- Shopping Bag ( 0 items )
A comprehensive review of recent work on chromatin and non-histone proteins, this book arises from the interactions of a multidisciplinary group of scientists involved in the study of acetylation. This area of research opens up new and exciting possibilities for drug design, and so the final chapters in the book examine some of the potential applications in the treatment of various diseases.
Yanming Wang, Wolfgang Fischle, Wang Cheung, Steven Jacobs, Sepideh Khorasanizadeh and C. David Allis
Laboratory of Chromatin Biology, The Rockefeller University, New York, New York 10021 and * Department of Biochemistry and Molecular Genetics, University of Virginia Health System, Charlottesville, Virginia 22908, USA
Abstract. Chromatin is the physiological carrier of not only genetic information, encoded in the DNA, but also of epigenetic information including DNA methylation and histone modifications. As such histone modifications are involved in many aspects of nuclear processes including gene regulation and chromosome segregation. Recently, a 'histone code' hypothesis was put forward to explain how patterns of histone modification may function in downstream processes. In support of the 'histone code' hypothesis, we found in vivo and in vitro evidence that effector proteins, HP1 (heterochromatin protein 1) and Pc (Polycomb) can discriminate and 'read' histone methylation marks on K9 and K27, respectively. Moreover, we propose a 'binary switch' model and suggest that binding and release of effector proteins to their cognate sites can be regulated by modifications on adjacent or nearby residues. Thus, combinations of adjacent histone modifications would function differently from singular modification, and static modifications(e.g. Lys methylation) may well be regulated by dynamic modifications (e.g. phosphorylation). Finally, we describe a novel histone phosphorylation event linking the function of Mst1 kinase and H2B Ser14 phosphorylation with apoptotic chromatin condensation in vertebrates. As this modification is not found during mitotic chromosome condensation, these findings suggest the intriguing possibility that a unique 'death' mark exists for chromatin condensation during apoptosis. 2004 Reversible protein acetylation. Wiley, Chichester (Novartis Foundation Symposium 259) p 3-21
The human genome is estimated to contain 30 000-40 000 unique genes. The DNA sequence and the chromatic location of most of these genes has been determined and they are publicly available (Lander et al 2001, Venter et al 2001). The central challenge now facing the biomedical community is how to derive valuable medical knowledge about the function of these genes from DNA sequence data, and to answer questions such as how the expression of these genes is orchestrated to carry out normal cellular functions and for responses to environmental and physiological changes. The genetic information encoded by the DNA sequence determines the sequence of RNAs and proteins. However, it is becoming clear that 'epigenetic' information plays a major role in determining when, where, and to what level the genetic information should be utilized. Epigenetic information refers to differential and inheritable changes of gene expression potentials that are not caused by mutations in DNA itself (Jaenisch & Bird 2003). Recent studies have focused on several molecular mechanisms of epigenetic gene regulation that include DNA methylation, histone modifications and small nuclear RNAs (SnRNAs) or RNA interference (RNAi). Whereas emerging evidence suggests that the three mechanisms are coordinated and affect each other, the discussion below will mainly focus on histone modifications as they relate to epigenetic-based forms of gene regulation.
Histone proteins as 'messengers' of epigenetic information
The human body contains multiple organs and diverse cell types, and every gene exists within every cell. However, only a small percentage of genes are activated in any given cell type, and each type of cell has its unique gene expression profile. These different expression profiles are formulated during early development in a multicellular organism, when cell division, cell differentiation, tissue and organ formation occur rapidly (Francis & Kingston 2001). Moreover, these gene expression potentials can be memorized and inherited after mitosis and even meiosis. To regulate this genetic information efficiently and in an epigenetic manner, nature has evolved a sophisticated system that controls access to specific genes. This system relies on packaging DNA into a DNA-histone complex called chromatin, which is the physiological substrate of all cellular processes involving the DNA (Felsenfeld & Groudine 2003). The dynamic change of the three dimensional architecture of chromatin makes certain genes more readily accessible to transcription factors and other machineries engaging the genetic template (Lomvardas & Thanos 2002). Because parental DNA and associated histones are divided and incorporated into the newly duplicated chromatin during S phase of the cell cycle, it is possible that the epigenetic modifications carried by DNA and histones can be passed to the daughter cells after M phase and cell division, making DNA and histone proteins also the attractive messengers of epigenetic information (Fig. 1).
Recently, the chromatin field has witnessed an explosion of literature documenting the involvement of various histone modifications, such as methylation, phosphorylation, acetylation, ubiquitination and ADP ribosylation in essentially all DNA-templated processes. In the coming of a new epigenomic era, the regulation of the enzymes responsible for adding or subtracting these covalent marks are poised to take centre stage in the study of gene expression regulation, and understanding the molecular aetiology of human diseases such as cancer. Identification of altered DNA methylation and histone modification activities in a range of human cancers supports the involvement of epigenetic mechanisms during cancer development (Kondo et al 2003). Thus, it is important to investigate the role of epigenetic regulatory proteins and the way that epigenetic regulation works in order to get a more in-depth picture of pathways leading to oncogenesis and to assist the development of new therapeutic strategies.
