Telomerase Reverse Transcriptase Has an Extratelomeric Function in Somatic Cell Reprogramming*

Background: The contribution of TERT to cellular reprogramming is unclear. Results: KO of TERT reduces the reprogramming efficiency of somatic cells, and continuous growth induces chromosome abnormalities; enzymatically inactive TERT rescues the efficiency. Conclusion: TERT has a telomerase-independent role in somatic cell reprogramming. Significance: Understanding the cellular reprogramming process is crucial for the clinical applications of induced pluripotent stem cells. Reactivation of the endogenous telomerase reverse transcriptase (TERT) catalytic subunit and telomere elongation occur during the reprogramming of somatic cells to induced pluripotent stem (iPS) cells. However, the role of TERT in the reprogramming process is unclear. To clarify its function, the reprogramming process was examined in TERT-KO somatic cells. To exclude the effect of telomere elongation, tail-tip fibroblasts (TTFs) from first generation TERT-KO mice were used. Although iPS cells were successfully generated from TERT-KO TTFs, the efficiency of reprogramming these cells was markedly lower than that of WT TTFs. The gene expression profiles of iPS cells induced from TERT-KO TTFs were similar to those of WT iPS cells and ES cells, and TERT-KO iPS cells formed teratomas that differentiated into all three germ layers. These data indicate that TERT plays an extratelomeric role in the reprogramming process, but its function is dispensable. However, TERT-KO iPS cells showed transient defects in growth and teratoma formation during continuous growth. In addition, TERT-KO iPS cells developed chromosome fusions that accumulated with increasing passage numbers, consistent with the fact that TERT is essential for the maintenance of genome structure and stability in iPS cells. In a rescue experiment, an enzymatically inactive mutant of TERT (D702A) had a positive effect on somatic cell reprogramming of TERT-KO TTFs, which confirmed the extratelomeric role of TERT in this process.


Reactivation of the endogenous telomerase reverse transcriptase (TERT) catalytic subunit and telomere elongation occur during the reprogramming of somatic cells to induced pluripotent stem (iPS) cells. However, the role of TERT in the reprogramming process is unclear. To clarify its function, the reprogramming process was examined in TERT-KO somatic cells. To exclude the effect of telomere elongation, tail-tip fibroblasts (TTFs) from first generation TERT-KO mice were used. Although iPS cells were successfully generated from TERT-KO TTFs, the efficiency of reprogramming these cells was markedly lower than that of WT TTFs. The gene expression profiles of iPS cells induced from TERT-KO TTFs were similar to those of WT iPS cells and ES cells, and TERT-KO iPS cells formed teratomas that differentiated into all three germ layers.
These data indicate that TERT plays an extratelomeric role in the reprogramming process, but its function is dispensable. However, TERT-KO iPS cells showed transient defects in growth and teratoma formation during continuous growth. In addition, TERT-KO iPS cells developed chromosome fusions that accumulated with increasing passage numbers, consistent with the fact that TERT is essential for the maintenance of genome structure and stability in iPS cells. In a rescue experiment, an enzymatically inactive mutant of TERT (D702A) had a positive effect on somatic cell reprogramming of TERT-KO TTFs, which confirmed the extratelomeric role of TERT in this process.
The potential roles of induced pluripotent stem (iPS) 4 cells in clinical regenerative medicine and the mechanisms involved in their induction by the reprogramming of somatic cells have been studied previously (1). However, despite a wealth of research, the precise mechanisms underlying the reprogramming process are not fully understood (2). Overexpression of the four reprogramming factors, Klf4, Sox2, Oct3/4, and c-Myc, which are also known as the Yamanaka factors, can induce pluripotency in somatic cells (1,3); however, only a minority of somatic cells can be reprogrammed (4). Recent reports suggest that these four transcription factors can induce other cell types in addition to pluripotent cells (5,6). Moreover, stochastic mechanisms are also thought to be involved in the generation of iPS cells (7,8). Therefore, somatic cell reprogramming appears to be the result of multiple processes.
