**4. Neuroprotection by ASCs — Demyelination, stroke, spinal cord injury**

Early translational studies do suggest that ASCs can be safely administered to nervous tissue injuries and that functional improvement is noted. Transplanted ASCs have been reported to improve functional deficits following middle cerebral occlusion or ischemic stroke [152]-[154], spinal cord contusion injury [155] and peripheral nerve gaps [156], [157]. Histologic analysis of these injury sites has suggested that ASC differentiation into neurons and/or glial cells may play a role in the functional recovery, with transplanted cells staining positively for MAP2 [153], GFAP, Tuj-1 and an oligodendrocyte marker [155]. However, this functional improve‐ ment may be due to paracrine actions on the host more than ASC differentiation, as less then 1% of transplanted ASCs can be found within a spinal contusive injury model, with those remaining appearing to be oligodendrocytes [158]. In addition, extremely low levels ASC differentiation into mature neurons is noted in a model of cerebral cortex injury [159]. However, both of these studies note significant changes in the host tissue with Nakada et al. observing improvements in microvasculature and Zhang et al. measuring increases in host oligodendrocyte formation. Therefore, like wound healing and IR models, ASCs are likely to exert paracrine actions within nervous tissue.

In 2002, Zhao et al. suggested that functional recovery in ischemic brain injury was not due to MSC differentiation but to secreted paracrine factors that act on the host [160]. A similar hypothesis has been put forth by bone marrow MSC groups who have noted increased survival and differentiation of Tuj1+ve neurons and neuroblastoma cells in co-cultures [161] and increased neuronal viability and glial cell differentiation using MSC conditioned media [162]. Consistent with this, ASC/Matrigel constructs implanted into models of mice limb re-inner‐ vation stimulate the regeneration of nerves and induce axon growth, likely through the expression numerous neurotrophins [163]. Moreover, enhanced nerve fiber growth is observed if the ASCs are pre-induced toward the neural lineage thus enhancing their production of brain-derived neurotrophic factor (BNDF). BDNF secretion (together with nerve growth factor/NGF and glial cell-derived neurotrophic factor/GDNF) by ASCs pre-differentiated toward a Schwann Cell (SC) phenotype is thought to be the basis for axonal regeneration in sciatic nerve gap models - although these authors speculate that this regeneration is likely due to the neuroprotective function of these three neurotrophins [164]. In support of this, studies using ASC-conditioned media appeared to further strengthen this theory. Protection against cortical and hippocampal volume loss in rats can be achieved through the infusion of ASCconditioned medium [165]. ASC-conditioned medium containing VEGF, BDNF and NGF is shown to have a protective effect against glutamate excitotoxicity on PC12 cells (a key factor implicated in stroke and neurodegenerative diseases) and increase PC12 viability 94. Condi‐ tioned media from pre-differentiated ASCs infused over one week into a rat model of ischemic stroke 8 days after stroke induction increases the number of CD31+ve cells [166]. Finally, functional deficits in a model of middle cerebral artery occlusion can be dramatically improved using ASC transduced to overexpress BDNF [153].

While these neurotrophic factors may act to protect neurons, ASCs may also play roles in decreasing inflammation and gliosis (i.e. glial cell-mediated scar formation) – two critical events that specifically affect healing in the both the central and peripheral nervous system. Systemic transplantation of human ASCs can attenuate cerebral degeneration in rats, reducing both brain atrophy and glial proliferation [79]. Rats implanted with ASC-derived SCs show significant locomotor function recovery compared with untreated ASCs and also reduction in gliosis [152]. Pre-differentiated canine ASCs in Matrigel scaffolds show better functional recovery and reduced fibrosis and inflammation when implanted into spinal cord injuries [167]. Decreased gliosis is also noted upon intrathecal administration of ASCs in a model of IR neuronal damage in rabbits – an event accompanied by increased expression of BDNF within the first 72 hours following ASCs administration [168]. Finally, a possible anti-inflammatory role for ASCs in sciatic nerve repair might be seen in a recent model describing possible immunosuppression of xenogeneic acellular nerve matrices combined with autologous ASCs [169]. Implantation of this construct does not result in host rejection, making it possible that peripheral nerves repair can be accomplished using commercial nerve matrices combined with the patient's own ASCs.

