Stem cells self-renew and generate specialized progeny through differentiation, but vary in the range of cells and tissues they generate, a property called developmental potency. Pluripotent stem cells produce all cells of an organism, while multipotent or unipotent stem cells regenerate only specific lineages or tissues. Defining stem-cell potency relies upon functional assays and diagnostic transcriptional, epigenetic and metabolic states. Here we describe functional and molecular hallmarks of pluripotent stem cells, propose a checklist for their evaluation, and illustrate how forensic genomics can validate their provenance.
At a glance
Stem cells, defined by dual hallmark features of self-renewal and differentiation potential, can be derived from embryonic and postnatal animal tissues and are classified according to their developmental potency (Fig. 1). The zygote and blastomeres are totipotent1, denoting potential to give rise to all embryonic and extra-embryonic tissues, but their developmental potential has not been captured in vitro. Mouse embryonic stem cells exemplify a quintessential pluripotent stem (PS) cell that can form all tissues of the body, but provides only limited contributions to the extra-embryonic membranes or placenta. As described in greater detail below, PS cells manifest distinct functional properties depending upon the conditions under which they are derived and cultured. Multipotent stem cells, such as the paradigmatic haematopoietic stem cell, are restricted to generating the mature cell types of their tissue of origin, but under normal physiologic circumstances will not differentiate into unrelated lineages. Unipotent stem cells, such as spermatogonial stem cells (SSCs), share the capacity for self-renewal yet exhibit limited developmental potential, giving rise to only a single cell type, such as sperm.
Human PS cells correspond to a stable state allowing propagation of immortal pluripotent cells that can generate any cell within the body. Nuclear reprogramming, via somatic cell nuclear transfer and transcription factor transduction, demonstrates that the specialized state of a somatic cell can be reversed to a totipotent or pluripotent state, respectively2, 3. The generation of induced pluripotent stem (iPS) cells from somatic cells via transcription factor expression constitutes a facile route to generate patient-specific PS cells, and has opened new paths to model diseases and new prospects for regenerative medicine. Given their versatility for medical applications, PS cells command considerable attention; therefore, defining the hallmarks of pluripotency has practical as well as fundamental value to biomedical research.
In this technical review, we describe the hallmark characteristics of PS cells, propose a checklist of assays for assessing the function and molecular state of pluripotency, and outline forensic genomic approaches to validate the provenance of reprogrammed cell lines.
Defining pluripotent stem cells
PS cells are self-renewing cells with the capacity to form representative tissues of all three germ layers of the developing embryo—ectoderm, mesoderm and endoderm, as well as the germ lineage, but typically provide little or no contribution to the trophoblast layers of placenta. PS cells can be derived from numerous sources (Table 1). The first PS cells cultured in vitro were derived from teratocarcinomas, a tumour of germ cell origin4. Later, derivation of PS cells from the murine blastocysts proved that pluripotent cells could be propagated as immortalized, non-transformed cell lines5, 6. PS cells have also been derived from non-human primate and human embryos7, 8, and from various stages of development, including the post-implantation epiblast and the germ line9, 10, 11, 12, 13, 14. Finally, somatic cells can be reprogrammed to pluripotency by ectopic expression of select sets of transcription factors3.
PS cells manifest distinct properties depending on derivation and maintenance conditions. PS cells established from pre-implantation embryos are known as ES cells, whereas those generated from slightly later embryonic epiblast stages are called epiblast stem cells (EpiSCs)9, 10. Their distinct culture requirements, gene expression programs and epigenetic features may reflect the dynamic development of pluripotency in the embryo. The terms ‘naive’ and ‘primed’ were introduced to describe early and late phases of epiblast ontogeny and respective ES cell and EpiSC derivatives15. PS cells from various sources have been classified accordingly (Table 1). Conventional human PS cells exhibit molecular attributes similar to EpiSCs and are classified as ‘primed’. Evaluation of naive pluripotency in humans by formation of human chimaeras is restricted on ethical grounds in many jurisdictions, but as conventional non-human primate ES cells fail to chimaerize pre-implantation embryos, traditional human ES cells are also probably primed by this criterion16.
Molecular hallmarks of pluripotency
PS cells are characterized by molecular mechanisms that sustain self-renewal and suppress differentiation while maintaining key differentiation genes in a quiescent yet ‘poised’ state reflective of their incipient developmental potential.
A select set of core transcription factors in combination governs and thereby defines pluripotency: OCT4 (also known as POU5F1), SOX2 and NANOG (collectively, OSN). OCT4 and NANOG are designated as core transcription factors based on their specific expression pattern in PS cells and early embryos, and genetic screens identifying their essential role in establishing pluripotency in mice and humans3, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26. OCT4 functions as a heterodimer with SOX2, placing SOX2 among the core regulators22. The generation of mouse and human iPS cells by ectopic expression of OCT4 and SOX2 highlights the pre-eminent role of OCT4/SOX2 in establishing pluripotency. Although NANOG is not required for mouse PS-cell maintenance25, and is expressed at low or absent levels in mouse EpiSCs, it stabilizes PS cells, is necessary for in vivo pluripotency to develop in the inner cell mass (ICM)26, and extensively co-localizes with OCT4 and SOX2 throughout the mouse and human PS cell genome. While the core transcription factors define and govern pluripotency, in special circumstances PS cells can tolerate loss of SOX2 or NANOG or substitution with other factors, suggesting flexibility in pluripotency governance. Among the core transcription factors, OCT4 has proven most indispensable and remains the preeminent pluripotency factor.
Mapping of OSN targets supports a model of regulatory control whereby OSN sustains self-renewal while restricting differentiation. OSN cooperatively bind their own promoters, forming an interconnected auto-regulatory loop17, 18. OSN activate a substantial fraction of protein-coding, miRNA, and non-coding RNA genes in ES cells, while also occupying genes encoding lineage-specific regulators27, 28. The promoters of many lineage regulators harbour both active (H3K4me3) and repressive (H3K27me3) histone marks, a bivalent state thought to facilitate activation of development genes upon exit from pluripotency29. The capacity of OSN to activate genes necessary for maintaining ES cells, while repressing lineage-specifying regulators, chiefly accounts for the dual hallmark features of self-renewal and differentiation potential.
