当前位置:首页 >> 生物学 >>

Histone modification patterns and epigenetic codes


Biochimica et Biophysica Acta 1790 (2009) 863–868

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b b a g e n

Review

Histone modi?cation patterns and epigenetic codes
Andreas Lennartsson, Karl Ekwall ?
Department of Biosciences and Medical Nutrition, Karolinska Institutet, NOVUM, 14157 Huddinge, Sweden

a r t i c l e

i n f o

a b s t r a c t
The eukaryotic DNA is wrapped around histone octamers, which consist of four different histones, H2A, H2B, H3 and H4. The N-terminal tail of each histone is post-transcriptionally modi?ed. The modi?cation patterns constitute codes that regulate chromatin organisation and DNA utilization processes, including transcription. Recent progress in technology development has made it possible to perform systematic genome-wide studies of histone modi?cations. This helps immensely in deciphering the histone codes and their biological in?uence. In this review, we discuss the histone modi?cation patterns found in genome-wide studies in different biological models and how they in?uence cell differentiation and carcinogenesis. ? 2009 Elsevier B.V. All rights reserved.

Article history: Received 29 August 2008 Received in revised form 22 December 2008 Accepted 29 December 2008 Available online 8 January 2009 Keywords: Epigenetic Chromatin Histone code Histone modi?cation Cell differentiation Cancer

1. Introduction Histones are the small basic proteins that together with DNA form chromatin structures in the cell nucleus. Until the beginning of the 1990's histones were generally considered to be packaging material for the DNA with no role in regulation of the genes [1]. Since then, it has become very clear that histones play important roles not only in regulation of gene expression but also in DNA damage repair, DNA replication and recombination. It is also becoming clear that histones are key players in epigenetic regulation i.e. regulation of heritable chromatin states, for example, clonally heritable states of gene expression that are not encoded in the DNA itself. This also makes histones crucial regulators of the cell differentiation process. Histones are covalently modi?ed in many ways, including phosphorylation, ubiquitination, acetylation and methylation. According to the histone code hypothesis, ‘distinct histone modi?cations, on one or more tails, act sequentially or in combination to form a ‘histone code’ that is read by other proteins to bring about distinct downstream events’ [2]. Such histone codes may be transient in nature or more stable, in which case they constitute true epigenetic codes that can be inherited [3]. In this review, we focus on recent studies that map several different histone modi?cations in a genome to systematically search for histone codes. Such studies have been performed on different genetic models e.g. yeast, Drosophila and different human cell types. We compare and discuss the histone modi?cation patterns that have been found in these different species. Finally, we address how histone modi?cation patterns are involved in the regulation of cell differentiation and disease.
? Corresponding author. Tel.: +4686089133; fax: +4687743358. E-mail address: karl.ekwall@ki.se (K. Ekwall). 0304-4165/$ – see front matter ? 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.bbagen.2008.12.006

2. Dynamics and complexity of histone codes Histones are modi?ed at several different amino acid residues and with many different modi?cations. For example, peptide mass ?ngerprinting mass spectrometry of calf histones that detected methylation, acetylation, phosphorylation and ubiquitination revealed 13 modi?cation sites in histone H2A, 12 modi?cation sites in histone H2B, 21 modi?cation sites in histone H3 and 14 modi?cation sites in histone H4 [4]. Regarding combinations of combinations, each site can, of course, either be unmodi?ed or modi?ed. In addition, some lysine residue can either be methylated or acetylated and there are three different possibilities for each methylated site (mono, di or tri). Therefore, the number of possible combinations of histone modi?cations is enormous. Histone codes may be transiently altered by the cell environment [5]. The existence of such transient histone codes is the result of the exact cell physiological state and surrounding signals, etc. that can vary over time. A heritable histone code is de?ned as an epigenetic code [3]. Heritable in this context means a code that is passed on from one cell to another i.e. a histone code that is somatically heritable within different tissues of a multicellular organism. A pioneering study on the behaviour of histone modi?cation patterns throughout the cell cycle is that of the HNF-1, HNF-4 and albumin genes in hepatocytes [6]. Cells were synchronised and a cell cycle time-course experiment was conducted which revealed that H3K4me2/me3, H3K79me2, H3 and H4 acetylation remained stable over time at the promoters of these genes. Furthermore, the maintenance of these “active” histone modi?cation patterns did not require transcription and the chromatin state was preserved in cells blocked in mitosis. This suggests that some modi?cation patterns are indeed epigenetic since they are transferred from one cell to another on mitotic chromosomes. Further support to this notion came from the ?ndings that in HeLa cells GAPDH, IL-2,

