Review Article, J Regen Med Vol: 1 Issue: 1
From Nanotechnology to Epigenomics and Regenerative Medicine
Dashzeveg Bayarsaihan* | |
Center for Regenerative Medicine and Skeletal Development, University of Connecticut Health Center, USA | |
Corresponding author : Dashzeveg Bayarsaihan Center for Regenerative Medicine and Skeletal Development, Department of Reconstructive Sciences, University of Connecticut Health Center, 262 Farmington Avenue, Farmington, CT 06030, USA Tel: 860-679-5456; Fax: 860-679-2910 E-mail: dashzeveg@uchc.edu |
|
Received: June 28, 2012 Accepted: August 28, 2012 Published: August 30, 2012 | |
Citation: Bayarsaihan D (2012) From Nanotechnology to Epigenomics and Regenerative Medicine. J Regen Med 1:1. doi:/10.4172/2325-9620.1000101 |
Abstract
From Nanotechnology to Epigenomics and Regenerative Medicine
The objective of regenerative medicine is to repair tissues and organs that are damaged by trauma, disease or aging. The stem cell based technologies have promised to advance therapeutic applications including organ transplants and tissue repair. Although more studies are now being conducted to reveal mechanisms regulating stem cell fate, epigenetic events that contribute to cell differentiation are still relatively poorly defined. Next generation sequencing combined with chromatin immunoprecipitation (ChIPseq) or bisulfite sequencing (BS-seq) is the most common tool to study histone modifications and DNA methylation. However, it requires an abundance of research material and has limited ability to detect different epigenetic signatures in the same cell type. The recent development of the single chromatin analysis at the nanoscale (SCAN) using the nanofluidic single-molecule sorter allows the simultaneous detection of multiple epigenetic marks in a small number of cells.
Keywords: Nanotechnology; Nanofluidics; Epigenome; Chromatin; Chip-seq; Regenerative medicine; Next generation sequencing
Keywords |
|
Nanotechnology; Nanofluidics; Epigenome; Chromatin; Chip-seq; Regenerative medicine; Next generation sequencing | |
Introduction |
|
Regeneration is a highly regulated process throughout the life cycle of an organism for maintaining or restoring the normal function of cells and tissues. Regenerative medicine deals with the processes of replacing or regenerating human tissues or organs to restore normal function [1]. One of the main obstacles in regenerative medicine is to develop technologies that allow optimal recovery of damaged tissues. In this respect bioengineering of artificial organs might solve some of the problems associated with poor tissue regeneration. Stem cell research has emerged as one of the most promising areas in cell transplantation including rapid advancement of personalized tissue regeneration therapy to reprogram adult cells into induced pluripotent stem cells (iPSCs) [2]. These cells carry many properties of embryonic stem cells (ESCs) and capable of differentiating into any type of body cell. Thus, there is little doubt that progress in stem cell therapy will enable an organism to regenerate damaged tissues or organs with new cells derived from a patientâ??s own cells. | |
The main quality of ESCs is a capacity to differentiate into the three primary germ layers: ectoderm, endoderm and mesoderm, which makes them an ideal cell source for translational studies [3,4]. The ESCs regulatory circuitry is modulated by a â??coreâ?? set of transcription factors OCT4, NANOG and SOX2 [5,6], which play a dual role: in silencing the expression of developmental regulators and, at the same time maintaining the expression of self-renewal and pluripotency genes. Although different regulatory pathways participate in keeping the balance between self-renewal and fate specification [7], histone modifications and DNA methylation are the most important landmarks for setting up differential epigenetic switches at the early stages of stem cell lineage commitment [8]. | |
Histone Modifications |
|
The basic unit of chromatin, the nucleosome, consists of 167 bp DNA fragment wrapped around histone octamer made up of two copies of H2A, H2B, H3 and H4 to provide a rigid chromatin structure [9]. The nucleosome is a subject to a variety of post-translational modifications such as methylation of lysines and arginines, acetylation, phosphorylation, ubiquitination and SUMOylation, which, in most cases, occur at the N-terminal tails of the core histones [10]. | |
Histone acetyltransferases | |
