Abstract
Mesenchymal stem cells (MSCs) are well known for their immunomodulatory capabilities. In particular, their immunosuppressive property is believed to permit their allogeneic or even xenogeneic transplantation into immunocompetent recipients without the use of immunosuppressants. Adipose-derived stem cell (ADSC), owing to its ease of isolation from an abundant tissue source, is a promising MSC for the treatment of a wide range of diseases. ADSC has been shown to lack major histocompatibility complex-II expression, and its immunosuppressive effects mediated by prostaglandin E2. Both preclinical and clinical studies have shown that allogeneic transplantation of ADSCs was able to control graft-versus-host disease. In regard to xenotransplantation a total of 27 preclinical studies have been published, with 20 of them performed with the investigators' intent. All 27 studies used ADSCs isolated from humans, possibly due to the wide availability of lipoaspirates. On the other hand, the recipients were mouse in 13 studies, rat in 11, rabbit in 2, and dog in 1. The targeted diseases varied greatly but all showed significant improvements after ADSC xenotransplantation. For clinical application in human medicine, ADSC xenotransplantation offers no obvious advantage over autotransplantation. But in veterinary medicine, xenotransplantation with porcine ADSC is a practical alternative to the costly and inconvenient autotransplantation.
Introduction
F
Early Allogeneic and Xenogeneic Transplantation Studies with BMSC
In year 2000, Liechty et al. [6] reported that human BMSCs exhibited tissue engraftment and site-specific cell differentiation when transplanted into immunocompetent fetal sheep. These authors concluded that BMSCs might possess immunologic properties that allow their persistence in a xenogeneic environment. In 2001, Devine et al. [7] reported that intravenously injected baboon BMSCs were capable of homing to the bone marrow of allogeneic recipients and persisted for at least 76 days. In the following year, a similar group of researchers reported that baboon BMSCs did not elicit a proliferative response from allogeneic lymphocytes in a mixed lymphocyte reaction [8]. Further, an independent group of researchers reported that human BMSCs were not just noninductive but actually capable of suppressing allogeneic T-lymphocyte proliferation [8].
Because the above-mentioned xenotransplantation experiment was performed with fetal recipients, Saito et al. [9] went a step further to test whether xenotransplantation could succeed in fully immunocompetent recipients. Adult rats were IV injected with mouse BMSCs and, 1 week later, underwent coronary artery ligation. Twelve weeks later, mouse cells were found to engraft into the bone marrow cavities of rats with or without myocardial infarction while circulating mouse cells were detected only in rats with myocardial infarction. Moreover, mouse cells were found in the infarcted myocardium where they appeared to differentiate into immature cardiac cells or integrate into newly formed blood vessels. Three years later, a similar group of researchers reported again the successful transplantation of mouse BMSCs into infarcted rat heart [10]. However, 2 other studies, both first-authored by Grinnemo, reported unsuccessful transplantation of human BMSCs into infarcted rat heart [11,12]. The discrepancy could perhaps be explained by differences in immunological properties between human and mouse BMSCs. First, murine BMSCs, unlike their human equivalent, lack major histocompatibility complex (MHC) class II expression [13], and second, T-cell inhibition by BMSCs requires cell contact in mice but is mediated by soluble factors in humans [14,15]. In any event, reports of successful xenotransplantation with BMSCs from various species continue to appear frequently; for example, rat BMSC for bone formation in rabbit [16], human BMSC for spinal cord injury in rat [17,18], and human BMSC for bone formation in mice [19].
ADSC as an Ideal MSC for Therapy
MSCs are increasingly believed to reside in the vasculature [2,20 –23]; therefore, tissues rich in blood vessels, particularly microvessels (capillaries), are ideal for the isolation of MSCs in large quantities for clinical applications. The adipose tissue is endowed with an extensive capillary network [24] and is one of the rare tissues that can be partially removed from a living person without causing harm. In fact, this partial removal is desired by many patients seeking to improve their image and/or health. In addition, its superficial location makes it easily excisable with virtually no health consequence. As such, unlike bone marrow, the removal of which is not only a health risk but also desired by none, the adipose tissue is routinely abandoned as “medical wastes.” As evidence, a worldwide survey published in 2002 shows that between 1994 and 2000 zero death was reported on 66,570 liposuction procedures with a serious adverse event rate of only 0.068% [25]. Further, in the lipoaspirate approximately 2% of nucleated cells can be recovered as MSCs, as compared with 0.002% in the bone marrow aspirate [26]. Thus, while ADSC and BMSC are virtually identical in their therapeutic potential, their difference in clinical applicability is obvious.
