SEARCH

SEARCH BY CITATION

Keywords:

  • Reprogramming;
  • Pluripotent stem cells;
  • Induced pluripotent stem cells;
  • ROS;
  • Mitochondrion

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  9. References
  10. Supporting Information

The generation of induced pluripotent stem cells (iPSC) has enormous potential for the development of patient-specific regenerative medicine. Human embryonic stem cells (hESC) are able to defend their genomic integrity by maintaining low levels of reactive oxygen species (ROS) through a combination of enhanced removal capacity and limited production of these molecules. Such limited ROS production stems partly from the small number of mitochondria present in hESC; thus, it was important to determine that human iPSC (hiPSC) generation is able to eliminate the extra mitochondria present in the parental fibroblasts (reminiscent of “bottleneck” situation after fertilization) and to show that hiPSC have antioxidant defenses similar to hESC. We were able to generate seven hiPSC lines from adult human dermal fibroblasts and have fully characterized two of those clones. Both hiPSC clones express pluripotency markers and are able to differentiate in vitro into cells belonging to all three germ layers. One of these clones is able to produce fully differentiated teratoma, whereas the other hiPSC clone is unable to silence the viral expression of OCT4 and c-MYC, produce fully differentiated teratoma, and unable to downregulate the expression of some of the pluripotency genes during the differentiation process. In spite of these differences, both clones show ROS stress defense mechanisms and mitochondrial biogenesis similar to hESC. Together our data suggest that, during the reprogramming process, certain cellular mechanisms are in place to ensure that hiPSC are provided with the same defense mechanisms against accumulation of ROS as the hESC. STEM CELLS 2010;28:661–673


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  9. References
  10. Supporting Information

Embryonic stem cells (ESC) have the capacity to differentiate almost into any cell type in the adult organism, including germ line cells (reviewed in [1]) and therefore are commonly referred to as pluripotent cells. We and others have shown that ESC cultures are endowed with superior maintenance and repair systems to ensure genomic stability over multiple generations that allow ESC derived from murine blastocysts to suffer a 100-fold lower mutation frequency compared with embryonic fibroblasts derived from the same mouse strain [2–4]. This is achieved through a combination of low levels of stress generation, high activities of stress defense, and high activity and fidelity of repair mechanisms coupled to efficient elimination of ESC that accumulate mutations or DNA damage [2].

Reactive oxygen species (ROS) generated as a by-product of normal mitochondrial respiration are one of the principal sources of DNA damage. Murine and human ESC cultures possess highly effective antioxidant defense mechanisms whose activities diminish during differentiation and these rely upon expression of high levels of antioxidant enzymes [3, 4]. Furthermore, ESC cultures generate fewer ROS than most somatic cell types because of lower reliance upon oxidative phosphorylation and limited mitochondrial biogenesis [3, 4]. Human ESC cultures contain very few mitochondria [5, 6] but their numbers increase substantially during differentiation, suggesting that mitochondrial activity might play a role not only in the balance between pluripotency and differentiation but also in meeting an increased demand for Adenosine-5′-triphosphate (ATP) in the differentiated cell types. These changes have profound consequences for the cells since increased levels of ROS and DNA damage accumulate in ESC-derived differentiated cells, suggesting that they are less able to protect themselves against oxidative damage compared to the ESC themselves [3, 4].

The generation of induced pluripotent stem cells (iPSC) from murine and human adult somatic cells [7–9] is a recent and important development in the field of stem cell biology because of their potential therapeutic applications, drug screening, and disease modeling. These possibilities hold considerable promise for the development of patient-specific tissue replacement because iPSC lines are genetically identical to the somatic cells from which they were derived and therefore somatic cells differentiated from these are unlikely to be subjected to immune rejection after transplantation into the patient. In addition, iPSC derived from sufferers of diseases caused by known mutations can generate valuable in vitro disease models but to fulfill the therapeutic promise of iPSC technology their safety and comparability to ESC must be confirmed.

A major research focus has been to investigate the mechanism by which epigenetic reprogramming of the genome occurs during the iPSC derivation process. To date, there has been no comparison of the antioxidant defenses of human iPSC (hiPSC) with hESC. The control of mitochondrial biogenesis and respiratory activity in iPSC is of considerable interest given that these cells are derived from somatic cell types that normally possess a much larger complement of mitochondria than ESC cultures. The pre-implantation embryo does not engage in mitochondrial biogenesis [10]; therefore, the mitochondria present in the oocyte become progressively diluted at each cleavage division and by the blastocyst stage individual cells will possess only a small fraction of the original mitochondria. ESC are derived from the inner cell mass of the blastocyst, which helps to explain why they do not contain large numbers of mitochondria, but it was not clear if iPSC would share this characteristic since there was little evidence to suggest that mitochondrial biogenesis would be repressed during the iPSC derivation process. Mitochondrial DNA replication requires the translocation of transcription and replication factors encoded by the nuclear genome into the mitochondria. The principal proteins that regulate this process in mammals are mitochondrial transcription factor A (TFAM) and the heterotrimeric polymerase γ, which comprises a catalytic subunit (POLG) and two accessory subunits (POLG2) needed to recognize the RNA primer that initiates mitochondrial DNA replication.

