Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2007 May 29;104(23):9673–9678. doi: 10.1073/pnas.0703407104

A high-throughput siRNA library screen identifies osteogenic suppressors in human mesenchymal stem cells

Yuanxiang Zhao 1,*, Sheng Ding 1,†
PMCID: PMC1887565  PMID: 17535907

Abstract

Tissue-specific (or adult) stem/progenitor cells are regarded as the source for normal tissue homeostasis and tissue repair. They also provide tremendous promise for regenerative medicine because of their capacity to proliferate and differentiate into a variety of mature cell types. Human mesenchymal stem cells (hMSCs) can differentiate into osteocytes, adipocytes, chondrocytes, muscle cells, and neurons. However, the molecular mechanisms underlying these differentiation processes are poorly understood. We screened a synthetic siRNA library targeting 5,000 human genes to identify the endogenous repressors of osteogenic specification, which when silenced could initiate differentiation of hMSCs into osteoblasts. This screen yielded 53 candidate suppressors, and 12 of those were further confirmed for their dynamic roles in suppressing osteogenic specification in hMSCs. Furthermore, cAMP was identified to play opposing roles in osteogenesis vs. adipogenesis. This study provides a basis for further elucidation of the genetic network controlling osteogenesis and, potentially, the molecular rationale for treating bone diseases.

Keywords: adipogenic differentiation, osteogenic differentiation, high-throughput RNAi screen

RNAi is a highly conserved gene-silencing mechanism functioning through targeted destruction of individual mRNA by a homologous double-stranded siRNA (1). siRNAs generated by both chemical synthesis and in vitro or in vivo transcription through vector-based expression systems have been proven very useful tools in studying gene loss-of-function in mammalian cells (2–10). Although high-throughput screens using genome-scale siRNA libraries have been successfully carried out in mammalian cells (11–13), effective application of arrayed synthetic siRNA library in stem cells has not been reported. Human mesenchymal stem cells (hMSCs) can be easily isolated from adults and expanded rapidly in vitro. Because of their ability to differentiate into various mature cell types (14–21), hMSCs have been of great interests to researchers exploring cell-based therapies for degenerative diseases including bone disorders (18, 20–23). Cell fate transition from stem cell self-renewal to differentiation involves not only positive regulators but also negative regulators that normally suppress differentiation. Evidence supporting the latter scenario in mesenchymal stem cells includes demonstration of enhanced osteoblast activity and increased bone size in the tob and smurf1 mutant mice as well as ectopic bone formation in humans carrying inactivating mutations in the GNAS gene locus (24, 25). To gain insight into the molecular mechanisms controlling the differentiation of hMSCs into bone cells, we screened an arrayed synthetic siRNA library containing 10,000 unique sequences, with two sequences per gene, to identify the endogenous suppressors of osteogenic specification, which when silenced by the corresponding siRNA could initiate osteogenic differentiation of hMSCs.

Results

High-Throughput siRNA Screen in hMSCs.

To use the large-scale arrayed siRNA library, a reverse transfection protocol was developed by using the lipofection method that provides >90% transfection efficiency and minimum cellular toxicity in hMSCs [supporting information (SI) Fig. 4] (also see Materials and Methods for details). This highly effective siRNA transfection method was then implemented into a high-throughput screen that was based on enzymatic assay of alkaline phosphatase (ALP), an early marker for osteogenic differentiation (26). Fifty-five hits that gave rise to a significant increase of ALP activity on day 7 after siRNA transfection in hMSCs were identified and confirmed (Fig. 1a and SI Table 1). Each image was taken from a representative field of the whole well (and the same applies to all other cell culture images thereafter).

Fig. 1.

Fig. 1.

Open in a new tab

The identification and confirmation of siRNA hits that induced osteogenic differentiation of hMSCs. (a) ALP activity, as indicated by the red staining, was up-regulated in the hit siRNA-transfected or OS-treated cells compared with the control siRNA-transfected cells. (b) Expression of the early (Cbfa1 and Osx but not Dlx5) and late (Bsp) osteogenic markers was differentially induced in different hit siRNA- or OS-treated samples prepared 3 days after transfection or OS treatment. Osx, osterix. (c) ALP activity was reduced in the hit siRNA- and siCbfa1-cotransfected cells compared with the hit siRNA- and siCon-cotransfected cells. (d) Expression knockdown of targeted genes was observed at 36 h after siRNA transfection. Pictures were taken 1 week after transfection.

Among the primary siRNA hits, the corresponding genes encode proteases, kinases, ion channels, protein receptors, ligands, transcription factors, extracellular matrix proteins, hypothetical proteins, etc., some of which are members of the same gene family (integrin family, angiopoietin family, adenylate cyclase family, and olfactory receptor family) (SI Table 1). Although the majority of the identified genes have not been implicated in bone development, two genes, TBX3 (T-box 3) and GNAS, have been found to cause skeletal abnormalities when mutated in mouse and human, respectively (24, 25, 27–29). To verify the screen, we picked 12 targeted genes (SI Fig. 5), including GNAS (human GNAS complex locus, transcript variant 2, isoform b of the alpha subunit of Gs; NM_080426), ADCY8 (adenylate cyclase 8; NM_001115), ADK (adenosine kinase; NM_001123), P2RY11 (purinergic receptor P2R, G protein coupled, 11; NM_002566), TBX3 (T-box 3 or ulnar mammary syndrome; NM_005996), BIRC4 (baculoviral IAP repeat-containing 4; NM_001167), BCL2l2 (BCL2-like 2; NM_004050), SLC12A2 (solute carrier family 12, member 2; NM_001046), KCNT1 (potassium channel, subfamily T, member 1; XM_029962.2), GDBR1 (putative glial blastoma cell differentiation-related; NM_016172), DUSP6 (dual specificity phosphatase 6; NM_001946), and MJD (Machado-Joseph disease or ataxin 3; NM_004993), to further characterize their function in osteogenic differentiation of hMSCs.

