Induced pluripotent stem iPS cells derived from human somati

Induced pluripotent stem (iPS) HG-9-91-01 derived from human somatic cells by expression of defined transcription factors represent a powerful novel system for disease modeling (Park et al., 2008; Takahashi et al., 2007). Moreover, the recent development of highly efficient genome editing tools has greatly facilitated the use of corrected iPS cell-derived products for autologous tissue replacement (Mali and Cheng, 2012). Most relevant to A-T, preclinical studies have started to define sets of transcription factors that promote differentiation of mouse ES cells (Muguruma et al., 2010; Tao et al., 2010) and human ES and iPS cells (Muguruma et al., 2015; Wang et al., 2015) to Purkinje neurons. Furthermore, these cells have some engraftment capability (Muguruma et al., 2010; Wang et al., 2015), suggesting that A-T iPS cells could similarly represent a source of neuronal cells for disease modeling and ultimately for regenerative therapy. Previous work has shown that A-T iPS cells generated by expression of Yamanaka factors in A-T fibroblasts (Fukawatase et al., 2014; Lee et al., 2013; Nayler et al., 2012) or T cells (Lin et al., 2015) are viable. However, all fibroblast-based protocols employed integrating viral vectors and feeder layers (Fukawatase et al., 2014; Lee et al., 2013; Nayler et al., 2012). Moreover, consistent with impaired reprogramming in fibroblasts deficient for other DSB repair factors (Gonzalez et al., 2013; Tilgner et al., 2013), the efficiency of reprogramming from A-T fibroblasts was decreased about 100-fold relative to control fibroblasts from healthy individuals (Fukawatase et al., 2014; Lee et al., 2013; Nayler et al., 2012). Furthermore, the efficiency of reprogramming from A-T carrier fibroblasts was also markedly decreased in one study (Nayler et al., 2012), suggesting a gene dose effect. Patient-derived circulating T lymphocytes were recently shown to represent an alternative to fibroblasts (Lin et al., 2015). However, the fact that they often harbor clonal pre-leukemic rearrangements involving antigen receptor loci (Taylor et al., 1996) complicates their use for disease modeling and therapy. In this context, reprogramming of nonlymphoid mononuclear cells (Chou et al., 2015; Dowey et al., 2012; Hu et al., 2011) could provide a robust yet safe approach for A-T patients and carriers. More specifically, the erythroid compartment is no or minimally affected in A-T (Boder and Sedgwick, 1958).
Materials and Methods
Results
Discussion A-T, a monogenic disease presenting with multi-organ dysfunction early in childhood, is a candidate for regenerative medicine after gene defect correction. However, previous strategies to generate iPS cells from patient-derived somatic cells were hampered by very low reprogramming efficiency (fibroblasts) or possible contamination of the source with premalignant cells (T cells). Here, we show that circulating erythroid cells provide a robust and safe alternative for the generation of A-T iPS cells. Moreover, we find that reprogramming corrects defects in chromosomal integrity and telomere length observed in A-T somatic cells, suggesting that patient-derived iPS cells rather than somatic cells represent the best substrate for gene defect correction. This observation is not unique to A-T because a previous report demonstrated that the abnormal ring chromosome 17 causing Miller Dieker Syndrome (MDS) is also corrected by reprogramming (Bershteyn et al., 2014). The use of patient peripheral blood for reprogramming has several advantages. First, the small volume of blood (30cm3 or less) employed here can be obtained from virtually any patient by venipuncture, obviating the need for specialized medical care and the discomfort associated to skin biopsies. Indeed, frozen material stored at a blood bank can be used. Secondly, the addition of BCl-xL to Yamanaka factors markedly increases the efficiency of reprogramming over previous findings using Yamanaka factors alone and fibroblasts (Fukawatase et al., 2014; Lee et al., 2013; Nayler et al., 2012). Importantly, the A-T line generated here has a normal karyotype, indicating that improved reprogramming efficiency does not result from unchecked proliferation of cells harboring chromosomal aberrations. Thirdly, our protocol takes advantage of the fact that, unlike the lymphoid compartment, the erythroid compartment is no or minimally affected in A-T, minimizing potential carry-over of abnormalities from parental cells. Finally, unlike most previous studies that employed feeder layers and/or viral vectors (Fukawatase et al., 2014; Lee et al., 2013; Lin et al., 2015; Nayler et al., 2012), our experiments were conducted using our previously described xeno-free episomal-based protocol (Chou et al., 2015), further supporting their clinical applicability.