TY - JOUR
AU - Guicheux, Jérôme
AB - Abstract Degenerative disc disease (DDD) primarily affects the central part of the intervertebral disc namely the nucleus pulposus (NP). DDD explains about 40% of low back pain and is characterized by massive cellular alterations that ultimately result in the disappearance of resident NP cells. Thus, repopulating the NP with regenerative cells is a promising therapeutic approach and remains a great challenge. The objectives of this study were to evaluate the potential of growth factor-driven protocols to commit human adipose stromal cells (hASCs) toward NP-like cell phenotype and the involvement of Smad proteins in this differentiation process. Here, we demonstrate that the transforming growth factor-β1 and the growth differentiation factor 5 synergistically drive the nucleopulpogenic differentiation process. The commitment of the hASCs was robust and highly specific as attested by the expression of NP-related genes characteristic of young healthy human NP cells. In addition, the engineered NP-like cells secreted an abundant aggrecan and type II collagen rich extracellular matrix comparable with that of native NP. Furthermore, we demonstrate that these in vitro engineered cells survived, maintained their specialized phenotype and secretory activity after in vivo transplantation in nude mice subcutis. Finally, we provide evidence suggesting that the Smad 2/3 pathway mainly governed the acquisition of the NP cell molecular identity while the Smad1/5/8 pathway controlled the NP cell morphology. This study offers valuable insights for the development of biologically-inspired treatments for DDD by generating adapted and exhaustively characterized autologous regenerative cells. Intervertebral disc, Regenerative medicine, Stem cells, Smad-driven nucleopulpogenic differentiation Significance Statement In the present manuscript, we investigated whether human adipose stromal cells (hASCs) can be a clinically relevant source of stem cells for the generation of phenotypically stable and biologically active NPCyte-like cells. We successfully generated NPCyte-like cells from hASCs using a reproducible, robust and accurate growth factor-based induction protocol. We also demonstrated by experimental organogenesis in nude mice subcutis, that NPCyte-like cells seeded in an instructive hydrogel survived, maintained their specialized phenotype and presented secretory activities in vivo. In addition to highlighting the robustness and reproducibility of our protocol, we also document the temporal role of Smad pathways towards NP commitment by using specific chemical inhibitors. These data strengthen the specificity of the NP cell commitment by demonstrating new insights into the role of Smad proteins. Introduction The nucleus pulposus (NP) is the central part of the intervertebral disc (IVD) and plays a key role in the management of biomechanical overload of the spine. Recent lineage tracing studies in the mouse have shown that NP cells are derived from the embryonic notochord [1, 2]. At birth, two cell types populate the NP: notochord cells (NTCs) and nucleopulpocytes (NPCytes). A natural growth and maturation process occurs during childhood that is marked by the progressive depletion of NTCs. This disappearance of NTCs is concomitant with the appearance of the resident mature NPCytes [3, 4]. The NPCyte phenotype has been elucidated and a comprehensive molecular signature including PAX1, OVOS2, CD24, CA12, ACAN, and COL2A1 markers has been provided to accurately distinguish the human NPCyte phenotype from that of articular chondrocytes [5]. The NPCytes are considered the executive cells of the NP responsible for the production of a highly hydrated extracellular matrix (ECM). This ECM is mainly composed of type II collagen and aggrecan, a proteoglycan with numerous chains of anchored sulfated glycosaminoglycans (sGAG). With aging, NPCytes progressively become less proliferative, more subject to apoptotic cell death and less prompt to produce ECM [6, 7]. This alteration of NPCytes survival and activity is considered one of the primary events that initiate an irreversible degenerative cascade in the NP ultimately leading to the loss of IVD biomechanical properties. This degenerative cascade is marked by the production of inflammatory cytokines by NPCytes that enhances ECM-degrading enzyme activity [8-10]. This degenerative disc disease (DDD) is recognized as being the first cause of low back pain (LBP), which triggers major disabilities worldwide [11]. Current treatments of discogenic LBP are only symptomatic and do not address etiological issues. The recent improvement of our fundamental knowledge on IVD degeneration, however, enables us to develop biologically inspired therapies to address DDD earlier in the degenerative cascade. In this context, the supplementation of the NP with regenerative cells has already been investigated. Clinical studies with intradiscal injections of undifferentiated autologous bone marrow-derived mesenchymal stromal cells (BM-MSCs) have revealed benefits for both pain relief and IVD hydration [12-14]. In vitro cell commitment before transplantation has thereafter been proposed to improve stem cell regenerative potential. Among the relevant strategies, one may reside in the development of a safe growth factor-driven differentiation protocol to produce regenerative NPCyte-like cells. BM-MSCs have been widely contemplated for the regeneration of skeletal tissues [15]. In contrast, only a limited number of studies have examined the use of BM-MSCs to generate NPCyte-like cells [16, 17]. Adipose tissue is an accessible source of multipotent stromal cells that can differentiate into a wide range of cell types [18-20]. Results from two independent studies have suggested that human adipose stromal cells (hASCs) could be more suitable than BM-MSCs to generate NPCyte-like cells [5, 21]. To address the clinical issue of disc degeneration, the development of a reproducible protocol that differentiates hASCs toward cells with a defined NPCyte identity and biological activity remains however a prerequisite. Analysis of transforming growth factor β (TGF-β) receptor II knockout mice has highlighted that this signaling pathway plays a role in the development of IVD [22]. TGF-β1 has also been shown to stimulate the anabolic activities of human NPCytes by up-regulating the expression of COL2A1 and ACAN genes [23]. In addition, TGF-β1 has been shown to trigger the expression of ACAN and COL2A1 by hASCs in vitro upon low oxygen concentration [24, 25]. The growth differentiation factor 5 (GDF5) belongs to the bone morphogenetic protein (BMP) family of the TGF-β superfamily and has been shown to be involved in IVD growth and homeostasis [26, 27]. It has also been demonstrated that GDF5 is expressed in human NP cells (hNP cells) and stimulates their anabolic activities by increasing type II collagen and aggrecan synthesis [28]. In addition, two clinical trials investigating the therapeutic effect of a single intradiscal injection of GDF5 are currently ongoing (https://clinicaltrials.gov, cases NCT01182337 and NCT01124006). Finally, two studies have suggested a potential effect of GDF5 in the differentiation of BM-MSCs toward the NP cell lineage [16, 17]. Considering the respective role of the growth factors TGF-β1 and GDF5 in IVD biology and in human MSC differentiation, we postulate that they could cooperatively support the generation of NPCytes-like cells from hASC. Thus, the first goal of this study was to examine the effects of the combination of TGF-β1 and GDF5 factors on the hASC fate decision and nucleopulpogenic differentiation. Second, we addressed the in vivo biological activity of the engineered NPCytes by assessing their behavior after transplantation in nude mice subcutis. Finally, we aimed to gain new insights into the differentiation mechanisms by deciphering the respective role of Smad2/3-mediated TGF-β1 and Smad1/5/8-mediated GDF5 signaling. Materials and Methods In Vitro Experiments hASC Isolation and Culture hASCs were isolated from lipoaspirates obtained from nine different patients (aged 26–67 years with an average body mass index of 25.7 ± 1.33) who had given their informed consent. Lipoaspirates were treated as described previously [29, 30] and the phenotype of collected cells was characterized at passage 2 by flow cytometry using anti-CD105, -CD73, -CD90, -CD45, and -CD34 antibodies. Cells were positive for CD105, -73, and -90 and negative for CD34 and -45 as previously reported [30]. For all the subsequent experiments, hASCs were used at passage 2. Human NP Cells Isolation Native hNP cells were isolated from four healthy NPs of a 12-years-old patient. Briefly, NP tissue was dissected and enzymatically digested by hyaluronidase (0.05%, H4272, Sigma-Aldrich/Saint-Louis/Missouri/www.sigmaaldrich.com) at 37 °C for 15 minutes, then by trypsin (0.2%, T9935, Sigma-Aldrich) for 30 minutes and by type II collagenase (0.2%, C5138, Sigma-Aldrich) for 1 hour as previously reported [31]. Isolated cells were immediately homogenized in lysis buffer before total RNA isolation. hASC Differentiation For hASC differentiation, 5 × 105 cells were seeded in 15 ml polypropylene tubes, centrifuged at 250g for 5 minutes and placed in a humid chamber at 37 °C with 21% O2 for 48 hours in the presence of amplification medium composed of Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% foetal calf serum (FCS) and 1% Penicillin/Streptomycin (P/S). After 48 hours, hASC pellets were formed and immediately lysed or fixed for reverse transcriptase quantitative polymerase chain reaction (RT-qPCR) and histological analyses (day 0 condition), respectively, or cultivated without FCS upon 5% oxygen concentration during the indicated differentiation times. Distinct experimental conditions were examined. The control group was cultivated in the presence of a basic medium (BM) composed of DMEM supplemented in 1% P/S, sodium L-ascorbate (50 nM, A4034, Sigma-Aldrich), and human insulin-transferrin-sodium selenite (6.25 µg/ml, 6.25 µg/ml, 6.25 ng/ml, I1884-1VL, Sigma-Aldrich). For the BM TGF-β1 condition, the BM was supplemented with recombinant human TGF-β1 (10 ng/ml, 100-21, Peprotech/Rocky Hill/New Jersey/www.peprotech.com), while in the BM GDF5 condition, the BM was supplemented with recombinant human GDF5 (100 ng/ml, 120-01, Peprotech). The BM TGF-β1+GDF5 medium was composed of BM supplemented with both growth factors at the same concentration. The latter conditions were also supplemented with dexamethasone (DEX; 10−8 M, D4902, Sigma-Aldrich) and referred to as chondrogenic medium (CHM) for BM TGF-β1 DEX or nucleopulpogenic medium (NPM) for BM TGF-β1+GDF5 DEX. For the inhibition of Smad2/3 and Smad1/5/8 phosphorylation, hASCs were cultivated in the NPM supplemented with SB505124 or dorsomorphin (5 µM, S4696 and P5499, respectively, Sigma-Aldrich). SB505124 blocks the kinase domains of ALK4, -5, and -7, thereby preventing Smad2/3 phosphorylation and dorsomorphin blocks the kinase domains of ALK1,-2, -3, and -6, thereby preventing Smad1/5/8 phosphorylation. Gene Expression Analyses Ten pellets (5 × 105 cells/pellet) per condition were homogenized in lysis buffer and total RNA was prepared using the Nucleospin RNA isolation kit (Macherey Nagel/Hoerdt/France/www.mn-net.com). Five hundred nanograms of total RNA were retrotranscribed (SuperScript III First-Strand Synthesis System for RT-PCR, Life Technologies/Carlsbad/California/www.lifetechnologies.com). PCR analysis of the target genes was performed on 15 ng of cDNA using TaqMan gene expression assays and on 2 ng of cDNA using custom TaqMan low-density arrays (Applied Biosystems/Foster City/California/www.appliedbiosystems.com). NCBI reference sequences and TaqMan references of the analyzed genes are summarized in the Supporting Information Table 1. Relative quantification was determined using the Livak method (RQ = 2−ΔΔCq) and represented using a base 2 logarithm. For the assessment of the specificity of the chondrogenic and nucleopulpogenic commitment (Fig. 4), the level of gene expression was quantified using the Livak method (RQ=2−ΔΔCq with ΔΔCq = ΔCqCHM or NPM-ΔCqBM) and represented using a base 10 logarithm. Peptidylprolyl Isomerase A (PPIA) gene was used as an internal reference for normalization. Histology and Immunohistochemistry Alcian blue staining was performed on whole pellets by incubating them overnight with alcian blue (0.1%, Sigma-Aldrich) in HCl (0.1 M, Sigma-Aldrich). Pellets were considered as spheres, and their volume was calculated using the following formula: volume = 4/3πr3 with r = radius of pellet. Pellets of cultured hASCs, explants collected from nude mice subcutis or the NP of a 11-year-old patient were fixed with 4% paraformaldehyde in phosphate buffer saline (PBS, Life Technologies). Four micrometer paraffin sections of pellets and explants were prepared and stained with alcian blue and Masson's trichrome, as described previously [32]. For each patient, alcian blue stained surface was quantified by using Fiji software over four areas per pellet for a total surface equivalent of 0.8 mm2. Cellular density was determined by counting counterstained nuclei over two areas per pellet for a total surface equivalent of 0.4 mm2. Immunohistochemical reactions against type II collagen and aggrecan were conducted as previously published [30]. In the immunochemical reactions, antigen retrievals were performed using 0.01 M citrate buffer for 20 minutes at 95 °C for OVOSTATIN 2 and PAX1, 1 mM EDTA pH 9.0 for 20 minutes at 95 °C for CD24, and 0.01 M citrate buffer for 2 hours at room temperature for P-Smad2, P-Smad1/5/8. Sections were then incubated overnight at 4 °C with primary antibodies for OVOSTATIN 2 (1:400 dilution, Ab110548, Abcam/Cambridge/United Kingdom/www.abcam.com), PAX1 (1:300 dilution, NBP1-19767, Novus Biologicals/Littleton/Colorado/www.novusbio.com), P-Smad2 (1:200 dilution, 3101, Cell Signaling Technology/Danvers/Massachusetts/www.cellsignal.com), P-Smad1/5/8 (1:200 dilution, 9511, Cell Signaling Technology), and CD24 (1:20 dilution, LS-C87659, LSbio/Seattle/Washington/www.lsbio.com). Biotinylated goat anti-rabbit and goat anti-mouse secondary antibodies (1:300 dilution, E0432 and E0433 respectively, Dako/Les Ulis/France/www.dako.com) were used. Staining was performed using a diaminobenzidine chromogen. For each differentiation condition, the positive P-Smad2 and P-Smad1/5/8 staining fraction was calculated over two areas per pellet for a total surface equivalent of 0.4 mm2 per patient. Western Blotting Confluent hASCs were placed in a humid chamber at 37 °C with 21% O2 for 12 hours in the presence of DMEM supplemented in 1% P/S and 0.5% fetal bovine serum and then stimulated with TGF-β1 (10 ng/ml) or GDF5 (100 ng/ml) alone or in combination. As positive controls, hASCs were stimulated during 30 minutes with BMP2 (100 ng/ml, PHC7145, Invitrogen/Carlsbad/California/www.thermofisher.com) for P-Smad1/5/8, 2 hours with Activin A (100 ng/ml, 120-14, Peprotech) for P-Smad2, 30 minutes with inorganic phosphate (10 mM) for P-Erk and UV treatment for P-p38 and P-jnk. Following stimulation with either reagent, cells were lysed by adding RIPA buffer (50 mM TrisHCl pH8, 150 mM NaCl, 1% NP40, 0.5% C24H39O4Na, and 0.1% CH3(CH2)11OSO3Na) supplemented with protease and phosphatase inhibitor cocktails (Sigma-Aldrich). Thirty micrograms of total proteins were resolved in PROTEAN TGX polyacrylamide gels and proteins were transferred to a PVDF membrane (Biorad/Hercules/California/www.bio-rad.com). Primary antibodies for P-Smad2, P-Smad1/5/8, P-Erk P-p38, and P-jnk (1:1,000 dilution, 9511, 3101, 9101, 9211, and 9251, respectively, Cell Signaling Technology) were detected using goat anti-rabbit HRP-conjugated secondary antibodies (1:2,000 dilution, 7074, Cell Signaling Technology). Primary antibodies for β-Actin (1:5,000 dilution, A5316, Sigma-Aldrich) were detected using goat anti-mouse HR-conjugated secondary antibodies (1:80,000 dilution, A9917, Sigma-Aldrich). The blots were visualized by enhanced chemiluminescence development using the ChemiDoc MP System (Biorad). Statistical Analyses Data are represented by compiling the results for each patient. Statistical significance was assessed by the non-parametric Kruskal-Wallis test. Significance was determined with a p-value ≤ 0.05. The high number of parameters (n = 3) and the low number of patient (n = 3) prevented us to perform statistical analysis for NPCytes marker expression study (Fig. 2) and for the P-Smad quantification analysis (Fig. 6). In Vivo Experiments Mice In vivo transplantations were performed using female nude mice at 5 weeks of age. The animal experiments were approved by the European Community guidelines for the care, accommodation, and use of laboratory animals (01310.02). Preparation of Si-HMPC Silanized hydroxypropylmethylcellulose (Si-HPMC) powder was synthesized by grafting 14.24% GPTMS onto E4M-Methocel (Dow Chemical Compagny supplied by Colorcon/Dartford/United Kingdom/www.colorcon.com) in heterogeneous process as described previously [33]. Si-HPMC powder (3% w/v) was solubilized in 0.2 M NaOH under constant stirring for 48 hours. To allow the formation of a reticulated hydrogel, the solution was mixed with 0.5 volume of 0.26 M HEPES buffer. The final product was a viscous liquid at pH 7.4, which allowed cell incorporation. Cell/Si-HPMC hydrogel mixture cross-linking occurred within 30 minutes, as described previously [34]. NPCytes-Like Cell Isolation After 21 days of differentiation, hASCs pellets