FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2069536617> ?p ?o ?g. }

• W2069536617 endingPage "R236" @default.
• W2069536617 startingPage "R233" @default.
• W2069536617 abstract "Over the past few years we have seen an odd change, or extension, in the use of the word ‘epigenetic’ when describing matters of gene regulation in eukaryotes. Although it may generally be that it is not worth arguing over definitions, this is true only insofar as the participants in the discussion know what each other means. I believe the altered use of the term carries baggage from the standard definition that can have misleading implications. Here I wish to probe our use of language in this way, and to show how such a discussion leads to some more general considerations concerning gene regulation. Over the past few years we have seen an odd change, or extension, in the use of the word ‘epigenetic’ when describing matters of gene regulation in eukaryotes. Although it may generally be that it is not worth arguing over definitions, this is true only insofar as the participants in the discussion know what each other means. I believe the altered use of the term carries baggage from the standard definition that can have misleading implications. Here I wish to probe our use of language in this way, and to show how such a discussion leads to some more general considerations concerning gene regulation. To begin, consider an example of an epigenetic change (I'll formally define the term in a moment) which involves a feedback loop. In this case a transcriptional activator — perhaps in response to an extra-cellular signal — binds DNA and turns on transcription of a second transcriptional activator; if the latter activates expression of its own gene, the new state of gene expression will be maintained even in the absence of the original activator. Such feedback loops are found widely in nature from phage lambda, to bacteria, to Drosophila and higher eukaryotes. Because the eukaryotic transcriptional machinery is so complicated — with its co-activators and co-repressors, histone modifiers and so on — it is rather easy to lose sight of the logic of regulatory systems: genes are turned on and off, often in feedback and interacting loops, by specific DNA-binding proteins. And these specific binding proteins typically work by recruiting the various components of the machinery to specific genes. These specific DNA binding proteins are expressed according to internal programs and, sometimes, in response to extracellular signals [1Ptashne M. Gann A. Genes and Signals. Cold Spring Harbor Press, 2002Google Scholar]. It is easy to see how the kind of feedback loop described above fits the classical definition of an epigenetic change in gene expression: “… a change in the state of expression of a gene that does not involve a mutation, but that is nevertheless inherited in the absence of the signal (or event) that initiated the change”. As a glance at the literature will reveal, however, histone modifications — acetylation, phosphorylation, methylation, and so on — are now often explicitly called ‘epigenetic modifications’. This despite the fact that, so far as I am aware, no histone modification has been shown to be heritable. (Eukaryotic DNA is of course wrapped in histones to form strings of nucleosomes, and in what follows I will refer interchangeably to modifications of histones and nucleosomes.) Certain histone modifications — methylation and acetylation, for example — are often found at actively transcribed genes. The enzymes that effect these modifications have no inherent specificity for one group of nucleosomes — such as those at a gene to be activated — over the nucleosomes found throughout the genome. Rather, like the transcriptional machinery itself, these modifying enzymes must be recruited to specific genes by specific DNA-binding proteins — transcriptional activators, for example. And, where examined, such modifying enzymes dissociate from the gene when the activator stops working, and the modifications are (actively or passively) lost [2Katan-Khaykovich Y. Struhl K. Dynamics of global histone acetylation and deacetylation in vivo: rapid restoration of normal histone acetylation status upon removal of activators and repressors.Genes Dev. 2002; 16: 743-752Crossref PubMed Scopus (163) Google Scholar]; these enzymes, of course, then work non-specifically and at a lower rate as part of the background. A similar picture holds for gene repression. For example, silent mating type loci in yeast are maintained ‘off’ by a group of Sir proteins, bound to the chromatin, one of which bears a required histone deacetylase (HDAC) activity that removes acetyl groups from nucleosomes. This complex does not form spontaneously, neither is it self-perpetuating: formation of the complex requires specific DNA-binding proteins that recruit the Sir proteins, and loss of those recruiters causes the complex to fall apart [3Cheng T. Gartenberg M. Yeast heterochromatin is a dynamic structure that requires silencers continuously.Genes Dev. 2000; 15: 452-463Google Scholar]. Another example is provided by those cases in which RNA interference (RNAi) triggers silencing of transcription with concomitant recruitment of nucleosome modifiers. In some of these cases, specific DNA-binding proteins are not required — the silencing machinery is specifically recruited by the RNAi machinery. Here the continual production of specific RNA (and hence of RNAi) is required for maintenance of silencing and of those nucleosome modifications [4Bühler M. Verdel A. Moazed D. Tethering RITS to a nascent transcript initiates RNAi- and heterochromatin-dependent gene silencing.Cell. 2006; 125: 873-886Abstract Full Text Full Text PDF PubMed Scopus (303) Google Scholar]. The transient nature of nucleosome modifications should not be surprising, because there is no known mechanism for self-propagation of such modifications. That is, there are no known enzymes that recognize a specific modification on a nucleosome and ensure that that modification will be transmitted to the next generation. This situation might change, of course (a possibility I return to later), but in the meantime the use of the word epigenetic implies a fundamental property of the system — self-propagation — that is, so far as we now know, not true of nucleosome modifications. I wish now, partly in light of what I have said, to comment briefly on the possible roles of histone modifications in regulating transcription of genes, and on the words sometimes used to describe the roles of these modifications. It is often said that genes “are regulated by chromatin modifications”. Isn't this rather misleading? The simple fact that the nucleosome modifiers have no inherent specificity — beyond that for one or another residue on any of the many nucleosomes in the cell — tells us that these modifying enzymes cannot in any informative way be said to ‘regulate’ gene expression. Genes are typically regulated, in the meaningful sense, by specific DNA binding proteins that recruit transcribing or repressing complexes — including nucleosome modifiers. As noted, it is true that certain nucleosome modifications are typically found associated with active genes. But so is RNA polymerase, an observation that speaks to the events that occur during activation, but not to the key regulatory step. One way nucleosome modifiers are sometimes said to regulate genes is by “opening or closing the chromatin structure”, thereby allowing (or preventing) access of the transcriptional machinery to the DNA. Such a view is problematical on the face of it: how would the nucleosome modifiers ‘know’ which genes to pack or unpack unless instructed by a specific DNA binding recruiter? Moreover, the example we discussed — the silent mating type loci in yeast — suggests that even in ‘silenced’ regions (called heterochromatic regions), the DNA must be accessible to specific DNA binding proteins. Recall that, in that case, continual recruitment by the specific DNA-binding regulator is required to maintain the silenced state. The straightforward conclusion would be that this heterochromatic region is silent, not because of lack of access to specific DNA-binding proteins, but rather because the specific proteins that bind to the relevant sites recruit proteins, such as Sirs, which somehow diminish the efficiency with which an activator, binding to the same general region, can recruit the transcriptional machinery. Indeed, strong activators have been shown to overcome silencing in yeast, flies and in mammalian cells [5Aparicio O. Gottschling D. Overcoming telomeric silencing: a trans-activator competes to establish gene expression in a cell cycle-dependent way.Genes Dev. 1994; 8: 1133-1146Crossref PubMed Scopus (237) Google Scholar, 6Ahmad K. Henikoff S. Modulation of a transcription factor counteracts heterochromatic gene silencing in Drosophila.Cell. 2001; 104: 839-847Abstract Full Text Full Text PDF PubMed Scopus (122) Google Scholar, 7Memedula S. Belmont A. Sequential recruitment of HAT and SWI/SNF components to condensed chromatin by VP16.Curr. Biol. 2003; 13: 241-246Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar, 8Janicki S. Tsukamoto T. Salghetti S. Tansey W. Sachidanandam R. Prasanth K. Ried T. Shav-Tal Y. Bertrand E. Singer R. et al.From silencing to gene expression: real-time analysis in single cells.Cell. 2004; 116: 683-698Abstract Full Text Full Text PDF PubMed Scopus (516) Google Scholar], and more direct experiments show that regulatory protein binding sites are accessible in heterochromatic regions in yeast [9Sekinger E. Gross D. Silenced chromatin is permissive to activator binding and PIC recruitment.Cell. 2001; 105: 403-414Abstract Full Text Full Text PDF PubMed Scopus (122) Google Scholar, 10Chen L. Widom J. Mechanism of transcriptional silencing in yeast.Cell. 2005; 120: 37-48Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar]. And, of course, regions silenced by RNAi must be accessible to the RNA polymerase that transcribes the RNA necessary to maintain repression. Whatever nucleosome modifications do — ‘loosen the chromatin’ as mentioned, or, another idea, provide binding sites to help in the recruitment of other proteins — it is very unlikely that such modifications suffice for efficient gene activation. Rather, it seems that an activator must contact the transcriptional machinery to recruit its various parts (directly or indirectly), and these interactions cannot be dispensed with for efficient expression. Thus, for example, nucleosome-depletion experiments in yeast (admittedly a messy business) show that for most genes, depletion suffices for no more than a modest increase in basal transcription [11Wyrick J. Holstage F. Jennings E. Causton H. Shore D. Grunstein M. Lander E. Young R. Chromosomal landscape of nucleosome-dependent gene expression and silencing in yeast.Nature. 