Here, we demonstrate that the generation of ROS in hiPSC is broadly similar to that of hESC and that differentiation results in increasing ROS levels in parallel to hESC differentiation. Mitochondrial DNA copy number is low in the two independent hiPSC lines examined and moreover represents only a small percentage of the number of mitochondrial genomes present in the adult parental fibroblasts from which the hiPSC were derived. These data could have profound consequences for future iPSC-based therapies that are heavily dependent on the quality of the differentiated cells and their ability to maintain an intact genome.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  9. References
  10. Supporting Information

Derivation of iPSC

Adult human dermal fibroblasts in the log phase of growth (Lonza Group Ltd., Basel, Switzerland, http://www.lonza.com) were transduced with retroviral particles as described in [11]. Briefly, the plasmids pMXs-hNANOG, pMXs-hOCT4, pMXs-hSOX2, pMXs-hKLF4, and pMXs-hcMYC (Addgene Inc., Cambridge, MA, http://www.addgene.org) were packaged into retroviral particles by transfection into Phoenix Amphotropic cells using the Calcium Phosphate Transfection Kit (Sigma Aldrich, Poole, Dorset, U.K. http://www.sigmaaldrich.com). Retroviral transductions were performed two times at a 24-hour interval. Forty-eight hours after the first transduction, the adult dermal fibroblasts were disaggregated to single cells by trypsinization (0.05% Trypsin; Invitrogen, Carlsbad, CA, http://www.invitrogen.com) and then plated onto feeder layers of mitotically inactivated mouse embryonic fibroblasts in hESC culture medium at a density of 8,000 cells per well of a six-well plate. The feeder plates with retrovirus-treated cells were maintained at 37°C/5% CO2 for 21 days or until colonies of cells with a morphology similar to that of ESC appeared. These were mechanically dissected into several pieces and plated onto fresh feeder cells to develop additional colonies for characterization. We were able to obtain seven hiPSC clones from 50,000 transduced adult dermal fibroblasts, which results in 0.014% efficiency of reprogramming. This is similar to what has been published by other groups using the same constructs as ours [11]. Two of the hiPSC clones, named hiPSC clones 1 and 4 were characterized further and used for all analyses described in this study.

Culture and Differentiation of hESC/hiPSC

Wisconsin hESC line H9 and two hiPSC lines (iPSC clones 1 and 4) (all 46, XX) were cultured as discrete colonies on feeder layers of mitotically inactivated mouse embryonic fibroblasts as described in [12]. For more information refer to the supporting information Annex 1.

Quantitative Reverse Transcription-Polymerase Chain Reaction

Total RNA was extracted using TRIzol reagent (Invitrogen, Paisley, U.K.) according to manufacturer's instructions. Following DNaseI treatment using RQ1 DNaseI (Promega, Mannheim, Germany, http://www.promega.com), cDNA was synthesized using SuperScript Reverse Transcriptase (Invitrogen). Quantitative reverse transcription-polymerase chain reaction (RT-PCR) analysis was carried out using SYBR Green PCR master mix (Sigma Aldrich) and the primers are listed in supporting information Table 1. All samples were analyzed using an AB7900HT real-time analyzer and were normalized to GAPDH, RPL13A, and SDHA expression.

DNA Fingerprinting

To confirm that hiPSC clones 1 and 4 were of identical origin to adult fibroblasts, we carried out DNA fingerprinting. Total genomic DNA was extracted from all three samples and amplified with 11 microsatellite markers: D3S1358, vWA, D16S539, D2S1338, Amelogenin, D8S1179, D21S11, D18S51, D19S433, TH01, and FGA and analyzed on an ABI 377 sequence detector using Genotype software (Applied Biosystems, Foster City, CA, http://www.appliedbiosystems.com).

Immunocytochemistry

Cells were fixed in 4% (w/v) paraformaldehyde for 20 minutes. An additional permeabilization step (0.2% (v/v) Triton X-100 in PBS) was performed prior to staining with antibodies for internal cell markers. Blocking step was performed by incubation in 1% (w/v) bovine serum albumin or alternatively in 10% (v/v) goat serum. Cells were incubated with primary antibodies for 1 hour and secondary antibodies for 30–60 minutes. Primary antibodies used in this study are anti-OCT-4 (1:100; Millipore, Billerica, MA, http://www.millipore.com, catalogue no. MAB4419), anti-SOX2 (1:100; BD Biosciences, San Diego, http://www.bdbiosciences.com, catalogue no. 560301), anti-SSEA-4 (1:100; BD Pharmingen, San Diego, http://www.bdbiosciences.com, catalogue no. 560073), anti-TRA-1-60 (1:100; Millipore, catalogue no. MAB4360), anti-TRA-1-81 (1:100; Millipore, catalogue no. MAB4381), anti-CD31 (PECAM1) (1:100; BD Pharmingen, catalogue no. 558068), anti-β-III-Tubulin (1:100; Covance, Princeton, NJ, http://www.covance.com, catalogue no. MMS-435P), anti-AFP (1:100; Sigma-Aldrich, catalogue no. A8452), and anti-mouse IgG-FITC conjugated (1:200; Sigma Aldrich, catalogue no. F2012). The nuclei were counterstained with 10 μg/ml Hoechst 33342 (Molecular Probes [Invitrogen], catalogue no. H3570). The bright-field and fluorescent images were obtained using a Zeiss microscope and the AxioVision software (Carl Zeiss, Jena, Germany, http://www.zeiss.com).

