By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
The Rosalind & Morris Goodman Cancer Research Center, Montréal, Québec, Canada
Department of Anatomy & Cell Biology, Montréal, Québec, Canada
Department of Medicine, Montréal, Québec, Canada
Department of Biochemistry, McGill University, Montréal, Québec, Canada
McGill University Health Center, Montréal, Québec, Canada
Correspondence: Xiang-Jiao Yang, Ph.D., Department of Medicine, McGill University Health Center, Montréal, Québec, Canada. Telephone: +1–514-398-5883; Fax: +1–514-398-6769; e-mail: firstname.lastname@example.org
Sumoylation adds a small ubiquitin-like modifier (SUMO) polypeptide to the ε-amino group of a lysine residue. Reminiscent of ubiquitination, sumoylation is catalyzed by an enzymatic cascade composed of E1, E2, and E3. For sumoylation, this cascade uses Ubc9 (ubiquitin conjugating enzyme 9, now officially named ubiquitin conjugating enzyme E2I [UBE2I]) as the sole E2 enzyme. Here, we report that expression of endogenous Ubc9 increases during reprogramming of mouse embryonic fibroblasts (MEFs) into induced pluripotent stem (iPS) cells. In addition, this E2 enzyme is required for reprogramming as its suppression dramatically inhibits iPS cell induction. While Ubc9 knockdown does not affect survival of MEFs and immortalized fibroblasts, Ubc9 is essential for embryonic stem cell (ESC) survival. In addition, we have found that Ubc9 knockdown stimulates apoptosis in ESCs but not in MEFs. Furthermore, the knockdown decreases the expression of the well-known pluripotency marker Nanog and the classical reprogramming factors Klf4, Oct4, and Sox2 in ESCs. Together, these observations indicate that while dispensable for fibroblast survival, the sole SUMO E2 enzyme Ubc9 plays a critical role in reprogramming fibroblasts to iPS cells and maintaining ESC pluripotency. Stem Cells2014;32:1012–1020
Different from many other post-translational modifications such as phosphorylation and acetylation, sumoylation adds a small ubiquitin-like modifier (SUMO) polypeptide to the ε-amino group of lysine residues [1–3]. As the name implies, SUMO polypeptides are similar to ubiquitin, sharing approximately 18% amino acid sequence identity and homologous 3D structural folds [2, 3]. In humans, there are four SUMO proteins, SUMO1, 2, 3, and 4. SUMO2 and SUMO3 are highly homologous to each other (95% identical) and also display significant similarity to SUMO1 (47% identical) [2, 3]. While SUMO1 confers mainly monosumoylation, SUMO2 and SUMO3 allow both monosumoylation and polysumoylation . The functional significance of SUMO4 remains to be determined. Reminiscent of ubiquitination, sumoylation is catalyzed by a three-enzyme cascade. For sumoylation, this cascade is composed of a single heterodimeric E1 activating enzyme (SAE1/SAE2), a single E2 conjugating enzyme and multiple E3 ligases [1, 3]. From yeast to humans, Ubc9 (ubiquitin conjugating enzyme 9, now officially named ubiquitin conjugating enzyme E2I [UBE2I]) is the single SUMO E2 enzyme and thus provides a convenient intervening point for interrogating globally how sumoylation regulates various cellular and biological processes in different organisms.
Since the initial description of induced pluripotent stem (iPS) cell formation by ectopic expression of only four transcription factors, Oct4 (also known as Pou5f1), Sox2 (SRY [sex determining region Y]-box 2), KLF4 (Krüppel-like factor 4), and c-Myc , there have been intensive efforts to apply this technology to disease modeling in vitro and eventual autologous cell therapy [6–8]. A better understanding of the underlying molecular and cellular mechanisms is important for improvement and optimization of this technology. Post-translational modifications such as sumoylation are crucial for various transcription factors [9–11], but it remains largely unknown how such modifications may affect iPS cell induction. Among the four classical reprogramming factors initially identified , Oct4, Sox2, and KLF4 have been shown to be sumoylated [12–15]. In addition, two KLF4 homologs, KLF2 and KLF5 play a role in reprogramming . While KLF5 has been shown to be sumoylated [17, 18], we found that KLF2 possesses two sumoylation sites and is sumoylated . Furthermore, estrogen-related receptors are subject to phosphorylation-dependent sumoylation and play a role in reprogramming [20, 21]. Together, these observations raise the intriguing possibility that sumoylation regulates iPS cell induction and stem cell pluripotency. This has led us to investigate how the sole SUMO E2 enzyme Ubc9 may play the role in fibroblast reprogramming to iPS cells and maintaining the pluripotency of embryonic stem cells (ESCs). The results indicate that while dispensable for survival of fibroblasts, Ubc9 plays a critical role in inducing and maintaining stem cell pluripotency.