1999; 402: 418-421Crossref PubMed Scopus (320) Google Scholar]. Moreover, in yeast, various fusion proteins bearing a DNA-binding domain covalently attached to one or another component of the mediator (a protein complex required for efficient transcription of many genes) can activate transcription to a very high level (by directly recruiting to the gene a component of the transcriptional machinery), but only much smaller effects have been seen with fusions of a DNA-binding domain to one or another component of a complex that modifies nucleosomes [12Barberis A. Pearlberg J. Simkovich N. Farrell S. Reinagel P. Bamdad C. Sigal G. Ptashne M. Contact with the Pol II holoenzyme suffices for gene activation.Cell. 1995; 81: 359-368Abstract Full Text PDF PubMed Scopus (233) Google Scholar]. Mutations in histone residues subject to modification have small effects on gene activation in yeast (Mary Ann Osley, personal communication), and it is now suspected that at least some of these modifications are involved in transcriptional elongation [13Carey M. Li B. Workan J.I. RSC exploits histone acetylation to abrogae the nucleosomal block to RNA polymerase II elongation.Mol. Cell. 2006; 24: 481-487Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar]. The notion that removing inhibitory proteins from DNA does not, in general, suffice for efficient gene expression is not unprecedented. In bacteria, for example, removal of a specific repressor rarely suffices for full transcription of a gene. Most bacterial genes are controlled by activators as well as by repressors, and removal of the repressor, in the absence of the activator, elicits in general only rather low (basal) expression [1Ptashne M. Gann A. Genes and Signals. Cold Spring Harbor Press, 2002Google Scholar]. The surmise that histone modifications per se do not suffice to activate or repress a gene is consistent with the findings that purported modified histone–target protein interactions are weak. Interactions that are weak can nevertheless be important, of course, but in each case their significance would remain to be shown. For example, the interesting idea that the interaction between a Drosophila Polycomb group protein and a methylated histone suffices to direct the Polycomb group protein to specific sites has not survived critical tests [14Ringrose L. Paro R. Polycomb/trithorax reponse elements and epigenetic memory of cell identity.Development. 2007; 134: 223-232Crossref PubMed Scopus (361) Google Scholar]. It is perhaps worth emphasizing again that some nucleosome modification might turn out — contrary to what we now know — to be self-perpetuating. One part of a possible mechanism is suggested by the phenomenon of ‘spreading’ of silencing in yeast. That effect — which can extend silenced regions over long stretches of DNA — involves the Sir proteins mentioned above, and in particular the HDAC activity (also referred to above) of one of these proteins. The idea is that a recruited Sir deacetylates an adjacent nucleosome and thereby increases the affinity of that nucleosome for an additional Sir complex, and so on [15Yang B. Kirchmaler A.L. Bypassing the catalytic activity of SIR2 for SIR protein spreading in Sacchsromyces cerevisiae.Mol. Biol. Cell. 2006; 17: 5287-5297Crossref PubMed Scopus (42) Google Scholar]. The Sirs, as we noted, require a recruiter to bind and to maintain deacetylation of histones. But one might imagine some other modification that, in the absence of a recruiter, would be self-reinforcing; and, we could imagine, that that modification would be inherited as nucleosomes were distributed to daughter chromosomes upon division. This remains to be seen. But if some such scenario holds, our current understanding requires that any such modification would have to be triggered by specific DNA binding proteins or by RNAi. It is interesting in this context to consider DNA methylation. In contrast to histone modifications, DNA methylation — a modification found in higher eukaryotes and bacteria, but not in yeast and flies — can be self-sustaining: maintenance methylases recognize hemi-methylated DNA, the product of replication of fully methylated DNA, and add methyl groups to the unmethylated DNA strand. It is believed that tumor suppressor genes are often turned off by aberrant methylation and kept off by epigenetic inheritance of that state. Assuming this is so, the question then arises as to what triggered methylation of these genes in the first place. Did the methylase spontaneously hop on and begin to work at (many) specific place? Or was there a genetic event, that for example, placed a methylase recruiting site nearer the gene, or caused over-production of a methylase recruiter? Another possibility arises from the suggestion that a specific histone modification can trigger DNA methylation [16Ooi S.K. Bestor T.H. DNMT3l connects unmethylated histone H3 lysine to DNA cytosine-5 methylation.Nature. 