probe set, after statistical analysis using a P value cutoff of 0.05. A pairwise comparison of d 4 vs. 5 of differentiation is shown in supplemental Tables 1 and 2 (published as supplemental data on The Endocrine Society’s Journals Online web site at http://mend.endojournals.org). At d 4, some of the pluripotency markers, such as Oct4 and nodal are still found in the gene set. Additionally, early differentiation markers are highly expressed (supplemental Table 1), such as nestin, {alpha} cardiac myosin heavy chain, and members of the canonical Wnt pathway (Wnt8a, cyclin D2, LPR). Mesodermal markers or genes known to induce mesoderm formation are up-regulated at d 5, such as TGFß2 or fibroblast growth factor receptor 3 (supplemental Table 2). Furthermore, genes implicated in early osteogenesis and matrix formation are already increasingly expressed compared with d 4, for example procollagen type I in its {alpha}1 and {alpha}2 isoform, FosB and twist. Cbfa1/runx2 was significantly up-regulated 1.7-fold at d 25. However, because of the 2-fold cutoff that we chose, it was not incorporated into our final data set. Instead, we noted a high increase in another isoform, Cbfa2T1 (AML1T1) at d 5, 6, and 8 (7.73-fold up-regulation at d 8).

Across the five time points d 6–30, where osteogenic treatment was compared with spontaneously differentiated controls, a total of 460 genes were found to be up-regulated. Supplemental Table 3 (published as supplemental data on The Endocrine Society’s Journals Online web site at http://mend.endojournals.org) summarizes their n-fold expression above spontaneously differentiated controls. Figure 2Go depicts a pie chart representing the composition of the regulated genes at the respective days according to their GO terminology. Cell adhesion-related genes including Caecam, integrins, and cadherin 2 increased from 5–20% and calcium binding molecules from 0–3%. Additionally, various collagens, collagen-synthesizing or -metabolizing enzymes or other matrix molecules, for instance Col I, Col III, Col IV, Col XVII, MMP-13, TIMP, and Sparc, were up-regulated as well. Alongside various bone markers, such as tescalcin, aquaporin 4, and activin A, we also found evidence for chondrocyte development in the cultures, such as Hsp47, cartilage link protein or cartilage-associated protein (Ctrap), arguing in favor of endochondral bone formation induced by VD3 (5).


Figure 2
View larger version (77K):
[in this window]
[in a new window]

  		Fig. 2. Clustering of Up-Regulated Genes

A, Clustering according to n-fold up-regulation. B, Pie chart depicting classification of identified probes according to their GO terminology. Shown are individual probe sets for d 6–30. The percentage of calcium binding and cell adhesion-related genes increases over the time course of differentiation due to the secretion and calcification of an extracellular matrix.

 
Interestingly, BMP family members are expressed throughout the differentiation process. However, BMP-2, which we used in our initial experiments, is expressed at later stages of differentiation on d 25 and 30 only (phase IV). RA family members are increasingly expressed starting at d 4 and not d 3, as had been previously used as supplementation period for RA. The expression of some members of this pathway, such as cRAbp1 and STRA13 is down-regulated after d 6, whereas others (Rbp4, Dusp6) increase with time.

In addition to the pathways that we had previously examined, five other pathway families were identified in the cell signaling group, representing another significant portion of up-regulated genes. The signaling molecules in those families are IGF, prostaglandin, follistatin, TGFß2, and Wnts (Fig. 3Go). IGF signaling appears to be regulated in the early stages of differentiation, peaking at d 16 and dropping back down afterward. The same expression pattern can be found for follistatins, and the TGF family. Prostaglandins, in contrast, are up-regulated during phase IV of differentiation at d 25 and 30. The three members of the Wnt pathway, Wnt8a, Wnt5a, and Wnt4 all follow a different expression pattern. Wnt8a is the one that is expressed first and is superseded by Wnt5a at d 5. Wnt4 is then taking over at later stages.


Figure 3
View larger version (36K):
[in this window]
[in a new window]

  		Fig. 3. Temporal Expression Patterns of the Identified Families of Signaling Molecules or Transcription Factors

The expression of each gene identified to be up-regulated in VD3-treated cultures was plotted over a 30-d period (d 4, 5, 6, 8, 16, 25, and 30). Graphs show that BMP-2 was increasingly expressed over controls at later stages of differentiation at d 16–25 and 25–30, respectively. RA seems to have a role in earlier differentiation because genes induced by RA are incrementally expressed starting on d 5 and 6 of culture. The y-axis shows the mean normalized hybridization signals of the three biological triplicates of VD3-induced EBs.