were rinsed with PBS before enzymatic digestion with the crude type 1A collagenase (0.5 mg/ml) for 1.5 hours at 37 °C under gentle agitation. The digest product was filtered through a 70-µm nylon mesh filter and then centrifuged at 250g for 5 minutes. A total of 2 × 106 cells were collected in 20 µl of PBS. Cell/Si-HPMC Hydrogel Mixture Transplantation General anesthesia was obtained in an induction chamber with isoflurane (2%) delivered in O2 and prolonged through an individual mask. Cells were mixed with Si-HMPC at 4 × 106 cells per milliliter and two subcutaneous injections (250 µl) were performed on the lateral back of the mice. A group of five animals was used for the BM condition and two groups of eight animals were used for the CHM and NPM conditions. After 6 weeks of implantation, mice were euthanized by an overdose of CO2. Implants were collected and histology and immunohistochemistry analysis were performed as described above. Results TGF-β1 and GDF5 Combination for the Nucleopulpogenic Differentiation of hASCs To address whether a combination of TGF-β1 and GDF5 favors the nucleopulpogenic differentiation of hASCs, we compared the ability of ASCs isolated from two patients to differentiate in the presence of defined culture media. After 28 days of differentiation in 5% oxygen concentration, the presence of sGAG was analyzed in hASC pellets obtained in distinct media. Basic medium (BM) composed of sodium l-ascorbate and human insulin-transferrin-sodium selenite did not induce sGAG production by hASCs. The results also showed that the culture of hASCs in BM enriched with TGF-β1 or with GDF5 alone or in combination induced a slight sGAG synthesis. DEX is a glucocorticoid that is known to promote the growth factor-induced chondrogenesis of MSCs in vitro [35]. Its potential effect on nucleopulpogenic differentiation of hASCs was analyzed. hASCs cultivated in the BM enriched with either TGF-β1 or GDF5 alone, supplemented or not with DEX, produced a comparable level of sGAG. In contrast, pellets of hASCs cultivated in the presence of the BM containing both TGF-β1 and GDF5 and supplemented with DEX produced the highest amount of sGAG. A strong peripheral sGAG accumulation was observed in these hASC pellets after 28 days of differentiation (Fig. 1A). Figure 1 Open in new tabDownload slide Human adipose stromal cell (hASC) differentiation in various combinations of transforming growth factor-β1 (TGF-β1), growth differentiation factor 5 (GDF5), and dexamethasone (DEX). hASCs derived from patients 1 and 2 were cultivated in the presence of the basic medium enriched with TGF-β1, GDF5, and DEX (one representative patient) for 28 days. (A): Whole pellets were stained with alcian blue. Black arrowheads indicate the peripheral crown of hASC pellets cultivated with TGF-β1, GDF5, and DEX. Scale bar = 1 mm. (B): Pellets were histologically prepared and sectioned for alcian blue staining. Scale bar = 100 µm. hASCs derived from patients 2 and 3 were cultivated in the presence of the BM, the BM TGF-β1 DEX, BM GDF5 DEX, and the BM TGF-β1+GDF5 DEX for 28 days. (C): The surface stained for alcian blue was quantified in hASCs pellets. (D): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was analyzed by reverse transcriptase quantitative polymerase chain reaction. Black crosses on the x-axis represent a Cq value > 38 cycles. Abbreviations: BM, basic medium; DEX, dexamethasone; GAG, glycosaminoglycan; GDF5, growth differentiation factor 5; TGF-β1, transforming growth factor-β1. Figure 1 Open in new tabDownload slide Human adipose stromal cell (hASC) differentiation in various combinations of transforming growth factor-β1 (TGF-β1), growth differentiation factor 5 (GDF5), and dexamethasone (DEX). hASCs derived from patients 1 and 2 were cultivated in the presence of the basic medium enriched with TGF-β1, GDF5, and DEX (one representative patient) for 28 days. (A): Whole pellets were stained with alcian blue. Black arrowheads indicate the peripheral crown of hASC pellets cultivated with TGF-β1, GDF5, and DEX. Scale bar = 1 mm. (B): Pellets were histologically prepared and sectioned for alcian blue staining. Scale bar = 100 µm. hASCs derived from patients 2 and 3 were cultivated in the presence of the BM, the BM TGF-β1 DEX, BM GDF5 DEX, and the BM TGF-β1+GDF5 DEX for 28 days. (C): The surface stained for alcian blue was quantified in hASCs pellets. (D): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was analyzed by reverse transcriptase quantitative polymerase chain reaction. Black crosses on the x-axis represent a Cq value > 38 cycles. Abbreviations: BM, basic medium; DEX, dexamethasone; GAG, glycosaminoglycan; GDF5, growth differentiation factor 5; TGF-β1, transforming growth factor-β1. These observations were confirmed on hASC pellet sections stained with alcian blue (Fig. 1B). The combination of both TGF-β1 and GDF5 in the presence of DEX directed hASC differentiation toward highly sGAG secreting cells. In addition, we noted an obvious change in cell morphology in this condition (Fig. 1B). Cells were large and anchored within a capsule. To decipher the effect of TGF-β1 and GDF5 on hASCs differentiation, cells from two patients were then cultured in four different conditions including BM, BM TGF-β1 DEX, BM GDF5 DEX, and BM TGF-β1+GDF5 DEX. In addition to the quantification of sGAG accumulation, we monitored the expression of specific NPCyte transcripts including CD24, OVOS2, and PAX1 (Fig. 1C, 1D). The quantification of the alcian blue staining in the BM or in the BM GDF5 DEX condition revealed that less than 0.01 mm2 of pellet section was alcian blue positive (Fig. 1C). In addition, low levels of NPCyte-related transcripts were detected in both conditions (Fig. 1D). In the BM TGF-β1 DEX condition, the quantification of sGAG accumulation showed that 0.025 mm2 of pellet section was stained with alcian blue. An increased expression of the two chondrogenic markers COL2A1 and ACAN was observed as previously reported [32]. The expression of CD24 transcript was also induced. In contrast, in the BM TGF-β1+GDF5 DEX condition, the quantification revealed that 0.045 mm2 of pellet section was stained. Moreover, the expression of the five NPCyte-related markers was strongly increased. These results clearly indicated a synergistic effect of the combination of TGF-β1 and GDF5 on hASCs nucleopulpogenic differentiation. According to this first set of experiments, this differentiation medium will hereafter be referred to as nucleopulpogenic medium (NPM) for all the subsequent experiments and compared with the BM enriched with TGF-β1 and DEX referred to as standard CHM. To further distinguish and characterize the nucleopulpogenic differentiation of hASCs, a time course analysis of the differentiation of hASCs was thus performed in the three conditions: BM, CHM and NPM (Fig. 2). The culture of hASCs in the BM did not show sGAG synthesis and the level of NPCyte-related transcripts remained comparable with D0 (Fig. 2A, 2B). In the CHM condition, the differentiating cells produced a substantial level of sGAG from day 14 accompanied with the increased expression of ACAN (at day 7) and COL2A1 (at day 14). Differentiated cells exhibited the previously described cell morphology from day 21. It was also noted that CD24 and OVOS2 transcripts were induced from day 14, whereas PAX1 expression was stable. Interestingly, in the NPM condition, sGAG synthesis was detected as early as day 7 and a high accumulation was observed from day 14. In addition, differentiating cells expressed high levels of ACAN, COL2A1, CD24, and OVOS2 from day 14 and PAX1 from day 21. Together these results demonstrated that the culture of hASCs in the NPM condition induced a cell differentiation process distinguishable from the chondrogenic condition. Figure 2 Open in new tabDownload slide Time course analysis of the chondrogenic and nucleopulpogenic differentiation of human adipose stromal cells (hASCs). hASC pellets from patients 2, 3, and 4 were cultivated in the presence of basic medium, chondrogenic medium, and nucleopulpogenic medium for the times indicated. (A): Pellets were histologically prepared and sectioned for alcian blue staining. Scale bar = 100 µm. (B): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was analyzed using the Taqman low-density array. Abbreviations: BM, basic medium; CHM, chondrogenic medium; NPM, nucleopulpogenic medium; T0, time of pellet formation. Figure 2 Open in new tabDownload slide Time course analysis of the chondrogenic and nucleopulpogenic differentiation of human adipose stromal cells (hASCs). hASC pellets from patients 2, 3, and 4 were cultivated in the presence of basic medium, chondrogenic medium, and nucleopulpogenic medium for the times indicated. (A): Pellets were histologically prepared and