New insights into the 'histone code' hypothesis
It is clear that the regulatory signals, either extracellular or intracellular, ultimately impinge on chromatin, which can be viewed as a gigantic signalling platform for integrating and recording these signalling events (Cheung et al 2000). The epigenetic information carried by the chromatin can in turn impact on most of the chromatin-templated processes with far-reaching consequences for cell fate decisions and for normal and pathological development (Jenuwein & Allis 2001, Fischle et al 2003). As mentioned above, epigenetic information is inheritable through the cell cycle and through meiosis from one sexual generation to the next. We and others have proposed that an epigenetic indexing system for our genome, a 'histone' or 'epigenetic' code, works as a fundamental regulatory mechanism in addition to the DNA and the genetic information itself (Strahl & Allis 2000, Turner 2000, Jenuwein & Allis 2001). The original histone code hypothesis proposed that 'distinct covalent histone modifications, acting alone, sequentially, or in combination, form a "histone code" that is then read by effector proteins to bring about distinct downstream events' (Strahl & Allis 2000). Although this hypothesis has received much attention and some strong experimental support (Agalioti et al 2002, Kanno et al 2004), it has been hard to derive definitive rules from our current knowledge of the 'code'.
In this meeting, I will expand on this general concept by proposing the 'methyl/ phos' (methylation/phosphorylation) switch hypothesis with 'predictive rules' that may govern the binding and release of effector proteins and complexes that engage the chromatin polymer. On the histone H3 tail, several clear examples of adjacent Lys residues and Ser/Thr residues exist, such as K9S10 and K27S28, that can be modified by methylation and phosphorylation, respectively (Fig. 2A). I will present recent work suggesting that methylation- and site-specific effector proteins exist, and their function is likely regulated by phosphorylation of adjacent residues (see below, Jacobs & Khorasanizadeh 2002, Fischle et al 2003a). Thus, these adjacent KT/S sites may form 'binary switches' to regulate the binding of effector proteins (Fig. 2B and discussion below). Importantly, our ideas provide an explanation for several long-standing questions embedded in the existing literature, and are open to experimental tests. In addition, I will present a newly discovered histone phosphorylation event and the responsible kinase, which link histone modification to apoptotic chromosome condensation.
Molecular basis for discrimination of repressive methyl-lysines in the histone H3 tail
On the histone H3 tail, lysines 9 and 27 are well-known methylation sites, and are often associated with epigenetic repression (Lys 27) and heterochromatinmediated gene silencing (Lys 9) (Cao et al 2002, Jacobs et al 2001). Although these two sites are involved in different epigenetic events, it is remarkable that both 'target' lysines are embedded within a highly related sequence motif: TAR[K.sup.9]S versus AAR[K.sup.27]S (Fig. 2A). Moreover, as predicted by the histone code hypothesis, emerging evidence shows that Lys9 and Lys27 methylation sites are 'read' by distinct effector binding proteins: heterochromatin protein 1 (HP1) and Polycomb (Pc), respectively. Both HP1 and Pc are the prototype proteins in which the chromodomain was identified (Singh et al 1991). Recent work suggests that chromodomains serve as methyl-lysine recognition and binding modules (Jacobs & Khorasanizadeh 2002, Nielsen et al 2002).
Our knowledge of the organization and function of heterochromatin has been greatly advanced by the study of HP1 and the histone H3 K9 methyltransferase, Su(var)3-9. The fact that the chromodomain of HP1 can recognize and bind the K9 methyl site generated by the Su(Var)3-9 offers a mechanistic insight on the epigenetic gene silencing phenomena associated with heterochromatin, such as position effect variegation (PEV) (Lachner et al 2001, Bannister et al 2001, Jacobs et al 2001, Rea et al 2000). On the epigenetic gene expression side, recent studies found that the E(z) (Enhancer of Zeste) complex can methylate histone H3 K27, and H3 K9 to a lesser extent, in in vitro enzymatic assays (Cao et al 2002, Czermin et al 2002, Kuzmichev et al 2002, Muller et al 2002). Likewise, the chromodomain of Pc was proposed to be able to 'read' both of these methylation sites. However, these promiscuous activities are paradoxical to the in in vivo observation that Pc and HP1 are involved in different pathways, and that the chromodomain of HP1 and Pc is responsible to target these protein to different destination in the nucleus (Messmer et al 1992, Platero et al 1995).