As a rule, acetylation of histones facilitates a more â??openâ? or â??relaxedâ? chromatin to favor the initiation of transcription [9,11]. The acetylated N termini protruding from the nucleosome core provide reduced affinity for the DNA allowing the chromatin to adopt a more relaxed structure for facilitating the recruitment of the basic transcription machinery. The acetylated histone marks H3K4ac, H3K27ac and H3K39ac are associated with transcriptional activation [12-15]. CBP and its close homolog p300 histone acetyltransferases (HATs) carry out the majority of these modifications [16,17]. | |
Histone deacetylases | |
A more condensed chromatin state is mediated by histone deacetylases (HDACs), a class of enzymes that reverse HAT activity and spread the gene silencing [18]. HDAC1 and HDAC2 participate in the regulation of chromatin structure as the core catalytic components of Sin3A, NuRD and CoREST co-repressor complexes. Ablation of HDAC1, but not HDAC2, is accompanied by significant reduction of histone hypoacetylation and increase in H3K56ac [19]. HDAC1-deficient embryoid bodies are smaller and display increased expression of cardiomyocyte and neuron-specific genes. | |
NuRD-mediated deacetylation of histone H3K27 enables Polycomb repressive complex PRC2 recruitment and subsequent H3K27 trimethylation at the NuRD target promoters [20]. Therefore, in ESCs NuRD is required for fine tuning the expression of genes essential for embryonic development by controlling the balance between acetylation and methylation of histones. Methyl-CpG binding domain protein 3 (MBD3), another component of NuRD repressive complex, functions as a dual epigenetic regulator by restricting differentiation of ESCs towards the trophectoderm lineage and maintaining cell pluripotency [21]. Specifically, Mbd3 knockdown in mouse ESCs leads to an increased acetylation in promoters of trophectoderm markers, including Cdx2, Eomesodermin, and Hand1. | |
SIRT1, a founding member of the Sirtuin family of histone deacetylases, is a subunit of the Polycomb repressive complex PRC4 involved in cellular resistance to stress, metabolism, hematopoiesis, aging and tumor suppression [22,23]. SIRT1 blocks nuclear translocation of cytoplasmic p53 in response to endogenous oxygen and triggers mitochondrial-dependent apoptosis in mouse ESCs [24]. Moreover, SIRT1 controls the human ESC fate by the p53-mediated regulation of NANOG [25] and prepares the PTEN/JNK/FOXO1 pathway response to reactive oxygen species [26]. Down-regulation of SIRT1 causes reactivation of the epigenetically repressed developmental genes DLL4, TBX3 and PAX6 [25]. Inhibition of SIRT1 by nicotinamide initiates differentiation of human ESCs toward motor neurons [27]. | |
Although HDAC activity is primarily linked to gene repression, recent studies have found that HDACs are also associated with active genes [28,29]. HDAC1 binds to many developmentally important genes including regulators of ESC and TSC self-renewal. HDAC1 colocalizes with MBD3, a subunit of the NuRD complex, at the promoters of Oct4, Nanog and Kfl4 [28]. | |
Histone methyltransferases | |
Histone methylation maintains chromatin either in an active or a silenced state. H3K4me3 and H3K36me3 define open chromatin associated with transcriptionally active loci [30]. By contrast, H3K9me3 and H3K27me3 facilitate chromatin compaction and gene repression [31]. Bivalent chromatin enriched with H3K4me3 and H3K27me3 defines the promoters of developmentally important genes in ESCs or multipotent progenitor cells in the poised state, which will be activated as differentiation proceeds [32,33]. | |
The Polycomb Repressive Complexes 1 and 2 (PRC1 and PRC2) are two histone methyltransferase complexes originally defined as repressors of the HOX genes [34-36]. There are six major groups of PRC1 complexes, each containing a distinct PCGF subunit, H2A monoubiquitin ligases RING1A/B and a unique set of associated proteins [37]. These PRC1 complexes differ in their genomic localization, and only a small subset colocalizes with H3K27me3. EZH2, EED and SUZ12 are the core subunits of PRC2 [35,38]. EZH2, an enzymatic component of PRC2, methylates histone H3 on lysine 27 to establish gene repression during ESC differentiation. For example, PRC2 affects the balance of NANOG expression in mouse ESCs by regulating the epigenetic status of the Nanog promoter and, therefore, the equilibrium between self-renewal and differentiation [39]. | |
