Adipose-derived stromal/stem cells

Abstract

Until recently, the complexity of adipose tissue and its physiological role was not well appreciated. This changed with the discovery of adipokines such as leptin. The cellular composition of adipose tissue is heterogeneous and changes as a function of diabetes and disease states such as diabetes. Tissue engineers view adipose tissue as a rich source of adult stromal/stem cells isolated by collagenase digestion. In vitro and in vivo studies have documented that adipose stromal/stem cells are multipotent, with the ability to differentiate along the adipocyte, chondrocyte, osteoblast and other lineage pathways. The adipose stromal/stem cells secrete a wide range of cytokines and growth factors with potential paracrine actions. Furthermore, adipose stromal/stem cells exert immunomodulatory functions when added to mixed lymphocyte reactions, suggesting that they can be transplanted allogeneically. This review article focuses on these mechanisms of adipose stromal/stem cell action and their potential utility as cellular therapeutics.

Introduction

Over the past three decades, the concept of stem and progenitor cells in adult tissues has evolved considerably. Historically, scientists hypothesized that the number of cells within a mature organ, such as adipose tissue or the central nervous system, was limited sometime shortly after birth. Consequently, subsequent regeneration was restricted due to insufficient numbers of reparative cells capable of differentiation and renewal. Nevertheless, evidence from pathological conditions challenged this paradigm since some patients with either stroke or liver resection were able to recover function and tissue mass. In the case of adipose tissue, the rare pediatric malady known as “Progressive Osseous Heteroplasia” (POH) which manifests as ectopic endochondral bone formation in the subcutaneous tissue secondary to a mutation in GNASα, provided evidence of a multipotent stromal/stem cell in adipose depots.¹ Pioneering work by Alexander Friedenstein at Moscow University along with his colleague Maureen Owen at Oxford University documented the presence of bone marrow derived fibroblasts or stromal cells capable of differentiation into multiple lineages.^2-4 In the early 1990s, Arnold Caplan at Case Western University, coined the term “Mesenchymal Stem Cell” (MSC) to describe those bone marrow derived cells with the ability to both self renew and differentiate along classical mesenchymal lineages, including adipocytes, chondrocytes, myocytes, osteoblasts, and tenocytes.^5-7 These advances led to the hypothesis that the transplantable bone marrow-derived hematopoietic stem and progenitor cells (HSPCs) are not the sole regenerative cell found within adult mammals. In 2006, the International Society for Cell Therapy (ISCT) formalized the acronym MSC as “Multipotent Stromal Cells” which could be identified based on expression of a subset of both positive stromal (CD44, CD73, CD90, CD105) and negative hematopoietic (CD11b, CD14, CD19, CD34, CD45, CD79α, HLA-DR) cell-associated surface antigens as well as their ability to differentiate along the adipocyte, chondrocyte, and osteoblast lineages in vitro.^8,9 MSC action in vivo has been attributed to a number of mechanisms including (A) differentiation potential, (B) release of paracrine factors influencing the microenvironment, (C) scavenging of reactive oxygen species, (D) immunomodulatory function and (E) fusion and rejuvenation of resident committed progenitor cells.^10,11 Regardless of mechanism, the potential availability of adult tissue-derived stem-like cells capable of self renewal, differentiation, and transplantation has contributed to the theoretical framework underpinning the developing fields of tissue engineering and regenerative medicine.

