|Home | About | Journals | Submit | Contact Us | Français|
Direct reprogramming of human fibroblasts to induced pluripotent stem cells (iPS) has been achieved by ectopic expression of defined transcription factors. Derivation of human fibroblasts however is a time consuming process and requires punch biopsies or isolation of patient foreskin. Here we use a polycistronic vector encoding Oct4, Klf4, Sox2 and c-Myc to generate iPS cells from from frozen peripheral blood of several donors. Genomic DNA analyses indicated that iPS cells were derived from mature T cells as well as myeloid donor cells. Inducing pluripotency in peripheral blood would allow utilization of easy to get samples from the adult and, more importantly, provide convenient access to numerous patient samples stored in blood banks. The latter is of major interest as frozen blood samples, when reprogrammed to iPS cells, would allow the retrospective molecular analyses of rare diseases.
Embryonic stem cells are pluripotent cells derived from the inner cell mass of the developing embryo that have the capacity to differentiate into every cell type of the adult (Evans and Kaufman, 1981; Martin, 1981; Martin and Evans, 1975; Thomson et al., 1998). The generation of patient-specific pluripotent cells is therefore an important goal of regenerative medicine. A major step to achieve this was the recent discovery that ectopic expression of defined transcription factors induces pluripotency in somatic cells (Lowry et al., 2008; Park et al., 2008b; Takahashi et al., 2007). Until now, the most common source to derive human iPS cells is skin fibroblasts (Lowry et al., 2008; Park et al., 2008a; Park et al., 2008b; Takahashi et al., 2007; Yu et al., 2009). However, the requirement for skin biopsies and the need to expand fibroblast cells for several passages in vitro represent a hurdle which must be overcome to make iPS technology broadly applicable. Peripheral blood can be utilized as an easily accessible source of patient tissue for reprogramming. Here we derived iPS cells from frozen human peripheral blood samples. Some of the iPS cells had rearrangements of the T cell receptor (TCR) indicating that T cells can be reprogrammed to pluripotency.
Recently, granulocyte colony stimulating factor (G-SCF) mobilized CD34+ blood cells have been used as a source to derive iPS cells (Loh et al., 2009). However, this requires the subcutaneous injection of G-CSF, a process that can only be applied if the donor is in good medical condition. Also, the negative effects of treatment of patients with growth factors such as erythropoietin (Miller et al., 2009) and G-CSF are still being investigated. Of concern is the use of G-CSF as this cytokine is a growth factor for myeloid cell precursors (Touw and van de Geijn, 2007) and because G-CSF treatment of patients with severe congenital neutropenia (SCN) can result in a truncated G-CSF receptor allele and acute myeloid leukemia transformation (Touw, 1997). Derivation of iPS cells from peripheral mononuclear blood cells would circumvent all these issues; in addition, peripheral blood is the most accessible adult tissue and permits access to numerous frozen samples already stored at blood banks. Such samples could be expanded in culture and reprogrammed to iPS cells, which in turn allows studying the molecular mechanism underlying blood and other disorders. We show here the derivation of iPS clones from mature peripheral blood T- and myeloid cells.
Mononuclear (MNC) blood cells were isolated from several donors by Ficoll-Hypaque density gradient centrifugation (Ferrante and Thong, 1980; Vissers et al., 1988). Samples were frozen and thawed days to several months after freezing and expanded in IL-7 or in G-CSF, GM-CSF, IL-6 and IL-3 for 5 days. In our initial experiments we used the FUW-M2rtTA and the individual doxycycline inducible lentiviruses encoding Oct4, Sox2, c-Myc or Klf4 (Brambrink et al., 2008). However, in ten independent experiments we were not able to reprogram peripheral blood cells using this system. One possibility for failure to obtain iPS cells is that peripheral blood cells are difficult to infect reducing the probability to obtain cells carrying the four factors as well as the FUW-M2rtTA construct. Also, we find that the efficiency of blood reprogramming (0.001-0.0002%) is approximately 10-50 times lower than that of human fibroblast reprogramming.
To increase the infection efficiency we used a doxycycline-inducible lentivirus encoding all four factors Oct4, Klf4, Sox2 and c-Myc from a polycistronic expression cassette (pHAGE2-TetOminiCMV-hSTEMCCA) (Sommer et al., 2010). Blood cells were simultaneously infected with a constitutively active lentivirus encoding the reverse tetracycline transactivator (FUW-M2rtTA) (Hockemeyer et al., 2008) as well as the polycistronic vector. Infected blood cells were transferred onto feeder layers of mouse embryonic fibroblasts (MEFs) and cultured in the presence of IL-7 or G-CSF, GM-CSF, IL-6 and IL3 and 2 μg/ml doxycycline (Dox) for an additional four days (Figure 1A). At day 5 after Dox induction, the cells were transferred to human ES medium containing 2 μg/ml Dox and 25 - 40 days later colonies were picked and expanded.