 
Validation of Microarray Data
Semiquantitative PCR experiments for 10 genes were performed to confirm the array results. In particular, genes with low expression levels and a close to 2-fold up-regulation were chosen to prove the reliability of our approach. Supplemental Figure 2 (published as supplemental data on The Endocrine Society’s Journals Online web site at http://mend.endojournals.org) shows the results of representative RT-PCR results using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a reference. The results might not all quantitatively reflect the fold difference of the microarray data, but they all agree with the microarray results in the trend they show. Because the cultures represent mixed populations of cells, we validated three of the genes by immunocytochemistry (supplemental Fig. 3, published as supplemental data on The Endocrine Society’s Journals Online web site at http://mend.endojournals.org). Wnt4, Sox 7, and Sox 17 all showed the exact expression patterns as was found with the microarray. No staining for Wnt4 can be found before d 19. Sox 7 and Sox 17 are expressed from d 15 onward. Moreover, staining occurred in cells that were mineralized as shown in phase contrast. This underscores the validity of our effort, because the identified probes could actually be localized to the osteoblasts in the culture.

RA Acts on Early Progenitors in Phase III of Osteogenesis
As described above, genes that can be associated with RA signaling, synthesis, or metabolism are up-regulated no earlier than d 5. Hence, we explored the importance of timing of supplementation by comparing effects of RA administration (1 x 10–7 M) starting on d 5 or 6, respectively, compared with the previous experiment, when the cultures were supplemented from d 3 onward. Figure 4Go shows the morphology of the cultures as well as their molecular characterization. We noticed that, in VD3-induced cultures, mineralization of the cells mostly occurred in the center of the EB (Fig. 4AGo, black arrows). This was also the case for cultures treated with RA from d 3 on (data not shown). Once we started to supplement with RA later in differentiation, we saw mineral buildups additionally in the outer ring of the EBs as indicated by the white arrows. Although the time of administration was only postponed about 2 d, the increase in mineralization as measured by calcium content of mature cultures (differentiation d 30) was significant (Fig. 4BGo). Compared with VD3 treatment, calcium content was increased 1.8-fold (RA d 5) and 2-fold (RA d 6), respectively. Osteocalcin expression niveaus went up from a 6-fold induction above spontaneously differentiated controls (RA d 3) to 24-fold (RA d 5) and 29.5-fold (RA d 6) as depicted in Fig. 4CGo. Similarly, BSP expression was induced by later administration of RA.


Figure 4
View larger version (62K):
[in this window]
[in a new window]

  		Fig. 4. Effect of Exogenous RA and BMP-2 Administration

RA is involved in early and late osteogenesis. A, RA given at d 5 and 6 onward, respectively, improves osteogenesis by forming endochondral bone on the outside of the EB (white arrow) in contrast to VD3-induced bone, which is seen in the center of the EB (black arrow). B and C, Calcium content as well as osteocalcin and BSP expression are significantly increased when RA was given together with VD3 from d 5 and 6 onward. n = 3 ± SD. *, P < 0.1; **, P < 0.01; ***, P < 0.001. BMP-2 acts as an osteoinducer in late differentiation. D, Morphology of EBs treated with BMP-2 from the indicated time points on forward. The white arrow indicates mineralization in the center of the EBs that is usually seen after VD3 induction. The black arrows points to mineralization in the outer layer of the EB. E, Calcium content of cultures is measured as a means to evaluate the degree of mineralization and is depicted as percentage compared with VD3 induction. F, Mean expression of osteocalcin and BSP ± SD of independent triplicates is quantified by quantitative PCR and normalized to GAPDH expression. Spontaneously differentiated controls are set as 1. **, P < 0.01; ***, P = 0.001. When medium was supplemented with BMP-2 during late development, osteogenesis was enhanced instead of decreased as was seen with the supplementation at d 3–30 (see Fig. 1Go).

 
BMP-2 Acts as an Osteoinducer in Phase IV of Differentiation
BMP-2, which has been implicated in osteogenesis, was found to be up-regulated in the later stages of VD3-induced osteogenesis (d 25 and 30) compared with nontreated controls. No significant up-regulation was found earlier than d 25. However, as time points between d 16 and 25 were not performed in this analysis, it is not possible to draw conclusions on what is occurring earlier than d 25. Thus, we supplemented EBs with BMP-2 (10 ng/ml) starting from d 17, 19, 21, 23, and 25, onward to d 30. We detected a similar phenomenon as was seen when RA was administered at the appropriate time point. In addition to the mineralization of the center of the EBs (see Fig. 1AGo), matrix-secreting cells were found in the outer ring of the EBs when BMP-2 was given at later stages of the culture (Fig. 4DGo, black arrow). This finding was verified with the results noted in the calcium assay (Fig. 4EGo). Calcium content was increased in cultures that were treated with BMP-2 at later stages as with a maximal increase from d 23–30 (1.9-fold). Expression of osteocalcin and BSP followed the same pattern. Osteocalcin expression was up-regulated 2.2-fold, BSP expression was increased 2.85-fold when BMP-2 administration started on d 23 compared with d 3 (Fig. 4FGo).