Regulatory Environment for Stromal/Stem Cells

Primary cultures of adult stromal/stem cells offer advantages for basic research and clinical translation.¹¹ In contrast to cell lines, human tissue derived stromal/stem cells have not undergone immortalization and may reflect more accurately tissue function and metabolism in vitro. Primary human adult stromal/stem cells should be evaluated in vivo in a stepwise manner advancing from small animal (rodents) to large animal, up to and including non-human primates, before advancing to clinical translation . All studies should be conducted with Institutional Animal Care and Use Committee (IACUC) review and approval. Furthermore, all human tissues should be obtained using an Institutional Review Board (IRB) approved informed patient consent mechanism. Due to the risk of blood borne pathogens, personnel should employ universal precautions whenever working with primary human cell cultures. Each laboratory should establish standard operating procedures that minimize the technician’s risk of infection due to sharps exposure, live cell sorting by flow cytometry, biohazard waste handling, and other procedures. The primary human cell products should meet quality control/quality assurance criteria outlined by national and international regulatory authorities. These include safety studies confirming that the cells (A) are not tumorigenic, migratory, or causes of ectopic tissue differentiation in an in vivo animal transplant model, (B) are free of bacterial, endotoxin, mycoplasma, and viral contaminants and (C) have not been altered significantly if more than minimally manipulated due to isolation procedure or by combination with bioscaffolds and/or genetic materials. Additionally, regardless of mechanism, the stromal/stem cells must exhibit some enhanced level of efficacy above and beyond that of current standard or care. Finally, it is advantageous if the stromal/stem cell therapy offers a financial benefit with respect to the long-term treatment of the underlying disease and its economic health care burden.

Adipose Tissue Composition

For decades, adipose tissue was viewed as a passive organ relegated to serve primarily as an energy reservoir.^12,13 The discovery of leptin as a circulating adipose-derived adipokine with systemic actions led to the re-classification of adipose tissue as a bona fide endocrine organ.¹⁴ Adipose tissue is found throughout the adult human body in bone marrow, intra-articular, subcutaneous, and visceral depots, as well as ectopic sites such as intra-hepatic and intra-muscular. The biology of adipose tissue has received increased international attention due to the obesity epidemic. Today, > 30% of adults in the United States are obese (body mass index or BMI > 30) and, based on trends in the pediatric population, these numbers are expected to increase further in coming years.¹⁵ Mature adipocytes within adipose depots have been organized recently as follows:

1. White adipocytes—energy storage depot with adipokine secretory function morphologically characterized in vivo by the presence of large lipid vacuoles.

2. Brown adipocytes— energy storage depot with non-shivering thermogenic function associated with the expression of the mitochondrial membrane Uncoupling Protein 1 (UCP1) and morphologically characterized in vivo by the presence of multiple small lipid vacuoles. Brown adipocytes have developmental links most closely to skeletal muscle rather than white adipocyte progenitor cells.^16-18

3. Beige adipocytes (also identified as “brite” or “brown/white”)—energy storage depot with the potential to express UCP1 but most closely linked developmentally to white adipocytes.¹⁹ Some have suggested that white adipocyte progenitors can trans-differentiate into beige adipocytes.

Although mature adipocytes comprise the bulk of adipose tissue’s volume, there is considerable cellular heterogeneity. The various cell types can be visualized by direct immunohistochemical detection of fixed or unfixed adipose tissue sections. Alternatively, their numbers can be quantified using flow cytometry. Adipose tissue obtained as excised surgical specimens or as lipoaspirates are digested with bacterially-derived collagenase enzyme in the presence of calcium to release the individual cell components (Fig. 1).^20,21 Subsequently, differential centrifugation is used to separate the mature adipocytes, which float, from the remaining cells, which form a Stromal Vascular Fraction (SVF) pellet.²¹ The SVF cell population includes endothelial cells, fibroblasts, B- and T-lymphocytes, macrophages, myeloid cells, pericytes, pre-adipocytes, smooth muscle cells, and the culture adherent adipose stromal/stem cells (ASC). After 4 to 6 d in culture with medium containing 10% fetal bovine serum, a single milliliter of human lipoaspirate will yield between 0.25 to 0.375 X 10⁶ ASCs capable of differentiating along the adipocyte, chondrocyte and osteoblast lineages in vitro.^22,23 Since > 400,000 patients in the US routinely undergo liposuction annually, often resulting in > 1 L of tissue, it is feasible to generate hundreds of million ASCs from a single donor within a single in vitro cell culture passage. These yields are sufficient to support regenerative medical applications at the clinical level. In contrast to the SVF cells, ASCs are relatively homogeneous based on their expression profile of surface antigens. Recently, the ISCT and the International Federation for Adipose Therapeutics and Science (IFATS) have established minimal criteria defining SVF cells and ASC based on functional and quantitative criteria, similar to but distinct from those identifying bone marrow MSCs.²⁴ Several companies have developed closed system devices designed to isolate SVF cells.²⁵ These automated devices are capable of reproducible outcomes under current Good Manufacturing Practice guidelines in a clinical setting and are at various stages of regulatory review internationally. At present, issues relating to the use of collagenase digestion remain to be resolved before surgeons can routinely employ machines at the point of care.

