|Home | About | Journals | Submit | Contact Us | Français|
Mesenchymal stromal cells (MSCs) originally isolated from bone marrow have multipotent differentiation potential, and favorable immunomodulatory and anti-inflammatory properties that make them very attractive for regenerative cellular therapy. Cells with similar phenotypic characteristics have now been derived from almost all fetal, neonatal and adult tissues; furthermore, more recently similar cells have also been generated from human embryonic stem cells (ESCs). Generation of MSCs from human ESCs provides an opportunity to study the developmental biology of human mesenchymal lineage generation in vitro. Generation of bone and cartilage from human ESC-derived MSCs and their functional characterization, both in vitro and in vivo, is also an active area of investigation as ESCs could provide an unlimited source of MSCs for potential repair of bone and cartilage defects. MSCs from adult sources are being investigated in numerous phase I-III clinical trials for a wide variety of indications, mainly based on their immunomodulatory properties. Our group and others have shown MSCs derived from human ESCs possess immunomodulatory properties similar to bone marrow-derived MSCs. Immunomodulatory properties of ESC-derived MSCs could prove to be highly valuable for their potential clinical applications in the future as derivatives of human ESCs have already entered clinical arena in the context of phase-I clinical trials.
Mesenchymal stromal/stem cells (MSCs) , originally isolated from bone marrow more than four decades ago, are multipotent cells capable of differentiating into cells of mesenchymal lineage such as bone, fat and cartilage 1,2. These cells are present at a very low frequency in unprocessed bone marrow and thus, there are no well-accepted criteria for their direct isolation and furthermore, their true identity in vivo is still not well known. Instead, what is essentially known as MSCs are a heterogeneous population of fibroblast like cells that are plastic adherent and are generated through ex vivo culture expansion. Among this population only a minority fulfills the strict criteria for stem cell definition, i.e. self-renewal and multi-lineage differentiation potential 3. Although MSCs were originally isolated from bone marrow, populations of cells with similar phenotypic characteristics have also been isolated from a wide variety of other adult tissues such as adipose tissue 4, skeletal muscle 5 and dental pulp 6; from neonatal tissues such as placenta 7 and umbilical cord blood 8; and from fetal tissues such as liver, blood and bone marrow 9. Due to uncertainties regarding how to define and/or characterize MSCs, a group of experts set forth the criteria for MSCs based on a combination of their culture properties, cell surface phenotypic markers and functional differentiation potential 10. According to these widely accepted criteria, MSCs grow as plastic adherent cells, do not express hematopoietic markers such as CD34 and CD45, express a number of cell surface molecules including CD73, CD90 and CD105, and have a tri-lineage differentiation potential into fat, cartilage and bone in vitro. The differentiation potential of MSCs into other mesenchymal and non-mesenchymal cell types, a highly debated topic, has not been included as part of these criteria.
MSCs have generated a lot of attention in the field of regenerative medicine not only due to their potential to differentiate into a variety of cell types but also because of their ability to migrate to sites of injury or inflammation after intravenous infusion and participate in tissue regeneration. For example, MSCs stimulate proliferation and differentiation of resident progenitor cells, and promote recovery of damaged tissues through secretion of a variety of cytokines and growth factors, and secretion of extracellular matrix components 11-14. Another intriguing property of ex vivo expanded MSCs is their ability to modulate the immune response in vitro and in vivo through interaction with a broad range of immune cells including T-lymphocytes, B-lymphocytes, natural killer and dendritic cells 15-18. A variety of in vitro studies, have shown that MSCs possess the ability to suppress activation and proliferation of T-lymphocytes 19,20. MSCs are also assumed to be not targets of cytotoxic T cells or NK cells 21,22 and thus, they could be potentially tolerated when transplanted over major histocompatibility complex barriers in humans 23. The mechanisms of action of MSCs in immunomodulation are through both direct cell-cell contact and through secretion of a variety of cytokines and paracrine factors. One of the more intriguing properties of ex vivo expanded MSCs is their ability to modulate the immune system and attenuate tissue damage caused by excessive inflammation 24; recent data strongly suggest that MSCs possess potent anti-inflammatory effects through interaction with the innate immune system cells such as macrophages 25,26. Again, many of these interactions are mediated through paracrine effects and secretion of anti-inflammatory cytokines 27-29. Thus, in contrast to pharmacological drugs or biological therapeutics, such as monoclonal antibodies, MSCs exert their beneficial effects through many different mechanisms which make them potentially beneficial for a wide range of therapies.