sectioned for alcian blue staining. Scale bar = 100 µm. (B): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was analyzed using the Taqman low-density array. Abbreviations: BM, basic medium; CHM, chondrogenic medium; NPM, nucleopulpogenic medium; T0, time of pellet formation. Reproducible and Specific Generation of NP-Like Cells from hASCs To further characterize the robustness of nucleopulpogenic differentiation, we analyzed the reproducibility and the specificity of the nucleopulpogenic differentiation protocol. hASCs from patient 4, 5, 6, and 7 were cultivated for 28 days in the presence of the BM, CHM, and NPM (Fig. 3). We observed that the culture of hASCs in the presence of BM did not induce the expression of ACAN, COL2A1, CD24, OVOS2, and PAX1, while the expression of ACAN and COL2A1 was significantly increased in the CHM condition. A significant increase in CD24 expression was also noted in the CHM condition, as recently reported [36]. Confirming the results obtained with the first patients, hASCs cultivated in the NPM condition expressed significantly higher levels of ACAN, COL2A1, CD24, OVOS2, and PAX1 transcripts even though inter-individual variability was found (Fig. 3A). Interestingly, the level of NPCytes-related markers expression in hASCs cultivated in the NPM condition was found similar to the expression levels measured in native hNP cells isolated from healthy hNPs. hASC pellets cultivated with NPM were 10 times larger than pellets cultivated in the BM (Fig. 3B), while cellular density was lower in the NPM condition compared with the BM (Fig. 3C). High ECM accumulation likely accounts for the increase in pellet volume in the NPM condition. The phenotype of the BM-, CHM-, and NPM-cultivated hASCs was thus further analyzed by histological analyses and immunostainings. The results showed a greater accumulation of sGAG, aggrecan, and type II collagen proteins in NPM pellets compared with CHM pellets. While a moderate level of expression of CD24 transcripts was found in the CHM condition, the corresponding protein was never detected. The CD24, OVOSTATIN 2, and PAX1 proteins were detected exclusively in the hASC pellets cultivated in the NPM (Fig. 3D). Figure 3 Open in new tabDownload slide Reproducibility of the nucleopulpogenic differentiation of human adipose stromal cells (hASCs). hASC pellets from patients 4, 5, 6, and 7 were cultivated in the presence of basic medium (hASC-BM), chondrogenic medium (hASC-CHM), and nucleopulpogenic medium (hASC-NPM) for 28 days. (A): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was analyzed in the three hASC-cultivated conditions and in native human nucleus pulposus cells (represented by a red circle) using the Taqman low-density array. *, p ≤ 0.05 versus BM and **, p ≤ 0.01 versus BM. Relative expression was represented when Cq value < 38 cycles. (B): The volume of the BM-, CHM-, and NPM-derived hASC pellets was measured. **, p ≤ 0.01 versus BM. (C): The cell density was measured in BM-, CHM-, and NPM-derived hASC pellets. *, p ≤ 0.05 versus BM. (D): Pellets were histologically prepared, stained with alcian blue, and immunostained for aggrecan, type II collagen, CD24, OVOSTATIN 2, and PAX1. Scale bar = 100 µm. Abbreviations: BM, basic medium; CHM, chondrogenic medium; hASC, human adipose stromal cell; hNP, human nucleus pulposus; NPM, nucleopulpogenic medium. Figure 3 Open in new tabDownload slide Reproducibility of the nucleopulpogenic differentiation of human adipose stromal cells (hASCs). hASC pellets from patients 4, 5, 6, and 7 were cultivated in the presence of basic medium (hASC-BM), chondrogenic medium (hASC-CHM), and nucleopulpogenic medium (hASC-NPM) for 28 days. (A): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was analyzed in the three hASC-cultivated conditions and in native human nucleus pulposus cells (represented by a red circle) using the Taqman low-density array. *, p ≤ 0.05 versus BM and **, p ≤ 0.01 versus BM. Relative expression was represented when Cq value < 38 cycles. (B): The volume of the BM-, CHM-, and NPM-derived hASC pellets was measured. **, p ≤ 0.01 versus BM. (C): The cell density was measured in BM-, CHM-, and NPM-derived hASC pellets. *, p ≤ 0.05 versus BM. (D): Pellets were histologically prepared, stained with alcian blue, and immunostained for aggrecan, type II collagen, CD24, OVOSTATIN 2, and PAX1. Scale bar = 100 µm. Abbreviations: BM, basic medium; CHM, chondrogenic medium; hASC, human adipose stromal cell; hNP, human nucleus pulposus; NPM, nucleopulpogenic medium. We then confirmed the specificity of the hASC nucleopulpogenic commitment by analyzing the expression of chondrocyte-, osteoblast-, and adipocyte-related markers as well as other markers associated with degenerative and inflammatory processes (Fig. 4). Our results showed that the expression pattern of hASCs cultured in the CHM condition is comparable between patients 4, 5, and 6 with a high level of expression of chondrogenic markers (ACAN, COL2A1, MGP, and COMP) accompanied by a low level of expression of osteoblastic and adipogenic markers. Chondrogenic differentiation was less efficient in the case of patient 7 as shown by the low levels of COL2A1 and MGP transcripts. Pellets of hASCs cultured in the NPM condition expressed high levels of the NPCyte-related transcripts (ACAN, COL2A1, CD24, OVOS2, and PAX1) and low levels of chondrogenic and adipogenic related markers. We noted a relatively high expression of the osteoblastic marker ALPL in three out of four patients accompanied by a similar expression of OCN and RUNX2 compared with their expression in the CHM condition. Except for the ALPL transcript, the expression profile of these differentiated cells was found consistent with the expression profile of native hNP cells (Supporting Information Fig. S1). Figure 4 Open in new tabDownload slide Specificity of the nucleopulpogenic differentiation of human adipose stromal cells (hASCs). hASC pellets from four patients were cultivated in the presence of chondrogenic medium or nucleopulpogenic medium for 28 days [(A) patient 4, (B) patient 5, (C) patient 6, and (D) patient 7]. The expression of nucleopulpogenic, chondrogenic, osteogenic, adipogenic, and various markers were analyzed using the Taqman low-density array. Abbreviations: AD, adipogenic; CH, chondrogenic; CHM, chondrogenic medium; NP, nucleopulpogenic; NPM, nucleopulpogenic medium; OB, osteogenic. Figure 4 Open in new tabDownload slide Specificity of the nucleopulpogenic differentiation of human adipose stromal cells (hASCs). hASC pellets from four patients were cultivated in the presence of chondrogenic medium or nucleopulpogenic medium for 28 days [(A) patient 4, (B) patient 5, (C) patient 6, and (D) patient 7]. The expression of nucleopulpogenic, chondrogenic, osteogenic, adipogenic, and various markers were analyzed using the Taqman low-density array. Abbreviations: AD, adipogenic; CH, chondrogenic; CHM, chondrogenic medium; NP, nucleopulpogenic; NPM, nucleopulpogenic medium; OB, osteogenic. Taking into account that GDF5 may also act as a chondrogenic factor on mesenchymal stromal cells [37], we examined the expression levels of the two chondrocyte-related markers, MGP and COMP [38, 39]. The expression of COL1A1, which is associated with pathological fibrosis was also analysed [40]. The results showed that the culture of hASCs with the NPM did not induce a higher level of MGP, COMP, and COL1A1 expression compared with the CHM condition (Supporting Information Fig. S2A, S2B). Finally, we analyzed the expression of MMP13 gene involved in matrix degradation during NP degeneration and osteoarthritis [41, 42]. As described in the literature, MMP13 transcript was detected in hASCs differentiated in CHM, contrasting with the lower level of transcript found in hASCs differentiated in NPM (Supporting Information Fig. S2C). In conclusion, we observed that hASCs isolated from all patients cultured in the NPM condition were efficiently differentiated toward the NPCyte lineage. Despite the robustness of the differentiation, inter-individual variability in the biological response exists, as illustrated by the over all expression pattern observed for patient 7 (Fig. 4D) and the COL1A1 expression observed for patient 6 (Supporting Information Fig. S2C). In Vivo Assessment of NPCyte-Like Cells Biological Behavior To address whether NPCytes-like cells can survive, maintain their specialized phenotype, and ECM secretory activity in vivo, we performed cell transplantation experiments in nude mice subcutis. Based on the aforementioned in vitro experiments, we aimed to transplant cells cultivated for 21 days in BM, CHM, or NPM in association with a silanized hydroxylpropylmethylcellulose (si-HPMC) hydrogel as a scaffolding biomaterial. Previous studies demonstrated