#### **4.1. Controlled release from ASCs — ASCs as a cellular biopump**

**4. Neuroprotection by ASCs — Demyelination, stroke, spinal cord injury**

Early translational studies do suggest that ASCs can be safely administered to nervous tissue injuries and that functional improvement is noted. Transplanted ASCs have been reported to improve functional deficits following middle cerebral occlusion or ischemic stroke [152]-[154], spinal cord contusion injury [155] and peripheral nerve gaps [156], [157]. Histologic analysis of these injury sites has suggested that ASC differentiation into neurons and/or glial cells may play a role in the functional recovery, with transplanted cells staining positively for MAP2 [153], GFAP, Tuj-1 and an oligodendrocyte marker [155]. However, this functional improve‐ ment may be due to paracrine actions on the host more than ASC differentiation, as less then 1% of transplanted ASCs can be found within a spinal contusive injury model, with those remaining appearing to be oligodendrocytes [158]. In addition, extremely low levels ASC differentiation into mature neurons is noted in a model of cerebral cortex injury [159]. However, both of these studies note significant changes in the host tissue with Nakada et al. observing improvements in microvasculature and Zhang et al. measuring increases in host oligodendrocyte formation. Therefore, like wound healing and IR models, ASCs are likely to

In 2002, Zhao et al. suggested that functional recovery in ischemic brain injury was not due to MSC differentiation but to secreted paracrine factors that act on the host [160]. A similar hypothesis has been put forth by bone marrow MSC groups who have noted increased survival and differentiation of Tuj1+ve neurons and neuroblastoma cells in co-cultures [161] and increased neuronal viability and glial cell differentiation using MSC conditioned media [162]. Consistent with this, ASC/Matrigel constructs implanted into models of mice limb re-inner‐ vation stimulate the regeneration of nerves and induce axon growth, likely through the expression numerous neurotrophins [163]. Moreover, enhanced nerve fiber growth is observed if the ASCs are pre-induced toward the neural lineage thus enhancing their production of brain-derived neurotrophic factor (BNDF). BDNF secretion (together with nerve growth factor/NGF and glial cell-derived neurotrophic factor/GDNF) by ASCs pre-differentiated toward a Schwann Cell (SC) phenotype is thought to be the basis for axonal regeneration in sciatic nerve gap models - although these authors speculate that this regeneration is likely due to the neuroprotective function of these three neurotrophins [164]. In support of this, studies using ASC-conditioned media appeared to further strengthen this theory. Protection against cortical and hippocampal volume loss in rats can be achieved through the infusion of ASCconditioned medium [165]. ASC-conditioned medium containing VEGF, BDNF and NGF is shown to have a protective effect against glutamate excitotoxicity on PC12 cells (a key factor implicated in stroke and neurodegenerative diseases) and increase PC12 viability 94. Condi‐ tioned media from pre-differentiated ASCs infused over one week into a rat model of ischemic stroke 8 days after stroke induction increases the number of CD31+ve cells [166]. Finally, functional deficits in a model of middle cerebral artery occlusion can be dramatically improved

While these neurotrophic factors may act to protect neurons, ASCs may also play roles in decreasing inflammation and gliosis (i.e. glial cell-mediated scar formation) – two critical

exert paracrine actions within nervous tissue.

60 Regenerative Medicine and Tissue Engineering

using ASC transduced to overexpress BDNF [153].

It is possible that the paracrine action of ASCs may be "fine-tuned" so that the ASC secretes a desired factor, hence turning the ASC into a "cellular biopump". This is not a recent concept as the engineering of numerous cell types to secrete a variety of factors has been reported in the literature for over a decade. In the field of stem cell research, bone marrow MSCs have been modified to secrete various factors, including BMP2 [170], [171], bFGF [172], IFN-β [173] and IL12 [174]. Similar to these studies, ASCs have been engineered for the delivery of BMP4 [175], BMP2 [176], [177], and BMP6 [178] in several bone regeneration models. Delivery of TGFβ2 by ASCs for the induction of chondrogenesis has been reported [179]. Adenovirallymediated VEGF secretion by ASCs has been used to induce vascular growth in a bone defect model [180] and adipose tissue grafts [181]. Finally, as described above, BDNF delivery by transduced ASCs into a model of middle cerebral artery occlusion improves functional deficits when compared to control ASCs [153].

However, a more exciting idea might be in the engineering of ASCs in the treatment of disease. In 2007, ASCs engineered to express cytosine deaminase were found to decrease the growth of colon carcinoma cells [182]. ASCs have recently been described in the delivery of an oncolytic myxoma virus that will specifically target gliomas [183]. ASC viability is not impacted with transduction and successful cross-infection of gliomablastoma cells is observed upon 3D coculture with glioblastoma cells, leading to their cell death. More importantly, rat survival is increased with this myxoma virus delivery, with the size of the gliomas significantly decreas‐ ing upon injection of transduced ASCs in comparison to non-transduced ASCs controls. Localization of ASCs and increased apoptosis within tumors has also been reported following intravenous or subcutaneous injection of ASCs engineered to express TRAIL, having no effect on the surrounding healthy tissue [184]. Finally, this approach may have far-reaching effects on autoimmune diseases through the delivery of interleukins and interferons. ASCs engi‐ neered to overexpress IL4 and administered at the time of T cell priming attenuate autoimmune encephalomyelitis and reduce peripheral T cell responses shifting the host pro-inflammatory response to an anti-inflammatory one [71]. With the development of inducible viral systems, there is the possibility that the ASC cellular biopump could be controlled not only at the dose level through the number of cells delivered but at the temporal level, giving clinicians more precise control over their therapeutic regimen.

