- Top of page
- MATERIALS AND METHODS
Background: Adipose tissue–derived stem cells (ADSCs) can be easily obtained from subcutaneous adipose tissue, and ADSCs can be demonstrated to display multilineage developmental plasticity. In this study, using TNBS-induced colitis rats, we show the feasibility of repairing injured intestinal mucosa with adipose tissue–derived stem cells.
Methods: The subcutaneous adipose tissue of F344 rats was obtained and digested by collagenase. The digested tissue was cultured in DMEM containing 10% FBS for 1 month. ADSCs were confirmed to differentiate under appropriate conditions into various lineages of cells, including bone, neural cells, adipocytes, and epithelial cells. HGF, VEGF, TGF-β, and adiponectin in the culture supernatants of ADSCs were determined by ELISA. ADSCs (107 cells) were injected into the submucosa of the colon to examine their capacity to repair intestinal mucosa injured by TNBS.
Results: In the experimental colitis model, the injection of ADSCs facilitated colonic mucosal repair and reduced the infiltration of inflammatory cells. High levels of HGF, VEGF, and adiponectin were detected in the culture supernatants of ADSCs. Moreover, injected ADSCs distributed to several layers of the colon, and some of them differentiated into mesodermal lineage cells.
Conclusions: ADSCs can accelerate the regeneration of injured regions in experimental colitis. HGF, VEGF, and adiponectin might be responsible for the regeneration of injured regions in the colon.
Crohn's disease is characterized by chronic relapsing inflammation of the gastrointestinal tract. Evidence suggesting that various immune, genetic, and environmental factors influence both the initiation and progression of colitis has been accumulated. However, the precise mechanism or mechanisms underlying the development of this disease have yet to be clarified.
In recent years, several groups have reported the ubiquitous distribution of adult stem cells in various tissues and organs, including bone marrow, muscle, brain, skin, and, more recently, even in subcutaneous fat.1 Among these adult stem cells, those in the subcutaneous fat, termed adipose tissue–derived stem cells (ADSCs), can be easily obtained with a relatively lower burden on donors; they can be easily harvested from subcutaneous adipose tissue by lipoaspiration. These ADSCs have been demonstrated to display multilineage developmental plasticity.2 Furthermore, ADSCs have been reported to have less heterogeneity in their immunophenotype and multilineage differentiation ability3, 4 than do bone marrow–derived mesenchymal stem cells. Because of these advantages, clinical use of ADSCs for not only fat, bone, and myocardium reproduction but also spinal cord regeneration, vascularization, and the treatment of intractable ulcers has been carried out, and the development of further clinical applications is expected. Garcia-Olmo et al5 reported in a phase I clinical trial of Crohn's disease that after an operation to close fistula, cell therapy using autologous ADSCs can facilitate fistula repair. However, ADSCs may have, in general, a potential to treat Crohn's disease and inflammatory bowel disease (IBD) in addition to the treatment for fistula observed in a subpopulation of patients with Crohn's disease. Therefore, the potential ability of ADSCs to improve mucosal healing of inflammatory lesions and the biological mechanism underlying the repair function of ADSCs should be examined. In the present study, using Crohn's disease models in rats in which the colonic mucosa is injured by TNBS, we attempted to treat the mucosal injury by a submucosal injection of ADSCs to facilitate mucosal recovery; we expected this to be a simple and safe procedure. It is noted that ADSCs injected intravenously might be trapped by the reticuloendothelial system in the liver or lung6; therefore, they could not be recruited to the intestinal mucosa. Thus, in this study, ADSCs were directly injected into the submucosa, and ADSCs actually facilitated the regeneration of the injured region (by TNBS injection). Furthermore, we also examined the potency of ADSCs to differentiate into various lineage cells and their production of growth factors such as HGF, VEGF, TGF-β, and adiponectin that may accelerate mucosal regeneration.
- Top of page
- MATERIALS AND METHODS
In the present study, we have shown that ADSCs can facilitate colonic mucosal repair and reduce the infiltration of inflammatory cells. The ADSCs used in the present study were characterized as CD90+CD45− cells,15, 16 being similar to bone marrow–derived mesenchymal stem cells (MSCs),1, 3, 4, 7 and they were cytohistochemically confirmed to be functional MSCs (Figs. 4 and 5) because of their capacity to differentiate into cells with adipogenic, osteogenic, neurogenic, and epitheliogenic lineages when lineage-specific induction factors have been added to the culture, as has been previously reported.2, 9, 17, 18 Therefore, the ADSCs used in the present study have similar features to MSCs from the bone marrow. The proliferative capacity of the ADSCs was also similar to that of MSCs.4, 7, 15, 16 The doubling time of ADSCs remained unchanged until 10 passages (Fig. 3). Therefore, along with the ADSCs being easily obtained by lipoaspiration,19 ADSCs can be a useful candidate for the source of cell therapy.