In research laboratories, the most commonly used procedure for ADSC isolation involves mincing the adipose tissue sample and centrifugation to separate the fatty content from the stromal vascular fraction (SVF) that forms a reddish pellet at the bottom. The SVF can be directly used for therapy or further processed for the isolation of ADSC. In future clinical applications, adipose tissue mincing will undoubtedly be substituted by liposuction, and the SVF isolation handled by all-in-one devices that are now commercially available [5,27,28]. The adoption of these automation procedures has resulted in the high availability of human ADSCs, which could perhaps explain why many preclinical studies chose human ADSCs for transplantation in non-human animals such as rats and mice, thus, intentionally or unintentionally demonstrating ADSC's xenotransplantation potential (Tables 1 and 2).
Human ADSCs were used as donor cells in all studies.
ADSCs, adipose-derived stem cell; ALDH, aldehyde-3-dehydrogenase; BrdU, 5-bromo-2-deoxyuridine; CFSE, carboxyfluorescein diacetate succinimidyl ester; DAPI, 4′,6-diamidino-2-phenylindole; EBA, endothelial barrier antigen; FISH, fluorescence in situ hybridization; GFAP, glial fibrillary acidic protein; GFP, green fluorescence protein; MAP2, microtubule-associated protein 2; ND, not determined; NG2, oligodendrocyte precursor marker; UnD, undetectable; vWF, Von Willebrand factor.
Human ADSCs were used as donor cells in all studies.
Col-II, collagen-II.
Evidence for ADSC's Immunomodulatory Capacity: Cell Culture Studies
The first report of ADSC's immunomodulatory and immunosuppressive properties appeared in 2005; specifically, its in vitro experiments showed that ADSC did not provoke alloreactivity and was able to suppress mixed lymphocyte reaction [29]. Moreover, the immunosuppressive effect appeared to require cell–cell contact. However, in 2 separate studies the cell–cell contact requirement was corroborated [30] and disputed [31], respectively. Regardless of this disagreement, the immunosuppressive effect of ADSC has been consistently observed in all subsequent relevant studies [32 –36]. Further, in a comparative study, ADSC and BMSC were found to exhibit the same pattern of immunologically relevant surface markers (MHC-I, MHC-II, CD40, and CD40L) [37]. Importantly, both BMSC and ADSC lacked expression of MHC-II, and both did not stimulate allogeneic peripheral blood mononuclear cells. Moreover, these characteristics were retained in both cell types during osteogenic differentiation. As such, it was concluded that allogeneic transplantation of BMSC and ADSC could be employed for tissue engineering [37]. In another study the lack of MHC-II expression in ADSCs was corroborated [31].
While it remains controversial whether cell–cell contact is required for ADSC's immunosuppressive effects [29 –31], several studies have demonstrated the importance of soluble factors, among which the most frequently identified being prostaglandin E2 (PGE2) [31,38 –41]. Specifically, inhibition of PGE2 by indomethacine effectively abolished ADSC's immunosuppressive effects. In addition, specific inhibition of indoleamine 2, 3 dioxygenase [39] or neutralization of leukemia inhibitory factor [42] has also been shown to abolish ADSC's immunosuppressive effects. Further, ADSC's immunosuppressive activity appears to be mediated through an interleukin-6 (IL-6)-dependent inhibition of dendritic cell differentiation and downregulation of MHC-II, CD40, and CD86 on mature dendritic cells [38]. A subsequent study further showed that ADSC was more potent than BMSC in suppressing dendritic cell differentiation and downregulation of costimulatory molecules on the surface of dendritic cells [43].
Rheumatoid arthritis (RA) is due to a loss in immunological self-tolerance that leads to the activation of autoreactive T cells against joint components. In a 2009 study Gonzalez-Rey et al. [44] found that allogeneic ADSCs were able to suppress the antigen-specific response of T cells from patients with RA. Specifically, ADSC inhibited the proliferative response and the production of inflammatory cytokines by collagen-activated CD4 and CD8 T cells. In addition, ADSC treatment significantly increased the numbers of IL-10-producing T cells and monocytes. ADSC also stimulated the generation of regulatory T cells that can suppress collagen-specific T-cell responses. Together, these findings suggest that allogeneic ADSC transplantation could treat RA by suppression of T-cell and inflammatory responses and by generation and/or activation of antigen-specific regulatory T cells. This dual immunomodulatory effect of suppressing overall T-cell proliferation while promoting the generation of regulatory T cells has also been observed more recently in cocultures of allogeneic ADSC and T cells [45].