Alkaline Phosphatase Staining

The alkaline phosphatase (AP) staining was carried out using the AP Detection Kit (refer to supporting information Annex 1).

Measurement of Mitochondrial Superoxide, Membrane Potential, and Mitochondrial Mass

These measurements were performed as described in [3, 4]. For more information please refer to supporting information Annex 1.

Karyotype Analysis of Human ESC

The karyotype of human ESC was determined by the standard G-banding procedure.

Teratoma Formation

To evaluate the developmental potential of hiPSC, cells were injected subcutaneously in adult SCID mice and maintained for 6–12 weeks. Subcutaenous injection is the most straightforward method of transplantation and does not compromise the welfare of the animals. Moreover, this recognized method produces teratomas containing cell types representative of all three primordial germ layers and allows the developing teratomas to be assessed by palpation [13]. Approximately 5 × 105 hiPSC, representative of either clones 1 or 4, were transplanted into the right flank of each animal. Similarly, for a positive control, approximately 5 × 105 human ESC (H9 cell line) were injected subcutaneously into adult SCID male mice. All cells were co-transplanted with 50 μl of Matrigel (BD Biosciences), which is known to enhance the success rate of teratoma formation. Four to six animals were injected in each group. After 6–12 weeks, mice were sacrificed, and tissues were dissected, fixed in Bouins overnight, processed and sectioned according to standard procedures, and counterstained with either hematoxylin and eosin or Massons Trichrome stain. Sections (5–8 μm) were examined using bright-field light microscopy and photographed as appropriate.

Relative Estimation of Mitochondrial Number by Q-PCR

Relative amounts of the mitochondrial gene(s) ND1, ND5, and MT-CYB (primers for these genes are shown in supporting information Table 1) were analyzed by Q-PCR analysis and normalized to the level of GAPDH/2 to normalize for the two copies of GAPDH in the genome compared to the single copies of genes of the mitochondrial genome. Each Q-PCR was carried out against a range of genomic standards of known concentration, to allow for relative quantitation and also to estimate primer efficiencies.

Statistical Analysis

One way analysis of variance (ANOVA test) was performed for the differences between undifferentiated hESC/hiPSC cells and adult dermal skin fibroblasts using SigmaStat 3.5 (Systat Software, Inc., Chicago, http://www.systat.com). For the increase in dihydrorhodamine (DHR) during differentiation a regression analysis was performed where the slope of regression was analyzed by Student's t test.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  9. References
  10. Supporting Information

Characterization of hiPSC Lines

Of the seven hiPSC lines derived according to the protocol detailed in Materials and Methods, two independent cell lines that were clonally derived from separate colonies (selected on the basis of morphology and high nucleus-to-cytoplasm ratio) were analyzed. These are named hiPSC clones 1 and 4 and were randomly selected for analysis. Both hiPSC clones 1 and 4 show morphology typical of hESC (Fig. 1A). Both hiPSC lines are karyotypically normal females and DNA fingerprinting confirms that they share a genome identical to that of the adult human dermal fibroblasts from which they were derived (Fig. 1B, 1C). Moreover, immunocytochemical staining of both hiPSC clones with SSEA4, SOX2, OCT4, TRA-1-60, TRA-1-81, and AP showed the same staining pattern as hESC (Fig. 2A). To investigate whether the exogenous expression of OCT4, SOX2, NANOG, KLF4, and c-MYC has occurred, we carried out RT-PCR analysis with primers that can pick up endogenous and exogenous transcripts and the total amount of each transcript (Fig. 2B). This analysis showed that all five factors are expressed endogenously in both hiPSC clones. Endogenous transcripts were also detected for KLF4 and c-MYC in the adult human dermal fibroblasts in addition to a low level of SOX2. These expressions have been observed by other investigators [11]; however, we also detected low-level endogenous OCT4 transcription in the parent fibroblasts, which has not been reported previously. This may result from enhanced sensitivity of the PCR primers used in this study. The different primers used to amplify total transcripts (endogenous plus viral transcript) may be less sensitive, which could account for our inability to detect transcripts other than c-MYC in the adult human dermal fibroblasts. In addition, we noticed that in hiPSC clone 1 the viral OCT4 and c-MYC have not been completely silenced; however, in hiPSC clone 4, only viral OCT4 is present. This corroborates with previously published data [11].

thumbnail image

Figure 1. Generation of hiPSC from adult human dermal fibroblasts (AHDF). (A): The morphology of hiPSC clones 1 and 4 at days 10 and 20 after plating on feeder cells. Scale bar, 100 μm. (B): DNA fingerprinting of hiPSC clones 1 and 4 and AHDF showing identical DNA genetic profiles. (C): Karyotype of hiPSC clones 1 and 4 and AHDF showing a normal 46, XX karyotype in all three cell types. Abbreviations: iPSC, induced pluripotent stem cells.