that this biomaterial efficiently carried hASCs in nude mice subcutis, in rabbit cartilage defects, and in rat infarcted myocardium [25, 43]. Before cell transplantation, the expression of chondrogenic and nucleopulpogenic markers was assessed to ensure the commitment of hASCs toward specific lineages (Fig. 5A). After 6 weeks of implantation, explants from hASCs cultured in the BM failed to exhibit alcian blue or Masson's trichrome staining demonstrating the lack of sGAG and collagen-enriched matrix formation (Fig. 5B). Further analyses of the biological behavior of transplanted cells will not be shown for the BM condition. Preconditioning of hASCs in CHM and NPM led to the survival of differentiated cells 6 weeks after transplantation and the secretion of sGAG and collagen containing ECM (Fig. 5B). Remarkably, we noted diffused alcian blue staining surrounding cells preconditioned in NPM, as found in the native hNP. These data indicated that cells were able to secrete sGAG through si-HPMC hydrogel. Type II collagen was found to be secreted by hASCs preconditioned in both CHM and NPM, whereas OVOSTATIN 2 and PAX1 proteins were distinctly produced by cells preconditioned in the NPM (Fig. 5C) consistent with the native hNP cells abilities. Figure 5 Open in new tabDownload slide In vivo biological behavior of human adipose stromal cell (hASC)-derived nucleopulpocytes. hASC pellets from patients 2, 3, and 8 were cultivated in the presence of basic medium, chondrogenic medium, and nucleopulpogenic medium for 21 days before their association with the si-HPMC hydrogel and their transplantation in nude mice subcutis for 6 weeks. (A): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was assessed before transplantation by reverse transcriptase quantitative polymerase chain reaction. Black crosses on the x-axis represent a Cq value > 38 cycles. (B): Explanted samples and a human nucleus pulposus (hNP) were histologically prepared and sectioned for alcian blue and Masson's trichrome staining. (C): Explanted samples and a hNP were histologically prepared and sectioned for type II collagen, OVOSTATIN 2, and PAX1 immunostainings. Scale bar = 10 µm. Abbreviations: BM, basic medium; CHM, chondrogenic medium; NPM, nucleopulpogenic medium; hNP, human nucleus pulposus. Figure 5 Open in new tabDownload slide In vivo biological behavior of human adipose stromal cell (hASC)-derived nucleopulpocytes. hASC pellets from patients 2, 3, and 8 were cultivated in the presence of basic medium, chondrogenic medium, and nucleopulpogenic medium for 21 days before their association with the si-HPMC hydrogel and their transplantation in nude mice subcutis for 6 weeks. (A): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was assessed before transplantation by reverse transcriptase quantitative polymerase chain reaction. Black crosses on the x-axis represent a Cq value > 38 cycles. (B): Explanted samples and a human nucleus pulposus (hNP) were histologically prepared and sectioned for alcian blue and Masson's trichrome staining. (C): Explanted samples and a hNP were histologically prepared and sectioned for type II collagen, OVOSTATIN 2, and PAX1 immunostainings. Scale bar = 10 µm. Abbreviations: BM, basic medium; CHM, chondrogenic medium; NPM, nucleopulpogenic medium; hNP, human nucleus pulposus. These data showed that the in vivo transplanted NPCytes-like cells within the hydrogel of si-HPMC survive, have a secretory activity and maintain their specialized phenotype after 6 weeks. Role of R-Smads Pathways in the Nucleopulpogenic Differentiation of hASCs To gain further knowledge on the mechanisms underlying nucleopulpogenic differentiation, we assessed the respective role of TGF-β1 and GDF5 signaling pathways. We first showed that Smad1/5/8 and Smad2 were the main downstream effectors activated in hASC upon TGF-β1+GDF5 stimulation (Fig. 6A). Second, to gain further insight into the role of Smad pathways during the chondrogenic and nucleopulpogenic differentiation of hASCs, we evaluated the total number of cells with a positive immunostaining for the phosphorylated Smad2 and Smad1/5/8 proteins. At the end of the pellet formation period (day 0, Pellet formation), positive cells for phosphorylated Smad1/5/8 and Smad2 proteins were observed (Fig. 6B). Early during the induction period, the total number of positive cells for Smad1/5/8 and Smad2 phosphorylation was found to be similar in both differentiation conditions (day 7). At day 14, both Smad1/5/8 and Smad2 proteins were found phosphorylated mainly in the CHM condition (Fig. 6B, 6C). During the last 2 weeks of induction, positive cells for phosphorylated Smad2 proteins was observed to be similarly in the two differentiation conditions, positive cells for phosphorylated Smad1/5/8 proteins was found to be higher in the NPM condition (Fig. 6C). In summary, these results indicated that Smad2 phosphorylation was detected all along the nucleopulpogenic differentiation process while a clear increase in Smad1/5/8 phosphorylation was observed during the maturation phase. This increase was never observed in the CHM condition. To further decipher the respective role of the Smad1/5/8 and Smad2/3 pathways in the nucleopulpogenic differentiation of hASCs, we analyzed the effect of two specific chemical inhibitors, respectively dorsomorphin and SB505124. We evaluated the nucleopulpogenic differentiation of hASCs isolated from three patients and cultivated for 28 days in BM, CHM, and NPM supplemented with dorsomorphin or SB505124 (Fig. 7). The results showed that ACAN, COL2A1, CD24, and OVOS2 expression was not significantly altered in the NPM supplemented with dorsomorphin condition. In contrast, SB505124 supplementation of NPM led to a significant reduction in ACAN, CD24, and OVOS2 expression (Fig. 7A). We thus performed histological analyses of hASC pellets after 28 days of differentiation in BM, CHM and NPM supplemented with dorsomorphin or SB505124 (Fig. 7B). Alcian blue and Masson's trichrome staining revealed that sGAG accumulation was drastically reduced, while collagen deposition was detected at a higher level in the NPM supplemented with dorsomorphin condition compared with NPM alone. In addition, we observed a change in cell shape in the presence of dorsomorphin. Indeed, cell morphology appeared similar to the morphology of CHM-cultivated hASCs. In the NPM supplemented with the SB505124 condition, sGAG and collagens were found at barely detectable levels and larger cells resembling adipocytes were observed. The detection of a high level of the adipocyte-related transcripts PPARγ and FABP4 confirmed the widespread differentiation toward adipocytes in the NPM supplemented with the SB505124 condition (Supporting Information Fig. S3). Taken together, these results indicated that the Smad2/3 and Smad1/5/8 signaling pathways are essential for hASC differentiation toward the NPCyte-like phenotype. Figure 6 Open in new tabDownload slide Smad proteins phosphorylation during differentiation of human adipose stromal cells (hASCs). Undifferentiated hASCs were stimulated for 30 minutes with transforming growth factor-β1 and growth differentiation factor 5 alone or in combination. (A): The phosphorylation of Smad1/5/8, Smad2, Erk, p38, and jnk proteins was analyzed using Western blot. (B): The percentage of phosphorylated Smad2- and Smad1/5/8-positive cells was calculated during hASC chondrogenesis and nucleopulpogenesis. The dotted line represents the induction time of hASCs differentiation. Error bars indicate SEM between the three patients. (C): hASCs derived from patients 2, 3, and 4 were cultivated in chondrogenic medium and nucleopulpogenic medium conditions for 28 days and the pellets prepared and sectioned for phospho-Smad1/5/8 and phospho-Smad2 immunostainings at days 14 and 28. Scale bar = 20 µm. Abbreviations: CHM, chondrogenic medium; GDF5, growth differentiation factor 5; NPM, nucleopulpogenic medium; TGF-β1, transforming growth factor-β1. Figure 6 Open in new tabDownload slide Smad proteins phosphorylation during differentiation of human adipose stromal cells (hASCs). Undifferentiated hASCs were stimulated for 30 minutes with transforming growth factor-β1 and growth differentiation factor 5 alone or in combination. (A): The phosphorylation of Smad1/5/8, Smad2, Erk, p38, and jnk proteins was analyzed using Western blot. (B): The percentage of phosphorylated Smad2- and Smad1/5/8-positive cells was calculated during hASC chondrogenesis and nucleopulpogenesis. The dotted line represents the induction time of hASCs differentiation. Error bars indicate SEM between the three patients. (C): hASCs derived from patients 2, 3, and 4 were cultivated in chondrogenic medium and nucleopulpogenic medium conditions for 28 days and the pellets prepared and sectioned for