Autologous and allogeneic party bone marrow-derived MSCs have been investigated extensively in the context of hematopoietic stem cell transplantation either for enhancement of HSC engraftment 30,31 or for prevention and/or treatment of graft versus host disease (GVHD). Indeed, the most compelling results have come from phase II studies in steroid-refractory acute GVHD 32, and MSCs are closest to regulatory approval for use in acute GVHD pending the results of the now fully-accrued phase III trials 24. Interestingly, experience in using MSCs from third party donors have shown that use of MSCs from a donor different than that of the donor of hematopoietic stem cells (third party MSCs) is safe and as efficacious as MSCs from original donors or haplo-identical donors 32. This has opened the way for use of third party MSCs from universal donors for treatment of other immune/inflammatory mediated disorders such as Crohn’s disease 33. So far, culture-expanded MSCs derived from(autologous, allogeneic, or third party) bone marrow, and to a more limited extent, MSCs derived from other tissues such as fat have been used in numerous phase I-II trials for a variety of non-hematological indications including treatment of patients with metachromatic leukodystrophy and Hurler’s disease 34, osteogenesis imperfecta 35, myocardial infarction 36,37, chronic obstructive pulmonary disease 38, amyotrophic lateral sclerosis 39, stroke 40, refractory wounds 41, diabetes mellitus 42, systemic sclerosis 43, and systemic lupus erythematosus 44 among others. In all of these trials MSCs demonstrated an impressive safety record when used over a wide range of cell doses and administration frequencies 45. Since many of these trials proved that MSCs from third-party donors could be used without any HLA-typing it makes it theoretically feasible that a few lines of clinical-grade human ESCs could be used as a renewable source for generation of sufficient MSCs for clinical applications for a large number of patients.
Upon completion of my clinical and research fellowship at National Heart, Lung and Blood Institute I joined the Bone Marrow Transplant program at University of Wisconsin-Madison as a junior faculty. A major reason for this move to Madison, the birth place for human embryonic stem cell research46, was the fact that I intended to establish a laboratory research program in this new field to specifically investigate human hematopoiesis using this highly relevant model system. This coincided with the first publication about the potential of third party bone marrow derived MSCs for treatment of refractory graft versus host disease 47, a major complication after bone marrow transplantation. The fact that HLA-matching is not needed for MSC transplantation and thus a universal bone marrow donor can provide cells for a number of recipients sparked an interest in me to embark on laboratory experiments to derive mesenchymal stromal cells from human ESCs. The National Blood Foundation (NBF) award, coinciding with my K08 award, affirmed a sense of confidence in my ability to garner approval or my hypotheses and research plans from a group of colleague scientists. Indeed, NBF award was my first peer reviewed funding specifically directed towards my research on ESC-derived MSCs; now several years later my whole laboratory and translational research is focused on MSCs. As a bone marrow transplant physician with an interest in novel cellular therapies my goal was to develop a methodology for generation of MSCs from human ESCs that could facilitate its translation into a potential clinical product, considering the regulatory requirements by FDA for cellular therapy products. For example, although lack of animal-derived feeder cells is not a mandatory requirement in cell production it could greatly facilitate transition to the clinic. Thus, with the help of NBF seed money soon our group reported generation of cells from human ESCs that could be defined as MSCs according to the ISCT criteria for MSCs 10, without the use of any animal feeder cells 48.
In general there are three methods for differentiation of ESCs. In the embryoid body (EB) method ESCs are grown into three-dimensional cellular aggregates which can then be differentiated into ectodermal, mesodermal and endodermal tissues based on differentiation medium and growth factors added. However, such differentiations are usually non-uniform and heterogeneous with multiple differentiated cell lineages present simultaneously. Another popular methodology is to use feeder cells to direct differentiation of ESCs towards specific lineages. For example, OP9 cells (a murine bone marrow stromal cell line), is commonly used for differentiation of ESCs towards mesodermal cells such as hematopoietic cells 49. Another two-dimensional differentiation culture method is to use extracellular matrix (ECM, such as Matrigel, collagen or gelatin) in combination with specific cytokines and growth factors in culture medium.
