In January 2014, it was reported that strong external stimuli, such as a transient low-pH stressor, was capable of inducing the reprogramming of mammalian somatic cells, resulting in the generation of pluripotent cells. This cellular reprograming event was designated ‘stimulus-triggered acquisition of pluripotency’ (STAP) by the authors of these reports. However, after multiple instances of scientific misconduct in the handling and presentation of the data were brought to light, both reports were retracted. To investigate the actual scientific significance of the purported STAP phenomenon, we sought to repeat the original experiments based on the methods presented in the retracted manuscripts and other relevant information. As a result, we have concluded that the STAP phenomenon as described in the original studies is not reproducible.
Cellular reprograming is a biological event in which a differentiated metazoan cell is induced to revert to a state functionally resembling that of cells at earlier developmental stages1,2. Full reprograming of somatic cells results in the acquisition of the ability to give rise to an entire organism, or totipotency; this can be achieved by somatic cell nuclear transfer3. Pluripotency in contrast is the ability of a cell to differentiate into all somatic cell lineages. It has been shown that the artificial expression of pluripotency-associated transcription factors results in reprogramming of somatic cells to a state of pluripotency, such cells are referred to as as induced pluripotent stem (iPS) cells4.
Mouse pluripotent stem cells share common features. Authentic pluripotent stem cells are embryonic stem (ES) cells derived from pre-implantation embryos5,6. Under optimized culture conditions, these maintain self-renewal by giving rise to pluripotent daughter cells via cell division. Leukemia inhibitory factor (LIF) is a well-known factor sufficient to maintain the pluripotency of mouse pluripotent stem cells in vitro7. Such cells express a unique set of genes associated with pluripotency, such as a transcription factor Oct3/48, and contribute to embryo development when transferred into pre-implantation embryos, resulting in the formation of germline chimeras9. These properties are shared by iPS cells derived from somatic cells4. Therefore, acquisition of pluripotency by somatic cells via reprograming is typically assessed based on such criteria.
In 2014, Obokata et al. reported that sublethal external stimuli, such as exposure to a transient low-pH stressor, reprogrammed mammalian somatic cells, resulting in the generation of pluripotent cells1,2. In these reports, this cellular reprograming event was designated ‘stimulus-triggered acquisition of pluripotency’ (STAP). The reports also described how the primary pluripotent cells, STAP cells, were able to give rise to two types of pluripotent stem cells in a culture-condition-dependent manner. However, both reports were subsequently retracted due to multiple intsances of scientific misconduct10,11. To investigate the scientific significance of the STAP phenomenon, we repeated the reported experiments based on the methods presented in the retracted manuscripts and other relevant information subsequently obtained. We examined the expression of pluripotency-associated genes in cell aggregates obtained in cultures of somatic cells treated with transient low-pH, and the ability of such cell aggregates to contribute to chimeric embryos after injection into pre-implantation embryos. The results of this reevaluation indicate that the previously reported STAP phenomenon is not reproducible.
Adenosine triphosphate (ATP)-mediated transient low-PH treatment enhances formation of characteristic cell aggregates
In the original report, transient low-pH stress induced by addition of hydrochloric acid (HCl) caused massive cell death of dissociated somatic cells around 1–2 days after treatment (Fig. 1d of Obokata et al. 20141); the surviving cells formed aggregates (Fig. 1b of Obokata et al.1). In the present study, we examined the effect of HCl treatment on dissociated cells derived from spleen, liver and heart of 4–9-day-old mice. The amount of the diluted HCl solution used to achieve optimized low-pH condition was set at 10 µl based on the titration assay (Fig. 1a). This amount resulted in massive cell death routinely although it gave higher pH than 5.7 that was indicated in the previous manuscript (Figure S1a of Obokata et al.1), and 12 µl often resulted in complete cell death. However, although massive cell death was observed at two days after treatment, aggregate formation was rarely observed in any cell type (Fig. 1b). Occasional formation of aggregates was also observed in the culture of non-treated cells, suggesting that low-pH treatment does not enhance the formation of cell aggregates.