The genome-wide mapping studies revealed that PRC1, PRC2 and H3K27me3 co-occupy the same genomic loci [34,35,38]. Although the chromatin recognition mode by PRC2 is not completely understood, JARID2, a member of the Jumonji family, facilitates PRC2 binding to chromatin [40]. JARID2 anchors PRC2 at the Polycomb responsive sites through the DNA-binding domain ARID. However, additional mechanisms have also evolved to mediate PRC2 recruitment including the non-coding RNAs and the large GC-rich elements [41,42]. A class of short RNAs of approximately 50-200 nucleotides in length, transcribed from the 5â?? end of PRC2 target genes was proposed to play a critical role in the Polycomb recruitment [43]. Short RNA transcription, associated with RNA Polymerase II (Pol II) and H3K4me3, occurs in the absence of mRNA transcription and is independent of Polycomb activity. | |
The loss of PRC2 binding results in a global increase in H3K27ac at the PcG occupancy sites [44] catalyzed by p300/CBP. This switch from methylation to acetylation at H3K27 correlates with the transcriptional activation of the PcG-associated genes during ESC differentiation. This model predicts that in stem cells PCR2 represses transcription by excluding p300/CBP from the target genes. | |
The PRC1 binding to the genomic targets is achieved through association with PRC2 or by the REST-dependent mechanism. REST (also called NRSF) is a zinc finger protein associated with different repressor complexes including CoREST, Sin3A, G9a, LSD1 and HDACs [45]. REST associates with CBX and RING1B, subunits of PRC1, and interacts with the OCT4-SOX2-NANOG transcriptional network. Loss of functional REST does not abrogate the stem cell potential although ablation of REST in ESCs and in differentiating neurons changes PRC1 occupancy at the REST-binding sites (RE1 elements) [46-50]. Therefore, the gene-specific control of PRC1 during ESC differentiation depends upon REST binding to the RE1 sequence. REST also mediates PRC1 recruitment to a subset of Polycomb regulated neuronal genes but limits the PRC2 binding at CpG islands [51]. | |
Phosphorylation of the carboxy-terminal domain of Pol II is associated with transcription initiation, elongation and termination. Sites of active transcription are generally characterized by hyperphosphorylation at Ser 2 residues, whereas inactive or poised genes may lack Pol II or may bind Ser 5-phosphorylated Pol II at promoter proximal regions. Bivalent genes possess poised Pol II configuration (Ser 5 phosphorylation) enforced by PcG-mediated ubiquitination of H2A [52]. Conditional deletion of RING1A and RING1B leads to gene de-repression by the sequential loss of H2A ubiquitination and subsequent release of poised Pol II. | |
Endogenous retroviruses (ERVs) are widely dispersed in the mammalian euchromatin, comprising approximately 10% of the mouse genome [53]. It was noticed that H3K9me3 marks ERVs in ESCs but not in mouse embryonic fibroblasts. H3K9 methyltransferase Setdb1 and the Kr?¼ppel-associated box (KRAB) co-repressor KAP1 are required for H3K9me3 and proviral silencing during the early stages of embryogenesis. Setdb1 inactivation in mice upregulates a set of genes that are distinct from those derepressed by the deficiency in the DNA methyltransferase genes Dnmt1, Dnmt3a, and Dnmt3b, with the exception of a small number of germline-specific genes [54]. Numerous ERVs lose H3K9me3 and are concomitantly derepressed exclusively in Setdb1 knockout mouse ESCs. It was proposed that H3K9me3 might maintain ERVs in a silent state by directly inhibiting deposition of active covalent histone marks [55]. | |
The ablation of Setdb1 results in peri-implantation lethality in mice [56]. The loss of Setdb1 in mouse ESCs leads to a decrease of H3K9 methylation, differentiation of trophoblast-like cells and upregulation of trophectoderm-specific genes Cdx2g and Tcfap2g [57-59]. Genome-wide mapping revealed that Setdb1 occupies key developmental genes marked by bivalent chromatin. Polycomb and Setdb1 work together to repress these genes, and the loss of either one can destabilize the pluripotency state [60]. The stability of Setdb1-PRC2 is achieved through interactions with OCT4, which anchors the repressive complex at the trophoblast-specific regulatory regions [58]. | |