Figure 1. Isolation of Adipose-Derived Cells. Lipoaspirate tissue (1) is washed in buffered saline solution (2) and subjected to collagenase digestion with rotation (3) prior to centrifugation and isolation of the stromal vascular fraction (SVF) pellet (4). The SVF cells are then incubated in culture (5) and subsequently differentiated along the adipocyte (Oil Red O) and osteoblast (Alizarin Red) lineages (6).

The presence of hematopoietic lineages suggests that adipose tissue plays a potential role in innate immunity and as an HSPC niche.^26-28 Obesity and diabetes are known to alter adipose tissue’s cellular composition. Diabetes is associated with sterile inflammation of adipose tissue. This has been characterized by the appearance of “crown” cells, a ring of macrophage-like cells surrounding an adipocyte undergoing apoptosis and capable of releasing inflammatory cytokines such as interleukin 6 and tumor necrosis factor α.^29-31 Similarly, obesity influences the T-lymphocyte population of adipose depots.^32,33 While the number of CD4⁺ and CD8⁺ T cells within subcutaneous adipose tissue of lean individuals (BMI < 25) is well below 1%, these values approach 4–5% in comparable tissues from morbidly obese subjects (BMI > 40).³² This correlates with increased adipose tissue levels of inflammatory T-lymphocyte derived cytokines.³² Studies in murine models have suggested that the SVF contains putative HSPCs.²⁶ When transplanted into lethally irradiated mice, adipose-derived SVF cells performed equivalently to bone marrow transplants. In contrast to the untreated controls which expired prior to 4 weeks post-irradiation, ~30–40% of the SVF and bone marrow transplant recipients survived for up to 10 weeks.²⁶ Consistent with these findings, human ASCs support the proliferation and differentiation in vitro of co-cultured human umbilical cord-derived hematopoietic progenitors.^34,35

Immunomodulatory Role of ASC

In recent years, considerable attention has focused on the paracrine and immunomodulatory mechanism of SVF cell and ASC reparative function. Multiple independent studies have used the mixed lymphocyte reaction (MLR) as an in vitro approach.^36-38 The MLR monitors the proliferative response of a non-irradiated “responder” lymphocyte when exposed to an irradiated “stimulator” lymphocyte no longer capable of a mitotic response (Fig. 2A and B). If the responder lymphocyte displays a histocompatible antigen mismatch with the stimulator cell, a robust proliferative response occurs. When irradiated SVF cells are used as a stimulator cell with a non-identical peripheral blood lymphoid cell, a positive MLR was observed (Fig. 2B). This is consistent with the presence of the HLA-DR antigen on the SVF cell surface. In contrast, with progressive passage in vitro, ASCs from the same donor lose their HLA-DR expression and fail to elicit an MLR when co-cultured with the same peripheral blood lymphoid cells. This reproducible finding suggests that ASC, unlike SVF cells, can be transplanted across classical immunogenic barriers with reduced risk of rejection.^36,37

Figure 2. Mixed Lymphocyte Reactions (MLR). (A) Inactive: Peripheral blood monocytes (PBNs) isolated from a single donor (blue) are divided into two populations. One is irradiated to prevent proliferation and used as a stimulator cell. The non-irradiated cells are used as a responder population. When the two populations are mixed together, the responder cells fail to undergo a mitotic response since the histocompatibility surface antigens seen on the stimulator cells are identical to those of the responder cell. (B) Active: When PBNs from an unrelated donor (red) are used as stimulator cells, the responder cells undergo a robust mitotic response due to the mismatched surface histocompatibility antigens.