Telomeres are repetitive TTAGGG sequences located at the end of chromosomes that protect against abnormal fusions, recombination, and degradation (9). Telomere shortening occurs with each round of cell division (10), and this process may cause genomic instability. Telomeres can be elongated by the addition of TTAGGG repeat sequences, a process that is mediated by telomerase reverse transcriptase (TERT), the catalytic subunit of the enzyme telomerase. Reprogramming of somatic cells requires multiple cell divisions; therefore, iPS cells must be able to sustain telomerase activity and maintain elongated telomeres (11). Although TERT is inactivated in most mature somatic cells, its constitutive activation in stem cells and germ cells allows life-long cellular proliferation (12,13). Telomeres play a role in the proliferation and differentiation of cells (14), and endogenous TERT expression is induced during somatic cell reprogramming (15). During this process, somatic cells acquire indefinite proliferative capacity, as well as the abil-ity to differentiate into the three germ layers as follows: ectoderm, endoderm, and mesoderm. Some reports suggest that TERT increases the efficiency of somatic cell reprogramming; for example, the induction of TERT enhances the generation of human iPS cells from fetal, neonatal, and adult primary cells, as well as those from dyskeratosis congenita patients (16,17). By contrast, another study using telomerase RNA component (TERC)-KO somatic cells showed that elongation of the telomere does not affect the reprogramming efficiency of somatic cells when the telomere length is not already shortened at the beginning of the reprogramming process (11).
It is possible that TERT plays a role in the reprogramming of somatic cells that is independent of telomere elongation. To examine this hypothesis, reprogramming experiments were performed using somatic cells from first generation (F1) TERT-KO mice, which have long telomeres. TERT-KO somatic cells could be reprogrammed to iPS cells by introducing the four reprogramming factors; however, the efficiency of reprogramming was lower than that of WT somatic cells. In rescue experiments, an enzymatically inactive mutant of TERT (D702A) improved the reprogramming efficiency of TERT-KO somatic cells. These data suggest that TERT has extratelomeric activity during the reprogramming of somatic cells.

Induction of iPS Cells from Adult Tail-tip Fibroblasts (TTFs)-
The induction of iPS cells from adult mouse TTFs was performed as described previously (18). To estimate the status of cellular reprogramming, a retroviral DsRed vector and a lentiviral early transposon Oct4 and Sox2 enhancer (EOS) vector were introduced into the cells to monitor the silencing activity of the retrovirus vector and the promoter activity of Oct3/4 and Sox2, respectively. Four days after induction, cells were reseeded on STO feeder layers, and the numbers of colonies were counted from day 11 of the culture. For the TERT-KO TTF rescue experiment, WT TERT or enzymatically inactive TERT (D702A mutant) was introduced into TERT-KO TTFs 1 day prior to induction by the four reprogramming factors. The TERT-WT and TERT D702A lentiviruses were generated using HEK 293T cells, as described previously (18).
Cell Culture-STO feeder cells were treated with 40 g/ml mitomycin C for 2 h and plated at a density of 1 ϫ 10 6 cells per 55 cm 2 . The iPS cells were cultured on mitomycin C-treated STO cells in knockout DMEM (Invitrogen) containing 15% FBS, ESGRO (Millipore), L-glutamine, nonessential amino acids, ␤-mercaptoethanol, 50 units/ml penicillin/streptomycin, and 20 g/ml ascorbic acid (19). For RNA extraction, feeder cells were depleted by two rounds of incubation on a 0.2% gelatin-coated dish.
EdU Assay-Cell cycle entry was analyzed using Click-iT EdU flow cytometry assay kits (Invitrogen), according to the manufacturer's instructions. Briefly, WT and TERT-KO iPS cells were seeded into 6-well plates at a density of 1 ϫ 10 5 cells per well. The following day, the cells were treated with 10 M EdU for 1.5 h and then washed with 1% BSA in PBS. The cells were fixed with Click-iT fixative at room temperature for 15 min. After a further wash, the cells were permeabilized by incubation with Click-iT saponin-based permeabilization and wash reagent for 15 min. Click-iT reaction mixture was then added to the permeabilized cells. Finally, the cells were washed with 1ϫ Click-iT saponin-based permeabilization and wash reagent, and then analyzed using the FACSCalibur platform (BD Biosciences).