A few years prior to studies in which generation of cells with bona fide characteristics of MSCs from ESCs were reported a few studies reported generation of mesenchymal-like cells that now in retrospect can be considered to be MSCs despite the fact that those cells were not fully characterized or labeled as MSCs 50-52. For example, Stojkovic et al. showed that fibroblast-like cells that formed spontaneously in human ESC cultures, designated human ESC-derived fibroblasts, express MSC cell surface markers such as CD44 and CD90 but their differentiation potential into mesenchymal lineages was not further investigated 51. However, Barberi et al. were the first to show mesenchymal-looking cells developed after 40 days of co-culturing human ESCs with OP9 cells, had characteristics typical of MSCs 53 based on expression of cell surface markers CD29, CD44, CD73, CD105, and their differentiating into bone, cartilage and fat. Later, Olivier et al. reported the Raclure method to generate MSCs from human ESCs, a method that did not need co-culturing with OP9, or any other supportive feeder cells 54. Lian et al. also reported a method to expand MSCs from the CD105+/CD24− population of cells differentiated from human ESCs 55, without a co-culture method. Although originally we used OP9 cells to derive MSCs from ESCs 56 later we developed a methodology to generate MSCs from human ESCs that did not include use of OP9 cells or any other animal feeder layer 48; such a modification removes a potential hurdle for their ultimate clinical application in the future. Not surprisingly, cells reported by different groups do not have the exact same phenotypic characteristics because even minor differences in the culture methodology can cause minor or major differences in the phenotype of generated cells. For example, cells generated by Olivier et al. 54 expressed typical markers of MSCs such as CD73 and CD105 but unlike our ESC-derived MSCs 48,56, those cells also expressed SSEA4 46. Presence of SSEA4 could pose a challenge as this marker is expressed on undifferentiated human ESCs and thus could confound the assessment of the purity of MSCs. Similarly, MSCs generated by Hwang et al. 57 exhibited many markers of MSCs including CD29 and CD105, but lacked CD73, a typical marker expressed by bone marrow derived MSCs 10. This variability in cell surface marker expression could be due to many known and unknown factors. It will be very challenging to determine if, and how, such phenotypic differences in cell surface markers could impact the functional characteristics of MSCs. Aside from the potential clinical utility of human ESC-derived MSCs (discussed below) one of the obvious applications of this development is that it provides a new platform to study basic human developmental processes by allowing investigation of the development of mesodermal, or any other tissues, in vitro. Recapitulating development of tissues in culture dishes under our direct observation provides an unprecedented opportunity to study the earliest stages of human development in vitro 49.
ESC-derived MSCs are of huge interest as they could be used for many non-clinical, and in the future, potentially clinical applications. For example, conventionally, human ESCs are derived and propagated on feeder cells, mainly murine embryonic fibroblasts (MEF) 46. Indeed, human ESC derived MSCs have now shown to be capable of functioning as autogenic or allogeneic feeder layers for derivation and/or expansion of human ESCs 58. Indeed, the ability to derive and use MSC feeder layers derived from human ESCs or from human adult bone marrow 59, instead of animal-derived feeder cells (i.e. MEF) is advantageous and potentially provides a safer source of cells for clinical applications.
In addition to studies on generation of MSCs from human ESCs, there has been a flurry of studies in which generation of derivative of MSCs, i.e. osteoblasts and chondrocytes, from human ESCs have been described 60-65. However, it is not known if in these methodologies MSCs were only transiently present in the culture and thus went undetected or bone or cartilage lineages were generated without going through an MSC intermediate phase. In any case these studies provide unique insight into the osteogenesis and chondorgensis in vitro and a potentially unlimited source for reconstructive surgery applications, including bone or cartilage repair 66. However, it can be also envisioned that ESC-derived MSCs could be directly implanted into bone or cartilage assuming the host tissue environment provides necessary signals towards their final differentiation into bone and cartilage 67. Although our laboratory was not the first to report the generation of MSCs from human ESCs our group was the first to report on the immunological characteristics of these cells 48. We showed that human ESC-derived MSCs possess immunological properties very similar to BM-derived MSCs, such as expression of HLA Class I antigens and lack of expression of HLA Class II antigens or co-stimulatory molecules, lack of capability to stimulate third party lymphocytes, and being able to suppress T-cell proliferation in mixed lymphocyte culture assays. Since our original report many other groups have expanded on our original observations and shown the immunosuppressive properties of ESC-derived MSCs both in vitro 68 and in vivo 69. Indeed the