This study demonstrates the reprogramming of somatic cells towards pluripotency in vivo without the generation of teratomas. We used hydrodynamic tail vein injection of plasmid DNA encoding the Yamanka factors to induce the in vivo reprogramming of adult hepatocytes into cells of enhanced pluripotency.
Induced pluripotent stem (iPS) cells that result from the reprogramming of somatic cells to a pluripotent state by forced expression of defined factors are offering new opportunities for regenerative medicine. Such clinical applications of iPS cells have been limited so far, mainly due to the poor efficiency of the existing reprogramming methodologies and the risk of the generated iPS cells to form tumors upon implantation.
We hypothesized that the reprogramming of somatic cells towards pluripotency could be achieved in vivo by gene transfer of reprogramming factors. In order to efficiently reprogram cells in vivo, high levels of the Yamanaka (OKSM) transcription factors need to be expressed at the target tissue. This can be achieved by using different viral or nonviral gene vectors depending on the target tissue. In this particular study, hydrodynamic tail-vein (HTV) injection of plasmid DNA was used to deliver the OKSM factors to mouse hepatocytes. This provided proof-of-evidence of in vivo reprogramming of adult, somatic cells towards a pluripotent state with high efficiency and fast kinetics. Furthermore no tumor or teratoma formation was observed in situ.
It can be concluded that reprogramming somatic cells in vivo may offer a potential approach to induce enhanced pluripotency rapidly, efficiently, and safely compared to in vitro performed protocols and can be applied to different tissue types in the future.
Ethical concerns about the use of embryonic stem cells have been one of the limitations in the development of stem cells research. However, the discovery by Yamanaka and colleagues that the forced expression of four transcription factors (namely: Oct3/4, Klf4, Sox2, c-Myc (OKSM)) could lead to in vitro reprogramming of somatic cells into pluripotent cells, named induced pluripotent stem (iPS) cells, has opened new opportunities. Viral1-5 and nonviral6-10 gene transfer, protein cytoplasmic translocation11,12, and miRNA13-16 transfection are among the various methods used today to generate iPS cells. Yet, such reprogramming methodologies suffer from various issues that are restricting their translation into the clinic, such as: a) severe limitations in efficiency of cell reprogramming; b) the predominant use of viral vectors; c) long and multi-step protocols of culturing conditions; and d) the risk of tumorigenicity by implantation of the in vitro generated iPS cells17-20.
Gene transfer of defined transcription factors3-5 by retroviruses is the most commonly used method to reprogram somatic cells. However, it contains the risks from the possibility of insertional mutagenesis, stable transduction and long-term proto-oncogene expression21,22. Nonviral gene transfer vectors such as plasmid DNA6,7,23 or RNA10 delivery using liposomes or electroporation have also been explored. While safer compared to viruses, those vectors offer significantly limited transduction and reprogramming efficiency24,25 .
One of the central dogmas of this emerging field is that in vivo implantation of iPS cells will lead to their uncontrolled differentiation and the formation of a tumor-like mass (teratoma), composed of various tissues from the three different germ layers. Therefore, the concept of reprogramming to pluripotency has been primarily focused on in vitro manipulations of primary extracted somatic cells (most commonly fibroblasts). However, as described above, this principle suffers from long and complex cell culture protocols, including multiple rounds of treatment (gene transfer, growth factors, antibiotics, antioxidants) that may themselves enhance the risks for teratoma formation or other forms of tumorigenesis upon implantation of the cells24-28.
We propose that reprogramming of adult somatic cells in vivo by the transient overexpression of the OKSM transcription factors does not lead to subsequent formation of teratomas29. The method to achieve that will depend on achieving high levels of transcription factor expression within the target tissue in the safest possible way. In this study, we selected an established virus-free gene transfer technology that has been shown to target efficiently liver in order to reprogram cells to pluripotency in vivo. This technology involves large-volume, rapid hydrodynamic tail vein (HTV) injection of plasmid DNA30,31. High expression levels of transfected genes can be achieved in adult hepatocytes following this nonviral gene transfer method. Moreover, this technology minimizes the potential risks from use of viral vectors (e.g. permanent genomic insertion, insertional mutagenesis).
