**3. Discussion**

Several preadipocytes culture systems by which cells are able to be maintained for several passages have been established so far. For example, human preadipocytes were cultured in a medium 199 or DMEM/F-12 medium supplemented with 10% fetal bovine serum (FBS) [10]. ADSCs from lipoaspirates were generally cultured in a DMEM medium containing 10% FBS [8,37]. However, we have noticed that such ADSCs, cultured in a medium containing 10% FBS, rapidly lost their growth potential and exhibited senescence-like morphology. Since fetal bovine serum is well-known to contain some components showing growth inhibitory effects [38], we attempted to find a serum-free culture condition suitable for ADSCs. Our medium was originally prepared for primate embryonic stem cells [39], and it significantly improved the growth of ADSCs in culture, which enabled us to examine their biological and physiological activities.

In HIV-infected patients, subcutaneous adipose-tissue damage is quite obvious among those receiving HAART [19–22]. In the current study, ADSCs were isolated from the less affected parts of the body of HIV-infected patients. However, the recovery rates of ADSCs from the lipoaspirates obtained from the HIV-infected patients were about one-tenth of those observed in HIV-uninfected patients, indicating that ADSCs obtained from those parts of body were notably damaged. Therefore, it was expected that physiological activity of ADSCs in those parts of the body was compromised. However, our present study clearly demonstrated that there was no detectable difference in the growth kinetics, saturation density, and cloning efficiency between ADSCs obtained from HIV-infected and HIV-uninfected patients (Table 2). We are able to conclude that ADSCs from HIV-infected patients are biologically normal. We then compared the potential of adipogenic differentiation of ADSCs, and we found that the formation of lipid droplets was completely normal in ADSCs obtained from HIV-infected patients, indicating that adipogenic conversion was not compromised in those ADSCs (Figure 2 and Table 3). The conclusion was also confirmed by the protein analysis in Figure 3, in which no difference was detectable in the expression of several proteins involved in adipogenesis. Thus, it is quite obvious that ADSCs from HIV-infected patients retain normal physiological activity. Since previous study has claimed that HIV-infection alone causes lipodystrophy through abrogation of mitochondrial function [32], we have examined whether ADSCs from HIV-infected patients show increased oxidative

stress levels or not. As shown in Figure 4, the results clearly showed no detectable change in oxidative stress level in all cases. While mitochondrial damage can occur in vivo, ADSCs isolated from the patients and expanded in culture might regain physiological mitochondrial function, suggesting that ADSCs used for therapy are better to allow time in culture to recover before transplantation.

Recent advances in plastic and reconstructive medicine have allowed autologous stem cell therapy [40–45]. As we recently reported [13], it is evident that autologous stem cell therapy is able to retrieve the abrogated function of skin. Furthermore, tissue regeneration using ADSCs involves indirect effects, which accelerate wound healing through secreting growth factors [46–52]. Thus, autologous ADSC therapy could be a promising solution for subcutaneous adipose-tissue damage, including lipodystrophy in the patients receiving HAART. So far, facial lipodystrophy has been recognized as a common side effect of HAART, and autologous fat grafting is an effective treatment modality. In this study, we proved that ADSCs derived from the patients receiving HAART retain sufficient physiological and biological activity for adipogenic differentiation. In this way, ADSCs from the HIV-infected patients are the ideal sources for autologous stem cell therapy for lipodystrophy. Although we should be cautious as a number of cases examined in this study was limited, ADSCs from the patients with lipodystrophy could have sufficient biological potentials, so that they could be used for autologous ADSCs-based regenerative therapy [53].
