Conceived and designed the experiments: MM ARM. Performed the experiments: MM OK MM ARM. Analyzed the data: MM GY ARM. Contributed reagents/materials/analysis tools: GY FHG ARM. Wrote the paper: MM FHG ARM. Financial Support for the project: FHG.
The authors have declared that no competing interests exist.
Genetic reprogramming of somatic cells to a pluripotent state (induced pluripotent stem cells or iPSCs) by over-expression of specific genes has been accomplished using mouse and human cells. However, it is still unclear how similar human iPSCs are to human Embryonic Stem Cells (hESCs). Here, we describe the transcriptional profile of human iPSCs generated without viral vectors or genomic insertions, revealing that these cells are in general similar to hESCs but with significant differences. For the generation of human iPSCs without viral vectors or genomic insertions, pluripotent factors Oct4 and Nanog were cloned in episomal vectors and transfected into human fetal neural progenitor cells. The transient expression of these two factors, or from Oct4 alone, resulted in efficient generation of human iPSCs. The reprogramming strategy described here revealed a potential transcriptional signature for human iPSCs yet retaining the gene expression of donor cells in human reprogrammed cells free of viral and transgene interference. Moreover, the episomal reprogramming strategy represents a safe way to generate human iPSCs for clinical purposes and basic research.
Genetic reprogramming to a pluripotent state of mouse somatic cells was first achieved by ectopic expression of four factors (Oct4, Sox2, Klf4 and c-Myc) using retroviruses
Viral vectors are known to affect the transcriptional profile from target cells, altering their behavior and sometimes inducing apoptosis
Several attempts were made to generate a viral-free, integration-free iPSCs. The generation of iPSCs with later excision of reprogramming factors was recently achieved; still, the genome continues to be affected by random solo-LTR insertions from viral vectors
The timing of the reprogramming and the factors required seem to vary depending on cellular context
Our starting material was a multipotent, karyotypically normal, c-Myc-immortalized human NSC line derived from a tissue sample of human midbrain (10 weeks of gestation). Our rationale was that the elevated expression of c-Myc and Sox2 in these cells might prompt them to reprogram more easily than reported for other types. First, we examined whether the combination of Oct4 and Nanog would reprogram these cells to a pluripotent state
A, Morphology of human fetal NSCs before lentiviral infection. Inset: after 3 days post-infection with Lenti-Oct4 and Lenti-Nanog, individual cells expressed alkaline phosphatase (AP). B, Example of infected plates stained for AP at 14 days post-infection showing several AP-positive colonies. Control infection did not result in any AP-positive colonies. C and D, Aspect of colonies 14 days after infection growing in MEFs. E, Established human iPSC colonies, with well-defined borders and compact cells, are morphologically similar to hESCs. F, Typical image of iPSCs growing in feeder-free conditions. G, Representative immunofluorescence analysis of iPSCs growing on matrigel. Clear expression of pluripotent markers is observed. Bar = 150 µm.
To generate human iPSCs without the use of viral delivery vectors or genomic insertions, the Oct4 and Nanog cDNAs were independently cloned under the CMV promoter into a plasmid (pCEP) with the
A, Aspect of human NSCs after plasmid electroporation and plating on MEFs. B and C, Some selected colonies display a strong differentiation tendency in feeder-free conditions. D, Established iPSC lines are morphologically similar to hESCs. E, iPSCs have a large nucleus-to-cytoplasm ratio and prominent nucleoli when compared to original NSCs (F). G, Immunofluorescence analysis of iPSCs growing on matrigel showed clear expression of typical ESC markers. H,
These three iPSC colonies expressed several pluripotent markers and were able to form embryoid bodies (EBs)
A and B, PCR analyses for plasmid integration in genomic DNA from the iPSC clones. Controls: (−) water; (+) pCEP4 plasmid. Primers were designed to specifically amplify plasmid backbone (A) or transgenes (B) (see
We then analyzed if myc levels from these iPSCs derived from NSCs would change after reprogramming. Interestingly, despite the fact that the NSCs were immortalized with ectoptic expression of myc, the transcriptional activity of myc is higher in iPSCs compared to NSCs. Moreover, iPSCs clones have similar myc transcriptional levels to hESCs (
Next, we asked if the global molecular signatures of two plasmid-free iPSC lines (iPSC1, iPSC2) resembled those of available hESC lines, namely HUES6 and Cyt25. Gene expression profiles measured using human genome Affymetrix Gene Chip arrays were grouped by hierarchical clustering, and correlation coefficients were computed for all pair-wise comparisons (
A, Hierarchical clustering and correlation coefficients of microarray profiles of triplicate iPSC1, iPSC2, CytES (Cyt25 hESC), Hues6 and NSC. Color bar indicates the level of correlation (from 0 to 1). Panel below illustrates marker genes implicated in pluripotency of NSCs, with color bar reporting log2 normalized expression values (green/red indicates high/low relative expression). B, Refseq-annotated genes that were insufficiently induced in iPSCs relative to hESCs (yellow/blue indicates high normalized log2 expression). C, Refseq-annotated genes that were insufficiently silenced in iPSCs relative to hESCs.