Exogenous Administration of RA and BMP-2 according to Their Endogenous Expression Patterns Enhances VD3-Induced Osteogenesis
Osteogenesis was then induced with VD3 plus RA from d 6 and BMP-2 from d 23–30 on Col I-coated plasticware, corresponding to the time points that were identified to increase bone markers maximally. Figure 5AGo shows the increased mineralization of EBs under this treatment compared with VD3 alone and VD3 plus RA or VD3 plus BMP-2, respectively. Again, RA and BMP-2 induced the formation of osteoblasts in the outer ring of cells in the EBs in addition to mineralization in the center. The degree of mineralization in the combined group (VD3 plus RA plus BMP) was significantly improved as the entire EB showed mineralization on d 30 (Fig. 5AGo, lower right corner). Image analysis of black areas in the pictures as a quantification method for mineralization revealed up-regulations of 418-fold for VD3 plus RA, 313-fold for VD3 plus BMP, and 740-fold for the VD3 plus RA plus BMP mixture (VRB). Compared with VD3 alone, the VRB mixture amplified the osteogenic response: osteocalcin expression increased 16-fold (P < 0.001), whereas BSP expression was augmented 5-fold (Fig. 5BGo). The calcium content of the EBs treated with all three supplements in the correct order increased 6.5-fold, whereas ALP activity increased 3.5-fold (Fig. 5Go, C and D). Using flow cytometry and an antibody against the membrane-bound enzyme ALP, the osteoblast yield was quantified over time (Fig. 5EGo). Because ALP is also used in stem cell research as a marker for undifferentiated cells, it is not surprising to find higher levels during early differentiation. On d 5 of differentiation, ALP levels have declined down to under 1%. With beginning osteoinduction at d 7, elevated ALP levels could be found in VD3-treated EBs due to formation of osteoprogenitors, which express ALP. Addition of RA and BMP-2 to the culture medium resulted in an enhancement from 43.65% osteoblasts (VD3) to 90.47% (VRB).


Figure 5
View larger version (58K):
[in this window]
[in a new window]

  		Fig. 5. Enhancement of Osteogenic Yield with RA and BMP-2

Osteogenesis was induced with VD3 plus RA from d 6 and BMP-2 from d 23–30 on Col I-coated plasticware. A, Morphology of EBs over the time course of differentiation under this treatment compared with VD3 alone and VD3 plus RA or VD3 plus BMP-2, respectively. Increased mineralization as represented by black pixels was quantified using image analysis. B–D, Up-regulation of osteogenic markers compared with VD3 alone. Osteocalcin (OCN) expression is increased 15-fold, and BSP expression is up-regulated 5-fold. Calcium content and ALP activity of the EBs treated with all three supplements in the correct order increased 6.5- and 3.5-fold, respectively. Values are means ± SD; n = 3. *, P < 0.1; **, P < 0.01; ***, P < 0.001. E, FACS analysis for ALP comparing control medium (ctrl) to VD3 alone and VD3 plus RA plus BMP-2. ALP levels decreased during early differentiation and increased again in osteoblast containing EBs. Supplementation of VD3 medium with RA and BMP-2 enhances the osteogenic yield from 43.65–90.47% of the cell population.

 
CatnB Levels Fluctuate during Osteogenesis and Are Regulated by VD3, RA, and BMP-2
BMP-2 and RA have previously been found to attenuate CatnB signaling in various cell systems (7, 8). Because our microarray analysis revealed many examples of genes targeting CatnB, we examined endogenous CatnB expression levels over time using RT-PCR. We found evidence for up-regulated CatnB expression in osteoblast cultures induced with VD3 toward the end of the 30-d differentiation period, whereas CatnB expression was up-regulated in spontaneously differentiating controls during early differentiation (Fig. 6Go, A and B). Upon RA treatment at d 6, CatnB levels are decreased, whereas BMP-2 induced CatnB expression at the RNA level (Fig. 6BGo). However, CatnB is heavily regulated at the protein, not the RNA level. Inside the cell, it can either regulate transcription of target genes in the nucleus together with lymphoid enhancer factor (LEF)/T cell factor (TCF) transcription factors or participate in cell adhesion in a complex together with axin. Thus, we examined the cellular localization of this molecule by immunocytochemistry. In control cultures that are spontaneously differentiating, CatnB is colocalized with the nuclear 4',6'-diamidino-2-phenylindole staining. In VD3- and RA-induced cultures, CatnB is localized to the plasma membrane. This suggests that CatnB is rerouted from the nucleus to the plasma membrane by both agents possibly participating in cell adhesion. Under BMP-2 treatment, however, CatnB remains in the nucleus. Furthermore, we obtained an LEF/TCF-green fluorescent protein (GFP) reporter cell line from Dr. I. Weissman (Stanford University, Stanford, CA) and performed quantitative PCR as well as Western blot analysis for GFP expression under the various treatments (Fig. 6Go, D–F). Here, GFP reporter activity indicates activity of TCF/LEF transcription factors, typical cofactors of CatnB-activated transcription. The results clearly indicate that GFP expression is lower than in the controls in VD3- and RA-treated cultures and increased in BMP-2- and LiCl2-treated cultures, which served as a positive control.