While OCT4 and SOX2 are expressed in all PS cells, PS cells can be classified into different states of pluripotency based on a complement of diagnostic molecular signatures that delineate proximity to the pre-implantation ICM or post-implantation epiblast, respectively (Fig. 1c). In mice, four key distinctions amongst the various pluripotent states have been described to date: (1) X chromosome status in female cells; (2) global levels of DNA methylation; (3) Oct4 enhancer utilization; and (4) expression levels of a select group of regulators designated as ‘naive’ transcription factors: Klf4, Klf2, Esrrb, Tfcp2l1, Tbx3 and Gbx2 (refs 10, 26, 30, 31, 32, 33). These naive transcription factors, along with Nanog, are expressed at low levels or are absent in primed PS cells and can reset primed PS cells in conjunction with naive pluripotency culture conditions. The capacity of ‘naive’ transcription factors to reset primed PS cells suggests a regulatory intersection between naive transcriptional circuitry and epigenetic resetting of the DNA methylome and X chromosome.
A molecular ‘ground state’ in mouse ES cells can be enforced by cultivating cells in leukaemia inhibitory factor and small molecule inhibitors of Mek and Gsk3 kinases (2i/LIF conditions), which stabilizes the diagnostic signatures of pluripotency in the pre-implantation blastocyst30, 34, 35. Ground state ES cells exhibit two active X chromosomes in female cells, low levels of DNA methylation, preferential utilization of the Oct4 distal enhancer, and naive transcription factor expression. In contrast, an alternative primed state is favoured by cultivation in FGF/ACTIVIN. Primed EpiSCs exhibit X-chromosome inactivation in female cells, high levels of DNA methylation, preferential utilization of the Oct4 proximal enhancer, and naive transcription factor repression. The molecular changes observed when ground state ES cells transition to primed EpiSCs in vitro appear to mirror changes during maturation of pre-implantation epiblast to post-implantation epiblast in vivo14, 36.
Both naive and primed PS cells exhibit heterogeneity at the level of state markers and single cells, which we briefly discuss below. While serum-cultivated mouse PS cells form chimaeras capable of germline transmission (a functional hallmark of naive pluripotency), such PS cells also bear high DNA methylation levels reminiscent of post-implantation epiblast30. EpiSCs also exhibit heterogeneity that can be altered via signalling pathway modulation. For example, region-specific EpiSCs (rsEpiSCs) preferentially engraft into posterior epiblasts and bear diagnostic markers of the post-implantation state, consistent with their status as primed PS cells37. Yet, rsEpiSCs possess higher cloning efficiency, a feature typically associated with naive PS cells. Thus, like serum-cultivated ES cells, rsEpiSCs manifest features associated with different phases of pluripotency. Cumulatively, these observations suggest mouse pluripotency encompasses a spectrum of functional and molecular states, highlighting the imprecision of nomenclature in the face of biological complexity.
A caveat to the concept of ground state PS cells arises from single-cell studies suggesting inherent metastability in PS cells. Heterogeneous single-cell gene expression profiles, flow cytometry, and replating experiments indicate the coexistence of distinct molecular and functional states in serum-cultivated mouse ES cells38. Even individual cells in more homogeneous ground state cultures have been reported to exhibit variable pluripotency transcription factor expression39 and, while the origin and consequence of such heterogeneity are yet to be elucidated, the dynamic nature of pluripotency cannot be disregarded when classifying PS cell states. The markers distinguishing ground state from alternative PS cells remain relevant for evaluating novel PS cell types, especially claims of ground state human PS cells.
Adding additional nuance to the definitions of pluripotency, functional and molecular states are not always correlated. Mouse PS cells maintain molecular features of pluripotency, including expression of the core transcription factors, even when DNA methylation and H3K27 methylation are ablated40, 41, 42, 43, but cannot differentiate, and thereby lack functional pluripotency44. Thus, while molecular signatures can suggest pluripotency, only functional tests can establish the true developmental potential of a cell. Unlike mouse ES cells, conventional ‘primed’ human ES cells cannot tolerate DNMT1 deletion, emphasizing the functional differences between mouse and human ES cells, which we discuss in detail below45. The observation that naive cells tolerate depletion of epigenetic regulators supports the concept of naive pluripotency as a configuration with a reduced requirement for epigenetic repression compared to primed PS cells and somatic cells.
Functional assessment of pluripotency
A range of assays can be employed to reveal the developmental potential of PS cells: (1) in vitro differentiation; (2) teratoma formation; (3) chimaera formation; (4) germline transmission; (5) tetraploid complementation; and (6) single-cell chimaera formation. A summary of these assays along with their advantages and disadvantages is provided in Box 1.
Box 1: Functional assays for pluripotency.
a, Overview of functional tests to assess developmental potency of PS cells. Blue boxes indicate assays that are restricted using human cells.
b, Functional assays for pluripotency, their grades of functional stringency, and ethical permissibility when using human cells. Analysis of in vitro characteristics, such as self-renewal capacity, colony morphology (CFU, colony-forming unit), and differentiation capacity in vitro, comprise a basic layer of pluripotency characterization. In vivo assays that measure differentiation capacity are taken as more robust indicators of potency.
Mouse PS-cell potency evaluation includes aggregate in vivo assays (that is, teratoma formation, embryo chimaeras (non-gestation), germline transmission, 2n/4n gestational complementation) and single-cell in vivo assays (that is, single-cell chimaeras and single-cell input gestations). 4n tetraploid complementation and single-cell chimaera formation are taken as more stringent functional assays for pluripotency.