Characterizations.

To confirm that the induced ALP activity was derived from the bone-specific isozyme ALPL (30), RT-PCR analysis using the ALPL-specific primers was carried out on hMSC samples collected on day 4 after siRNA transfection. As shown in Fig. 1b, similar to samples treated with the optimal osteogenic inducing medium (OS) (0.05 mM ascorbic acid 2-phosphate/10 mM glycophosphate/0.1 μM dexamethasone in cell culture medium) (17), samples transfected with the hit siRNAs, except for siTBX3 and siMJD, generated increased ALPL transcripts compared with the control sample transfected with nonspecific siRNAs (siCon). To further confirm the osteogenic identity of these hit siRNA-transfected hMSCs, we also examined the expression of additional early (CBFA1, DLX5, Osterix/OSX) and late [bone-specific sialoprotein (BSP)] stage osteogenic markers (31–34). Except for DLX5, which appeared unchanged among all tested samples, CBFA1, OSX, and BSP were differentially up-regulated in the different hit siRNAs and OS-treated samples compared with the siCon-treated sample (Fig. 1b). These results suggested that the induced osteogenic differentiation by particular hit siRNAs may be restricted at different levels or progress at different paces. CBFA1/RUNX2, a master transcription factor required for bone cell fate determination and maturation in mouse (33), is normally expressed in hMSCs (Fig. 1b). To examine whether the hit siRNA induced osteogenic specification requires the function of CBFA1, each hit siRNA was cotransfected with either the CBFA1-specific siRNA or the siCon in hMSCs, and ALP activity was examined. Whereas the cotreatment with the siCon did not cause any noticeable change in ALP activity compared with the single hit siRNA treatment, the cotreatment with the CBFA1 siRNA reduced the level of ALP activity induced by the hit siRNAs or OS (Fig. 1c and data not shown), suggesting that the hit siRNA-induced osteogenic cell fate commitment in hMSCs also requires the function of CBFA1.

To confirm that the induced ALPL expression was not caused by off-target effect from the transfected hit siRNAs, RT-PCR was performed on the corresponding siRNA targeted genes with the RNA samples prepared at 36 h after siRNA transfection. Compared with the control samples, the reduced transcript level of the targeted gene in the corresponding hit siRNA-transfected hMSCs affirmed the specificity (Fig. 1d). In addition, the knockdown of the targeted genes was also observed at 72 h after siRNA transfection (data not shown).

Bone cell maturation is accompanied by extracellular matrix mineralization (35). To examine whether the osteogenic differentiation of hMSCs induced by the hit siRNAs can further proceed to the maturation stage, hit siRNA-treated cells were cultured for 20 days. No extracellular matrix mineralization was observed based on the alizarin red staining that detects calcium phosphate deposits, a major component of mineralized extracellular matrix, indicating that the gene knockdown by the hit siRNAs primarily functions to induce early osteogenic specification of hMSCs and the late-stage osteocyte maturation may require additional factors. Because OS-treated hMSCs undergo osteocyte maturation within 2 weeks, we tested whether sequential treatments by the hit siRNAs (initial 2–4 days) followed by the OS (additional 5–9 days) could facilitate this process. As demonstrated in SI Fig. 6, such sequential treatments of hMSCs with siADK and OS medium clearly enhanced the intensity of alizarin red staining compared with the control treatment. When the OS treatment was started 3 days after siRNA transfection and continued for 9 days, among the 12 hits tested, six (siADK, siGNAS, siP2RY11, siGDBR1, siSLC12a2, and siKCNT1) dramatically enhanced the intensity of alizarin red staining, two (siDUSP6 and siBCL2l2) provided modest enhancement, two (siBIRC4 and siADCY8) provided weak enhancement, and two (siTBX3 and siMJD) had no significant effect (Fig. 2a). These results suggest that the targeted genes of the 10 effective hit siRNAs may play inhibitory roles not only in osteogenic specification of hMSCs but also in bone cell maturation. Combined treatment of two siRNA hits, such as siSLC12A2 with siKCNT1, could further enhance the osteogenic differentiation process (SI Fig. 7 and data not shown). It has been suggested that the osteogenic and adipogenic differentiation of hMSCs are two inverse processes, with one process inhibiting the other (36). Furthermore, in osteoporotic patients, the number of adipocytes was increased in the marrow, implying a change of cell fate disposition from osteoblast lineage to adipocyte lineage (37, 38). We therefore examined the effect of our hit siRNAs on adipogenic differentiation of hMSCs by combined treatment with an adipogenic inducing mixture [100 μg/ml 3-isobutyl-1-methylxanthine (IBMX)/1 μM dexamethasone/10 μg/ml insulin]. Consistent with previous studies, our hit siRNAs inhibited adipogenic differentiation of hMSCs at various degrees, except for siGNAS (Fig. 2b).

Fig. 2.

Fig. 2.

Open in a new tab

The majority of the osteogenic hit siRNAs enhanced bone cell maturation and inhibited the adipogenic differentiation of hMSCs. (a) Ten selected hit siRNAs enhanced bone cell maturation when combined with the OS treatment, as determined by alizarin red staining. (b) Eleven selected hit siRNAs inhibited adipogenic differentiation of hMSCs under adipogenic inducing medium, as determined by Oil Red O staining. 3+9OS, OS treatment was started 3 days after siRNA transfection and continued for 9 days.

GNAS was identified in our screen as an osteogenic cell fate suppressor in hMSCs. The major GNAS gene product, G protein α-subunit (Gsα), couples transmembrane receptors to adenylyl cyclase and is required for the receptor-stimulated intracellular cAMP production. Inactivating mutations in this gene have been found in two types of disorders (progressive osseous hetroplasia and Albright's hereditary osteodystrophy) that are characterized by heterotopic ossification (24, 25, 39), whereas activating mutations in this gene were found in patients with McCune–Albright syndrome which is characterized by fibrous dysplasia of bone (27), suggesting that Gsα is a critical negative regulator of osteogenic commitment. This is consistent with our result as well as that of others that reduced GNAS expression switches on osteogenic cell fate in hMSCs (40). Coincidentally, our screen identified not only Gsα but also several proteins that are closely involved in cAMP production, including ADCY8, ADK, and P2RY11, suggesting a close linkage between cAMP signaling and osteogenic differentiation in hMSCs. Because the targeted genes are all positively involved in intracellular cAMP signaling, we tested the effect of two compounds of opposing function on cAMP production on osteogenic differentiation of hMSCs, ADK inhibitor 5-iodotubercidin and ADCY activator forskolin (41, 42). Treatment with 5-iodotubercidin mimicked the effect of the hit siADK and significantly enhanced the osteogenic differentiation of hMSCs when combined with the OS treatment, whereas treatment with forskolin inhibited this process (Fig. 3a). Furthermore, combined treatment with a cell-permeable cAMP analog such as 8-CPT-cAMP and N6,2′-dibutyryl-cAMP (DB-cAMP), or IBMX (a potent cyclic nucleotide phosphodiesterase inhibitor), with dexamethasone was sufficient to differentiate hMSCs into mature adipocytes (Fig. 3b). Moreover, 8-CPT-cAMP, as well as IBMX or prostaglandin E (PGE) (a cAMP production inducer), could switch hMSC fate specification from osteogenesis to adipogenesis under the OS treatment (Fig. 3c). ATP has been shown as an agonist for P2RY11 receptor (43). Treatment with ATP partially inhibited the OS-induced osteogenesis of hMSCs (SI Fig. 8) and slightly increases the number of cells undergoing adipogenic differentiation in a dosage-dependent manner (data not shown), which is consistent with the notion that P2RY11 acts as an osteogenic suppressor in hMSCs. These results suggested that reducing intracellular cAMP level promotes osteogenic specification but inhibits adipogenesis in hMSCs, and vice versa.

Fig. 3.

Fig. 3.

Open in a new tab

cAMP signaling plays opposing roles in osteogenic vs. adipogenic differentiation of hMSCs. (a) Adenylyl cyclase agonist forskolin inhibited the osteogenic effect of OS on hMSCs in a dose-dependent manner (first column), whereas adenosine kinase inhibitor 5-iodotubercidin enhanced the osteogenic effect of OS (second and third columns) and inhibited the adipogenic differentiation of hMSCs in a dose-dependent manner (fourth column). First and second columns, ALP staining; third column, alizarin red staining; fourth column, Oil Red O staining. (b) Treatment with cAMP analogs (DB-cAMP and CPT-cAMP) or IBMX promoted adipogenic differentiation of hMSCs, especially in the presence of dexamethasone (DEX). (c) CPT-cAMP, IBMX, or PGE2 could switch hMSC fate specification from osteogenesis to adipogenesis under OS treatment. Adipo, adipogenic inducing medium. CPT-cAMP, 0.5 mM; DB-cAMP, 0.5 mM; IBMX, 100 μg/ml; PGE, 100 μM; DEX, 0.1 μM. Pictures were taken after 14 days of treatment.

To find out whether the cAMP response element binding protein (CREB), a common downstream effector of cAMP signaling pathway, is involved in mediating cAMP signaling-controlled differentiation of hMSCs, we examined the expression level of the active form of CREB proteins in cells 3 days after siP2RY11, siADK, siGNAS, siSLC12a2, or siCon transfection. Interestingly, in these hit siRNAs or OS-treated samples, the level of pCREB was increased compared with the siCon-treated or untreated samples (SI Fig. 9a), suggesting that CREB activity might be involved in osteogenic differentiation of hMSCs. Because transcription factors of the CREB/CRE modulator (CREM) family mediate cAMP-dependant gene regulation by binding to CRE sites (44, 45), we further examined the role of CRE in osteogenesis and adipogenesis. A synthetic 24-mer CRE decoy oligonucleotide that can compete with endogenous CRE enhancers for binding proteins was transfected into hMSCs (46), which were then subjected to the treatment of osteogenic or adipogenic induction medium. Compared with the 24-mer CRE mismatch control oligonucleotide, the CRE decoy inhibited cells from undergoing not only adipogenic differentiation but also osteogenic differentiation (SI Fig. 9b), suggesting that CRE-directed gene transcription is involved in both osteogenesis and adipogenesis.

Although the screen revealed multiple players involved in the cAMP signaling pathway that controls osteogenic differentiation of hMSCs, combinatorial siRNA treatment study also suggested that different targeted genes (when knocked down) might have used different mechanisms to control or contribute to the osteogenic process, one of which is to turn on positive regulators of osteogenesis. Several proteins, including BMP-2, PLZF, and TAZ, have been shown to promote osteogenic differentiation in multipotent mesenchymal precursor cells (47–49). To examine their roles in our siRNA-induced osteogenic differentiation, RT-PCR was performed on these genes. Although the expression level of BMP-2 and TAZ was not significantly altered among the different samples, expression of PLZF was detected in the siMJD and OS-treated samples, but not in the rest of the hit siRNA-treated samples (SI Fig. 10a), suggesting that MJD may normally suppress the expression of PLZF in hMSCs. In addition, MAPK signaling pathways have been shown to be activated in the OS-treated hMSCs, and inhibition of these pathways suppresses OS-induced osteogenesis (SI Fig. 10b) (50). DUSP6 is a dual-specificity phosphatase that has been shown to directly dephosphorylate ERK1/2 or p38 kinase (51). We therefore examined the activation status of ERK1/2 and p38 in the siDUSP6-treated hMSCs along with other siRNA-treated samples by Western blot using antibodies against the phosphorylated form of the proteins at 72 h after transfection (SI Fig. 10c). The increased level of the active p38 but not ERK1/2 (data not shown) in the siDUSP6-treated sample suggests that DUSP6 may normally inhibit osteogenic differentiation of hMSCs at least partly by inhibiting the activation of p38 signaling pathway. Our initial probing of the molecular mechanisms underlying the osteogenic induction effect of the identified suppressors suggests that different suppressors may function through discrete pathways to suppress osteogenic differentiation in normal hMSCs.

Discussion

Stem cell fate determination is a balanced act of signaling molecules. Our screen identified at least part of the signaling network involved in controlling osteogenic cell fate specification of hMSCs. Of the identified osteogenic suppressors, there appear to be two distinct types: the cell fate-specific type and the nonspecific type. In the cell fate-specific type, these suppressors only brake on the osteogenic initiation but not the adipogenic differentiation. Rather, the expression of these suppressors is often required for the optimal induction of adipogenic differentiation. Consequently, release of suppression imposed by these osteogenic-specific suppressors compromises the adipogenic differentiation of hMSCs. From a similar screen aiming to identify endogenous adipogenic suppressors in hMSCs, we uncovered a different set of siRNA hits with minimum overlap to those from the osteogenic screen (unpublished data). Consistently, most of the siRNA hits from the adipogenic screen also suppressed the osteogenic differentiation while inducing adipogenic differentiation of hMSCs (unpublished data), indicating the presence of cell fate-specific suppressors in adipogenic differentiation as well. In the cell fate-nonspecific type, however, the suppressors appear to brake on both osteogenic and adipogenic differentiation in hMSCs, because releasing the brake results in enhanced differentiation of hMSCs into both cell lineages. One of such molecules identified in both of our osteogenesis and adipogenesis screens is cartilage oligomeric matrix protein (COMP), which promotes both osteogenic and adipogenic differentiations when knocked down by its siRNA (SI Fig. 11). This suggests a class of suppressors in hMSCs whose role is to maintain the self-renewal of hMSCs and prevent differentiation in general. Defects in COMP are the cause of human pseudoachondroplasia characterized by short stature and early onset osteoarthrosis (52). It is possible that such mutations in COMP may partly function to promote mesenchymal stem/progenitor cells to differentiate during the development in pseudoachondroplasia patients, which could result in decreased adult mesenchymal stem/progenitor cell reservoir required for normal tissue growth, leading to growth retardation.

Several molecules linked to cAMP signaling were identified in our osteogenesis screen. Our study further revealed that the cAMP signaling plays opposing roles in the osteogenic vs. adipogenic differentiation of hMSCs. Interestingly, whereas inhibiting cAMP signaling through siRNA or a compound antagonist induces/promotes osteogenic differentiation of hMSCs, blocking CRE-mediated gene transcription by CRE decoy oligonucleotides inhibits both adipogenic and osteogenic differentiation of hMSCs. This result indicates the complexity of signaling downstream of cAMP involved in hMSC fate determination and that CRE-mediated gene regulation is important not only for adipogenesis but also osteogenesis. A recent study by Chen et al. (53) indicates that PKA signaling via CREB controls myogenesis induced by noncanonical Wnt signaling in mouse embryo. CREB may be an important player involved in various cellular differentiations, potentially through interacting with different modulating cofactors available in different cellular context. It is also noteworthy that, although it has been shown that cAMP signaling pathway promotes osteogenesis in murine cellular systems (54), its suppressive role in hMSC osteogenesis may be species and context specific, similar to the Wnt signaling pathway, which was also demonstrated to have opposite effects on osteogenic differentiation in hMSCs vs. murine 10T1/2 cells (55, 56).

Self-renewal of stem cells can be regarded as a combined phenotypic outcome of cellular proliferation and inhibition of cell differentiation, senescence, and death. It has been shown that proliferative and/or antiapoptotic genes (e.g., AKT, BCL2) can contribute to embryonic stem cell self-renewal (57). Of interest, in our osteogenesis screen, several antiapoptotic genes were found to suppress osteogenic differentiation of hMSCs, including TBX3, BCL2, BCL2L2, and BIRC4. BIRC4/XIAP is known to inhibit apoptosis by binding to cell-death proteases caspases 3, 7, and 9 (58). Although BIRC4 knockout mice appear morphologically normal, it is not clear whether they would have normal development of bone and bone marrow stromal cells. However, consistent with the BIRC4 knockdown phenotype in hMSCs, which may be caused partly by increased caspase 3 activity, mutant mice of caspase 3 have decreased bone density because of attenuated osteogenic capacity of bone marrow stromal cells (59). One possibility would be that, in bone marrow stromal stem cells or hMSCs, certain components of the apoptotic signaling are involved in self-renewal of stem cells and their activation would stop cells from self-renewing and “by default” promote cells to undergo osteogenic differentiation. Further studies are required to clarify the relationship between apoptotic signaling and osteogenic differentiation in hMSCs.

Compared with embryonic stem cells that are associated with concerns of tumor formation and ethical controversy, adult stem/progenitor cells are ideally suited for cell-based therapy and tissue engineering. An additional therapeutic advantage of hMSCs lies in its avoidance of allogeneic rejection demonstrated in both humans and animal models (15). It has been shown that hMSCs offer feasible posttransplantation therapy for children with osteogenesis imperfecta (23). Understanding the molecular mechanisms underlying the differentiation processes of hMSCs into various mature cell types would not only provide insights into the causes of disorders originating in mesenchymal stem/progenitor cells but also help develop safer, more specific and effective therapeutic applications. Our study employing a high-throughput screen of an arrayed synthetic siRNA library provided an example of highly efficient gene identification and functional dissection of the genetic network regulating stem cell fate. Further studies on the identified genes will not only help us understand the critical molecular switches between self-renewal and osteogenic/adipogenic differentiations in hMSCs but also facilitate the characterization of the genetic basis of bone diseases as well as the development of new therapies to treat them.

Materials and Methods

Cell Culture, Transfection, and High-Throughput Screen.

The hMSCs (PT-2501) were purchased from Cambrex (East Rutherford, NJ) and cultured as instructed by the supplier (see www.cambrex.com for detailed characterization and scientific citation of these cells). Cells were expanded to passage 5 before being used for siRNA library screen. Briefly, diluted Xtreme-siRNA transfection reagent (Roche, Indianapolis, IN) (0.12 μl of Xtreme in 14 μl of DMEM) was mixed with prespotted siRNA (14 ng per well) in each well of 384-well plates for 1 h. Cells (4,000 per well) in 60 μl of medium (no antibiotics) were then added by using an automated dispenser (Multidrop 384; Thermo LabSystems, Waltham, MA). Transfection was allowed to continue for 8 h before first medium renewal. After 7 days of culture, cells were stained for ALP activity (catalog no. 86R-1KT; Sigma–Aldrich, St. Louis, MO). Postscreen assays were done in 96-well plates with appropriate scaling. Control siRNA (siCon) used in the screen contains a pool of 50 scrambled siRNA sequences that have at least four mismatches to known human transcripts and ESTs, and at least two to three mismatches to the whole human genome. siTOX was purchased from Dharmacon (Lafayette, CO) (D-001500-01-05).

siCBFA1 is as follows: UAGUAGAGAUAUGGAGUGCtg (antisense); GCACUCCAUAUCUCUACUAtt (sense).

CRE Decoy Transfection.

CRE decoy oligonucleotide and control oligonucleotide (phosphorothioate bond modification on all nucleotides) (46) were transfected into cells similarly as siRNA by using Xtreme-siRNA transfection reagent, except that the final DNA concentration was 90 nM. Cells were then treated with adipogenic or osteogenic inducing medium 24 h after transfection.

CRE decoy 24-mer palindrome is as follows: 5′-TGACGTCATGACGTCATGACGTCA-3′.

CRE mismatch control is as follows: 5′-TGTGGTCATGTGGTCATGTGGTCA-3′.

Calcium Phosphate Staining.

Cells were rinsed with PBS twice, stained in alizarin red solution (catalog no. vw3611-2; VWR, West Chester, PA) for 8 min at room temperature, washed with PBS again, fixed in 10% formalin solution (catalog no. HT-5014; Sigma–Aldrich) for 20 min, and then rinsed with water and air dried.

Oil Red O Staining.

Cells were rinsed with PBS twice, fixed in 10% formalin solution for 20 min, rinsed with PBS twice and distilled water once, washed with propylene glycol for 5 min, and stained with Oil Red O in propylene glycol (I2722A; Newcomer Supply, Middleton, WI) for 30 min. Cells were then washed with 85% propylene glycol for 5 min, rinsed with distilled water three times, and preserved in 50% glycerol.

Compound Treatment.

8-CPT-cAMP, Na (catalog no. 116812), 5-iodotubercidin (catalog no. 407900), PGE2 (catalog no. 538904), SB 202190 (catalog no. 559388), and U0126 (catalog no. 662005) were purchased from Calbiochem (La Jolla, CA); N6,2′-O-dibutyryl-cAMP (catalog no. D0260-5MG), forskolin (catalog no. F6886), ascorbic acid 2-phosphate (catalog no. 49752-10G), β-glycerophosphate (G-6251), dexamethasone (D8893), IBMX (I5879), and insulin (I9278) were purchased from Sigma–Aldrich/Fluka (Buchs, Switzerland).

RNA Preparation and RT-PCR.

Total RNA from each sample was prepared from ≈5 × 104 cells by using RNeasy mini kit from Qiagen (Hilden, Germany) and further treated with Turbo DNA-Free (catalog no. 1907; Ambion, Austin, TX) to prevent DNA contamination. Reverse transcription was carried out by using SuperScript First-Strand Synthesis System for RT-PCR (catalog no. 11904-018; Invitrogen, San Diego, CA) or Qiagen OneStep RT-PCR kit (catalog no. 210210) as instructed. Primer sequences are listed in SI Materials and Methods.

Western Blot.

Cell lysis buffer contains 20 mM Tris (pH 8.0)/1 mM EDTA/150 mM NaCl/0.5% Nonidet P-40 protease inhibitor mixture (catalog no. p8340; 1:100 dilution; Sigma–Aldrich) and phosphatase inhibitor mixture 2 (catalog no. p5726; 1:100 dilution; Sigma–Aldrich). Proteins (60 μg per well) were separated in 12% Novex Tris·glycine gel (catalog no. EC6008 box; Invitrogen) and transferred to nitrocellulose membrane (catalog no. LC2001; Invitrogen) by using the XCell II blot module system (catalog no. EI9051; Invitrogen). The membrane was blocked in 5% milk/PBST (0.05% Tween 20 in PBS) for 1 h at room temperature, incubated overnight at 4°C with primary antibody, washed three times with PBST, followed by 1-h incubation with secondary antibody in 5% milk/PBST and three washes with PBST at room temperature. For antibody against pCREB, however, PBST was substituted with PBS in all steps except for the last one. Antibody-bound proteins were detected by using ECL Western blotting detection reagent (Amersham Biosciences, Piscataway, NJ). Anti-phospho-CREB was purchased from Upstate Biotechnology (Lake Placid, NY) (catalog no. 06-519); anti-phospho-p38 was purchased from Cell Signaling Technology (Beverly, MA) (catalog no. 9910).

Supplementary Material

Supporting Information
pnas_0703407104_index.html (27.3KB, html)

Acknowledgments

We thank Ms. Feng Yan for her excellent technical assistance and Drs. Tony Orth, Suhaila White, Sumit Chanda, Serge Batalov, Pedro Aza-blanc, and Peter Schultz at Genomics Institute of the Novartis Research Foundation (San Diego, CA) for generous support in this work. This is Scripps Research Institute manuscript no. 17986-CH.

Abbreviations

hMSC

human mesenchymal stem cell

ALP

alkaline phosphatase

OS

osteogenic inducing medium

BSP

bone-specific sialoprotein

DB-cAMP

N6,2′-dibutyryl-cAMP

IBMX

3-isobutyl-1-methylxanthine

PGE

prostaglandin E

CREB

cAMP response element binding protein.

Footnotes

The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/cgi/content/full/0703407104/DC1.

References

  • 1.Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC. Nature. 1998;391:806–811. doi: 10.1038/35888. [DOI] [PubMed] [Google Scholar]
  • 2.Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T. Nature. 2001;411:494–498. doi: 10.1038/35078107. [DOI] [PubMed] [Google Scholar]
  • 3.Brummelkamp TR, Bernards R, Agami R. Science. 2002;296:550–553. doi: 10.1126/science.1068999. [DOI] [PubMed] [Google Scholar]
  • 4.Caplen NJ, Parrish S, Imani F, Fire A, Morgan RA. Proc Natl Acad Sci USA. 2001;98:9742–9747. doi: 10.1073/pnas.171251798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Miyagishi M, Taira K. Nat Biotechnol. 2002;20:497–500. doi: 10.1038/nbt0502-497. [DOI] [PubMed] [Google Scholar]
  • 6.Lee NS, Dohjima T, Bauer G, Li H, Li MJ, Ehsani A, Salvaterra P, Rossi J. Nat Biotechnol. 2002;20:500–505. doi: 10.1038/nbt0502-500. [DOI] [PubMed] [Google Scholar]
  • 7.Paul CP, Good PD, Winer I, Engelke DR. Nat Biotechnol. 2002;20:505–508. doi: 10.1038/nbt0502-505. [DOI] [PubMed] [Google Scholar]
  • 8.Yu JY, DeRuiter SL, Turner DL. Proc Natl Acad Sci USA. 2002;99:6047–6052. doi: 10.1073/pnas.092143499. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Sui G, Soohoo C, Affar el B, Gay F, Shi Y, Forrester WC, Shi Y. Proc Natl Acad Sci USA. 2002;99:5515–5520. doi: 10.1073/pnas.082117599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Paddison PJ, Caudy AA, Bernstein E, Hannon GJ, Conklin DS. Genes Dev. 2002;16:948–958. doi: 10.1101/gad.981002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Zheng L, Liu J, Batalov S, Zhou D, Orth A, Ding S, Schultz PG. Proc Natl Acad Sci USA. 2004;101:135–140. doi: 10.1073/pnas.2136685100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Paddison PJ, Silva JM, Conklin DS, Schlabach M, Li M, Aruleba S, Balija V, O'Shaughnessy A, Gnoj L, Scobie K, et al. Nature. 2004;428:427–431. doi: 10.1038/nature02370. [DOI] [PubMed] [Google Scholar]
  • 13.Berns K, Hijmans EM, Mullenders J, Brummelkamp TR, Velds A, Heimerikx M, Kerkhoven RM, Madiredjo M, Nijkamp W, Weigelt B, et al. Nature. 2004;428:431–437. doi: 10.1038/nature02371. [DOI] [PubMed] [Google Scholar]
  • 14.Prockop DJ. Nat Biotechnol. 2002;20:791–792. doi: 10.1038/nbt0802-791. [DOI] [PubMed] [Google Scholar]
  • 15.Ryan JM, Barry FP, Murphy JM, Mahon BP. J Inflamm (London) 2005;2:8. doi: 10.1186/1476-9255-2-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Campagnoli C, Roberts IA, Kumar S, Bennett PR, Bellantuono I, Fisk NM. Blood. 2001;98:2396–2402. doi: 10.1182/blood.v98.8.2396. [DOI] [PubMed] [Google Scholar]
  • 17.Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, Moorman MA, Simonetti DW, Craig S, Marshak DR. Science. 1999;284:143–147. doi: 10.1126/science.284.5411.143. [DOI] [PubMed] [Google Scholar]
  • 18.Pittenger MF, Martin BJ. Circ Res. 2004;95:9–20. doi: 10.1161/01.RES.0000135902.99383.6f. [DOI] [PubMed] [Google Scholar]
  • 19.Woodbury D, Schwarz EJ, Prockop DJ, Black IB. J Neurosci Res. 2000;61:364–370. doi: 10.1002/1097-4547(20000815)61:4<364::AID-JNR2>3.0.CO;2-C. [DOI] [PubMed] [Google Scholar]
  • 20.Dezawa M, Kanno H, Hoshino M, Cho H, Matsumoto N, Itokazu Y, Tajima N, Yamada H, Sawada H, Ishikawa H, et al. J Clin Invest. 2004;113:1701–1710. doi: 10.1172/JCI20935. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Dezawa M, Ishikawa H, Itokazu Y, Yoshihara T, Hoshino M, Takeda S, Ide C, Nabeshima Y. Science. 2005;309:314–317. doi: 10.1126/science.1110364. [DOI] [PubMed] [Google Scholar]
  • 22.Prockop DJ. Science. 1997;276:71–74. doi: 10.1126/science.276.5309.71. [DOI] [PubMed] [Google Scholar]
  • 23.Horwitz EM, Gordon PL, Koo WK, Marx JC, Neel MD, McNall RY, Muul L, Hofmann T. Proc Natl Acad Sci USA. 2002;99:8932–8937. doi: 10.1073/pnas.132252399. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Eddy MC, Jan de Beur SM, Yandow SM, McAlister WH, Shore EM, Kaplan FS, Whyte MP, Levine MA. J Bone Miner Res. 2000;15:2074–2083. doi: 10.1359/jbmr.2000.15.11.2074. [DOI] [PubMed] [Google Scholar]
  • 25.Shore EM, Ahn J, Jan de Beur S, Li M, Xu M, Gardner RJ, Zasloff MA, Whyte MP, Levine MA, Kaplan FS. N Engl J Med. 2002;346:99–106. doi: 10.1056/NEJMoa011262. [DOI] [PubMed] [Google Scholar]
  • 26.Rodan GA, Harada S. Cell. 1997;89:677–680. doi: 10.1016/s0092-8674(00)80249-4. [DOI] [PubMed] [Google Scholar]
  • 27.Weinstein LS, Shenker A, Gejman PV, Merino MJ, Friedman E, Spiegel AM. N Engl J Med. 1991;325:1688–1695. doi: 10.1056/NEJM199112123252403. [DOI] [PubMed] [Google Scholar]
  • 28.Davenport TG, Jerome-Majewska LA, Papaioannou VE. Development (Cambridge, UK) 2003;130:2263–2273. doi: 10.1242/dev.00431. [DOI] [PubMed] [Google Scholar]
  • 29.Bianco P, Riminucci M, Majolagbe A, Kuznetsov SA, Collins MT, Mankani MH, Corsi A, Bone HG, Wientroub S, Spiegel AM, et al. J Bone Miner Res. 2000;15:120–128. doi: 10.1359/jbmr.2000.15.1.120. [DOI] [PubMed] [Google Scholar]
  • 30.Weiss MJ, Henthorn PS, Lafferty MA, Slaughter C, Raducha M, Harris H. Proc Natl Acad Sci USA. 1986;83:7182–7186. doi: 10.1073/pnas.83.19.7182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Nakashima K, Zhou X, Kunkel G, Zhang Z, Deng JM, Behringer RR, de Crombrugghe B. Cell. 2002;108:17–29. doi: 10.1016/s0092-8674(01)00622-5. [DOI] [PubMed] [Google Scholar]
