CAPÍTULO II MARCO TEÓRICO
ENSEÑAR A DECIDIRSE
Abstract
Wnts (wingless and int-related proteins) are a family of secreted cysteine-rich glycoproteins, expressed in a variety of tissues in developing embryos, thought to be involved in cell fate specification and stem cell commitment. To identify the specific Wnts involved in osteoblastic differentiation of human mesenchymal stem cells (hMSCs), we performed degenerative RT-PCR cloning method to amplify Wnt- encoding cDNAs expressed during osteoblastic differentiation of hMSCs in vitro and during hMSC- directed ectopic osteogenesis in the severe combined immunodeficient (SCID) mouse host. Wnt5a was found to be the dominant Wnt expressed during osteoblastic differentiation of hMSCs both in vitro and in vivo. RT-PCR further revealed that hWNT5A and its receptor Frizzled family member 5 (hFZD5) was up- regulated during osteoblastic differentiation compared to uncommitted hMSCs. To evaluate the function of Wnt5a in osteoprogenitor cells, calvarial cells were obtained from Wnt5a-/-, Wnt5a+/-, and wild type mice. Wnt5a-/- cells showed significantly slower cell proliferation when compared to Wnt5a+/- and wild type cells. Gene expression profiles of the Wnt5a-/- calvarial cells as compared to wild type cells were evaluated using microarray analysis. 255 genes exhibited at least 2-fold changes in expression. In addition, genes regulating osteoblastic differentiation including Runx2, osterix, and alkaline phosphatase (ALP) were shown to be down-regulated in Wnt5a-/- cells. In conclusion, Wnt5a expression occurs in lineage specific manner. The knock-out model suggests a role for Wnt5a signaling during osteogenesis. Knock- out cells reveal Wnt5a expression affects cell cycle progression and is required for cell growth.
Introduction
Osteoblastic differentiation is a complex process involving a variety of signaling molecules. Prominent among them are the bone morphogenic protein (BMP)-mediated signals, runt-related transcription factor 2 (Runx2) regulation of gene expression and osterix directed transcription (Ducy et al 1997; Komori et al 1997; Nakashima et al 2002; Otto et al 1997; Skillington Choy and Derynck 2002). In addition, numerous growth factors as well as systemic hormones play important roles in the definition of
the osteoblast phenotype (Aubin and Bonnelye 2000). Developmental signaling in limb as well as craniofacial development invokes signals from a wider range of molecules including msh homeobox (MSX), distal-less homeobox (DLX) and wingless-type MMTV integration site family (Wnt) members (Carlson Bryant and Gardiner 1998; Church and Francis-West 2002; Robledo et al 2002).
The Wnt secreted proteins are a family of cysteine-rich glycoproteins that regulate embryonic development, cell differentiation, proliferation, and migration. (Brandon Eisenberg and Eisenberg 2000; Church and Francis-West 2002; Li et al 2002; Lickert et al 2001). Wnt signaling is initiated by binding to two receptors: the Frizzled family (FZD) and the lipoprotein receptor-related proteins 5 or 6 (LRP5/6). The Wnt ligands are divided into two categories based on the downstream mediated signaling. The canonical Wnts are β-catenin-dependent, including Wnt1, Wnt2, Wnt3, Wnt3a, Wnt8 and Wnt8b (Katoh 2002). The other class is the so-called “non-canonical” Wnts such as Wnt4, Wnt5a, Wnt5b, Wnt6, Wnt7a, and Wnt11, which function independently of or inhibit β-catenin signaling (Church and Francis-West 2002; Katoh 2002).
The well-understood canonical Wnt signaling pathway is mediated by β-catenin activity. In the absence of a Wnt ligand, β-catenin, is bound in a cytosolic protein complex containing Axin, adenomatous polyposis coli gene product (APC), glycogen synthase kinase-3β (GSK-3β), and other proteins. Thereby, it is ubiquitinated by β-transducin repeat-containing homologue protein (βTrCP) and degraded in the proteasome. Binding of Wnt protein to Frizzled (FZD) and LRP5/6 coreceptors leads to the activation of Dishevelled (DVL) protein, which then inhibits GSK-3β-mediated phosphorylation of β- catenin. Cytosolic β-catenin is then released from the protein complex and translocated into nucleus, where it binds to T-cell factor (TCF)/lymphoid enhancer factor (LEF) transcription factors and leads to activation of the target gene expression (Gordon and Nusse 2006; Nusse 2005; Reya and Clevers 2005). These target genes consequently affect cell behavior. Some of the target genes also have feedback effects on the Wnt/β-catenin signaling and regulate cell-cell communication
Two β-catenin-independent pathways have been demonstrated previously. Some Wnts signal via the small GTPases Rho and Cdc42 to c-Jun N-terminal kinase (JNK) (Boutros et al 1998; Ryu and Chun 2006; Yamanaka et al 2002). Other Wnts can stimulate the release of intracellular Ca2+, activating protein kinase C (PKC), nuclear factor associated with T cells (NFAT) and Ca2+/calmodulin-dependent kinase II (CamkII) (Kuhl et al 2000; Sheldahl et al 1999). Wnt5a has been shown to signal through both the Wnt/JNK pathway and the Wnt/Ca2+ pathway (Kuhl et al 2000; Ryu and Chun 2006; Yamanaka et al 2002).