2007; (in press)Google Scholar]. The idea might then be that the hypothetical mutational event caused aberrant histone modification that, in turn, triggered DNA methylation. Experiments with 5-AzaC cast doubt on the ‘spontaneous methylation’ idea. Treatment with this methylation inhibitor typically results in de-methylation. But when one removes the 5-AzaC the aberrant DNA methylation typically reappears [17Bestor T. Unanswered questions about the role of promoter methylation in carcinogenesis.Ann. N.Y. Acad. Sci. 2003; 987: 22-27Crossref Scopus (51) Google Scholar, 18McGarvey K. Fahrner J. Greene E. Martens J. Jenuwein T. Baylin S. Silenced tumor suppressor genes reactivated by DNA demethylation do not return to a fully euchromatic chromatin state.Cancer Res. 2006; 66: 3541-3549Crossref PubMed Scopus (240) Google Scholar]. Thus, whatever mechanism (presumably requiring a genetic change) that initiated the methylation remains in place to restore the methylation. In bacteria there are several cases in which DNA methylation of crucial sites occurs spontaneously unless blocked by specific DNA-binding proteins. A mutant lacking such a protein would display aberrant methylation, a pattern that would be — as for the hypothetical scenarios I outlined above for tumor suppression — caused by a genetic event [19Casadesús J. Low D. Epigenetic gene regulation in the bacterial world.Microbiol. Mol. Biol. Rev. 2006; 70: 830-856Crossref PubMed Scopus (341) Google Scholar]. Unfortunately, especially in dealing with eukaryotes, there can be a very high bar to actually proving that the event that initiated self-propagating aberrant methylation did not involve a genetic change. One needs, in effect, to demonstrate that no mutation that might have initiated the changes occurred, something very hard to do. Why might one be pre-disposed to misuse the term epigenetic? The term is sometimes used in the context of “maintaining stable states of gene expression”, as though some ‘locked in’ mechanism, involving histone and/or DNA modifications, were required for stable states of gene expression in eukaryotes. But, in the first instance, there would seem to be no such special requirement: lambda lysogens are essentially infinitely stable in the absence of the specific signal that inactivates the repressor, and this system — of course — involves no histones. Beyond that, there is evidence against the idea that special irreversible mechanisms are required in higher organisms. Thus, for example, fusion of a human muscle cell with a rat fibroblast produces a ‘cybrid’ which expresses rat muscle genes, a kind of experiment that has been repeated with an array of different cell types [20Blau H. Blakely B. Plasticity of cell fate: Insights from heterokaryons.Cell Dev. Biol. 1999; 10: 267-272Crossref Scopus (56) Google Scholar]. The surmise is that these highly differentiated cells bear, in their cytoplasms, factors that turn on (or in other cases off) transcription of crucial genes, and there is no special ‘lockdown’ that prevents these specific DNA-binding proteins from working. And, as already noted, there are several examples of inactivated genes being reactivated by strong transcriptional activators in an array of eukaryotes. Whatever elaborations might be discovered, we have one clear picture for how patterns of gene expression are usually maintained as cells divide. In eukaryotes, as in bacteria, regulatory proteins (typically specific DNA-binding proteins) are distributed to progeny cells, and in the absence of changed signals (or new transcriptional programs initiated by the inherited regulators) these patterns, reinforced by feedback loops, will be maintained. Patterns of histone modification are often strongly correlated with patterns of inherited gene expression, and it has often been assumed that these modifications must cause them — hence the questionable use of the term ‘epigenetic’ to describe all such modifications. However, there are no examples where these modifications have been shown to be self-propagating, and there are explicit examples where these modifications are shown not to be self-propagating. Histone modifiers can play roles in gene expression, but they (as well as enzymes that trigger DNA methylation, in some cases at least) must be recruited to genes by specific DNA binding proteins. This general description is of course not complete: in mammals, for example, the phenomena of X-chromosome inactivation and imprinting show that genes can be maintained in an off state even in the presence of activators sufficient to turn on genes in homologous chromosomes (see also [21Xu E. Zawadzki K. Single-cell observations reveal intermediate transcriptional silencing states.Mol. Cell. 2006; 23: 219-229Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar, 22Grewal S. Klar A. Chromosomal inheritance of epigenetic states in fission yeast during mitosis and meiosis.Cell. 1996; 86: 95-101Abstract Full Text Full Text PDF PubMed Scopus (182) Google Scholar] for examples of similar phenomena in yeast). How these special cases arise and are maintained is not yet understood. I thank, for help, Alex Gann, Sandy Johnson, Marc Gartenberg, Dan Gottschling, Steve Henikoff, Mary Anne Osley, Nicole Francis, Tim Bestor, Danesh Moazed, Bob Weinberg, Rob Martienssen, Kevin Struhl, Michael Grunstein and members of my lab." @default.
• W2069536617 created "2016-06-24" @default.
• W2069536617 creator A5082391932 @default.
• W2069536617 date "2007-04-01" @default.
• W2069536617 modified "2023-10-11" @default.
• W2069536617 title "On the use of the word ‘epigenetic’" @default.
• W2069536617 cites W1658331573 @default.
• W2069536617 cites W1915481228 @default.
• W2069536617 cites W1981354816 @default.
• W2069536617 cites W1997818262 @default.
• W2069536617 cites W2008284328 @default.
• W2069536617 cites W2010383826 @default.
• W2069536617 cites W2018822795 @default.
• W2069536617 cites W2029940868 @default.
• W2069536617 cites W2040249554 @default.
• W2069536617 cites W2047500542 @default.
• W2069536617 cites W2069233620 @default.
• W2069536617 cites W2073869073 @default.
• W2069536617 cites W2085090468 @default.
• W2069536617 cites W2103009708 @default.
• W2069536617 cites W2108839057 @default.
• W2069536617 cites W2109546612 @default.
• W2069536617 cites W2146470265 @default.
• W2069536617 cites W2161976991 @default.
• W2069536617 cites W2163550224 @default.
• W2069536617 cites W2171256063 @default.
• W2069536617 doi "https://doi.org/10.1016/j.cub.2007.02.030" @default.
• W2069536617 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/17407749" @default.
• W2069536617 hasPublicationYear "2007" @default.
• W2069536617 type Work @default.
• W2069536617 sameAs 2069536617 @default.
• W2069536617 citedByCount "315" @default.
• W2069536617 countsByYear W20695366172012 @default.
• W2069536617 countsByYear W20695366172013 @default.
• W2069536617 countsByYear W20695366172014 @default.
• W2069536617 countsByYear W20695366172015 @default.
• W2069536617 countsByYear W20695366172016 @default.
• W2069536617 countsByYear W20695366172017 @default.
• W2069536617 countsByYear W20695366172018 @default.
• W2069536617 countsByYear W20695366172019 @default.
• W2069536617 countsByYear W20695366172020 @default.
• W2069536617 countsByYear W20695366172021 @default.
• W2069536617 countsByYear W20695366172022 @default.
• W2069536617 countsByYear W20695366172023 @default.
• W2069536617 crossrefType "journal-article" @default.
• W2069536617 hasAuthorship W2069536617A5082391932 @default.
• W2069536617 hasBestOaLocation W20695366171 @default.
• W2069536617 hasConcept C104317684 @default.
• W2069536617 hasConcept C138885662 @default.
• W2069536617 hasConcept C41091548 @default.
• W2069536617 hasConcept C41895202 @default.
• W2069536617 hasConcept C54355233 @default.
• W2069536617 hasConcept C70721500 @default.
• W2069536617 hasConcept C78458016 @default.
• W2069536617 hasConcept C86803240 @default.
• W2069536617 hasConcept C90805587 @default.
• W2069536617 hasConceptScore W2069536617C104317684 @default.
• W2069536617 hasConceptScore W2069536617C138885662 @default.
• W2069536617 hasConceptScore W2069536617C41091548 @default.
• W2069536617 hasConceptScore W2069536617C41895202 @default.
• W2069536617 hasConceptScore W2069536617C54355233 @default.
• W2069536617 hasConceptScore W2069536617C70721500 @default.
• W2069536617 hasConceptScore W2069536617C78458016 @default.
• W2069536617 hasConceptScore W2069536617C86803240 @default.
• W2069536617 hasConceptScore W2069536617C90805587 @default.
• W2069536617 hasIssue "7" @default.
• W2069536617 hasLocation W20695366171 @default.
• W2069536617 hasLocation W20695366172 @default.
• W2069536617 hasOpenAccess W2069536617 @default.
• W2069536617 hasPrimaryLocation W20695366171 @default.
• W2069536617 hasRelatedWork W1828691184 @default.
• W2069536617 hasRelatedWork W1903732681 @default.
• W2069536617 hasRelatedWork W1991523530 @default.
• W2069536617 hasRelatedWork W2002128513 @default.
• W2069536617 hasRelatedWork W2020824267 @default.
• W2069536617 hasRelatedWork W2044401234 @default.
• W2069536617 hasRelatedWork W2057739827 @default.
• W2069536617 hasRelatedWork W2075354549 @default.
• W2069536617 hasRelatedWork W4310701073 @default.
• W2069536617 hasRelatedWork W2092874662 @default.
• W2069536617 hasVolume "17" @default.
• W2069536617 isParatext "false" @default.
• W2069536617 isRetracted "false" @default.
• W2069536617 magId "2069536617" @default.
• W2069536617 workType "article" @default.