Download figure to PowerPoint

thumbnail image

Figure 2. Characterization of hiPSC. (A): Immunocytochemical analysis of human induced pluripotent stem cells (hiPSC) clones 1 and 4 for the expression of SSEA4, SOX2, OCT4, TRA-1-81, TRA-1-60, and alkaline phosphatase staining. Scale bar, 100 μm. (B): Reverse transcription-polymerase chain reaction analysis for the expression of total, endogenous, and exogenous OCT4, SOX2, KLF4, NANOG, and c-MYC. The housekeeping gene, RPL13A, is shown in the left-hand panel. Abbreviations: AHDF, adult human dermal fibroblasts; AHDF; IPSC1, hiPSC clone 1; IPSC4, hiPSC clone 4; NC, negative control.

Download figure to PowerPoint

To investigate whether continuous expression of some of the exogenous factors can affect the differentiation potential, both hiPSC clones were removed from the feeder layers and placed either in suspension culture to make EBs or attached to gelatin-coated plates in differentiation media as described in Materials and Methods (Fig. 3A). Quantitative RT-PCR analysis suggested that OCT4 expression was significantly downregulated during the 30-day EB differentiation time course similar to that of hESC (Fig. 3B). Similarly, the expression of mesodermal marker KDR (Fig. 3B), endodermal marker AFP (data not shown), and ectodermal marker NESTIN was upregulated during the differentiation time course in a pattern similar to that of differentiating hESC. Similar results were obtained by immunohistochemistry (Fig. 3C). We also tested the in vitro differentiation potential of hiPSC clones using monolayer differentiation conditions as described in the Materials and Methods section. Although there are differences in upregulation of various markers during the 4-week time course, it is clear that both hiPSC clones can give rise to differentiated cells expressing markers of endoderm (as shown by AFP expression), mesoderm (as shown by BRACHYURY expression), trophoectoderm (as shown by CDX2 expression), primitive endoderm (as shown by GATA6 expression), and primitive ectoderm (as shown by PAX6 expression; supporting information Fig. 1). Both hESC and hiPSC clone 4 also show gradual upregulation of primitive ectoderm marker, FGF5, whereas very little expression of this marker is seen during differentiation of hiPSC clone 1, suggesting impaired differentiation toward this extra-embryonic lineage (supporting information Fig. 1).

thumbnail image

Figure 3. In vitro differentiation of human induced pluripotent stem cells (hiPSC) clones. (A): Microphotographs showing hiPSC clone 1 differentiating under monolayer conditions and hiPSC clone 4 under embryoid body conditions. Scale bar, 100 μm. (B): Quantitative reverse transcription-polymerase chain reaction for the expression of OCT4, KDR, and NESTIN during the differentiation of H9 hESC line and hiPSC clones 1 and 4. (C): Immunocytochemistry for the expression of β3-tubulin, CD31, and AFP during the differentiation of hiPSC clones 1 and 4. Scale bar, 100 μm. Abbreviations: d10, day 10 of differentiation; d20, day 20 of differentiation; d30, day 30 of differentiation; EB, embryoid body.

Download figure to PowerPoint

To prove the pluripotency of hiPSC clones, we carried out teratoma formation assays for both hiPSC clones 1 and 4 and also control hESC. Human ESC (H9) grafted into SCID mice developed into teratomas that were restricted to the site of transplantation. Histological examination of teratomas revealed advanced differentiation of structures representative of all three embryonic germ layers (Fig. 4A). Similarly, hiPSC clone 4 produced teratomas containing a diverse range of differentiated structures and examples of tissues from each germ layer, including neuroepithelium, kidney, intestine, and cartilage (Fig. 4B). In contrast, only one of the transplants for hiPSC clone 1 produced a small tissue growth that did not possess the characteristic heterogeneous structure of a fully differentiated teratoma (Fig. 4C). The success rate of tumor formation was 75% for the human ESC, 33% for hiPSC clone 4, and 16% for hiPSC clone 1 (a similar number of human ESC and hiPSC clones were transplanted as detailed in Materials and Methods).

thumbnail image

Figure 4. In vivo differentiation of hiPSC clones. (A): Histological analysis of teratomas formed from grafted colonies of hESC (H9, positive control) in adult SCID male mice. Structures positively identified during histological analysis included the following: (A) neuroepithelium; (B) kidney; (C) intestine; (D) cartilage. Histological staining: Hematoxylin and eosin (A–C); Masson's trichrome (D). Scale bars: (A) 100 μm; (B–D) 200 μm. (B): Histological analysis of tumor tissues derived from grafted colonies of hiPSC clone 4 in adult SCID male mice. Tissues identified that possessed morphology typical of certain tissues commonly found in hESC-derived teratomas, including the following: (A) neuroepithelium; (B) kidney; (C) intestine; (D) cartilage. Histological staining: Hematoxylin and eosin (A–D). Scale bars: (a,b) 100 μm; (C) 750 μm; (D) 200 μm. (C): Histological analysis of tissue mass derived from grafted colonies of hiPSC clone one in adult SCID male mice. (A): Analysis revealed a small tissue nodule approximately 3 mm in diameter that lacked the diversity of differentiated structures normally found in a teratoma derived from pluripotent stem cells. (B): Higher magnification images showed that the tissue mass contained primarily connective tissue and a few gland-like structures that have been found in an ESC derived teratoma. Histological staining: Hematoxylin and eosin (a, b). Scale bars: (A) 750 μm; (B) 100 μm. (D): Quantitative reverse transcription-polymerase chain reaction analysis for expression of pluripotency markers in hESC and hiPSC. Data are presented as mean ± SEM (n = 3). The value for H9 hESC was set to 1 and all other values were calculated with respect to that. Abbreviations: hiPSC, human induced pluripotent stem cells.

