Human IPS Cells Generated From Adult Peripheral Blood Cells and Purified CD34+ Cells by a Non-Integrating Plasmid.

Abstract 1589

Human induced pluripotent stem cells (iPSCs) that are functionally similar to embryonic stem cells (ESCs) hold great potential for cell and gene therapies, disease modeling and drug development. The earliest success was achieved by using adherent fibroblastic cells and retroviral vectors that transduce fibroblasts very efficiently. It is also highly desirable to reprogram postnatal blood cells, including those from cord blood (CB) and adult peripheral blood (PB), which are easily accessible and less exposed to environmental mutagens. In 2009, we and others have achieved the reprogramming of human postnatal blood cells using the 4 Yamanaka factors delivered by retroviral vectors. We also found that reprogramming efficiencies of CB and PB CD34+ cells are higher than age-matched fibroblasts or MSCs. This may result from an epigenetic profile of hematopoietic CD34+ cells that appears closer to iPSCs/ESCs than that of fibroblasts/MSCs to iPSCs/ESCs. To generate integration-free iPSCs that produce hematopoietic progeny efficiently, we attempted to reprogram adult PB as well as CB cells by OriP/EBNA1 episomal vectors, which were used previously to reprogram foreskin fibroblasts albeit at a low efficiency (Yu/Thomson, 2009). When one of the best combinations (#6, 3 plasmids) was used, 1–3 candidate iPSC clones per 1 million cells were obtained as reported (Yu/Thomson, 2009). We and others found that the efficiency of generating iPS clones was even lower with human adult somatic cells by the 3 vectors. To improve, we constructed a new episomal reprogramming vector system using 1–2 OriP/EBNA1 plasmids. One (pEB-C5) expresses 5 factors (OCT4/SOX2/KLF4/Myc/LIN28), and the second expresses SV40 T antigen (Tg). CB and adult PB CD34+ cells were first cultured for 4 days and expanded ≥4-folds. The expanded cells (1 million) were then transfected once by the 1–2 new OriP/EBNA1 plasmids we constructed. Fourteen days later, we obtained on average 250 and 9 TRA-1-60+, iPSC-like colonies from CB and adult PB cells, respectively, when both pCB-C5 and pEB-Tg were used. A single plasmid (pEB-C5) can also generate iPSCs although the efficiency is ∼4-folds lower. Five characterized iPSC lines derived from CB and adult PB CD34+ cells (with or without Tg) are karyotypically normal and pluripotent. After successful reprogramming and expansion, episomal DNA is gradually lost in proliferating iPSCs. After serial expansions for 11–12 passages, vector DNA was undetectable either as episomes or in the genome of the 5 iPSC lines. We next extended this approach to reprogram un-fractionated adult PB mononuclear cells (PBMCs) including those from a sickle cell patient (SCDB003). To achieve better cell proliferation that is critical to iPSC production, we used a culture condition that favors the formation and proliferation of erythroblasts from PBMCs. PBMCs purified by standard Ficoll gradient were cultured in a serum-free condition with cytokines SCF, EPO and IL-3. Although cell death was observed and cell number decreased significantly in the first 4 days, equal or more cells than input were obtained by day 8. The expanded cells morphologically resemble pro-erythroblast cells, and express high-level CD71. Less than 1.5% of them express markers of T cells (CD3, CD2, CD4 and CD8) and B cells (CD19 and CD20). When 2×10⁶ expanded SCDB003 cells (achievable from PBMCs in 1 ml or less PB) were transfected by the 2 OriP/EBNA1 plasmids and reprogrammed in the presence of butyrate, we observed 8 colonies at day 14 that are TRA-1-60+ and iPSC-like. The second plasmid (pEB-Tg) was not essential although it enhanced the efficiency by ∼4 folds. We picked and characterized 3 iPSC-like colonies derived from PBMCs with or without Tg. All of them express pluripotency markers and behave as typical iPSCs. So far we do not have evidence if they are derived from committed T or B cells that somatic mutations altered and rearranged their genomes. We are currently examining karyotypes, in vivo pluripotency, and status of episomal vectors in 3 PBMC-derived iPSCs. As compared to recent studies using viruses that preferentially reprogram human T cells with a rearranged genome, our method of using 1–2 plasmids is virus-free and genomic alteration-free. The ability to obtain integration-free human iPSCs from a few ml PB by 1–2 plasmids will greatly accelerate uses of iPSCs in both research and future clinical applications, epically for blood disease modeling and treatment.