The teratoma assay is the gold standard functional assay for assessing human PS-cell developmental potential. Chimaerism assays of human PS cells in murine embryos, as well as formation of primary human embryo chimaeras (non-gestation), are permissible under international stem-cell research guidelines110 after rigorous scientific and ethical review. Potency evaluation of primary human chimaeras by in vivo gestational complementation in humans is ethically impermissible.
The assays for totipotency are: (1) gestation from a single input cell; and (2) gestational complementation experiments from a single cell that demonstrate contribution to all tissues of the body and high-grade placenta contribution. Note that it is not necessarily the case that if a test cell performs well in a more stringent test, that it will definitely pass a less stringent test. For example, it is unclear if totipotent cells form teratomas.
In vitro differentiation to derivatives of all three embryonic germ layers—ectoderm, mesoderm and endoderm—represents the lowest hurdle for establishing pluripotency. Typically, culture conditions that maintain pluripotency are replaced by cocktails of differentiation-inducing cytokines, morphogens or chemicals, and markers of specific target tissues are then surveyed.
The teratoma formation assay assesses the spontaneous generation of differentiated tissues from the three germ layers following the injection of cells into immune-compromised mice. Histologic analysis of teratomas is neither quantitative nor capable of assessing every possible cell type. Incompletely reprogrammed cells can generate masses that superficially resemble teratomas yet lack terminal three-germ-layer differentiation, potentially leading to misinterpretation46. Moreover, co-injection with matrices or scaffolds can elicit inflammatory or foreign-body reactions that can be misinterpreted as evidence of tissue differentiation, necessitating the use of lineage tracing or marker analysis to distinguish donor cells from reactive host tissue93. Because teratomas are not generated from single cells, the teratoma assay assesses developmental potency at a population-based level.
A third differentiation assay, blastocyst chimaera formation, measures whether test cells can re-enter development when introduced into host embryos at either of two pre-implantation stages: by aggregation with cleavage-stage morulas or by injection into blastocysts47. High-quality PS cells support normal development and generate high-grade chimaeras with extensive colonization of all embryonic tissues including the germ line, whereas less-potent PS cells produce either low chimaerism or reduced embryo viability.
A fourth assay, germline transmission, entails breeding chimaeras to produce all-donor PS cell-derived offspring, which thus demonstrates the capacity of test cells to generate functional gametes. The integration of donor cells into all tissues of viable late-stage embryos, postnatal or adult mice, followed by germline transmission, is a robust indicator of chromosomal integrity and of functional pluripotency.
A fifth assay applied to mouse cells, tetraploid complementation, measures the capacity of test PS cells to direct development of an entire organism. Donor PS cells are introduced into tetraploid (4n) host blastocysts, which are generated by electrofusion of blastomeres at the two-cell stage. Because 4n blastocysts cannot sustain normal embryonic development beyond mid-gestation, while tetraploid extra-embryonic tissues develop normally and support donor cells48, any resulting embryos are derived essentially entirely from donor PS cells.
A sixth, highly stringent assay is to inject single-donor mouse PS cells into a morula or blastocyst49. Genuine pluripotency is a property of a single cell and therefore chimaeras with widespread contribution from a single injected cell provide the clarity of clonal analysis. Both single-cell chimaerism and tetraploid complementation assays suffer from higher failure rates, but can be interpreted as the most definitive ways of demonstrating pluripotency.
Finally, while primed EpiSCs generate tri-lineage differentiation in vitro and form teratomas, EpiSCs rarely form chimaeras upon introduction into pre-implantation blastocysts. However, EpiSCs contribute to all germ layers when introduced into early post-implantation embryos in whole-embryo culture37, 50, although pluripotency of single cells has not yet been demonstrated.
Human pluripotent stem cells
Conventional human PS cells exhibit molecular hallmarks of primed state pluripotency, including preferential utilization of the OCT4 proximal enhancer, pronounced levels of DNA methylation, and a propensity for X chromosome inactivation in female cell lines51. Reports of human naive PS cells prompted some groups to attempt to assess potency by blastocyst chimaerism52, 53, 54, constrained by the widespread acceptance that culture of human embryos for more than 14 days of development in vitro, or past the point of primitive streak formation (whichever is first), is ethically impermissible. Nevertheless, both primed and altered human PS cells have been introduced into mouse pre-implantation embryos52, 53, 54, 55. Human naive PS cells engraft into the mouse ICM52, 54, although contribution to cross-species chimaeras has been minimal52 or not detectable53, 54. By contrast, region-specific human PS cells engraft into the posterior epiblast of cultured murine post-implantation embryos, indicating limited cross-species chimaerism37.
More compelling evidence for cross-species blastocyst chimaerism has been reported following injection of primate naive iPS cells into mouse blastocysts, leading to clonal contribution to solid tissues56. Whereas primate ICM cells have thus far failed to form blastocyst chimaeras, unlike mouse ICM cells16, aggregation of primate blastomeres (totipotent cells) does produce chimaerism16. Nonetheless, a recent study described altered primate PS cells that can incorporate into host embryos and develop into chimaeric fetuses with low-grade contribution to all three germ layers and early germ cell progenitors57. As in mice, high-grade contribution and germline transmission remain as more stringent tests to demonstrate naive pluripotency in primate ES cells.
Given the distinct behaviour of primate PS cells in chimaera studies, and lingering uncertainties about interspecies chimaerism, injecting human cells into mouse embryos needs additional validation before being accepted as a routine assay for stem-cell potency. Lacking robust functional assays for human stem-cell potency, transcriptional and epigenetic similarity of hypothetical ground state PS cells to the pluripotent cells in human pre-implantation embryos will remain the molecular standard for designation of human ground state PS cells (Fig. 1).