Embryoid Body (EB) Formation-For the formation of EBs, iPS cells were dissociated to form a single cell suspension and reseeded into Petri dishes at a density of 1 ϫ 10 6 cells/ml.
Plasmids-The retroviral plasmids for iPS cell induction and the DsRed, EOS, and PGK lentiviral plasmids have been described previously (18). The CSII-EF-mTERT-IRES2-Venus plasmid was a gift from Dr. Hiroyuki Miyoshi (RIKEN Bioresource Center, Tsukuba, Japan). This plasmid carries a point mutation in TERT and was created using the following primers designed using the QuikChange Primer Design Program (Agilent Technologies): forward, 5Ј-gtactttgttaaggcagctgtgaccggggcctatg-3Ј; reverse, 5Ј-cataggccccggtcacagctgccttaacaaagtac-3Ј. PCR was performed using KOD-Plus polymerase (Toyobo) and the following conditions: initial denaturation at 95°C for 30 s, followed by 18 cycles of 95°C for 30 s, 55°C for 1 min, and 68°C for 15 min. The PCR product was digested with DpnI and then transformed into Escherichia coli. The mutation was confirmed by sequencing with the following primer, 5Ј-gctacgggagctgtcacaag-3Ј.
Gene Expression Analysis-Primers and probes specific for the four endogenous and retroviral reprogramming factors and the pluripotent cell markers (Nanog, ECAT1, and ERas) were designed as described previously (18). Transcript levels were determined using the 7500 Fast Real Time PCR system (Applied Biosystems).
Telomere Length Assay-Telomere lengths were analyzed using the Telo TAGGG telomere length assay kit (Roche Applied Science), according to the manufacturer's instructions. Briefly, 1.5 g of genomic DNA was digested with HinfI and RsaI. The digested DNA was hybridized to a digoxigenin-labeled telomere probe and detected using an anti-digoxigenin antibody.
Karyotype Analysis-Cell karyotypes were generated and analyzed at the Central Institute for Experimental Animals in Japan. Chromosomes from 50 metaphases were counted for each genotype and at each passage. Representative characteristics for each chromosome were aligned to identify fusion events.
Microarray Analysis-TERT-KO TTFs were infected with lentivirus containing mock-Venus, TERT-WT-Venus, or TERT-D702A-Venus. Two days later, infected cells were identified by the level of Venus fluorescence. One day after infection of cells with the retroviral reprogramming factors, cells were collected by Venus fluorescence. The experimental design is summarized in Fig. 6a. Expression profiles were analyzed using a whole mouse genome 44K3D-Gene Mouse Oligo chip 24K (Agilent Technologies). Fluorescence intensities were detected using a Scan-Array Life Scanner (PerkinElmer Life Sciences), and photomultiplier tube levels were adjusted to achieve 0.1-0.5% pixel saturation. Each TIFF image was analyzed with GenePix Pro software version 6.0 (Molecular Devices). The data were filtered to remove low confidence measurements and normalized globally per array such that the median signal intensity was set at 50. The signatures were calculated using genes that were at least 2-fold up-or down-regulated. To determine the gene expression signature of TERT-KO iPS cells at passage 43, genes that were up-and down-regulated genes compared with the other iPSCs we examined were determined. To characterize the molecular backgrounds of the signature genes, an enrichment analysis of canonical pathways and gene ontology biological processes (c2-cp and c5-bp gene sets in MsigDB version 3.0 (16) was performed using the gene ontology term finder (17). p Ͻ 5% was used as a threshold.

Preparation and Characterization of TTFs from TERT-KO
Mice-To analyze the function of TERT in somatic cell reprogramming, adult TTFs were prepared from TERT-KO mice. To exclude the effects of shortened telomeres, TTFs from F1 mice were used, as described in a previous study that analyzed the telomere RNA component (TERC) mutant (11). The efficiency of reprogramming is related to proliferative capacity; therefore, the viability and growth properties of the TERT-KO TTFs were examined (4). The numbers of viable TTFs obtained from single TERT-KO and WT mouse tails were comparable (Fig. 1a), and there was no significant difference after thawing stress (Fig. 1b).
To clarify the role of TERT in the reprogramming of somatic cells, the four reprogramming factors, Oct3/4, Sox2, Klf4, and c-Myc, were introduced into WT and TERT-KO TTFs as retroviruses. The two types of cells showed similar cell cycle patterns both before (Fig. 1c) and after (Fig. 1d) the introduction of the reprogramming factors. These data indicate that there were no significant differences between the viabilities and growth properties of the TERT-KO and WT TTFs.
KO of TERT Decreases the Efficiency of Somatic Cell Reprogramming of TTFs-To estimate the status of cellular reprogramming after the introduction of the four reprogramming factors, the endogenous activities of Oct3/4 and Sox2 were monitored using a lentiviral EOS vector, and the silencing of retroviral genes was monitored using a retroviral DsRed vector. Reprogrammed cells were identified as those that were EOSpositive and DsRed-negative ( Fig. 1e) (18,20). The colonies began to merge ϳ2 weeks after induction, and the numbers of colonies were counted at days 11-23 ( Fig. 1, f-l). 23 days after transfection, the number of silenced colonies of WT TTFs was ϳ10-fold higher than the number of silenced colonies of TERT-KO TTFs (Fig. 1l), indicating that TERT contributes to somatic cell reprogramming.
Although the efficiency of induction was low, candidate colonies from EOS-positive/DsRed-negative reprogrammed cells were successfully obtained from TERT-KO TTFs. These colonies were maintained and expanded, and their gene expression profiles and differentiation capacities were determined. Real time PCR analyses revealed that the endogenous mRNA levels of the two of the reprogramming factors, Oct3/4 and Sox2, were up-regulated in both WT and TERT-KO reprogrammed cells compared with parental TTFs (Fig. 2, a-d). In addition, compared with WT TTFs, pseudo colony cells (cells that were infected with the four Yamanaka reprogramming factors but were not morphologically ES cell-like), and STO cells, the mRNA levels of three other pluripotent cell markers, namely Nanog, ECAT1, and ERas, were also up-regulated in TERT-KO reprogrammed cells, and their expression levels were comparable with those in ES cells (Fig. 2, e-g). Because silencing of retroviral genes is an important requirement for iPS cells, the expression levels of the retroviral reprogramming factors were also analyzed. Both WT and TERT-KO reprogrammed cells had very low expression levels of all four viral genes (Fig. 2, h-k). These data indicate that TERT-KO-reprogrammed cells have comparable gene expression profiles to WT iPS cells.
Next, the differentiation capacity of TERT-KO reprogrammed cells was examined by injecting the cells into nude mice. After 3-4 weeks, teratomas containing tissues from all three germ layers were formed (Fig. 2l). These results demonstrate that TERT-KO reprogrammed cells are capable of forming all three germ layers, and taken together, these data indicate that TERT-KO TTFs are able to generate iPS cells.

Continuous Passaging Causes a Transient Growth Defect but Has No Effect on the Gene Expression Patterns or Differentiation
Capacity of TERT-KO iPS Cells-To evaluate the effect of TERT deficiency on the maintenance of pluripotency, WT and TERT-KO iPS cells were cultured continuously, and the numbers of cells and their cell cycle stages were compared at several time points. At passages 14 -18, the proliferation of TERT-KO iPS cells was comparable with that of WT iPS cells (Fig. 3, a and  d). The proliferative capacity of TERT-KO iPS cells was lower than that of WT iPS cells after 40 passages (Fig. 3, b and e); however, the numbers of each cell type were comparable after 60 passages (Fig. 3, c and f). At each passage, the gene expression profiles of the endogenous pluripotency genes and the silencing status of the four retroviral reprogramming factors were maintained (Fig. 2, a-k). The differentiation capacities of the WT and TERT-KO iPS cells were also analyzed by injecting iPS cells at various passages into nude mice (Fig. 3, g-i). At passage 43, WT iPS cells formed teratomas, but TERT-KO iPS cells did not ( Fig. 3h and Table 1). Surprisingly, TERT-KO iPS cells at passage 60 did produce teratomas, which might be due to escape from the proliferative crisis (Figs. 3i and 2m) (21).
To investigate the transient defects in growth and teratoma formation further, and to identify changes at the molecular level, microarray analyses of reprogrammed WT and TERT-KO iPS cells at various passages were performed. A cluster analysis identified 792 and 157 genes that were up-regulated and downregulated in TERT-KO iPS cells at passage 43, respectively (Fig.  3j and supplemental Table S1). A gene ontology analysis also revealed that the terms "cell adhesion" and "morphogenesis" were down-regulated in TERT-KO iPS cells at passage 43 ( Table  2). The array data were then compared with the results of a recently published analysis of Mbd3-deficient cells (22), which reported close to 100% efficient iPS cell reprogramming upon Mbd3 depletion and identified deterministic gene expression patterns for this reprogramming. Genes that were commonly down-regulated in TERT-KO iPS cells at passage 43 (compared with WT cells at the same passage) and up-regulated in Mbd3-KO cells at day 8 (compared with WT cells at day 11) were identified (supplemental Table S1). These inversely correlated genes may be responsible for the deficiencies in the proliferation and differentiation potential of TERT-KO iPS cells at passage 43. To determine whether the failure of these cells to form teratomas was caused by a loss of differentiation capacity, iPS cells at passages higher than 40 were analyzed using an in vitro EB differentiation assay. TERT-KO iPS cells at passage 41 formed EBs and had high expression levels of genes that repre-sent the differentiation of the three germ layers (Fig. 3, k-m).
These data indicate that, around passage 40, TERT-KO iPS cells display defective teratoma formation due to a decrease in proliferative capacity rather than a loss of differentiation capacity.
Chromosomal Instability in TERT-KO iPS Cells-A passagedependent decrease in growth rate and recapture of proliferative capacity have previously been reported for TERC-KO ES cells (21); therefore, the telomere length and chromosome stability of TERT-KO iPS cells were examined. Compared with WT iPS cells, TERT-KO iPS cells displayed a passage-dependent reduction in telomere length and relative signal intensity (Fig. 4a). In addition, a karyotype analysis identified a passagedependent accumulation of chromosome fusions and the appearance of characteristically long chromosomes formed as a result of chromosomal fusions in TERT-KO iPS cells (Fig. 4, b-f). At passages 6 and 58, most WT iPS cells had a normal complement of chromosomes (n ϭ 40) (Fig. 4e); however, after 30 passages, it was difficult to find TERT-KO cells with a normal complement of chromosomes, and most cells had 38 chromosomes or less (Fig. 4f). This accumulation of chromosome fusions increased as the cells were grown continuously up to around passage 60.
TERT Has an Extratelomeric Function in Somatic Cell Reprogramming-A deficiency in TERC, the RNA gene that serves as a template for the telomere repeat sequence (TTAGGG) elongated by TERT, is another cause of defects in telomere elongation; however, the reprogramming of somatic   (11). Because the TERT-KO TTFs used in this study were generated from F1 mice, the telomere length was long enough to exclude an effect on reprogramming efficiency. Despite this fact, the reprogramming efficiency of TERT-KO TTFs was markedly lower than that of WT TTFs, which suggests an extratelomeric function of TERT in somatic cell reprogramming. To investigate this possibility, a rescue experiment was performed by introducing WT TERT or the TERT D702A mutant (23) into TERT-KO TTFs. The D702A mutant is rendered enzymatically inactive by a point mutation of aspartic acid 702 to alanine (Fig. 5a). Prior to transfection of the TERT-KO TTFs with the four reprogramming factors, the cells were pre-infected with lentiviral vectors carrying WT or mutant TERT. The expression of both WT and mutant TERT was detected 1 day after infection (Fig. 5b), and colony numbers were counted 17 days after the induction of reprogramming. As expected, the colony formation efficiency of TERT-KO TTFs was rescued by the introduction of WT TERT (Fig. 5c). Interestingly, the TERT D702A mutant also improved the colony formation efficiency 19 days after induction (Fig. 5d). Moreover, the colonies generated from the rescue of TERT-KO TTFs with WT TERT or TERT D702A showed similar expression pat-terns of endogenous pluripotency genes, silencing of the introduced reprogramming factors, and differentiation capacities for all three germ layers through teratoma formation (Fig. 5, e-m). Taken together, these results indicate that the colonies generated after rescue were iPS cells and that TERT plays an extratelomeric role in somatic cell reprogramming.

Identification of Putative TERT Target and Response Genes-
To elucidate the molecular mechanism by which TERT improves the efficiency of somatic cell reprogramming, a mouse whole genome microarray analysis was performed. The gene expression profiles in mock (vector only)-infected TERT-KO TTFs were compared with those in TERT-KO TTFs infected with WT or D702A mutant TERT. The expression profiles were examined both before and after the induction of reprogramming by infection of the cells with the Klf4, Sox2, Oct3/4, and c-Myc retroviruses. An overview of the experimental design is shown in Fig. 6a. The numbers of genes that were up-or down-regulated in noninduced and reprogrammed cells compared with mock-infected TERT-KO cells are shown in Fig.  6, b and c, respectively. Prior to induction, 75 and 1571 genes were commonly up-and down-regulated, respectively, in both WT TERT-and TERT D702A-infected cells (Fig. 6b and supplemental Table S2); these genes are putative targets of enzymati- cally inactive TERT. When we focused specifically on transcription factors and epigenetic modifiers, histone modifiers involved in the maintenance of heterochromatin were identified as down-regulated genes, including Ezh2, Suv39h2, and HDAC9 (Table 3). TERT might relax the chromatin structure to facilitate its access by reprogramming factors. Furthermore, when compared with the array data of Mbd3-KO cells at day 4 (22), a number of positively correlated genes were identified (supplemental Table S3), including two members of the Prdm family (Prdm6 and Prdm1). Prdm6 is a known transcriptional repressor that associates with EHMT2/G9a (24), whereas Prdm1 is a known transcriptional repressor (25). In a recent study, overexpression of Prdm1 impaired embryonic stem cell maintenance (26). After induction by the four reprogramming factors, 538 and 77 genes were commonly up-or down-regulated, respectively, in both WT TERT and TERT D702A-infected TERT-KO cells compared with mock-infected TERT-KO cells (Fig. 6c and supplemental Table S4); these genes are putative TERT-response genes for reprogramming and included a number of transcription factors associated with somatic cell reprogramming, such as Esrrb and Klf17 (Table 4). Esrrb and Klf17 were enriched upon comparison with the Mbd3-KO data (supplemental Table S5) (22). Esrrb is a key regulatory factor for reprogramming, and Klf17 is reported to suppress epithelial mesenchymal transition in breast cancer (27,28). Because mesenchymal epithelial transition is a major event during the early phase of reprogramming, suppression of the process would be expected to inhibit its progression (29,30). Overall, the microarray data suggest that, upon induction, TERT up-regulates key genes involved in cellular reprogramming.

DISCUSSION
The reprogramming of somatic cells allows unipotent differentiated cells to become pluripotent. TERT, the enzyme involved in telomere synthesis, is expressed at low levels in somatic cells; however, during the reprogramming process, TERT is reactivated and telomeres are re-elongated (11). In addition to telomere elongation, TERT has other functions, including RNA-dependent RNA polymerase activity and modulation of ␤-catenin signaling (31,32). This study demonstrates that TERT also plays a role in somatic cell reprogramming that is independent of telomere elongation.
Reprogramming and Enzymatic TERT Activity-The reprogramming efficiency of TERT-KO TTFs was approximately one-tenth that of WT TTFs; however, despite this reduction in efficiency, the reprogrammed cells from TERT-KO TTFs showed characteristics typical of iPS cells, including similar gene expression profiles, silencing of retroviral genes, and the ability to differentiate into the three germ layers. These data indicate that TERT plays an important but nonessential role in somatic cell reprogramming. A previous study demonstrated that TERC-KO mice begin to show stem cell defects related to telomeric shortening from the fifth generation onward (33). Because the TERT-KO TTFs used in this study were generated from F1 KO mice, the telomeres were not shortened, and defects were not observed ( Fig. 4) (23). Shortened telomeres are related to a significant decrease in the reprogramming efficiency of TERC-KO somatic cells; however, if telomeres are sufficiently long at the beginning of reprogramming, telomere shortening may not affect the efficiency of reprogramming (11). Taken together, the current evidence suggests that the low reprogramming efficiency of TERT-KO TTFs is not caused by shortened telomeres. Because another study reported that TERT functions as an RNA-dependent RNA polymerase (31), it would be interesting to elucidate the role played by RNA-dependent RNA polymerase in somatic cell reprogramming.
Interestingly, the reprogramming efficiency of TERT-KO TTFs was improved by infection of the cells with the TERT D702A mutant, indicating that enzymatically inactive TERT also has reprogramming activity. Enzymatically inactive TERT D702A induces epithelial cell proliferation through the ␤-catenin pathway (34), and the inhibition of GSK-3␤, which leads to ␤-catenin activation, is one of the major signals involved in maintaining pluripotency (35). It is possible that the TERT D702A mutant activates ␤-catenin to rescue the reprogramming efficiency. Recent reports suggest that Esrrb is a downstream effector of GSK-3␤ inhibition that is required to maintain pluripotency (36,37). Intriguingly, the microarray analysis described here indicated that Esrrb is a putative TERT target gene involved in reprogramming. A recent study demonstrated that TERT binds several proteins (31), and these TERTbinding proteins are candidate effectors for potentiating the efficiency of reprogramming. Because some of these candidate proteins are reported to have functions related to pluripotency (38), it would be interesting to investigate their potential roles in somatic cell reprogramming. Furthermore, Esrrb plays an important role at the late phase of reprogramming (39), and retrovirus silencing also occurs as a late event during this process (40,41). Here, TERT-KO TTFs produced more nonsilenced colonies than silenced colonies (Fig. 1, h and k). Therefore, it will be important to analyze the relationship between TERT, Esrrb, and silencing.

Down-regulated genes Transcription factors Dlx5
Distal-less homeobox 5 Tbx10 T-box 10 Adipoq Adiponectin, C1Q, and collagen domain containing Myf6 Myogenic factor 6 Zfp57 Zinc finger protein 57 Tlx1 T-cell leukemia, homeobox 1 Eomes Eomesodermin homolog (Xenopus laevis) Smad9 MAD homolog 9 (Drosophila) Hoxa2 Homeobox A2 Esx1 Extraembryonic, spermatogenesis, homeobox 1 Klf1 Kruppel-like factor 1 (erythroid) Epigenetic modifiers No gene Continuous Passaging of Pluripotent TERT-KO iPS Cells Leads to Abnormal Chromosomes-The number of abnormal chromosomes in TERT-KO iPS cells increased in proportion to the passage number. After 30 passages, no cells with a normal complement of chromosomes were observed (Fig. 4f). As predicted from the function of telomeres in protecting chromosomal ends, the majority of abnormalities in TERT-KO iPS cells were caused by chromosomal end fusions (Fig. 4d). However, the expression profiles of pluripotency-associated genes were maintained regardless of the passage number. In addition, TERT-KO iPS cells retained the capacity to differentiate into all three germ layers, even at passage numbers greater than 60. However, at around passage 40, TERT-KO iPS cells showed a transient decrease in the proliferation rate and a failure to form teratomas, even though the cells were able to differentiate into the three germ layers through EB formation in vitro (Fig. 3,  k-m). Although the mechanism underlying the transient defect is unknown, this interesting phenomenon was also previously reported in TERC-KO ES cells and TERT-knockdown ES cells (21,42).
In pluripotent cells such as ES cells and iPS cells, the cell cycle checkpoint mediated by p53 is nonfunctional (43). Therefore, pluripotent cells are exposed to continuous genomic mutations, which leads to a potential risk of tumorigenicity. If these cells are to be used for clinical applications, it is important that such a risk is avoided. However, in this study, the presence of abnormal chromosomes did not affect the gene expression pattern in the undifferentiated state nor the differentiation capacity of iPS cells. Furthermore, in addition to chromosomal fusions, amplifications were also detected, as observed previously in ATM-KO iPS cells (18). Therefore, analysis of chromosomal abnormalities is essential for the future clinical applications of iPS cells, and it will also be important to establish a culture method capable of maintaining the genomic stability of these cells.