864

A. Lennartsson, K. Ekwall / Biochimica et Biophysica Acta 1790 (2009) 863–868

hsp70 promoters maintain H3K4me2/me3 as well as H3 and H4 acetylation in mitotic cells [7] and from studies of the β-globin locus in Murine erythroleukemia cells which maintain H3K79me, H3K4me2, H3 and H4 acetylation in mitosis [8]. How these patterns at active gene promoters are replicated and maintained is not clear, but the ?nding that at the cyclin B1 promoter H3K4me3 is increased in mitosis [7] argues that K-methyltransferases (KMT) are active at this point in the cell cycle in HeLa cells. A mechanism that propagates heterochromatic (H3 and H4 deacetylated and H3K9me) histone modi?cation patterns was recently discovered in ?ssion yeast. During each S phase, this pattern gets re-established by a cell cycle dependent mechanism that involves a pulse of transcription by RNA pol II and subsequent recruitment of histone modifying enzymes via the RNAi interference pathway [9]. Although it remains to be demonstrated, it is plausible that similar mechanisms could contribute to heritable repressed histone codes at individual gene promoters. 3. Finding and understanding histone modi?cation patterns? Today, the only technique that enables detection of histone modi?cation patterns at a given locus is chromatin immunoprecipitation (ChIP). In this method, a chromatin extract is prepared and antibodies speci?c for each histone modi?cation is used for immunoprecipitation of chromatin that carries the modi?cation in question. By carrying out several parallel ChIP experiments from the same extract, one can then draw conclusions regarding the combinations that existed in that chromatin region in vivo. It is, of course, important to use the same chromatin extracts for all antibodies to obtain a snapshot view, otherwise transient codes may escape detection. By adding a micrococcal nuclease step, the smallest unit of chromatin i.e. individual nucleosomes can also be analysed (MNase-ChIP). To systematically search for histone modi?cation patterns, genome wide applications of ChIP have been used, such as ChIP-chip (microarrays) and ChIP-seq (Solexa sequencing). It is important to remember that the obtained data represents the average situation in the cell population that was used as starting material. However, it does not tell us whether the modi?cations existed on the same histone molecule or the same nucleosome at the same time since millions of cells are typically used for each experiment. A useful validation of histone codes is therefore ‘topdown’ mass spectrometry approaches that allow detection of combinations on the same histone polypeptide [10]. For example, recently ChIP-chip in combination with quantitative mass spectrometry was used in ?ssion yeast to provide direct evidence for antagonistic effects of H3K36me and H3K27ac both at gene promoters in vivo and on isolated histone peptides (Buchanan, L., Sinha, I., R?nnerblad, M., Ekwall, K., Shevchenko, A. and Stewart, A.F., unpublished data).

To detect patterns in the genome wide datasets, different strategies have been used. The different approaches may suffer from intrinsic biases as discussed in [11]. Kurdistani et al. ChIP–chip mapped 11 acetylation sites in S. cerevisiae and used hierarchical clustering after variance normalization, which resulted in identi?cation of many different patterns. In contrast, in the study by Liu et al., mapping 12 modi?cations by MNase-ChIP–chip also used hierarchical clustering but this time in combination with principal component analysis, and only identi?ed two main patterns of histone modi?cation in budding yeast. The lack of concurrence between these studies could, in part, be due to the fact that different modi?cations were studied in each case or differences in the bioinformatics approaches used. Despite these differences, both studies do agree on the presence of both H3K9ac and H3K18ac in promoters of many active genes (Table 1). In a very comprehensive human T-cell ChIP-seq study, where as many as 39 different modi?cations were mapped, identi?cation of patterns was performed using sequence Tag count statistics, and resulted in the description of 13 main patterns present in more than 62 gene promoters in each [12]. Interestingly, taken together the ChIP–chip and ChIP-seq mapping studies carried out so far in yeast, Drosophila and human cells clearly show that only a small fraction of the myriad of possible modi?cation patterns actually exist in vivo. One approach to investigate the downstream events that histone codes represent is to use bioinformatics gene ontology search tools. The study of Kurdistani et al. identi?ed 53 different histone acetylation patterns in S. cerevisiae and 12 of these matched signi?cantly to gene ontology terms [13]. For example, the combination of H3K18ac, H3K9ac and H3K27ac (IGR cluster 1) was linked to genes involved in protein synthesis. In human T-cells, the gene ontology analysis showed that genes implicated in cellular physiology and metabolism are enriched among the active genes, that all have the backbone of 17 modi?cations and some additional modi?cations (Table 1; class III) whereas genes involved in cell signalling and other non T-cell functions such as development and synaptic signalling were enriched in the inactive promoters (Table 1; class I) [12]. Thus, both in yeast and in human T-cells it is clear that the most common histone modi?cation patterns found in both active and silent promoters are linked to groups of biologically related genes that are involved in the same cellular process as revealed by the gene ontology terms. In the study of Wang et al., a backbone of 17 histone modi?cations was detected at 3266 of 12,541 promoter regions. Many of the genes with this backbone are highly expressed in human T-cells. The backbone is a combination of the histone variant H2A.Z (which replaces H2A) and H2BK5ac, H2BK12ac, H2BK20ac, H2BK120ac, H3K4ac, H3K4me1, H3K4me2, H3K4me3, H3K9ac, H3K9me1, H3K18ac, H3K27ac, H3K36ac, H4K5ac, H4K8ac and H4K91ac. Inter-

Table 1 A comparison of histone modi?cation patterns in gene promoters in different species Gene expression Human CD4+ T-cells [12] CHIP-seq; Single single nucleosome mapping III; high B,H2AK9ac,H2BK5me1,H3K79me1,me2, me3,H4K12ac,H4K16ac,H4K20me1:62a B,H2AK9ac,H2BK5me1,H3K79me1,me2, me3,H4K16ac,H4K20me1:68a B,H2BK5me1,H3K79me1,me2,me3, H4K16ac,H4K20me1:74a B,H4K16ac:67a B:64a H3K36me3:135a H2AZ,H3K4me1,me2,me3,H3K9me1, H3K27me3:77a H2AZ,H3K4me2,me3, H3K9me1,H3K27me3:70a H2AZ,H3K4me3, H3K27me3:74a H3K4me3,H3K27me3:167a H3K4me3:85a H3K27me3:630a Human ES cells Flies [61] CHIP–chip [60] CHIP–chip H3K9ac, H3K14ac and H3K4me3 Budding yeast [13] CHIP–chip (only a few representative patterns are shown) Budding yeast [14] CHIP–chip; Single single nucleosome mapping

H3K18ac, H4K12ac, Hyperacetylated for H3 H3K18ac, H3K9ac, H3K27ac (IGR cluster 1) H3K18ac, H3K27ac (IGR cluster 2) H3K18ac, H3K9ac, H3K14ac H4K5ac and H4 and H3K4me H3K9ac, H2AK7ac (IGR cluster 37) H3K18ac, H3K4me3 H3K79me H3K23ac, H3K27ac (IGR cluster 45)

II; intermediate I; low

H3K4me3 low levels

Inactive genes being hypomethylated and deacetylated at the same residues

H4K8ac, H4K16ac, H2BK11ac, H2BK16ac (IGR cluster 6) H3K23ac, H2BK16ac (IGR cluster 29) H4K16ac (IGR cluster 30) H4K8ac, H4K16ac H2AK7ac (IGR cluster 46)

Inactive genes being hypomethylated and deacetylated at the same residues

B = backbone of H2A.Z, H2BK5ac, H2BK12ac, H2BK20ac, H2BK120ac, H3K4ac, H3K4me1, H3K4me2, H3K4me3, H3K9ac, H3K9me1, H3K18ac, H3K27ac, H3K36ac, H4K5ac, H4K8ac and H4K91ac. a The number of promoters carrying the combination of histone modi?cations.

