According to the Berlin definition of ARDS, the term acute lung injury (ALI) has been eliminated since 2011 , however, the majority of the field still uses ALI as a valid denomination. For the purpose of this article, we still use the term ALI. Given that excessive inflammatory responses play a crucial role in the pathophysiology of ALI/ARDS , immunomodulatory therapy is an attractive new approach in the treatment of such afflictions [28, 29]. Multipotent stromal cells provide unique opportunities for developing novel treatments for a large array of inflammatory diseases, because MSCs possess potent immunomodulatory properties and are often considered to be hypoimmunogenic or 'immunoprivileged' [10, 30]. An emerging body of data indicates that either human or mouse bone marrow-derived MSCs (BMSCs) alleviate tissue damage and suppress inflammatory reactions in different models of ALI [13–15, 31]. As an alternative source of mesenchymal stem cells, adipose tissue is abundant and easily accessible, and high yields of adipose-derived MSCs (ASCs) are easier to isolate with less invasive procedures. ASCs have been reported to have similar immune regulatory capabilities as BMSCs [20, 32]. While the use of BMSCs to treat ALI/ARDS is promising, there is little known about the potential therapeutic effects of ASCs in this disease setting. In this study, the therapeutic effect of human-and mouse-derived ASCs was investigated in a mouse model of LPS-induced ALI.
Endotoxin (bacterial LPS)-induced ALI in mice is a widely used experimental model to investigate the pathogenesis and treatment of ALI/ARDS. However, it should be noted that no single ALI animal model can completely reproduce all the pathophysiological features of ALI/ARDS in humans. The airway LPS administration model employed in these studies results in massive alveolar neutrophil infiltration, which is also seen in human ALI/ARDS [33, 34].
In this study the relatively noninvasive OA method  was used to administer both LPS and ASCs. A previous study from our group showed a comparable beneficial effect using the IV, intraperitoneal (IP), and OA methods to deliver human BMSCs to LPS-challenged mice .
Taking advantage of the cross-species nature of hASCs and the eGFP marker of mASCs in our experiments, the persistence of the cells in the lungs after delivery through OA was confirmed by real-time RT-PCR for human GAPDH and eGFP. It is expected that the cells detected reflect persistence of live cells since mRNA for human GAPDH and eGFP was detected as markers for human and mouse ASCs, respectively.
These results demonstrated that treatment of ALI with either hASCs or mASCs decreased the body weight loss following ALI induction, which suggests lower disease severity following treatment. Both treatments decreased LPS-induced lung damage and neutrophil infiltration substantially, as shown by H&E staining. This was further corroborated by decreased levels of total protein and albumin in BALF, indicating reduced endothelial and epithelial permeability in the lung following either hASCs or mASCs treatment.
Because neutrophil activation and transmigration into lung interstitium and broncheoalveolar space plays a crucial role in the development of ALI , it can be inferred that decreasing infiltration of neutrophils using ASCs administration may have clinically relevant therapeutic potential. Activated neutrophils secrete proteinases and potent enzymes such as MPO that generate oxidants, all of which can damage lung tissue leading to edema and loss of lung function . Decreased MPO activity after cell treatment indicates that both hASCs and mASCs are capable of decreasing the lung inflammatory responses that ultimately cause the neutrophil-associated tissue damage common in ALI.
Nemeth et al. and Anderson et al. showed that mouse BMSCs and ASCs were able to reprogram macrophages to increase their IL-10 production, and the administration of these cells protected mice from sepsis [12, 32]. In this study, both hASC and mASC treatments led to a reduction of the expression of various proinflammatory cytokines/chemokines, while increasing the production of the anti-inflammatory cytokine IL-10. These data suggest that the anti-inflammatory effects of hASC and mASC in this model were mediated, at least in part, by the increased production of IL-10 in the lung. The decrease of the production of proinflammatory cytokines/chemokines at the protein level was found to be not as pronounced as that at the mRNA level 24 hours after LPS exposure. This may be explained by the time lag between translation and transcription.