Download figure to PowerPoint

Three new markers, ABCG2, REX1, and DNMT3B, have been suggested as potential tools for distinguishing between fully and partially reprogrammed hiPSC colonies [14]. In view of this, we carried out quantitative RT-PCR analysis for expression of a range of pluripotency markers including the three above-mentioned genes. In contrast to what has been described recently by Chan et al. [14], we found that ABCG2 and REX1 were expressed at similar levels in hiPSC clone 1 and hiPSC clone 4 when compared to human ESC (Fig. 4D); hence, this clone cannot be classified as partially reprogrammed. Other pluripotency markers such as DPPA3, DPPA2, DPPA5, LEFTY2, GDF3, and DNMT3B were expressed at lower levels in both hiPSC clones when compared to hESC, whereas SOX2 showed the reverse pattern. Notwithstanding this, both hESC and hiPSC clone 4 were able to downregulate the expression of all the markers investigated here during the 4-week time course of differentiation, whereas hiPSC clone 1 was unable to fully downregulate the expression of ABCG2 and REX1 and to a smaller extent the expression of TERT, LEFTY2, and DPPA4 (Fig. 4D). For therapeutic purposes the most desirable cell lines are the hiPSC clones that can differentiate in vitro into cell types representative of three germ layers without forming teratomas [15]. For this reason it is imperative to fully study these clones along with the ones that have the ability to give rise to teratoma and hESC. In view of this, we used both hiPSC clone 1 (non-teratoma-forming) and hiPSC clone 4 (teratoma-forming) in all subsequent experiments shown in this article.

Human iPSC Clones Show ROS Defense Mechanisms to Those of hESC

We used the fluorescent dye dihydrorhodamine (DHR) to quantify changes in cellular ROS levels in hiPSC clones 1 and 4 in addition to Wisconsin hESC line H9 and the normal adult human dermal fibroblasts (AHDF) from which hiPSC clones were derived. DHR is oxidized by cellular peroxides and peroxynitrite to fluorescent rhodamine 123, which is measured by flow cytometry (FL3 detector), and this probe shows not only that the ROS levels of undifferentiated H9 and hiPSC clones 1 and 4 are broadly similar but also that all three pluripotent cell types show similar increases in ROS concentrations during differentiation (Fig. 5A). The ROS levels in hiPSC clones 1 and 4 are significantly lower than that present in normal adult dermal fibroblasts (AHDF), which have even higher levels than the cells obtained from either H9 or hiPSc clones 1 and 4 after 4 weeks of differentiation.

thumbnail image

Figure 5. Characterization of intracellular superoxide levels in hiPSC clones and during differentiation. (A): DHR 123 (Molecular Probes/Invitrogen) was used to measure intracellular peroxides during the 4-week differentiation time course. One way analysis of variance (ANOVA test) was performed for the differences between undifferentiated hESC/hiPSC cells and ADHF. p < .001 (*) using Sigmastat 3.5 (Systat Software Inc.). For the increase in DHR during differentiation a regression analysis was performed where the slope of regression was analyzed by t test. The respective significance levels are as follows: for H9, p = .0508; for hiPSC clone 1, p = .0131; for hiPSC clone 4, p = .0009. Values were normalized to ADHF (100%) to compare different experiments (n = 2). (B): Quantitative reverse transcription-polymerase chain reaction for the expression of key genes involved in reactive oxygen species defense. Data are presented as mean ± SEM (n = 3). The value for AHDF was normalized to 1 and all other data were calculated with respect to that. Abbreviations: AHDF, adult human dermal fibroblasts; hiPSC, human induced pluripotent stem cells.

Download figure to PowerPoint

We performed quantitative RT-PCR analysis for a number of genes involved in cellular response to stress during the differentiation process to investigate whether there are differences in gene regulation between hiPSC and hESC (Fig. 5B). It is interesting to note that both OCT4 and NANOG levels were drastically downregulated during the differentiation process in both hiPSC clones in a fashion similar to that of the H9 hESC line. Both hiPSC clones show downregulation of GSR (glutathione reductase), SOD2 (superoxide dismutase 2), three transcript variants of MGST1 (microsomal glutathione S-transferase 1), and MAPK26 (mitogen-activated kinase 26; Fig. 5B) in a fashion similar to hESC. In addition to this, the hiPSC clone 4 shows downregulation of GSTA3 (glutathione S transferase), GPX2 (glutathione peroxidase 2), and HSPA1B (heat shock protein 1B) and upregulation of HSPB1 (heat shock protein 1) in a manner similar to hESC, corroborating with our previously published data [4]. Notwithstanding this, the hiPSC clone 1 showed different trends in regulation of expression of the four above-mentioned genes during the differentiation process. Although the global difference between hESC and hiPSC could be attributed to genetic variation between the individuals from whom the respective cell lines were derived, the differences between the two hiPSC lines are more surprising since these are clonally derived lines from the same fibroblast source and can only be attributed to differences in the copies of integrated transgenes and the level of reprogramming that may control at the epigenetic level the expression and regulation of these individual genes during the differentiation process.