#### **4.2. ASC uses in the clinic**

In light of their differentiative capacity and paracrine actions, there is great interest in the use of ASCs within the clinic. As source of regenerative stem cells, the ASC may have no equal. Bone marrow aspirates yield on average 6x106 nucleated cells per ml, of which, only 0.001 to 0.01% are thought to be stem cells [185], [186]. In comparison, approximately three-fold more cells can be obtained per gram adipose tissue [187] [188] with 10% of these cells thought to be stem cells [188], [189]. The abundance of ASCs within adipose tissue, combined with the relative ease of its harvest and isolation also makes the ASC a good choice for clinical work. Patient's could conceivably have their adipose tissue harvested relatively painlessly a few weeks prior to their procedure in a simple outpatient procedure, the ASCs isolated and expanded under good manufacturing protocols and then used for regenerative purposes. With the confirmed absence of HLA/MHC class II proteins and continuing xenogeneic animal models, the patient may not even need to use their own stem cells. Donated allogeneic ASC lines could be used in lieu of autologous cells without the fear of immunorejection or inflam‐ matory complications. Such a situation might be perfect in the case of myocardial infarct treatment where a delay in treatment could have serious consequences.

The first published article using ASCs in a clinical setting was in 2004, in which freshly harvested SVF cells were combined with fibrin glue and used in the repair of a traumatic calvarial injury [190]. Three months after reconstruction, CT scans showed new bone formation within the injury. However, it is important to point out that the cells used in this study were not ASCs, purified through plastic adherence and culture time, but the SVF - a heterogenous mixture of ASCs, endothelial cells, pre-adipocytes, pericytes, fibroblasts and red blood cells. Therefore, it is difficult to attribute the observed healing to the action of the ASC itself. Since that time, other clinical studies using the SVF have been attempted [191] and a review by Casteilla et al. does an excellent job of summarizing these works [192]. It is worth noting that with the exception of some cysts and microcalcifications being observed upon breast recon‐ struction [193], the use of SVFs clinically has not resulted in any serious complications.

Because of its heterogeneity, clinical studies using purified ASCs have also been performed for the treatment of such disorders as critical limb ischemia and radiation therapy ([194], [195] – for a more comprehensive review, see [192]). Bone regeneration using ASCs has recently been reported in 2009 with the reconstruction of the maxilla being induced using ASC in combina‐ tion withBMP2 [196].BonyhealingusingBMPshas beendocumentedinnumerous translation‐ al animal models [197]-[201], making this clinical study an exciting addition to the ways bone regenerationandhealingcanbebroughtaboutinthe clinic.However,manyofthese translation‐ al models fail to report the appropriate control – the amount of bone being formed just by the BMPitself.Thefirsttranslational studytocombineASCsandaBMP(i.e.BMP2)failedtomeasure any significant improvement in bone formation when BMP2 and ASC+BMP2 groups were compared [197]. Since this study, others have appeared to suggest that BMP2 may not pro‐ mote the in vivo osteogneic capacity of the ASC [202] but may, in fact, may have a deleterious effect on bone regeneration [203]. Since it is not possible to perform similarly controlled studies clinically, it remains unknown if the addition of ASCs to BMP-treated scaffolds provides any more advantage. However, It is worth noting that, as with the use of SVFs, administration of ASCs into human patients has not been associated with any adverse effects [204].

response to an anti-inflammatory one [71]. With the development of inducible viral systems, there is the possibility that the ASC cellular biopump could be controlled not only at the dose level through the number of cells delivered but at the temporal level, giving clinicians more

In light of their differentiative capacity and paracrine actions, there is great interest in the use of ASCs within the clinic. As source of regenerative stem cells, the ASC may have no equal. Bone marrow aspirates yield on average 6x106 nucleated cells per ml, of which, only 0.001 to 0.01% are thought to be stem cells [185], [186]. In comparison, approximately three-fold more cells can be obtained per gram adipose tissue [187] [188] with 10% of these cells thought to be stem cells [188], [189]. The abundance of ASCs within adipose tissue, combined with the relative ease of its harvest and isolation also makes the ASC a good choice for clinical work. Patient's could conceivably have their adipose tissue harvested relatively painlessly a few weeks prior to their procedure in a simple outpatient procedure, the ASCs isolated and expanded under good manufacturing protocols and then used for regenerative purposes. With the confirmed absence of HLA/MHC class II proteins and continuing xenogeneic animal models, the patient may not even need to use their own stem cells. Donated allogeneic ASC lines could be used in lieu of autologous cells without the fear of immunorejection or inflam‐ matory complications. Such a situation might be perfect in the case of myocardial infarct