The establishment of H3K9me3 marks is also a characteristic feature of pericentric heterochromatin. The association of Setdb1 with heterochromatin protein HP1 and chromatin assembly factor CAF1 results in H3K9 methylation in the pericentric regions [61]. G9a-GLP is another H3K9 methyltransferase complex [62]. The G9a-GLP-mediated stable gene silencing is accompanied by the recruitment of DNMTs and LSH, a member of the SNF2 family of chromatin remodeling ATPases, to specific genomic loci. In the absence of LSH, DNA methylation is lost or significantly reduced along a large set of promoters causing changes in the expression of corresponding genes during differentiation [63]. Activation of protein kinase A (PKA) determines timing of early differentiation by regulating the expression of G9a along with H3K9 methylation and DNA methylation across Oct3/4 and Nanog gene promoters. Deletion of G9a completely abolishes PKA-elicited differentiation, results in the prolonged expressions of Oct3/4 and Nanog at embryonic day 7.5 and delays development [64]. | |
A mutually exclusive role was found for PRC1 and Suv39H1 in regulation of H3K9me3 and early lineage specification [65]. The exclusion of PRC1 and Pol II complexes from the bivalent genes upon trophoblast lineage commitment leads to down-regulation of developmental genes. For example, expression HOX, GATA and SOX is repressed by the Suv39H1-mediated H3K9 trimethylation and de novo DNA methylation. | |
Trithorax (TrxG) group restricts the Polycomb-associated repression to maintain the balance between ESC self-renewal and differentiation [8]. TrxG-mediated methylation of H3K4 or H3K36 spreads transcriptionally active regions across the genome. WDR5 (WD repeat domain 5), a subunit of TrxG, regulates key self-renewal factors by cooperating with the OCT4-SOX2-NANOG circuitry [66]. DPY-30 is another member of TrxG group critical during stem cell fate decisions [67]. Ablation of DPY-30 does not alter stem cell self-renewal but causes significant impact on the ESC differentiation, which is accompanied by changes in H3K4 trimethylation and gene activation across a subset of key developmental regulatory gene loci. ASH1, another member of TrxG group, antagonizes the Polycomb-induced silencing through the H3K36-specific dimethyltransferase activity [68]. | |
Histone demethylases | |
Compared to histone methyltransferases, relatively little is known about the function of histone demethylases during the early stages of stem cell specification. Histone demethylase LSD1, also known as KDM1A and AOF2, is a flavin adenine dinucleotide (FAD)-containing enzyme that removes mono- and di-methyl groups from H3K4me2 as part of a repressive complex that consists of CoREST, RCOR2 co-repressor and HDACs [69,70]. LSD1 maintains the balance between H3K4 and H3K27 methylation by binding to the bivalent promoters of key developmental genes [71]. Deletion of Lsd1 in mouse ESCs changes the appropriate timing of developmental genes through a global increase in H3K56 acetylation [72,73] suggesting a possible link between LSD1-dependent demethylation of H3K4me2 and p300 mediated H3K56 acetylation. Moreover, LSD1 is essential in decommissioning of enhancers during ESC differentiation, which is essential for the complete shutdown of the ESC gene expression program and the transition to new cell states [74]. | |
Pasini et al. showed a functional interplay between H3K4me3 demethylase JARID1A (RBP2) and PRC2 [75]. By recruiting JARID1A to its target genes, PRC2 sustains repressive activity during ESC differentiation. JARID1B (KDM5B), an activator of the OCT4-SOX2-NANOG regulatory network, is another H3K4me2/3 demethylase [76]. JARID1B activates self-renewal-associated gene expression by repressing a cryptic intrageneic initiation and maintaining an H3K4me3 gradient important for productive transcriptional elongation. JARID1B occupies the H3K36me3-rich genomic regions via an association with the chromodomain protein MRG15. Depletion of JARID1B or MRG15 increases H3K4me3 and cryptic transcription and inhibits transcriptional elongation of JARID1B target genes. JARID1B colocalizes with PRC2 and H3K4me3 at the transcription start sites of developmental regulators and during neural differentiation, depletion of JARID1B leads to inefficient silencing of stem and germ cell-specific genes [77]. | |
The JmjC domain-containing protein UTX (ubiquitously transcribed tetratricopeptide repeat, X chromosome), as well as the related JMJD3, were identified as H3K27-specific demethylases [78-80]. It was noted that TGFβ/Nodal signaling, by recruiting JMJD3 to chromatin with the assistance of Smad2/3, counteracts the PRC2-mediated repression [81]. | |