phospho-Smad1/5/8 and phospho-Smad2 immunostainings at days 14 and 28. Scale bar = 20 µm. Abbreviations: CHM, chondrogenic medium; GDF5, growth differentiation factor 5; NPM, nucleopulpogenic medium; TGF-β1, transforming growth factor-β1. Figure 7 Open in new tabDownload slide Role of Smad2/3 and Smad1/5/8 pathways during nucleopulpogenesis of human adipose stromal cells (hASCs). hASCs derived from patients 6, 7, and 9 were cultivated in the presence of basic medium (BM), chondrogenic medium, and nucleopulpogenic medium supplemented with dorsomorphin or SB505124 for 28 days. (A): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was analyzed using the Taqman low-density array. *, p ≤ 0.05 versus BM. Relative expression was represented when Cq value < 38 cycles. (B): Pellets were histologically prepared and sectioned for alcian blue and Masson's trichrome staining. Scale bar = 100 µm. Black arrowheads indicate adipocyte-like cells. Scale bar = 100 µm. Abbreviations: BM, basic medium; CHM, chondrogenic medium; NPM, nucleopulpogenic medium. Figure 7 Open in new tabDownload slide Role of Smad2/3 and Smad1/5/8 pathways during nucleopulpogenesis of human adipose stromal cells (hASCs). hASCs derived from patients 6, 7, and 9 were cultivated in the presence of basic medium (BM), chondrogenic medium, and nucleopulpogenic medium supplemented with dorsomorphin or SB505124 for 28 days. (A): The expression of ACAN, COL2A1, CD24, OVOS2, and PAX1 was analyzed using the Taqman low-density array. *, p ≤ 0.05 versus BM. Relative expression was represented when Cq value < 38 cycles. (B): Pellets were histologically prepared and sectioned for alcian blue and Masson's trichrome staining. Scale bar = 100 µm. Black arrowheads indicate adipocyte-like cells. Scale bar = 100 µm. Abbreviations: BM, basic medium; CHM, chondrogenic medium; NPM, nucleopulpogenic medium. Discussion This study demonstrates for the first time that the use of TGF-β1 associated with GDF5 in the presence of DEX synergistically induces the differentiation of hASCs toward NPCyte-like cells. We assessed the success of the differentiation of hASCs toward NPCyte-like cells using markers with physiologic importance in accordance with the emergent consensus guidelines to define a “healthy NP cell phenotype” proposed by Risbud et al. [44]. hASCs acquired the molecular signature of healthy NPCytes by expressing ACAN, COL2A1, CD24, OVOS2, and PAX1. As previously reported in BM-MSC, the terminal chondrogenic differentiation is characterized by an increased expression of ALPL, RUNX2, and MMP13 [45]. Interestingly, hASCs-derived NPCytes, that expressed high levels of ALPL transcripts, also expressed low levels of RUNX2 and MMP13 strongly suggesting that those hASCs-derived NPCytes are different from terminally differentiated chondrogenic cells. We also found that from 21 days of differentiation, cells displayed a particular morphology. They were larger and anchored within a capsule, a cellular arrangement resembling that of native NPCytes [46]. Furthermore, at 28 days, differentiated cells produced an aggrecan- and type II collagen-rich ECM comparable with that of native NP. The nucleopulpogenic differentiation was successful in nine patients and the data suggest that neither age nor body mass index affects this process. The generation of NPCyte-like cells from all the patients, which express similar level of NPCytes-related transcripts to young native hNP cells, highlights the robustness of our differentiation protocol. In addition, the low expression level of genes associated with chondrogenic, osteogenic, or adipogenic differentiation by NPCyte-like cells attests to the specificity of the hASC commitment. The development and validation of this growth factor-driven differentiation protocol is an improvement in the control of the differentiation process when compared with co-culture protocols [47-49]. In addition, the use of clinical grade growth factors enables us to consider this differentiation protocol for clinical translation. A previous study reported that the intradiscal injections of undifferentiated MSCs in rabbits led to the formation of osteophytes, probably because injected MSCs experienced an osteo-chondrogenic differentiation process within the degenerated NP [50]. The in vitro preconditioning of cells in nucleopulpogenic conditions may prove to overcome this drawback. The propensity of these committed cells to produce an abundant and functional ECM, resembling that of the NP, remains to be examined in vivo. Taking the first step forward, we evaluated the biological activity of NPCyte-like cells after transplantation in nude mice subcutis by analyzing their fate and behavior in this environment with poor nutrient supply, as they may experience after intradiscal transplantation. We demonstrated that in vitro preconditioned hASCs in NPM (a) survived after in vivo transplantation within si-HPMC hydrogel, (b) exhibited a specific secretory activity, and (c) maintained their specialized phenotype. These results clearly demonstrate that the differentiation of hASCs toward NPCyte-like cells only occurred in the presence of TGF-β1 associated with GDF5 and DEX. This differentiation process may be driven by different mechanisms. The notochord-derived NPCytes originate from a type of mesoderm distinct from the type generating the ASCs. Thus, the generation of NPCyte-like cells may require a reprogramming step to bring the hASCs to a more undifferentiated state. However, no change in the level of expression of the pluripotency markers NANOG, POU5F1 and SOX2 was observed during the course of nucleopulpogenic differentiation (RT-qPCR, data not shown). This suggests that the hASCs in the NPM condition did not gain a greater capacity to differentiate toward the nucleopulpogenic lineage but rather the combination of TGF-β1 and GDF5 directly enhanced the acquisition of NPCyte-like phenotype. The data herein also showed that Smad1/5/8 and Smad2 were the main effectors activated upon TGF-β1 and GDF5 stimulation. We postulated that the nucleopulpogenic commitment is due to the differential activation of their signaling pathways. DEX is a glucocorticoid with the ability to form a complex with its intracellular receptor and has been shown to enhance the TGF-β1 mediated expression of ACAN and COL2A1 transcripts during chondrogenesis. A local accumulation of sGAG in induced BM-MSC pellets has also been reported [51]. Consistent with these observations, our data showed a substantial and localized sGAG synthesis by hASCs cultivated in BM TGF-β1 DEX compared with a slight and homogenous sGAG synthesis occurring without DEX. An interaction between the glucocorticoid receptors and the Smad proteins has also been reported to exert a transcriptional activity [52] and could contribute to explaining the effect of DEX on hASCs chondrogenesis and nucleopulpogenesis. Alternatively, potential physical interaction between TGF-β1 and GDF5 and formation of a heterocomplex may explain their synergistic effect on hASC commitment toward the nucleopulpogenic lineage. Preliminary experiments using surface plasmon resonance showed that these TGF-β superfamily members are able to interact together with a moderate affinity (Kd = 10−7 M, data not shown). The impact of the TGF-β1/GDF5 physical interaction on the hASCs biological response remains to be further characterized. In accordance with the data obtained from developmental studies and in vitro analysis of chondrogenesis [45, 53, 54], our results showed that the chondrogenic differentiation of hASCs is initiated by both pathways and regulated by the Smad2 pathway, which is active throughout the differentiation process. Knowledge of the role played by the Smad pathways in NP formation during embryogenesis is limited. During postnatal maturation, it has been demonstrated that both pathways are associated with ECM synthesis. Studies in the mouse showed evidence of an upregulation of Smad2/3 phosphorylation during NP growth from birth to 6 months and the activation of the Smad1/5/8 pathway, during the NP maturation phase, from birth to 2 months [55, 56]. This study has demonstrated for the first time that during the course of nucleopulpogenic differentiation of hASCs, positive cells for phosphorylated Smad1/5/8 and Smad2 proteins are found at the initiation phase of induction. A drop in phosphorylation of both protein complexes follows the initiation phase. Positive cells for phosphorylated Smad1/5/8 and Smad2 proteins were found in the second phase of the differentiation process with positive cells for phosphorylated Smad1/5/8 reaching their highest level. Whether this could be dependent upon a tuned expression of TGF-β1 and GDF5 receptors is under investigation. To gain further knowledge on the respective roles of