The first published article using ASCs in a clinical setting was in 2004, in which freshly harvested SVF cells were combined with fibrin glue and used in the repair of a traumatic calvarial injury [190]. Three months after reconstruction, CT scans showed new bone formation within the injury. However, it is important to point out that the cells used in this study were not ASCs, purified through plastic adherence and culture time, but the SVF - a heterogenous mixture of ASCs, endothelial cells, pre-adipocytes, pericytes, fibroblasts and red blood cells. Therefore, it is difficult to attribute the observed healing to the action of the ASC itself. Since that time, other clinical studies using the SVF have been attempted [191] and a review by Casteilla et al. does an excellent job of summarizing these works [192]. It is worth noting that with the exception of some cysts and microcalcifications being observed upon breast recon‐ struction [193], the use of SVFs clinically has not resulted in any serious complications.

Because of its heterogeneity, clinical studies using purified ASCs have also been performed for the treatment of such disorders as critical limb ischemia and radiation therapy ([194], [195] – for a more comprehensive review, see [192]). Bone regeneration using ASCs has recently been reported in 2009 with the reconstruction of the maxilla being induced using ASC in combina‐ tion withBMP2 [196].BonyhealingusingBMPshas beendocumentedinnumerous translation‐ al animal models [197]-[201], making this clinical study an exciting addition to the ways bone regenerationandhealingcanbebroughtaboutinthe clinic.However,manyofthese translation‐ al models fail to report the appropriate control – the amount of bone being formed just by the BMPitself.Thefirsttranslational studytocombineASCsandaBMP(i.e.BMP2)failedtomeasure any significant improvement in bone formation when BMP2 and ASC+BMP2 groups were

treatment where a delay in treatment could have serious consequences.

precise control over their therapeutic regimen.

**4.2. ASC uses in the clinic**

62 Regenerative Medicine and Tissue Engineering

The first phase I clinical trials using ASCs were not conducted on bone formation or even fat grafting but in the healing of chronic fistulae in Crohn's disease [205]-[210]. In 2005, nine rectovaginal fistulae in four patients were treated with ASCs, purified and cultured for up to one month. Of the eight fistulae examined, six showed complete healing in 8 weeks [206]. These fistulae had previously failed to heal using conventional surgical treatments, thus justifying progression to more comprehensive phase II trials. In 2009, a larger phase II trial using patients with and without Crohn's fistulae were treated with ASCs [211]. As seen with their earlier clinical trial, the majority of Crohn's and non-Crohn's fistulae were healed completely using ASCs in comparison to controls. Currently, there are three phase II clinical trials recruiting for the use of ASCs in Crohn's disease fistulae (Clinicaltrials identifiers: NCT01011244, NCT01157650, NCT00999115, http://clinicaltrials.gov/ct2/results?term=adipose+derived +cells), in addition to one phase III trial (NCT00475410) recently completed [212].

One of the reasons ASCs are considered in the treatment of Crohn's disease is their ability to suppress inflammation. This review includes numerous examples of how the ASC may be capable of suppressing the immune system and recent clinical trials have attempted to take direct advantage of this quality. The treatment of multiple sclerosis (MS) with SVFs, containing ASCs, has been described by Riordan and colleagues in 2009, with the 3 enrolled patients showing improvement in numerous functional categories including balance and coordination [213]. The use of culture expanded ASCs in autoimmune diseases like hearing loss, MS and rheumatoid arthritis was recently discussed in 2011 [214]. Prior to this, ASCs have been proposed as a viable therapy for suppression of graft vs. host disease (GVHD) [215]-[218]. Each of these studies report favorable functional outcomes and propose ASCs, or their SVF coun‐ terpart, for the treatment of immune system disorders.