A. Lennartsson, K. Ekwall / Biochimica et Biophysica Acta 1790 (2009) 863–868

865

estingly, a part of this backbone e.g. H3K18ac, H3K9ac, H3K27ac and the histone variant H2A.Z is also found in chromatin of many gene promoter genes in yeast, including promoters of highly expressed genes [13–15]. A systematic cross-species comparison of histone modi?cation patterns is currently hampered by the limited number of modi?cations that have been mapped in all systems (Table 1). Nevertheless, it seems as if the histone modi?cation patterns are at least partially conserved from yeast to human. The presence of H2A.Z at active gene promoters in combination with at least three speci?c H3 acetylation sites, at K9, K18 and K27, may be of special importance for gene activation [15,12]. It is also clear from Table 1 that repressed genes generally lack acetylation of H3 and H4 except H4K8 and H4K16, which are acetylated at some repressed gene promoters in yeast. Instead, other species speci?c modi?cations patterns seem to exist, for example, H3K27me is found at repressed genes in human cells but is not present in yeast. 4. The role of histone codes in cell differentiation Regulation of histone modi?cations plays an important role in the control of many physiological processes, including cell differentiation. No systematic genome wide study on histone modi?cations has been reported that compares differently maturated cell populations. However, speci?c histone modi?cations have been mapped in various systems. Most focus has been on post-transcriptional modi?cations on histones 3 and 4. The histone modi?cations on H2A and H2B were recently reviewed by Wyrick and Parra [16]. It has been suggested that a global high level of histone acetylation is important to maintain cells in an undifferentiated and pluripotent state. Several studies have demonstrated that treatment with HDAC inhibitors reduces differentiation and induces expansion of the stem cell/progenitor pool [17,18]. However, treatment of primary cells [19] or cancer cells [20,21] with HDAC inhibitor has been shown to induce differentiation. Thus, it seems like the effect of HDAC inhibitors vary with cell type and context. Brain development and memory formation are also regulated with epigenetic mechanisms [22]. Hence, epigenetic regulation may be a general mechanism for cell maturation. Which speci?c histone modi?cations or combinations of modi?cations control the maturation process? As mentioned above, a high histone acetylation level has been suggested to be important for maintaining an undifferentiated state in cells. In contrast, the levels of the repressive marks K3K9me2/3 increase during embryogenesis [17,23] (Fig. 1). However, a special set of developmental genes needs to be repressed to maintain pluripotency. This repression is regulated by PolyComb (PcG) complexes, which associate with H3K27me3 [24]. Hence, PcG proteins are required for normal cell development

[24–26]. H4K20 methylation is also linked to transcriptional silencing. However, recently it was suggested that trimethylation of H4K20 is not directly involved in the regulation of gene expression, but plays a crucial role in DNA damage control [27]. Nevertheless, H4K20 methylation is proposed to vary during cell differentiation. In proliferating neural cells from mouse embryo, the level of H4K20me1 is high, while H4K20me3 is low. H4K20me3 increases progressively throughout development [28]. In mouse embryonic stem cells (ES), many developmental regulatory genes have “bivalent domains”, consisting of the transcription repressing H3K27me3 and transcription activating H3K4me3 [29]. During cell differentiation, the bivalent imprint disappears and at individual genes, either H3K27me3 or H3K4me3 remains (Fig. 1). The bivalently marked genes have low expression, thus H3K27me3 seems to be the dominant mark. H3K27me3 silences developmental genes in ES, while H3K4me3 keeps the genes poised for activation [29]. Kimura et al. also found that both transcribed and silenced genes in ES cells have H3K4me3 [30]. Hence, H3K4me3 seems to lead to a transcriptionally activated permissive state, but does not necessarily lead to transcription. 5. Context effects of histone modi?cations When colorectal cancer cells are treated with the cytidine methylation blocker, 5-aza cytidine, transcription of human mutL homolog 1 (hMLH1) is induced. The induction resulted in the depletion of H3K9me1/2 and enhancement of H3K9ac and H3K4me2. However, the transcribed gene kept the silencing marks, H3K9me3 and H3K27me3 in its promoter [31]. Thus, in this context H3K27me3 was not a dominant mark. These results emphasize the importance of studying combinations of different histone modi?cations. The readout of a single modi?cation may be different in various epigenetic environments. Neighbouring histone modi?cations may affect each other's interaction with secondary effector proteins. This phenomenon is exempli?ed by phosphoswitch [32] and methylation of H3K79 by the KMT Dot1 [33,34]. Studies in yeast have demonstrated that H4K16ac prevents binding of the repressor Sir3, while it allows binding of Dot1. Not only binding, but also H2BK123 ubiquitination is required for Dot1 to become enzymatically fully active (reviewed in [35]). Hence, different histone modi?cations are connected in a complex network to regulate each other's outcome. H3K9me3 causes silencing through its interaction with the heterochromatin protein 1 (HP1). Phosphorylation of H3S10 evicts HP1 from H3K9me3. This implies that transcription is induced, although the repressive mark, H3K9me3, is still present. Accordingly, HP1 binding to H3K9me3 decreases when hMLH1 transcription is induced in colorectal cancer cells [31]. Not only the histone modi?cations themselves but also their locations have proven to be important. In acute leukemic cell lines and medulloblastomas, it was demonstrated that over-expressed oncogenes, such as MEIS1 and HOXA9, are enriched in H3K9ac, H3K4me2 and surprisingly also the repressive H3K9me3 mark [36]. However, H3K9me3 is located downstream of the transcription start site and not in the promoter region, where H3K9me3 often is concentrated in repressed genes [36]. In addition, H3K9me3 was found within transcribed regions in expressed genes [37]. The HP1 isoform, HP1γ, was shown to be associated with both H3K9me3 and phosphorylated RNA polymerase II in transcribed genes, suggesting a role for HP1γ in elongation of transcription. This demonstrates the dif?culty of drawing conclusions from global analysis of single histone modi?cations. The histone code needs to be read carefully in regard to both location and combinations of modi?cations. 6. Altered histone codes in diseases Histone modi?cations regulate many different physiological mechanisms. It is, therefore, not surprising that its deregulation is