UTX directly binds to the HOXB1 locus and is required for its activation [80]. Inactivation of JMJD3 orthologue in Caenorhabditis elegans results in abnormal gonad development. Therefore, it was suggested that UTX/JMJD3-regulated H3K27me3 demethylation is essential for normal development. Moreover, UTX associates with Trithorax group proteins MLL2/3 [82]. These findings suggest a highly coordinated mechanism for transcriptional activation in which H3K4 methylation by MLL2/3 are linked with demethylation of H3K27 through UTX. | |
DNA Methylation and Demethylation |
|
DNA methylation of the mammalian genome has a profound impact on chromatin structure, gene expression and maintenance of cellular identity [83]. DNA methylation, catalyzed by DNA methyltransferases (DNMTs), is a transfer of the methyl group from S-adenosylmethionines to a cytosine [83-85]. | |
DNA methyltransferases | |
ESCs have a unique DNA methylation signature responsible for the control of self-renewal, pluripotency and cell fate commitment [86,87]. Although mouse ESCs and epiblast stem cells (mEpiSC) are two closely related classes of pluripotent stem cells, these cells show differences in DNA methylation pointing to the importance of epigenetic modifications in regulating pluripotency [88]. | |
In response to changing culture conditions DNA methylation alters in human ESCs [89]. After adaption to new environment, the transcriptome and DNA methylation become quite stable for additional passaging. However, upon reversion to the original culture conditions, DNA methylation pattern across many genes becomes irreversible indicating that ESCs harbor a memory of culture history. | |
Differential DNA methylation of pluripotency-associated genes such as NANOG and OCT4 has been observed in pluripotent and differentiated cells [90]. DNA methylation of the OCT4 promoter is a highly coordinated process and depends on the cooperation between DNMT3A, DNMT3B, DNMT1 and G9a [91]. On the other hand, DNMT3B participates in recognition of a higher-order chromatin structure in ESCs by preferential association with histone deacetylase SIRT1 and histone H1-rich heterochromatin [92]. | |
Interestingly, the two isoforms of DNMT3A display a dual role in mouse ESCs by working as the negative and positive regulators of transcription. The N-terminal 219 amino acid region, unique for the DNMT3A1 isoform, has a specific DNA-binding activity [93]. The recruitment of DNMT3A1 to the inactive OCT4 promoter prevents DNA methylation and activates OCT4 expression. At the same time, DNMT3A1 silences the promoter of actively transcribed Vtn (vitronectin) gene. | |
DNA demethylases | |
Although the regulatory cues orchestrated by DNMTs are relatively well investigated, the class of enzymes known as DNA demethylases is not sufficiently well explored group of epigenetic modulators [94]. DNA methylation at the C-5 position of cytosine (5mC) in the mammalian genome is a key epigenetic event critical for various cellular processes. In contrast to 5mC, which is under-represented at gene promoters and CpG islands, 5-hydroxymethylcytosine (5hmC) mostly associates with euchromatin and active transcription sites [95]. 5hmC enrichment correlates with a pluripotent state indicating that genomic 5hmC is an epigenetic feature of cellular pluripotency [96,97]. High levels of 5hmC were found within transcription start sites around CpG islands of developmental regulator genes (e.g. Hox genes) and long interspersed nuclear elements (LINE1) [98]. The ratio of 5hmC to 5mC is different across different genomic regions, which reflects a local state of chromatin in a given cell type. TET proteins, mammalian homologs of the trypanosome proteins JBP1 and JBP2, are 2-oxoglutarate (2OG)- and Fe(II)-dependent enzymes that catalyze conversion of 5mC to 5hmC [99]. TET1 and TET2 have an important role in mouse ESCs by maintaining the expression of pluripotency marker genes [100,101]. TET proteins bind to the CpG-rich sequences at promoters of both transcriptionally active and Polycomb-repressed genes [102] The knockdown of TET factors reduces the expression of Nanog, Esrrb, Klf2 and other pluripotency genes by increasing DNA methylation at their promoters, committing ESCs toward trophectoderm lineage. The TET1-mediated promoter hypomethylation is not only required for maintaining the transcriptionally active state but it is also involved in silencing gene targets of the Polycomb repression by facilitating the recruitment of PRC2 to the CpG-rich gene promoters [91,101,103]. The majority of TET1 binding sites, in contrast to 5mC, are enriched at the CpG-rich promoter regions and within the gene bodies to prevent unwanted DNA methyltransferase activity [103,104]. MBD3, methyl CpG-binding protein 3 and BRG1, ATP-dependent helicase are required for normal levels of 5hmC in vivo. MBD3 preferentially binds to 5hmC in the TET1-dependent manner and depletion of Mbd3 preferentially affects expression of the 5hmC-primed genes [105]. | |