The most obvious application of the ASC clinically should be in breast reconstruction. In the lab, the combination of ASC-containing SVFs with fat grafts through a protocol called cellassisted lipotransfer has enjoyed success [151]. Clinically, treatment of facial lipoatrophy has been reported [219] and two recent trials overseas has suggested that the ASCs within the SVF are capable of increasing breast volume and improving contour 6 months post-surgery [193], [220]. However, the use of ASCs in breast reconstruction is being pursued carefully in light of recent findings that link stem cells to cancer. Bone marrow MSCs have been found to increase proliferation of breast cancer cell lines [221] and subcutaneous injection of MSCs with tumor cells can favor their growth [222]. Similar to this, ASCs can increase tumorigenesis of estab‐ lished breast cancer lines [223]. In this study, ASCs not only promote the growth of metastatic pleural effusion cells both in vitro and in vivo but the ASC also secretes adipsin and leptin – both of which are known to promote breast cancer growth [224]. Additional work in MSCs has documented their ability to secrete large amounts of IL-6 and the corresponding increase in the growth of estrogen receptor alpha-positive cell lines [225]. Increased expression of IL4 and IL10 have also been reported by ASCs isolated from breast cancer tissue [226], leading many to speculate that the ASC may be capable of altering the immune environment within the breast, resulting in the "protection" of the cancerous cells. Such a possibility could have farreaching effects in the development of breast cancer and in its possible reoccurrence if ASCs are used in reconstruction. However, it is encouraging to find that cultured ASCs are resistant to the chemotherapies cisplatin, vincristine or comptothecin and that they still retain their stem cell characteristics [227]. Such findings could make it possible for a more natural reconstruction of the breast if ASCs are found not to contribute to the cancer itself.

#### **4.3. "Paracrines gone wild" — ASCs and adipose disorders**

With the proposed paracrine function of ASCs now well accepted, a re-examination of certain disorders and how the ASC might play a role might now be in order. The most obvious of these disorders would be obesity. However, studying the ASC might allow more information into lesser known dysfunctions such as lipedema and rare adipose disorders (RADs) like Dercum's and Madelung's disease. Normal fat has been described as having an anti-inflam‐ matory milieu with adipocytes storing lipid, regulating energy metabolism, and, together with resident macrophages, secreting anti-inflammatory mediators such as IL-10 and adiponectin to protect against the possible development of inflammation-driven obesity [228]-[230]. However, with chronic nutrient overload, existing adipocytes increase their fat storage to become hypertrophic and resident pre-adipocytes (or ASCs) are thought to undergo increased differentiation to increase adipocyte number (i.e. hyperplasia). The hypertrophic adipocytes increase their secretion of "adipokines" - soluble factors known to affect angiogenesis and inflammation [231], [232]. Specifically, these adipocytes shift their adipokine production from anti-inflammatory to inflammatory, producing a series of feedback cascades that ultimately manifests in obesity [232].

Obesity has been recognized since the 1950s as a chronic state of low-level inflammation associated with excess accumulation of adipose tissue [233]. This inflammation is now thought to be a complex response to cellular events, such as hypoxia and oxidative stresses within the adipocyte. Figure 1 outlines the possible interacting events underlying obesity starting with the creation of hypertrophic adipocytes. These adipocytes become too large to be adequately supplied by the existing vasculature in the adipose depot, resulting in localized areas of hypoxia. This hypoxic state induces the production of numerous pro-inflammatory adipokines (e.g. IL1Rα, IL6, IL8, TNFα, MCP-1, leptin) and decreases the secretion of several key antiinflammatory factors (e.g. IL10, adiponectin). Excellent reviews on these adipokines in obesity can be found in Fain et al. 2010 and Balistreri 2010. In these hypertrophic adipocytes, hypoxia is thought to induce oxidative stress [234], [235]. Oxidative stress is defined as an imbalance in the levels of reaction oxygen species (ROS) relative to the tissue's antioxidant capacity, resulting in the accumulation of oxidative products such as superoxide and hydroxyl radicals, reactive nitrogen species (RNS) and hydrogen peroxide [236]. Excess nutrients and hypertro‐ phic adipocytes can produce ROS through: the nicotinamide dinucleotide phosphate oxidase (NOX) system [237], incomplete mitochondrial respiration due to excess free fatty acids [238] and endoplasmic reticulum (ER) stress due to excess lipid storage [239]. Both mitochondrial and ER dysfunction have been demonstrated to increase the secretion of inflammatory adipokines [239], [240] and numerous studies in obesity models and obese subjects now exist linking hypoxia, oxidative stress and inflammation (reviewed in [236]). Concomitant with the development of hypertrophic adipocytes, there is a shift within the adipose tissue from M2 macrophages, found in normal adipose tissue, to a more pro-inflammatory M1 macrophage subset [241]-[243]. This shift is likely, in part, a consequence of the production of pro-inflam‐ matory adipokines by adipocytes – such as MCP-1, but this infiltration is also likely to be due to the death of these adipocytes [244]. Consistent with this, "crown-like" structures of macro‐ phages are known to be associated with necrotic adipocytes in obese murine adipose tissue [242]. These macrophages may directly contribute to the production of inflammatory agents within obese adipose tissue [245]. However, they may also augment adipokine production by the adipocyte through possible cross-talk mechanisms. While these mechanisms are unclear at this point, there are many who postulate that adipocyte-macrophage interaction is the key factor in inflammation and resulting obesity [230], [246], [247].