Figure 6
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 6. BMP-2 Increases CatnB Levels

A, Endogenous CatnB expression as examined by RT-PCR. B, RA and BMP-2 decrease and increase CatnB RNA expression levels, respectively. C, Localization of CatnB within the cell visualized by immunocytochemistry for CatnB. VD3 and RA reroute CatnB to the plasma membrane. D–F, Indirect measure of nuclear CatnB in d 6 osteoinductions using a D3 ESC line containing a TCF/LEF-GFP reporter gene. D, Quantitative PCR for GFP. GFP expression was normalized to GAPDH and expressed as n-fold induction compared with control cells (spontaneously differentiating). E, Western blot for the GFP protein. F, Green fluorescing cells are detectable in cultures treated with BMP-2 and LiCl2 and to a lesser extent in controls. ctrl, Control; DAPI, 4',6'-diamidino-2-phenylindole; GPAA, ß-glycerophosphate and ascorbic acid.

 
Low nuclear CatnB activity early on in differentiation (phase III) and high nuclear CatnB activity in phase IV thus seem to support the formation of osteoblasts. These findings provide some evidence as to why BMP-2 given together with RA and VD3 for the entire culture period has an opposing effect on osteogenesis. Together, these observations suggest that a biphasic expression of CatnB is important in ESC osteoinduction and that perturbation of CatnB via the ill-timed addition of influencing factors can have a negative effect on differentiation.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
With this study, we demonstrated that gene array analysis using a mixed cell population can indeed lead to the identification of new regulatory genes of a specific lineage. In particular, we succeeded in identifying temporal expression patterns of osteogenic inducers and their downstream targets. Many of the changes between the early stages of osteoblast development and the later stages occurred in the growth factor and signaling molecule categories. A number of transcripts in diverse functional categories were differentially regulated with ESC development. Such steady-state patterns reflect the continuum of differentiation among each stage of development, further validating the usefulness of the ESC model to study the molecular mechanisms governing the progression through specific stages of osteoblast maturation.

CatnB and Skeletal Development
The CatnB molecule acts as a transcription factor at the end of the canonical Wnt signaling cascade. When the Wnt ligand binds to its coreceptors frizzled and LRP5/6, CatnB is stabilized via members of the canonical Wnt signaling cascade. Consequently, cytoplasmic CatnB levels increase, and CatnB accumulates in the nucleus (9), where it forms complexes with the TCF and LEF family of transcription factors (10, 11), triggering downstream gene transcription (12). Several lines of evidence have recently implicated canonical Wnt/CatnB signaling as having a central role in osteoblastogenesis (for review, see Ref. 13). For example, disruption of the Wnt coreceptor LRP5 results in phenotypes similar to patients with osteoporosis-pseudoglioma syndrome, having delayed osteoblast differentiation and reduced osteoblast proliferation (14). Likewise, dermo-Cre mice, which have a targeted deletion of CatnB in mesenchymal precursors of chondrogenesis and osteogenesis, show a reduction in all osteogenesis markers and an absence of bone (but not cartilage) in the developing embryo (15).

Canonical Wnt signaling has also been found to promote osteoblast differentiation from mesenchymal progenitor cells in vitro (16, 17) with both LRP5 and -6 being up-regulated during in vitro osteoblast differentiation (16). In our experiments, we observed that CatnB expression increases in the later stages of ESC osteoinduction. This result is consistent with the in vivo expression of CatnB, which is found to increase in differentiating osteoblasts (15).

The Role of RA in Osteogenesis and Chondrogenesis
We observe that CatnB expression is regulated in a biphasic manner during ESC osteoinduction and suggest that the reduction of CatnB expression is critical for ESC osteoinduction to occur. Indeed, both VD3 and RA have been found to down-regulate CatnB activity in colorectal cancer cells (9). We observed that the addition of VD3 or RA to d 6 EBs had a suppressive effect on both CatnB expression and canonical Wnt signaling. It is thought that VD3 blocks CatnB nuclear activity by redirecting the molecule from the nucleus to the cytoplasm (18), which we have shown is the case during osteogenesis in ESCs, but the ligand-activated vitamin D receptor can also compete with TCF-4 for CatnB binding, thereby blocking transcriptional activation of target genes. This is contradictory to studies described by Phillips et al. (1) in which no effect of RA supplementation was noted when treatment commenced after EB attachment (i.e. in phase III). The VD3 receptor as well as the RA receptor are nuclear receptors that have been postulated to regulate CatnB activity and vice versa. Nuclear receptor ligands, VD3, trans/cis RA, glucocorticoids, and thiazolidines, all induce dramatic changes in the physiology of cells harboring high Wnt/CatnB/Tcf activity (reviewed in Ref. 19). This might mechanistically explain why exposing either VD3 or RA to EBs is necessary and sufficient to induce an osteogenic cascade in ESCs. Consistent with this hypothesis, even dexamethasone, which also has been used an ESC osteoinducer during phase III (3, 20), has recently been found to reduce canonical Wnt signaling in osteoblasts (21, 22). Hence, we suggest that, with all of these ESC osteoinduction methods, a common mechanism may be occurring, which is to reduce canonical Wnt signaling through CatnB in a specific phase of differentiation.