Furthermore, ASCs exert a potential immunosuppressive function (Fig. 3).³⁹ Studies of human and murine ASCs have documented their ability to release paracrine factors associated with suppression of immune response. These include: indoleamine 2,3 dioxygenase (IDO), a rate limiting enzyme in tryptophan metabolism that is associated with suppression of T lymphocyte expansion; prostaglandin E2 (PGE2) and; transforming growth factor β.^38,40-44 When co-cultured with peripheral blood monocytes, ASCs support the expansion of CD25 positive T-regulatory cells, which in turn are responsible for suppression of immune responses.⁴¹ This interaction can be further enhanced under low oxygen culture conditions in vitro that mimic the physiological oxygen tension with adipose tissue in vivo (T. Frazier, B. Rowan, manuscript in preparation).

Figure 3. Immunosuppression of Mixed Lymphocyte Reactions by Adiopose-Derived Stromal/Stem Cells. While the addition of SVF cells to an active MLR does not alter the mitotic response, the addition of adipose-derived stromal/stem cells (ASC) suppresses proliferation.

Based on these findings, multiple investigators have begun to explore the utility of the ASC’s immunosuppressive properties in autoimmune models. Independent teams have reported positive outcomes in murine models of allergic rhinitis, arthritis, graft vs. host disease, and systemic lupus erythematosis.^45-48 The murine model of multiple sclerosis is experimental autoimmune encephalitis (EAE), where mice injected with a myelin-derived peptide develop motor function loss progressing to limb paralysis and death within a 4 to 5 week period. Injection of ASCs intravenously at the time of exposure to the antigen was found to reduce the clinical severity of EAE as compared with untreated murine controls.⁴⁹ This correlated with increased levels of cytokines associated with Th2 helper cells, such as interleukin 4 and 10 ⁴⁹. Both murine and human SVF cells and ASCs delivered intraperitoneally at the time of antigen exposure offered comparable protection in the murine EAE model (B. Bunnell, manuscript in preparation). Further studies will be required in large animal models prior to clinical translation of these findings

Clinical Applications

The clinical translation of SVF cells and ASCs remains a work in progress.⁵⁰ Nevertheless, there have been some very promising outcomes reported in the peer reviewed literature. One example relates the use of autologous human ASCs to repair craniofacial defects.⁵¹ Finnish collaborators at the Universities of Helsinki and Tampere reported the successful treatment of a hard palate defect in a recovering cancer patient. Loading ASCs derived from the patient’s subcutaneous adipose tissue onto a custom designed ceramic scaffold, they created ectopic bone within an intramuscular implant that was subsequently transplanted to the defect site while maintaining an intact vascular supply. Ultimately, the newly formed hard palate integrated sufficiently to allow the incorporation of dental implants with full recovery of oral function. Independent groups have shown similar outcomes.^52,53

Keith Jacobs, Graduate Student Radiation Oncology Washington University

In a past project I had found evidence that ASCs may differentiate into osteoclasts, despite osteoclasts being part of the hematopoietic and not the mesenchymal lineage. I am therefore curious if you have found any evidence of differentiation of ASCs into cells other than chondrocytes, adipocytes, and osteoblasts.

