Currently there is limited information

Currently, there is limited information on the stability of the mitochondrial genome during expansion of adult cells. A very limited number of studies have reported the detection of mutations in the mtDNA of human colonic crypt stem irak pathway as well as in adult and embryonic primate stem cells (Taylor et al., 2003; Gibson et al., 2006). More recently, it has been suggested that stem cells may have an inherent mechanism to efficiently repair most mtDNA lesions, or they are simply not subjected to enough oxidative damage in their niche (Alison et al., 2010). In this context, one important aspect to consider is whether this scenario could be changed during ex vivo expansion. In this work, we recurrently found mutations occurring within the D-loop region D310. The latter has been regarded as a mutational hotspot in primary tumors, and has been described as extremely sensitive to oxidative damage and electrophilic attack (Sanchez-Cespedes et al., 2001; Peng et al., 2011). Yet, the findings that D310 mutations are also present in non-cancerous cells (albeit at a lower frequency), and that alterations are typically confined to the polymorphic length range, ultimately suggest that abnormalities in this region may be concurrent, but not causative agents of a cancerous state (Legras et al., 2008; Chatterjee et al., 2011). Curiously, an association can be established between variations in the D310 region, and the presence of a family of hot-spot motifs based around the degenerate 13-mer CCNCCNTNNCCNC. The latter has been implicated in recruiting meiotic crossover events in the human genome, as well as in promoting the mitochondrial common deletion (Myers et al., 2008). On the basis of our observation, that the D310 region matches the degenerate consensus, we hypothesize that the presence of the full 13-mer motif (or internal sub-motifs) might facilitate the occurrence of slippage events at this locus, thus enhancing the mutation rate. Two aspects of this work merit some discussion. First, we were unable to analyze BMSCs from young donors. Nevertheless, it has been shown previously that similar down-regulation of DNA repair genes and onset of MSI under hypoxia occurs in samples from donors as young as 19years old (Rodriguez-Jimenez et al., 2008). Secondly, one important caveat with our procedure is that hypoxic cells were passaged under normoxic conditions. This limitation is also shared with the majority of published studies that deal with low oxygen tensions. In this sense, we must recognize the need to further clarify the real impact of transiently handling hypoxia expanded cells under non-hypoxic conditions, and if this has somehow biased the conclusions of the many existing publications. Although dedicated equipment is commercially available for this purpose, its high cost has greatly limited a wider application. One interesting detail unveiled by our data concerns the response of ASCs to hypoxic environments. Although studies focusing on the role of hypoxia on human ASCs are not abundant, these point out that low oxygen tensions do not seem to maintain stem cell characteristics or enhance proliferation in a similar fashion to what has been described for human BMSCs (Ma et al., 2009). Our results, both in terms of gene expression and mitochondrial response to hypoxia, collectively point to a similar conclusion: ASCs react to hypoxic environment more slowly than BMSCs. The reasons behind this behaviour are still unclear, but it is possible that the different characteristics of each cell niche (e.g. degree of vascularization, oxygen tension, cell-cell interactions), determine distinct sensitivities to hypoxia ex vivo. In this sense, it becomes important, from the bioengineering point of view, to clarify in more detail if the use of hypoxia during SC expansion (particularly ASC), brings any clear benefit in the typical time frames used to achieve clinically relevant cell numbers.