Th17 lymphocytes are a subset of CD4+ T cells that produce the proinflammatory cytokine IL-17. These cells have been found to play important roles in the pathogenesis of many autoimmune diseases, including RA and systemic lupus erythematosus (SLE). Thus, it has been proposed that controlling Th17 cells or neutralizing IL-17 may offer therapeutic benefits for these autoimmune diseases [46,47]. In 2011, Lai et al. [48] investigated the effects of allogeneic ADSCs on Th17 cells by coculturing ADSCs with peripheral blood mononuclear cells of SLE patients. The results showed that ADSCs from passage 3 decreased the proportion of Th17 cells and suppressed their production of IL-17; however, ADSCs from passage 8 had the opposite effects. Thus, while allogeneic ADSC may have therapeutic potential toward SLE, their prolonged culturing should be avoided.
The above-mentioned studies suggest that, due to its immunomodulatory capability, ADSC might be suitable for allogeneic transplantation for the treatment of various diseases. However, a recent study showed that, despite being immunosuppressive, ADSCs were susceptible to lysis by allogeneic CD8+ T cells and NK cells [49]. Indeed, in an earlier study these authors also showed that ADSCs induced explosive T-cell proliferation [50]. Thus, whether allogeneic transplantation of ADSC offers therapeutic benefits requires further testing, especially in preclinical and clinical settings.
Immunomodulatory Therapy
In 2006, Yanez et al. [33] reported that allogeneically transplanted ADSCs were able to control experimentally induced graft-versus-host disease (GVHD) in mice. In the same year, Fang et al. [51] reported successful treatment of GVHD in a female patient by using ADSC isolated from an unrelated male donor. In the following year, Fang et al. reported a total of 7 cases of successful GVHD treatment with allogeneic ADSCs in 9 patients [52 –54]. Two additional years later Fang et al. published 2 other clinical studies. In one study, a male patient was successfully treated with ADSC isolated from his brother, resulting in the resolution of refractory chronic autoimmune thrombocytopenic purpura [55]. In the other study, 2 patients were successfully treated with ADSCs isolated from unrelated donors, resulting in the resolution of refractory pure red cell aplasia due to major ABO-incompatible hematopoietic stem cell transplantation [56].
Allotransplantation
In a canine spinal cord injury model, injection of allogeneic ADSC into the injured site resulted in significant improvement in both hind limb function and nerve conduction [57]. Histological examination identified expression of neural markers GFAP, Tuj-1, and NF160 in the transplanted cells, suggesting neural differentiation. In another study, allogeneic ADSC seeded on a biomaterial scaffold were found to accelerate spinal fusion in a rat model of lumbar compression fracture [58]. In the recipient rats T-cell priming was undetectable, but significant antibody responses were observed [59]. However, the antibodies were determined to be noncytotoxic and thus not expected to impede the prospective implementation of allogeneic ADSC for spinal fusion. In an allergic rhinitis mouse model IV injected ADSCs were found to migrate to the nasal mucosa, reduce allergic symptoms, and inhibit eosinophilic inflammation in the nasal mucosa [60]. ADSCs also significantly decreased the allergen-specific IgE level and IgG1/IgG2a ratio, suggesting the inhibition of eosinophilic inflammation was due to a shift from a Th2 immune response to a T-helper response.
Xenotransplantation
A total of 27 studies have performed xenotransplantation with ADSCs. Twenty of these studies mentioned the investigators' intent to conduct xenotransplantation while the other 7 did not (Tables 1 and 2). Without exception, all 27 studies used ADSCs isolated from humans, possibly due to the wide availability of lipoaspirates. On the other hand, the recipients were mouse in 13 studies, rat in 11, rabbit in 2, and dog in 1.
The first ADSC xenotransplantation study was published in 2003, in which intracerebral transplantation of human ADSCs was found to improve neurological functions in a cerebral ischemic rat model [61]. Interestingly, despite being a xenogeneic transplantation, the investigators observed no evidence of inflammation or rejection. They thus offered several explanations including the brain being partially immunoprivileged and ADSCs being lacking MHC-II expression. Two years later, another study conducted a more detailed examination of ADSC's potential for xenogeneic transplantation [62]. First, in cell culture experiments human ADSCs were shown not to induce proliferation of murine splenocytes; then, in animal transplantation experiments human ADSCs were found not to cause murine CD3 lymphocyte infiltration. This study's main purpose, however, was to show that ADSC had the potential for treating muscular dystrophy. Specifically, transplantation of human ADSCs into a murine model of muscular dystrophy resulted in high-level expression of human dystrophin and long-term engraftment of the transplanted cells. In a more recent study of using human ADSCs to treat experimentally induced atrial injury in dogs, intravenous administration of human ADSCs caused virtually no changes in the composition of peripheral blood lymphocytes of the recipient dogs, thus indicating immunocompatibility [63].
The feasibility of using ADSC (and BMSC) for xenogeneic transplantation was specifically investigated in a 2008 study [64]. Undifferentiated and osteogenically differentiated ADSCs (and BMSCs) were subcutaneously injected into immunocompetent mice and then tracked at 4 and 8 weeks postinjection. The results show that undifferentiated ADSCs/BMSCs survived for at least 8 weeks while osteogenically differentiated ADSCs/BMSCs were eliminated at 4 weeks. The authors thus concluded that undifferentiated ADSCs/BMSCs were suitable for xenogeneic transplantation. However, in another study differentiated insulin-producing cells from human ADSCs were found to survive for a remarkable period of 1 year following subrenal capsule transplantation into streptozotocin-induced diabetic mice [65]. Thus, it appears that under certain conditions both undifferentiated and differentiated ADSCs could be used for xenotransplantation.
Prospective Clinical Application of Allogeneic and Xenogeneic ADSCs
Liposuction is a well-established clinical practice in human medicine, and processing the liposuction material into SVF can be done expeditiously with commercially available devices [5,27,28]. Therefore, the future application of ADSCs in human medicine will be conducted mostly, if not exclusively, in an autologous fashion. On the other hand, liposuction is not a standard procedure in veterinary medicine, and the manual isolation of SVF or ADSC from each dog or cat is an impractical proposition for most veterinary clinics. Thus, the allogeneic or xenogeneic application of ADSCs in veterinary medicine is worthy of consideration.
In the mainstream media and in the Internet there have been thousands of claims about ADSC's “miraculous” therapeutic efficacy in treating animal diseases, especially canine OA. In the scientific literature, there have been 3 studies that used autologous ADSCs to treat canine OA [66 –68], and the results all indicated ADSC's efficacy in ameliorating OA symptoms. However, the adoption of this novel OA treatment into veterinarian practice faces many challenges, including (1) most veterinary clinics lack the equipment and expertise for ADSC isolation, (2) excision of adipose tissue causes donor site morbidity, (3) individually made ADSC isolation is costly and time-consuming, and (4) at least 2 veterinarian appointments are needed for adipose tissue procurement and ADSC injection. However, these obstacles can be overcome if the therapeutic ADSCs are from an allogeneic or xenogeneic source. For example, canine and porcine ADSCs can be prepared in commercial-scale quantities, portions of which are stored in liquid nitrogen or further propagated. Upon receiving an order from a vet clinic, the cells can be shipped in a syringe via an express courier; and upon its arrival, the cell preparation can be injected by the veterinarian into the diseased joint of a patient dog. Thus, there is no need for the veterinary clinic to purchase expensive equipment or hire cell-isolation technicians. The demand on the veterinarian is minimal as well.
From an immunological point of view, allotransplantation is perhaps a better choice than xenotransplantation. However, from an ethical point of view, the harvest of canine tissues for commercial purpose is definitely less acceptable than the harvest of porcine tissues. Thus, xenotransplantation of porcine ADSC for veterinary uses is expected to have a better chance to succeed. In addition, the practice of porcine organ transplantation in humans has been extensively investigated with the establishment of strict guidelines [69]. Thus, the breeding of donor pigs and the harvest of their adipose tissue can follow these established guidelines. It should be further pointed out that, in more than 2 centuries of investigation there has been no documentation of transfer of viruses from donor pig tissues to recipient humans [70]. Thus, it is reasonable to expect that porcine ADSC will be safe for transplantation into dogs and cats. If so, xenotransplantation of porcine ADSC should provide great therapeutic benefits to our best friends.
Footnotes
Author Disclosure Statement
No competing financial interests exist.