Wnt expression in development is particularly interesting because it appears to function in control of cell fate specification and stem cell commitment. Wnt5a and Wnt11 modulate the diversification of hematopoietic progenitor cells (Austin et al 1997; Brandon Eisenberg and Eisenberg 2000; Lako et al 2001). Wnt10b signaling was indicated to serve as a molecular switch that governs adipogenesis. Disruption of Wnt signaling caused the transdifferentiation of adipocytes to myoblasts in vitro (Ross et al 2000). Inhibition of Wnt signaling can predispose mesenchymal stem cells (MSCs) to cell cycle entry and inhibition of osteogenesis (Gregory et al 2003; Gregory et al 2005). Limb developmental research has further suggested the prominent role of Wnt signaling in mesenchymal stem cell fate (Chimal-Monroy et al 2002; Church and Francis-West 2002; Hartmann and Tabin 2000).
Given the acknowledged role of the mesenchymal stem cells (MSCs) in the multipotential generation of tissue forming cells, the possible role of Wnt signaling leading to MSCs commitment and differentiation should be defined. In the current study, we have demonstrated the unique expression pattern of Wnt5a during hMSCs differentiation along the osteoblastic lineage. The potential role of Wnt5a in cell cycle regulation of osteoprogenitor cells was further suggested by the molecular phenotype of calvaria-derived cells from Wnt5a-/- mice.
Materials and methods
Human MSCs were obtained from iliac crest bone marrow aspirates provided from 4 donors, under the Institutional Review Board (IRB) approved informed consent using procedures previously described in detail (Jaiswal et al 1997; Pittenger et al 1999). Expanded cells were cryopreserved and subsequently thawed and grown to confluence in 150 mm2 dishes. The cells were then split once, and plated into 100 mm2 dishes in Dulbecco's modified Eagle's medium (DMEM) containing 10% (v/v) fetal bovine serum (FBS).
Cells were grown to confluence (Day 0) and then induced to differentiate along the osteogenic, adipogenic and chondrogenic lineages using the corresponding established induction assay (Jaiswal et al 1997; Pittenger et al 1999). For osteoblastic differentiation, cells were treated with osteogenic supplemented (OS) medium containing 0.1 μM dexamethasone, 0.05 mM ascorbic acid, and 10 mM β- glycerophosphate in addition to DMEM-low glucose, 10% FBS and antibiotics. Cells were cultured for 1- 14 days with OS medium change every other day.
Adipogenic differentiation was stimulated by three cycles of induction/maintenance (Janderova et al 2003; Pittenger et al 1999). Each cycle consisted of three days of induction by feeding cells with adipogenic induction medium (containing 1 μM dexamethasone, 0.2 mM indomethacin, 0.01 mg/ml insulin, 0.5 mM 3-isobutyl-1-methyl-xanthine in addition to DMEM-high glucose, and 10% FBS) for 3 days, followed by 2 days of culture in adipogenic maintenance medium (containing 0.01 mg/ml insulin in addition to DMEM-high glucose, and 10% FBS).
The micromass culture system was applied to induce chondrogenic differentiation (Mackay et al 1998). 2.5 x 105 cells were pelleted and cultured in 14 ml Falcon tubes by feeding with chondrogenic induction medium (containing 0.1 μM dexamethasone, 1 mM sodium pyruvate, 0.17 mM ascorbic acid, 0.35 mM proline, 6.25 μg/ml bovine insulin, 6.25 μg/ml transferrin, 6.25 μg/ml selenous acid, 5.33 μg/ml linoleic acid, 1.25 mg/ml bovine serium albulim (BSA) and 0.01 μg/ml TGF-β3 in addition to DMEM- high glucose and 10% FBS). Cells were cultured for 1-14 days with replacing medium ever other day.
An ectopic model of bone formation was employed (Cooper et al 2001). Human MSCs were cultured in the DMEM containing 10% FBS. Upon confluence, cells were loaded into 3 mm hydroxyapatite tricalcium phosphate (HA/TCP) cubes, and implanted subcutaneously in severe combined immunodeficient (SCID) mice for ectopic bone formation. After 3-14 days, the implants were harvested, and frozen in liquid nitrogen for RNA extraction.
RNA extraction and degenerate RT-PCR
Total RNA was extracted using Trizol Reagent (Invitrogen, Carlsbad, CA), followed by DNase treatment (Clontech, Palo Alto, CA). Messenger RNAs were reverse transcribed to cDNA using Superscript II Reverse Transcriptase (Invitrogen, Carlsbad, CA). The implicated Wnt genes were then amplified by degenerate PCR using HotStarTaq DNA Polymerase (Qiagen, Valencia, CA). The degenerate primers were designed to have the ability to amplify all 19 members of human Wnts (Forward primer: nnn gtc gac gct tgy aar tgy cay gg; Reverse primer: nnn gtt aac tac gtr rca rca cca rtg). PCR products were purified using Qiaquick PCR purification microcentrifuge kit (Qiagen, Valencia, CA), and then cloned into pCR2.1 vector using TA cloning kit (Invitrogen, Carlsbad, CA).