It is important to note that while cytokine and chemokine levels were quantified in the lungs of mASC-treated mice, the cellular source of the cytokines or chemokines (that is, lung epithelial cells, alveolar macrophages, inflammatory cells or administered ASCs) was not able to be determined. Our previous study showed that, when stimulated with LPS or BALF, mASCs did secrete some cytokines and chemokines (for example IL-1α and MIP-1α) in vitro (data not shown). These data suggest that decreased cytokine levels in animals treated with mASCs may actually be lower than recorded since the mASC may act as an exogenous source of these cytokines. However, the expression of cytokines and chemokines from human ASCs could not be detected since human cytokines and chemokines would not be detected by mouse specific primers. Therefore, using human ASCs offers the possibility of determining how the ASCs are responding to the injured lung, because human specific primers or antibodies can be used.
To the best of our knowledge, this is the first report to demonstrate the beneficial effects of ASCs for the treatment of LPS-induced acute lung injury in vivo. Moreover, the therapeutic effects of human and mouse ASCs have been compared side by side. While both hASCs and mASCs significantly attenuated the lung injury and inflammation, some notable differences between the two cell types were observed. For most of the parameters examined, mASCs had a more beneficial effect than xenogeneic hASCs. A possible explanation for this observation is that the cross-talk between mASCs and the injured mouse lung is more effective than in the xenotransplantation model using hASCs. In addition, studies have demonstrated species variation in the mechanisms of bone marrow-derived MSC-mediated immunosuppression. Ren et al. showed that under the same culture conditions, immunosuppression by human or monkey BMSCs was mediated by indoleamine 2,3-dioxygenase, while mouse BMSCs utilized nitric oxide . The species variation may also be an explanation to the different therapeutic effects of hASCs and mASCs in this ALI model.
To advance the use of stem cell therapy in the treatment of ALI/ARDS it is essential that the molecular mechanisms behind any observed beneficial effects be understood. As discussed above, there is increasing evidence that both mouse and human BMSCs have similar anti-inflammatory properties, however the mechanisms mediating these outcomes may be unique to each cell type. Similarly, our unpublished data suggest that human and mouse ASCs suppress inflammation in the lung by different mechanisms. Therefore, it is essential that human cells be used in preclinical studies employing animal models. However concerns that the protective factors produced by human MSCs may not be effective in other species have limited such studies. Results from this study demonstrate potent anti-inflammatory effects of both human and mouse ASCs in wild-type mice exposed to LPS, a well-characterized model of ALI. While syngeneic mouse ASCs were expectedly more effective at suppressing inflammation, our results clearly demonstrate the feasibility of using immunocompetent mice to interrogate the molecular mechanisms by which human ASCs suppress inflammation.
There are currently more than 60 clinical trials registered with clinicaltrial.gov testing ASCs in a variety of disorders, including Crohn's fistula, type 2 diabetes, and chronic obstructive pulmonary disease. As discussed above, ASCs are easy to isolate and expand in vitro, and they are also 'immunoprivileged' , which makes allogeneic transplantation to humans feasible. Despite the promising results we obtained in the mouse model, clinicians should use caution when considering the use of ASCs as a therapy for ALI/ARDS in human subjects. A recent study showed that in both humans and mice, intravenous infusion of ASCs did not induce any serious adverse events related to ASC transplantation or any tumor development (even in nude mice) . However, other studies have indicated that ASCs enhance tumorigenesis and metastasis of breast cancer cell lines and primary breast cancer samples [39, 40]. Our recently published data showed that ASCs isolated from the subcutaneous abdominal adipose tissue of obese subjects have enhanced invasion toward breast cancer cells . Prantl et al. demonstrated that ASCs promote prostate tumor growth . Therefore, before use in human subjects, the safety of ASCs should be fully investigated and evaluated, particularly in patients with breast cancer and prostate cancer.