Next we examined the effect of adenosine triphosphate (ATP) as a transient low-pH stressor based on personal communication with the authors of the original study. The amount of the diluted ATP solution to achieve optimized low-pH (~5.7) was adjusted (Fig. 1a) and experiments were repeated several times. Massive cell death was again observed at two days after treatment (Fig. 2a); however, we found that liver cells reproducibly gave rise to cell aggregates morphologically similar to those shown in the previous report, whereas spleen and heart cells only occasionally formed similar cell aggregates (Fig. 1b). The efficiency of aggregate formation was clearly higher for ATP-treated cells than for HCl-treated or non-treated cells, especially in the case of liver cells.
In the case of liver cells, when 5 × 105 cells were seeded in a well of 12 well plate, 20–30 aggregates were observed after seven days on average (Fig. 2b). Addition of fibroblast growth factor (Fgf)-2, based on personal communication with the authors of the 2014 reports, slightly enhanced cell agregate formation. Since the culture medium contains leukemia inhibitory factor (LIF), which shows differential action on ES cells derived from different genetic backgrounds12, we suspected that the genetic background might affect aggregate formation. However, as shown in Fig. 2c, although it was known that the 129 background confers a dominant effect in obligating the LIF signal input to maintain pluripotency12, there was no difference between C57BL6 and C57BL6 × 129 F1 (either C57BL6/129 or 129/C57BL6) in the observed frequency of aggregate formation in the present study.
Induced cell aggregates show poor induction of pluripotency-associated markers
To test the induction of pluripotency markers in cell aggregates obtained from ATP-treated liver cells, we assessed the expression of pluripotency-associated genes. Oct3/4 is a well-defined marker of pluripotent stem cells. Using a primer pair to detect Oct3/4 transcript from the Pou5f1 allele, but not pseudo-genes13, we did not find a detectable level (above 0.1% of the expression level in mouse ES cells, relative to the expression levels of Gapdh) of the transcript by quantitative polymerase chain reaction (Q-PCR) using a total RNA sample prepared from all cells in the culture (Fig. 3a), indicating that extremely few or no cells expressing Oct3/4 were present. Interestingly, expression of Gfp from the Oct3/4-GFP transgene (GOF18)14 was detected in liver cells cultured for seven days irrespective of ATP treatment, suggesting leaky expression of this transgene.
We next performed qPCR on individual cell aggregates isolated from culture. Aggregates were selected and RNA samples were prepared separately. These RNAs were reverse-transcribed and qPCR was performed. We found that some aggregates expressed a comparable amount—more than 10% of the expression level in ES cells—of pluripotency-associated genes, including Oct3/4 (Fig. 3b). Since the cell aggregates consist of ~10 cells, such expression level indicated possible existence of the cell(s) expressing pluripotency-associated genes at the equivalent level to that in ES cells. Klf4 expression was detected in all samples, which may reflect its expression in liver cells, and thus serves as a positive control in this assay. Of cell aggregates derived from liver cells treated with ATP, 19% expressed the amount of Oct3/4 comparable to ES cells (Fig. 3c). These data suggest that some proportion of cells in the aggregates express pluripotency-associated genes at comparable levels to those of ES cells.
To examine the proportion of the cells expressing Oct3/4 in the aggregates, we next applied immuno-staining using a specific antibody against Oct3/4 we raised and assessed previously15. Cell aggregates derived from low-PH treated liver cells were fixed, stained by anti-Oct3/4 antibody, and observed using confocal microscopy. We stained morula-stage mouse embryos as positive controls. By comparison with these positive controls, we found that some of the cell aggregates contained cells expressing Oct3/4 at comparable levels (Fig. 4a). In the case of cell aggregates derived from liver cells treated by ATP, 20% of cell aggregates contained Oct3/4-positive cells (Fig. 4b), which is consistent to the proportion of cell aggregates expressing the amounts of Oct3/4 comparable to ES cells detected by QPCR (Fig. 3c). In contrast, cell aggregates derived from liver cells treated by HCl included Oct3/4-positive cells at a frequency comparable to that of non-treated cells. The presence of Oct3/4-positive cells in the cell aggregates found only 1 in 14 cases in cultures of non-treated liver cells suggests that such cells are derived from Oct3/4-positive cells present in the liver cell population, that in vitro culture may itself be a source of the stress to the cells, or that the immuno-staining technique may produce some non-specific signal. In cell aggregates derived from ATP-treated liver cells, the Oct3/4-positive cells were typically positioned in the center of the cell aggregates and exhibit large nuclei, and were surrounded by Oct3/4-negative cells with small nuclei at the peripheries of the cell aggregates (Fig. 4c).