Despite the global similarity between iPSCs and hESCs, the profiles were not completely indistinguishable, which led us to study what the molecular differences were. Four independent (A versus B) group-wise comparisons were performed to identify differentially expressed genes: (i) iPSC versus hESC (1,952 Refseq-annotated genes were significantly enriched in iPSCs versus hESCs; 1,072 genes were enriched in hESCs versus iPSCs at P<0.01 after correcting for multiple hypotheses testing); (ii) iPSC versus NSC (3,347 genes were significantly enriched in iPSCs versus NSCs; 2,959 genes were enriched in NSCs versus iPSCs); (iii) hESC versus NSC (2,376 genes were significantly enriched in hESCs versus NSCs; 2,541 genes were enriched in NSCs versus hESCs); (iv) iPSC and hESC versus NSC (3,730 genes were significantly enriched in iPSCs and hESCs, versus NSCs and 3,638 genes were enriched in NSCs versus iPSCs and hESCs (
Panel illustrates marker genes implicated in pluripotency of NSCs, with color bar reporting log2 normalized expression values (green/red indicates high/low relative expression).
Next, we repeated the transient transfection using NSCs derived from the H1 hESC line that contains the EGFP reporter cassette knocked in the endogenous Oct4 gene by homologous recombinantion
A, Undifferentiated H1 Oct4-EGFP hESC line expresses the EGFP reporter gene that is gradually turned off during NSC differentiation. NSCs are morphologically distinct from hESCs. B, Small iPSC colonies can be detected 10 days after transfection with pCEP-Oct4 and pCEP-Nanog. C, Typical number of iPSC colonies obtained with electroporation of pCEP-Oct4 and Nanog or with Oct4 alone. Bar = 150 µm.
Using a simple methodology (
Episomal plasmids carrying reprogramming factors are transfected into NSCs and cells are plated on MEFs. On the following day, medium is changed to the hESC condition. Resistant selection is kept for a week. After 14 days, iPSC colonies are visible and can be transferred to a feeder-free condition. Individual colonies are expanded and ready for characterization. At this time, no evidence of plasmid integration is found.
Our results support earlier observations that viral integration is dispensable for genetic reprogramming
All animal work was conducted according to relevant national and international guidelines. Protocols were previously approved by the University of California San Diego Institutional Animal Care and Use Committee, the Institutional Review Board and the Embryonic Stem Cell Research Oversight Committee.
Human fetal NSCs (ReNCell VM, Chemicon) were cultured on laminin-coated dishes in ReNcell maintenance medium (Chemicon) in the presence of basic fibroblast growth factor 2 (bFGF2), following the manufacturer's instructions. The hESC Cyt25 (Cythera, San Diego) and HUES6 cell lines were cultured as previously described
Lentiviral vectors containing the Oct4 and Nanog human cDNAs from Yamanaka's group were obtained from Addgene. The cDNAs were then subcloned into the pCEP4β episomal plasmid (Invitrogen). Plasmid transfections were done by electroporation of equimolar amounts of pCEP-Oct4 and pCEP-Nanog (5 µg each) using the nucleofactor for rat NSCs, following the manufacturer's instructions (Lonza/Amaxa Biosystem). Lentiviruses were produced by triple transfection of HEK293T cells followed by ultracentrifugation as previously described elsewhere
Cells were briefly fixed in 4% paraformaldehyde and then permeabilized with 0.5% Triton-X in PBS. Cells were blocked in 0.5% Triton-X with 5% donkey serum for 1 hour before incubation with primary antibody overnight at 4°C. After 3 washes in PBS, cells were incubated with secondary antibodies (Jackson ImmunoResearch) for 2 hours at room temperature. Fluorescent signals were detected using a Zeiss inverted microscope and images were processed with Photoshop CS3 (Adobe Systems). Primary antibodies used in this study are SSEA-4, TRA-1-60, TRA-1-81 (1∶100, Chemicon) and Lin28 (1∶500 R&D Systems). Alkaline phosphatase activity was detected in live cells using the Vector Red Alkaline Phosphatase substrate kit (Vector Laboratories).
Genomic DNA was isolated and prepared using standard molecular techniques. The PCR primers were designed to recognize the pCEP4 episomal vector (Invitrogene). The primers pairs used to amplify the plasmid back bone were: CEP19-F:
Total cellular RNA was extracted from ∼5×106 cells using the RNeasy Protect Mini kit (Qiagen, Valencia, CA), according to the manufacturer's instructions, and reverse transcribed using the SuperScript III First-Strand Synthesis System RT-PCR from Invitrogen. The cDNA was amplified by PCR using Accuprime Taq DNA polymerase system (Invitrogene). The primer sequences were: hNanog-Fw:
Around 1−3×106 cells were injected subcutaneously into the dorsal flanks of nude mice (CByJ.Cg-Foxn1nu/J) anesthetized with isoflurane. Five to 6 weeks after injection, teratomas were dissected, fixed overnight in 10% buffered formalin phosphate and embedded in paraffin. Sections were stained with haematoxylin and eosin for further analysis.
Adult Sprague-Dawley male rats (320–350 g; n = 6) were anesthetized with isoflurane (1.5–2% maintenance; in room air), placed into a spinal unit apparatus (Stoelting, Wood Dale, IL, USA) and a partial Th12–L1 laminectomy performed using a dental drill (exposing the dorsal surface of L2–L5 segments). Using a glass capillary (tip diameter 80–100 µm) connected to a microinjector (Kopf Instruments, Tujunga, CA), rats were injected with 0.5 µl (10, 000 cells per injection) of the iPS (n = 3) or proliferating H9 cells in DMEF/F12 media. The duration of each injection was 60 s followed by 30 s pause before capillary withdrawal. The center of the injection was targeted into the dorsal horn (distance from the dorsal surface of the spinal cord at L3 level: 0.5–0.7 mm). Ten injections (500–800 µm rostrocaudally apart) were made on each side of the lumbar spinal cord. After injections, the incision was cleaned with penicillin-streptomycin solution and sutured in two layers. Three or four weeks after cell grafting, rats were deeply anesthetized with pentobarbital and phenytoin and transcardially perfused with 200 ml of heparinized saline followed by 250 ml of 4% paraformaldehyde in PBS. The spinal cords were dissected and postfixed in 4% formaldehyde in PBS overnight at 4°C and then cryoprotected in 30% sucrose PBS until transverse sections (30 µm thick) were cut on a cryostat and mounted on Silane-Prep slides (Sigma). Sections were stained with H&E or immunostained overnight at 4°C with primary human specific (h) or non-specific antibodies made in PBS with 0.2% Triton-X100: mouse anti-nuclear matrix protein/h-nuc (hNUMA; 1:100; Millipore, Temecula, CA, USA); goat anti-doublecortin (DCX; 1:1000; Millipore); mouse anti-Nestin (hNestin; Chemicon). After incubation with primary antibodies, sections were washed 3× in PBS and incubated with fluorescent-conjugated secondary donkey anti-mouse, or donkey anti-goat antibodies (Alexa 488, 546; 1:250; Invitrogen Corp., Carlsbad, CA, USA) and DAPI for general nuclear staining. Sections were then dried at room temperature, covered with Prolong anti-fade kit (Invitrogen Corp., Carlsbad, CA, USA) and analyzed with confocal microscopy (Olympus, Fluoview 1000).
DNA fingerprinting analysis was performed by Cell Line Genetics (Madison, WI).
The Affymetrix Power Tools (APT) suite of programs and Affymetrix HG-U133 Plus 2.0 library files and annotation were obtained from
Sustained expression using episomal vectors. A, Human fetal NSCs were electroporated with an episomal plasmid carrying the EGFP reporter gene. Transfection efficiency was around 95%. B, Percentage of cells expressing EGFP in the presence or not of hygromycin. Bar = 150 μm.
(1.00 MB TIF)
Integration-free iPSC colonies are genetically identical to the original human fetal NSCs. DNA fingerprinting analysis at 16 independent loci indicated that both iPSCs generated by lentivirus infection (iPSC colony 19) and by transient transfection with episomal vectors (iPSC colony 1) and the original human fetal NSCs (ReNCell VM) shared all alleles investigated and were different from commonly available hESC lines.
(0.81 MB TIF)
Development of teratomas after spinal injections of iPSCs into lumbar gray matter. Lumbar spinal cord sections were stained with H&E at 3 weeks after grafting (A, B). The presence of rosette-like structures (A, yellow arrow) and ectoderm-derived squamous epithelium was identified (B, yellow arrow). Staining with human-specific nestin (green) and DCX (red) antibody show well organized nestin positive cells in primitive neuronal tube and numerous postmitotic DCX-positive neurons at the periphery of grafts (C, D).
(9.63 MB TIF)
Myc levels in neural stem cells before and after reprogramming. The myc levels in iPSCs are similar to hESCs.
(0.27 MB TIF)
IPSC-enriched probes in IPSC versus ES. Probesets enriched in group-wise comparisons: Column headings are probeset identifiers, T-statistic, P-value, Fold-Change (log2), Refseq identifier and Description of the gene. (NA indicates no Refseq annotation).
(5.43 MB DOC)
ES-enriched probes in IPSC versus ES. Probesets enriched in group-wise comparisons: Column headings are probeset identifiers, T-statistic, P-value, Fold-Change (log2), Refseq identifier and Description of the gene. (NA indicates no Refseq annotation).
(2.66 MB DOC)
IPSC-enriched probes in IPSC versus NSC. Probesets enriched in group-wise comparisons: Column headings are probeset identifiers, T-statistic, P-value, Fold-Change (log2), Refseq identifier and Description of the gene. (NA indicates no Refseq annotation).
(8.67 MB DOC)
NSC-enriched probes in IPSC versus NSC. Probesets enriched in group-wise comparisons: Column headings are probeset identifiers, T-statistic, P-value, Fold-Change (log2), Refseq identifier and Description of the gene. (NA indicates no Refseq annotation).
(9.01 MB DOC)
ES-enriched probes in ES versus NSC. Probesets enriched in group-wise comparisons: Column headings are probeset identifiers, T-statistic, P-value, Fold-Change (log2), Refseq identifier and Description of the gene. (NA indicates no Refseq annotation).
(6.06 MB DOC)
NSC-enriched probes in ES versus NSC. Probesets enriched in group-wise comparisons: Column headings are probeset identifiers, T-statistic, P-value, Fold-Change (log2), Refseq identifier and Description of the gene. (NA indicates no Refseq annotation).
(7.45 MB DOC)
IPSC, ES-enriched probes in IPSC, ES versus NSC. Probesets enriched in group-wise comparisons: Column headings are probeset identifiers, T-statistic, P-value, Fold-Change (log2), Refseq identifier and Description of the gene. (NA indicates no Refseq annotation).
(10.18 MB DOC)
NSC-enriched probes in IPSC, ES versus NSC. Probesets enriched in group-wise comparisons: Column headings are probeset identifiers, T-statistic, P-value, Fold-Change (log2), Refseq identifier and Description of the gene. (NA indicates no Refseq annotation).
(12.08 MB DOC)
We would like to thank M. P. Hefferan for his help with confocal microscopy and M.L. Gage for editorial comments.