Materials and Methods
The following lentiviral shuttle plasmids were purchased from Addgene: pLOVE (15948), pLOVE-Klf4 (15950), pLOVE-N-Myc (15951), pSin-EF2-Sox2-Pur (16577), and pSin-EF2-Oct4-Pur (16579). The plasmids for Flag-tagged wild-type and mutant proteins were constructed in the lab. Lentiviral shuttle vectors were also prepared on pENTR11 for homologous recombination with pLOVE via the Gateway system (Invitrogen, Carlsbad, CA, http://www.invitrogen.com).
293FT cells (Invitrogen) were maintained in a medium containing 400 µg/ml neomycin (Geneticin, Invitrogen, 11811-098) per manufacturer's instructions and incubated in the antibiotic-free medium for at least 8 hours prior to Lipofectamine 2000-mediated transfection. 10 µg of expression plasmid was mixed with 6.5 µg psPAX2, 3.5 µg pMD2.G, 50 µl Lipofectamine 2000, and 1 ml Opti-MEM to transfect 8 × 106 293FT cells in a 10-cm dish according to the manufacturer's instructions (Invitrogen). The medium was collected as the viral supernatant everyday for 3 days and was subjected to centrifugation at 76,000g for 1.5 hours. The viral pellet was then dissolved in fresh complete Dulbecco's modified Eagle's medium (DMEM) and rotated overnight at 4°C. The resulting virus was used to infect cells directly or was flash-frozen in aliquots on dry ice for long-term storage at −80°C.
shRNA (small hairpin RNA) Knockdown
Lentiviruses expressing shRNAs against mouse Ubc9 were generated as described above (Lentivirus preparation). The lenviral vector coexpressing a Ubc9 shRNA and green fluorescent protein (GFP) was kindly provided by X. Liu , whereas the three puromycin-resistant lenviral vectors were purchased from Sigma-Aldrich (St. Louis, http://www.sigmaaldrich.com; TRCN00000040839, TRCN00000040840 and TRCN00000040842), designated as shUbc9-1, −3, and −4, respectively. For lentiviral transduction, cells were infected with concentrated lentiviruses for 48 hours in the presence of 8 µg/ml polybrene.
Mouse Embryonic Fibroblast Preparation and iPS Cell Generation
Mouse embryonic fibroblasts (MEFs) were derived from 13.5-day postcoitus mouse embryos as described previously [19, 23]. Drug selection-free iPS cell generation has been described . Briefly, on the day before infection, MEFs (passage 2 or 3) were plated at 0.08–0.1 × 106 cells per well in a 12-well plate or 0.3 × 106 cells per well in a six-well plate and incubated overnight in DMEM containing 10% fetal bovine serum (FBS) and penicillin/streptomycin (50 µg/ml each) inside a 37°C CO2 incubator for 48 hours. On the next day of infection, the medium was replaced with a concentrated virus mixture (dissolved in DMEM/10% heat-inactivated FBS) plus polybrene (8 µg/ml) and incubated in a 37°C CO2 incubator. The cells were then washed with PBS (phosphate buffered saline) and cultured with mESC medium (DMEM high glucose, 1% nonessential amino acids [100× stock, Invitrogen], 1% sodium pyruvate [100× stock, Invitrogen], 0.1 mM β-mercaptoethanol, 15% FBS, penicillin and streptomycin [50 µg/ml each], and 1,000 U leukemia inhibitory factor/ml [Millipore, Billerica, MA, http://www.millipore.com]). The medium was changed every other day. iPS cell colonies normally appeared 5–6 days after infection. For alkaline phosphatase (ALP) staining, a detection kit (Millipore, SCR004) was used according to the manufacturer's instructions.
ESC Survival and Self-Renewal
ESCs initially isolated from FVB and C57BL/6 mice were infected with the lentivirus expressing shRNA against Ubc9 or with the corresponding lentivirus lacking the shRNA. Survival of ESC colonies were monitored by microscopic analysis of colonies or ALP staining (Millipore, SCR004). Western blotting analysis was used to detect the expression of ESC-specific factors.
Nuclear and Apoptotic Body Staining with DAPI
Cells were fixed with 2% paraformaldehyde and permeabilized with 0.2% Triton X-100 solution/PBS for 10 minutes. Permeabilized cells were incubated with DAPI staining (final concentration of 0.5 µg/ml) for 10 minutes and washed two times with Nano-pure H2O. Fluorescence microscopy was used to detect nuclei and apoptotic bodies.
Western blot analysis was performed as described [25, 26]. Briefly, cells were washed two times with PBS and lysed in buffer K (20 mM sodium phosphate, pH 7.0, 150 mM KCl, 30 mM sodium pyrophosphate, 0.1% Nonidet P40, 5 mM EDTA, 10 mM NaF, 0.1 mM Na3VO4, and protease inhibitors). Soluble protein extracts were separated by SDS-PAGE for immunoblotting with the following antibodies: anti-Nanog (Bethyl Laboratories, Montgomery, TX 77356 USA, http://www.bethyl.com/ BL1662), anti-KLF4 (Santa Cruz Biotechnology, Santa Cruz, CA, http://www.scbt.com; H-180), anti-Caspase-3 (Cell Signaling Technology, Beverly, MA, http://www.cellsignal.com; 9662), anti-Ubc9 (Santa Cruz Biotech., C-12), anti-Sox2 (Santa Cruz Biotech., V-17), anti-α-tubulin (Sigma-Aldrich, T5168), and mouse anti-neuronal class III β-tubulin (Tuj1) (Covance, http://www.covance.com/products/nonclinical/antibody/, Montreal H3B 1R1 Canada MMS-435P).
GFP and indirect immunofluorescence microscopic analyses were performed as previously described [26, 27].
Data are presented as means ± S.D. Unpaired two-tailed Student's t tests were performed to calculate p value. p < .05 was considered statistically significant.
Requirement of Ubc9 for iPS Cell Induction
To investigate how sumoylation may affect iPS cell induction, we first examined the expression of Ubc9 during the process. For this, MEFs were infected with lentiviruses expressing GFP or the four classical reprogramming factors, Oct4, Sox2, KLF4, and c-Myc (OSKM). Afterward, cells were collected on each day for 5 days to prepare soluble protein extracts for detection of endogenous Ubc9 by Western blotting. As shown in Figure 1, while transduction with the control GFP-expressing lentivirus had minimal effects on the Ubc9 level, infection of lentiviruses expressing the OSKM reprogramming factors gradually increased the level of endogenous Ubc9 as the reprogramming proceeded. The increased level of Ubc9 upon OSKM expression suggested a potentially important role of Ubc9 in fibroblast-to-iPS cell reprogramming.
We next analyzed how shRNA knockdown of Ubc9 might affect the reprogramming process. As reported , Ubc9 knockdown in MEFs by the shRNA-expressing lentivirus was efficient (Fig. 2A). We then infected MEFs with the lentiviruses expressing the OSKM factors along with the anti-Ubc9 shRNA lentivirus or a control lentivirus. Two days after infection, MEFs were transferred into a typical ESC medium. On day 7, cells were fixed and stained to detect ALP-positive colonies. Detection of ALP+ colonies is a typical convenient assay for assessing efficiency at the early stages of reprogramming . As shown in Figure 2B, upon Ubc9 knockdown by shRNA, far fewer ALP+ colonies were obtained, indicating that the reprogramming was dramatically inhibited. Quantification of the reprogramming efficiency in independent experiments confirmed that this is the case (Fig. 2C). A few colonies did appear in the plates with Ubc9 knockdown, but those colonies exhibited flat morphology and were sensitive to further passage and expansion (Fig. 2D and data not shown). This is in stark contrast to typical mouse iPS or ESC clones, which are dome-shaped and can be expanded by mild trypsinization and repeated passages. In addition, two other shRNAs against Ubc9 inhibited iPS cell formation (Fig. 2E), indicating that the knockdown effect is specific. These results support that Ubc9 knockdown inhibits reprogramming and Ubc9 is required for the process.
We also investigated how ectopic expression of Ubc9 may affect reprogramming. Immunoblotting indicated that exogenous Ubc9 was expressed as expected (Fig. 3A). Ubc9 expression yielded slightly more ALP+ clones but that did not appear to be statistically significant (Fig. 3B-C). Thus, ectopic expression of Ubc9 had minimal impacts on the reprogramming process. This and the above results also suggest that while Ubc9 is essential for the process, its constitutive expression by lentiviral transduction may not be able to stimulate the process significantly.
Essential Role of Ubc9 in Maintaining ESC Pluripotency
Because iPSCs possess pluripotency as ESCs [6–8], the above findings raise the intriguing possibility that Ubc9 is also required for maintaining the pluripotency of ESCs. Consistent with this contention, two elegant studies of constitutive and conditional Ubc9−/− mice have recently identified an essential role of Ubc9 in survival and proliferation of the inner cell mass (ICM) within blastocysts and adult intestinal stem cells [28, 29]. All ESCs are established cell lines that were directly derived from the ICM [30, 31], but it has not been formally tested whether Ubc9 is required for survival of ESCs. As such, we decided to investigate whether this is indeed the case. For this, mouse ESCs were infected with the lentivirus expressing the shRNA against mouse Ubc9. As shown in Figure 4A, shRNA knockdown of Ubc9 eliminated ESC colonies on day 3 postinfection, confirming the requirement of Ubc9 for ESC cell survival. To substantiate this and also avoid potentially off-target effects of the shRNA that was used, we used three additional shRNA-expressing vectors (shUbc9-1, −3, and −4) to knockdown Ubc9. As shown in Figure 4B, all three yielded efficient knockdown of Ubc9 in MEFs. Interestingly, Ubc9 knockdown had no obvious effects on the survival of regular MEFs (Fig. 4C) or three established mouse cell lines, 3T3-L1, NIH 3T3, and C2C12 (Supporting Information Fig. S1). By contrast, phase-contrast microscopic analysis revealed that Ubc9 knockdown by all three shRNA vectors progressively inhibited formation of ESC colonies although the impact by shUbc9-3 was less dramatic (Fig. 5A). Moreover, this inhibitory effect of the knockdown was confirmed by staining ALP+ colonies (Fig. 4D-E). Similar results were reproduced with another mouse ESC line (Fig. 5B-D). Together, these results establish that Ubc9 is essential for the survival of ESCs but not MEFs or three immortalized somatic cell lines.
Ubc9 Knockdown Promotes Apoptosis in ESCs but Not MEFs
Two potential mechanisms responsible for affecting ESC survival resulting from the absence of Ubc9 are ESC differentiation and apoptosis. Based on our observations, the first mechanism was unlikely, because no additional cell types appeared in the culture and the only remaining cells were the irradiated MEFs that were used to form the feeder layers (Figs. 4D, 5). A previous study showed that genetic deletion of the mouse Ubc9 gene induces cell death by nuclear envelope disruption of the ICM, followed by apoptosis . We also observed from day 3 after Ubc9 knockdown, ESCs progressively underwent nuclear breakdown and the apoptotic bodies appeared in the culture dish (Fig. 6A-B). This effect was ESC specific as MEFs appeared normal and disruption of nuclear envelope was not detectable in these cells (Fig. 6A, Supporting Information Fig. S2). To test whether caspase-dependent mechanism triggered nuclear breakdown in Ubc9-defcient ESCs, we determined caspase-3 activation in MEFs and ESCs following shRNA-mediated knockdown of Ubc9 (Fig. 6C). Cleaved caspase-3 was detected in ESCs but not MEFs on day 3 after Ubc9 knockdown, which coincided with appearance of apoptotic bodies (Fig. 6C). Therefore, shRNA knockdown resulted in cell death (Figs. 4, 5), activation of caspase 3 (Fig. 6C), and formation of apoptotic bodies (Fig. 6A-B) in ESCs.
Then we wondered about the molecular mechanism whereby Ubc9 knockdown causes apoptosis in ESCs. One possibility is that sumoylation regulates the expression of pluripotency markers. To investigate this possibility, we performed Western blotting analysis to examine the levels of Nanog, Oct4, Sox2, and Klf4. As shown in Figure 7A, Ubc9 knockdown decreased the endogenous levels of these transcription factors, suggesting that the knockdown downregulates their expression. We also investigated whether Ubc9 knockdown promoted differentiation, which in turn inhibits the expression of pluripotency markers. Related to this, the knockdown did not induce expression of lineage-specific markers such as Tuj1 (Fig. 7B). Therefore, Ubc9 appeared to promote ESC survival by decreasing expression of pluripotency protein markers such as Nanog, Oct4, Sox2, and Klf4. Whether this is due to protein degradation remains to be determined.
We have demonstrated herein that Ubc9 is required for iPS cell formation (Figs. 1, 2). Related to this essential role of Ubc9 in reprogramming, it is noteworthy that sumoylation of KLF4, Oct4, and Sox2 exerted inhibitory effects . Moreover, sumoylation of the nuclear orphan receptor Nr5a2 also inhibits reprogramming (Supporting Information Fig. S3) . These findings suggest that sumoylation of an unidentified factor(s) positively regulates reprogramming. Related to this, a recent study has identified many reprogramming factors , raising the interesting question whether sumoylation plays a role in these new factors during reprogramming.
We have also found that Ubc9 is essential for ESC survival in vitro (Figs. 4, 5). By contrast, Ubc9 is not required for survival of MEFs (Fig. 4C) and immortalized fibroblasts such as NIH3T3 cells (Supporting Information Fig. S1C, S1D). Moreover, Ubc9 knockdown did not induce apoptosis in these cells (Supporting Information Fig. S2). Consistent with what we observed (Supporting Information Fig. S1A, S1B, S1E, S1F), two other groups have reported that Ubc9 is not essential for survival of C2C12 myoblasts  or 3T3-L1 preadipocytes . Thus, it appears that Ubc9 is specifically required for maintenance of pluripotent ESCs. In agreement with this, Ubc9−/− mouse blastocysts are viable, but their inner cell mass fails to expand and shows signs of apoptosis , indicating that Ubc9 is required for the pluripotent ESC population during early development of mouse embryos. Related to this, Ubc9 promotes cancer cell survival , perhaps due to an effect on cancer stem cells. Moreover, conditional deletion of mouse Ubc9 revealed its requirement for the intestinal stem cell compartment . It will be interesting to examine whether this is the case for other stem cell compartments in mice. These findings suggest that at least in mice, sumoylation is not only important for iPS cell formation but also plays a crucial role in embryonic and adult stem cell maintenance. This is further supported by enrichment of sumoylation components (e.g., SUMO1 and PIAS2) in a global gene expression survey of ESCs and iPS cells .
Our findings indicate that as the sole SUMO conjugating enzyme, Ubc9 is required for inducing and maintaining pluripotency in murine cells. This is good agreement with published studies on Ubc9−/− mouse blastocysts . It has been shown that global sumoylation occurs during torpor in ground squirrels and that overexpression of Ubc9 in cell lines and cortical neurons protects against oxygen and glucose deprivation . In addition, Ubc9 levels changes during adipocyte differentiation . Therefore, these studies reiterate the dynamic and active roles of Ubc9 and sumoylation in specific but diverse cellular programs.
Together with the published study on Ubc9−/− mouse blastocysts , our findings indicate that as the sole SUMO conjugating enzyme, Ubc9 is required for inducing and maintaining pluripotency in murine cells. Whether this is also the case in human cells is an important question awaiting further investigation.
We thank Dr. Yojiro Yamanaka for valuable advice on isolation of ESCs from C57BL/6 blastocysts. This project was supported by studentships from McGill University (to S.T. and G.G.) and Natural Sciences and Engineering Research Council of Canada (NSERC, to P.S.) and by operating grants from NSERC, Canadian Institutes of Health Research and Ministère du Développement Économique, Innovation et Exportation du Québec (to X.J.Y.).
S.T.: initiated the project; S.T. and M.G.: performed the most of the experiments; P.S.: made Ubc9 constructs; G.G.: provided technical help; X-J.Y.: wrote the manuscript. S.T. and M.G. contributed equally to this article.
Disclosure of Potential Conflicts of Interest
The authors indicate no potential conflicts of interest.