BMPs and Skeletogenesis
BMP-2, which is known to enhance canonical Wnt signaling (7, 23), enhanced CatnB expression and canonical Wnt signaling in d 6 EBs and had a significant negative effect on ESC osteoinduction when added at this early stage. However, we observe that the addition of BMP-2 later, when CatnB levels are rising, can actually enhance ESC osteoinduction. Our results suggest that timing of factor addition may be crucial when optimizing ESC differentiation protocols. It is through our high-fidelity microarray experiments that we discovered BMP-2 and canonical Wnt signaling to be up-regulated in the later stages of ESC osteoinduction. Consistent with our observations, BMP-2 has been found to increase the level of CatnB in the nucleus of preosteoblastic cells and induces the expression of Wnt 15, 3a, 5b, and 7 (24). Phillips et al. (1) also did not detect BMP-2 expression before d 15 in the ESC system. Hence, our observation of up-regulated BMP-2 expression later in ESC osteoinduction is consistent with these observations suggesting that BMP-2 acts later in skeletogenesis. Our data now suggest another regulation layer in osteogenesis and may help to explain the complexity of cellular responses to BMP signaling. This is only part of the picture, because we are investigating several other factors that appear in our array study.

Conclusion
In conclusion, the temporal gene expression profile of VD3-induced differentiation of ESCs is most likely to mirror an attempt of the ESC to lay down an extracellular matrix, as a prerequisite for osteogenesis. The weaving of genetic pathways as identified throughout the course of differentiation helped to identify factors, which increased the efficiency of specific gene expression profiles, as was shown for RA and BMP-2, in a time-dependent manner. Osteogenic yield was highly dependent on the timing of supplementation. Exogenous administration of the identified factors increased osteogenic yield in ESCs from 40–60% to about 90%. The identified effect of RA and BMP-2 were due to regulation of CatnB localization within the cell by both molecules, with a maximal effect evoked by low CatnB nuclear activity during early differentiation (d 6) and elevated CatnB levels during late osteogenesis (d 23). The advantage that the ES model has over other in vitro models is that it encompasses all of the in vivo developmental steps, even those that usually are not accessible in vivo. As such, we could finally show that BMP-2 acts as an inhibitor in early stages of osteogenesis and as an inducer later on.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
ESCs and Their Differentiation
Murine ESCs of the line D3 were maintained as described previously (25). Differentiation was initiated from a cell suspension in hanging drops as shown previously (2). On d 5 of culture, EBs were plated onto adhesive Primaria- or Col I-coated flasks (both Falcon; BD Biosciences, Mountain View, CA). Medium was supplemented with 10 mM ß-glycerophosphate, 50 ng/ml ascorbic acid, and 5x10–8 M 1,25-(OH)2 VD3 for induction of osteoblasts unless stated elsewhere and was changed every second day after d 10. Spontaneously differentiating time-corresponding controls were grown in nonsupplemented control medium.

RNA Isolation, Semiquantitative and Real-Time Quantitative PCR
Total RNA was isolated using the Qiagen RNeasy Midi kit, and yield was determined using the RiboGreen RNA quantitation reagent (Molecular Probes, Eugene, OR). cDNA was synthesized from 625 ng RNA in a total volume of 25 µl with Superscript II as suggested by the manufacturer (Invitrogen, San Diego, CA). One microliter of the first-strand reaction was used for quantitative PCR for bone-specific markers essentially as described previously (2). A two-step quantitative PCR was performed in an iCycler iQ system using a SYBR Green PCR master mix (Bio-Rad, Hercules, CA), with the following cycle conditions: 95 C for the initial 5-min denaturation step and 40 rounds of cycling between 30 sec at 95 C and 45 sec at the respective annealing temperatures. A melting curve was obtained for each PCR product after each run to confirm the presence of a single amplicon. Two microliters of the first-strand reaction was used for PCR validation of microarray data with gene-specific primers, which were generated with primer 3 (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) and blasted for their specificity. Primer sequences were designed for an annealing temperature of 60 C and are listed in Table 1Go. GAPDH was used as a standard, which has been shown to be a nonchanging housekeeping gene during ESC differentiation (26).