Fig. 1. Histone modi?cations during cell differentiation. A global high histone acetylation level has been suggested to important to keep cells in a pluripotent and immature state. The histone acetylation seems to decrease during cell differentiation and instead repressive marks, such as H3K9me3 and H4K20me3 increase. Furthermore, many developmental regulatory genes have a “bivalent domain”, consisting of the transcriptional repressive H3K27me3 mark and the transcriptionally activating H3K4me3 mark. This keeps the transcription in a silent state. During cell maturation the mark on each individual gene becomes either H3K27me3 or H3K4me3.

866

A. Lennartsson, K. Ekwall / Biochimica et Biophysica Acta 1790 (2009) 863–868

involved in various complications and diseases, such as mental disorders [38,39] and malaria [40]. However, by far the most studied and well characterised “epigenetic disease” is cancer. Feinberg, Ohlsson and Henikoff, proposed that tumors develop in three stages [41]. a) Epigenetic perturbation of stem/progenitor cells, mediated by an aberrant regulation of tumor progenitor genes (TPG). TPG was de?ned as genes that mediate epigenetic expansion of progenitor cells and increase their cancer proneness and their ability for self-renewal and pluripotency. b) Genetic changes in tumor suppressor and oncogenes. c) Genetic and epigenetic instability, implying increased tumor evolution. The order of events still needs to be further investigated, but it is clear that epigenetic rearrangements play a major role in tumor development. DNA methylation and various histone modi?cations contribute to silence tumor suppressor genes or destabilize the genome, thereby facilitating further chromosomal rearrangement and deregulation and leading to cancer. Embryonal carcinoma cells differ from normal ES cells in that they add two repressive markers, H3K9me2/3, to the bivalent state [42]. The added marks make the promoters prone to become hypermethylated and silenced. When key tumor suppressor genes are silenced by the additional modi?cation, a tumor is more likely to develop. The silencing mechanism is further enhanced by diminished H3K4 methylation in adult cancer cells; thus, the genes obtain a fully repressive state [43]. The tumor suppressor p16INKA and p14ARF loci are associated with H3K9 hypermethylation and H3K4 hypomethylation, and are silenced in some cancers [44,45]. A study in gastric adenocarcinoma showed that H3K9me3 positively correlated with tumor stage, lymphovascular invasion and cancer reoccurrence [46]. The importance of combinatory effects of histone modi?cations in cancer prognosis was shown by Seligson et al. [47] (Table 2). They analysed global levels of ?ve different histone modi?cations in prostate cancer, H3K18ac, H3K9ac, H3K4me2, H4K12ac and H4R3me2, all associated with active transcription. When they looked at each modi?cation separately, no prognostic tendency could be discerned. However, when analysing them collectively two different groups were de?ned with clearly different prognoses. Especially H3K4me2 together with H3K18ac showed a high prognostic value. The group with lower levels of H3K4me2 and H3K18ac had poorer prognoses than the patients with higher levels of these histone modi?cations [47]. Moreover, it has been demonstrated that the level of histone modi?cations in prostate cancer differ between benign, pre-neoplastic and cancer cells [48]. Thus, much more detailed and precise information can be obtained when using a combinatory approach. The promising results of using global analysis of histone modi?cations for cancer prognosis has been further highlighted in other reports (Table 2). A general reduction of H4K16ac and H4K20me3 has been demonstrated during tumor development in a wide array of cancer types [49]. A loss of H4K20me3 impairs DNA damage control

Table 2 Global histone modi?cation patterns that correlate with cancer Global increase of combinatory or single histone modi?cations H3K9me2/3 H3K4me H4K16ac and H4K20me3 H4K20me3 Global decrease of combinatory or single histone modi?cations Cancer type

H3K9me3 High levels of H3K4me2, H2AK5ac and H3K9ac individually correlate with good prognosis

Embryonal carcinoma cells [42] adenocarcinoma, [46] Many cancer types [43] Many cancer types [49,54] Hepatocarcinogenesis [55] Non-small-cell lung cancer (NSCLC) [57]