immunomodulatory and anti-inflammatory properties of MSCs is one of the most common reasons for their use in clinical trials. It is estimated that several thousand patients have so far have received BM-derived MSCs 70. MSCs derived from other sources such as adipose tissue, placenta, umbilical cord and even fetal tissues have also been used but to a much more limited extent 45. It can be argued that human ESCs could provide an unlimited source of MSCs for clinical application and thus obviate the need for invasive procedures to obtain MSCs from adult human volunteers. However, it should be noted that due to proliferative expansion of MSCs from adult sources usually small amounts of starting material is enough for treating potentially hundreds of patients. Also, tissues such as fat that will be otherwise discarded after surgery could be a source of MSCs. It should be also emphasized that although use of third party MSCs has been shown to be safe for specific indications such as GVHD, there will be certain clinical scenarios in which long-term engraftment of MSCs will be desirable, such as for repair of bone defects. In these situations use of autologous MSCs derived from adult tissues or MSCs derived from HLA-matched human ESC lines could prove to be advantageous. Nevertheless, ESC-derived MSCs remain as a potential therapeutic option for unique reasons. It can be envisioned that MSCs derived from a pluripotent stem cell line (either embryonic stem cell or induced pluripotent stem cells) could be used to enhance engraftment of other types of cells derived from those lines. Similar to utilization of BM-MSCs for enhancement of hematopoietic engraftment ESC-derived MSCs could be also used for potential enhancement of ESC-derived HSCs. MSCs are also being investigated as potential facilitators of solid organ transplantation 71,72. So, it could be also envisioned that MSCs derived from a pluripotent stem cell line could be used to enhance/promote engraftment of other cells, such as cardiomyocyte or pancreatic islet cells, derived from the same cell line.
However, before human ESC-derived MSCs reach clinic there are many issues that need to be resolved. For example, similar to the situation with adult tissue derived MSCs it is not clear if ESC-derived MSCs at different passages (passage-0 arbitrarily defined as the population that is homogenously looking like adult MSCs) possess different biological effects and what are the optimal passage of MSCs before they lose their biological properties. Furthermore, in the case of human ESCs as the source material for MSC derivation there will be an added layer of complexity as there will be ESC-inherent variables such as the passage number of human ESCs or the way they were derived that could have an effect on the biological properties of their derivatives (i.e. MSCs).
Infusion of ex vivo culture expanded MSCs from adult human bone marrow have been shown an impressive safety record of MSCs in all clinical trials thus far 45. Furthermore, transplantation of MSCs derived from other adult, neonatal and fetal tissues have been shown to be safe too due to the fact that these cells are considered to be not only non-immunogenic but actually elicit an immunomodulatory effect in the recipient. Based on the similar profile of immunomodulatory properties of human ESC-derived MSCs shown by our group and others it can be assumed that if MSCs can be generated from a few human ESC lines under clinically acceptable conditions they could provide an alternative to adult tissue-derived MSCs and could be used as universal donor cells. However, use of any derivate of human ESCs carries some major concerns. For example, undifferentiated human ESCs, by definition, generate teratomas when transplanted into immunodeficient animals 73, a major concern for any clinical application of derivatives of human ESCs 74. However, derivation of MSCs from human ESCs generates highly pure population of cells in contrast to other types of cells75-77. One reason might be the fact that generation and expansion of MSCs from human ESCs, and any other sources, involves repeated passaging of cells in culture conditions that are not permissible for growth of human ESCs. This is in contrast to other derivatives of ESCs, such as cells of cardiac or neural lineages, in which repeated passaging is not part of their differentiation processes. In these circumstances persistence of undifferentiated ESCs could pose a major challenge to the safety of cellular therapy product.
Derivative of human ESCs provides the next frontier for novel cellular therapies. MSCs derived from human ESCs could provide an unlimited source of cells for potential clinical applications. These cells could be used based on their immunomodulatory properties, similar to bone marrow derived MSCs, or could be further differentiated into bone and cartilage before implantation. Furthermore, immunomodulatory and tissue protective effects of these cells provide a rationale for their co-transplantation with cells/tissues derived from the same pluripotent stem cell lines. Generation of MSCs from ESCs using clinically applicable methodologies could expand and facilitate use of derivatives of ESCs in the field of regenerative medicine.
The author is recipient of National Blood Foundation Scientific Research Grant and National Institutes of Health/NHLBI HL081076 K08 grant.