Erasure and resetting of DNA methylation is a molecular hallmark in mammalian pre-implantation and germline development. Human pre-implantation embryos have hypomethylated genomes. In contrast, ICM outgrowths undergo genomic remethylation and established human ES cells maintain pronounced DNA hypermethylation, similar to mouse primed PS cells58, 59. Such epigenetic resetting appears to be controlled by a unique regulatory network present in pre-implantation embryos and the germ line. KLF4, TFCP2L1, ESRRB, TBX3 and GBX2, transcription factors implicated in mouse naive pluripotency, have been detected in human pre-implantation epiblast and are transcriptionally repressed in derived human ES cells, similarly to mouse EpiSCs60. However, the transcripts of certain murine naive transcription factors, such as KLF2, have not been detected in the human pre-implantation epiblast, revealing complexity. Additional species-specific differences also remain unresolved. The timing of X chromosome inactivation in human embryos is contentious61, 62 and ‘epigenetic erosion’ of the X chromosome in primed human ES cells complicates our understanding of X chromosome regulation63, 64. Therefore, by current standards, we identify human ground state or naive PS cells according to molecular criteria used to delineate mouse ground state pluripotency, accepting that these criteria are tentative and subject to revision.
Acknowledging such caveats, a growing number of studies have demonstrated the feasibility of altering human PS cells towards a ‘metastable’ naive state of pluripotency52, 65, 66, 67. More convincingly, PS cells generated by the Jaenisch and Smith laboratories express transcription factors implicated in the governance of mouse ground state ES cells53, 54. While the X chromosome was inactive in human PS cells generated in the Jaenisch laboratory, we note again the uncertain significance of X chromosome status in human pluripotency52, 53, 61, 62, 63, 64. Cells ‘reset’ in the Smith laboratory exhibit a meaningful reduction in DNA methylation to levels approaching human pre-implantation embryos. However, the unclear activation of the OCT4 distal enhancer, and lack of detailed characterization of transgene-independent cell lines leaves open the question of whether the reset state is stable54.
More experimental understanding of the transition from totipotency to pluripotency in the intact human or primate embryo will be needed to truly define the human ground state PS cell. Direct derivation of ground state ES cells from human embryos would be a landmark, highlighting the continued relevance of human ES cell research.
Potency in native somatic cells
As an organism progresses from the earliest embryonic stages to adulthood its cells become progressively restricted in developmental potency, and acquire epigenetic modifications that present barriers to dedifferentiation. However, germ cells, responsible for perpetuating the species, retain a unique chromatin state receptive to reprogramming to a naive pluripotent state by signalling pathway modulation alone. Cultivation of primordial germ cells in 2i/LIF, among other culture conditions, generates chimaera-competent naive pluripotent cells68.
By contrast, acquisition of naive pluripotency from somatic cells requires the prolonged, combinatorial action of reprogramming transcription factors and ES cell growth conditions3. An exception to this principle is chemical reprogramming, suggesting that culture conditions alone can fully reverse the differentiated state to pluripotency69. Notably, the final stage of chemical reprogramming is also induced by 2i/LIF. In contrast to mouse cells, our current capacity to generate human PS cells by signalling pathway modulation alone is more limited. The pluripotency of human embryonic germ cells and adult testis-derived human PS cells, both generated by culture of human germ cells, remains contentious, and small-molecule-based reprogramming of human somatic cells to pluripotency has not yet been demonstrated70, 71, 72.
Alterations in cellular identity can accompany human disease. Chronic exposure to stomach acid from gastro-oesophageal reflux converts stratified squamous epithelium of the distal oesophagus to goblet-cell containing columnar epithelia more typical of the intestine, a condition termed Barrett’s oesophagus, which predisposes to adenocarcinoma. Metaplasia and other forms of tissue ectopias, where aberrant tissues form in unusual locations, suggest cell identity conversion occurs in the body. Thus it is intriguing to consider various claims of pluripotency for cells isolated from perinatal or somatic tissues, such as multipotent adult progenitor cells73, 74, very small embryonic-like cells75, multi-lineage differentiating stress-enduring cells76, and endogenous pluripotent stem cells77. When considering novel claims of expanded potency a strict criterion is demonstration that a single cell can differentiate into different cell types, a standard of clonal analysis lacking in most studies.
Evaluating totipotency features
A robust, bidirectional capacity to form both embryonic lineages and extra-embryonic trophoblast layers of the placenta, as well as yolk sac derivatives, distinguishes totipotency from pluripotency. While somatic cells are reset to totipotency following nuclear transfer into oocytes, to date no lab (to our knowledge) has claimed to propagate in vitro cells with totipotency equivalent to zygotes or blastomeres. Below, we briefly review previous claims of placental differentiation capacity in PS cells and propose how one might evaluate claims of totipotency (Table 2).
The most stringent demonstration of totipotency requires that a single cell produce a term birth under experimental conditions78, 79, 80, 81, a standard achieved in rodents and in non-human primates for single blastomeres extracted from pre-implantation embryos1, 82. Later-stage blastomeres may contribute to all embryonic and extra-embryonic tissues, and yet fail to support a viable conceptus because of reduced cell numbers at the blastocyst stage. Thus, an alternate and less stringent test of totipotency is the potential of genetically marked single cells to contribute extensively to both embryonic and extra-embryonic lineages after introducing donor cells into pre-blastocyst-stage embryos. In the mouse, for example, only isolated two-cell blastomeres can generate an entire conceptus82, but single blastomeres at the eight-cell stage still manifest totipotency in aggregation chimaeras1. Sister blastomeres of a four-cell stage human embryo can develop individually into blastocysts with ICM and trophectoderm cells83. An essential feature of these functional tests of totipotency is demonstration of developmental capacity at the single-cell level.
Mouse PS cells with bidirectional developmental capacity for extra-embryonic and somatic fates have been claimed following specific genetic (for example, Dnmt1 knockout84) or cell culture modifications (for example, ground state39, 85) (summarized in Table 2). ‘In vivo reprogrammed’ iPS cells purportedly contribute to the placenta, unlike ES cells or in vitro reprogrammed iPS cells86. These studies reported differentiation into trophoblast-stem-like cells and the formation of blastocyst-like structures. However, the in vivo chimaera potential of trophoblast-stem-like cells was not assessed. Further, single cells did not yield robust high-grade contribution to the placenta39. Thus, the definitive functional criterion for establishing totipotency, single-cell contribution to the trophoblast and ICM lineages, has not yet been demonstrated. The molecular changes associated with acquisition of totipotent-like developmental potential have differed across studies and include the expression of ‘2C-specific’ genes, morula-specific genes, and extra-embryonic transcription factors. Therefore, by current standards, accepting that the relevance of these molecular criteria are tentative and subject to revision, the essential criterion of totipotency remains functional, whereby a single cell generates both ICM and trophectoderm fates in a transplantation assay. Ideally, detailed assessment of embryonic and extra-embryonic tissues should be made late in gestation, so that extensive and functional contribution can be demonstrated.
Conventional primed human PS cells reportedly form both trophectoderm and primitive endoderm-like derivatives in vitro87. However, confirmation of the identity of these derivatives has proven challenging88. Injection of human naive PS cells into mouse embryos has not resulted in contribution to ICM and trophectoderm lineages. Future claims of mouse totipotent stem cells will require stringent functional and molecular validation, while in humans, molecular criteria and comparison to primate species will have to suffice to establish plausibility.
Assessing provenance and potency via genomics
Advanced sequencing platforms have allowed researchers to generate a multitude of genomic and epigenomic data (for example, RNA sequencing (RNA-seq), chromatin immunoprecipitation sequencing (ChIP-seq) and bisulfite sequencing), enabling a more comprehensive description of cellular identity. Systems-level analyses have confirmed that direct reprogramming of somatic cells largely re-establishes molecular signatures associated with ES cells89, 90. These analyses also detected low-fidelity reprogramming, such as in intermediates and cells with epigenetic memory89, 90. Recently, genomic analyses have proven instrumental in defining ground state pluripotency. Thus, while not required for routine characterization of PS cells, genomic analyses play a critical role for benchmarking novel claims of reprogramming and PS cells (Box 2).
Box 2: Forensic genomics for potency and provenance of PS cells.
DNA sequencing also provides genetic fingerprints that can eliminate cell contamination as a confounder of reported results. Because cell line contamination is widespread, applying such genotyping methods to confirm cell line provenance is appropriate91. In the case of the STAP cell phenomenon, the authors reported acid-reprogrammed PS cells with features of totipotency. Our re-analysis of genomic data revealed unexpected mismatches in sex and genotype between donor somatic cells and converted STAP cells92. Further analysis of a STAP-derived cell line, Fgf4-induced stem cells, revealed a mixture that contained trophoblast stem cells, explaining the high-grade placenta colonization reported for Fgf4-induced stem cells. These findings are consistent with and extend the results of an extensive whole-genome sequencing analysis of STAP-related samples for the RIKEN investigation93, which found contamination of purported STAP stem-cell lines with embryonic stem cells of a different genetic background94.
By contrast, forensic genomics applied to sequencing data from two reports of nuclear-transfer-derived human ES cells (NT-hESCs) have confirmed cell line provenance95, 96. Inferred genome-wide single nucleotide variants (SNVs) from exome sequencing data classified samples generated in the Egli laboratory as genetically similar or dissimilar (Fig. 2). Parental donor fibroblasts and NT-hESCs possessed similar SNV profiles, consistent with nuclear transfer origin. Independently sourced in vitro-fertilization-derived ES cells and parthenogenetic ES cells manifest distinct genetic provenance from parental donor fibroblasts and NT-hESCs, as expected. SNV genotyping also confirmed previously reported patterns of recombination in human parthenogenetic ES cells, concordant with observations in mouse parthenogenetic ES cells97, 98. Matching genotypes between parental fibroblasts and reprogrammed NT-hESCs were also confirmed in RNA-seq data generated in the Mitalipov laboratory (Supplementary Fig. 1). Collectively, these analyses support appropriate provenance of NT-hESCs and exclude a parthenogenetic origin for NT-hESCs.
Reproducibility of computational analyses
As genomic analyses can validate the provenance and confirm molecular signatures of novel PS cells, we advocate posting of relevant genomic data, metadata, and full details of computational analysis upon manuscript publication. Deposition of sequencing data to public repositories such as the Gene Expression Omnibus (GEO; http://www.ncbi.nlm.nih.gov/geo/) and Short Read Archive (SRA; http://www.ncbi.nlm.nih.gov/sra) is required by most peer-reviewed journals, but enforcement of sharing policies is highly variable, and complicated by the complexities of experimental design and data. Consequently, verification that full data and associated metadata have been deposited often requires expertise and time beyond what is available during peer review. Greater compliance by the stem-cell community in depositing all relevant genomic data and metadata as well as consistent enforcement by journals will promote reproducibility of results. We also recommend deposition of ‘intermediate’ data, the key steps and results obtained in the data analysis process. For full reproducibility of computational analysis, we also advocate release of the computer code, through a supplementary website or open source code management tools. We note that genomic analysis and availability of data, metadata, and methods are especially important for novel claims of reprogramming and altered stem-cell states.
Conclusion and future prospects
Here, we articulate a consensus definition of pluripotency predicated on both functional assessments of differentiation potential and diagnostic molecular signatures. Such an integration of functional and molecular hallmarks of pluripotency provides for a robust set of criteria against which to validate claims of pluripotency achieved by novel experimental strategies. Given the central role of core transcription factors in reprogramming somatic cells and maintaining the pluripotent state, failure to observe ES-cell-like levels of these transcription factors in studies asserting functional pluripotency from novel sources should merit scepticism and should be accompanied by strong evidence for alternative gene regulatory networks and mechanisms that maintain the unique pluripotent state of the mammalian genome. Another example of uncoupling between molecular and functional hallmarks is a report that overexpression of cell adhesion molecules such as E-cadherin can endow primed PS cells with the capacity to chimaerize the pre-blastocyst, with no evidence of resetting to naive pluripotency99. Conversely, recent reports suggesting that reprogramming transitions through a transient state that molecularly resembles naive pluripotency, but without functional hallmarks of naive pluripotency, might not comprise bona fide naive pluripotency100. While most labs deriving PS cells for routine use need not employ the comprehensive set of assays reviewed here, claims of novel states of potency or new means of deriving PS cells necessitate more comprehensive characterization and documentation.
Documentation of PS-cell states that span the continuum between ground state pluripotency and primed pluripotency provokes the question of how to define the human ground state. Further, reports that human PS cells can be ‘reset’ imply the feasibility of generating PS cells with bona fide totipotency. Ultimately, refined molecular benchmarking of reprogramming and more predictable experimental capture of altered pluripotent states requires a more sophisticated understanding of human pre-implantation development.
For lasting scientific impact, claims of reprogramming and altered states of pluripotency should be broadly applicable to more than one experimental model and be independently replicated by multiple laboratories. Before publication, we encourage that researchers claiming landmark reprogramming advances first demonstrate replication by independent laboratories and incorporate forensic genomic analyses to confirm appropriate cell provenance. Science is ultimately a self-correcting process where the scientific community plays a crucial and collective role.
- Studies of the developmental potential of 4- and 8-cell stage blastomeres. J. Exp. Zool. 200, 365–376 (1977)
- The developmental capacity of nuclei taken from intestinal epithelium cells of feeding tadpoles. J. Embryol. Exp. Morphol. 10, 622–640 (1962)
A pioneering study that demonstrated that somatic cells can be reset to an early embryonic state via nuclear transplantation into eggs.
- Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126, 663–676 (2006)
The landmark paper establishing that four transcription factors can reprogram somatic cells to a pluripotent state.
- Studies on transplantable testicular teratomas of strain 129 mice. J. Natl. Cancer Inst. 20, 1257–1275 (1958)
- Establishment in culture of pluripotential cells from mouse embryos. Nature 292, 154–156 (1981)
A paper establishing that pluripotent stem cells can be isolated from mouse blastocysts and be propagated in vitro as continuously growing cell lines.
- 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)
A paper establishing that pluripotent stem cells can be isolated from mouse blastocysts and be propagated in vitro as continuously growing cell lines.
- Embryonic stem cells derived from human blastocysts. Science 282, 1145–1147 (1998)
The landmark paper establishing that pluripotent stem cells can be isolated from human blastocysts.
- Isolation of a primate embryonic stem cell line. Proc. Natl Acad. Sci. USA 92, 7844–7848 (1995) et al.
- Derivation of pluripotent epiblast stem cells from mammalian embryos. Nature 448, 191–195 (2007)
One of two pioneering studies that established that an alternative pluripotent state can be isolated from post-implantation mouse embryos resembling conventional human ES cells, suggesting that human ES cells might correspond to a post-implantation state.
- New cell lines from mouse epiblast share defining features with human embryonic stem cells. Nature 448, 196–199 (2007)
One of two pioneering studies that established an alternative pluripotent state from post-implantation mouse embryos that resembles conventional human ES cells, suggesting that human ES cells might correspond to a post-implantation state.
- Derivation of pluripotential embryonic stem cells from murine primordial germ cells in culture. Cell 70, 841–847 (1992) , &
- Generation of pluripotent stem cells from neonatal mouse testis. Cell 119, 1001–1012 (2004) et al.
- Induction of pluripotency in adult unipotent germline stem cells. Cell Stem Cell 5, 87–96 (2009) et al.
- The ability of inner-cell-mass cells to self-renew as embryonic stem cells following epiblast specification. Nature Cell Bio. 16, 516–528 (2014) et al.
- Naive and primed pluripotent states. Cell Stem Cell 4, 487–492 (2009) &
- Generation of chimeric rhesus monkeys. Cell 148, 285–295 (2012) et al.
- Core regulatory circuitry in human embryonic stem cells. Cell 122, 947–956 (2005)
A study that affirmed the principle that OCT4, SOX2, and NANOG constitute a core regulatory circuitry that explains the self-renewal and differentiation capacity of PS cells.
- The Oct4 and Nanog transcription network regulates pluripotency in mouse embryonic stem cells. Nature Genet. 38, 431–440 (2006) et al.
- Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 95, 379–391 (1998) et al.
- Functional expression cloning of Nanog, a pluripotency sustaining factor in embryonic stem cells. Cell 113, 643–655 (2003) et al.
- The homeoprotein Nanog is required for maintenance of pluripotency in mouse epiblast and ES cells. Cell 113, 631–642 (2003) et al.
- Multipotent cell lineages in early mouse development depend on SOX2 function. Genes Dev. 17, 126–140 (2003) et al.
- Pluripotency governed by Sox2 via regulation of Oct3/4 expression in mouse embryonic stem cells. Nature Cell Biol. 9, 625–635 (2007) et al.
- Distinct lineage specification roles for NANOG, OCT4, and SOX2 in human embryonic stem cells. Cell Stem Cell 10, 440–454 (2012) , , , &
- Nanog safeguards pluripotency and mediates germline development. Nature 450, 1230–1234 (2007) et al.
- Nanog is the gateway to the pluripotent ground state. Cell 138, 722–737 (2009) et al.
- Integration of external signaling pathway with the core transcriptional network in embryonic stem cells. Cell 133, 1106–1117 (2008) et al.
- An extended transcriptional network for pluripotency of embryonic stem cells. Cell 132, 1049–1061 (2008) , , , &
- A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell 125, 315–326 (2006) et al.
- Naive pluripotency is associated with global DNA hypomethylation. Nature Struct. Mol. Biol. 20, 311–316 (2013)
The first study to describe the association between 2i cultivation and DNA hypomethylation, linking in vitro naive pluripotency with the DNA hypomethylation observed in pre-implantation embryos and the germ line.
- Klf4 reverts developmentally programmed restriction of ground state pluripotency. Development 136, 1063–1069 (2009)
The first paper to convert EpiSCs to naive pluripotency, affirming the concept that the naive and primed states of pluripotency are interconvertible.
- Epigenetic reversion of post-implantation epiblast to pluripotent embryonic stem cells. Nature 461, 1292–1295 (2009) et al.
- Defining an essential transcription factor program for naive pluripotency. Science 344, 1156–1160 (2014) , , , &
- The ground state of embryonic stem cell self-renewal. Nature 453, 519–523 (2008)
The first study to report the remarkable synergism between MEK and GSK3 inhibition and the first articulation of the concept of a pluripotent ‘ground state’.
- The transcriptional and epigenomic foundations of ground state pluripotency. Cell 149, 590–604 (2012) et al.
- A unique regulatory phase of DNA methylation in the early mammalian embryo. Nature 484, 339–344 (2012) et al.
- An alternative pluripotent state confers interspecies chimeric competency. Nature 521, 316–321 (2015) et al.
- Deconstructing transcriptional heterogeneity in pluripotent stem cells. Nature 516, 56–61 (2014) et al.
- Totipotent embryonic stem cells arise in ground-state culture conditions. Cell Rep. 3, 1945–1957 (2013) et al.
- Polycomb repressive complex 2 is dispensable for maintenance of embryonic stem cell pluripotency. Stem Cells 26, 1496–1505 (2008) , &
- DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99, 247–257 (1999) , , &
- Target mutation of the DNA methyltransferase gene results in embryonic lethality. Cell 69, 915–926 (1992) , &
- Maintenance of self-renewal ability of mouse embryonic stem cells in the absence of DNA methyltransferases Dnmt1, Dnmt3a, and Dnmt3b. Genes Cells 11, 805–814 (2006) et al.
- Loss of methylation activates XIST in somatic but not in embryonic cells. Genes Dev. 9, 2325–2334 (1995) , &
- Targeted disruption of DNMT1, DNMT3A, and DNMT3B in human embryonic stem cells. Nature Genet. 47, 469–478 (2015) et al.
- Live cell imaging distinguishes bona fide human iPS cells from partially reprogrammed cells. Nature Biotechnol. 27, 1033–1037 (2009) et al.
- Embryonic stem cells alone are able to support fetal development in the mouse. Development 110, 815–821 (1990) , et al.
- Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc. Natl Acad. Sci. USA 90, 8424–8428 (1993) , , , &
- At most three ES cells contribute to the somatic lineages of chimeric mice produced by ES-tetraploid complementation. Dev. Biol. 275, 192–201 (2004) &
- In vivo differentiation potential of epiblast stem cells revealed by chimeric embryo formation. Cell Rep. 2, 1571–1578 (2012) , , &
- Human embryonic stem cells with biological and epigenetic characteristics similar to those of mouse ESCs. Proc. Natl Acad. Sci. USA 107, 9222–9227 (2010) et al.
- Derivation of novel human ground state naive pluripotent stem cells. Nature 504, 282–286 (2013) et al.
- Systematic identification of culture conditions for induction and maintenance of naive human pluriptoency. Cell Stem Cell 15, 471–487 (2014) et al.
- Resetting transcription factor control circuitry toward ground-state pluripotency in human. Cell 158, 1254–1269 (2014) et al.
- Contribution of human embryonic stem cells to mouse blastocyst. Dev. Biol. 295, 90–102 (2006) , , &
- Generation of naive induced pluripotent stem cells from rhesus monkey fibroblasts. Cell Stem Cell 15, 488–496 (2014) et al.
- Generation of cynomolgus monkey chimeric fetuses using embryonic stem cells. Cell Stem Cell 17, 116–124 (2015) et al.
- The DNA methylation landscape of human early embryos. Nature 511, 606–610 (2014) et al.
- DNA methylation dynamics of the human preimplantation embryo. Nature 511, 611–615 (2014) et al.
- Single-cell RNA-seq profiling of human preimplantation embryos and embryonic stem cells. Nature Struct. Mol. Biol. 20, 1131–1139 (2013) et al.
- Eutherian mammals use diverse strategies to initiate X-chromosome inactivation during development. Nature 472, 370–374 (2011) et al.
- Tracking the progression of the human inner cell mass during embryonic stem cell derivation. Nature Biotechnol. 30, 278–282 (2012) et al.
- X-chromosome inactivation and epigenetic fluidity in human embryonic stem cells. Proc. Natl Acad. Sci. USA 105, 4820–4825 (2008) , , &
- Molecular signatures of human induced pluripotent stem cells highlight sex differences and cancer genes. Cell Stem Cell 11, 75–90 (2012) et al.
- Induction of a human pluripotent state with distinct regulatory circuitry that resembles preimplantation epiblast. Cell Stem Cell 13, 663–675 (2013) et al.
- Derivation of naive human embryonic stem cells. Proc. Natl Acad. Sci. USA 111, 4484–4489 (2014) et al.
- Platform for induction and maintenance of transgene-free hiPSCs resembling ground state pluripotent stem cells. Stem Cell Rep. 2, 366–381 (2014) et al.
- Rebuilding pluripotency from primordial germ cells. Stem Cell Rep. 1, 66–78 (2013) et al.
- Pluripotent stem cells induced from mouse somatic cells by small-molecule compounds. Science 341, 651–654 (2013) et al.
- Derivation of pluripotent stem cells from cultured human primordial germ cells. Proc. Natl Acad. Sci. USA 95, 13726–13731 (1998) et al.
- Generation of pluripotent stem cells from adult human testis. Nature 456, 344–349 (2008); retraction 512, 338 (2014) et al.
- Human adult germline stem cells in question. Nature 465, E1 (2010) et al.
- Characterization of multipotent adult progenitor cells, a subpopulation of mesenchymal stem cells. Ann. NY Acad. Sci. 938, 231–235 (2001) &
- Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 418, 41–49 (2002) et al.
- A population of very small embryonic-like (VSEL) CXCR4+SSEA-1+Oct4+ stem cells identified in adult bone marrow. Leukemia 20, 857–869 (2006) et al.
- Isolation, culture, and evaluation of multilineage-differentiating stress-enduring (Muse) cells. Nature Protocols 8, 1391–1415 (2013) et al.
- Rare somatic cells from human breast tissue exhibit extensive lineage plasticity. Proc. Natl Acad. Sci. USA 110, 4598–4603 (2013) et al.
- Experiments on developing rats: II. The development of isolated blastomeres and fused eggs. J. Exp. Zool. 90, 441–459 (1942) &
- Experiments on the development of isolated blastomeres of mouse eggs. Nature 184, 1286–1287 (1959)
- Attempts to produce monozygotic quadruplets in cattle by blastomere separation. Vet. Rec. 108, 211–213 (1981) &
- Monozygotic twinning in rhesus monkeys by manipulation of in vitro-derived embryos. Biol. Reprod. 66, 1449–1455 (2002) et al.
- Postimplantation development of blastomeres isolated from 4- and 8-cell mouse eggs. J. Embryol. Exp. Morphol. 36, 283–290 (1976)
- The four blastomeres of a 4-cell stage human embryo are able to develop individually into blastocysts with inner cell mass and trophectoderm. Hum. Reprod. 23, 1742–1747 (2008) , , , &
- Epigenetic restriction of embryonic cell lineage fate by methylation of Elf5. Nature Cell Biol. 10, 1280–1290 (2008) et al.
- Embryonic stem cell potency fluctuates with endogenous retrovirus activity. Nature 487, 57–63 (2012) et al.
- Reprogramming in vivo produces teratomas and iPS cells with totipotency features. Nature 502, 340–345 (2013) et al.
- BMP4 initiates human embryonic stem cell differentiation to trophoblast. Nature Biotechnol. 20, 1261–1264 (2002) et al.
- BRACHYURY and CDX2 mediate BMP-induced differentiation of human and mouse pluripotent stem cells into embryonic and extraembryonic lineages. Cell Stem Cell 9, 144–155 (2011) et al.
- Dissecting direct reprogramming through integrative genomic analysis. Nature 454, 49–55 (2008) et al.
- Epigenetic memory in induced pluripotent stem cells. Nature 467, 285–290 (2010) et al.
- A resource for cell line authentication, annotation, and quality control. Nature 520, 307–311 (2015) et al.
- Failure to replicate the STAP cell phenomenon. Nature http://dx.doi.org/10.1038/nature15513 (this issue) et al.
- Report on STAP Cell Research Paper Investigation. http://www3.riken.jp/stap/e/c13document52.pdf (2014) .
- STAP cells are derived from ES cells. Nature http://dx.doi.org/10.1038/nature15366 (this issue) et al.
- Human embryonic stem cells derived by somatic cell nuclear transfer. Cell 153, 1228–1238 (2013)
This study was the first to demonstrate the feasibility of somatic cell nuclear transfer to reset human cells to totipotency.
- Human oocytes reprogram adult somatic nuclei of a type I diabetic to diploid pluripotent stem cells. Nature 511, 177–183 (2014) et al.
- Histocompatible embryonic stem cells by parthenogenesis. Science 315, 482–486 (2007) et al.
- Recombination signatures distinguish embryonic stem cells derived by parthenogenesis and somatic cell nuclear transfer. Cell Stem Cell 1, 346–352 (2007) et al.
- E-cadherin promotes incorporation of mouse epiblast stem cells into normal development. PLoS ONE 7, e45220 (2012) , &
- Integrative analyses of human reprogramming reveal dynamic nature of induced pluripotency. Cell 162, 412–424 (2015) et al.
- The role of the paternal genome in the development of the mouse germ line. Curr. Biol. 7, 881–884 (1997) , &
- Efficient establishment of mouse embryonic stem cell lines from single blastomeres and polar bodies. Stem Cells 25, 986–993 (2007) et al.
- Long-term proliferation of mouse germ cells in culture. Nature 359, 550–551 (1992) , , &
- Generation of functional multipotent adult stem cells from GPR125+ germline progenitors. Nature 449, 346–350 (2007) et al.
- Isolation of pluripotent embryonic stem cells from reprogrammed adult mouse somatic cell nuclei. Curr. Biol. 10, 989–992 (2000) et al.
- A comparative study of eight cell lines derived from human testicular teratocarcinoma. Int. J. Cancer 26, 269–280 (1980) , , , &
- Patient-specific stem cell lines derived from human parthenogenetic blastocysts. Cloning Stem Cells 9, 432–439 (2007) et al.
- Human embryonic stem cell lines derived from single blastomeres. Nature 444, 481–485 (2006) , , , &
- Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131, 861–872 (2007) et al.
- Guidelines for the conduct of human embryonic stem cell research. http://www.isscr.org/docs/default-source/hesc-guidelines/isscrhescguidelines2006.pdf (2006) .
- A bioinformatics assay for pluripotency in human cells. Nature Methods 8, 315–317 (2011) et al.
- CellNet: network biology applied to stem cell engeineering. Cell 158, 903–915 (2014) et al.
- Guide to human cell line authentication (2012. http://standards.atcc.org/kwspub/home/the_international_cell_line_authentication_committee-iclac_/Authentication_SOP.pdf .
We would like to thank B. Johannesson and D. Egli for providing sequencing data from nuclear-transfer-derived human embryonic stem cells. We would also like to thank P. J. Tesar, S. Byrne, A. Urbach, Y.-H. Loh, R. Zhao, K. Tsankov, J. Powers, T. Schlaeger, L. Daheron, N. Shyh-Chang, Y. S. Chan, and other members of the Daley laboratory for critical reading of this manuscript. G.Q.D. is an Investigator of the Howard Hughes Medical Institute.
- Supplementary Information (580 KB)
This file contains Supplementary Figure 1, Supplementary Notes and additional references.