**Table 3.** Secretion of Pro-inflammatory Cytokines by ASCs

the growth of estrogen receptor alpha-positive cell lines [225]. Increased expression of IL4 and IL10 have also been reported by ASCs isolated from breast cancer tissue [226], leading many to speculate that the ASC may be capable of altering the immune environment within the breast, resulting in the "protection" of the cancerous cells. Such a possibility could have farreaching effects in the development of breast cancer and in its possible reoccurrence if ASCs are used in reconstruction. However, it is encouraging to find that cultured ASCs are resistant to the chemotherapies cisplatin, vincristine or comptothecin and that they still retain their stem cell characteristics [227]. Such findings could make it possible for a more natural reconstruction

With the proposed paracrine function of ASCs now well accepted, a re-examination of certain disorders and how the ASC might play a role might now be in order. The most obvious of these disorders would be obesity. However, studying the ASC might allow more information into lesser known dysfunctions such as lipedema and rare adipose disorders (RADs) like Dercum's and Madelung's disease. Normal fat has been described as having an anti-inflam‐ matory milieu with adipocytes storing lipid, regulating energy metabolism, and, together with resident macrophages, secreting anti-inflammatory mediators such as IL-10 and adiponectin to protect against the possible development of inflammation-driven obesity [228]-[230]. However, with chronic nutrient overload, existing adipocytes increase their fat storage to become hypertrophic and resident pre-adipocytes (or ASCs) are thought to undergo increased differentiation to increase adipocyte number (i.e. hyperplasia). The hypertrophic adipocytes increase their secretion of "adipokines" - soluble factors known to affect angiogenesis and inflammation [231], [232]. Specifically, these adipocytes shift their adipokine production from anti-inflammatory to inflammatory, producing a series of feedback cascades that ultimately

Obesity has been recognized since the 1950s as a chronic state of low-level inflammation associated with excess accumulation of adipose tissue [233]. This inflammation is now thought to be a complex response to cellular events, such as hypoxia and oxidative stresses within the adipocyte. Figure 1 outlines the possible interacting events underlying obesity starting with the creation of hypertrophic adipocytes. These adipocytes become too large to be adequately supplied by the existing vasculature in the adipose depot, resulting in localized areas of hypoxia. This hypoxic state induces the production of numerous pro-inflammatory adipokines (e.g. IL1Rα, IL6, IL8, TNFα, MCP-1, leptin) and decreases the secretion of several key antiinflammatory factors (e.g. IL10, adiponectin). Excellent reviews on these adipokines in obesity can be found in Fain et al. 2010 and Balistreri 2010. In these hypertrophic adipocytes, hypoxia is thought to induce oxidative stress [234], [235]. Oxidative stress is defined as an imbalance in the levels of reaction oxygen species (ROS) relative to the tissue's antioxidant capacity, resulting in the accumulation of oxidative products such as superoxide and hydroxyl radicals, reactive nitrogen species (RNS) and hydrogen peroxide [236]. Excess nutrients and hypertro‐ phic adipocytes can produce ROS through: the nicotinamide dinucleotide phosphate oxidase (NOX) system [237], incomplete mitochondrial respiration due to excess free fatty acids [238]

of the breast if ASCs are found not to contribute to the cancer itself.

**4.3. "Paracrines gone wild" — ASCs and adipose disorders**

manifests in obesity [232].

64 Regenerative Medicine and Tissue Engineering

So obesity results from a complex series of cellular events that ultimately increases the production of inflammatory adipokines within the tissue. These adipokines are known to further increase adipocyte hypertrophy producing a positive feedback system. This feedback system could be augmented further by the secretory activity of non-fat cells – i.e. the preadipocyte and even the ASC. Pre-adipocytes and adipocytes secrete many of the same proinflammatory factors listed above - with the exception of leptin and adiponectin, factors secreted by the adipocyte (reviewed in [235]). Furthermore, a review of the current literature

**Figure 1.** Possible interactions in obesity. Excess energy leads to development of hypertophic adipocytes. Hypertro‐ phic adipocytes lead to the development of cellular stresses and hypoxia, via HIFI1α signaling, which can induce the adipocyte to release numerous pro-inflammatory cytokines. Hypoxia can also result in the death of adipocytes, induc‐ ing infiltration by pro-inflamatory/M1 macrophages into the adipose tissue. Paracrine activity by macrophages could affect the release of inflammatory cytokines from the adipocytes. In addition the macrophage may also release these cytokines directly. The resulting inflammation is likely to set up a feedback loop to enhance hypertrophic adipocyte development. The role of the ASC remains unknown in obesity but possible points of interaction could be the differ‐ entiation of ASCs, leading to adipocyte hyperplasia and the release of similar pro-inflammatory cytokines. Paracrine activity is shown as solid arrows.