Dr. Jeffrey Gimble, Professor, Stem Cell Biology Laboratory, Pennington Biomedical Research Institute

We have not looked at osteoclasts. I have wanted to look at that question due to prior findings in my laboratory on hematopoiesis and stromal support function. I think Dr. Adam Katz explored that when he was on the faculty at the University of Virginia and reported these findings at meetings; however, I do not believe the work was published. If I remember correctly, he found an osteoclast population in co-cultures. I don’t know of any data that says that we can transdifferentiate an ASC to the macrophage lineage. We have been able to get our cells in vitro to differentiate along hepatic lineage based on biochemical markers as well as the adipose, bone, and cartilage lineages.⁵⁴ Other groups have reported cardiomyocyte and skeletal myocyte differentiation as well as tendon like cells and epithelial like cells.^55-58 Those are the only lineages I have seen; I have not seen anybody who has consistently shown that the ASC is capable of giving rise to hematopoietic cells. However, the only thing that I think could be consistent with what you are saying is that there is evidence from the adipose biology literature that there are many genes expressed in macrophages that are also expressed in adipocytes. Some of Tom Burris’s early work showed that if you took a macrophage cell and stimulated it with adipogenic ligands, usually used to stimulate adipogenesis, it would start expressing adipocyte like markers such as PPAR gamma or fatty acid binding protein.⁵⁹ So there may be elements of comparability there. Louis Castella’s group I think, would be one group with findings that would support your finding that there is a capability of an ASC differentiating into a myeloid lineage.⁶⁰ But I think we need to see a lot more data before you could get a hematologist to buy into that.

Dr. Linda Sandell, Mildred B. Simon Research Professor of Orthopedic Surgery, Washington University

What is the latest feeling about implanting the cells to form a tissue. Do you want to go as close to the tissue as possible? Like if it is cartilage or bone, or renal cells or whatever. Really functioning tissues or do you want to implant them a little bit earlier so that they can use the microenvironment, as Arnie (Caplan) would say, to further differentiate.

Dr. Gimble

I would think that most people would say that they would want to use a micro environment as much as possible if for no other reason than to get the cells to integrate into their natural surroundings and architecture. I think the best work I have seen recently is out of Mike Longaker’s group at Stanford University where he has been looking at the calvaria model with Ben Levi.^61,62 They implant directly into the defect that they have created in the calvaria above the dura but immediately adjacent to the edges so that the native tissue can integrate with the new tissue.

Dr. Sandell

What about this idea that I see floated around but I haven’t really seen a lot of studies on it, of injecting either MSCs or adipose stem cells into the joint to repair meniscus.

Dr. Gimble

Likewise, I have heard a good deal about that, I think, going back to Frank Barry and Mary Murphy’s work (now at National University of Ireland-Galway) when they were still at Osiris Therapeutics.⁶³ Their studies on the goat used bone marrow derived MSCs and injected them into a damaged knee, seeing what looked like the formation of a new meniscus by cell injection alone. This has led some people to suggest that this approach might serve as a therapeutic. I think we need to see more but I would predict that Frank and Mary probably are generating that kind of data at their institute in Ireland.

Dr. Sandell

I see Mary at every meeting, and often Frank, but you don’t see a whole lot new on it. So I was wondering if anybody else was trying it. Because you don’t see people in the clinic who want to do it and certainly it would be available to do if they wanted to.

Dr. Gimble

Right, I have heard bits and pieces, but again nothing published that I could point out.

Dr. Marc Hammerman, Chromalloy Professor of Renal Diseases in Medicine, Washington University

I have three questions. First, how much variability do you see in the cell populations isolated from liposuction procedures from donor to donor or liposuction site to liposuction site?

Dr. Gimble

We generally look at about 8 or 9 flow cytometry surface markers on our cells. And we have looked most recently at about 60 of our subjects in a paper we published back in 2010 and the numbers are incredibly tight for classical MSC markers like CD73, CD90, CD105 used by the International Society for Cell Therapy definition.^23,24 What is variable is the level of expression of CD34. So some of these markers are variable but others are quite solid between donor to donor.

Dr. Hammerman

Second, do you have any evidence that the human cells engraft following intraperitoneal injection in mice?