Smad1/5/8 and Smad2/3 signaling in nucleopulpogenic differentiation, we examined the consequences of the inhibition of these pathways. We observed that following the inhibition of Smad1/5/8 phosphorylation, cells loosed their typical morphology even though they maintained their NPCyte-like molecular expression pattern. In contrast, the inhibition of Smad2/3 phosphorylation led to a complete abrogation of the nucleopulpogenic differentiation of hASCs. Thus, these data indicate that the Smad2/3 pathway is essential for the acquisition of NPCyte molecular identity in the early phases of differentiation. They also showed that the Smad1/5/8 pathway is required for the maturation phase with the acquisition of NPCyte-like morphology. Conclusion This study provides evidence of a promising and innovative autologous regenerative cell source and has established a robust protocol for in vitro differentiation toward the nucleopulpogenic lineage. We generated NPCyte-like cells in vitro that display a molecular and morphological phenotype similar to the native phenotype. These engineered NPCyte-like cells produce an aggrecan- and type II collagen-rich ECM in vitro and maintain their biological activity after 6 weeks of in vivo transplantation in nude mice subcutis. These data fulfilled the necessary condition to move on to the next step of proof of concept: using NPCyte-like cells associated with si-HPMC hydrogel to test their therapeutic efficacy in a relevant animal model of disc degeneration. Acknowledgments We thank Drs. F. Lejeune and B. Le Fourn (Clinique Brétéché, Nantes) for harvesting human lipoaspirates and L. Nothrup for her careful reading and editing of the manuscript. This work was supported by INSERM and grants from the Fondation pour la Recherche Médicale (FRM pionnier de la recherche “Bioingénierie”), the AO Foundation (S-12-14G), the Fondation de l'Avenir pour la Recherche Médicale Appliquée (FARMA ET3-683), the région Pays de la Loire (BIOREGOS II and BIODIV “nouvelle équipe-nouvelle thématique” program), the Agence Nationale pour la Recherche (ANR générique 2014 REMEDIV), and doctoral grant from the AXA Research Fund (to P.C.). Author Contributions P.C.: conception and design, collection and assembly data, data analysis and interpretation, manuscript writing, final approval of manuscript; J.C.: conception and design, financial support, data analysis and interpretation, final approval of manuscript; C.B., L.L., and P.W.: data analysis and interpretation, final approval of manuscript; M.R., G.B., J.L., and B.H.F. collection and assembly data; A.M.: data analysis and interpretation; A.C.: conception and design, data analysis and interpretation, manuscript writing, final approval of manuscript; J.G.: conception and design, financial support, data analysis and interpretation, manuscript writing, final approval of manuscript. A.C. and J.G. contributed equally to this article. Disclosure of Potential Conflicts of Interest The authors indicate no potential conflicts of interest. References 1 Choi KS , Cohn MJ, Harfe BD. Identification of nucleus pulposus precursor cells and notochordal remnants in the mouse: Implications for disk degeneration and chordoma formation . Dev Dyn 2008 ; 237 : 3953 - 3958 . Google Scholar Crossref Search ADS PubMed WorldCat 2 McCann MR , Tamplin OJ, Rossant J et al. Tracing notochord-derived cells using a Noto-cre mouse: Implications for intervertebral disc development . Dis Model Mech 2011 ; 5 : 73 - 82 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Colombier P , Camus A, Lescaudron L et al. Intervertebral disc regeneration: A great challenge for tissue engineers . Trends Biotechnol 2014 ; 32 : 433 - 435 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Roberts S , Evans H, Trivedi J et al. Histology and pathology of the human intervertebral disc . J Bone Joint Surg Am 2006 ; 88 : 10 - 14 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 5 Minogue BM , Richardson SM, Zeef LA et al. Characterization of the human nucleus pulposus cell phenotype and evaluation of novel marker gene expression to define adult stem cell differentiation . Arthritis Rheum 2010 ; 62 : 3695 - 3705 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Erwin WM , Ashman K, O'Donnel P et al. Nucleus pulposus notochord cells secrete connective tissue growth factor and up-regulate proteoglycan expression by intervertebral disc chondrocytes . Arthritis Rheum 2006 ; 54 : 3859 - 3867 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Erwin WM , Islam D, Inman RD et al. Notochordal cells protect nucleus pulposus cells from degradation and apoptosis: Implications for the mechanisms of intervertebral disc degeneration . Arthritis Res Ther 2011 ; 13 : R215 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Hoyland JA , Le Maitre C, Freemont AJ. Investigation of the role of IL-1 and TNF in matrix degradation in the intervertebral disc . Rheumatology 2008 ; 47 : 809 - 814 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Le Maitre CL , Freemont AJ, Hoyland JA. The role of interleukin-1 in the pathogenesis of human intervertebral disc degeneration . Arthritis Res Ther 2005 ; 7 : R732 - 745 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Mavrogonatou E , Angelopoulou MT, Dimitris K. The catabolic effect of TNFalpha on bovine nucleus pulposus intervertebral disc cells and the restraining role of glucosamine sulfate in the TNFalpha-mediated up-regulation of MMP-3 . J Orthop Res 2014 . 32 : 1701 - 1707 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Andersson GB . Epidemiological features of chronic low-back pain . Lancet 1999 ; 354 : 581 - 585 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Haufe SM , Mork AR. Intradiscal injection of hematopoietic stem cells in an attempt to rejuvenate the intervertebral discs . Stem Cells Dev 2006 ; 15 : 136 - 137 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Orozco L , Soler R, Morera C et al. Intervertebral disc repair by autologous mesenchymal bone marrow cells: A pilot study . Transplantation 2011 ; 92 : 822 - 828 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Yoshikawa T , Ueda Y, Miyazaki K et al. Disc regeneration therapy using marrow mesenchymal cell transplantation: A report of two case studies . Spine 2010 ; 35 : 475 - 480 . Google Scholar Crossref Search ADS WorldCat 15 Murphy MB , Moncivais K, Caplan AI. Mesenchymal stem cells: Environmentally responsive therapeutics for regenerative medicine . Exp Mol Med 2013 ; 45 : 54 . Google Scholar Crossref Search ADS WorldCat 16 Gantenbein-Ritter B , Benneker LM, Alini M et al. Differential response of human bone marrow stromal cells to either TGF-beta(1) or rhGDF-5 . Eur Spine J 2011 ; 20 : 962 - 971 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Stoyanov JV , Gantenbein-Ritter B, Bertolo A et al. Role of hypoxia and growth and differentiation factor-5 on differentiation of human mesenchymal stem cells towards intervertebral nucleus pulposus-like cells . Eur Cell Mater 2011 ; 21 : 533 - 547 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Bonora-Centelles A , Jover R, Mirabet V et al. Sequential hepatogenic transdifferentiation of adipose tissue-derived stem cells: Relevance of different extracellular signaling molecules, transcription factors involved, and expression of new key marker genes . Cell Transplant 2009 ; 18 : 1319 - 1340 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Chavez-Munoz C , Nguyen KT, Xu W et al. Transdifferentiation of adipose-derived stem cells into keratinocyte-like cells: Engineering a stratified epidermis . PLoS One 2013 ; 8 : e80587 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Lee J , Baek JH, Choi KS et al. Cyclin-dependent kinase 4 signaling acts as a molecular switch between syngenic differentiation and neural transdifferentiation in human mesenchymal stem cells . Cell Cycle 2013 ; 12 : 442 - 451 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Clarke LE , McConnell JC, Sherratt MJ et al. Growth differentiation factor 6 and transforming growth factor-beta differentially mediate mesenchymal stem cell differentiation, composition and micromechanical properties of nucleus pulposus constructs . Arthritis Res Ther 2014 ; 16 : R67 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Baffi MO , Moran MA, Serra R. Tgfbr2 regulates the maintenance of boundaries in the axial skeleton . Dev Biol 2006 ; 296 : 363 - 374 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Park JY , Yoon YS, Park HS et al. Molecular response of human cervical and lumbar nucleus pulposus cells from degenerated discs following cytokine treatment . Genet Mol Res 2013 ; 12 : 838 - 851 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Merceron C , Portron S, Vignes-Colombeix C et al. Pharmacological modulation of human mesenchymal