Low levels of H3K4me2 and H3K18ac correlate bad prognosis

Prostate cancer [47]

and may thereby increase the risk for mutations and cancer development. H4K16ac seems to be a unique histone mark. It modulates high order chromatin structures and functional interactions between non-histone proteins and chromatin [50]. Deacetylation of H4K16 is associated with transcriptional activation in budding and ?ssion yeast [13,51] and at least in ?ssion yeast, deacetylation was shown to be in the coding region [51]. In Drosophila is the localisation of H4K16ac regulated by a crosstalk with H3K36me2/3 [52]. This interaction has been proposed to enable dynamic regulation of chromatin condensation during transcriptional elongation [52]. However, in dosage compensation activates H4K16ac transcription [53]. The effect of H4K16ac seems hence, to vary in different chromatin environments. As discussed above, H4K16ac participates in networks, involving also other histone modi?cations, which regulate the biological outcome. The reduction of H4K16ac and H4K20me3 seems to be an early event that progresses throughout tumor development in many different cancer types [49,54]. The reduction of H4K16ac and H4K20me3 mostly occurs in repetitive DNA sequences in association with global loss of DNA methylation, often seen in cancer [49,55]. The reduced H4K16ac was suggested to be caused by a loss of the Kacetyltransferases KAT6A (MOZ), KAT6B (MORF) and KAT8 (MOF), at repetitive sequences. Cells with reduced H4K16ac demonstrate an increased genomic instability [56], which facilitates mutations and chromosomal rearrangements that may cause cancer. Since the more malignant human breast cancer cell line MDA-MB-231 has globally a more severe loss of H4K20me3 than the less malignant MCF-7, it has been suggested that the reduction of H4K20me3 correlates with increased malignant properties [54]. A loss of H4K20me3 was also observed early in hepatocarcinogenesis, followed by an increase of H3K9me3 at a later stage of the tumor development [55]. Those studies support that perturbed epigenetics is an early event in tumor development. The decreased histone acetylation seems to be speci?c for H4K16, since H4K5, H4K8 and H4K12 were not hypoacetylated. Therefore, Fraga et al. suggested that the loss of H4K16ac and H4K20me3 could be considered as an almost universal epigenetic marker for cancer, in a similar way to global DNA hypomethylation and CpG island hypermethylation [49]. Park et al. reported recently that the prognosis of gastric adenocarcinoma did not seem to be signi?cantly affected by the level of H4K20me3, but H3K9me3 correlated with lower survival rate [46]. H4K16ac only demonstrated a weak tendency to correlate with better prognosis [46]. Thus, the strong correlation between reduced H4K20me3, H4K16ac and cancer [47] was not discovered in gastric adenocarcinoma [46]. The complexity of epigenetic regulation and cancer was further emphasized by the ?nding that thymic lympohomas have globally increased levels of H4K16ac compared to normal tissues [56]. This contradicts previous results, which showed decreased H4K16ac in cancer cells [49]. In our opinion, the H4K16ac data demonstrates that cancer is a very heterogeneous disease and further studies are required before we can fully bene?t from the opportunities that the histone codes provide us for therapy and prognostic purpose. Also, other histone modi?cations have shown promising prognostic value. In a recent study in non-small-cell lung cancer (NSCLC), the levels of H3K4me2, H2AK5ac and H3K9ac could be used to predict the prognosis in different subgroups of NSCLC [57]. In different subgroups, individual high level of H3K4me2, H2AK5ac or H3K9ac correlates with a higher survival rate compared to patients with low levels of respectively modi?cation. H3K27me3, which often associates with PolyComb (PcG) complex binding, also shows potential to be used as a prognostic marker. PcG complex proteins are often upregulated in aggressive tumors, thus associating H3K27me3 and PcG with cancer [58]. Yu et al. showed that H3K27me3 marked genes in metastatic prostate cancer cells correspond to a high degree with PcG silenced genes in stem cells [59]. Fourteen of those genes were used to develop a “PcG repression signature” that could cluster the prostate cancer

A. Lennartsson, K. Ekwall / Biochimica et Biophysica Acta 1790 (2009) 863–868

867

patients into two groups, with good or bad prognosis. The signature was proven valid for multiple cancer types including breast cancer. Thus, despite that only a limited number of histone modi?cations have been characterized in cancer samples and without any detailed genomic localisation mapping, it is already clear that the epigenetic signature provided additional prognostic information beyond standard clinical and pathological methodologies. 7. Future directions Recent research has proven the power of using histone codes as prognostic markers and in therapy. To be able to discern the altered epigenetic patterns in various diseases, we ?rst need to better understand the role of histone codes and their consequences for gene regulation and genome stability in normal cells. Comprehensive genomewide mapping and analysis of histone modi?cations in model organisms is therefore of great importance. Such pioneer studies using ChIP-seq or ChIP–ChIP are starting to reveal the histone code [12,13,57,60,61]. To restore the altered histone codes in various diseases, we need to therapeutically target the corresponding epigenetic regulatory enzymes. An important goal is therefore to identify and determine the substrate speci?city of the enzymes that regulate histone modi?cations. Genome-wide analysis of histone deacetylase (HDAC) function has been carried out in yeast, reviewed in [62]. Such studies would be valuable also for other groups of modifying enzymes, e.g. K-Acetyltransferases (KATs), K-Methyltransferases (KMTs) and K-Demethylases (KDMs) and in other model organisms, e.g. mammalian systems. To date, of the modifying enzymes HDACs have received most attention in epigenetic therapy. The histone deacetylase inhibitor (HDACi) Valproic acid (VPA) is already used for treatment of various diseases. A signi?cant list of synthetic and natural HDACi exists and many have been shown to induce cell cycle arrest, growth inhibition, chromatin decondensation, differentiation and apoptosis in several cancer cell types [63]. Several HDACi show promising results in clinical trials for various cancer types and the HDACi Vorinostat is already approved by the FDA for treatment of cutaneous T-cell lymphoma. However, the HDACi on the market today have a very broad speci?city and target whole classes of HDACs. This complicates the usage of HDACi in therapy. VPA induces apoptosis in acute myeloid leukemic cells, but it also expands the leukemic stem cell population [64]. Hence, not until we have more knowledge of the speci?city of HDACs and other modi?ers will we be able to use the full potential of epigenetics in prognosis and tailored therapy. Acknowledgements K.E. is supported by the Swedish Cancer Society, Swedish Research Council (VR) and the EU ‘The Epigenome’ NoE network. We thank our group members for critical reading of this manuscript. References
[1] M. Grunstein, Histones as regulators of genes, Sci. Am. 267 (1992) 68–74B. [2] B.D. Strahl, C.D. Allis, The language of covalent histone modi?cations, Nature 403 (2000) 41–45. [3] B.M. Turner, Histone acetylation and an epigenetic code, BioEssays 22 (2000) 836–845. [4] L. Zhang, E.E. Eugeni, M.R. Parthun, M.A. Freitas, Identi?cation of novel histone post-translational modi?cations by peptide mass ?ngerprinting, Chromosoma 112 (2003) 77–86. [5] A. Bird, Perceptions of epigenetics, Nature 447 (2007) 396–398. [6] A. Kouskouti, I. Talianidis, Histone modi?cations de?ning active genes persist after transcriptional and mitotic inactivation, EMBO J. 24 (2005) 347–357. [7] E. Valls, S. Sanchez-Molina, M.A. Martinez-Balbas, Role of histone modi?cations in marking and activating genes through mitosis, J. Biol. Chem. 280 (2005) 42592–42600. [8] L. Xin, G.L. Zhou, W. Song, X.S. Wu, G.H. Wei, D.L. Hao, X. Lv, D.P. Liu, C.C. Liang, Exploring cellular memory molecules marking competent and active transcriptions, BMC Mol. Biol. 8 (2007) 31. [9] E.S. Chen, K. Zhang, E. Nicolas, H.P. Cam, M. Zofall, S.I. Grewal, Cell cycle control of