We obtained iPS colonies from several donors of different age (25-65 years) (Supplemental Table 1). We found that cells cultured in the presence of IL-7 expanded and reprogrammed more efficiently than cells grown in the presence of G-CSF, GM-CSF, IL-3 and IL-6. Southern blot analyses probing for M2rtTA vector integrations indicated that several iPS lines of donor 3 (Figure S1A) and 4 (and data not shown) were derived from independent cells. Colonies were expanded into stable, Dox-independent iPS lines that were not dependent on exogenous factor expression (Figure S1B). iPS cells which displayed the morphology characteristic of human ES cells, had a normal karyotype (Figure S1C) and stained positive for the pluripotency markers Oct4, Nanog and Tra1-81 (Figure 1B). Blood derived iPS cells were cultured up to 35 passages and had elongated telomeres as shown by Southern blot analyses using an 800bp TTAGGG repeat probe (de Lange, 1992) (Figure 1C).
To assess the in vitro differentiation capacity of the iPS lines, the cells were differentiated into embryoid bodies (EBs). Quantitative RT-PCR analyses of mesodermal (Msx-1), endodermal (AFP) and ectodermal (NCAM) markers demonstrated that all three lineage markers were up regulated in the differentiated EBs as compared to the undifferentiated iPS lines (Figure 1D). To evaluate their in vivo differentiation potential the iPS cells were injected subcutaneously into NOD-SCID mice and tumors were removed after 6-8 weeks. Histological analyses revealed that cell types characteristic for all three germ layers including ectoderm (neural rosette, neural epithelium), mesoderm (cartilage, bone) and endoderm (intestinal epithelium) were present (Figure 1E) indicative of pluripotency.
Reactivation of Oct4 locus through demethylation of its promoter during reprogramming is a hallmark of iPS cells. We therefore determined the methylation patterns of the Oct4 promoter region in iPS cells and peripheral blood cells. iPS cells had mostly non-methylated promoter regions characteristic for the active Oct4 gene, whereas peripheral blood samples showed the expected highly methylated promoter (Figure 1F). The results shown in Figure 1 indicate that peripheral blood derived iPS cells are pluripotent and show the molecular and morphological characteristics of human ES cells.
T-cell development involves sequential genetic DNA rearrangements of the T cell receptor (TCRD > TCRG > TCRB > TCRA) or immunoglobulin loci (IGH > IGL > IGK), respectively (Davis and Bjorkman, 1988; Kisielow and von Boehmer, 1995; Rajewsky, 1996; Tonegawa, 1983). This allowed us to retrospectively assess whether the iPS cells were derived from mature T-cells (TCR gene rearrangements), B-cells (IG gene rearrangements) or myeloid cells (no TCR/IG gene rearrangements). During normal development, T-cells mature in the thymus and migrate into the periphery as fully differentiated cells (Kisielow and von Boehmer, 1995). Detection of any TCR gene rearrangement in iPS cells derived from peripheral blood of healthy donors is, therefore, indicative of a mature T-cell. PCR analyses were performed to detect potential TCR delta (TCRD), TCR gamma (TCRG), or TCR beta (TCRB) rearrangements using TCR primer sets designed by the BIOMED-2 consortium (van Dongen et al., 2003) and purchased from InVivoScribe Technologies. The primer mixes target conserved regions within the variable (V), diversity (D) and the joining regions (J) of the TCRB, TCRD or the V and J regions of TCRG, respectively. In a clonal cell population, amplification of this region results in a PCR product within a predictable size range.
We identified bands within the valid size range for TCRB (Figure 2A, S1D) and TCRG (Figure 2B, S1E) gene rearrangements for all iPS clones derived from donor 3 and 4. All clones analyzed tested negative for TCRD rearrangements (Figure 2C and data not shown). Sequencing analyses further identified the specific nature of productive TCRB and productive or unproductive TCRG gene rearrangements as shown in Figure 2D. Some PCR conditions led to amplification of additional bands inside and outside the valid size range. We cloned numerous PCR products that were in close proximity to the expected size range and confirmed that they reflect unspecific amplicons (data not shown).
