PERSPECTIVES research12 (BOX 2). Nevertheless, many OPINION important aspects regarding MSC biology are still unclear. In this Opinion article, we The elusive nature and function of discuss the well­established properties of MSCs cultured in vitro, and focus on how these properties relate to recent studies that mesenchymal stem cells are beginning to uncover MSC localization and function in vivo. César Nombela-Arrieta, Jerome Ritz and Leslie E. Silberstein Studying MSCs in vitro Abstract | Mesenchymal stem cells (MSCs) are a diverse subset of multipotent Defining MSCs. Stem cells are classically precursors present in the stromal fraction of many adult tissues and have drawn defined by their multipotency and self­ intense interest from translational and basic investigators. MSCs have been renewal (BOX 1). Based on these criteria, the operationally defined by their ability to differentiate into osteoblasts, adipocytes central and most disputed issue is the use and chondrocytes after in vitro expansion. Nevertheless, their identity in vivo, of the term MSC. This term was first used heterogeneity, anatomical localization and functional roles in adult tissue to refer to a hypothetical postnatal, multi­ potent and self­renewing precursor derived homeostasis have remained enigmatic and are only just starting to be uncovered. from an original embryonic MSC, the func­ tion of which was to maintain the turnover Stem cells are clonogenic cells that have two colony­forming unit fibroblasts (CFU­Fs)4, in of skeletal tissues in homeostasis or tissue remarkable features, the ability to differenti­ analogy to their haematopoietic counterpart) repair during adulthood. As described for ate into multiple mature cell types (multi­ that could be selected by adherence to plastic HSCs, MSCs would lie at the top of the potency) and the ability to simultaneously surfaces and expanded in vitro. Further stud­ mesenchymal cell hierarchy and progress replenish the stem cell pool (self­renewal), ies substantiated the ability of cultured cells through discrete stages of differentiation that allow them to sustain tissue develop­ derived from single CFU­Fs to proliferate in an orderly manner to give rise to func­ ment and maintenance1 (BOX 1). The surge of while preserving the ability to differentiate tionally and pheno typically mature tissues, stem cell research arose when it became clear to osteoblasts, adipocytes and chondrocytes including bone, smooth muscle, tendons that all blood cell components are derived in vitro5. Multilineage capacity and prolifera­ and cartilage6. from a rare subset of bone marrow (BM)­ tion in vitro were interpreted at the time as This theoretical model provided an residing haematopoietic stem cells (HSCs), indicative of in vivo multi potency and self­ attractive conceptual framework in which which can be isolated and assayed in vitro renewal, the hallmarks of ‘stemness’. Thus, the stromal multipotent precursor cell and in vivo. Currently, HSCs are the para­ the term mesen chymal stem cell (MSC) was described by Friedenstein was rapidly digmatic and best­characterized example coined and gained acceptance to refer to regarded as the prototypical MSC, despite of tissue­specific stem cells. these newly identified precursor cells6,7. Since the fact that such a cell had not been shown Soon after the discovery of HSCs, studies their original description, stromal cells cat­ to strictly fulfil the attributes conveyed in by Friedenstein and colleagues2 reported that egorized as MSCs based on trilineage (osteo­ the term MSC. First, the biological proper­ the BM stroma can generate bone, fat cells, blast, adipo cyte and chondrocyte) potential ties of multipotent progenitors had mostly cartilage and reticular cells following hetero­ in vitro have been isolated from the adherent been inferred from the analysis of clonal topic transplantation (that is, transplantation fraction of many adult and embryonic tissues and non­clonal in vitro­expanded popu­ into a different tissue from that of origin) in in multiple species8–11 (FIG. 1). lations owing to the inability to isolate and mice. This suggested the existence of non­ Multipotentiality in vitro, as well as assay them directly from tissues. Until haematopoietic BM multipotent precursor ease of isolation and expansion, rapidly recently, the multipotency and self­renewal cells with skeletal and adipose potential2,3. positioned MSCs as promising therapeutic of un cultured progenitors had not been fully It was later shown that these precursors were agents in regenerative medicine and made probed using stringent in vitro and in vivo a subset of fibroblast­like cells (defined as them the subject of intensive clinical assays. Furthermore, the existence of a com­ mon postnatal ‘mesenchymal’ progenitor has been questioned, as bone and muscle derive Box 1 | Criteria used to define stem cells: HSCs and BM‑resident MSCs from different progenitors during embryonic Multipotency development, and because whether MSCs The first criterion used to define stem cells is their ability to differentiate into multiple types give rise to muscle cells in vivo has not been of functionally mature specialized cells. Haematopoietic stem cells (HSCs) can give rise to all convincingly demonstrated to date. For this blood cell components, including neutrophils, lymphocytes, natural killer cells, dendritic cells, reason, alter native names such as osteogenic macrophages and monocytes. Bone marrow (BM)-resident mesenchymal stem cells (MSCs) can or skeletal stem cells have been suggested. 1 give rise to osteoblasts, chondrocytes, adipocytes and reticular stroma . Regardless of its inaccuracy13,14, the term Self-renewal MSC has remained prevalent to date to Stem cells also possess the capacity of self-renewal; that is, the ability to undergo numerous cell designate stromal precursors with triline­ divisions while retaining their stem cell identity. HSCs have been shown to self-renew: HSCs age potential isolated from the BM and, by transplanted into irradiated mice can reconstitute the haematopoietic system, including giving extension, from any other mammalian tis­ rise to a population of HSCs that can be serially re-transplanted1. BM-resident MSCs have also sue. Of note, the common use of the name been shown to possess this ability, as their transplantation gives rise to ‘ossicles’ composed of bone, cartilage and reticular stroma, as well as to a population of serially re-transplantable MSCs1. MSC to indistinctively refer to both in vivo precursors and their in vitro­expanded 126 | FEBRUARY 2011 | VOLUME 12 www.nature.com/reviews/molcellbio © 2011 Macmillan Publishers Limited. All rights reserved PERSPECTIVES progeny has frequently led to misconcep­ tions in the field. The International Society for Cellular Therapy, Vancouver, Canada, has recommended the use of the name multipotent mesenchymal stromal cell (also $QPGOCTTQY -KFPG[ 2CPETGCU #FKRQUGVKUUWG 5MGNGVCNOWUENG $TCKP .KXGT abbreviated to MSCs, although not in this Opinion article) for the in vitro cultured cells, restricting the term stem cell to 2GTKXCUEWNCTOWNVKRQVGPV RTGEWTUQTt/5%! des ig nate the proposed in vivo precursors %& or stem cells15,16. 0) 2&)(4β Characterization of mesenchymal stromal cells. Beyond their ability to generate osteo­ blasts, adipocytes and chondrocytes in vitro5, mesenchymal stromal cells give rise to bone %(7( and cartilage after ectopic implantation in vivo17,18 and have been documented to contribute to bone regeneration in animal 19 Multipotent mesenchymal models of genetic bone disorders . Many stromal cell studies have further reported mesenchymal stromal cell differentiation into multiple other cell types of mesodermal and non­ mesodermal origin, including endothelial &GZCOGVJCUQPG +UQDWV[NOGVJ[NZCPVJKPG 6TCPUHQTOKPIITQYVJ cells20, cardiomyocytes21, hepato cytes22 and .CUEQTDKECEKFCPF FGZCOGVJCUQPG HCEVQTβCPFFGZCOGVJCUQPG 23,24 KPQTICPKERJQURJCVG KPFQOGVJCEKPGCPFKPUWNKP neural cells . Nevertheless, such multi­ ! potential capabilities of mesenchymal stro­ mal cells are not universally accepted. There are concerns because of the lack of globally 1UVGQDNCUVU #FKRQE[VGU %JQPFTQE[VGU 1VJGTEGNNV[RGU standardized methods for their isolation, Figure 1 | MSCs and multipotent mesenchymal stromal cells. The plastic-adherent cellular frac- expansion and identification, as well as the tion of many organs contains stromal progenitor cells that 0CVWTcan giveG4G riseXKGYU to ^colonies/QNGEWNCT%GNN$KQNQI[ of fibroblastic range of assays used to define terminally morphology. This cellular subset, known as colony-forming unit fibroblasts (CFU-Fs), totally or partially differentiated, functionally mature popu­ corresponds to a proposed multipotent progenitor cell population, most probably heterogeneous in lations. Claims for in vivo differentiation nature and origin, that resides in the proximity of blood vessels in all tissues studied so far and has been into other cell types are equally controver­ shown to express pericyte-specific markers (CD146, NG2 (also known as CSPG4) and platelet-derived growth factor receptor (PDGFR )). When cultured under the appropriate cell densities, colonies sial, as BM­derived mesenchymal stromal ­β β derived from single CFU-Fs can be isolated and expanded after multiple passages in vitro (indicated cell cultures have been shown to contribute by the curved arrow) without losing their multipotent mesenchymal capacity. These cultured cells, to many tissues following transplantation classically referred to as mesenchymal