[10]

[11] [12]

[13] [14]

[15]

[16]

[17] [18]

[19]

[20]

[21]

[22] [23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35] [36]

[37]

centromeric repeat transcription and heterochromatin assembly, Nature 451 (2008) 734–737. B.A. Garcia, J. Shabanowitz, D.F. Hunt, Characterization of histones and their posttranslational modi?cations by mass spectrometry, Curr. Opin. Chem. Biol. 11 (2007) 66–73. F. van Leeuwen, B. van Steensel, Histone modi?cations: from genome-wide maps to functional insights, Genome Biol. 6 (2005) 113. Z. Wang, C. Zang, J.A. Rosenfeld, D.E. Schones, A. Barski, S. Cuddapah, K. Cui, T.Y. Roh, W. Peng, M.Q. Zhang, K. Zhao, Combinatorial patterns of histone acetylations and methylations in the human genome, Nat. Genet. 40 (2008) 897–903. S.K. Kurdistani, S. Tavazoie, M. Grunstein, Mapping global histone acetylation patterns to gene expression, Cell 117 (2004) 721–733. C.L. Liu, T. Kaplan, M. Kim, S. Buratowski, S.L. Schreiber, N. Friedman, O.J. Rando, Single-nucleosome mapping of histone modi?cations in S. cerevisiae, PLoS Biol. 3 (2005) e328. R.M. Raisner, P.D. Hartley, M.D. Meneghini, M.Z. Bao, C.L. Liu, S.L. Schreiber, O.J. Rando, H.D. Madhani, Histone variant H2A.Z marks the 5′ ends of both active and inactive genes in euchromatin, Cell 123 (2005) 233–248. J.J. Wyrick, M.A. Parra, The role of histone H2A and H2B post-translational modi?cations in transcription: a genomic perspective, Biochim. Biophys. Acta 1789 (2009) 37–44. J.H. Lee, S.R. Hart, D.G. Skalnik, Histone deacetylase activity is required for embryonic stem cell differentiation, Genesis 38 (2004) 32–38. L. De Felice, C. Tatarelli, M.G. Mascolo, C. Gregorj, F. Agostini, R. Fiorini, V. Gelmetti, S. Pascale, F. Padula, M.T. Petrucci, W. Arcese, C. Nervi, Histone deacetylase inhibitor valproic acid enhances the cytokine-induced expansion of human hematopoietic stem cells, Cancer Res. 65 (2005) 1505–1513. S. Snykers, T. Vanhaecke, A. De Becker, P. Papeleu, M. Vinken, I. Van Riet, V. Rogiers, Chromatin remodeling agent trichostatin A: a key-factor in the hepatic differentiation of human mesenchymal stem cells derived of adult bone marrow, BMC Dev. Biol. 7 (2007) 24. P.N. Munster, T. Troso-Sandoval, N. Rosen, R. Rifkind, P.A. Marks, V.M. Richon, The histone deacetylase inhibitor suberoylanilide hydroxamic acid induces differentiation of human breast cancer cells, Cancer Res. 61 (2001) 8492–8497. S. Minucci, C. Nervi, F. Lo Coco, P.G. Pelicci, Histone deacetylases: a common molecular target for differentiation treatment of acute myeloid leukemias, Oncogene 20 (2001) 3110–3115. J.M. Levenson, J.D. Sweatt, Epigenetic mechanisms: a common theme in vertebrate and invertebrate memory formation, Cell Mol. Life Sci. 63 (2006) 1009–1016. E. Meshorer, D. Yellajoshula, E. George, P.J. Scambler, D.T. Brown, T. Misteli, Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells, Dev. Cell 10 (2006) 105–116. T.I. Lee, R.G. Jenner, L.A. Boyer, M.G. Guenther, S.S. Levine, R.M. Kumar, B. Chevalier, S.E. Johnstone, M.F. Cole, K. Isono, H. Koseki, T. Fuchikami, K. Abe, H.L. Murray, J.P. Zucker, B. Yuan, G.W. Bell, E. Herbolsheimer, N.M. Hannett, K. Sun, D.T. Odom, A.P. Otte, T.L. Volkert, D.P. Bartel, D.A. Melton, D.K. Gifford, R. Jaenisch, R.A. Young, Control of developmental regulators by Polycomb in human embryonic stem cells, Cell 125 (2006) 301–313. A.P. Bracken, N. Dietrich, D. Pasini, K.H. Hansen, K. Helin, Genome-wide mapping of Polycomb target genes unravels their roles in cell fate transitions, Genes Dev. 20 (2006) 1123–1136. L.A. Boyer, K. Plath, J. Zeitlinger, T. Brambrink, L.A. Medeiros, T.I. Lee, S.S. Levine, M. Wernig, A. Tajonar, M.K. Ray, G.W. Bell, A.P. Otte, M. Vidal, D.K. Gifford, R.A. Young, R. Jaenisch, Polycomb complexes repress developmental regulators in murine embryonic stem cells, Nature 441 (2006) 349–353. S.L. Sanders, M. Portoso, J. Mata, J. Bahler, R.C. Allshire, T. Kouzarides, Methylation of histone H4 lysine 20 controls recruitment of Crb2 to sites of DNA damage, Cell 119 (2004) 603–614. V.L. Biron, K.J. McManus, N. Hu, M.J. Hendzel, D.A. Underhill, Distinct dynamics and distribution of histone methyl-lysine derivatives in mouse development, Dev. Biol. 276 (2004) 337–351. B.E. Bernstein, T.S. Mikkelsen, X. Xie, M. Kamal, D.J. Huebert, J. Cuff, B. Fry, A. Meissner, M. Wernig, K. Plath, R. Jaenisch, A. Wagschal, R. Feil, S.L. Schreiber, E.S. Lander, A bivalent chromatin structure marks key developmental genes in embryonic stem cells, Cell 125 (2006) 315–326. H. Kimura, M. Tada, N. Nakatsuji, T. Tada, Histone code modi?cations on pluripotential nuclei of reprogrammed somatic cells, Mol. Cell. Biol. 24 (2004) 5710–5720. K.M. McGarvey, J.A. Fahrner, E. Greene, J. Martens, T. Jenuwein, S.B. Baylin, Silenced tumor suppressor genes reactivated by DNA demethylation do not return to a fully euchromatic chromatin state, Cancer Res. 66 (2006) 3541–3549. W. Fischle, B.S. Tseng, H.L. Dormann, B.M. Ueberheide, B.A. Garcia, J. Shabanowitz, D.F. Hunt, H. Funabiki, C.D. Allis, Regulation of HP1-chromatin binding by histone H3 methylation and phosphorylation, Nature 438 (2005) 1116–1122. F. Frederiks, M. Tzouros, G. Oudgenoeg, T. van Welsem, M. Fornerod, J. Krijgsveld, F. van Leeuwen, Nonprocessive methylation by Dot1 leads to functional redundancy of histone H3K79 methylation states, Nat. Struct. Mol. Biol. 15 (2008) 550–557. R.K. McGinty, J. Kim, C. Chatterjee, R.G. Roeder, T.W. Muir, Chemically ubiquitylated histone H2B stimulates hDot1L-mediated intranucleosomal methylation, Nature 453 (2008) 812–816. B.M. Turner, Simplifying a complex code, Nat. Struct. Mol. Biol. 15 (2008) 542–544. J.K. Wiencke, S. Zheng, Z. Morrison, R.F. Yeh, Differentially expressed genes are marked by histone 3 lysine 9 trimethylation in human cancer cells, Oncogene 27 (2008) 2412–2421. C.R. Vakoc, S.A. Mandat, B.A. Olenchock, G.A. Blobel, Histone H3 lysine 9

