SEARCH

SEARCH BY CITATION

Keywords:

  • Embryonic stem cells;
  • Induced pluripotent Stem cells;
  • Signal transduction;
  • Bone

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Marfan syndrome (MFS) is a hereditary disease caused by mutations in the gene encoding Fibrillin-1 (FBN1) and characterized by a number of skeletal abnormalities, aortic root dilatation, and sometimes ectopia lentis. Although the molecular pathogenesis of MFS was attributed initially to a structural weakness of the fibrillin-rich microfibrils within the extracellular matrix, more recent results have documented that many of the pathogenic abnormalities in MFS are the result of alterations in TGFβ signaling. Mutations in FBN1 are therefore associated with increased activity and bioavailability of TGF-β1, which is suspected to be the basis for phenotypical similarities of FBN1 mutations in MFS and mutations in the receptors for TGFβ in Marfan syndrome-related diseases. We have previously demonstrated that unique skeletal phenotypes observed in human embryonic stem cells carrying the monogenic FBN1 mutation (MFS cells) are faithfully phenocopied by cells differentiated from induced pluripotent-stem cells (MFSiPS) derived independently from MFS patient fibroblasts. In this study, we aimed to determine further the biochemical features of transducing signaling(s) in MFS stem cells and MFSiPS cells highlighting a crosstalk between TGFβ and BMP signaling. Our results revealed that enhanced activation of TGFβ signaling observed in MFS cells decreased their endogenous BMP signaling. Moreover, exogenous BMP antagonized the enhanced TGFβ signaling in both MFS stem cells and MFSiPS cells therefore, rescuing their ability to undergo osteogenic differentiation. This study advances our understanding of molecular mechanisms underlying the pathogenesis of bone loss/abnormal skeletogenesis in human diseases caused by mutations in FBN1. STEM CELLS 2012;30:2709–2719


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

A heritable disorder of fibrous connective tissue, Marfan syndrome (MFS) shows striking pleitropism and clinical variability [1, 2]. MFS is a life-threatening, autosomal dominant genetic disease affecting approximately 10,000 individuals worldwide [3]. The French physician Antoine Marfan in 1896 first described the skeletal features of this disorder in a 5.5-year-old girl. He referred the condition as dolichostenomelia (long, thin limbs). The cardinal features of MFS occur in three systems: skeletal, cardiovascular, and ocular. The clinical features of skeletal system are increased height, disproportionately long limbs and digits (arachnodactyly). MFS is caused by a mutation in the Fibrillin-1 (FBN1) gene encoding a 350-kD protein which is a major constituent of microfibrils found in many tissues. FBN1 is a multidomain cysteine-rich glycoprotein containing 43 calcium-binding epidermal growth factor-like domains and 78 cysteine-containing transforming growth factor β binding protein-like domain motifs [4] and has been shown to interact with latent TGFβ binding proteins and control TGF-β bioavailability [5–7]. The MFS gene was linked to the chromosome 15q21 region [8, 9], and the gene involved, FBN1, was identified by a candidate gene cloning approach [10, 11]. Several unique and independent missense, nonsense, and deletion mutations of FBN1 gene have been reported [8, 12–16], and about one-quarter of affected individuals arise as new mutations [12, 15].

Selected manifestations of MFS reflect excessive signaling by the TGFβ family of cytokines [17–20]. Enhanced TGFβ signaling is a major contributor to the pathology of MFS. A model has been proposed in which FBN1 mutations perturb the normal microfibril regulation of latent TGFβ and, thereby, contribute to MFS pathogenesis [1, 6]. Furthermore, it has been showed that the increased TGFβ signaling in MFS can be prevented by TGFβ antagonists such as TGFβ neutralizing antibody or the angiotensin II type receptor blocker, losartan [17, 19–21]. This constitutive activation of TGFβ-mediated signaling shares similarity with other syndrome such as, the Loeys-Dietz syndrome, an autosomal dominant aortic-aneurysm syndrome presenting with cardiovascular and skeletal manifestations consistent with those seen in MFS, along with other features not present in MFS. Typical Loeys-Dietz syndrome is characterized by a mutation in either TβRI or TβRII [22–25].

The emerging of human embryonic stem cells (hESCs) and induced pluripotent stem cells (iPSCs) has shed light on a new approach to study both early development and disease pathology. Our recent work on MFS embryonic stem cells and MFS-induced pluripotent stem cells (MFSiPS) cells, derived from two MFS patients, revealed that the enhanced activation of TGFβ signaling present in these cells caused inhibition of osteogenic differentiation while promoting chondrogenesis in a TGFβ cell-autonomous fashion [20]. Moreover, blocking of enhanced TGFβ signaling either with the specific inhibitor SB431542 or TGFβ neutralizing antibody rescued osteogenic differentiation in MFS embryonic stem cells and MFSiPS cells [20].

In this study, we have further investigated the underlying mechanism(s) responsible for the impaired osteogenic differentiation observed previously in MFS embryonic stem cells and MFSiPS cells. Our results indicate that enhanced activation of TGFβ signaling lead to reduced bone morphogenetic protein (BMP) signaling which in turn impaired the osteogenic differentiation in MFS and MFSiPS cells. Moreover, addition of exogenous BMP-2 overcame the inhibitory effect exerted by TGFβ signaling, leading to proper osteogenic differentiation. These data suggest a crosstalk between TGFβ and BMP signaling in MFS and MFSiPS cells. Thus, in sifting through the possible mechanisms and dissecting relevant player pathways in MFS, this study guided us to an interesting example of a mutated gene affecting selectively one of the TGFβ superfamily signaling pathways that can potently impact a signaling pathway mediated by another member of the TGFβ superfamily.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Cell Derivation and Characterization

Embryonic MFS stem cells and MFSiPS lines were established and characterized following protocols that have been described in detail previously [20]. Briefly, MFS hESCs, referred as to MFS cells in this manuscript were derived from a donated in vitro fertilization blastocyst determined by preimplantation genetic diagnosis as having a mutation (c.1747delC) in the 5′ region of the FBN1 gene [20].

As previously described [20], wild-type hESCs (referred as to wild type [WT] cells) were derived from a blastocyst donated for research under informed consent. Both, MFS and WT embryonic stem cells were derived as follows: the zona pellucida was removed by briefly incubating the blastocyst in an acid tyrodes solution. Six days later, the blastocyst was plated on mitotically inactivated mouse embryonic fibroblast feeders (MEFs) and fed daily with 50% W8 50% hESC conditioned media. At day 12, putative trophectoderm cells were scraped away from the initial outgrowth. Three weeks after plating the blastocyst, an outgrowth with hESC morphology was manual dissected onto MEFs.

iPS cells referred as to MFSiPS in this manuscript were derived from MFS fibroblasts harboring a FBN1 splice-site mutation (c.3839-1 g>t) that causes skipping of exon 31 (FB1121) [26]. Wild-type human iPS cells (referred as to WTiPS cells) were derived from normal human fibroblasts as previously described [20]. All cells were maintained on inactivated MEFs in medium consisting of knockout Dulbecco's modified Eagle's medium (DMEM)/F12 supplemented with 20% knockout serum replacer, 2 mM L-glutamine, 0.1 mM nonessential amino acids, 0.1 mM β-mercaptoethanol and 10 ng/ml recombinant human fibroblast growth factor-2 (Invitrogen, Carlsbad, CA, http://www.invitrogen.com). Medium was changed every day. Cells were routinely passaged using 1 mg/ml type IV collagenase (Invitrogen). The results are presented as means ± SD of three independent experiments.

Fluorescence-Activated Cell Sorting

As we previously described [20], in order to identify skeletogenic phenotypes of MFS, cells positive for CD73, one of the mesenchymal cellular markers [27, 28] representative of cells differentiating toward both osteogenic and chondrogenic fates, were isolated. Cells were dissociated by incubation with collagenase IV and then with accutase. Single cells were stained using phycoerythrin-conjugated anti-human CD73 (Becton Dickinson, Franklin Lakes, NJ, http://www.bd.com) and analyzed and sorted with FACS Aria II and FACS Diva software (Becton Dickinson). Cells positive for CD73 were plated in DMEM supplemented with +20% fetal bovine serum (FBS) (Gibco Life Technologies and Invitrogen Corporation) on Matrigel growth factor reduced-coated plates and supplemented with ROCK inhibitor Y0503 (Sigma Aldrich, St. Louis, MO, http://www.sigmaaldrich.com) to promote cell survival. After first passage, cells were passed with accutase onto 0.25% gelatin-coated plates and maintained in DMEM+10% FBS. Three independent fluorescence-activated cell sorting (FACS) products of CD73+ MFS and CD73+ WT cells were analyzed. For MFSiPS and WTiPS cells, three independent FACS products of CD73+ cells derived from two different clones were analyzed. All products gave similar results.

Proliferation Assay

Analysis of cell proliferation was performed by growth curve assay and 5-bromo-2-deoxyuridine (BrdU) incorporation. For the growth curve assay, cells were seeded at a density of 8,000 cells per well in a 12-well plate in DMEM+10% FBS (Gibco Life Technologies and Invitrogen Corporation). The number of cells per well was counted in triplicate using a hemacytometer at days 2, 4, 6, and 8. The data presented are representative of three individual experiments with similar results. For the BrdU assay, cells were seeded at 1,000 per well in a 96-well plate. After 24 hours, cell cultures were washed twice with sterile phosphate buffered saline (PBS) and incubated in fresh DMEM + 1% FBS. BrdU incorporation was carried out for 24 and 48 hours using a kit according to the manufacturer's instructions (Roche Diagnostics, Indianapolis, IN, http://www.roche-applied-science.com). Photometric detection was done with an ELISA reader at 370-nm wavelength. Each time point was run in triplicate. Where required, SB-431542 or neutralizing mouse monoclonal anti-human TGF-β1, TGF-β2, and TGF-β3 antibodies (clone 1D11 #MAB1835, R&D Systems, Minneapolis, MN, http://www.rndsystems.com) were added at concentration of 10 μM and 2 μg/ml, respectively. For proliferating cell nuclear antigen (PCNA) immunodetection, subconfluent cell cultures were harvested, cell lysates were prepared, and protein concentration was quantified by bicinchoninic acid. Fifty microgram of total protein of each sample were analyzed by immunoblotting procedure as described below using rabbit anti-PCNA antibody (FL-261, sc7907, 1:200; Santa Cruz Biotechnology, Santa Cruz, CA, http://www.scbt.com), mouse anti-CYCLIN D1 (DCS-6, 1:500), mouse anti-CYCLIN E (HE12, 1:500) (Cell Signaling, Danvers, MA, http://www.cellsignal.com) and anti-β-ACTIN (ab8227 Rabbit Polyclonal Antibody 1:5,000; Abcam, Cambridge, MA, http://www.abcam.com).

Stimulation with Exogenous BMP-2

Subconfluent cells were starved for 12 hours in DMEM + 1% FBS. Cells were then incubated for the indicated period of time with fresh medium supplemented with 100 ng/ml of recombinant human BMP-2 protein (R&D Systems). After stimulation, the cells were washed three times with cold PBS, lysated with RIPA buffer, and processed for immunoblotting analysis of phosphoSmad1/5.

Induction of Osteogenic Differentiation

For osteogenic differentiation, cells were seeded at high density in six-well plates at 5 × 105 per well in order to let them to reach confluence in 18/20 hours. After 18 hours, cells were incubated in osteogenic medium (day 0). Osteogenic differentiation was induced by culturing cells in DMEM supplemented with 5% FBS, 100 IU/ml penicillin, 100 IU/ml streptomycin, 10 mM β-glycerophosphate, and 100 μg/ml ascorbic acid (Sigma-Aldrich) (osteogenic differentiation medium). Osteogenic differentiation in presence of BMP-2 was performed by adding 100 ng/ml of recombinant human BMP-2 protein (R&D Systems) to the osteogenic medium. Where required, SB-431542 (R&D Systems) was added at concentration of 10 μM. Medium was changed every 2 days.

RNA isolation, Reverse-Transcriptase Polymerase Chain Reaction and Quantitative PCR Analysis

RNA isolation, reverse transcription, and quantitative real time polymerase chain reaction (qPCR) were described previously [29–31]. Briefly, qPCR was performed using the ABI Prism 7900 Sequence Detection System, TaqMan Gene Expression Master Mix, and TaqMan Gene Expression Assays (Applied Biosystems, Foster City, CA, http://www.appliedbiosystems.com). PCNA, CYCLIN D1 (CCND1), and CYCLIN E (CNNE) primers sequences are as follows: PCNA (Forward) 5′-GCCGAGATCTCAGCCATATT-3′, (Reverse) 5′-ATGTACTTAGAGGTACAAAT-3′; CCND1 (Forward) 5′-ACCTGAGGAGCCCCAACAA-3′, (Reverse) 5′-TCTGCTCCTGGCAGGCC-3′; CCNE (Forward) 5′-GTCCTGGCTGAATGTATACATGC-3′, (Reverse) 5′-CCCTATTTTGTTCAGACAACATGGC-3′. Other primers sequences were previously described [20, 32, 33]. Cycling conditions were initial denaturation at 95°C for 3 minutes, followed by 30 cycles consisting of a 15 second denaturation interval at 95°C and a 30 second interval for annealing and extension at 60°C. The relative mRNA level in each sample was normalized to its GAPDH content. Values are given as relative to GAPDH expression. The results are presented as means ± SD of three independent experiments.

Alkaline Phosphatase Enzymatic Activity

Alkaline phosphatase (ALPL) enzymatic activity was performed on cell lysates obtained from cultures at day 10 of differentiation as previously described [34].

Mineralization Assay

Alizarin red staining was performed to evaluate mineralization of extracellular matrix as previously described [34]. Briefly, cells were washed with PBS twice, fixed with 70% ethanol at 4°C for 30 minutes, and then washed with deionized water. Cells were stained with 0.2% Alizarin red for 1 hour at room temperature and then washed with PBS three times. Cell culture plates were air dried and evaluated by light microscopy using an inverted microscope. Images were acquired using a ScanJet 5370C scanner (Hewlett-Packard Company, Palo Alto, CA). Alizarin red staining was quantified by colorimetric assay, incubating stained cells with a 20% methanol 10% acetic acid solution for 30 minutes to elute all calcium-bound stain. Supernatants were collected and optical density was determined at 450 nm. All values were normalized against protein concentration obtained from duplicate wells.

Immunoblotting Analysis

Immunoblotting analysis was performed using the following primary rabbit antibodies: anti-phosphorylated SMAD1/5 (pSMAD1/5Ser463/465), anti-SMAD5, anti-phosphorylated SMAD2 (pSMAD2Ser465/467), anti-SMAD2 (1:1,000; Cell Signaling) and anti-β-ACTIN (1:5,000; ab8227, Abcam). Fifty microgram of cell lysate protein isolated from subconfluent cells cultured for 24 hours in DMEM + 1% FBS were resolved by 12% Tris-HCl sodium dodecyl sulfate -polyacrylamide gel. Proteins were transferred to a polyvinylidene fluoride membrane (Bio-Rad, Inc., Hercules, CA, http://www.bio-rad.com). Membranes were probed with specific antibody. A horseradish peroxidase-conjugated secondary anti-rabbit was used (1:2,000; Cell Signaling). Immunoblotted proteins were visualized by enhanced chemiluminescence (Amersham Biosciences, Buckinghamshire, U.K., http://www.amersham.com). To assess for the total amount of endogenous SMAD5 and to control for equal loading and transfer of the samples, the membranes were stripped and reprobed with anti-SMAD5 and/or anti-SMAD2 antibodies, and anti-β-ACTIN antibody. Densitometry analysis of electrophoretic bands was performed using the ImageJ software program, (NIH, Bethesda, MA). The density of each phosphorylated bands was normalized to the loading controls (β-ACTIN) and presented as percentage increase. The results are the mean ±SD of three independent experiments.

Enzyme-Linked Immunosorbent Assay

Endogenous levels of CNND1 and CNNE proteins were analyzed on MFS, MFSiPS and their relative WT control cell lysates (50 μg) by enzyme-linked immunosorbent assay (ELISA) (Quantikine human BMP-2 and human BMP-4 from R&D Systems (R&D Systems) according to the manufacturer's instruction. Photometric detection was done with an ELISA reader at 370-nm wavelength. Each sample was run in triplicate. The assay was repeated twice.

Statistical Analysis

Data are expressed as mean ± SD of at least three independent samples. Statistical comparisons between groups were performed with a two-tailed Student's t test, *, p ≤ .05; **, p ≤ .01 was considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

MFS and MFSiPS Cells Have Higher Proliferate Activity

TGFβ1 is a multifunctional cytokine with diverse biological effects on many cellular processes, including cell proliferation [35–37]. Our previous study revealed higher threshold levels of active TGFβ1 in MFS and MFSiPS cells compared to WT controls [20]. As first step, in this study we determined the proliferation rate of MFS and corresponding WT control cells. The proliferation activity was initially assessed by a time course growth curve assay. Cells were counted every other day for 10 days. As shown in Figure 1A, both MFS and MFSiPS cells proliferated at significantly higher rate than WT controls. To further confirm the difference in growth rate, we performed BrdU incorporation at different time points. As revealed by the growth curve assay, BrdU incorporation also indicated that MFS cells proliferated more than their corresponding controls (Fig. 1B). Significant differences in BrdU incorporation between MFS and WT controls cells were observed as early as 12 hours (data not shown). Immunoblotting analysis of PCNA showed higher amount of this protein in MFS and MFSiPS cells than their controls (Fig. 1C). Further support for this came from qPCR analysis of cell cycle regulatory markers CNND1 and CNNE (G1/S related). In MFS and MFSiPS cells, mRNA expression of CNND1 and CNNE was increased as well as mRNA expression of other pro-proliferative molecules such as PCNA (Fig. 1D). Moreover, immunoblotting analysis revealed increased CNND1 and CNNE also at protein level in both MFS and MFSiPS cells (supporting information Fig. 1). Collectively, these results demonstrate higher proliferative activity in both MFS and MFSiPS cells.

thumbnail image

Figure 1. MFS and MFSiPS cells proliferate more than WT control cells. (A): Cell proliferation measured by growth curve assay revealed higher proliferative activity in MFS and MFSiPS cells as compared to WT control cells. Cells were cultured in Dulbecco's modified Eagle's medium for the indicated periods. The values are presented as means ± SD of triplicates. (B): Cell proliferation was also assessed by 5-bromo-2-deoxyuridine incorporation assay. The data presented are representative of two individual experiments with similar results. (C): Endogenous levels of PCNA protein as assessed by immunoblotting analysis using anti-PCNA antibody indicated higher amount of protein in MFS and MFSiPS cells compared to their controls. Membranes were stripped and re-probed with β-ACTIN antibody as loading control. Histogram below represents quantification of PCNA protein obtained by Image J program. (D): Gene expression analysis of CNND1, CNNE, and PCNA examined by Quantitative real-time polymerase chain reaction revealed upregulation of these genes in MFS and MFSiPS cells compared to corresponding controls. The relative mRNA level in each sample was normalized to its GAPDH content. Values are given as relative to GAPDH expression. Abbreviations: MFS, Marfan syndrome; MFS-iPS, MFS-induced pluripotent stem cells; PCNA, proliferating cell nuclear antigen; WT, wild type; WTiPS, WT-induced pluripotent stem cells.

Download figure to PowerPoint

Endogenous Activation of TGFβ Promotes Proliferation of MFS and MFSiPS Cells

To directly determine whether the endogenous activated TGFβ signaling previously observed in MFS and MFSiPS cells [20] was responsible for their high proliferative activity, growth curve, BrdU and qPCR assays were performed in the presence of either 10 μM SB431542 or 2 μg/ml anti-TGFβs neutralizing antibody. Treatment with SB431542, a specific inhibitor of TGFβ signaling [38], dramatically reduced proliferation of MFS and MFSiPS cells to levels similar to that of control cells, as indicated by growth curve and BrdU incorporation assay (Fig. 2A–2D). Similarly, TGFβs neutralizing antibody decreased the proliferation of MFS cells, albeit the level of inhibition was not as high as in cells treated with SB431542 (Fig. 2A–2D). As endogenously produced TGFβ may act in an autocrine fashion, SB431542 acting in the cytoplasm may block TGFβ signaling more efficiently than TGFβs antibody. qPCR analysis also confirmed that treatments either with SB431542 or TGFβs neutralizing antibody abolished the higher proliferative activity of MFS and MFSiPS cells, as indicated by downregulation of CNND1, CNNE, and PCNA gene expression (Fig. 2E, 2F). These results indicate that indeed, the sustained endogenous TGFβ signaling directly promoted proliferation of MFS and MFSiPS cells.

thumbnail image

Figure 2. Inhibition of TGFβ signaling decreases proliferation in MFS and MFSiPS cells. (A, B): Growth curve assay of MFS and MFSiPS cells treated either with SB431542 (10 μM) or anti-TGFβs neutralizing antibody (2 μg/ml) showing a decrease in proliferation to levels similar of untreated WT controls. (C, D): 5-Bromo-2-deoxyuridine incorporation assay performed on cells treated as above also revealed a dramatic decrease of proliferation in MFS and MFSiPS. (E, F): Quantitative real-time polymerase chain reaction analysis revealed that the decreased proliferation of treated MFS and MFSiPS cells was paralleled by downregulation of CNND1, CNNE, and PCNA genes. The relative mRNA level in each sample was normalized to its GAPDH content. Values are given as relative to GAPDH expression. Abbreviations: MFS, Marfan syndrome; MFSiPS, MFS-induced pluripotent stem cells; PCNA, proliferating cell nuclear antigen; TGFβ, transforming growth factor β; WT, wild type; WTiPS, WT-induced pluripotent stem cells.

Download figure to PowerPoint

Inhibition of BMP-Mediated Signaling in MFS and MFSiPS Cells and Crosstalk Between TGFβ and BMP Signaling

Thus, the above results demonstrated that MFS and MFSiPS cells proliferate more than their controls due to the presence of higher activation of TGFβ-mediated signaling. Conversely, this enhanced activation of TGFβ-mediated signaling impaired osteogenic differentiation in MFS and MFSiPS cells [20]. Because BMP signaling is a crucial player of osteogenesis [39], we reasoned to investigate the status of this signaling in MFS and MFSiPS cells. When endogenous phosphorylation of BMP-specific R-SMADs in MFS and MFSiPS cells was analyzed by immunoblotting, using phospho-SMAD1/5 specific antibody, we found that SMAD1/5 phosphorylation was strongly reduced in MFS and MFSiPS cells compared to their corresponding controls, suggesting downregulation of BMP signaling (Fig. 3A). To verify whether the inactive BMP signaling observed in MFS and MFSiPS cells could be due to absence and/or low levels of BMP ligands and their receptors compared to WT controls, we analyzed by qPCR the expression of BMP-2 and BMP-4 ligands as well BMPRIA and BMPRIB receptors. This analysis did not reveal significant differences in the expression level either of BMP ligands or their receptors between MFS cells and WT controls (supporting information Fig. 2A). Moreover, ELISA assay detected similar amount of BMP-2 and BMP-4 proteins in both, MFS cells and WT controls (supporting information Fig. 2B). Interestingly, upon SB431542 treatment for 24 hours, SMAD1/5 phosphorylation in MFS and MFSiPS cells achieved levels similar to that observed in untreated WT control cells (Fig. 3A). The effective inhibition of endogenous TGFβ signaling induced upon SB431542 treatment was confirmed by immunoblotting analysis using phospho-SMAD2-specific antibody. As shown in Figure 3B, treatment with 10 μM SB431542 reduced dramatically SMAD2 phosphorylation in MFS cells whereas, increased phospho-SMAD1/5 was observed (Fig. 3A). Next, we sought to investigate the profile of phospho-SMAD1/5 in MFS and MFSiPS cells during the osteogenic differentiation assay. Immunoblotting analysis at different time points revealed downregulation of phospho-SMAD1/5 in both MFS and MFSiPS cells throughout the entire assay, while in WT control cells were observed increased levels of phospho-SMAD1/5 (Fig. 3C, 3D). Treatment with SB431542 induced phospho-SMAD1/5 in MFS and MFSiPS cells, therefore restoring their ability to undergo osteogenic differentiation, as previously demonstrated [20] and showed again herein (Fig. 3E, 3F). These results demonstrate a dramatic downregulation of BMP signaling in MFS cells and suggest the presence of crosstalk between TGFβ and BMP signaling pathways in MFS and MFSiPS cells.

thumbnail image

Figure 3. Bone morphogenetic protein (BMP)-mediated signaling is dramatically inhibited in MFS and MFSiPS cells: cross-talk between TGFβ and BMP signaling. (A): Extent of endogenous SMAD1/5 phosphorylation measured by blotting cell lysates with specific anti-pSMAD1/5 antibody revealed a remarkable inhibition of BMP signaling in MFS and MFSiPS cells as compared to their controls. Treatment with SB431542 (10 μM) for 24 hours induced SMAD1/5 phosphorylation. Membranes were stripped and reprobed with β-ACTIN antibody to control for equal loading. SMAD1/5 were detected by specific antiSMAD1/5 antibody. Histogram below represents quantification of phosphorylated SMAD1/5 protein obtained by Image J program. (B): Immunoblotting analysis of phospho-SMAD2 with specific anti-pSMAD2 antibody to assess the effective inhibition of TGFβ signaling upon SB431542 treatment showed decreased levels of phospho-SMAD2 in treated cells. Membranes were stripped and reprobed with β-ACTIN antibody to control for equal loading. SMAD2 was detected by specific anti-SMAD2 antibody. Histogram below represents quantification of phosphorylated SMAD2 protein obtained by Image J program. (C, D): Immunoblotting analysis of phospho-SMAD1/5 profile during osteogenic differentiation assay showing low levels of phosphorylation in MFS and MFSiPS cells. Treatment with SB431542 stimulated phospho-SMAD1/5 in MFS and MFSiPS cells. Contrary, higher levels of SMAD1/5 phosphorylation were observed during differentiation of untreated WT controls. Membranes were stripped and reprobed with β-ACTIN antibody to control for equal loading. (E, F): Quantification of alizarin red staining at day 21 revealed robust osteogenic differentiation in MFS and MFSiPS cells treated with SB431542. Abbreviations: MFS, Marfan syndrome; MFSiPS, MFS-induced pluripotent stem cells; TGFβ: transforming growth factor β; pSMAD, phosphorylated SMAD; WT, wild type; WTiPS, WT-induced pluripotent stem cells.

Download figure to PowerPoint

Activation of BMP-Mediated Signaling upon Treatment with BMP-2

The observation of BMP signaling inhibition in MFS and MFSiPS cells prompted us to investigate whether treatment with BMP-2 could stimulate phosphorylation of BMP-specific R-SMADs in both MFS and MFSiPS cells therefore, triggering BMP signaling. When MFS and MFSiPS cells were incubated in the presence of 100 ng/ml recombinant human BMP-2 for different period of time, we observed increased phosphorylation of SMAD1/5 (Fig. 4A, 4B). MFS and MFSiPS cells responded effectively, stimulation of SMAD1/5 phosphorylation by BMP-2 was observed as early as after 15 minutes. Phospho-SMAD1/5 levels remained sustained at late time points of stimulation. Furthermore, during the osteogenic differentiation assay, when phosphorylation of SMAD1/5 was examined in MFS and MFSiPS cells, we found that sustained phosphorylated SMAD1/5 was detected up to day 21 of osteogenic differentiation. In contrast, phosphoSMAD1/5 levels in untreated cells were significantly lower (Fig. 4C, 4D).

thumbnail image

Figure 4. Activation of BMP signaling upon BMP-2 stimulation. (A): Time-course stimulation of SMAD1/5 phosphorylation in MFS incubated with BMP-2 (100 ng/ml) for the indicated period of time. Prior stimulation cells were starved for 18 hours in Dulbecco's modified Eagle's medium–1% fetal calf serum. Immunoblotting with pSMAD1/5 antibody showed a strong and sustained phosphorylation of SMAD1/5 in MFS cells. Histogram below represents quantification of phosphorylated SMAD1/5 protein obtained by Image J program. The relative intensity of each band was normalized to respective β-ACTIN loading controls. (B): Time-course stimulation of SMAD1/5 phosphorylation in MFSiPS cells as described in (A). (C): Profile of phospho-SMAD1/5 during osteogenic differentiation of MFS treated with or without BMP-2 (100 ng/ml). Levels of phosphoSMAD1/5 are strongly increased in treated cells. Histogram below represents quantification of phosphorylated SMAD1/5 obtained by Image J program. The relative intensity of each band was normalized to respective β-ACTIN loading controls (D), profile of phospho-SMAD1/5 during osteogenic differentiation of MFSiPS cells also detected higher levels of phosphorylation in BMP-2-treated cells. Histogram below represents quantification of phosphorylated SMAD1/5 obtained by Image J program. Abbreviations: BMP, bone morphogenetic protein; MFS, Marfan syndrome; MFSiPS, MFS-induced pluripotent stem cells; pSMAD, phosphorylated SMAD.

Download figure to PowerPoint

Exogenous BMP-2 Rescues the Osteogenic Differentiation of MFS and MFSiPS Cells

Having assessed an inhibition of BMP-mediated signaling in MFS and MFSiPS cells and the ability of exogenous BMP-2 to overcome it, we investigated whether treatment with BMP-2 protein could rescue the osteogenic differentiation previously observed in both MFS and MFSiPS cells. For this purpose, an osteogenic differentiation assay was performed in the presence of 100 ng/ml BMP-2 protein. The osteogenic profile was analyzed at different time points both, at gene expression and biochemical levels. We also measured the expression levels of mRNA known to be associated with osteogenesis. qPCR analysis revealed upregulation of osteogenic markers such as RUNX-2 (early marker), ALPL (intermediate marker), and osteocalcin (BGLAP) (late marker) in BMP-2-treated MFS and MFSiPS cells compared to untreated cells (Fig. 5A). The ongoing osteogenesis in BMP-2-treated MFS and MFSiPS cells was further confirmed by ALPL enzymatic activity (Fig. 5B) and mineralization of extracellular matrix detected by alizarin red staining, as well (Fig. 5C). Taken together, these results demonstrate that exogenous BMP-2 protein could indeed overcome the impaired osteogenesis in MFS and MFSiPS cells.

thumbnail image

Figure 5. Exogenous BMP-2 rescues osteogenic differentiation in MFS and MFSiPS cells. (A): Effect of BMP-2 on expression of osteogenic marker genes. Quantitative real-time polymerase chain reaction analysis showing that under osteogenic conditions, BMP-2 treatment markedly upregulated RUNX2, ALPL, and osteocalcin (BGLAP) in MFS and MFSiPS cells as compared to untreated cells. The relative mRNA level in each sample was normalized to its GAPDH content. Values are given as relative to GAPDH expression. (B): ALPL enzymatic activity was determined at different time points in cells cultured in ODM in the absence or presence of 100 ng/ml of BMP-2. Addition of BMP-2 increased ALPL activity in MFS and MFSiPS cells. (C): Effect of BMP-2 on extracellular matrix mineralization assessed by Alizarin red staining and its quantification. Cells were maintained in ODM supplemented with or without BMP-2 (100 ng/ml) for 21 days. Prominent matrix mineralization was detected in MFS and MFSiPS cells cultured in ODM with BMP-2 but not in MFS cells cultured without. Lower panel represents quantification of Alizarin red staining. Abbreviations: ALPL, alkaline phosphatase; BMP, bone morphogenetic protein; MFS, Marfan syndrome; MFSiPS, MFS-induced pluripotent stem cells; ODM, osteogenic differentiation medium.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

One important hallmark of MFS and MFSiPS cells is a constitutive activation of endogenous TGFβ signaling, which impairs osteogenesis but leads to chondrogenesis in a TGFβ-cell autonomous manner [20]. This study extends the knowledge of molecular signaling(s) controlling the unique skeletal phenotype observed in MFS embryonic stem cells and MFSiPS cells.

TGFβ are multifunctional growth factors that are capable of inducing a spectrum of cellular events including proliferation [35–37]. Therefore, we sought to investigate whether differences in cell proliferation between MFS and their corresponding WT controls might exist. Proliferation assays clearly indicated that MFS cells proliferated significantly more than controls. This finding was consolidated by PCNA and CYCLIN D and −E gene expression analysis showing their upregulation in MFS and MFSiPS cells. Thus, the MFS cells' proliferative capacity correlated inversely with their osteogenic ability [20]. Collectively, these results are consistent with studies showing that TGFβ can promote proliferation and inhibit osteogenesis of mesenchymal cells [40–42].

The higher proliferative activity and inhibition of osteogenic differentiation elicited by the augmented activation of endogenous TGFβ in MFS and MFSiPS cells could suggest that these cells may be more difficult to differentiate into any cell type, compared to WT controls. However, our previous study ruled out this possibility, as MFS and MFSiPS cells can readily differentiate along the chondrogenic lineage without exogenous TGFβ, whereas WT controls do not [20]. Therefore, although MFS and MFSiPS cells may be more rapidly cycling cells, when cultured in a chondrogenic milieu undergo to chondrogenesis in a TGFβ-cells autonomous fashion, whereas in a osteogenic milieu fail to undergo to osteogenesis [20].

In contrast with what we previously observed in MFS and MFSiPS cells relative to SMAD2 phosphorylation, which was sustained, in this study we observed a poor phosphorylation of SMAD1/5. These results indicated, conversely to what happens for the TGFβ signaling, the occurrence of BMP signaling inactivation in MFS and MFSiPS cells (Fig. 6).

thumbnail image

Figure 6. Schematic proposed model of cross-talk between TGFβ and BMP signaling in MFS embryonic stem cells and MFS-induced pluripotent stem cells (MFSiPS) cells. Continuous enhanced activation of TGFβ signaling impairs osteoblastic differentiation of MFS and MFSiPS cells via inhibition of BMP signaling. Inhibition of endogenous TGFβ signaling by SB431542 treatment and/or treatment with exogenous BMP-2 protein stimulates activation of BMP signaling, therefore rescuing osteogenic differentiation in MFS and MFSiPS cells. Abbreviations: BMP, bone morphogenetic protein; MFS, Marfan syndrome; TGFβ, transforming growth factor.

Download figure to PowerPoint

TGFβ and BMP signaling pathways exhibit antagonistic activities during the development of many tissues. Notably, BMPs act on mesenchymal progenitor cells inducing osteoblastic differentiation through the early commitment phase and the late maturation phase [33, 41, 43, 44]. Inhibition of this key osteogenic signaling pathway might be responsible for the impaired osteogenesis previously observed in MFS and MFSiPS cells [20], and indeed stimulation of BMP signaling might rescue osteogenesis. This hypothesis was validated by analysis of phosphoSMAD 1/5 as well as osteogenic differentiation assays performed in the presence of exogenous BMP-2 protein. Upon BMP-2 treatment, MFS and MFSiPS cells showed increased levels of phosphoSMAD 1/5 and could differentiate into osteoblast lineage likely their WT controls. Therefore, activation of BMP-signaling overcame the osteogenic inhibitory effect triggered by enhanced endogenous TGFβ-signaling in MFS and MFSiPS cells. The latter observation would suggest the occurrence of a crosstalk between TGFβ-signaling and BMP signaling. Our hypothesis is strongly supported by the fact that treatment with SB-431542, a specific inhibitor of TGFβ signaling, rescued phosphorylation of SMAD1/5 in both MFS and MFSiPS cells. Moreover, blockade of TGF-β signaling by SB-431542 exhibited effects similar to those induced by exogenous BMP-2, as shown in the current and previous study [20].

Members of TGFβ superfamily of cytokines play decisive roles in the differentiation of mesenchymal stem cells (MSCs) into osteoblasts. TGFβ promotes recruitment and proliferation of osteoprogenitors and expression of matrix proteins but inhibits late osteoblast differentiation and mineralization [41, 45, 46], while BMPs are potent regulators of osteoblast proliferation and differentiation [39]. A proper balance between these two signaling determines commitment of MSCs to differentiate toward the osteoblast lineage and the efficiency of bone formation. Our study unveiled a severe unbalance between these two signaling in MFS and MFSiPS cells, which was not observed in WT control cells. This unbalance was at the expense of activation of BMP signaling which was dramatically downregulated, therefore preventing MFS cells to differentiate into osteoblasts (Fig. 6).

Increasing number of reports have surfaced over recent years describing both synergistic and antagonistic effects between TGFβ and BMP signaling pathways depending on tissue/developmental context [47–51]. It has been reported that exogenous administration of recombinant human BMP-7 antagonized profibrogenic events that are induced by TGFβ in cultured mesangial cells [47], while another report demonstrated that BMP-7 opposed the TGFβ1-stimulated collagen synthesis in mouse pulmonary myofibroblastic cells by inducing Id2 and Id3 [52]. Keller et al. [49] showed that BMP and TGFβ signaling can have antagonistic effects on chondrocyte proliferation and differentiation in vivo. Collectively, our results also indicate that indeed, these two cognate pathways have antagonistic effects and seem to be intertwined in MFS cells.

It is known that BMP signaling negatively regulates stem cell proliferation in many niches through suppression of Wnt signaling. For instance, He et al. [53], using a conditional BmpR1a mutation in mice, have reported that BMP signaling inhibits intestinal stem cells self-renewal through suppression of Wnt-β-catenin-signaling. Similarly, Kobielak et al. [54] have shown that when the BmpR1a gene is conditionally ablated, quiescent hair follicle stem cells, in the bulge, are activated to proliferate, causing an expansion of the niche and loss of slow-cycling cells. Thus, in both of these cases, BMP signaling appears to suppress Wnt signaling in the context of stem cell niches. Several reports also indicated that noggin, an inhibitor of BMP signaling, maintains pluripotency of human stem cells [55–57]. Conversely, it has been demonstrated that TGFβ1 induces rapid nuclear translocation of β-catenin in adult MSCs in a Smad3-dependent manner leading to their proliferation [54]. Therefore, it would be of interest to investigate whether a similar condition occurs in our MFS cells. In this context, it is worth mentioning that CYCLIN D, one of the canonical Wnt signaling targets, was upregulated in MFS and MFSiPS cells. The latter result is suggestive of a potential active Wnt signaling in MFS and MFSiPS cells, in addition to TGFβ signaling. This possibility is of considerable interest and deserves further studies.

In addition to mutations in the genes encoding the TGF-β receptors type I and II (TGFBR1, TGFBR2) shown to be associated with a number of diseases with significant clinical similarity to MFS [22, 23], other mutations in the genes encoding TGFβ superfamily pathway components are linked to an increasing number of human pathologies [39, 58]. In the context of genetic mutations affecting components of TGFβ superfamily signaling pathway, a pathological condition opposite to MFS is represented by the fibrodysplasia ossificans progressiva (FOP). FOP is a rare disabling autosomal dominant disease characterized by heterotopic ossification for which there is currently no treatment available [59]. Point mutations in the BMP type I receptor ALK2 have been linked as the causative mutation in FOP patients with classic clinical features [60]. FOP patients show enhanced BMP signaling and osteogenic differentiation [61] Thus, FOP represents somehow the counterpart of MFS. In this context, it would be of interest to investigate whether a crosstalk between BMP and TGFβ signaling exists also in this pathology and whether TGFβ treatment could mitigate the enhanced osteogenesis occurring in the connective tissue progenitor cells of FOP patients.

The generation of disease-specific iPS cells from patients with incurable diseases is a promising approach for studying disease mechanisms and drug screening [62, 63]. Such innovation leads to autologous cell sources for use in regenerative medicine.

This study provides evidences that constitutive active TGFβ-mediated signaling presents in MFS cells, as indicated by SMAD2 phosphorylation, and inhibition of osteogenesis [20], correlated with inhibition of BMP-mediated signaling, as indicated by poor phosphorylation of SMAD1/5. Furthermore, exogenous activation of BMP signaling counteracted the osteogenic inhibitory effect elicited by the endogenous enhanced TGFβ signaling on MFS cells. This study represents an interesting example of how a mutated gene induces selective activation of one signaling pathway that can potently impact another signaling pathway.

Our results shed new light on the signaling crosstalk occurring in MFS cells. This study is an important step toward determining the biological potency of clinically relevant, functional cell types derived from MFS embryonic stem cells and MFSiPS cells. These findings could contribute to development of effective novel therapeutic strategies targeting both, TGFβ and BMP pathways for treatment of MFS patients.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

To conclude, we have provided evidence showing that the inhibition of osteogenic differentiation by enhanced and sustained activation of TGFβ signaling present in MFS and MFSiPS cells is mediated via downregulation of endogenous BMP signaling. Furthermore, exogenous stimulation of BMP signaling overcomes the inhibitory effect of active TGF β signaling, rescuing osteogenesis in MFS and MFSiPS cells. Altogether, our results identify the existence of a crosstalk between TGFβ and BMP signaling in a cell model of MFS.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

This work was supported by the National Institutes of Health NIH-U01 HL099776, RC1 HL100490, RC2 DE020771 (MTL) and the California Institute Regenerative Medicine #RL1-00662-1 grant.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSIONS
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
sc-12-0555_sm_SupplFigure1.tif295KSupplemental Figure 1. Endogenous levels of CYCLIN D1 and CYCLIN E proteins in MFS and MFSiPS cells. Immunoblotting analysis using anti-CNND1 and anti-CNNE antibodies show higher amounts of endogenous CYCLIN D1 and CYCLIN E proteins in MFS and MFSiPS cells compared to their controls. Membranes were stripped and reprobed with β.ACTIN antibody as loading control. Histograms below represent quantification of CYCLIN D1 and CYCLIN E proteins obtained by Image J program.
sc-12-0555_sm_SupplFigure2.tif2963KSupplemental Figure 2. A, Expression profile of BMP ligands and their receptors. qPCR analysis of BMP-2, BMP-4 and BMPRIA and BMPRIB receptors showing similar levels of expression between MFS, MFFiPS cells and their corresponding WT controls. The relative mRNA level in each sample was normalized to its GAPDH content. Values are given as relative to GAPDH expression. B, ELISA assay does not detect significant differences of BMP-2 and -4 ligands between MFS, MFSiPS cells and their WT controls.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.