Here, we present new data to show that the chromodomain proteins Pc and HP1 are highly discriminatory for binding to these sites both in vivo and in vitro. Using newly developed methyl- and site-specific antibodies, we showed that trimethyl-Lys27 and Pc are colocalized and both excluded from heterochromatic areas that are enriched in di- and trimethyl-Lys9 and HP1 in Drosophila S2 cells and on polytene chromosomes. In addition, swapping of the chromodomain regions of Pc and HP1 is sufficient for switching the nuclear localization patterns of these repressors, indicating a role for their chromodomains in both target site binding and discrimination (Fischle et al 2003b).
To better understand the molecular basis for the selection of methyl-Lys binding sites, we have recently solved the 1.8A structure of the Pc chromodomain in complex with a trimethyl-Lys27 H3 tail and compared it with our previously determined structure of the HP1 chromodomain complexed with a trimethyl-Lys9 H3 tail (Jacobs & Khorasanizadeh 2002). The structures show clear differences in how two chromodomains that are highly related in sequence and structure effectively distinguish methylation sites on the H3 tail (Fischle et al 2003). Whereas both the HP1 and the Pc chromodomains form aromatic 'cages' that bind the positively charged methylammonium ion, they distinguish the two binding sites by discriminating residues N-terminal to the common ARKS motif, which differ between the two target sites. The Pc chromodomain has evolved an extended 'groove' that provides more contact surfaces to engage five more residues to a modest degree in addition to the AR[K.sup.27]S. Together, these surfaces provide enough additional binding complimentarily to generate enhanced recognition and binding anity (Fischle et al 2003b). On the HP1 side, it seems the residue T6 in front of AR[K.sup.9]S is critical for the binding specificity (Jacobs & Khorasanizadeh 2002, Fischle et al 2003).
HP1 and Pc proteins themselves have been implicated in fundamental nuclear processes including heterochromatin-mediated gene silencing, homeotic gene expression and chromosome dynamics (Simon & Tamkun 2002). The above studies of HP1 and Pc offer supportive evidence to a central tenet of the 'histone code hypothesis', that the covalent marks are docking sites for effector proteins that in turn bring about distinct downstream events (Strahl & Allis 2000). It is quite intriguing from an evolutionary aspect that the two effector proteins with the similar functional domain can recognize two binding sites embedded in the similar sequence context, and are evolved to participate in two different silencing pathways.
Binary switches as part of the histone code?
The density of modifiable residues on the histone tail, for example H3, is very striking (Fig. 2A). Recently, mass spectrometry analyses suggest that modification of two adjacent sites does coexist (C. D. Allis, D. F. Hunt, unpublished data). Many site- and modification-specific antibodies have been developed and have greatly benefited the field to tackle the histone modification problem. However, as many immunological tools were developed against specific histone modification sites, a recurring question to us and others is whether adjacent histone modifications might affect the epitope recognition by antibodies. Similarly, it is equally intriguing to know whether adjacent modifications, if they exist, may affect the binding of effector proteins that normally recognize single modifications, such as HP1 and Pc.
On the basis of the presence of close dual modification sites on the histone tails, we wish to extend the histone code hypothesis and to propose the concept of 'binary switches' (Fig. 2B). We hypothesize that binary switches in the histone tails regulate the 'ON/OFF' state for the binding of effector proteins, such as HP1 and Pc.
Excerpted from Reversible Protein Acetylation, No. 259 Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Chair's Introduction (E. Verdin).
Beyond the double helix: writing and reading the histone code (Y. Wang, et al.).
The indexing potential of histone lysine methylation (G. Schotta, et al.)
A model for step-wise assembly of heterochromatin in yeast (D. Moazed, et al.).
H2B Ubiquitylation and deubiquitylation in gene activation (A. Wyce, et al.).
Structural and chemical basis of histone acetylation (R. Marmorstein).
Phosphorylation and acetylation of histone H3 at inducible genes: two controversies revisted (L. Mahadevan, et al.).
HDAC7 regulates apoptosis in developing thymocytes (E. Verdin, et al.).
Dual roles of histone deacetylases in the control of cardiac growth (T. McKinsey and E. Olson).
Chromatin modifications as clues to the regulation of antigen receptor assembly (D. Ciccone and M. Oettinger).
General discussion I Histone modifications in X inactivation.
The HDAC complex and cytoskeleton (J. Kovacs, et al.).
Tat acetylation: a regulatory switch between early and late phases in HIV transcription elongation (M. Ott, et al.).
Dynamics of the p53 acetylation pathway (W. Gu, et al.).
Regulation of the NF-kB action by reversible acetylation (W. Greene and L. Chen).
General discussion II p300 and DNA repair.
Reversal of gene silencing as a therapeutic target for cancer - roles for DNA methylation and interdigitation with chromatin (S. Baylin).
Transcription regulation by histone deacetylases (S. Wang, et al.).
Molecular and cellular basis for the anti-proliferative effects of the HDAC inhibitor LAQ824 (P. Atadja, et al.).
Histone deacetylase inhibitors: development as cancer therapy (P. Marks, et al.).
General discussion III PML-RAR hypermethylation in leukemia.
Index of Contributors.