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

  		Table 1. Sequences for the Primer Sets Used in RT-PCR for Validation of the Microarray Data Set

 
Microarray Hybridization
Total RNA was sent to the Stem Cell Network Facility (Ottawa, Ontario, Canada) for probe preparation and hybridization to Affymetrix chips. RNA quality was checked with the Agilent 2100 Bioanalyzer system. Double-stranded cDNA was generated from samples only that passed the quality control whereupon a 5' T7 RNA polymerase promoter was affixed to the cDNA during second-strand synthesis. Biotinylated cRNA was derived by T7 transcription in the presence of biotinylated UTP. Resulting biotinylated cRNAs were fragmented and hybridized to individual M430A mouse genomic chips. Signals were detected using a two-step signal amplification protocol and scanning at 488 nm.

Data Analysis
Raw Affymetrix data files that were generated by the Stem Cell Network Facility were examined using Affymetrix Data Mining Tool (DMT) and the GeneSpring software package. Three independent runs were performed for each biological time point. Intensity signals were normalized across chips and a mean was calculated for each biological triplicate. Signals from VD3-treated cultures were compared with untreated controls for the d 6, 8, 16, 25, and 30 time points. Cultures from d 4 and 5 were only treated with control ESC medium and thus compared with each other. A probe list was generated from genes that were showing a 2-fold or higher regulation with a value of P < 0.05 (t test and Mann-Whitney).

Immunocytochemistry
Staining with specific antibodies for the identified genes was performed on EBs cultured for varying durations. Immunostaining was performed with rat antimouse Wnt4 (MAB475; R&D Systems, Minneapolis, MN), rabbit antimouse Sox7 (sc-20093; Santa Cruz Biotechnology, Santa Cruz, CA), a goat antimouse Sox17 (sc-17355; Santa Cruz Biotechnology), and a goat anti-CatnB (sc-1496; Santa Cruz Biotechnology). Respective secondary Alexa Fluor-conjugated antibodies were used to visualize binding of the primary (antirat AF 350, A21093; antirabbit AF546, A11010; antigoat AF488, A11055; all Molecular Probes). Binding of the goat antimouse Wnt5a (sc-23698; Santa Cruz Biotechnology) was confirmed by an horseradish peroxidase-conjugated donkey antigoat secondary antibody and subsequent colorimetric detection with avidin-biotin complex substrate (Vectastain kit; Vector Laboratories, Burlingame, CA).

Ca2+ Determination
Matrix incorporated calcium was quantified using the purple substrate Arsenazo III (DCL, Toronto, Ontario, Canada). Calcium reacts with Arsenazo III at neutral pH to yield a blue-colored complex, whose intensity is proportional to the calcium concentration. Cells were intensively washed with PBS to remove any traces of medium. Values are calculated from a standard, which is measured along with the samples at 650 nm in a Benchmark Plus microplate spectrophotometer (Bio-Rad). Calcium content of EBs was normalized to the total protein content of the samples.

Protein Assay
EBs were rinsed twice in PBS and lysed in radioimmunoprecipitation assay buffer [150 mM NaCl, 10 mM Tris (pH 7.2), 0.1% sodium dodecyl sulfate, 1% Triton X-100, 1% deoxycholate, 5 mM EDTA] containing 1 mM phenylmethylsulfonyl fluoride, 10 mM benzamidine, 2 µg/ml leupeptin, 100 µM sodium orthovanadate, and 10 mM p-nitrophenylphosphate. After gentle rocking for 30 min, lysates were collected and centrifuged, and an aliquot of the supernatant was mixed with DC protein assay reagents from Bio-Rad. After 15-min incubation at room temperature, absorbance was read in a Benchmark Plus microplate spectrophotometer (Bio-Rad) at 750 nm. Protein quantities in samples were taken from a BSA standard curve. ALP enzyme activity was determined as described previously (2).

Image Analysis
The degree of mineralization of the EBs was quantified using Image J 1.33u available from the website of the National Institutes of Health (http://www.rsb.info.nih.gov/ij/). The mean black value per image was calculated from n = 3 images of each treatment group and expressed as n-fold up-regulation above d 6, which was set as 1.

Western Blotting
A TCF/LEF-GFP reporter ESC line (D3) was kindly provided by Dr. Irving Weissmann (Stanford University, Stanford, CA). ESCs were treated with indicated supplements on d 5 of differentiation and whole-cell protein extracts were prepared by lysing the cells after 24 h in radioimmunoprecipitation assay buffer with the addition of appropriate protease inhibitors (P-8340; Sigma-Aldrich, St. Louis, MO). Protein concentration was measured as described above. Fifty micrograms of protein were separated by SDS-PAGE on 12% acrylamide gels and transferred onto nitrocellulose membranes (Protran). Immunoblotting was performed using a GFP antibody (MAB3580; Chemicon, Temecula, CA) and a mouse anti-ß-actin antibody (kind gift from Dr. Steve Robbins, University of Calgary, Calgary, Alberta, Canada) to control for equal protein loading. Immune complexes were visualized by incubation with horseradish peroxidase-conjugated antibody and the ECL detection system (GE Healthcare, Piscataway, NJ).

Statistical Analysis
Significance of quantitative PCR results, Ca2+ determination, and Image analysis was determined with Student’s t test using a web-based program from the Physics Department of the College of Saint Benedict/Saint John’s University (http://www.physics.csbsju.edu/stats/t-test.html).


   ACKNOWLEDGMENTS
 
We thank Dr. Irving Weissman (Stanford University, Stanford, CA) for kindly supplying D3 ESCs carrying the LEF/TCF-GFP reporter and Dr. Steve Robbins (University of Calgary, Calgary, Alberta, Canada) for the ß-actin antibody.


   FOOTNOTES
 
This work was supported by the Canadian Stem Cell Network Centres of Excellence. N.I.z.N. is supported by a postdoctoral fellowship from the Alberta Heritage Foundation of Medical Research (AHFMR). L.A.D. holds a scholarship from the Canadian Stem Cell Network. F.D.P. was supported by a Markin-Flanagan project studentship. D.E.R. is an AHFMR Senior Scholar.

Present address for F.D.P.: Sprott Center for Stem Cell Research, Mail Box 511, 501 Smyth Road, Ottawa, Ontario, Canada K1H 8L6.

The authors have nothing to disclose.

First Published Online December 14, 2006

Abbreviations: ALP, Alkaline phosphatase; BMP-2, bone morphogenetic protein 2; BSP, bone sialoprotein; CatnB, ß-catenin; Col I, collagen type I; EB, embryoid body; ESC, embryonic stem cell; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GFP, green fluorescent protein; LEF, lymphoid enhancer factor; RA, retinoic acid; TCF, T cell factor; VD3, vitamin D3; VRB, VD3 plus RA plus BMP.

Received for publication October 31, 2005. Accepted for publication November 17, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Phillips BW, Belmonte N, Vernochet C, Ailhaud G, Dani C 2001 Compactin enhances osteogenesis in murine embryonic stem cells. Biochem Biophys Res Commun 284:478–484[CrossRef][Medline]
  2. zur Nieden NI, Kempka G, Ahr HJ 2003 In vitro differentiation of embryonic stem cells into mineralized osteoblasts. Differentiation 71:18–27[CrossRef][Medline]
  3. Buttery LDK, Bourne S, Xynos JD, Wood H, Hughes FJ, Hughes SPF, Episkopou V, Polak JM 2001 Differentiation of osteoblasts and in vitro bone formation from murine embryonic stem cells. Tissue Eng 7:89–99[CrossRef][Medline]
  4. Kramer J, Hegert C, Guan K, Wobus AM, Müller PK, Rohwedel J 2000 Embryonic stem cell-derived chondrogenic differentiation in vitro: activation by BMP-2 and BMP-4. Mech Dev 92:193–205[CrossRef][Medline]
  5. zur Nieden NI, Kempka G, Rancourt DE, Ahr HJ 2005 Induction of chondro-, osteo- and adipogenesis in embryonic stem cells by bone morphogenetic protein-2: effect of cofactors on differentiating lineages. BMC Dev Biol 5:1–15[CrossRef][Medline]
  6. Erdo F, Buhrle C, Blunk J, Hoehn M, Xia Y, Fleischmann B, Focking M, Kustermann E, Kolossov E, Hescheler J, Hossmann KA, Trapp T 2003 Host-dependent tumorigenesis of embryonic stem cell transplantation in experimental stroke. J Cereb Blood Flow Metab 23:780–785[CrossRef][Medline]
  7. Mbalaviele G, Sheikh S, Stains JP, Salazar VS, Cheng SL, Chen D, Civitelli R 2005 ß-Catenin and BMP-2 synergize to promote osteoblast differentiation and new bone formation. J Cell Biochem 94:403–418[CrossRef][Medline]
  8. Easwaran V, Pishvaian M, Salimuddin, Byers S 1999 Cross-regulation of ß-catenin-LEF/TCF and retinoid signaling pathways. Curr Biol 9:1415–1418[CrossRef][Medline]
  9. Tolwinski NS, Wieschaus E 2004 A nuclear function for Armadillo/ß-catenin. PLoS Biol 2:486–493
  10. van de Weterning M, Cavallo R, Dooijes D, van Beest M, van Es J, Loureiro J, Ypma A, Hursh D, Jones T, Bejsovec A, Peifer M, Mortin M, Clevers H 1997 Armadillo coactivates transcription driven by the product of the Drosophila segment polarity gene dTCF. Cell 88:789–799[CrossRef][Medline]
  11. Behrens J, von Kries JP, Kuhl M, Bruhn L, Wedlich D, Grosschedl R, Birchmeier W 1996 Functional interaction of ß-catenin with the transcription factor LEF-1. Nature 382:638–642[CrossRef][Medline]
  12. Topol L, Jiang X, Choi H, Garrett-Beal L, Carolan PJ, Yang Y 2003 Wnt-5a inhibits the canonical Wnt pathway by promoting GSK-3-independent ß-catenin degradation. J Cell Biol 162:899–908[Abstract/Free Full Text]
  13. Krishnan V, Bryant HU, Macdougald OA 2006 Regulation of bone mass by Wnt signaling. J Clin Invest 116:1202–1209[CrossRef][Medline]
  14. Kato M, Patel MS, Levasseur R, Lobov I, Chang BH, Glass II DA, Hartmann C, Li L, Hwang TH, Brayton CF, Lang RA, Karsenty G, Chan L 2002 Cbfa1-independent decrease in osteoblast proliferation, osteopenia, and persistent embryonic eye vascularization in mice deficient in Lrp5, a Wnt coreceptor. J Cell Biol 157:303–314[Abstract/Free Full Text]
  15. Day TF, Guo X, Garrett-Beal L, Yang Y 2005 Wnt/ß-catenin signaling in mesenchymal progenitors controls osteoblast and chondrocyte differentiation during vertebrate skeletogenesis. Developmental Cell 8:739–750[CrossRef][Medline]
  16. 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; Osteoporosis-Pseudoglioma Syndrome Collaborative Group 2001 LDL receptor-related protein 5 (LRP5) affects bone accrual and eye development. Cell 107:513–523[CrossRef][Medline]
  17. Bain G, Müller T, Wang X, Papkoff J 2003 Activated ß-catenin induces osteoblast differentiation of C3H10T1/2 cells and participates in BMP-2 mediated signal transduction. Biochem Biophys Res Commun 301:84–91[CrossRef][Medline]
  18. Palmer HG, Gonzalez-Sancho JM, Espada J, Berciano MT, Puig I, Baulida J, Quintanilla M, Cano A, de Herreros AG, Lafarga M, Munoz A 2001 Vitamin D3 promotes the differentiation of colon carcinoma cells by the induction of E-cadherin and the inhibition of ß-catenin signaling. J Cell Biol 154:369–387[Abstract/Free Full Text]
  19. Mulholland DJ, Dedhar S, Coetzee GA, Nelson CC 2005 Interaction of nuclear receptors with the Wnt/ß-catenin/Tcf signaling axis: Wnt you like to know? Endocr Rev 26:898–915[Abstract/Free Full Text]
  20. Sottile V, Thomson A, McWhir J 2003 In vitro osteogenic differentiation of human ES cells. Cloning Stem Cells 5:149–155[CrossRef][Medline]
  21. Ohnaka K, Tanabe M, Kawate H, Nawata H, Takayanagi R 2005 Glucocorticoid suppresses the canonical Wnt signal in cultured human osteoblasts. Biochem Biophys Res Commun 329:177–181[CrossRef][Medline]
  22. Smith E, Frenkel B 2005 Glucocorticoids inhibit the transcriptional activity of LEF/TCF in differentiating osteoblasts in a glycogen synthase kinase-3ß-dependent and -independent manner. J Biol Chem 280:2388–2394[Abstract/Free Full Text]
  23. Yang L, Yamasaki K, Shirakata Y, Dai X, Tokumaru S, Yahata Y, Tohyama M, Hanakawa Y, Sayama K, Hashimoto K 2006 Bone morphogenetic protein-2 modulates Wnt and frizzled expression and enhances the canonical pathway of Wnt signaling in normal keratinocytes. J Dermatol Sci 42:111–119[CrossRef][Medline]
  24. Rawadi G, Vayssiere B, Dunn F, Baron R, Roman-Roman S 2003 BMP-2 controls alkaline phosphatase expression and osteoblast mineralization by a Wnt autocrine loop. J Bone Miner Res 18:1842–1853[CrossRef][Medline]
  25. zur Nieden NI, Ruf LJ, Kempka G, Hildebrand H, Ahr HJ 2001 Molecular markers in embryonic stem cells. Toxicol In Vitro 15:455–461[CrossRef][Medline]
  26. Murphy CL, Polak JM 2002 Differentiating embryonic stem cells: GAPDH, but neither HPRT nor ß-tubulin is suitable as an internal standard for measuring RNA levels. Tissue Eng 8:551–559[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
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 zur Nieden, N. I.
Right arrow Articles by Rancourt, D. E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by zur Nieden, N. I.
Right arrow Articles by Rancourt, D. E.

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