One iPS line (D1MiPS #1) derived in IL-7 was negative in all TCR gene rearrangement assays (Figure S2A, S2B and S2C). We therefore investigated whether this clone originated from a B-lymphocyte or a myeloid cell. Using the framework 1-3 primer sets (van Dongen et al., 2003) we did not detect any IGH gene rearrangements (Figure S2D) suggesting that this clone may have originated from a myeloid cell. As expected, the two iPS clones (D2MiPS #1 and #2) derived in the presence of G-CSF, GM-CSF, IL-3 and IL-6 tested negative in all TCR (Figure S2A, S2B and S2C) or IGH gene (Figure S2D) rearrangement assays, whereas one iPS line (D2TiPS#1) derived from the same donor in IL-7 tested positive for TCRG gene rearrangement (Figure S2C). In summary, our results indicate that iPS cells can be derived from terminally differentiated adult peripheral T-cells.
Current protocols for reprogramming human cells are based on skin fibroblasts, keratinocytes or G-CSF mobilized CD34+ cells (Loh et al., 2009; Park et al., 2008b; Takahashi et al., 2007; Yu et al., 2009). Nuclear transfer and four factor mediated reprogramming experiments have demonstrated that pluripotency can be induced in terminally differentiated mouse lymphocytes (Hanna et al., 2008; Hochedlinger and Jaenisch, 2002; Hong et al., 2009). Here, we show that human peripheral blood T- and myeloid cells cultured in IL-7 or G-CSF, GM-CSF, IL-3 and IL-6, can be reprogrammed to a pluripotent state using a polycistronic vector encoding Oct4, Klf4, Sox2 and c-Myc. Due to sequential DNA rearrangements of TCR or IG genes during lymphocyte development (Kisielow and von Boehmer, 1995; Rajewsky, 1996) we were able to retrospectively assess that the majority of iPS cells were derived from peripheral blood T-cells. Two clones obtained in G-CSF, GM-CSF, IL-3 and IL-6 and one clone derived in IL-7 tested negative for TCR and IG gene rearrangements, suggesting that these iPS cells originated from myeloid cells. Proliferation of somatic cells is an important parameter of reprogramming (Hanna et al., 2009), which is consistent with the higher reprogramming efficiency of T-lymphoyctes as compared to myeloid cells because T cells have higher proliferation rates and better long-term growth potential in vitro than myeloid cells.
Our study demonstrates that peripheral blood can be utilized as an easily accessible source of patient tissue for reprogramming without the need to extensively maintain cell cultures prior to reprogramming experiments. This is an important step to make the iPS technology more broadly applicable. Importantly, reprogramming of peripheral blood samples will permit access to numerous frozen samples already stored at blood banks. These samples are often of restricted use for research, because limited cell numbers do not allow experimental manipulations. This is particularly relevant if the patient is deceased and new material cannot be obtained. Generation of iPS cells from such samples could provide cell numbers large enough to retrospectively screen for genetic factors and to study molecular mechanisms underlying myeloid and lymphoid blood disorders. The culture conditions used in the current study primarily expanded T-lymphocytes and myeloid cells and are likely the reason why B-lymphocyte iPS cells were not derived. To expand B lymphocytes more efficiently, prior to the reprogramming process cells would need to be cultured in IL-4/CD40 ligand in order to promote B-cell survival and expansion (von Bergwelt-Baildon et al., 2002). In summary, our study allows reprogramming of the most easily available adult lineage and provides a protocol to access samples stored at blood banks.
We thank Raaji Alaggapan and Ping Xu for support with tissue culture and Jessie Dausman, Ruth Flannery and Dongdong Fu for help with animal husbandry and teratoma processing. We thank all volunteers for donating blood, Catherine Ricciardi at CRC for taking blood, Dirk Hockemeyer for providing the 800bp TTAGGG telomere probe and advice with telomere Southern blots and members of the Jaenisch lab for critical reading of the manuscript. R.J. is supported by NIH grants 5-RO1-HDO45022, 5-R37-CA084198, and 5-RO1-CA087869. J.S. is a long-term HFSP postdoctoral fellow; M.M.D. is a Damon Runyon postdoctoral fellow. R.J. is an advisor to Stemgent and Fate Therapeutics. Blood donations were conducted at the Clinical Research Center (CRC) at the Massachusetts Institute of Technology, supported by Grant Number UL1 RR025758- Harvard Clinical and Translational Science Center, from the National Center for Research Resources.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.