Oct3/4-GFP transgene expression not detected in low-pH treated cells
In the original report, the authors used transgenic reporter gene expression as a marker of pluripotency1. This reporter consisted of the transcriptional regulatory element of Oct3/4 and the fluorescent marker GFP, designated GOF, which is silent in somatic cells and activated in pluripotent cells14. When we used the same transgenic mouse line as a source of dissociated cells, we found that they began to acquire strong auto-fluorescence in culture after ATP treatment. On observation with fluorescent microscopy, most aggregates showed both green and red fluorescence, a sign of auto-fluorescence (Fig. 5a) although this may also include the fluorescent signal from the GOF transgene, since we detected Gfp mRNA by qPCR in these cells after culture in vitro for seven days (Fig. 3a).
Specific detection of GFP fluorescence by fluorescence-activated cell sorting (FACS) was also performed. In spleen cells collected using Lympholyte, CD45-positive/E-cadherin-negative blood cells were enriched. The reprogramming of such cells to a state of pluripotency can be monitored by their conversion to CD45-negative/E-cadherin-positive cells and acquisition of GFP expression from the GOF transgene. However, although we again observed increased auto-fluorescence and some reduction of CD45 expression, neither a specific signal of GFP fluorescence nor an increase of E-cadherin expression was observed in the low-pH treated cells (Fig. 5b). Given these findings, we suggest that that there was no evident sign of reprogramming in low-pH treated spleen cells based on the expression of the GOF transgene.
Induced cell aggregates do not contribute to chimeric embryos after injection into pre-implantation embryos
In the original report, the authors showed that the cell aggregates obtained by the culture of low-pH treated cells contribute to chimeras when the cell aggregates were chosen by their morphologies under the microscopic observation, manually dissected and injected into blastocysts (Fig. 4a of Obokata et al.1). The frequency of obtaining chimeric mice from injected blastocysts reached 24% (64 chimeric mice from 264 injected blastocysts; Figure S7b of Obokata et al.1). We prepared cell aggregates from the liver cells dissociated from the livers of transgenic mice carrying CAG-EGFP16 or selected cell aggregates expressing GFP prepared from the liver cells derived from Alb-cre: Rosa-GFP double transgenic mice17,18 (Fig. 6, top) and repeated injections of these into morula and blastocysts eight times. However, we found no chimeric embryos carrying GFP-positive cells among 117 embryos derived from 244 injected embryos based on observation by fluorescent microscopy (Fig. 6, bottom), although the embryo manipulation technique of the animal facility should be excellent as the huge number of knockout mice were routinely generated by the injection of ES cells into blastosysts (http://www2.clst.riken.jp/arg/mutant_mice_generated_in_CDB.html and http://www2.clst.riken.jp/arg/publications.html). These data strongly suggest that the acquisition of pluripotency occurs in cell aggregates derived from low-pH treated liver cells only extremely rarely, if at all.
Induced cell aggregates do not give rise to stem cell lines
The original reports stated that two different types of stem cell lines could be established from cell aggregates obtained by the culture of low-pH treated cells: ES-like ‘STAP stem cells’ (Fig. 5 of Obokata et al.1) and trophoblast stem (TS)-like ‘FGF-induced (FI) stem cells’ (Fig. 2 of Obokata et al.2). To reevaluate these claims, we transferred cell aggregates derived from liver cells from various genetic backgrounds into the culture conditions for derivation of either ES-like or TS-like stem cells. In the case of the culture for ES-like stem cells in serum-free culture containing knockout serum replacement (KSR), adrenocorticotropic hormone (ACTH) and LIF19, most of the cell aggregates died without outgrowth, which may attributable to the absence of serum, while a small number aggregates gave rise to colonies containing small cells with large nuclei, resembling the morphology of embryonic stem cells. However, most of these cells ceased proliferation at day 7 and gradually regressed. Marked proliferation after day 7 was observed in only three of 492 cell aggregates and none of these gave rise to cell lines (Fig. 7a). In the case of the culture for TS-like stem cells containing FGF-4 and heparin20, many clumps showed outgrowth of fibroblastic cells, which may be due to the presence of FGF2 in the medium. Few of these (22 of 391 cell aggregates) gave rise to colonies of small stem cell-like cells, and one could be passaged three times (Fig. 7b). However, all ultimately regressed without giving rise to cell lines. These data showed that we are unable to derive stem cell lines from aggregates derived from low-pH treated liver cells.
In the present study, we investigated the properties of cell aggregates obtained by culture of liver cells transiently treated with low-pH stimulus, which was performed by the group directed by the author. We initially followed the protocol described in the original paper1 with the detail description in protocol exchange where HCl was applied to achieve low-pH condition. However, we merely obtained the cell aggregates expressing the pluripotency makers as described in this report even when it was combined with the culture in medium containing Fgf2, which was not described in the original protocol but subsequently suggested by the authors. However, when we used ATP instead of HCl, also based on a suggestion by the authors, a few cells in a subset of cell aggregates expressed the pluripotency marker Oct3/4 at levels comparable to those in ES cells that were reproducibly detected by QPCR (Fig. 3c) and immunostaining (Fig. 4b). However, the frequency was very low; 5 × 105 liver cells yielded only ~30 cell aggregates, in which about 20% of the cell aggregates contained 1–2 Oct3/4 positive cells, indicating a frequency per seeded liver cell of 0.0012–0.0024%. Moreover, the pluripotency of such cells was not confirmed by chimera formation assay, and they did not give rise to any stem cell lines. We thus conclude that such cell aggregates do not fulfill the definition for STAP cells proposed in the original studies. Moreover, since the frequency of Oct3/4-positive cells in the cell aggregates was quite low, it was impossible to determine whether they were selected from the original population or induced in culture, again highlighting the lack of evidence supporting the existence of the reported STAP phenomenon. An independent examination was made on chimeric potency of STAP-like cell aggregates that were generated by Haruko Obokata. Among 1154 embryos injected with the aggregates, 671 developed beyond E8.5; however, none of the aggregates made significant contribution to any tissues, the details of which was reported in a Biorixiv website (bioRxiv doi: http://dx.doi.org/10.1101/028472). These data are consistent to the recent report by De Los Angeles et al.22.
Materials and Methods
C57BL/6NJcl (CLEA Japan) and 129 X 1/SvJJmsSlc (Japan SLC) mice were purchased from suppliers. C57BL/6-Tg(CAG-EGFP)C14-Y01-FM131Osb transgenic mouse (CAG-GFP Tg) line was provided by Research Institute for Microbial Disease, Osaka University16. C57BL/6J-Tg(GOFGFP)11Imeg transgenic mouse (GOF-Tg) line was obtained from RIKEN Bio-Resource Center (RBRC00771)21. B6.Cg-Tg(Alb-cre)21Mgn/J transgenic mouse (Alb-cre Tg) line was supplied by Jackson Laboratory18. R26R-H2B-EGFP transgenic mouse (Rosa-GFP) line was generated by Laboratory for Animal Resources and Genetic Engineering (LARGE), RIKEN CDB17. All animal experiments were carried out in accordance with our Guidelines for the Care and Use of Laboratory animals and were approved by the Institutional Committee for Laboratory Animal Experimentation (RIKEN Kobe Institute).
Isolation of cells from mice
4–9-day-old mice were euthanized using carbon dioxide and then sterilized with 70% ethanol. For the isolation of spleen cells, excised spleen was minced with scissors and the tissue fragments were dissociated in phosphate buffered serine (PBS) by pipetting. The cell suspension was strained through a cell strainer followed by the collection of cells by centrifugation at 1,000 rpm for 5 min. The collected cells were re-suspended in 5 ml of Dulbecco’s Modified Eagle medium (DMEM; Life Technologies) and added to the same volume of Lympholyte® (Cedarlane), and then centrifuged at 1,000 g for 20 min. The lymphocyte layer was isolated and washed with PBS to obtain single cell suspension. For the isolation of liver cells, excised liver was minced with scissors and the tissue fragments were dissociated by incubation in Type I collagenase (Worthington Biochemical) solution (0.5 mg/ml in Hanks Balanced Salt Solution (HBSS, no calcium, no magnesium; Life Technologies)). Next, the cell suspension was strained through a cell strainer followed by the collection of cells by centrifugation at 1,000 rpm for 5 min. For the isolation of heart cells, the excised heart was minced with scissors and the tissue fragments were dissociated by incubation in Type II collagenase (Worthington Biochemical) solution (0.5 mg/ml in HBSS). The cell suspension was strained through a cell strainer followed by the collection of cells by centrifugation at 1,000 rpm for 5 min.
Low-pH treatment and culture of cell aggregates
Diluted HCl solution was prepared with 10 μl of 35% HCl (Nakarai) in 590 μl HBSS. Diluted ATP solution was prepared with ATP (Sigma) in distilled water at 200 mM. Titration of pH with various amount of diluted HCl or ATP was performed with 500 μl of HBSS containing 7 × 105 liver cells. As a routine method, 10 μl of either diluted HCl or ATP solution was added into 500 μl of cell suspension containing 5 × 105 cells in HBSS followed by incubation for 25 min at 37 °C, and then centrifuged at 1,000 rpm at room temperature for 5 min. After the supernatant was removed, precipitated cells were re-suspended and plated onto either adhesive or non-adhesive plates at cell density of 1–5 × 105 cells per well in 1 ml of the culture medium. The culture medium consists of DMEM/HamF12 (Life Technologies) supplemented with 1,000 U/ml of mouse LIF (home-made) and 2% of B27® Supplement (Life Technologies). Optionally, recombinant human Fgf2 (Wako) was added at final concentration of 10 ng/ml.
To quantify the levels of mRNA transcripts, total RNA was prepared by TRIzol® (Life Technologies). cDNA were synthesized from 1 μg of total RNA using SuperScript® III (Life Technologies), and quantified by real-time PCR using a CFX384 system (BioRad). All samples were tested in triplicate, and the mean relative amounts of each transcript were calculated by normalization to an endogenous control Gapdh.
Each cell aggregate was washed with PBS and transferred in 2 μl of PBS into 8 μl RealTime ready Cell Lysis Buffer (Roche) supplied with NP-40, RNAsin and RNase inhibitor. Then 3 μl of cell lysis solution was mixed with 1.5 μl of DNaseI solution (0.2 U/μl) to degradate genomic DNA followed by addition of 1.5 μl of 8 mM EDTA solution to stop the reaction. For reverse transcription of RNA, 3 μl of pre-mixture of SuperScript® VILO reverse transcriptase (Life Technologies) was added into 6 μl of DNaseI-treated cell lysate and incubated at 42 °C for 1 hour. The reverse-transcribed product was pre-amplified with Plutinum multiplex PCR master mix using pooled primer mixture using the reaction cycle (95 °C for 30 sec; 60 °C for 90 sec; 72 °C for 60 sec) for 14 cycles. The mixture was treated with Exonuclease I to remove the primers for pre-amplification, and quantitative PCR was performed with the primer pairs specific for each gene using Quantitest SYBR Green PCR mix (Qiagen) in BioRad CFX384 Real-Time System (Bio-Rad). All samples were tested in triplicate, and the mean relative amounts of each transcript were calculated by normalization to an endogenous control Gapdh or Gnb2l1.
Cells were fixed by 4% paraformaldehyde in PBS for 30 min at 4 °C and then permeabilized by 0.1% Triton X-100 in PBS for 15 minutes at room temperature (RT). After brief washing with PBS followed by blocking with PBS containing 2% FCS, the cells were incubated with the following primary antibodies: anti-Oct3/4 rabbit antiserum15 and anti-Nanog rat monoclonal antibody (R&D) for overnight at 4 °C. After washing with PBS, the cells were incubated with Alexa Fluor 488- or 633-conjugated donkey antibodies (Invitrogen) were used in a proper combination of species specificity as indicated in Figure legends. Fluorescent images were captured with an IX51 microscope with DP70 digital camera (Olympus) or a Leica SP8 confocal microscope (Leica).
For flow cytometric analyses, cell aggregates were harvested, washed by PBS, and incubated with TrypLETM Select (Life Technologies) for 5 min. After dilution with culture medium, aggregates were dissociated into single cells by gentle pipetting. Cells adhered to the culture substrate were also harvested following a standard method. These cells were mixed and collected as pellets by a centrifugation. For quantification of GFP-positive population, dissociated cells were re-suspended in 500 μl HBSS containing 1 μl of DRAQ7, a cell-nonpermeable DNA dye (for the detection of dead cells; Cell Signaling). When combined with a staining for CD45 antigen, cell pellets were suspended with 50 μl HBSS containing 10 μl of APC-conjugated rat anti-CD45 antibody (BD Pharmingen), and incubated for 30 min on ice. For co-staining with CD45/E-cadherin antibodies, cell pellets were suspended in culture medium, and then incubated for 30 min in CO2 incubator. The cells were harvested and suspended with 50 μl HBSS containing 5 μl biotin-labeled rat anti-E-cadherin antibody (ECCD2). After incubation for 30 min on ice, the stained cells were once washed by HBSS, re-suspended with 50 μl HBSS containing 1 μl PE-conjugated streptavidin (Life Technologies) and 10 μl APC-conjugated anti-CD45 antibody, and further incubated for 30 min on ice. These stained cells were once washed by HBSS and suspended with 500 μl HBSS.
After the cell suspension was passed through a filter mesh, the cells were analyzed using a FACSAria IIIu cell sorter (Becton Dickinson).
Injection into pre-implantation embryos
Cell aggregates obtained by the ATP treatment of liver cells were cut into smaller pieces using a laser (XYClone, Nikko Hansen & Co., Ltd) or shaped glass capillaries, and single piece was then microinjected into each 8-cell or blastocyst stage embryo from ICR mice (Charles River Laboratories Japan, Inc.). Injected embryos were transferred to the uterus of 2.5 dpc pseudopregnant ICR females (Charles River Laboratories Japan, Inc.) on or the next day of injection.
Culture for derivation of stem cells
The culture medium for derivation of ES-like stem cells consists of Glasgow-modified eagles medium (GMEM, Sigma), 15% KnockOut Serum Replacement® (KSR, Life Technologies), 1 × non-essential amino acids (NEAA, Nakarai), 1 × Sodium Pyruvate (Nakarai), 10−4 M 2-mercaptoethanol (Nakarai), 1,000 U/ml of LIF and 10 μM ACTH (Kurabo on consignment). We confirmed the medium is optimal for the culture of conventional ES cells. The culture medium for derivation of TS-like stem cells consists of GMEM, 20% FCS, 1 × NEAA, 1 × Sodium Pyruvate, 10−4 M 2-mercaptoethanol, 25 ng/ml of recombinant mouse Fgf4 (Wako) and 1 μg/ml of heparin (Wako). We confirmed the medium is optimal for the culture of conventional TS cells. To derive stem cells, cell aggregates were isolated under a microscope and transferred into a well of 96-well plate with 100 μl of the culture medium and 1,000 feeder cells. Feeder cells were prepared by treatment of mouse embryonic fibroblasts prepared from day 14 C57BL6 embryos with Mitomycin C (Wako) for 3 hours.
How to cite this article: Niwa, H. Investigation of the cellular reprogramming phenomenon referred to as stimulus-triggered acquisition of pluripotency (STAP). Sci. Rep. 6, 28003; doi: 10.1038/srep28003 (2016).
Obokata, H. et al. Stimulus-triggered fate conversion of somatic cells into pluripotency. Nature 505, 641–647 (2014).
Obokata, H. et al. Bidirectional developmental potential in reprogrammed cells with acquired pluripotency. Nature 505, 676–680 (2014).
Gurdon, J. B. Adult frogs derived from the nuclei of single somatic cells. Dev Biol 4, 256–273 (1962).
Takahashi, K. & Yamanaka, S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663–676 (2006).
Martin, G. R. Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc Natl Acad Sci USA 78, 7634–7638 (1981).
Evans, M. J. & Kaufman, M. H. Establishment in culture of pluripotential cells from mouse embryos. Nature 292, 154–156 (1981).
Smith, A. G. et al. Inhibition of pluripotential embryonic stem cell differentiation by purified polypeptides. Nature 336, 688–690 (1988).
Okamoto, K., Okazawa, H., Okuda, A., Sakai, M., Muramatsu, M. & Hamada, H. A novel octamer binding transcription factor is differentially expressed in mouse embryonic cells. Cell 60, 461–472 (1990).
Bradley, A., Evans, M., Kaufman, M. H. & Robertson, E. Formation of germ-line chimaeras from embryo-derived teratocarcinoma cell lines. Nature 309, 255–256 (1984).
Obokata, H. et al. Retraction: Stimulus-triggered fate conversion of somatic cells into pluripotency. Nature 511, 112 (2014).
Obokata, H. et al. Retraction: Bidirectional developmental potential in reprogrammed cells with acquired pluripotency. Nature 511, 112 (2014).
Ohtsuka, S. & Niwa, H. The differential activation of intracellular signaling pathways confers the permissiveness of embryonic stem cell derivation from different mouse strains. Development 142, 431–437 (2015).
Mizuno, N. & Kosaka, M. Novel variants of Oct-3/4 gene expressed in mouse somatic cells. J Biol Chem 283, 30997–31004 (2008).
Yeom, Y. I. et al. Germline regulatory element of Oct-4 specific for the totipotent cycle of embryonal cells. Development 122, 881–894 (1996).
Niwa, H. et al. Interaction between Oct3/4 and Cdx2 determines trophectoderm differentiation. Cell 123, 917–929 (2005).
Okabe, M., Ikawa, M., Kominami, K., Nakanishi, T. & Nishimune, Y. ‘Green mice’ as a source of ubiquitous green cells. FEBS Lett 407, 313–319 (1997).
Abe, T. et al. Establishment of conditional reporter mouse lines at ROSA26 locus for live cell imaging. Genesis 49, 579–590 (2011).
Postic, C. et al. Dual roles for glucokinase in glucose homeostasis as determined by liver and pancreatic beta cell-specific gene knock-outs using Cre recombinase. J Biol Chem 274, 305–315 (1999).
Ogawa, K., Matsui, H., Ohtsuka, S. & Niwa, H. A novel mechanism for regulating clonal propagation of mouse ES cells. Genes Cells 9, 471–477 (2004).
Tanaka, S., Kunath, T., Hadjantonakis, A. K., Nagy, A. & Rossant, J. Promotion of trophoblast stem cell proliferation by FGF4. Science 282, 2072–2075 (1998).
Ohbo, K. et al. Identification and characterization of stem cells in prepubertal spermatogenesis in mice. Dev Biol 258, 209–225 (2003).
De Los Angeles, A. et al. Failure to replicate the STAP cell phenomenon. Nature 525, E6–9 (2015).
We would like to thank the assistance of the members of the Scientific Validity Examination Team, Dr. Hiroshi Kiyonari and Mr. Kenichi Inoue for chimera production and animal breeding, and Laboratory of Animal Resources and Genetic Engineering for animal housing. We also thank Mr. Douglas Sipp for critical discussion of this report. This examination was supported by the grant for Scientific Validity Examination by RIKEN President.
The author declares no competing financial interests.
About this article
Cite this article
Niwa, H. Investigation of the cellular reprogramming phenomenon referred to as stimulus-triggered acquisition of pluripotency (STAP). Sci Rep 6, 28003 (2016). https://doi.org/10.1038/srep28003
Journal of Biological Chemistry (2019)
When the research is not reproducible: the importance of author-initiated and institution-driven responses and investigations
Accountability in Research (2018)
Stem Cell Reviews and Reports (2017)