turns up many studies that document the secretion of similar pro-inflammatory factors by ASCs (Table 3). It is possible that the secretion of inflammatory factors, like IL6 or TNFα, by ASCs may play a crucial role in inflammation and the development of obesity. Alternatively, it is possible that inflammation and obesity may result from "defective" ASCs that fail to secrete key anti-inflammatory factors such as IL-10 or have lost their ability to ameliorate oxidative stresses, thus allowing inflammation to go on unchecked. Unfortunately, the effect of inflam‐ mation and the ASC is under-represented in today's literature. Those studies that do exist document the inhibition of ASC adipogenesis under inflammatory conditions [248]. This is an interesting finding, as the ASC might be thought of as the logical source for adipocyte hyperplasia observed in obesity. However, if it is the paracrine activity of the ASC that plays a crucial role in the development or maintenance of obesity, then ASC differentiation capacity might be sacrificed in the name of maintaining this function. In light of what we know about adipocytes and pre-adipocytes in obesity, more in-depth studies on the ASC are certainly warranted.

A similar argument for ASCs could be made for other adipose disorders such lipedema and rare adipose disorders (RADs) such as Dercum's (aka Adiposa Dolorosa) [249] and Madelung's disease or Multiple Symmetric Lipomatosis (MSL) [250]. Lipedema (LD), or edema of the fat, is defined as the symmetrical accumulation of adipose tissue in the lower extremities [251]. Because the fat may also be painful as the disorder progresses, LD is often described in the same spectrum as Dercum's [252]. While lipedema and obesity share many similarities – leading to the misdiagnosis of lipedema in up to 15% of the population as obesity, there are some significant differences between LD and obesity. Specifically, excess fat accumulates almost exclusively in the lower limbs in LD and this adipose tissue is stubbornly resistant to loss through dieting [253]. LD is almost exclusively seen in women in their 30s or older, suggesting a hormonal component [251]. Despite these differences, the etiology of obesity and LD may share some commonalities, in that LD is thought to be mediated, in part, through hypoxia and the production of inflammatory cytokines (Figure 2). Like obesity, LD is initially characterized by adipocyte hypertrophy and hyperplasia [254], although the reason for this hypertrophy cannot be attributed to nutrient overload and currently remains unknown. This hypertrophy results in hypoxia, which is thought to result in inflammatory adipokine secretion and a putative positive feedback cascade as seen in obesity. Like obesity, LD fat is characterized by macrophage "crowns" in close association with hypertrophic and/or necrotic adipocytes [132]. These macrophages will almost certainly contribute to the inflammatory reactions occurring in LD fat. Furthermore, when examining adipose tissues isolated from Dercum's, similar immune infiltrations in association with perivascular cells and hypertrophic adipocytes are also seen, again, suggesting that LD and Dercum's may be points along the same spectrum [252]. In light of these commonalities with obesity, it would be logical to assume that the ASC would also play some critical role in mediating inflammation in LD or RADs through its production of paracrine factors. Unfortunately, these studies do not exist at this point.

Despite sharing many of the same characteristics, there are some important distinctions between obesity and LD that may also be at work. These distinctions are also likely to be found in RADs like Dercum's and Madelung's disease. Specifically, LD (and possibly Dercum's and Madelung's) is associated with defects in the microvasculature, together with lymphatic dysfunction [252]. Current theories propose that adipocyte hypertrophy leads to hypoxia, which results in increased angiogenesis. However, this angiogenesis is pathologic and the resulting capillaries are said to be "fragile" or "leaky" [255]. In support of this, perivascular cells, indicative of vascular damage, can be found in LD adipose tissue [254] and pathologic angiogenesis producing fragile capillaries have been found in many eye diseases [256], [257]. What produces this pathology is unknown but studies have shown that leptin can increase the number of fenestrations in capillaries [258] and increased plasma VEGF levels can be found in LD patients [259]. Increased plasma VEGF levels can also be found in LD patients [259], so it is possible that paracrine secretion from hypertrophic and hypoxic adipocytes could disrupt angiogenesis within LD adipose tissue. With studies showing ASCs capable of secreting

turns up many studies that document the secretion of similar pro-inflammatory factors by ASCs (Table 3). It is possible that the secretion of inflammatory factors, like IL6 or TNFα, by ASCs may play a crucial role in inflammation and the development of obesity. Alternatively, it is possible that inflammation and obesity may result from "defective" ASCs that fail to secrete key anti-inflammatory factors such as IL-10 or have lost their ability to ameliorate oxidative stresses, thus allowing inflammation to go on unchecked. Unfortunately, the effect of inflam‐ mation and the ASC is under-represented in today's literature. Those studies that do exist document the inhibition of ASC adipogenesis under inflammatory conditions [248]. This is an interesting finding, as the ASC might be thought of as the logical source for adipocyte hyperplasia observed in obesity. However, if it is the paracrine activity of the ASC that plays a crucial role in the development or maintenance of obesity, then ASC differentiation capacity

activity is shown as solid arrows.

66 Regenerative Medicine and Tissue Engineering

**Figure 1.** Possible interactions in obesity. Excess energy leads to development of hypertophic adipocytes. Hypertro‐ phic adipocytes lead to the development of cellular stresses and hypoxia, via HIFI1α signaling, which can induce the adipocyte to release numerous pro-inflammatory cytokines. Hypoxia can also result in the death of adipocytes, induc‐ ing infiltration by pro-inflamatory/M1 macrophages into the adipose tissue. Paracrine activity by macrophages could affect the release of inflammatory cytokines from the adipocytes. In addition the macrophage may also release these cytokines directly. The resulting inflammation is likely to set up a feedback loop to enhance hypertrophic adipocyte development. The role of the ASC remains unknown in obesity but possible points of interaction could be the differ‐ entiation of ASCs, leading to adipocyte hyperplasia and the release of similar pro-inflammatory cytokines. Paracrine

**Figure 2.** Lipedema. Development of lipedema may have numerous commonalities with obesity starting with the de‐ velopment of hypertrophic adipocytes. Howerver, causation for this is unknown at this time may involve the ASC. As with obesity, adipocyte hypertrophy can lead to the development of hypoxia and the release of inflammatory cyto‐ kines from the adipocyte. Possible release of these factors from the ASC due to hypoxia is also shown. In addition, adipocyte hypertrophy is also accompanied by the development of "leaky" capillaries and lymphatics. While the cause of pathologic angiogenesis remains unclear, a role for the gene Prox-1 is though to be involved in lymphatic patholo‐ gy. Increased filtration from capillaries, combined with poor lypmphatic drainage (due to hypertrophic adipocytes and the the leaking of lymph back from the lymphatic vessel) leads to an accumulation of protein rich fluids within the tissue. Fluid accumulation and hypoxia may induce pro-inflammatory cytokine release. Other mechanisms of obesity (e.g. macrophage infiltration) are also likely to be involved. Paracrine activity is shown as solid arrows.

numerous paracrine factors, including VEGF, and inducing endothelial differentiation and vessel formation, the question of whether the ASC plays a role in this vascular pathology should be asked. The fragile capillaries allow the filtration of protein-rich plasma into the interstitial space, driving the formation of edema [255]. In the early stages of LD, lymphatic drainage can keep up [260]. However with progression of the disorder, lymphatic drainage does decrease as the patient ages [253]. Added to this, the hypertrophic adipocytes are thought to physically restrict fluid drainage and the smaller lymphatic vessels themselves are thought to become "leaky", possibly through the appearance of microaneurysms in these vessels [253]. All of this results in the accumulation of lymph within the adipose tissue. Recent studies now suggest that "lymph can make you fat" [261]. In support of this, adipogenesis in vitro increases when cells are cultured in the presence of lymph [262], [263]. Furthermore, the removal of axillary lymphs nodes in individuals with breast cancer is frequently associated with increased fat deposition within the arm [263]. More recently, mice heterozygous for a mutation in the Prox1 gene not only exhibit leaky lymphatics, but develop obesity as they age [264]-[266]. What it is in the lymph that enhances adipogenesis is unclear. It simply could be the result of edema causing hypoxia, inflammation and adipocyte hypertrophy – not unlike obesity. Alternatively, factors in the lymph could directly induce the ASC to differentiate or the mature adipocyte to store more fat. Since lymph is interstitial fluid combined with emulsified fats, non-reabsorbed proteins and immunocompetent leukocytes, any of these factors could conceivably alter the behavior of the ASC. As it stands, more studies investigating the exact consequences of lymph accumulation on ASC and adipocyte behavior are needed.

So while the mechanisms may differ at points, at the basis of obesity, LD and RADs is inflammation. How the ASC participates in this inflammation remains to be seen, but the ASC could be used in the treatment of these disorders. If inflammation results in adipocyte hypertrophy, then ameliorating this process could decrease the size and number of these cells. In this regard, the anti-inflammatory, anti-oxidant properties of ASCs could be taken advant‐ age of and enhanced in the hopes of mitigating the damaging effects of inflammation in these adipose disorders. However, before this could be attempted, more information is definitely required on the exact roles the ASC plays in adipose tissue formation and how these roles can go wrong when adipose disorders develop.