stem cell chondrogenesis by a chemically oversulfated polysaccharide of marine origin: Potential application to cartilage regenerative medicine . Stem Cells 2012 ; 30 : 471 - 480 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Portron S , Merceron C, Gauthier O et al. Effects of in vitro low oxygen tension preconditioning of adipose stromal cells on their in vivo chondrogenic potential: Application in cartilage tissue repair . PLoS One 2013 ; 8 : e62368 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Chujo T , An HS, Akeda K et al. Effects of growth differentiation factor-5 on the intervertebral disc—In vitro bovine study and in vivo rabbit disc degeneration model study . Spine 2006 ; 31 : 2909 - 2917 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Li X , Leo BM, Beck G et al. Collagen and proteoglycan abnormalities in the GDF-5-deficient mice and molecular changes when treating disk cells with recombinant growth factor . Spine 2004 ; 29 : 2229 - 2234 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Le Maitre CL , Freemont AJ, Hoyland JA. Expression of cartilage-derived morphogenetic protein in human intervertebral discs and its effect on matrix synthesis in degenerate human nucleus pulposus cells . Arthritis Res Ther 2009 ; 11 : R137 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Estes BT , Diekman BO, Gimble JM et al. Isolation of adipose-derived stem cells and their induction to a chondrogenic phenotype . Nat Protoc 2010 ; 5 : 1294 - 1311 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Merceron C , Vinatier C, Portron S et al. Differential effects of hypoxia on osteochondrogenic potential of human adipose-derived stem cells. American Journal of Physiology . Cell Physiology 2010 ; 298 : 355 - 364 . Google Scholar Crossref Search ADS WorldCat 31 Poiraudeau S , Monteiro I, Anract P et al. Phenotypic characteristics of rabbit intervertebral disc cells. Comparison with cartilage cells from the same animals . Spine 1999 ; 24 : 837 - 844 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Merceron C , Portron S, Masson M et al. The effect of two and three dimensional cell culture on the chondrogenic potential of human adipose-derived mesenchymal stem cells after subcutaneous transplantation with an injectable hydrogel . Cell Transplantation 2011 ; 20 : 1575 - 1588 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Vinatier C , Magne D, Weiss P et al. A silanized hydroxypropyl methylcellulose hydrogel for the three-dimensional culture of chondrocytes . Biomaterials 2005 ; 26 : 6643 - 6651 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Vinatier C , Gauthier O, Fatimi A et al. An injectable cellulose-based hydrogel for the transfer of autologous nasal chondrocytes in articular cartilage defects . Biotechnol Bioeng 2009 ; 102 : 1259 - 1267 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Erickson GR , Gimble JM, Franklin DM et al. Chondrogenic potential of adipose tissue-derived stromal cells in vitro and in vivo . Biochem Biophys Res Commun 2002 ; 290 : 763 - 769 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Babadagli ME , Tezcan B, Yilmaz ST et al. Matrilin-3 as a putative effector of C-type natriuretic peptide signaling during TGF-beta induced chondrogenic differentiation of mesenchymal stem cells . Mol Biol Rep 2014 ; 41 : 5549 - 5555 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Yang X , Shang H, Katz A et al. A modified aggregate culture for chondrogenesis of human adipose-derived stem cells genetically modified with growth and differentiation factor 5 . Biores Open Access 2013 ; 2 : 258 - 265 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Clouet J , Grimandi Gl, Pot-Vaucel M et al. Identification of phenotypic discriminating markers for intervertebral disc cells and articular chondrocytes . Rheumatology 2009 ; 48 : 1447 - 1450 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Rutges J , Creemers LB, Dhert W et al. Variations in gene and protein expression in human nucleus pulposus in comparison with annulus fibrosus and cartilage cells: Potential associations with aging and degeneration . Osteoarthritis Cartilage 2010 ; 18 : 416 - 423 . Google Scholar Crossref Search ADS PubMed WorldCat 40 Pretheeban T , Lemos DR, Paylor B et al. Role of stem/progenitor cells in reparative disorders . Fibrogenesis Tissue Repair 2012 ; 5 : 20 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Bomer N , den Hollander W, Ramos YF et al. Underlying molecular mechanisms of DIO2 susceptibility in symptomatic osteoarthritis . Ann Rheum Dis 2014 ; 71 : 1254 - 1258 . Google Scholar OpenURL Placeholder Text WorldCat 42 Wang M , Tang D, Shu B et al. Conditional activation of beta-catenin signaling in mice leads to severe defects in intervertebral disc tissue . Arthritis Rheum 2012 ; 64 : 2611 - 2623 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Mathieu E , Lamirault G, Toquet C et al. Intramyocardial delivery of mesenchymal stem cell-seeded hydrogel preserves cardiac function and attenuates ventricular remodeling after myocardial infarction . PLoS One 2012 ; 7 : e51991 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Risbud MV , Schoepflin ZR, Mwale F et al. Defining the phenotype of young healthy nucleus pulposus cells: Recommendations of the Spine Research Interest Group at the 2014 Annual ORS Meeting . J Orthop Res 2014 ; 33 : 283 - 293 . Google Scholar Crossref Search ADS WorldCat 45 Hellingman CA , Davidson EN, Koevoet W et al. Smad signaling determines chondrogenic differentiation of bone-marrow-derived mesenchymal stem cells: Inhibition of Smad1/5/8P prevents terminal differentiation and calcification . Tissue Eng Part A 2011 ; 17 : 1157 - 1167 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Trout JJ , Buckwalter JA, Moore KC. Ultrastructure of the human intervertebral disc: II. Cells of the nucleus pulposus . Anat Rec 1982 ; 204 : 307 - 314 . Google Scholar Crossref Search ADS PubMed WorldCat 47 Lu ZF , Zandieh Doulabi B, Wuisman PI et al. Differentiation of adipose stem cells by nucleus pulposus cells: Configuration effect . Biochem Biophys Res Commun 2007 ; 359 : 991 - 996 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Richardson SM , Walker RV, Parker S et al. Intervertebral disc cell-mediated mesenchymal stem cell differentiation . Stem Cells 2006 ; 24 : 707 - 716 . Google Scholar Crossref Search ADS PubMed WorldCat 49 Sun Z , Liu ZH, Zhao XH et al. Impact of direct cell co-cultures on human adipose-derived stromal cells and nucleus pulposus cells . J Orthop Res 2013 ; 31 : 1804 - 1813 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Vadala G , Sowa G, Hubert M et al. Mesenchymal stem cells injection in degenerated intervertebral disc: Cell leakage may induce osteophyte formation . J Tissue Eng Regen Med 2012 ; 6 : 348 - 355 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Derfoul A , Perkins GL, Hall DJ et al. Glucocorticoids promote chondrogenic differentiation of adult human mesenchymal stem cells by enhancing expression of cartilage extracellular matrix genes . Stem Cells 2006 ; 24 : 1487 - 1495 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Smoak KA , Cidlowski JA. Mechanisms of glucocorticoid receptor signaling during inflammation . Mech Ageing Dev 2004 ; 125 : 697 - 706 . Google Scholar Crossref Search ADS PubMed WorldCat 53 Baffi MO , Slattery E, Sohn P et al. Conditional deletion of the TGF-beta type II receptor in Col2a expressing cells results in defects in the axial skeleton without alterations in chondrocyte differentiation or embryonic development of long bones . Dev Biol 2004 ; 276 : 124 - 142 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Retting KN , Song B, Yoon BS et al. BMP canonical Smad signaling through Smad1 and Smad5 is required for endochondral bone formation . Development 2009 ; 136 : 1093 - 1104 . Google Scholar Crossref Search ADS PubMed WorldCat 55 Dahia CL , Mahoney EJ, Durrani AA et al. Intercellular signaling pathways active during intervertebral disc growth, differentiation, and aging . Spine 2009 ; 34 : 456 - 462 . Google Scholar Crossref Search ADS PubMed WorldCat 56 Tran CM , Smith HE, Symes A et al. Transforming growth factor beta controls CCN3 expression in nucleus pulposus cells of the intervertebral disc . Arthritis Rheum 2011 ; 63 : 3022 - 3031 . Google Scholar Crossref Search ADS PubMed WorldCat © 2015 AlphaMed Press This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - TGF-β1 and GDF5 Act Synergistically to Drive the Differentiation of Human Adipose Stromal Cells toward Nucleus Pulposus-like Cells
JF - Stem Cells
DO - 10.1002/stem.2249
DA - 2016-03-01
UR - https://www.deepdyve.com/lp/oxford-university-press/tgf-1-and-gdf5-act-synergistically-to-drive-the-differentiation-of-u0HpqwHIFT
SP - 653
EP - 667
VL - 34
IS - 3
DP - DeepDyve
ER -