DNA glycosylases | |
Another group of enzymes, in addition to DNA methyltransferases and demethylases, has emerged as a critical regulator of stem cell fate decision. Thymine DNA glycosylase (TDG) is a member of the uracil DNA glycosylase family of DNA repair enzymes [106]. TDG was proposed to act against the mutability of 5mC deamination in the mammalian genome. The TDG-dependent DNA repair provides epigenetic stability by contributing to the maintenance of active (H3K4me2 and H3K4me3) and bivalent chromatin marks (H3K4me3/K27me3) during ESC differentiation [106]. TDG associates with nucleosomes at the promoters of developmental genes both in mouse embryo fibroblasts (MEFs) and in ESCs. MEFs derived from Tdg null embryos show impaired gene regulation, which is coincident with imbalanced histone modification and CpG methylation at the regulatory regions of affected genes. Depletion of TDG in mouse ESCs leads to accumulation of 5-carboxylcytosine (5caC), which is specifically recognized and excised by TDG [107]. Interestingly, loss of TDG compromises the association of CBP/p300 and Trithorax group protein MLL1 at the promoters of Hoxd13, Hoxa10, Sfrp2 and Twist2 [106]. | |
Analysis of Epigenetic Modifications |
|
Next generation sequencing (NGS) has dramatically increased our ability to survey epigenetic switches genome-wide. NGS combined with the chromatin immunoprecipitation (ChIP) and bisulfite conversion of genomic DNA (ChIP-seq and BS-seq) has become the most accepted technique for analyzing histone modifications and DNA methylation. In ChIP-seq assay, protein-DNA complexes are crosslinked, immunoprecipitated, purified, and amplified for the target sequence analyses using NGS [108]. In BS-seq method, treatment of DNA with bisulfite introduces specific changes in the sequence converting cytosine residues to uracil leaving 5mC residues unaffected, which provides the single nucleotide resolution information about the DNA methylation status [109]. However, both procedures require fairly high amount of starting material precluding the application of ChIP-seq and BS-seq in many biologically important but rare cell types. The recently described nano-ChIP-seq method combines a high-sensitivity small-scale ChIP assay and a tailored procedure for generating high-throughput sequencing libraries from scarce amounts of ChIP DNA, but even this approach requires at least 10,000 cells [110]. | |
To overcome these limitations, single chromatin analysis at the nanoscale (SCAN) has been developed by Cipriany et al. [111]. The protocol combines multicolor fluorescence microscopy and nanofluidic single-molecule sorter for detection of histone and DNA markers within individual chromatin fragments. Nanofluidics has emerged as a discipline of engineering, where a fluid flows in structures with at least one transversal dimension approaching the nanometer range [112]. The nanoscale technology based on nanofluidics employs the science of building microdevices with chambers and tunnels for the containment and flow of fluid samples with high precision and efficiency. Therefore, nanofluidic sorting and analysis of single molecules in real time has promised to provide the unprecedented opportunity to simultaneously explore multiple epigenetic changes on chromatin derived from a small number of cells. | |
For SCAN validation fluorescently labeled methylated and unmethylated DNA molecules were lined up single-file in a nanofluidic channel in the input section [113]. Methylated DNA molecules were automatically sorted into one channel, while the unmethylated molecules were separated into a different channel achieving up to 98% sorting accuracy. This procedure requires very small amount of DNA unlike NGS. Therefore, SCAN could allow the investigators to perform a multi-step analysis of epigenetic modifications quickly from a relatively limited amount of cells. | |
Acknowledgement |
|
The author wants to thank the Connecticut Stem Cell Grant 09-SCB-UCHC. | |
References |
|
|