1. In vivo Reprogramming of Liver Tissue by HTV Administration of pDNA
2. Liver perfusion and Isolation of Primary Hepactocytes for qRT-PCR Studies
3. Gene Expression Study of Isolated Hepatocytes for Upregulation of Pluripotent Genes
4. Flow Cytometry Analysis of Enhanced Pluripotency at the Protein Level
5. Immunostaining of Liver Sections for Markers of Pluripotency
6. Alkaline Phosphatase (ALP) Staining of Liver Tissue Sections
7. Quantification of Albumin and Liver Enzyme Levels in Serum to Assess Liver Toxicity
8. Hematoxylin and Eosin (H&E) and Periodic Acid-Schiff (PAS) Staining of Liver Sections to Assess Liver Toxicity
Figure 1 shows the overview of the procedure that involves the gene transfer of OKSM encoding plasmids to mice liver and the different techniques performed to observe in vivo cell reprogramming. Following HTV injection of the plasmids, a significant increase in gene expression of the transfected reprogramming factors (Oct3/4, Sox2, Klf4, and cMyc) at the mRNA level was observed on day 2 after injection. The expression of these factors decreases over time after injection, as shown in Figure 2a. Regarding the expression of endogenous pluripotency markers (Nanog, Ecat1 and Rex1), their levels were significantly upregulated compared to those in hepatocytes from saline-injected animals on day 2 and 4 after injection, and were back to baseline levels from day 8 onward (Figure 2b). At the same time dedifferentiation of the hepatocyte population was confirmed by the downregulation of hepatocyte-specific genes (Alb, Aat, and Trf) that was statistically significant on day 4 and reached baseline levels from day 8 onward (Figure 2d).
The expression of Oct3/4 and Nanog at the protein level was investigated by flow cytometry. As shown in Figure 2c, only Oct3/4 is expressed on day 1 after HTV injection, while for the expression of the endogenous pluripotency marker Nanog it was necessary to wait until day 4.
The occurrence of in vivo cell reprogramming was further confirmed by immunohistochemical analysis of liver tissues with anti-OCT4, anti-SOX2 and anti-Nanog antibodies and by specific staining for ALP activity. Positive cells for all markers and enzymatic activity are reproducibly found in the liver tissues from OKSM-injected animals but not in the saline-injected controls (Figure 3).
The possible toxicity side-effects and teratoma formation from in vivo cell reprogramming by HTV injection of pDNA were investigated by quantification of liver enzyme levels in serum, as well as H&E and PAS staining of liver tissue sections over a period of 120 days. Transient and not severe signs of tissue damage were observed until day 2, but not longer. No formation of teratomas or any sign of dysplasia or morphological alterations were observed for the period of study (Figure 4). There were no hepatic structural or functional abnormalities throughout the course of the study for any of the animals, as confirmed by albumin and liver enzyme levels and glycogen staining of tissue sections (Figures 4b and c).
Figure 1. The schematic overview of in vivo reprogramming procedure and its analysis. The protocol involves two main stages: (i) administration of reprogramming factors in vivo and (ii) tissue extraction and sample analysis. Click here to view larger figure.
Figure 2. In vivo overexpression of Yamanaka transcription factors in adult mouse liver. Balb/C mice were HTV injected with 0.9% saline alone, 75 μg of pCX-OKS-2A and 75 μg pCX-cMyc in 0.9% saline and at days 2, 4, 8, 12, 24, RT-qPCR analysis of hepatocytes was performed to determine the relative gene expression of: (a) transfected transcription factors (OKSM) and (b) endogenous pluripotency markers; (c) flow cytometry analysis of OCT3/4 positive and Nanog positive cells; (d) relative gene expression of hepatocyte markers as determined by RT-qPCR. All gene expression levels were normalized to saline HTV-injected group. (* p<0.05 indicates statistically significant difference compared to saline HTV-injected groups, obtained by the analysis of variance and Tukey's pairwise comparison). Figure adapted from Yilmazer et al.29 Click here to view larger figure.
Figure 3. In vivo cell reprogramming on adult mouse liver tissue by immunohistochemistry. Balb/C mice HTV injected with 0.9% saline alone, 75 μg of pCX-OKS-2A and 75 μg pCX-cMyc in 0.9% saline. At day 4, livers were collected and frozen tissue sections were stained with anti-OCT4, anti-SOX2 or anti-Nanog antibodies to assess immunoreactivity, or BCIP/NBT to determine ALP activity in the tissue (40X). Scale bars represent 100 μm. Figure adapted from Yilmazer et al.29 Click here to view larger figure.
Figure 4. The effect of in vivo cell reprogramming on liver functionality and histopathology. Balb/C mice HTV injected with either 75 μg of pCX-OKS-2A and 75 μg pCX-cMyc in 0.9% saline or 0.9% saline only. On days 2, 4, 8, 12, 50, and 120 liver tissues and sera were isolated and processed for: (a) H&E staining; (b) levels of liver enzymes; (c) levels of albumin; (d) PAS staining. Representative images were captured with light microscopy (10X). Scale bars represent 100 μm. Figure adapted from Yilmazer et al.29 Click here to view larger figure.
This study provides proof-of-principle evidence of in vivo cell reprogramming towards pluripotency following the efficient transfer and overexpression of the OKSM factors to the adult mouse liver. With the help of different techniques such as qRT-PCR, flow cytometry, IHC or serum analysis, the target tissue was investigated to observe in vivo cell reprogramming. Very rapidly from the induction of OKSM factor overexpression in the tissue, pluripotency markers were upregulated at the mRNA and protein levels (within 48 hr). Throughout these experiments, between 5-15% of cells extracted from the tissue were shown to be Nanog and Oct3/4 positive. Furthermore, no structural or functional abnormalities were observed in the liver throughout 120 days. Expression of transgenes by HTV injection of plasmid DNA in tissues other than liver (lung, spleen, kidney and heart) has been previously reported, however the levels of gene expression were at least 3 orders of magnitude lower than those in hepatocytes30. The data in this study indicates that reprogramming of somatic cells towards a pluripotent state in vivo is safe, fast and efficient. In view of the majority of existing in vitro cell reprogramming methodologies that generally achieve pluripotent (iPS) cell generation within 3 weeks with less than 1% of efficiency32, in vivo reprogramming to pluripotency may offer very interesting alternatives. We hypothesize that the concept of in vivo reprogramming to pluripotency could be applied to various types of somatic cells with high efficiency (dependent on the vector used), however the level of functional pluripotency achieved in the in vivo reprogrammed cells will need to be determined. Moreover, with the development of improved protocols and technologies, in vivo reprogrammed cells could potentially be extracted and utilized for regenerative medicine purposes.
There are some critical steps to follow in order to reprogram adult, somatic cells in vivo with high efficiency. Achieving high levels of gene expression in the target tissue is a prerequisite. In this study this was illustrated by utilizing the efficient transfer and expression of plasmid DNA to hepatocytes using HTV injection. The most important factors for HTV injection are the volume, which should be between 8-10% of the total body weight of the animal, and the speed of injection, which must be performed within 5-7 sec. Mice younger than 6 weeks old should not be used, as HTV injection can cause more tissue damage in the liver, which can affect the efficiency of cellular reprogramming. The genetic background of the animals had no impact in determining the efficiency of in vivo reprogramming, as we have previously reported similar levels of reprogramming in both Balb/C and TNG-A mice, the latter being a transgenic strain that carries the eGFP reporter inserted in the Nanog locus and is bred on Sv129 background29.
We speculate that in vivo cell reprogramming to a pluripotent state (that will need to be determined) can be applied to other tissue types provided that the OKSM factors are efficiently delivered to target cells, allowing high levels of gene expression. A variety of nonviral and viral gene delivery vectors have been studied in preclinical and clinical trials in order to target different cell and tissue types33. For example, for neural or ocular tissues adeno-associated viruses have been the preferred vector choice for efficient in vivo gene transfer34-37, whereas in muscle tissue naked plasmid DNA injections have been shown to result in relatively high levels of gene expression38,39. Therefore, the method of gene transfer vector should be carefully selected depending on the tissue type in order to achieve efficient and rapid in vivo cell reprogramming toward a pluripotent state.
The authors have nothing to disclose.
This work was partially supported by the Engineering and Physical Sciences Research Council (EPSRC, Swindon, United Kingdom) and the European Commission FP-7 Programs NANONEUROHOP (PIEF-GA-2010-276051) and ANTICARB (HEALTH-2008-20157). AY was supported by a full-time PhD scholarship from the Ministry of Education (Turkey) and IL acknowledges partial support from the British Council and the CAIXA Foundation (Spain). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Name of Reagent/Material | Company | Catalog Number | Comments |
pCX-OKS-2A pDNA | Addgene | 19771 | Obtained from Addgene as bacterial stab, plasmid production performed in Plasmid Factory (Germany) |
pCX-cMyc | Addgene | 19772 | Obtained from Addgene as bacterial stab, plasmid production performed in Plasmid Factory (Germany) |
0.22 μm filter | Milliopore | SLGP033RS | |
Isoflurane | Abbott | B505 | |
HBSS buffer | Sigma-Aldrich | H6648 | Ca2+ and Mg2+ free, with bicarbonate |
Liver Digest Medium | Gibco | 17703-034 | |
Hepatocyte Wash Medium | Gibco | 17704-024 | |
100 μm cell strainer | BD Biosciences | 352360 | |
Nucleospin RNA II kit | Macherey-Nagel | 740955.25 | Kit for RNA isolation from cells and tissues |
iScript cDNA synthesis kit | Bio-Rad | 170-8890 | |
iO SYBR Green Supermix | Bio-Rad | 170-8880 | |
Oct3/4 forward primer | Sigma | sequence given in comments | TGAGAACCTTCAGGAGATATGCAA |
Oct3/4 reverse primer | Sigma | sequence given in comments | CTCAATGCTAGTTCGCTTTCTCTTC |
Sox2 forward primer | Sigma | sequence given in comments | GGTTACCTCTTCCTCCCACTCCAG |
Sox2 reverse primer | Sigma | sequence given in comments | TCACATGTGCGACAGGGGCAG |
C-myc forward primer | Sigma | sequence given in comments | CAGAGGAGGAACGAGCTGAAGCGC |
C-myc reverse primer | Sigma | sequence given in comments | TTATGCACCAGAGTTTCGAAGCTGTTCG |
Nanog forward primer | Sigma | sequence given in comments | CAGAAAAACCAGTGGTTGAAGACTAG |
Nanog reverse primer | Sigma | sequence given in comments | GCAATGGATGCTGGGATACTC |
Ecat1 forward primer | Sigma | sequence given in comments | TGTGGGGCCCTGAAAGGCGAGCTGAGAT |
Ecat1 reverse primer | Sigma | sequence given in comments | ATGGGCCGCCATACGACGACGCTCAACT |
Rex1 forward primer | Sigma | sequence given in comments | ACGAGTGGCAGTTTCTTCTTGGGA |
Rex1 reverse primer | Sigma | sequence given in comments | TATGACTCACTTCCAGGGGGCACT |
Alb forward primer | Sigma | sequence given in comments | GTTCGCTACACCCAGAAAGC |
Alb reverse primer | Sigma | sequence given in comments | CCACACAAGGCAGTCTCTGA |
Aat forward primer | Sigma | sequence given in comments | CAGAGGAGGCCAAGAAAGTG |
Aat reverse primer | Sigma | sequence given in comments | ATGGACAGTCTGGGGAAGTG |
Trf forward primer | Sigma | sequence given in comments | ACCATGTTGTGGTCTCACGA |
Trf reverse primer | Sigma | sequence given in comments | ACAGAAGGTCCTTGGTGGTG |
B actin forward primer | Sigma | sequence given in comments | GACCTCTATGCCAACACAGT |
B actin reverse primer | Sigma | sequence given in comments | AGTACTTGCGCTCAGGAGGA |
BD Cytofix fixation buffer | BD Biosciences | 560585 | |
1x BD Perm/Wash buffer | BD Biosciences | 560585 | |
anti-mouse OCT4-PerCP-Cy5.5 | BD Biosciences | 560585 | |
anti-mouse Nanog-PE | BD Biosciences | 560585 | |
2-Methylbutane | Sigma-Aldrich | M32631 | |
X100-Triton | Sigma-Aldrich | X100-500ML | |
Goat serum | Sigma-Aldrich | G9023 | |
BSA | Sigma | A2153 | |
rabbit pAb anti-OCT4 | Abcam | ab19857 | Use at a concentration of 3 μg/ml |
rabbit pAb anti-SOX2 | Abcam | ab97959 | Use at a concentration of 1 μg/ml |
rabbit pAb anti-Nanog | Abcam | ab80892 | Use at a concentration of 1 μg/ml |
mouse mAb anti-SSEA1 | Abcam | ab16285 | Use at a concentration of 20 μg/ml |
goat pAb anti-rabbit IgG labeled with Cy3 | Jackson ImmunoResearch Laboratories Inc. | 111-165-003-JIR | Use at a concentration of 1/250 |
goat pAb anti-mouse IgG labeled with Cy3 | Jackson ImmunoResearch Laboratories Inc. | 115-165-003-JIR | Use at a concentration of 1/250 |
VECTASHIELD mounting medium with DAPI | Vector Laboratories | H-1200 | |
BCIP/NBT liquid substrate system | Sigma | B1911 | |
Aqueous mounting medium | Sigma | ||
16% Paraformaldehyde | Fisher | AA433689M | Stock solution is 16%, use it at 4% |
Hematoxylin and eosin | Sigma | GH53/HT110216 | |
PAS staining | Sigma | 395B |