In addition, one advantage of ADSCs is their potential to secrete growth factors, such as VEGF, HGF,20 and adiponectin,21 which facilitate the regeneration of injured tissue alone or synergistically.22 As has been reported, HGF regulates cell growth, motility, and morphogenesis of various types of cells, including epithelial cells and endothelial cells, and it also prevents fibrosis.23, 24 Recent studies have shown that the combination of HGF with VEGF increases neovascularization in the rat corneal assay25 and that HGF facilitates the repair of large colonic ulcers in TNBS-induced colitis in rats.26 Adiponectin has been reported to not only improve insulin resistance and prevent atherosclerosis, fatty liver, and liver fibrosis but also to exert several anti-inflammatory effects.27–30 Moreover, it has been reported that adiponectin reduces the attachment of monocytes to the endothelium by down-regulating the expression of vascular cell adhesion molecule–1, intercellular adhesion molecule-1, and E-selectin27 and that it inhibits phagocytic activity and production of TNF-α and IL-6 from cultured macrophages.31 These results indicate that HGF, VEGF, and adiponectin play a crucial role in intestinal mucosal wound healing. In the present study, significant amounts of HGF, VEGF, and adiponectin were detected in the culture supernatants of ADSCs (Fig. 6), indicating that ADSCs injected into the submucosa may secrete these cytokines in situ, resulting in the acceleration of regeneration of wound mucosa and also inhibition of the inflammation observed in TNBS-induced colitis. This possibility was actually confirmed by the in vivo examination in which ADSCs facilitated the repair of colonic ulcers (Fig. 8) and reduced inflammatory responses along with TNBS-induced colitis (Fig. 9). There was no statistically significant differences in colon length between the group submucosally injected with ADSCs and the group injected with PBS. This might have been a result of the large ulcer being detected in a limited area of colon in TNBS-induced colitis-model rats in contrast to pan-colitis-model rats.
However, the decreases in the levels of IL-8 in the colonic tissue after the injection of ADSCs clearly indicate the wound-healing effects of ADSCs (Fig. 9), as also shown in the analyses of colon weight as an index of edema (Fig. 7). Furthermore, it is noted that the reduced activity of MPO observed in the colonic tissue confirms decreased infiltration of neutrophils and strengthens the effect of ADSCs. From these findings, it can be speculated that ADSCs prevent both the accumulation and the activation of neutrophils, thereby attenuating the neutrophil-dominant inflammation in colonic tissue caused by TNBS through the reduction of IL-8 levels that play a pivotal role in the accumulation of circulating leukocytes in inflammatory foci.32 Furthermore, the effect of ADSCs might also be attributable to the anti-inflammatory effect of adiponectin, as shown in the previous study, in which adiponectin was found to exert its anti-inflammatory effect by inhibiting IL-8 production.30 In addition, adiponectin promotes angiogenesis in response to tissue ischemia.33 These data suggest that adiponectin secreted by ADSCs may also play a crucial role in TNBS-induced colitis.
As shown in Figure 10, cells derived from ADSCs, defined by Y-FISH, were detected in the mucosal layer under the basal membrane, the submucosal layer around the vessel, and the muscle layer; they were especially observed in the area of inflammation, such as the ulcer edge. They could be thought to differentiate into fibroblast-lineage (vimentin-positive cells), adipocyte-lineage (S-100-positive cells), and muscle-lineage cells (SMA-positive cells) when immunohistochemically examined. However, they were not detected in the epithelial layer, indicating that ADSCs (submucosally injected) did not differentiate into epithelial-lineage or neural-lineage cells, even though they differentiated into various lineage cells in vitro (Figs. 4 and 5). This is in contrast with the findings that bone marrow MSCs were detected in the epithelium of the stomach and intestine34 and differentiated into neural cells35 and that MSCs in bone marrow seem to be more primitive than ADSCs. Therefore, our results indicate that although ADSCs have the potential to differentiate into various lineage cells in vitro, ADSCs in vivo can differentiate into mesodermal lineage cells.
Examination in vitro showed that collectively, ADSCs have multiple differentiation ability and secrete growth factors such as VEGF and HGF in large quantities and also produce adiponectin. Therefore, although we have not measured the tissue levels of these growth factors after the injection of ADSCs, it is highly feasible that ADSCs injected in vivo (1 × 107 cells) produce them and that these soluble factors might be responsible for the regeneration of the injured regions observed in the TNBS-induced colitis. Moreover, some ADSCs may differentiate into the various components of the colon, such as smooth muscles, fibroblasts, and myofibroblasts. These mesodermal cells are essential for facilitating the regeneration of epithelial cells. Furthermore, ADSCs improved TNBS-induced colitis by ameliorating colonic injury induced by proinflammatory cytokines.
Importantly, the safety of high-dose lipoaspiration to obtain ADSCs in humans was already established,36 and thus ADSCs are a viable therapeutic option for amelioration of Crohn's disease or IBD by repairing injured intestinal mucosa. If the number of ADSCs obtained from the subcutaneous adipose tissues of the abdomens of patients themselves is enough to use for the cell therapy, ADSCs can be immediately applicable without cell culture, and the remaining ADSCs can be kept by cryopreservation for future use when a fistula recurs. In addition, treatment via endoscopy is considered a general and established method, thereby submucosally injecting ADSCs without an operation. It is feasible that in vitro cultured ADSCs inoculated into the submucosa are transformed into tumors. Therefore, we have to examine this possibility for a long life span after the treatment, and these types of experiments are now under investigation.
In conclusion, ADSCs can accelerate the regeneration of injured regions, and HGF, VEGF, and adiponectin produced by ADSCs might be responsible for the regeneration.