Together these data suggest that both the teratoma-forming and non-teratoma-forming hiPSC clones show regulation of defense to ROS similar to that of human ESC cultures, suggesting that, during the reprogramming process, certain cellular mechanisms are in place to ensure that hiPSC are provided with the same defense mechanisms against accumulation of ROS as the hESC.

hiPSC Clones Show Mitochondrial Regulation Similar to That of hESC

The vast majority of cellular ROS arises from superoxide anions generated in the mitochondria, but since hESC cultures typically have a smaller number of mitochondria compared to those of their differentiated progeny cells, lower levels of superoxide may be generated. Superoxide concentrations can be measured using the MitoSOX reagent, which is targeted toward mitochondria and reacts preferentially with superoxide in the mitochondrial matrix. The relatively small number of mitochondria present in hESC should not generate large amounts of mitochondrial superoxide [4, 6]; hence, it is not surprising that AHDF show significantly higher staining with the MitoSOX reagent than that of either H9 or the hiPSC clones 1 and 4, respectively (Fig. 6A). However, it is interesting that both hiPSC lines show lower mitochondrial superoxide levels than those of the hESC line H9 (Fig. 6A). Mitochondrial superoxide is converted to hydrogen peroxide within the mitochondria by Cu-Zn and manganese superoxide dismutases [16], and since the hydrogen peroxide may diffuse out of the mitochondria more readily than superoxide, it is the peroxide that has the greatest effect on DHR oxidation. In view of this, it is interesting to note that the mitochondrial superoxide level is apparently lower in the hiPSC lines than in hESC (H9), whereas all H9 and hiPSC clones 1 and 4 show similar ROS levels determined by DHR staining (Fig. 5A).

thumbnail image

Figure 6. Characterization of mitochondrial biogenesis in hiPSC and during their differentiation. (A): The mitochondrial dye mitosox (Molecular Probes/Invitrogen) was used to determine mitochondrial superoxide levels using a Partec Flow cytometer (Partec, Muenster, Germany). FL3 fluorescence was used and autofluorescence levels subtracted. One way analysis of variance (ANOVA test) was performed for the differences between undifferentiated hESC/hiPSC and AHDF using SigmaStat 3.5 (Systat Software Inc.). p < .001 (*).Values were normalized to AHDF (100%) to compare different experiments (n = 2). (B): Mitochondrial membrane potential was measured using JC1 (Molecular Probes/Invitrogen) and the ratio between FL3 (autofluorescence subtracted) and FL1 was taken. H9 and both hiPSC clones have a higher membrane potential that decreases during differentiation. Values were normalized to ADHF (100%) to compare different experiments (n = 2). (C): Mitochondrial mass was determined using Nonyl acridine orange (Molecular Probes/Invitrogen) and Fl1 fluorescence measured. ANOVA test was performed for the differences between undifferentiated hESC/hiPSC cells and AHDF. p < .001 (*) using Sigmastat 3.5. Values were normalized to AHDF (100%) to compare different experiments (n = 2). (D): Relative estimation of mitochondrial copy numbers in hESC, hiPSC, and AHDF using quantitative PCR. Data are presented as mean ± SEM (n = 3). Abbreviations: AHDF, adult human dermal fibroblasts; hiPSC, human induced pluripotent stem cells; PCR, polymerase chain reaction.

Download figure to PowerPoint

Measurements on isolated mitochondria and whole cells [17, 18] show a steep correlation between the mitochondrial membrane potential and the rate of superoxide generation; therefore, we measured the mitochondrial membrane potential of both hiPSC clones using the JC-1 dye as described in Materials and Methods, but we could not find statistically significant differences between these cell lines and hESC line H9 (Fig. 6B). Differentiation of hESC results in a decrease of mitochondrial membrane potential and this is reflected in similar decreases during hiPSC differentiation.

Mitochondrial mass was measured by staining with Nonyl acridine orange, which demonstrates that AHDF have substantially higher levels of mitochondria than those of all three pluripotent cell types and that there is no significant difference between the mitochondrial mass present in hESC and hiPSC clones 1 and 4, respectively (Fig. 6C). To confirm the higher number of mitochondria in adult fibroblasts compared to those in hiPSC, we used quantitative PCR amplification from genomic DNA of three genes encoded by the mitochondrial genome (ND1, ND5, and MT-CYB). These data indicate that hESC H9 and hiPSC clones 1 and 4 show broadly similar mitochondrial genome copy numbers and therefore are likely to contain a similar number of intact mitochondria (Fig. 6D). Conversely, AHDF demonstrate much higher copy numbers.

Generation of hiPSC requires reprogramming of the somatic genome to express a pattern of genes consistent with a pluripotent phenotype. The basis of this reprogramming phenomenon is probably epigenetic modification of the genome. However, since the mitochondrial genome is not subject to DNA methylation to the same extent as the nuclear genome (CpG sites are under-represented on the mtDNA), we chose to focus our analysis upon mitochondrial proteins whose genes are encoded in the nuclear genome. Quantitative RT-PCR was performed using hESC (H9), hiPSC clones 1 and 4, and AHDF to determine the relative mRNA expression levels for a range of genes mainly involved in mitochondrial biogenesis, cytochrome oxidase function, and transport of proteins across the mitochondrial membrane (Fig. 7). Many of these genes are expressed at similar levels in hESC and hiPSC and show similar changes in expression levels during differentiation. Three key factors involved in mitochondrial DNA replication are mitochondrial transcription factor A (TFAM), the mitochondrial-specific DNA polymerase gamma (POLG), and its primer recognition accessory factor POLG2. All three genes show a similar trend in expression during differentiation of hESC and hiPSC, albeit it is interesting to note that the TFAM1 expression remains significantly higher in the differentiated progeny of hiPSC clone 1 even after 4 weeks of differentiation. These data also correlate to a low mitochondrial mass in the hESC and hiPSc clones.

thumbnail image

Figure 7. Comparison of hESC and hiPSC with respect to genes involved in mitochondrial biogenesis. Expression of key genes involved in mitochondrial biogenesis during the differentiation of hESC and hiPSC using quantitative real time reverse transcription-polymerase chain reaction. Data are presented as mean ± SEM (n = 3). The value for AHDF was normalized to 1 and all other data were calculated with respect to that. Abbreviations: AHDF, adult human dermal fibroblasts; hiPSC, human induced pluripotent stem cells.

Download figure to PowerPoint

Transcription of mtDNA also requires a small number of nucleus encoding protein including RNA polymerase, POLRMT, auxiliary factors for promoter recognition (TFBM1 and TFBM2) and a termination factor (MTERF) in addition to mitochondrial transcription factor TFAM [19]. Whereas the expression of POLRMT and TFBM1 is downregulated during differentiation of hESC and hiPSC (Fig. 7), there are notable differences in expression of TFBM2 and MTERF with the hESC downregulating their expression during differentiation and hiPSC upregulating the expression. There are several other notable examples where expression levels differ significantly between hESC and hiPSC (Fig. 7) such as the mitochondrial membrane components UCP2 and UCP4 (mitochondrial uncoupling proteins 2 and 4) [20]. UCP2 is significantly overexpressed in both hiPSC clones, and although both show downregulation of UCP2 during differentiation, the gene is never expressed at the low level present in the differentiated cells obtained from H9 hESC. Uncoupling protein 4 (UCP4) shows less overexpression in hiPSC clones 1 and 4. In hiPSC clone 4, the UCP4 expression changes during differentiation are highly similar to those of H9 hESC. The hiPSC clone 1 does not seem to be able to achieve a similar repression (Fig. 7); however, since ectopic expression of this protein in mammalian cells is known to reduce mitochondrial membrane potential, the higher expression of this and perhaps other members of the UCP group could contribute to the lower membrane potential observed in hiPSC clones 1 and 4.

Nuclear respiratory factors (NRF1 and NRF3) that are involved in mitochondrial DNA transcription and replication are also expressed in hESC and both hiPSC clones. NRF3 is downregulated in both hESC and hiPSC differentiation (Fig. 7); however, the NRF1 expression is still high in hiPSC clone 1 (unlike the hESC and hiPSC clone 4 where the expression is downregulated during differentiation).

Taken together, these data suggest that there are similarities in regulation of mitochondrial number and mass between hESC and hiPSC clones.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  9. References
  10. Supporting Information

The generation of induced pluripotent stem cell lines from human somatic tissues has the potential to circumvent many of the problems of immune rejection that would arise following transplantation of somatic cell types differentiated from hESC. We have shown previously that both murine and human ESC are highly proficient in antioxidant defense [3, 4]; therefore, in this study we sought to characterize the oxidative stress resistance and mitochondrial biogenesis of hiPSC cell lines generated from adult human skin fibroblasts.

We used retroviral transduction of five factors (OCT4, SOX2, KLF4, C-MYC, and NANOG) to generate hiPSC lines from human adult dermal fibroblasts following already published protocols [11]. The efficiency of reprogramming was low (0.014%) but similar to other published studies [11]. In vitro analysis and morphological assessment showed that both hiPSC clones were very similar to hESC and expressed all pluripotency markers including TRA1-60, TRA1-81, OCT4, SOX2, NANOG, and SSEA4. Under in vitro conditions, both clones were differentiated into cells derived from all three germ layers. Notwithstanding these positive results, the viral expression of OCT4 and c-MYC was not suppressed in hiPSC clone 1 and viral OCT4 was also maintained at low levels in hiPSC clone 4. Although injection of hiPSC clone 4 into SCID mice resulted in the formation of teratoma composed of cells derived from all three germ layers, proving its pluripotency, hiPSC clone 1 gave rise to only one small tissue growth (one out of six trials) that did not show the typical characteristics of a fully differentiated teratoma. Furthermore, hiPSC clone 1 was unable to downregulate expression of several marker genes during the in vitro differentiation. However, this hiPSC clone expressed ABCG2, REX1, and DNMT3B and cannot be classified as a partially reprogrammed clone [14]. We are more inclined to believe that the lack of differentiation capability and maintenance of pluripotent gene expression can be explained by the inability to switch off the viral c-MYC gene expression. It is conceivable that iPSC clone 1 is perhaps locked into a self-renewal state and has lost some of the differentiation capabilities because of its inability to switch off the expression of pluripotent genes during the differentiation process. It is well known in the literature that c-MYC is able to inhibit cell differentiation through different mechanisms such as enhancing cell proliferation through inhibition of cell cycle inhibitors [21], repressing differentiation-inducing genes in its role as co-repressor [22], inducing expression of miRNAs that can attenuate ESC differentiation [23] or acting as chromatin remodeler [24]. Both hiPSC clones 1 and 4 show similar cell cycle characteristics and population doubling time (data not shown). Furthermore, there is a greater maintenance of pluripotent markers during differentiation of hiPSC clone 1, which leads us to speculate that the remaining viral c-MYC is inhibiting hiPSC differentiation, resulting in the lack of fully formed teratomas. Most importantly, the non-teratoma-forming hiPSC clone 1 is able to give rise to differentiated cells belonging to three germ layers under two different types of in vitro conditions, making this clone desirable for further clinically relevant studies. In view of this, we characterized both the teratoma-forming and non-teratoma-forming hiPSC clones with respect to ROS defense and mitochondrial regulation.

Our study shows that both of the hiPSC clones reduce their mitochondrial genome copy number to the levels typical of hESC and moreover are capable of mounting a similar oxidative stress defense to hESC. In spite of normal ROS stress defense in the partially reprogrammed hiPSC clone 1, there are differences between this clone and hiPSC clone 4 and hESC as far as the expression of some key antioxidant genes and genes involved in mitochondrial biogenesis (GSTA3, GPX2, HSPA1B, MAPK26, MGST1C, NRF1, UCO4, and TFAM1) is concerned during the differentiation process. These differences extend to several other genes involved in mitochondrial function (UCP2, UCP4), which are integral components of the inner mitochondrial membrane that may function to control the production of superoxide and other downstream ROS [20] and also function as mitochondrial transporter proteins that create proton leaks across the inner mitochondrial membrane, thus uncoupling oxidative phosphorylation from ATP synthesis. As a result, energy is dissipated in the form of heat. These marker genes are normally downregulated during the differentiation of pluripotent cell lines as shown by us previously [3, 4] and in this study. However, their expression is maintained during the differentiation of hiPSC clone 1. This could be due to either poorer differentiation of this clone or epigenetic changes in the promoters of those genes that do not enable the normal downregulation during the differentiation process.

The hiPSC clone 4 passes all the current tests of pluripotency, although it is notable that even this cell line has not completely repressed the viral OCT4 transgene. In view of its apparent pluripotency one might expect antioxidant gene expression similar to that of hESC, which is the case. Differentiation of hiPSC clone 4 results in an increase of intracellular peroxides and downregulation of key genes involved in mounting the stress defense against ROS in a manner similar to that of hESC. Most importantly, the process of generating hiPSC seems to be able to reduce the larger numbers of mitochondria present in the adult fibroblasts to numbers similar to those found in hESC, although the mechanism of how this occurs is currently unclear. There are however some minor differences in the level of expression of some of the genes involved in mitochondrial biogenesis between hiPSCs clone and hESC and their regulation during the differentiation process. This may be an important issue for further investigation and potential use of hiPSC in regenerative medicine since the formation of some cell types, for example, neurons and skeletal muscle cells, is heavily dependent upon mitochondrial biogenesis. It is obvious that the teratoma-forming hiPSC clone 4 shows more similarity to hESC with respect to regulation of key genes involved in mitochondrial DNA replication. However, the most important finding from our work is the reduction in mitochondrial mass and mitochondrial number during the reprogramming process, which is likely to result in reduced mitochondrial superoxide levels in hiPSC (both teratoma-forming and non-teratoma-forming clones) compared to those of the parent fibroblast population from which they were derived.

In summary, our studies show that both teratoma-forming and non-teratoma-forming hiPSC clones have antioxidant defense mechanisms similar to those of hESC. These data support the opinion that hiPSC are functionally equivalent to hESC on the basis of our current measures of pluripotency, but the observed differences in the expression of genes important for mitochondrial regulation suggests that further studies are necessary to confirm the functionality, safety, and efficacy of hiPSC as an alternative source of clinically useful cell types for hESC.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  9. References
  10. Supporting Information

This study was supported by MRC grant No. G0301182, Newcastle University, funds for research in the field of Regenerative Medicine through the collaboration agreement from the Conselleria de Sanidad (Generalitat Valenciana) and the Instituto de Salud Carlos III (Ministry of Science and Innovation) and Sir James Knott Trust. We thank Addgene for providing the retroviral constructs, ATCC for the Phoenix Amphotropic cell lines, Complement Genomics for their help with DNA fingerprinting, Dr. Jerome Evans for help with the karyotyping, and Dennis Kirk for help with technical issues.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  9. References
  10. Supporting Information

Additional supporting information available online.

FilenameFormatSizeDescription
STEM_307_sm_suppinfoAnnex1.pdf79KSupporting Information Annex 1
STEM_307_sm_suppinfofigure1.tif512KSupporting Information Figure 1
STEM_307_sm_suppinfoTable1.pdf43KSupporting Information Table 1

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.