Dr. Gimble

I think at the moment I have to say that when you are injecting the cells either IP or IV, or intrathecally, you are probably looking primarily at paracrine function rather than long-term engraftment. It may be that when they are transplanted within a scaffold you are more able to keep them concentrated in one site. This then allows you to track them well enough to say that they are maintaining themselves over a period of time. We have done some studies where we have been able to detect the human cells 12 weeks later, but that has been when the cells were implanted in some kind of scaffold such as alginate.⁶⁴ So I think it is more likely going to be the paracrine mechanisms that are at work. Some of the data looking at cytokines production would be consistent with that.

Dr. Hammerman

Finally, what can you tell us about results of adipocyte cell transplantation conducted outside of the USA regulatory net?

Dr. Gimble

There is quite a bit that is taking place and having been in this field from both a biotech and academic perspective, I would say that over the past 10 y Asia and Europe have really stepped ahead with this process. Some of the most exciting work I know of is the bone work I talked about in Finland.^51,52 There is equally exciting work that’s underway in a project that is in the Orthopedic and Trauma department in Basel, Switzerland under Dr. Ivan Martin.^65,66 He is using these cells to do a clinical long bone fracture model where they are going to conducting a very nicely controlled study comparing large cohorts of ASC treated and untreated patients side by side based on recovery of bone formation. There are a number of studies that are taking place in places like Korea and China in conjunction with Korean companies.^67,68 I think there is quite a bit that is underway that is going to be coming to the FDA's attention over time and it may or may not influence how we move ahead in the US.

Dr. Sanjay Jain, Associate Professor of Medicine and Pathology and Immunology, Washington University

The characterization of these fat cells has focused on those from liposuction. How do the molecular changes you observed in fat cells obtained from liposuction compare with those in neoplastic fat cells: Lipoma, Liposarcoma, Angiomyolipoma?

Dr. Gimble

That’s a good question, I wish I had done the experiments, but unfortunately our IRB protocol was only to look at freshly isolated tissue from healthy subjects. So we have not had the opportunity to do that. We have sent our cells as controls to the soft tissue tumor group at Memorial Sloan Kettering with Sam Singer and he has used it as a control to compare against liposarcomas in a microarray approach and there are differences in the observed expression profiles.⁶⁹ For flow cytometry comparisons, I don’t think he has done that and we certainly haven’t. I would like to but I think that the detailed differences there are probably more likely show up through the microarray approaches that he has been pursuing already.

Conclusion

Adipose tissue is a complex organ that plays a role in energy metabolism, endocrinology, and immunity. It is now appreciated that adipose tissue contains a heterogeneous cell population that can change as a function of obesity and diabetes. The adipose-derived SVF cells and ASCs offer distinct potential advantages for tissue engineering and regenerative medical applications. In part, these rely on the cells’ paracrine factor secretion and immunomodulatory function. Pre-clinical animal models suggest that both SVF cells and ASC will have utility in the treatment of human disorders relating to immune response, such as multiple sclerosis. Further studies addressing regulatory concerns need to be completed prior to clinical translation.

Disclosure of Potential Conflicts of Interest

Glossary

Abbreviations:

ASC

Adipokine, Adipose Stromal/stem Cells

EAE

Experimental Autoimmune Encephalitis

HSPC

Hematopoietic Stem and Progenitor Cell

IDO

Indoleamine 2,3 DiOxygenase

IACUC

Institutional Animal Care and Use Committee

IRB

Institutional Review Board

MSC

Mesenchymal Stem Cell or Multipotent Stromal Cell

POH

Progressive Osseous Heteroplasia

SVF

Stromal Vascular Fraction

UCP1

Uncoupling Protein 1

JMG and XW are the co-owners and co-founders and FS is an employee of LaCell LLC, a commercial biotechnology company focusing on the development of primary human stromal/stem cell products for research applications.

Note

Edited transcripts of research conferences sponsored by Organogenesis and the Washington University George M. O’Brien Center for Kidney Disease Research (P30 DK079333) are published in Organogenesis. These conferences cover organogenesis in all multicellular organisms including research into tissue engineering, artificial organs and organ substitutes and are participated in by faculty at Washington University School of Medicine, St. Louis Missouri USA.