  • 32.Acampora D, Merlo GR, Paleari L, Zerega B, Postiglione MP, Mantero S, Bober E, Barbieri O, Simeone A, Levi G. Development (Cambridge, UK) 1999;126:3795–3809. doi: 10.1242/dev.126.17.3795. [DOI] [PubMed] [Google Scholar]
  • 33.Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G. Cell. 1997;89:747–754. doi: 10.1016/s0092-8674(00)80257-3. [DOI] [PubMed] [Google Scholar]
  • 34.Wang D, Christensen K, Chawla K, Xiao G, Krebsbach PH, Franceschi RT. J Bone Miner Res. 1999;14:893–903. doi: 10.1359/jbmr.1999.14.6.893. [DOI] [PubMed] [Google Scholar]
  • 35.Bilezikian JP, Raisz LR, Rodan GA. Principles of Bone Biology. London: Academic; 2002. [Google Scholar]
  • 36.Beresford JN, Bennett JH, Devlin C, Leboy PS, Owen ME. J Cell Sci. 1992;102:341–351. doi: 10.1242/jcs.102.2.341. [DOI] [PubMed] [Google Scholar]
  • 37.Meunier P, Aaron J, Edouard C, Vignon G. Clin Orthop Relat Res. 1971;80:147–154. doi: 10.1097/00003086-197110000-00021. [DOI] [PubMed] [Google Scholar]
  • 38.Justesen J, Stenderup K, Ebbesen EN, Mosekilde L, Steiniche T, Kassem M. Biogerontology. 2001;2:165–171. doi: 10.1023/a:1011513223894. [DOI] [PubMed] [Google Scholar]
  • 39.Chan I, Hamada T, Hardman C, McGrath JA, Child FJ. Clin Exp Dermatol. 2004;29:77–80. doi: 10.1111/j.1365-2230.2004.01439.x. [DOI] [PubMed] [Google Scholar]
  • 40.Lietman SA, Ding C, Cooke DW, Levine MA. Clin Orthop Relat Res. 2005;434:231–238. doi: 10.1097/01.blo.0000153279.90512.38. [DOI] [PubMed] [Google Scholar]
  • 41.de Souza NJ, Dohadwalla AN, Reden J. Med Res Rev. 1983;3:201–219. doi: 10.1002/med.2610030205. [DOI] [PubMed] [Google Scholar]
  • 42.Cottam HB, Wasson DB, Shih HC, Raychaudhuri A, Di Pasquale G, Carson DA. J Med Chem. 1993;36:3424–3430. doi: 10.1021/jm00074a024. [DOI] [PubMed] [Google Scholar]
  • 43.Communi D, Govaerts C, Parmentier M, Boeynaems JM. J Biol Chem. 1997;272:31969–31973. doi: 10.1074/jbc.272.51.31969. [DOI] [PubMed] [Google Scholar]
  • 44.Montminy MR, Bilezikjian LM. Nature. 1987;328:175–178. doi: 10.1038/328175a0. [DOI] [PubMed] [Google Scholar]
  • 45.Foulkes NS, Borrelli E, Sassone-Corsi P. Cell. 1991;64:739–749. doi: 10.1016/0092-8674(91)90503-q. [DOI] [PubMed] [Google Scholar]
  • 46.Park YG, Nesterova M, Agrawal S, Cho-Chung YS. J Biol Chem. 1999;274:1573–1580. doi: 10.1074/jbc.274.3.1573. [DOI] [PubMed] [Google Scholar]
  • 47.Katagiri T, Yamaguchi A, Komaki M, Abe E, Takahashi N, Ikeda T, Rosen V, Wozney JM, Fujisawa-Sehara A, Suda T. J Cell Biol. 1994;127:1755–1766. doi: 10.1083/jcb.127.6.1755. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Ikeda R, Yoshida K, Tsukahara S, Sakamoto Y, Tanaka H, Furukawa K, Inoue I. J Biol Chem. 2005;280:8523–8530. doi: 10.1074/jbc.M409442200. [DOI] [PubMed] [Google Scholar]
  • 49.Hong JH, Hwang ES, McManus MT, Amsterdam A, Tian Y, Kalmukova R, Mueller E, Benjamin T, Spiegelman BM, Sharp PA, et al. Science. 2005;309:1074–1078. doi: 10.1126/science.1110955. [DOI] [PubMed] [Google Scholar]
  • 50.Jaiswal RK, Jaiswal N, Bruder SP, Mbalaviele G, Marshak DR, Pittenger MF. J Biol Chem. 2000;275:9645–9652. doi: 10.1074/jbc.275.13.9645. [DOI] [PubMed] [Google Scholar]
  • 51.Muda M, Theodosiou A, Gillieron C, Smith A, Chabert C, Camps M, Boschert U, Rodrigues N, Davies K, Ashworth A, et al. J Biol Chem. 1998;273:9323–9329. doi: 10.1074/jbc.273.15.9323. [DOI] [PubMed] [Google Scholar]
  • 52.Briggs MD, Mortier GR, Cole WG, King LM, Golik SS, Bonaventure J, Nuytinck L, De Paepe A, Leroy JG, Biesecker L, et al. Am J Hum Genet. 1998;62:311–319. doi: 10.1086/301713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Chen AE, Ginty DD, Fan CM. Nature. 2005;433:317–322. doi: 10.1038/nature03126. [DOI] [PubMed] [Google Scholar]
  • 54.Tsutsumimoto T, Wakabayashi S, Kinoshita T, Horiuchi H, Takaoka K. Bone. 2002;31:396–401. doi: 10.1016/s8756-3282(02)00839-6. [DOI] [PubMed] [Google Scholar]
  • 55.Rawadi G, Vayssiere B, Dunn F, Baron R, Roman-Roman S. J Bone Miner Res. 2003;18:1842–1853. doi: 10.1359/jbmr.2003.18.10.1842. [DOI] [PubMed] [Google Scholar]
  • 56.Boland GM, Perkins G, Hall DJ, Tuan RS. J Cell Biochem. 2004;93:1210–1230. doi: 10.1002/jcb.20284. [DOI] [PubMed] [Google Scholar]
  • 57.Yamane T, Dylla SJ, Muijtjens M, Weissman IL. Proc Natl Acad Sci USA. 2005;102:3312–3317. doi: 10.1073/pnas.0500167102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Deveraux QL, Takahashi R, Salvesen GS, Reed JC. Nature. 1997;388:300–304. doi: 10.1038/40901. [DOI] [PubMed] [Google Scholar]
  • 59.Miura M, Chen XD, Allen MR, Bi Y, Gronthos S, Seo BM, Lakhani S, Flavell RA, Feng XH, Robey PG, et al. J Clin Invest. 2004;114:1704–1713. doi: 10.1172/JCI20427. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
pnas_0703407104_index.html (27.3KB, html)
pnas_0703407104_1.pdf (353.7KB, pdf)
pnas_0703407104_2.pdf (846.3KB, pdf)
pnas_0703407104_3.pdf (262.6KB, pdf)
pnas_0703407104_4.pdf (354.2KB, pdf)
pnas_0703407104_5.pdf (363KB, pdf)
pnas_0703407104_6.pdf (945KB, pdf)
pnas_0703407104_7.pdf (1.2MB, pdf)
pnas_0703407104_8.pdf (674.9KB, pdf)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES