- Email: [email protected]
Environmental epigenetic modifications and reprogramming-recalcitrant genes
Stem Cell Research (2010) 4, 157–164
available at www.sciencedirect.com
www.elsevier.com/locate/scr
REVIEW
Environmental epigenetic modifications and reprogramming-recalcitrant genes Kazuhiro Sakurada ⁎ Sony Computer Science Laboratories Inc., Takanawa Muse Bldg, 3-14-13, Higashigotanda, Shinagawa-ku, Tokyo, 141-0022 Japan Received 13 October 2009; received in revised form 15 January 2010; accepted 15 January 2010 Abstract The term “environmental epigenetic modifications” refers to alterations in phenotype triggered by environmental stimuli via epigenetic mechanisms. Epidemiologic and animal model studies show that a subset of such environmental epigenetic marks may affect susceptibility to chronic diseases. A growing body of evidence regarding incompleteness of reprogramming indicates that the potential retention of pathogenic environmental epigenetics in human induced pluripotent stem cells (iPSCs) should be seriously considered. Given this possibility, the optimization of methods for the generation of human induc pluripotent stem cells may require the identification of epigenetically appropriate somatic cell sources. Similarly, techniques for controlling epigenetic modification by environmental factors may also play a critical role in the development of epigenetically stable sources of pluripotent stem cells. © 2010 Elsevier B.V. All rights reserved.
Contents Environmental epigenetic effects . . . . . . . . . . . . . . . . . . . . . Stem cells as generators and transmitters of environmental epigenetic The nature of reprogramming-recalcitrant genes . . . . . . . . . . . . . Challenges in the application of human iPSCs. . . . . . . . . . . . . . . Optimization of human iPSC generation . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Although basic biomedical research has yielded an enormous amount of information on disease mechanisms, the recent decline in the development of new drugs and treatments suggests that our fundamental understanding of pathogenesis remains insufficient (Butler, 2008). The major characteristics ⁎ Fax: +81 3 5448 4273. E-mail address: [email protected].
. . . . . . . . . modifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
159 160 161 161 161 163 163 163
of the outputs of scientific investigations and their applications in technology include predictability, reproducibility, and controllability, and, importantly, findings are expected to be generalizable in the sense that when two systems are uniform and constant, the causes of events in one system can serve as the basis for predictions in its counterpart. The human body, however, is exceedingly complex and diverse, which frustrates the direct use of predictions from models in
1873-5061/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.scr.2010.01.001
158 medical care. The genetic sequences of any two individuals will show several millions of single nucleotide substitutions, and the human phenotype also changes as a result of epigenetic mechanisms. It is thus understandable that predicting and controlling such a complex, diverse, dynamic, and irreversible organism as the human body represents a tremendous scientific challenge. In clinical studies, population-based statistics are used to make inferences about heterogeneous patient groups, and the uniformity of components of therapeutic entities in the face of this interindividual diversity remains a major challenge in cell-based medicine. The advent and advances in the study of human induced pluripotent stem cells (iPSCs) have raised considerable public interest, particularly with respect to their potential applications in stem-cell-based therapy and in vitro cellular models of disease (Takahashi et al., 2007; Yu et al., 2007; Park et al., 2008; Masaki et al., 2008; Lowry et al., 2008; Park et al., 2008; Dimos et al., 2008; Ebert et al., 2009; Soldner et al., 2009; Raya et al., 2009). This enthusiasm is partially rooted in the assumption that human iPSCs are equivalent to human embryonic stem (ES) cells. This assumption, however, remains to be substantiated. In mammalian embryonic development, chromosomal DNA undergoes two different physiological reprogramming events: a first round following soon after fertilization, and a second in primordial germ cells (Reik, 2007). Under physiological conditions, only germline-derived chromosomes are subject to these reprogramming processes (Fig. 1). Technologies that enable artificial reprogramming, such as somatic cell nuclear transfer (SCNT), ES cell-somatic cell fusion, and the induction of pluripotency in somatic cells, are thought to mimic the physiological reprogramming that
K. Sakurada takes places after fertilization, as such methods were developed to establish a post-fertilization environment for somatic chromosomes (Fig. 1) (Gurdon and Melton, 2008). Although SCNT has shown that the genome state can be reprogrammed to an early embryonic (zygotic) pattern, the use of somatic cell nuclei from adult animals is inefficient, and fewer than 6% of cloned mice embryos develop into births (Rideout et al., 2001; Kishigami and Wakayama, 2009). The process is also error-prone, and cloned mammals of different species exhibit a variety of abnormal phenotypes, including placental hyperplasia, large fetus syndrome, immune disorders, and shortened life span (Tanaka et al., 2001; Tamashiro et al., 2002; Ogonuki et al., 2002). Tissue-specific gene expression profiles in cloned mice are also extremely different from those generated by natural reproduction (Kohda et al., 2005). Interestingly, most of the abnormal phenotypes in cloned mice disappear after a single generation, suggesting epigenetic alterations as a possible cause of the abnormalities in gene expression profile and phenotype (Tamashiro et al., 2002). Lending credence to this notion is evidence that the epigenetic memory of active and repressed transcription in nuclear-donor somatic cells can be inherited through SCNT, in both mouse and Xenopus laevis (Ng and Gurdon, 2005; Boiani et al., 2002; Bortvin et al., 2003). Human iPSC colonies expressing Nanog and/or ALP (alkaline phosphatase) following the forced expression of Oct3/4, Sox-2, c-Myc and Klf4 show significant heterogeneity characterized by a number of hallmark features (Fig. 2) (Masaki et al., 2008). Such heterogeneity indicates the different degrees of reprogramming in Nanog-and ALPpositive colonies. Partially reprogrammed cells have also been observed in human iPSCs induced by the forced
Figure 1 Physiological and artificial reprogramming. In physiological conditions, chromosomal DNA of germ cells are the sole source for reprogramming. Somatic cells receive epigenetic modification during developmental and postnatal life. In physiological conditions, chromosomal DNA of somatic cells and these epigenetic modifications are not inherited by the next generation, which is called the Weismann Barrier. Environmental epigenetic modifications on chromosomes in germline cells and somatic cells must be different.
Reprogramming-recalcitrant genes
159
Figure 2 Heterogeneity of ALP-positive colonies induced by Oct3/4, Sox2, c-Myc and Klf4 genes from human dermis-derived cells. (A) Heterogeneity of ALP-positive colonies was identified by analyzing the gene expression profile of eight marker genes. Although Nanog gene expression was observed in 161 of 163 ALP-positive colonies, only four colonies expressed eight human ES cell marker genes. The gene expression profile of the remaining 159 colonies was significantly heterogeneous; however, it had a characteristic hierarchical feature. (B) The eight genes are aligned in the order of tendency to be expressed in ALP-positive colonies, Nanog N TDGF1 N Dnmt3b N Zfp42 N FoxD3 N GDF3 N CYP26A1 N TERT. Eight genes were not randomly expressed by their tendency but showed a hierarchical feature that expresses lower-tendency genes in colonies where higher-tendency genes are expressed. (C) Relation between the transcriptional factor binding sites and the tendency of marker gene expression. Data was obtained from reference (Masaki et al., 2008).
expression of a different set of genes (Yu et al., 2007). Recently, Chin et al. identified and described several important features of a number of differentially expressed genes in human ESCs and iPSCs (Chin et al., 2009). I will refer to these as “reprogramming-recalcitrant” genes, as they resist the induction of a transcriptional state identical to that seen in embryonic stem cells. The underlying causes can be categorized broadly as 1) insufficient induction of human ES cell-specific genes; 2) insufficient suppression of somatic cell-specific genes; and 3) induction of human iPSC-specific genes. Of these, 1 and 2 correspond roughly to the causes of recalcitrance to reprogramming by SCNT (Ng and Gurdon, 2005). It is important to recognize that genes that are not expressed in pluripotent cells cannot be detected by gene expression profiling in self-renewing human iPSCs, meaning that there may be more reprogramming-recalcitrant genes in human iPSCs than have been reported to date. The measurement of genome-wide epigenetic marks is one means of addressing this limitation. It was recently reported
that, in human iPSCs, the DNA methylation landscapes of chromosomes 12 and 20 are different from those of the same chromosomes in human ES cells (Deng et al., 2009). This raises the question of whether the incomplete resetting of the nuclear state is a hallmark of artificial or postfertilization reprogramming or, in other words, whether it is possible to achieve complete reprogramming by mimicking post-fertilization reprogramming. The mechanistic basis and functional roles of reprogramming-recalcitrant genes in the resetting of somatic nuclei is also of great interest for further study.
Environmental epigenetic effects Epigenetics refers to heritable modifications in gene function without alteration of DNA sequences (Goldberg et al., 2007). Chemical modifications to DNA and chromatin proteins are the underlying drivers of epigenetic modifications. Although
160 human DNA methylation occurs exclusively at cytosine residues in CpG dinucleotides in differentiated cells, nonCpG methylation, which is primarily observed at CpA nucleotides, has recently been detected in stem cells (Lister et al., 2009). Chromatin itself is composed of DNA and histone proteins, which are subject to more than 100 different types of modification, including methylation, acetylation, phosphorylation, and ubiquitination (Goldberg et al., 2007). The term “epigenetics” was coined by Waddington to describe the mechanisms necessary for the unfolding of genetic information in ontogenic development (Waddington, 1942). Riggs (Riggs, 1975) and Holiday and Pugh (Holliday and Pugh, 1975) proposed that the programmed methylation and demethylation of DNA might be the molecular mechanism behind Waddington's hypothesis, and it has since been shown that Xchromosome-inactivation, genome imprinting, and lineagespecific gene silencing all rely on DNA methylation. Epigenetics do not function solely in the establishment and maintenance of tissue-and cell-type-specific gene expression. In addition to tissue-specific DNA methylation regions (T-DMRs), alternative DNA methylation regions (ADMRs) have been identified by analyses of the DNA methylation profiles of human chromosomes 6, 20, and 22 (Eckhardt et al., 2006). In this study, DNA methylation of TDMRs was found to occur preferentially in evolutionarily conserved non-protein coding regions (ECRs) several kbp distant from core promoters, while A-DMRs, which exhibit cell mosaicism, were observed within the promoter. DNA methylation profiles of A-DMRs vary between different tissues in the same donor, as well as in the same tissues or organs in different individuals, indicating that such regions are acquired throughout life and depend on the specific environmental stimuli at the cellular level. Monozygotic (MZ) twins are highly similar in appearance, especially at birth, yet they are frequently divergent for several important phenotypes, including the onset of common diseases (Wong et al., 2005). The concordance of susceptibility to age-dependent diseases in MZ twins can be quite low: 15% for breast cancer, 25%-30% for multiple sclerosis, 25%-45% for diabetes, 50% for schizophrenia, and 40-70% for Alzheimer's disease (Wong et al., 2005; Petronis, 2006). Phenotypic divergence increases with age, and greater differences are seen between MZ twins raised and living in separate environments. Several different mechanisms contribute to these changes, one of which is epigenetic differences. To address this issue, Fraga et al. analyzed global and locus-specific DNA methylation and histone acetylation of a large cohort of MZ twins and showed that, although such twins are epigenetically indistinguishable in the early years of life, older MZ twins exhibit remarkable differences in the overall content and distribution of 5-methylcytosine and histone acetylation (Fraga et al., 2005). Findings from human epidemiological and animal studies indicate that environmental stimuli can induce permanent changes in phenotype, including metabolism and susceptibility to chronic disease, at critical periods during pre-and postnatal development, (Jirtle and Skinner, 2007). It has also been suggested that epigenetic mechanisms are likely involved in the developmental origins of health and disease (Waterland and Michels, 2007). Although this remains the subject of active discussion, environmental epigenetics
K. Sakurada may reflect an adaptive response to a given prenatal environment that serves to optimize the postnatal phenotype to meeting the challenges of that environment (Gluckman and Hanson, 2004; Huxley et al., 2007). It has been suggested, for example, mismatches between pre-and postnatal environments may be linked to increases in the risk of developing chronic diseases, including type II diabetes, hypertension, and coronary heart disease (Gluckman and Hanson, 2004). Although the molecular mechanisms by which environmental stimuli alter gene expression to achieve phenotypic diversity are largely unknown, changes in the concentration of fetal hormones, such as glucocorticoids, may serve as molecular cues. Glucocorticoids have been shown to alter DNA methylation at the promoter of the glucocorticoid receptor. Maternal behaviors in rat, for example, have been shown to influence the epigenetic modification of the promoter of the glucocorticoid receptor in neurons (Weaver et al., 2004). In this study, the glucocorticoid receptor gene promoter was epigenetically repressed in the hippocampus of the offspring of low “licking and arched-back nursing” mothers (Weaver et al., 2004). Another study has shown that chronic social defeat stress induces methylation of lysine 27 (K27) of the histone H3 adjacent to the promoter region of the brain-derived neurotrophic factor (BDNF) gene in hippocampal neurons (Tsankova et al., 2006). Developing organisms are also extremely sensitive to chemical perturbation. Environmental compounds, drug administration, and physiological stresses in embryogenesis and the early postnatal period have been demonstrated to influence the pathogenesis and course of adult onset diseases (Hellwig et al., 2000). For example, endocrine disruptors, such as vinclozolin and diethylstilbestrol, induce reproductive and endocrine defects (Hellwig et al., 2000; Klip et al., 2002). The frequency and reproducibility of such abnormal phenotypes strongly suggest that epigenetic modifications, not genetic mutations, are causative, and indeed, aberrant DNA methylation of several genes has been found in tissues and cells exposed to diethylstilbestrol (Bromer et al., 2009). Cultured cells exposed to benzopyrene additionally exhibited genome-wide alterations in histone H3 lys9 acetylation (Sadikovic et al., 2008). Clearly, exposure to different environmental factors, including nutritional, endocrine, and chemical perturbation throughout life, influences the epigenetic landscape of each individual. These epigenetic modifications can be distinguished with developmental epigenetics and is called environmental epigenetics.
Stem cells as generators and transmitters of environmental epigenetic modifications Environmental epigenetic modifications introduced into terminally differentiated cells of post-mitotic tissues or organs, such as neurons or cardiomyocytes, can play indispensable functional roles (Weaver et al., 2004; Tsankova et al., 2006; Zhang et al., 2002), but are not transmitted. In contrast, epigenetic modifications in somatic stem cells in mitotically active tissues can be passed on to progeny cells, meaning that somatic stem cell epigenetic modifications that occur at early developmental
Reprogramming-recalcitrant genes stages may be transmitted large, tissue-specific cellular populations or lineages over the life of the organism. Importantly, proliferating cells, including stem and progenitor cells, have a higher tendency to undergo epigenetic modification (Meissner et al., 2008). During DNA replication, the disruption, transfer, and de novo assembly of nucleosomes are coordinated to reproduce the epigenetic landscape (Groth et al., 2007). Because these genome-wide rearrangements of chromosome structure occur during DNA replication, the mitotic S phase may provide a unique opportunity for epigenetic modification to occur. As an example, culture-induced instability of DNA methylation in ES and somatic cells has been observed in high-CpG promoters (Meissner et al., 2008; Allegrucci et al., 2007). The DNA in somatic cells in mitotically active tissues also tends to be more highly methylated, which has been suggested as a possible cause of replicative aging (Chu et al., 2007). High-turnover tissues also show higher rates of cancer development (Rando, 2006; Sakurada et al., 2008). The cell mosaicism of A-DMRs described above supports the notion that stem cells can serve as generators and transmitters of environmental epigenetic modifications.
The nature of reprogramming-recalcitrant genes Given the evident importance of environmental epigenetic changes in ontogeny, physiology, and pathogenesis, it seems clear that the extent of reprogramming should be assessed with reference not only to developmental epigenetics, but environmental epigenetics as well. A large body of evidence on the inheritance of environmental epigenetics following physiological reprogramming has been compiled, and indicates that transgenerational epigenetic effects define phenotypes present in successive generations that are not genetically determined (Yongson and Whitelaw, 2008). Specifically, environmental epigenetic modifications in germline cells may play a major role in transgenerational phenotypes. A number of examples of adaptive and non-adaptive transgenerational effects have been found in mammals, including humans (Yongson and Whitelaw, 2008). Not all transgenerational epigenetic phenotypes can be explained by the inheritance of epigenetic marks that avoid zygotic and germline reprogramming. It has been shown, however, that a patient suffering hereditary nonpolyposis colorectal cancer (HNPCC) had abnormal DNA methylation of the DNA mismatch repair genes, MLH1, in all three germ layers inherited from parental germ cells that escaped zygotic reprogramming (Hitchins et al., 2007). The incomplete reprogramming observed in SCNT and iPSCs could similarly correspond to zygotic reprogramming. As suggested by the heterogeneity of A-DMRs (Eckhardt et al., 2006), environmental epigenetics may mirror common differences between germline and somatic cells as determined by differences in niche microenvironments. If so, this raises the possibility that the artificial reprogramming of somatic cells may generate disease phenotypes, which are usually not caused by the reprogramming recalcitrant genes inherited through physiological reprogramming of the germline cell nucleus.
161
Challenges in the application of human iPSCs Much has been written about potential applications for human iPSC technology in the study of disease etiology and patient-specific cell replacement therapy. The existence of reprogramming-recalcitrant genes, however, may signify important constraints on such applications. We have seen how epigenetic modifications can influence disease phenotypes by affecting gene expression levels, irrespective of DNA sequence. The causes of the phenotype differences observed in somatic cells differentiated from patientspecific iPSCs may be complex, as these may be the result of variations in DNA sequence or the activity of iPSC-specific reprogramming-recalcitrant genes. Environmental epigenetic marks inherited from somatic cell sources also exhibit cell mosaicism. This suggests that the only way to reconstitute the disease-specific phenotypic features resulting from inherited environmental epigenetic modifications in patient-specific human iPSCs may be to induce pluripotency in somatic cells from the affected tissue. Neural stem cells are currently being considered as potential therapies for neurological disorders. It was recently reported that a boy suffering ataxia telangiectasia who had received multiple injections of fetal neural cells into the brain developed a multifocal brain tumor derived from the donor cells (Amariglio et al., 2009). Importantly, the fetal neural cells were cultured before transplantation (Amariglio et al., 2009). Culture-induced DNA methylation instability has been observed in the epigenome of human ES cells as well as in that of somatic cells (Meissner et al., 2008; Allegrucci et al., 2007). It is conceivable, therefore, in this unfortunate case that the cancer-prone epigenetic phenotype was acquired during culture. The possibility that de novo epigenetic modification during the iPSC-generating process may also induce epigenetic phenotypic changes cannot be ruled out. The role of reprogramming-recalcitrant genes in human iPSCs needs to be considered seriously. Epigenetics is a latent change in gene function, which cannot be detected comprehensively by gene expression profile. At present, there is no technology able to accurately measure genomewide epigenetic modifications at the single-cell level, and thus it remains impossible to prospectively identify potentially small populations of cancer-prone cells in a cell therapy product. The use of sample-based statistics can help to address this epigenetic heterogeneity, and should be applied in the evaluation of cells intended for use in the study of disease etiology or cell-based therapy. Ideally, a larger number of samples than is typically used in clinical studies should be used to improve accuracy and statistical power.
Optimization of human iPSC generation The quality of human iPSCs should be evaluated in respect to their retention of environmental as well as developmental epigenetic marks. The definition of human iPSCs remains ambiguous, even in terms of true pluripotency. For example, most publications to date have examined the expression only of Nanog or ALP as a surrogate in determining increases in the efficiency of iPSC generation (Hong et al., 2009; Yoshida
162 et al., 2009). However, the presence of partially reprogrammed developmental genes in Nanog-and/or ALP-positive human colonies indicates that the increase in the efficiency of pluripotent cell generation through the use of newly developed reprogramming methods cannot be confirmed without evaluating the expression of multiple pluripotency genes, including Nanog, TDGF1, Dnmt3b, Zfp42, FoxD3, GDF3, Cyp26A1 and TERT, in all colonies generated in the primary induction culture (Masaki et al., 2008) (Fig. 3A). Absent such confirmatory evidence, increases in the number of Nanog-and/or ALP-positive colonies may only indicate an increase in the number of partially reprogrammed cells. To my knowledge, no evaluation of environmental epigenetic modifications, such as those caused by disease states in somatic cells of origin or cell culture, have been conducted on human iPSCs. It should be emphasized that functional analyses using the present criteria for pluripotency are insufficient to determine the quality of human iPSCs in terms of their environmental epigenetics. The differentiative potential of iPSCs does not necessarily reflect the full set of disease-associated epigenetic modifications that may be inherited from their somatic cells of origin. Without clear evidence of bona fide complete reprogramming, in which all unnecessary epigenetic marks in somatic cells is erased, it is essential to optimize the process of
K. Sakurada human iPSC generation by identifying epigenetically ideal somatic cell sources (Fig. 3B). Although the generation of pluripotent stem cells from lymphocytes, monocytes/macrophages, and albumin-or insulin-producing cells has been proposed, based on the assumption that all somatic cells, including terminally differentiated cells, have the ability to be reprogrammed to pluripotency, this may be a leap too far (Yamanaka, 2009). Studies of DNA methylation profiles in sperm and lymphocytes revealed relatively smaller differences, consistent with the similarities between their gene expression profiles (Eckhardt et al., 2006; Zeng et al., 2004). Monocytes and macrophages express multiple lineage genes by differentiation or by cell fusion (Zhao et al., 2003; Camargo et al., 2003). These characteristics are not observed in other terminally-differentiated cells. Additionally, the expression of tissue-specific genes, such as those encoding albumin or insulin, cannot serve as the basis for the reprogramming of terminally-differentiated cells, since it has been shown that these genes can also be expressed in somatic stem cells during wound healing and cell culture (Rountree et al., 2007). It has been shown that Bmi-1 modification is essential in self-renewing hematopoietic and neural stem cells in an Ink4a-dependent manner (Park et al., 2003; Molofsky et al., 2003). Thus, the increase in the efficiency of colony formation by inhibition of the Ink4/arf pathway means that somatic stem cells are favorable sources
Figure 3 Evaluation and optimization of degree of epigenetic reprogramming. Increase in the number of Nanog and/or ALP-positive colonies may reflect the generation of partially reprogrammed cells. The degree of differentiation in somatic cell sources will contribute to the degree of developmental epigenetic reprogramming. On the other hand, the epigenetic background of a donor, and culture-induced epigenetics will contribute to the degree of environmental epigenetic reprogramming.
Reprogramming-recalcitrant genes for iPSC reprogramming (Li et al., 2009). And indeed, the reprogramming of somatic stem cells has been shown to be more efficient than that of skin-derived cells (Silva et al., 2008; Eminli et al., 2009). Although human iPSCs have been established from elderly donors, their quality was not analyzed with respect to environmental epigenetics (Dimos et al., 2008). Global hypomethylation and local hypermethylation are commonly observed in cancer and senescent cells (Suzuki et al., 2006), and age-dependent hypermethylation is observed at A-DMRs, indicating that environmental epigenetics may increase with cell senescence. These observations indicate that young donors may be preferable to elderly donors as sources for somatic cells used in the generation of human iPSCs. Mitotically inactive tissue is also known to undergo less DNA methylation at A-DMRs (Chu et al., 2007). Given with these observations regarding DNA methylation and the instabilities associated with proliferation, it stands to reason that minimally passaged somatic stem cells derived from low-turnover tissues may represent optimal starting cell sources for use in the generation of human iPSCs.
Conclusions The available evidence indicates that reprogramming is incomplete in human iPSCs, and that reprogrammingrecalcitrant genes may influence both human iPSC function and the phenotypes of their differentiated progeny. It is thus necessary to determine the epigenetic state of “wildtype” pluripotent stem cells in regard to both their developmental and environmental epigenetics. Understanding the mechanisms of environmental epigenetics and transgenerational epigenetic inheritance will offer new methods for manipulating epigenetic marks. The use of chemical compounds that modify the function of epigenetic enzymes, such as histone deacetylase (HDAC), histone methylase (HMT), and DNA methyltransferase (Dnmt), may help to improve reprogramming efficiencies, and in fact, it has been shown that HDAC and Dnmt1 inhibitors do just that (Huangfu et al., 2008). However, these compounds affect genome stability by targeting various cellular genomic surveillance mechanisms (EotHoullier et al., 2009). Evidence that epigenetics processes are regulated by RNA signaling is also beginning to accumulate (Mattick et al., 2009). Compared to epigenetic enzymes and chromatin-modifying complexes (Polycombgroup and Trithorax-group), RNA-directed regulation of chromosome modification may offer the advantage of a high degree of sequence-and locus-specificity. Non-protein coding RNAs (ncRNAs) play a major role in this process. A better understanding of the mechanisms of ncRNA-mediated epigenetic modification may provide us with new tools for controlling epigenetic modifications associated with disease susceptibility and pathogenesis.
Acknowledgments The discussions described in this paper are the outcome of research activities at Sony Computer Science Laboratories. The author is grateful to Dr. Douglas Sipp for his helpful critique of this manuscript.
163
References Allegrucci, C., et al., 2007. Restriction landmark genome scanning identifies culture-induced DNA methylation instability in the human embryonic stem cell epigenome. Hum. Mol. Genet. 16, 1253–1268. Amariglio, N., et al., 2009. Donor-derived brain tumor following neural stem cell transplantation in an ataxia telangiectasia patient. PLoS Med. e1000029, 6. Boiani, M., et al., 2002. Oct4 distribution and level in mouse clones: consequences for pluripotency. Genes Dev. 16, 1209–1219. Bortvin, A., et al., 2003. Incomplete reactivation of Oct4-related genes in mouse embryos cloned from somatic nuclei. Development 130, 1673–1680. Bromer, J.G., Wu, J., Zhou, Y., Taylor, H.S., 2009. Hypermethylation of homeobox A10 by in utero diethylstilbestrol exposure: an epigenetic mechanism for altered developmental programming. Endocrinology 150, 3376–3382. Butler, D., 2008. Crossing the valley of death. Nature 453, 840–842. Camargo, F.D., Green, R., Capetanaki, Y., Jackson, K.A., Goodell, M.A., 2003. Single hematopoietic stem cells generate skeletal muscle through myeloid intermediates. Nat. Med. 9, 1520–1527. Chin, M.H., et al., 2009. Induced pluripotent stem cells and embryonic stem cells are distinguished by gene expression signatures. Cell Stem Cell 5, 111–123. Chu, M.W., et al., 2007. Lack of increases in methylation at three CpG rich genomic loci in non-mitotic adult tissues during aging. BMC Med. Genet. 8, 50. Deng, J., et al., 2009. Targeted bisulfate sequencing reveals changes in DNA methylation associated with nuclear reprogramming. Nat. Biotechnol. 27, 353–360. Dimos, J.T., et al., 2008. Induced pluripotent stem cells generated from patients with ALS can be differentiated into motor neurons. Science 321, 1218–1221. Ebert, A.D., et al., 2009. Induced pluripotent stem cells from a spinal muscular atrophy patient. Nature 457, 277–280. Eckhardt, F., et al., 2006. DNA methylation profiling of human chromosomes 6, 20 and 22. Nat. Genet. 12, 1378–1385. Eminli, S., et al., 2009. Differentiation stage determines potential of hematopoietic cells for reprogramming into induced pluripotent stem cells. Nat. Genet. 41, 968–976. Eot-Houllier, G., et al., 2009. Histone deacetylase inhibitors and genomic instability. Cancer Lett. 274, 169–176. Fraga, M.F., et al., 2005. Epigenetic differences arise during the lifetime of monozygotic twins. Proc. Natl. Acad. Sci. U.S.A. 102, 10604–10609. Gluckman, P.D., Hanson, M.A., 2004. Living with the past: evolution, development, and patterns of disease. Science 305, 1733–1736. Goldberg, A.D., Allis, C.D., Bernstein, E., 2007. Epigenetics: A landscape takes shape. Cell 128, 635–638. Groth, A., Rocha, W., Verreault, A., Almouzni, G., 2007. Chromatin challenges during DNA replication and repair. Cell 128, 721–733. Gurdon, J.B., Melton, D.A., 2008. Nuclear reprogramming in cells. Science 322, 1811–1815. Hellwig, J., et al., 2000. Pre-and postonatal oral toxicity of vinclozolin in Wistar and Long-Evans rats. Regul. Toxicol. Pharmacol. 32, 42–50. Hitchins, M.P., et al., 2007. Inheritance of a cancer-associated MLH1 germ-line epimutation. N. Engl. J. Med. 356, 697–705. Holliday, R., Pugh, J.E., 1975. DNA modification mechanisms and gene activity during development. Science 187, 226–232. Hong, H., et al., 2009. Suppression of induced pluripotent stem cell generation by the p53–p21 pathway. Nature 460, 1132–1135. Huangfu, D., et al., 2008. Induction of pluripotent stem cells by defined factors is greatly improved by small-molecule compounds. Nat. Biotechnol. 26, 795–797. Huxley, R., et al., 2007. Is birth weight a risk factor for ischemic heart disease in later life? Am. J. Clin. Nutr. 85, 1244–1250.
164 Jirtle, R.L., Skinner, M.L., 2007. Environmental epigenomics and disease susceptibility. Nat. Rev., Genet. 8, 253–262. Kishigami, S., Wakayama, T., 2009. Somatic cell nuclear transfer in the mouse. Methods Mol. Biol. 518, 207–218. Klip, H., Verloop, J., van Gool, J.D., Koster, M.E., Burger, C.W., van Leeuwen, F.E., 2002. Hypospadias in sons of women exposed to diethylstilbestrol in utero: a cohort study. Lancet 359, 1102–1107. Kohda, T., et al., 2005. Variation in gene expression and aberrantly regulated chromosome regions in cloned mice. Biol. Reprod. 73, 1302–1311. Li, H., et al., 2009. The Ink4/Arf locus is a barrier for iPS cell reprogramming. Nature 460, 1149–1153. Lister, R., et al., 2009. Human DNA methylomes at base resolution show widespread epigenomic differences. Nature 462, 315–322. Lowry, W.E., et al., 2008. Generation of human induced pluripotent stem cells from dermal fibroblasts. Proc. Natl. Acad. Sci. U.S.A. 105, 2883–2888. Masaki, H., et al., 2008. Heterogeneity of pluripotent marker gene expression in colonies generated in human iPS cell induction culture. Stem Cell Res. 1, 105–115. Mattick, J.S., et al., 2009. RNA regulation of epigenetic processes. BioEssays 31, 51–59. Meissner, A., et al., 2008. Genome-scale DNA methylation maps of pluripotent and differentiated cells. Nature 454, 766–770. Molofsky, A.V., et al., 2003. Bmi-1 dependence distinguishes neural stem cell self-renewal from progenitor proliferation. Nature 425, 962–967. Ng, R.K., Gurdon, J.B., 2005. Epigenetic memory of active gene transcription is inherited through somatic cell nuclear transfer. Proc. Natl. Acad. Sci. U.S.A. 102, 1957–1962. Ogonuki, N., et al., 2002. Early death of mice cloned from somatic cells. Nat. Genet. 30, 253–254. Park, I.H., et al., 2008. Disease-specific induced pluripotent stem cells. Cell 134, 877–886. Park, I.H., et al., 2008. Reprogramming of human somatic cells to pluripotency with defined factors. Nature 451, 141–146. Park, I.K., et al., 2003. Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells. Nature 423, 302–305. Petronis, A., 2006. Epigenetics and twins: three variations on the theme. Trends Genet. 22, 347–350. Rando, T.A., 2006. Stem cells, ageing and the quest for immortality. Nature 441, 1080–1086. Raya, A., et al., 2009. Disease-corrected haematopoietic progenitors from Fanconi anaemia induced pluripotent stem cells. Nature 460, 53–59. Reik, W., 2007. Stability and flexibility of epigenetic gene regulation in mammalian development. Nature 447, 425–432. Rideout, W.M., Eggan, K., Jaenisch, R., 2001. Nuclear cloning and epigenetic reprogramming of the genome. Science 293, 1093–1098. Riggs, A.D., 1975. X inactivation, differentiation, and DNA methylation. Cytogenet. Cell Genet. 14, 9–25.
K. Sakurada Rountree, C.B., et al., 2007. A CD133-expressing murine liver oval cell population with bilineage potential. Stem Cells 25, 2419–2429. Sadikovic, B., Andrews, J., Carter, D., Robinson, J., Rodenhiser, D.I., 2008. Genome-wide H3K9 histone acetylation profiles are altered in benzopyrene-treated MCF7 breast cancer cells. J. Biol. Chem. 283, 4051–4060. Sakurada, K., McDonald, F.M., Shimada, F., 2008. Regenerative Medicine and stem cell based drug discovery. Angew. Chem. Int. Ed. 47, 5718–5738. Silva, J., et al., 2008. Promotion of reprogramming to ground state pluripotency by signal inhibition. PLoS Biol. 6, e253. Soldner, F., et al., 2009. Parkinson's disease patient-derived induced pluripotent stem cells free of viral reprogramming factors. Cell 136, 964–977. Suzuki, K., et al., 2006. Global DNA demethylation in gastrointestinal cancer is age dependent and precedes genomic damage. Cancer Cell 9, 199–207. Takahashi, K., et al., 2007. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131, 861–872. Tamashiro, K.L., et al., 2002. Cloned mice have an obese phenotype not transmitted to their offspring. Nat. Med. 8, 262–267. Tanaka, S., et al., 2001. Placentomegaly in cloned mouse concepti caused by expansion of the spongiotrophoblast layer. Biol. Reprod. 65, 1813–1821. Tsankova, N.M., et al., 2006. Sustained hippocampal chromatin regulation in a mouse model of depression and antidepressant action. Nat. Neruosci. 9, 519–525. Waddington, C.H., 1942. The epigenotype. Endeavour 1, 18–20. Waterland, R.A., Michels, K.B., 2007. Epigenetic epidemiology of the developmental origins hypothesis. Annu. Rev. Nutr. 27, 363–388. Weaver, I.C., et al., 2004. Epigenetic programming by maternal behavior. Nat. Neurosci. 7, 847–854. Wong, A.H.C., Gottesman, I.I., Petronis, A., 2005. Phenotypic differences in genetically identical organisms: the epigenetic perspective. Hum. Mol. Genet. 14, R11–18. Yamanaka, S., 2009. Elite and stochastic models for induced pluripotent stem cell generation. Nature 460, 49–52. Yongson, N.A., Whitelaw, E., 2008. Transgenerational epigenetic effects. Annu. Rev. Genomics Hum. Genet. 9, 233–257. Yoshida, Y., Takahashi, K., Okita, K., Ichisaka, T., Yamanaka, S., 2009. Hypoxia enhances the generation of induced pluripotent stem cells. Cell Stem Cell 5, 237–241. Yu, J., et al., 2007. Induced pluripotent stem cell lines derived from human somatic cells. Science 318, 1917–1920. Zeng, W., et al., 2004. Transcript profile of CD4+and CD8+T cells from the bone marrow of acquired aplastic anemia patients. Exp. Hematol. 32, 806–814. Zhang, C.L., et al., 2002. Class II histone deacetylases act as signalresponsive repressors of cardiac hypertrophy. Cell 110, 479–488. Zhao, Y., Glesne, D., Huberman, E.A., 2003. human peripheral blood monocyte-derived subset acts as pluripotent stem cells. Proc. Natl. Acad. Sci. U.S.A. 100, 2426–2431.
available at www.sciencedirect.com
www.elsevier.com/locate/scr
REVIEW
Environmental epigenetic modifications and reprogramming-recalcitrant genes Kazuhiro Sakurada ⁎ Sony Computer Science Laboratories Inc., Takanawa Muse Bldg, 3-14-13, Higashigotanda, Shinagawa-ku, Tokyo, 141-0022 Japan Received 13 October 2009; received in revised form 15 January 2010; accepted 15 January 2010 Abstract The term “environmental epigenetic modifications” refers to alterations in phenotype triggered by environmental stimuli via epigenetic mechanisms. Epidemiologic and animal model studies show that a subset of such environmental epigenetic marks may affect susceptibility to chronic diseases. A growing body of evidence regarding incompleteness of reprogramming indicates that the potential retention of pathogenic environmental epigenetics in human induced pluripotent stem cells (iPSCs) should be seriously considered. Given this possibility, the optimization of methods for the generation of human induc pluripotent stem cells may require the identification of epigenetically appropriate somatic cell sources. Similarly, techniques for controlling epigenetic modification by environmental factors may also play a critical role in the development of epigenetically stable sources of pluripotent stem cells. © 2010 Elsevier B.V. All rights reserved.
Contents Environmental epigenetic effects . . . . . . . . . . . . . . . . . . . . . Stem cells as generators and transmitters of environmental epigenetic The nature of reprogramming-recalcitrant genes . . . . . . . . . . . . . Challenges in the application of human iPSCs. . . . . . . . . . . . . . . Optimization of human iPSC generation . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Although basic biomedical research has yielded an enormous amount of information on disease mechanisms, the recent decline in the development of new drugs and treatments suggests that our fundamental understanding of pathogenesis remains insufficient (Butler, 2008). The major characteristics ⁎ Fax: +81 3 5448 4273. E-mail address: [email protected].
. . . . . . . . . modifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
159 160 161 161 161 163 163 163
of the outputs of scientific investigations and their applications in technology include predictability, reproducibility, and controllability, and, importantly, findings are expected to be generalizable in the sense that when two systems are uniform and constant, the causes of events in one system can serve as the basis for predictions in its counterpart. The human body, however, is exceedingly complex and diverse, which frustrates the direct use of predictions from models in
1873-5061/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.scr.2010.01.001
158 medical care. The genetic sequences of any two individuals will show several millions of single nucleotide substitutions, and the human phenotype also changes as a result of epigenetic mechanisms. It is thus understandable that predicting and controlling such a complex, diverse, dynamic, and irreversible organism as the human body represents a tremendous scientific challenge. In clinical studies, population-based statistics are used to make inferences about heterogeneous patient groups, and the uniformity of components of therapeutic entities in the face of this interindividual diversity remains a major challenge in cell-based medicine. The advent and advances in the study of human induced pluripotent stem cells (iPSCs) have raised considerable public interest, particularly with respect to their potential applications in stem-cell-based therapy and in vitro cellular models of disease (Takahashi et al., 2007; Yu et al., 2007; Park et al., 2008; Masaki et al., 2008; Lowry et al., 2008; Park et al., 2008; Dimos et al., 2008; Ebert et al., 2009; Soldner et al., 2009; Raya et al., 2009). This enthusiasm is partially rooted in the assumption that human iPSCs are equivalent to human embryonic stem (ES) cells. This assumption, however, remains to be substantiated. In mammalian embryonic development, chromosomal DNA undergoes two different physiological reprogramming events: a first round following soon after fertilization, and a second in primordial germ cells (Reik, 2007). Under physiological conditions, only germline-derived chromosomes are subject to these reprogramming processes (Fig. 1). Technologies that enable artificial reprogramming, such as somatic cell nuclear transfer (SCNT), ES cell-somatic cell fusion, and the induction of pluripotency in somatic cells, are thought to mimic the physiological reprogramming that
K. Sakurada takes places after fertilization, as such methods were developed to establish a post-fertilization environment for somatic chromosomes (Fig. 1) (Gurdon and Melton, 2008). Although SCNT has shown that the genome state can be reprogrammed to an early embryonic (zygotic) pattern, the use of somatic cell nuclei from adult animals is inefficient, and fewer than 6% of cloned mice embryos develop into births (Rideout et al., 2001; Kishigami and Wakayama, 2009). The process is also error-prone, and cloned mammals of different species exhibit a variety of abnormal phenotypes, including placental hyperplasia, large fetus syndrome, immune disorders, and shortened life span (Tanaka et al., 2001; Tamashiro et al., 2002; Ogonuki et al., 2002). Tissue-specific gene expression profiles in cloned mice are also extremely different from those generated by natural reproduction (Kohda et al., 2005). Interestingly, most of the abnormal phenotypes in cloned mice disappear after a single generation, suggesting epigenetic alterations as a possible cause of the abnormalities in gene expression profile and phenotype (Tamashiro et al., 2002). Lending credence to this notion is evidence that the epigenetic memory of active and repressed transcription in nuclear-donor somatic cells can be inherited through SCNT, in both mouse and Xenopus laevis (Ng and Gurdon, 2005; Boiani et al., 2002; Bortvin et al., 2003). Human iPSC colonies expressing Nanog and/or ALP (alkaline phosphatase) following the forced expression of Oct3/4, Sox-2, c-Myc and Klf4 show significant heterogeneity characterized by a number of hallmark features (Fig. 2) (Masaki et al., 2008). Such heterogeneity indicates the different degrees of reprogramming in Nanog-and ALPpositive colonies. Partially reprogrammed cells have also been observed in human iPSCs induced by the forced
Figure 1 Physiological and artificial reprogramming. In physiological conditions, chromosomal DNA of germ cells are the sole source for reprogramming. Somatic cells receive epigenetic modification during developmental and postnatal life. In physiological conditions, chromosomal DNA of somatic cells and these epigenetic modifications are not inherited by the next generation, which is called the Weismann Barrier. Environmental epigenetic modifications on chromosomes in germline cells and somatic cells must be different.
Reprogramming-recalcitrant genes
159
Figure 2 Heterogeneity of ALP-positive colonies induced by Oct3/4, Sox2, c-Myc and Klf4 genes from human dermis-derived cells. (A) Heterogeneity of ALP-positive colonies was identified by analyzing the gene expression profile of eight marker genes. Although Nanog gene expression was observed in 161 of 163 ALP-positive colonies, only four colonies expressed eight human ES cell marker genes. The gene expression profile of the remaining 159 colonies was significantly heterogeneous; however, it had a characteristic hierarchical feature. (B) The eight genes are aligned in the order of tendency to be expressed in ALP-positive colonies, Nanog N TDGF1 N Dnmt3b N Zfp42 N FoxD3 N GDF3 N CYP26A1 N TERT. Eight genes were not randomly expressed by their tendency but showed a hierarchical feature that expresses lower-tendency genes in colonies where higher-tendency genes are expressed. (C) Relation between the transcriptional factor binding sites and the tendency of marker gene expression. Data was obtained from reference (Masaki et al., 2008).
expression of a different set of genes (Yu et al., 2007). Recently, Chin et al. identified and described several important features of a number of differentially expressed genes in human ESCs and iPSCs (Chin et al., 2009). I will refer to these as “reprogramming-recalcitrant” genes, as they resist the induction of a transcriptional state identical to that seen in embryonic stem cells. The underlying causes can be categorized broadly as 1) insufficient induction of human ES cell-specific genes; 2) insufficient suppression of somatic cell-specific genes; and 3) induction of human iPSC-specific genes. Of these, 1 and 2 correspond roughly to the causes of recalcitrance to reprogramming by SCNT (Ng and Gurdon, 2005). It is important to recognize that genes that are not expressed in pluripotent cells cannot be detected by gene expression profiling in self-renewing human iPSCs, meaning that there may be more reprogramming-recalcitrant genes in human iPSCs than have been reported to date. The measurement of genome-wide epigenetic marks is one means of addressing this limitation. It was recently reported
that, in human iPSCs, the DNA methylation landscapes of chromosomes 12 and 20 are different from those of the same chromosomes in human ES cells (Deng et al., 2009). This raises the question of whether the incomplete resetting of the nuclear state is a hallmark of artificial or postfertilization reprogramming or, in other words, whether it is possible to achieve complete reprogramming by mimicking post-fertilization reprogramming. The mechanistic basis and functional roles of reprogramming-recalcitrant genes in the resetting of somatic nuclei is also of great interest for further study.
Environmental epigenetic effects Epigenetics refers to heritable modifications in gene function without alteration of DNA sequences (Goldberg et al., 2007). Chemical modifications to DNA and chromatin proteins are the underlying drivers of epigenetic modifications. Although
160 human DNA methylation occurs exclusively at cytosine residues in CpG dinucleotides in differentiated cells, nonCpG methylation, which is primarily observed at CpA nucleotides, has recently been detected in stem cells (Lister et al., 2009). Chromatin itself is composed of DNA and histone proteins, which are subject to more than 100 different types of modification, including methylation, acetylation, phosphorylation, and ubiquitination (Goldberg et al., 2007). The term “epigenetics” was coined by Waddington to describe the mechanisms necessary for the unfolding of genetic information in ontogenic development (Waddington, 1942). Riggs (Riggs, 1975) and Holiday and Pugh (Holliday and Pugh, 1975) proposed that the programmed methylation and demethylation of DNA might be the molecular mechanism behind Waddington's hypothesis, and it has since been shown that Xchromosome-inactivation, genome imprinting, and lineagespecific gene silencing all rely on DNA methylation. Epigenetics do not function solely in the establishment and maintenance of tissue-and cell-type-specific gene expression. In addition to tissue-specific DNA methylation regions (T-DMRs), alternative DNA methylation regions (ADMRs) have been identified by analyses of the DNA methylation profiles of human chromosomes 6, 20, and 22 (Eckhardt et al., 2006). In this study, DNA methylation of TDMRs was found to occur preferentially in evolutionarily conserved non-protein coding regions (ECRs) several kbp distant from core promoters, while A-DMRs, which exhibit cell mosaicism, were observed within the promoter. DNA methylation profiles of A-DMRs vary between different tissues in the same donor, as well as in the same tissues or organs in different individuals, indicating that such regions are acquired throughout life and depend on the specific environmental stimuli at the cellular level. Monozygotic (MZ) twins are highly similar in appearance, especially at birth, yet they are frequently divergent for several important phenotypes, including the onset of common diseases (Wong et al., 2005). The concordance of susceptibility to age-dependent diseases in MZ twins can be quite low: 15% for breast cancer, 25%-30% for multiple sclerosis, 25%-45% for diabetes, 50% for schizophrenia, and 40-70% for Alzheimer's disease (Wong et al., 2005; Petronis, 2006). Phenotypic divergence increases with age, and greater differences are seen between MZ twins raised and living in separate environments. Several different mechanisms contribute to these changes, one of which is epigenetic differences. To address this issue, Fraga et al. analyzed global and locus-specific DNA methylation and histone acetylation of a large cohort of MZ twins and showed that, although such twins are epigenetically indistinguishable in the early years of life, older MZ twins exhibit remarkable differences in the overall content and distribution of 5-methylcytosine and histone acetylation (Fraga et al., 2005). Findings from human epidemiological and animal studies indicate that environmental stimuli can induce permanent changes in phenotype, including metabolism and susceptibility to chronic disease, at critical periods during pre-and postnatal development, (Jirtle and Skinner, 2007). It has also been suggested that epigenetic mechanisms are likely involved in the developmental origins of health and disease (Waterland and Michels, 2007). Although this remains the subject of active discussion, environmental epigenetics
K. Sakurada may reflect an adaptive response to a given prenatal environment that serves to optimize the postnatal phenotype to meeting the challenges of that environment (Gluckman and Hanson, 2004; Huxley et al., 2007). It has been suggested, for example, mismatches between pre-and postnatal environments may be linked to increases in the risk of developing chronic diseases, including type II diabetes, hypertension, and coronary heart disease (Gluckman and Hanson, 2004). Although the molecular mechanisms by which environmental stimuli alter gene expression to achieve phenotypic diversity are largely unknown, changes in the concentration of fetal hormones, such as glucocorticoids, may serve as molecular cues. Glucocorticoids have been shown to alter DNA methylation at the promoter of the glucocorticoid receptor. Maternal behaviors in rat, for example, have been shown to influence the epigenetic modification of the promoter of the glucocorticoid receptor in neurons (Weaver et al., 2004). In this study, the glucocorticoid receptor gene promoter was epigenetically repressed in the hippocampus of the offspring of low “licking and arched-back nursing” mothers (Weaver et al., 2004). Another study has shown that chronic social defeat stress induces methylation of lysine 27 (K27) of the histone H3 adjacent to the promoter region of the brain-derived neurotrophic factor (BDNF) gene in hippocampal neurons (Tsankova et al., 2006). Developing organisms are also extremely sensitive to chemical perturbation. Environmental compounds, drug administration, and physiological stresses in embryogenesis and the early postnatal period have been demonstrated to influence the pathogenesis and course of adult onset diseases (Hellwig et al., 2000). For example, endocrine disruptors, such as vinclozolin and diethylstilbestrol, induce reproductive and endocrine defects (Hellwig et al., 2000; Klip et al., 2002). The frequency and reproducibility of such abnormal phenotypes strongly suggest that epigenetic modifications, not genetic mutations, are causative, and indeed, aberrant DNA methylation of several genes has been found in tissues and cells exposed to diethylstilbestrol (Bromer et al., 2009). Cultured cells exposed to benzopyrene additionally exhibited genome-wide alterations in histone H3 lys9 acetylation (Sadikovic et al., 2008). Clearly, exposure to different environmental factors, including nutritional, endocrine, and chemical perturbation throughout life, influences the epigenetic landscape of each individual. These epigenetic modifications can be distinguished with developmental epigenetics and is called environmental epigenetics.
Stem cells as generators and transmitters of environmental epigenetic modifications Environmental epigenetic modifications introduced into terminally differentiated cells of post-mitotic tissues or organs, such as neurons or cardiomyocytes, can play indispensable functional roles (Weaver et al., 2004; Tsankova et al., 2006; Zhang et al., 2002), but are not transmitted. In contrast, epigenetic modifications in somatic stem cells in mitotically active tissues can be passed on to progeny cells, meaning that somatic stem cell epigenetic modifications that occur at early developmental
Reprogramming-recalcitrant genes stages may be transmitted large, tissue-specific cellular populations or lineages over the life of the organism. Importantly, proliferating cells, including stem and progenitor cells, have a higher tendency to undergo epigenetic modification (Meissner et al., 2008). During DNA replication, the disruption, transfer, and de novo assembly of nucleosomes are coordinated to reproduce the epigenetic landscape (Groth et al., 2007). Because these genome-wide rearrangements of chromosome structure occur during DNA replication, the mitotic S phase may provide a unique opportunity for epigenetic modification to occur. As an example, culture-induced instability of DNA methylation in ES and somatic cells has been observed in high-CpG promoters (Meissner et al., 2008; Allegrucci et al., 2007). The DNA in somatic cells in mitotically active tissues also tends to be more highly methylated, which has been suggested as a possible cause of replicative aging (Chu et al., 2007). High-turnover tissues also show higher rates of cancer development (Rando, 2006; Sakurada et al., 2008). The cell mosaicism of A-DMRs described above supports the notion that stem cells can serve as generators and transmitters of environmental epigenetic modifications.
The nature of reprogramming-recalcitrant genes Given the evident importance of environmental epigenetic changes in ontogeny, physiology, and pathogenesis, it seems clear that the extent of reprogramming should be assessed with reference not only to developmental epigenetics, but environmental epigenetics as well. A large body of evidence on the inheritance of environmental epigenetics following physiological reprogramming has been compiled, and indicates that transgenerational epigenetic effects define phenotypes present in successive generations that are not genetically determined (Yongson and Whitelaw, 2008). Specifically, environmental epigenetic modifications in germline cells may play a major role in transgenerational phenotypes. A number of examples of adaptive and non-adaptive transgenerational effects have been found in mammals, including humans (Yongson and Whitelaw, 2008). Not all transgenerational epigenetic phenotypes can be explained by the inheritance of epigenetic marks that avoid zygotic and germline reprogramming. It has been shown, however, that a patient suffering hereditary nonpolyposis colorectal cancer (HNPCC) had abnormal DNA methylation of the DNA mismatch repair genes, MLH1, in all three germ layers inherited from parental germ cells that escaped zygotic reprogramming (Hitchins et al., 2007). The incomplete reprogramming observed in SCNT and iPSCs could similarly correspond to zygotic reprogramming. As suggested by the heterogeneity of A-DMRs (Eckhardt et al., 2006), environmental epigenetics may mirror common differences between germline and somatic cells as determined by differences in niche microenvironments. If so, this raises the possibility that the artificial reprogramming of somatic cells may generate disease phenotypes, which are usually not caused by the reprogramming recalcitrant genes inherited through physiological reprogramming of the germline cell nucleus.
161
Challenges in the application of human iPSCs Much has been written about potential applications for human iPSC technology in the study of disease etiology and patient-specific cell replacement therapy. The existence of reprogramming-recalcitrant genes, however, may signify important constraints on such applications. We have seen how epigenetic modifications can influence disease phenotypes by affecting gene expression levels, irrespective of DNA sequence. The causes of the phenotype differences observed in somatic cells differentiated from patientspecific iPSCs may be complex, as these may be the result of variations in DNA sequence or the activity of iPSC-specific reprogramming-recalcitrant genes. Environmental epigenetic marks inherited from somatic cell sources also exhibit cell mosaicism. This suggests that the only way to reconstitute the disease-specific phenotypic features resulting from inherited environmental epigenetic modifications in patient-specific human iPSCs may be to induce pluripotency in somatic cells from the affected tissue. Neural stem cells are currently being considered as potential therapies for neurological disorders. It was recently reported that a boy suffering ataxia telangiectasia who had received multiple injections of fetal neural cells into the brain developed a multifocal brain tumor derived from the donor cells (Amariglio et al., 2009). Importantly, the fetal neural cells were cultured before transplantation (Amariglio et al., 2009). Culture-induced DNA methylation instability has been observed in the epigenome of human ES cells as well as in that of somatic cells (Meissner et al., 2008; Allegrucci et al., 2007). It is conceivable, therefore, in this unfortunate case that the cancer-prone epigenetic phenotype was acquired during culture. The possibility that de novo epigenetic modification during the iPSC-generating process may also induce epigenetic phenotypic changes cannot be ruled out. The role of reprogramming-recalcitrant genes in human iPSCs needs to be considered seriously. Epigenetics is a latent change in gene function, which cannot be detected comprehensively by gene expression profile. At present, there is no technology able to accurately measure genomewide epigenetic modifications at the single-cell level, and thus it remains impossible to prospectively identify potentially small populations of cancer-prone cells in a cell therapy product. The use of sample-based statistics can help to address this epigenetic heterogeneity, and should be applied in the evaluation of cells intended for use in the study of disease etiology or cell-based therapy. Ideally, a larger number of samples than is typically used in clinical studies should be used to improve accuracy and statistical power.
Optimization of human iPSC generation The quality of human iPSCs should be evaluated in respect to their retention of environmental as well as developmental epigenetic marks. The definition of human iPSCs remains ambiguous, even in terms of true pluripotency. For example, most publications to date have examined the expression only of Nanog or ALP as a surrogate in determining increases in the efficiency of iPSC generation (Hong et al., 2009; Yoshida
162 et al., 2009). However, the presence of partially reprogrammed developmental genes in Nanog-and/or ALP-positive human colonies indicates that the increase in the efficiency of pluripotent cell generation through the use of newly developed reprogramming methods cannot be confirmed without evaluating the expression of multiple pluripotency genes, including Nanog, TDGF1, Dnmt3b, Zfp42, FoxD3, GDF3, Cyp26A1 and TERT, in all colonies generated in the primary induction culture (Masaki et al., 2008) (Fig. 3A). Absent such confirmatory evidence, increases in the number of Nanog-and/or ALP-positive colonies may only indicate an increase in the number of partially reprogrammed cells. To my knowledge, no evaluation of environmental epigenetic modifications, such as those caused by disease states in somatic cells of origin or cell culture, have been conducted on human iPSCs. It should be emphasized that functional analyses using the present criteria for pluripotency are insufficient to determine the quality of human iPSCs in terms of their environmental epigenetics. The differentiative potential of iPSCs does not necessarily reflect the full set of disease-associated epigenetic modifications that may be inherited from their somatic cells of origin. Without clear evidence of bona fide complete reprogramming, in which all unnecessary epigenetic marks in somatic cells is erased, it is essential to optimize the process of
K. Sakurada human iPSC generation by identifying epigenetically ideal somatic cell sources (Fig. 3B). Although the generation of pluripotent stem cells from lymphocytes, monocytes/macrophages, and albumin-or insulin-producing cells has been proposed, based on the assumption that all somatic cells, including terminally differentiated cells, have the ability to be reprogrammed to pluripotency, this may be a leap too far (Yamanaka, 2009). Studies of DNA methylation profiles in sperm and lymphocytes revealed relatively smaller differences, consistent with the similarities between their gene expression profiles (Eckhardt et al., 2006; Zeng et al., 2004). Monocytes and macrophages express multiple lineage genes by differentiation or by cell fusion (Zhao et al., 2003; Camargo et al., 2003). These characteristics are not observed in other terminally-differentiated cells. Additionally, the expression of tissue-specific genes, such as those encoding albumin or insulin, cannot serve as the basis for the reprogramming of terminally-differentiated cells, since it has been shown that these genes can also be expressed in somatic stem cells during wound healing and cell culture (Rountree et al., 2007). It has been shown that Bmi-1 modification is essential in self-renewing hematopoietic and neural stem cells in an Ink4a-dependent manner (Park et al., 2003; Molofsky et al., 2003). Thus, the increase in the efficiency of colony formation by inhibition of the Ink4/arf pathway means that somatic stem cells are favorable sources
Figure 3 Evaluation and optimization of degree of epigenetic reprogramming. Increase in the number of Nanog and/or ALP-positive colonies may reflect the generation of partially reprogrammed cells. The degree of differentiation in somatic cell sources will contribute to the degree of developmental epigenetic reprogramming. On the other hand, the epigenetic background of a donor, and culture-induced epigenetics will contribute to the degree of environmental epigenetic reprogramming.
Reprogramming-recalcitrant genes for iPSC reprogramming (Li et al., 2009). And indeed, the reprogramming of somatic stem cells has been shown to be more efficient than that of skin-derived cells (Silva et al., 2008; Eminli et al., 2009). Although human iPSCs have been established from elderly donors, their quality was not analyzed with respect to environmental epigenetics (Dimos et al., 2008). Global hypomethylation and local hypermethylation are commonly observed in cancer and senescent cells (Suzuki et al., 2006), and age-dependent hypermethylation is observed at A-DMRs, indicating that environmental epigenetics may increase with cell senescence. These observations indicate that young donors may be preferable to elderly donors as sources for somatic cells used in the generation of human iPSCs. Mitotically inactive tissue is also known to undergo less DNA methylation at A-DMRs (Chu et al., 2007). Given with these observations regarding DNA methylation and the instabilities associated with proliferation, it stands to reason that minimally passaged somatic stem cells derived from low-turnover tissues may represent optimal starting cell sources for use in the generation of human iPSCs.
Conclusions The available evidence indicates that reprogramming is incomplete in human iPSCs, and that reprogrammingrecalcitrant genes may influence both human iPSC function and the phenotypes of their differentiated progeny. It is thus necessary to determine the epigenetic state of “wildtype” pluripotent stem cells in regard to both their developmental and environmental epigenetics. Understanding the mechanisms of environmental epigenetics and transgenerational epigenetic inheritance will offer new methods for manipulating epigenetic marks. The use of chemical compounds that modify the function of epigenetic enzymes, such as histone deacetylase (HDAC), histone methylase (HMT), and DNA methyltransferase (Dnmt), may help to improve reprogramming efficiencies, and in fact, it has been shown that HDAC and Dnmt1 inhibitors do just that (Huangfu et al., 2008). However, these compounds affect genome stability by targeting various cellular genomic surveillance mechanisms (EotHoullier et al., 2009). Evidence that epigenetics processes are regulated by RNA signaling is also beginning to accumulate (Mattick et al., 2009). Compared to epigenetic enzymes and chromatin-modifying complexes (Polycombgroup and Trithorax-group), RNA-directed regulation of chromosome modification may offer the advantage of a high degree of sequence-and locus-specificity. Non-protein coding RNAs (ncRNAs) play a major role in this process. A better understanding of the mechanisms of ncRNA-mediated epigenetic modification may provide us with new tools for controlling epigenetic modifications associated with disease susceptibility and pathogenesis.
Acknowledgments The discussions described in this paper are the outcome of research activities at Sony Computer Science Laboratories. The author is grateful to Dr. Douglas Sipp for his helpful critique of this manuscript.
163
References Allegrucci, C., et al., 2007. Restriction landmark genome scanning identifies culture-induced DNA methylation instability in the human embryonic stem cell epigenome. Hum. Mol. Genet. 16, 1253–1268. Amariglio, N., et al., 2009. Donor-derived brain tumor following neural stem cell transplantation in an ataxia telangiectasia patient. PLoS Med. e1000029, 6. Boiani, M., et al., 2002. Oct4 distribution and level in mouse clones: consequences for pluripotency. Genes Dev. 16, 1209–1219. Bortvin, A., et al., 2003. Incomplete reactivation of Oct4-related genes in mouse embryos cloned from somatic nuclei. Development 130, 1673–1680. Bromer, J.G., Wu, J., Zhou, Y., Taylor, H.S., 2009. Hypermethylation of homeobox A10 by in utero diethylstilbestrol exposure: an epigenetic mechanism for altered developmental programming. Endocrinology 150, 3376–3382. Butler, D., 2008. Crossing the valley of death. Nature 453, 840–842. Camargo, F.D., Green, R., Capetanaki, Y., Jackson, K.A., Goodell, M.A., 2003. Single hematopoietic stem cells generate skeletal muscle through myeloid intermediates. Nat. Med. 9, 1520–1527. Chin, M.H., et al., 2009. Induced pluripotent stem cells and embryonic stem cells are distinguished by gene expression signatures. Cell Stem Cell 5, 111–123. Chu, M.W., et al., 2007. Lack of increases in methylation at three CpG rich genomic loci in non-mitotic adult tissues during aging. BMC Med. Genet. 8, 50. Deng, J., et al., 2009. Targeted bisulfate sequencing reveals changes in DNA methylation associated with nuclear reprogramming. Nat. Biotechnol. 27, 353–360. Dimos, J.T., et al., 2008. Induced pluripotent stem cells generated from patients with ALS can be differentiated into motor neurons. Science 321, 1218–1221. Ebert, A.D., et al., 2009. Induced pluripotent stem cells from a spinal muscular atrophy patient. Nature 457, 277–280. Eckhardt, F., et al., 2006. DNA methylation profiling of human chromosomes 6, 20 and 22. Nat. Genet. 12, 1378–1385. Eminli, S., et al., 2009. Differentiation stage determines potential of hematopoietic cells for reprogramming into induced pluripotent stem cells. Nat. Genet. 41, 968–976. Eot-Houllier, G., et al., 2009. Histone deacetylase inhibitors and genomic instability. Cancer Lett. 274, 169–176. Fraga, M.F., et al., 2005. Epigenetic differences arise during the lifetime of monozygotic twins. Proc. Natl. Acad. Sci. U.S.A. 102, 10604–10609. Gluckman, P.D., Hanson, M.A., 2004. Living with the past: evolution, development, and patterns of disease. Science 305, 1733–1736. Goldberg, A.D., Allis, C.D., Bernstein, E., 2007. Epigenetics: A landscape takes shape. Cell 128, 635–638. Groth, A., Rocha, W., Verreault, A., Almouzni, G., 2007. Chromatin challenges during DNA replication and repair. Cell 128, 721–733. Gurdon, J.B., Melton, D.A., 2008. Nuclear reprogramming in cells. Science 322, 1811–1815. Hellwig, J., et al., 2000. Pre-and postonatal oral toxicity of vinclozolin in Wistar and Long-Evans rats. Regul. Toxicol. Pharmacol. 32, 42–50. Hitchins, M.P., et al., 2007. Inheritance of a cancer-associated MLH1 germ-line epimutation. N. Engl. J. Med. 356, 697–705. Holliday, R., Pugh, J.E., 1975. DNA modification mechanisms and gene activity during development. Science 187, 226–232. Hong, H., et al., 2009. Suppression of induced pluripotent stem cell generation by the p53–p21 pathway. Nature 460, 1132–1135. Huangfu, D., et al., 2008. Induction of pluripotent stem cells by defined factors is greatly improved by small-molecule compounds. Nat. Biotechnol. 26, 795–797. Huxley, R., et al., 2007. Is birth weight a risk factor for ischemic heart disease in later life? Am. J. Clin. Nutr. 85, 1244–1250.
164 Jirtle, R.L., Skinner, M.L., 2007. Environmental epigenomics and disease susceptibility. Nat. Rev., Genet. 8, 253–262. Kishigami, S., Wakayama, T., 2009. Somatic cell nuclear transfer in the mouse. Methods Mol. Biol. 518, 207–218. Klip, H., Verloop, J., van Gool, J.D., Koster, M.E., Burger, C.W., van Leeuwen, F.E., 2002. Hypospadias in sons of women exposed to diethylstilbestrol in utero: a cohort study. Lancet 359, 1102–1107. Kohda, T., et al., 2005. Variation in gene expression and aberrantly regulated chromosome regions in cloned mice. Biol. Reprod. 73, 1302–1311. Li, H., et al., 2009. The Ink4/Arf locus is a barrier for iPS cell reprogramming. Nature 460, 1149–1153. Lister, R., et al., 2009. Human DNA methylomes at base resolution show widespread epigenomic differences. Nature 462, 315–322. Lowry, W.E., et al., 2008. Generation of human induced pluripotent stem cells from dermal fibroblasts. Proc. Natl. Acad. Sci. U.S.A. 105, 2883–2888. Masaki, H., et al., 2008. Heterogeneity of pluripotent marker gene expression in colonies generated in human iPS cell induction culture. Stem Cell Res. 1, 105–115. Mattick, J.S., et al., 2009. RNA regulation of epigenetic processes. BioEssays 31, 51–59. Meissner, A., et al., 2008. Genome-scale DNA methylation maps of pluripotent and differentiated cells. Nature 454, 766–770. Molofsky, A.V., et al., 2003. Bmi-1 dependence distinguishes neural stem cell self-renewal from progenitor proliferation. Nature 425, 962–967. Ng, R.K., Gurdon, J.B., 2005. Epigenetic memory of active gene transcription is inherited through somatic cell nuclear transfer. Proc. Natl. Acad. Sci. U.S.A. 102, 1957–1962. Ogonuki, N., et al., 2002. Early death of mice cloned from somatic cells. Nat. Genet. 30, 253–254. Park, I.H., et al., 2008. Disease-specific induced pluripotent stem cells. Cell 134, 877–886. Park, I.H., et al., 2008. Reprogramming of human somatic cells to pluripotency with defined factors. Nature 451, 141–146. Park, I.K., et al., 2003. Bmi-1 is required for maintenance of adult self-renewing haematopoietic stem cells. Nature 423, 302–305. Petronis, A., 2006. Epigenetics and twins: three variations on the theme. Trends Genet. 22, 347–350. Rando, T.A., 2006. Stem cells, ageing and the quest for immortality. Nature 441, 1080–1086. Raya, A., et al., 2009. Disease-corrected haematopoietic progenitors from Fanconi anaemia induced pluripotent stem cells. Nature 460, 53–59. Reik, W., 2007. Stability and flexibility of epigenetic gene regulation in mammalian development. Nature 447, 425–432. Rideout, W.M., Eggan, K., Jaenisch, R., 2001. Nuclear cloning and epigenetic reprogramming of the genome. Science 293, 1093–1098. Riggs, A.D., 1975. X inactivation, differentiation, and DNA methylation. Cytogenet. Cell Genet. 14, 9–25.
K. Sakurada Rountree, C.B., et al., 2007. A CD133-expressing murine liver oval cell population with bilineage potential. Stem Cells 25, 2419–2429. Sadikovic, B., Andrews, J., Carter, D., Robinson, J., Rodenhiser, D.I., 2008. Genome-wide H3K9 histone acetylation profiles are altered in benzopyrene-treated MCF7 breast cancer cells. J. Biol. Chem. 283, 4051–4060. Sakurada, K., McDonald, F.M., Shimada, F., 2008. Regenerative Medicine and stem cell based drug discovery. Angew. Chem. Int. Ed. 47, 5718–5738. Silva, J., et al., 2008. Promotion of reprogramming to ground state pluripotency by signal inhibition. PLoS Biol. 6, e253. Soldner, F., et al., 2009. Parkinson's disease patient-derived induced pluripotent stem cells free of viral reprogramming factors. Cell 136, 964–977. Suzuki, K., et al., 2006. Global DNA demethylation in gastrointestinal cancer is age dependent and precedes genomic damage. Cancer Cell 9, 199–207. Takahashi, K., et al., 2007. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131, 861–872. Tamashiro, K.L., et al., 2002. Cloned mice have an obese phenotype not transmitted to their offspring. Nat. Med. 8, 262–267. Tanaka, S., et al., 2001. Placentomegaly in cloned mouse concepti caused by expansion of the spongiotrophoblast layer. Biol. Reprod. 65, 1813–1821. Tsankova, N.M., et al., 2006. Sustained hippocampal chromatin regulation in a mouse model of depression and antidepressant action. Nat. Neruosci. 9, 519–525. Waddington, C.H., 1942. The epigenotype. Endeavour 1, 18–20. Waterland, R.A., Michels, K.B., 2007. Epigenetic epidemiology of the developmental origins hypothesis. Annu. Rev. Nutr. 27, 363–388. Weaver, I.C., et al., 2004. Epigenetic programming by maternal behavior. Nat. Neurosci. 7, 847–854. Wong, A.H.C., Gottesman, I.I., Petronis, A., 2005. Phenotypic differences in genetically identical organisms: the epigenetic perspective. Hum. Mol. Genet. 14, R11–18. Yamanaka, S., 2009. Elite and stochastic models for induced pluripotent stem cell generation. Nature 460, 49–52. Yongson, N.A., Whitelaw, E., 2008. Transgenerational epigenetic effects. Annu. Rev. Genomics Hum. Genet. 9, 233–257. Yoshida, Y., Takahashi, K., Okita, K., Ichisaka, T., Yamanaka, S., 2009. Hypoxia enhances the generation of induced pluripotent stem cells. Cell Stem Cell 5, 237–241. Yu, J., et al., 2007. Induced pluripotent stem cell lines derived from human somatic cells. Science 318, 1917–1920. Zeng, W., et al., 2004. Transcript profile of CD4+and CD8+T cells from the bone marrow of acquired aplastic anemia patients. Exp. Hematol. 32, 806–814. Zhang, C.L., et al., 2002. Class II histone deacetylases act as signalresponsive repressors of cardiac hypertrophy. Cell 110, 479–488. Zhao, Y., Glesne, D., Huberman, E.A., 2003. human peripheral blood monocyte-derived subset acts as pluripotent stem cells. Proc. Natl. Acad. Sci. U.S.A. 100, 2426–2431.