868

A. Lennartsson, K. Ekwall / Biochimica et Biophysica Acta 1790 (2009) 863–868 methylation and HP1gamma are associated with transcription elongation through mammalian chromatin, Mol. Cell 19 (2005) 381–391. J. Peedicayil, The role of epigenetics in mental disorders, Indian J. Med. Res. 126 (2007) 105–111. D. Waggoner, Mechanisms of disease: epigenesis, Semin. Pediatr. Neurol. 14 (2007) 7–14. S.A. Ralph, A. Scherf, The epigenetic control of antigenic variation in Plasmodium falciparum, Curr. Opin. Microbiol. 8 (2005) 434–440. A.P. Feinberg, R. Ohlsson, S. Henikoff, The epigenetic progenitor origin of human cancer, Nat. Rev. Genet. 7 (2006) 21–33. J.E. Ohm, K.M. McGarvey, X. Yu, L. Cheng, K.E. Schuebel, L. Cope, H.P. Mohammad, W. Chen, V.C. Daniel, W. Yu, D.M. Berman, T. Jenuwein, K. Pruitt, S.J. Sharkis, D.N. Watkins, J.G. Herman, S.B. Baylin, A stem cell-like chromatin pattern may predispose tumor suppressor genes to DNA hypermethylation and heritable silencing, Nat. Genet. 39 (2007) 237–242. J.A. Fahrner, S. Eguchi, J.G. Herman, S.B. Baylin, Dependence of histone modi?cations and gene expression on DNA hypermethylation in cancer, Cancer Res. 62 (2002) 7213–7218. C.T. Nguyen, D.J. Weisenberger, M. Velicescu, F.A. Gonzales, J.C. Lin, G. Liang, P.A. Jones, Histone H3-lysine 9 methylation is associated with aberrant gene silencing in cancer cells and is rapidly reversed by 5-aza-2′-deoxycytidine, Cancer Res. 62 (2002) 6456–6461. Y. Kondo, L. Shen, J.P. Issa, Critical role of histone methylation in tumor suppressor gene silencing in colorectal cancer, Mol. Cell. Biol. 23 (2003) 206–215. Y.S. Park, M.Y. Jin, Y.J. Kim, J.H. Yook, B.S. Kim, S.J. Jang, The global histone modi?cation pattern correlates with cancer recurrence and overall survival in gastric adenocarcinoma, Ann. Surg. Oncol. 15 (2008) 1968–1976. D.B. Seligson, S. Horvath, T. Shi, H. Yu, S. Tze, M. Grunstein, S.K. Kurdistani, Global histone modi?cation patterns predict risk of prostate cancer recurrence, Nature 435 (2005) 1262–1266. M.A. Mohamed, P.A. Greif, J. Diamond, O. Sharaf, P. Maxwell, R. Montironi, R.A. Young, P.W. Hamilton, Epigenetic events, remodelling enzymes and their relationship to chromatin organization in prostatic intraepithelial neoplasia and prostatic adenocarcinoma, BJU Int. 99 (2007) 908–915. M.F. Fraga, E. Ballestar, A. Villar-Garea, M. Boix-Chornet, J. Espada, G. Schotta, T. Bonaldi, C. Haydon, S. Ropero, K. Petrie, N.G. Iyer, A. Perez-Rosado, E. Calvo, J.A. Lopez, A. Cano, M.J. Calasanz, D. Colomer, M.A. Piris, N. Ahn, A. Imhof, C. Caldas, T. Jenuwein, M. Esteller, Loss of acetylation at Lys16 and trimethylation at Lys20 of histone H4 is a common hallmark of human cancer, Nat. Genet. 37 (2005) 391–400. M. Shogren-Knaak, H. Ishii, J.M. Sun, M.J. Pazin, J.R. Davie, C.L. Peterson, Histone H4-K16 acetylation controls chromatin structure and protein interactions, Science 311 (2006) 844–847. M. Wiren, R.A. Silverstein, I. Sinha, J. Walfridsson, H.M. Lee, P. Laurenson, L. Pillus, D. Robyr, M. Grunstein, K. Ekwall, Genomewide analysis of nucleosome density histone acetylation and HDAC function in ?ssion yeast, EMBO J. 24 (2005) 2906–2918. [52] O. Bell, C. Wirbelauer, M. Hild, A.N. Scharf, M. Schwaiger, D.M. MacAlpine, F. Zilbermann, F. van Leeuwen, S.P. Bell, A. Imhof, D. Garza, A.H. Peters, D. Schubeler, Localized H3K36 methylation states de?ne histone H4K16 acetylation during transcriptional elongation in Drosophila, EMBO J. 26 (2007) 4974–4984. [53] A. Akhtar, P.B. Becker, Activation of transcription through histone H4 acetylation by MOF, an acetyltransferase essential for dosage compensation in Drosophila, Mol. Cell 5 (2000) 367–375. [54] V.P. Tryndyak, O. Kovalchuk, I.P. Pogribny, Loss of DNA methylation and histone H4 lysine 20 trimethylation in human breast cancer cells is associated with aberrant expression of DNA methyltransferase 1, Suv4-20h2 histone methyltransferase and methyl-binding proteins, Cancer Biol. Ther. 5 (2006) 65–70. [55] I.P. Pogribny, S.A. Ross, V.P. Tryndyak, M. Pogribna, L.A. Poirier, T.V. Karpinets, Histone H3 lysine 9 and H4 lysine 20 trimethylation and the expression of Suv420h2 and Suv-39h1 histone methyltransferases in hepatocarcinogenesis induced by methyl de?ciency in rats, Carcinogenesis 27 (2006) 1180–1186. [56] A. Gupta, T.G. Guerin-Peyrou, G.G. Sharma, C. Park, M. Agarwal, R.K. Ganju, S. Pandita, K. Choi, S. Sukumar, R.K. Pandita, T. Ludwig, T.K. Pandita, The mammalian ortholog of Drosophila MOF that acetylates histone H4 lysine 16 is essential for embryogenesis and oncogenesis, Mol. Cell. Biol. 28 (2008) 397–409. [57] F. Barlesi, G. Giaccone, M.I. Gallegos-Ruiz, A. Loundou, S.W. Span, P. Lefesvre, F.A. Kruyt, J.A. Rodriguez, Global histone modi?cations predict prognosis of resected non small-cell lung cancer, J. Clin. Oncol. 25 (2007) 4358–4364. [58] C.G. Kleer, Q. Cao, S. Varambally, R. Shen, I. Ota, S.A. Tomlins, D. Ghosh, R.G. Sewalt, A.P. Otte, D.F. Hayes, M.S. Sabel, D. Livant, S.J. Weiss, M.A. Rubin, A.M. Chinnaiyan, EZH2 is a marker of aggressive breast cancer and promotes neoplastic transformation of breast epithelial cells, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 11606–11611. [59] J. Yu, J. Yu, D.R. Rhodes, S.A. Tomlins, X. Cao, G. Chen, R. Mehra, X. Wang, D. Ghosh, R.B. Shah, S. Varambally, K.J. Pienta, A.M. Chinnaiyan, A polycomb repression signature in metastatic prostate cancer predicts cancer outcome, Cancer Res. 67 (2007) 10657–10663. [60] M.G. Guenther, S.S. Levine, L.A. Boyer, R. Jaenisch, R.A. Young, A chromatin landmark and transcription initiation at most promoters in human cells, Cell 130 (2007) 77–88. [61] D. Schubeler, D.M. MacAlpine, D. Scalzo, C. Wirbelauer, C. Kooperberg, F. van Leeuwen, D.E. Gottschling, L.P. O, Neill, B.M. Turner, J. Delrow, S.P. Bell, M. Groudine, The histone modi?cation pattern of active genes revealed through genome-wide chromatin analysis of a higher eukaryote, Genes Dev. 18 (2004) 1263–1271. [62] K. Ekwall, Genome-wide analysis of HDAC function, Trends Genet. 21 (2005) 608–615. [63] D. Marchion, P. Munster, Development of histone deacetylase inhibitors for cancer treatment, Expert Rev. Anticancer Ther. 7 (2007) 583–598. [64] G. Bug, K. Schwarz, C. Schoch, M. Kampfmann, R. Henschler, D. Hoelzer, O.G. Ottmann, M. Ruthardt, Effect of histone deacetylase inhibitor valproic acid on progenitor cells of acute myeloid leukemia, Haematologica 92 (2007) 542–545.

[38] [39] [40] [41] [42]

[43]

[44]

[45] [46]

[47]

[48]

[49]

[50]

[51]



相关文章:
更多相关标签: