Critical Review Heterochromatin Structure: Lessons from the Budding Yeast

Xin Bi

Department of Biology, University of Rochester, Rochester, NY, USA

Abstract The eukaryotic genome can be roughly divided into euchromatin and heterochromatin domains that are structurally and functionally distinct. Heterochromatin is characterized by its high compactness and its inhibitory effect on DNA transactions such as gene expression. Formation of heterochromatin involves special histone modifications and the recruitment and spread of silencing complexes and causes changes in the primary and higher order structures of chromatin. The past two decades have seen dramatic advances in dissecting the molecular aspects of heterochromatin because of the identification of the histone code for heterochromatin as well as its writers and erasers (histone-modifying enzymes) and readers (silencing factors recognizing histone modifications). How hetero-

chromatic histone modifications and silencing factors contribute to the special primary and higher order structures of heterochromatin has begun to be understood. The budding yeast Saccharomyces cerevisiae has long been used as a model organism for heterochromatin studies. Results from these studies have contributed significantly to the elucidation of the general principles governing the formation, maintenance, and function of heterochromatin. This review is focused on investigations into the structural aspects of heterochromatin in S. cerevisiae. Current understanding of other aspects of heterochromatin including how it promotes gene silencing and its epigenetic inheritance is briefly summarized. C 2014 IUBMB Life, 00(00):000–000, 2014 V

Keywords: heterochromatin; higher order chromatin structure; Saccharomyces cerevisiae; Sir proteins; transcriptional silencing

Introduction The eukaryotic genome is packed into chromatin, an ensemble of DNA and proteins including histones and nonhistone proteins. This not only helps to confine DNA in the nucleus and protects it from damage but also provides the basis for chromatin-mediated regulation of DNA transactions including gene transcription, DNA replication, recombination, and repair. The basic unit of chromatin is the nucleosome consisting of 147 base pairs of DNA wrapping around a protein core made of two each of histones H2A, H2B, H3, and H4. The degree of chromatin compaction/organization is not homoge-

C 2014 International Union of Biochemistry and Molecular Biology V

Volume 00, Number 00, Month 2014, Pages 00–00 Address correspondence to: Xin Bi, Department of Biology, University of Rochester, Rochester, NY 14627, USA. Tel: 1585-275-6922. Fax: 1585275-2070. E-mail: [email protected] Received 26 September 2014; Revised 12 October 2014; Accepted 14 October 2014 DOI 10.1002/iub.1322 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com)

IUBMB Life

neous across the genome, leading to the existence of interspersed highly condensed regions called heterochromatin and less condensed regions referred to as euchromatin (1). Heterochromatin was originally defined cytologically as the part of chromosome that stays heavily stained/condensed in both mitosis and interphase, as opposed to the rest of the chromosome that is only visible/condensed in mitosis but becomes decondensed in interphase in insect cells. Another unique feature of heterochromatin noticed in early studies is its tendency to localize at specific sites in the nucleus including the nuclear envelope and nucleolus. Later studies of heterochromatin in diverse organisms revealed its various structural and functional properties and uncovered mechanisms underlying its establishment, maintenance, and inheritance. Research in the past several decades has resulted in a detailed characterization of heterochromatin at the nucleosomal and molecular levels. This also led to the realization of the existence of heterochromatin loci in small lower eukaryotes where cytological observation of chromatin compaction is difficult, if not impossible (2,3). Such loci are exemplified by the HML and HMR loci in the budding yeast Saccharomyces cerevisiae. The HM loci are transcriptionally silent and are composed of chromatin that shares many structural and

1

IUBMB LIFE

FIG 1

Transcriptionally silent loci in S. cerevisiae. A schematic of chromosome III in MATa cells of S. cerevisiae is illustrated. The HML, MAT, HMR, and telomeric loci are shown. The E and I silencers flanking HML and HMR are indicated. See text for more description. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

functional characteristics of heterochromatin in higher eukaryotes. Studies of heterochromatin in S. cerevisiae have helped to define many concepts and principles regarding the structure, formation, maintenance, function, and inheritance of heterochromatin that are similar in principle to those in higher eukaryotes. The focus of this review is on investigations of the structural aspects of heterochromatin in the budding yeast. Heterochromatin-mediated transcriptional silencing and the epigenetic inheritance of heterochromatin are briefly discussed. Studies of heterochromatin in the fission yeast and metazoans have been the topic of many recent reviews and are only summarized briefly at the end.

Heterochromatin in S. cerevisiae: A Historical Perspective Studies of heterochromatin in S. cerevisiae stemmed from early genetic analyses of how the mating type of this singlecelled fungus was determined. Haploid yeast cells adopt either the a or the a mating type, and a and a cells can mate to form an a/a diploid. It was found that there are three loci in the haploid genome, MAT, HML, and HMR, that encode factors involved in mating type determination (Fig. 1). However, only the MAT locus is transcriptionally active, whereas HML and HMR are not transcribed. As a result, mating genes present at the MAT locus determine the mating type of haploid cells. A cell bearing the MATa allele (containing a1 and a2 genes) is an a cell, whereas a cell bearing MATa allele (containing a1 and a2 genes) is an a cell. The a1 and a2 genes are also present at the HML locus, and a1 and a2 genes are present at HMR (Fig. 1). As HML and HMR are not expressed, they do not contribute to the mating type of the host under normal circumstances, but serve as donors in a gene conversion event that result in mating type switching of haploid cells. Intrigued by the finding that the mating cassettes (a1–a2 and a1–a2) are transcribed when located at MAT but repressed/silenced at HML and HMR loci, many yeast biologists have investigated this phenomenon for several decades.

2

Genetic and molecular biological studies have identified cisacting elements and trans-acting factors involved in the silencing of the HM loci. The cis-acting elements are small sequences dubbed silencers that flank the HML and HMR loci, each containing a combination of two or three binding sites for the Abf1 and Rap1 proteins and origin recognition complex (ORC; Fig. 1). The trans-acting factors include silencerbinding proteins and the Sir2, Sir3, and Sir4 proteins that are strictly required for transcriptional silencing of the HM loci (Fig. 1). The first piece of evidence linking histones/chromatin to transcriptional silencing in yeast was the discovery that deletion of part of the N-terminus of histone H4 led to derepression of the HM loci (4). Later studies showed that mutating H4 Nterminus, especially lysines 5, 8, 12, and 16, derepresses HML to various degrees (5). It was thought that the lysines were important for silencing because of the positive charges they carry. However, changing all four lysines 5, 8, 12, and 16 to arginines abolished transcriptional silencing despite the fact that the positive charges were preserved. Given that lysine acetylation was known to be correlated with transcriptional activity in general, it was suggested that lysine acetylation at H4 N-terminus, or the lack of it, played a role in transcriptional silencing. This idea gained critical support when histones H3 and H4 at the silent HML and HMR loci were found to be hypoacetylated when compared with those at the active MAT locus (6). The groundbreaking discovery of the first histone acetyltransferase (HAT) and first histone deacetylase (HDAC) in 1996 dramatically impacted the studies of chromatin and chromatin-mediated gene regulation (7,8). The fact that the HAT Gcn5 is a known transcriptional coactivator and HDAC Rpd3 is a transcriptional repressor in yeast not only reaffirms the correlation between histone acetylation and gene activity but also strongly suggests that cells actively modulate the level of chromatin acetylation as a means of gene regulation. Since then, numerous HATs and HDACs have been identified in various eukaryotes. Importantly, the Sir2 protein essential for transcriptional silencing in yeast was found to be a HDAC (9). This, together with findings on the properties of the Sir2

Structural Aspects of Heterochromatin in the Budding Yeast

Special Primary and Higher Order Structures of Heterochromatin in S. cerevisiae

FIG 2

Proposed mechanism for the formation of heterochromatin in S. cerevisiae. The proposed propagation of SIR complex during the formation of heterochromatin is shown. A silencer or telomere binds factors that in turn recruit the SIR complex (Sir3/Sir4/Sir4). Sir2 deacetylates a nearby nucleosome, and the hypoacetylated nucleosome recruits another incoming SIR complex. Repeated cycles of histone deacetylation and SIR complex recruitment result in the linear propagation of SIR complex until a barrier element is encountered. Green circles, euchromatic nucleosomes; orange circles, heterochromatic nucleosomes. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

through Sir4 proteins detailed later in this review, suggests a mechanism for the formation of heterochromatin involving active deacetylation of histones (Fig. 2). In this mechanism, a silencer recruits silencer-binding proteins Abf1, Rap1, and ORC that in turn recruit the SIR complex (Sir2/Sir3/Sir4; Fig. 2). The SIR complex at the silencer then deacetylates histones in an adjacent nucleosome, which then associates with another SIR complex with high affinity. The nucleosome-bound SIR complex subsequently deacetylates the neighboring nucleosome, which in turn recruits a new SIR complex. Through repeated cycles of nucleosome deacetylation and SIR complex recruitment, SIR complex promotes its own propagation along the nucleosome array in chromatin and serves as an integral part of heterochromatin. In this model, interactions between the SIR complex and chromatin and the HDAC activity of Sir2 are key to the establishment and maintenance of silenced chromatin (2). Note that although the model in Fig. 2 depicts the formation of heterochromatin as a sequential/linear process, there is evidence suggesting that the SIR complex does not always propagate on a linear template (10). In addition to the HML and HMR loci, transcriptional silencing also exists at subtelomeric regions in the budding yeast (Fig. 1; ref. 2). Sir2, Sir3, and Sir4 are also essential for telomeric silencing, and formation of silenced telomeric heterochromatin uses the same mechanism as that governing heterochromatin formation at HM loci. The telomeric repeats bind multiple molecules of Rap1 that are responsible for recruiting Sir proteins and initiating the formation of telomeric heterochromatin.

Bi

Heterochromatin is generally inhibitory to expression of genes embedded in it, which is believed to be due to its special repressive structure that differs from euchromatin in many aspects. Heterochromatin is marked by special histone modifications, is associated with specialized silencing complexes, and is relatively inaccessible to DNA-modifying proteins. Nucleosomes are generally more regularly distributed in heterochromatin than in euchromatin. Moreover, heterochromatin folds into higher order structures. Heterochromatin formation also induces an increase in the negative supercoiling of DNA in it. Although heterochromatin is a stable structure, it is not static, but rather dynamic, and is subject to cell cycle regulation. Propagation of heterochromatin is restricted by special DNA elements called barriers that promote the formation of active chromatin and/or exclusion of nucleosomes. Heterochromatin loci tend to associate with each other and the nuclear envelope to form clusters where there is a high concentration of Sir proteins, which is believed to facilitate the formation and maintenance of heterochromatin.

Heterochromatin Is Marked by Special Histone Modifications In general, the core histones are hypoacetylated in yeast heterochromatin. Acetylation of a lysine residue removes its positive charge, thereby reducing the interaction between histone and the negatively charged DNA, making the nucleosome assume an “open” structure (11). In addition, histone H4-K16 acetylation inhibits the folding of chromatin fiber into higher order structures (12). As such, histone hypoacetylation in heterochromatin is believed to render nucleosomes less “open” and is inducive to the formation of higher order structures. Histone H3-K4 methylation and K79 methylation are associated with transcriptionally active chromatin. Heterochromatin is hypomethylated at H3-K4 and K79. Importantly, the SIR complex preferentially interacts with hypoacetylated and hypomethylated nucleosomes, which provides the basis for its specific targeting to heterochromatin. In fission yeast, Drosophila, and vertebrates, histones in heterochromatin are not only hypoacetylated and hypomethylated at H3-K4 and K79 but also highly methylated at H3-K9 (3). Methylated H3-K9 is the target of heterochromatin protein 1 (HP1), an essential component of heterochromatin in these organisms.

Heterochromatin Is Associated With Specialized Silencing Complexes The Sir2, Sir3, and Sir4 proteins associate with each other to form a heterotrimeric SIR complex (Sir2/Sir3/Sir4; ref. 2). Specifically, Sir2 and Sir3 both interact with Sir4 in the SIR complex, and Sir2/Sir4 can also exist as a separate complex. The fact that the SIR complex plays a direct role in heterochromatin formation was demonstrated by the finding that Sir3 and

3

IUBMB LIFE

binding to Esc1, a nuclear envelope-associated protein. The Sir2-interacting domain of Sir4 binds the N-terminal regulatory domain of Sir2 (Sir2N) and makes contact with the interface between Sir2N and Sir2 catalytic domain, which stimulates the deacetylase activity of Sir2 (20). The C-terminal coiled coil domain is involved in Sir4 homodimerization and its interaction with Sir3. Importantly, SIR complex preferentially binds deacetylated and demethylated histones/nucleosomes (15,16). Taken together, the multiple interactions among the Sir proteins, silencer/telomere-binding proteins, and the nucleosome and the ability of Sir2 to deacetylate histones form the basis for how the SIR complex binds chromatin and promotes the transition of euchromatin into heterochromatin (Fig. 2).

DNA in Heterochromatin Is Less Accessible Than That in Euchromatin

FIG 3

Modular structures of the Sir2, Sir3, and Sir4 proteins and their interacting partners. See text for description. (Adapted from ref. 2.). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Sir4 can bind histones/nucleosomes in vitro and that the SIR complex is associated with heterochromatin loci in vivo (13–16). Extensive investigations of the structures and functions of Sir2, Sir3, and Sir4 have yielded a plethora of information about their roles in the formation of heterochromatin. As a few recent reviews have already thoroughly discussed these important studies (e.g., ref. 2), a brief summary of the major findings is as follows. Sir2 is an NAD-dependent protein deacetylase consisting of a C-terminal catalytic domain and an Nterminal regulatory domain that interacts with Sir4 (Fig. 3). Sir2 deacetylates lysines in the N-terminal tails of histones (9). Sir3 contains regions that interact with Sir4, Rap1, and the nucleosome, as well as a homodimerization domain (Fig. 3). The N-terminal bromoadjacent homology (BAH) domain of Sir3 interacts with both the N-terminal tail of histone H4 and a surface of the nucleosome called loss of rDNA silencing (LRS) that include H3-K79 (17). Both the H4 N-terminal tail and LRS domain are important for transcriptional silencing. An AAA1 ATPase-like domain interacts with Sir4 and may also interact with the nucleosome (18). A C-terminal-winged helixturn-helix (wH) motif mediates Sir3 homodimerization, which is essential for heterochromatin formation (19). Sir4 interacts with many factors including Sir2, Sir3, Sir1, Yku70, Yku80, Rap1, Esc1, nucleosome, and DNA via several domains and is thought to have a scaffold function in heterochromatin assembly (Fig. 3; ref. 2). Sir1 is an ORC-binding protein, and the Yku70/Yku80 complex associates with chromosome ends. By binding to Sir1, Rap1, and Yku70/Yku80, Sir4 acts to initiate the assembly of SIR complex at the HM silencers and telomeres. The partitioning and anchoring domain (PAD) mediates the tethering of heterochromatin to nuclear periphery by

4

The fact that heterochromatin in metazoans appears to be more condensed than euchromatin suggested that DNA in heterochromatin may be less accessible to DNA-interacting factors. Loo and Rine (21) tested this idea in yeast by examining how efficiently the HO site-specific endonuclease could digest the HMR vs. MAT locus. The HML, HMR, and MAT loci each contain a single cutting site for HO. It was shown that HMR could not be cut in cells with intact SIR complex, but was cut efficiently in cells lacking any of the Sir proteins. On the other hand, MAT was readily cut independently of the presence of SIR complex. Similar results were obtained when restriction enzymes were used in the experiment. These results demonstrated clearly that heterochromatin shields HMR from digestion by endonucleases. In line with this finding, Gottschling (22) showed that unlike DNA in euchromatin, DNA within telomeric heterochromatin was refractory to modification by the Dam methyltransferase ectopically expressed in yeast. Therefore, DNA in heterochromatin is less accessible than that in euchromatin.

Primary and Higher Order Structures of Heterochromatin Formation of heterochromatin is accompanied by changes in the primary and higher order structures of chromatin. The primary structure of chromatin concerns the distribution of nucleosomes along DNA. Formation of a nucleosome involves the wrapping of 147 base pairs of DNA around the histone octamer, which is generally a sequence-independent process. However, nucleosomes in eukaryotic genome are mostly positioned at specific sites consistently (e.g., ref. 23). This is likely the consequence of at least three possible causes. (i) Although histone–DNA interaction is sequence nonspecific, the abilities of certain special DNA sequences to form nucleosomes may be higher or lower than random sequences. For example, a sufficiently long poly(dA:dT) sequence, or T-track, adopts a rigid structure that prevents it from wrapping into nucleosomes (24). (ii) An array of nucleosomes may be forced into a specific arrangement by DNA-binding proteins and chromatin remodelers. A strong DNA-binding protein may exclude histones, thereby establishing the border of a nucleosome array thereby

Structural Aspects of Heterochromatin in the Budding Yeast

FIG 4

Formation of heterochromatin induces changes in primary chromatin structure. Top: Schematic of the HML locus. The HML-E and -I silencers and the a2 and a1 genes are shown. The filled bar represents the sequence of a probe used in a chromatin mapping experiment. Bottom: Mapping HML chromatin in wild-type (SIR1) cells and cells lacking SIR complex (sir2). The result of a MNase digestion and indirect end-labeling experiment is shown. Permeabilized cells were treated with MNase. DNA was isolated, digested with Pvu II, as shown on the top, and subjected to electrophoresis. After Southern blotting, relevant DNA fragments were detected by a probe near the Pvu II site as shown on the top. Red circles indicate MNase sensitive sites unique to wildtype cells. Yellow diamonds are sites unique to cells lacking SIR complex. The inferred positions of nucleosomes are shown on the right. They are numbered in accordance with designations in ref. 29. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

determining the phasing of the nucleosomes. A chromatin remodeler has the ability to slide the nucleosome along the DNA. (iii) A nucleosome-binding protein may restrict the dynamics of the nucleosome and “fix” the nucleosome to one of many dynamic positions it can adopt without the protein (25). The first attempt to map chromatin at the HML and HMR loci revealed SIR-dependent changes in DNase I and micrococcal nuclease (MNase) digestion patterns, supporting the idea that silent HM loci assumed a special, silenced chromatin (26). Later, high-resolution mapping experiments revealed 20 positioned nucleosomes within HML and 12 within HMR bordered by the silencers (27,28). These nucleosomes are disrupted and/ or shifted to various degrees in cells deleted for Sir2, Sir3, or Sir4, demonstrating again that the SIR complex promotes the formation of a special heterochromatin structure (Fig. 4; ref. 27–29). Along this line, it was found that nucleosomes are

Bi

arranged with a characteristically high regularity in Drosophila heterochromatin, which is believed to be important for the folding of chromatin fiber into higher order structures (30). What promotes nucleosome rearrangement in the transition from euchromatin to heterochromatin has not been resolved. It may happen during DNA replication when new nucleosomes are formed at the silent loci. Alternatively, or in addition, preexisting nucleosomes may be repositioning independently of DNA replication. Given that de novo formation of heterochromatin can occur independently of DNA replication per se (31,32), repositioning of preexisting nucleosomes is likely the major activity involved in the establishment of the special primary structure of heterochromatin. The next question concerns the factor(s) responsible for altering the conformation and/or positions of nucleosomes. SIR complexes may bind nucleosomes, thereby limiting their dynamics, whereas chromatin-remodeling factors may remodel nucleosomes to make them adopt new conformations and/or positions. Many chromatin remodelers in various organisms have been found to contribute to transcriptional silencing and heterochromatin structure (33). A recent example is the Fun30 chromatin remodeler in yeast. Deletion of Fun30 reduces transcriptional silencing at HML to an intermediate level between fully silenced and fully derepressed states (34). Consistently, Fun30 deletion also changes HML heterochromatin to a configuration that is distinct from both the fully silenced state (in the presence of SIR complex) and fully derepressed state (in the absence of SIR complex; ref. 34). This not only demonstrates that Fun30 contributes to the primary structure of heterochromatin but also suggests that Fun30 is not the only force behind chromatin changes in heterochromatin formation. The SIR complex and/or other yet to be identified chromatinremodeling activities may also be involved. The primary chromatin structure can be further folded into higher order structures such as the 30-nm fiber, a reasonably well-defined secondary structure, and other poorly defined structures (35). Higher order structures of heterochromatin are believed to be important for its function in genome organization and function. However, confirming the existence and defining the nature of a higher order chromatin structure has proven to be generally technically challenging, especially in lower eukaryotes like the budding yeast. Nevertheless, there is increasing evidence pointing to the existence of higher order structures (or folding) of heterochromatin in the budding yeast. There was evidence suggesting that a loop is formed at telomeric heterochromatin where the telomere associated with Rap1 proteins contact subtelomeric heterochromatin (36). It was also found that the two silencers flanking the HMR locus are in close proximity, suggesting the existence of an HMR loop (37). Sperling and Grunstein (38) found that a 9-kb-long chromosome fragment containing the HML locus migrates more slowly in sucrose gradient, when the silent state of HML is disrupted, suggesting that HML heterochromatin is more compact. Consistently, SIR complex was shown to promote condensation of reconstituted chromatin template in vitro in a

5

IUBMB LIFE

is derepressed (43,44), which is consistent with the higher nucleosome density and lower histone acetylation in heterochromatin at HM loci. Therefore, the topology of DNA at a defined genomic locus can be used as a proxy of the state of local chromatin. This strategy has proven to be a very useful tool for examining the stability and kinetics of heterochromatin formation (43,45).

Heterochromatin Is a Dynamic Structure

FIG 5

Model for a higher order heterochromatin superstructure at HMR. In this model, HMR-E and -I silencers interact to form a chromatin loop associated with Sir proteins. There is also an internal loop that contains functional elements such as the promoter of the a1 and a2 genes and the HO cutting site. This loop is buried in the superstructure. (Adapted from ref. 40.). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

histone deacetylation-dependent manner (39). A recent highresolution chromatin immunoprecipitation coupled with nextgeneration sequencing analysis of SIR complex distribution yielded a surprising finding that SIR complexes appear to associate with HML and HMR loci in a heterogeneous fashion, concentrating in certain parts while avoiding other parts (40). This result is in contrast to the commonly held notion that SIR complexes “cover” HML and HMR regions uniformly as a result of their linear step-by-step propagation during the establishment of heterochromatin (Fig. 2), but is consistent with the idea that heterochromatin forms a tightly packed “superstructure” with functional elements such as the promoter of mating genes and the HO cutting site buried inside (Fig. 5). Taken together, the above findings support the notion that heterochromatin folds into a more compact, higher order structure.

DNA in Heterochromatin Assumes a Distinct Topology DNA in the eukaryotic genome is generally negatively supercoiled, as formation of each nucleosome introduces on average one negative supercoil into nucleosomal DNA when it wraps around the histone core (41). Histone acetylation reduces the number of supercoils constrained on nucleosomal DNA (42). Therefore, the topology of DNA within a particular genomic region is an indicator of the state of local chromatin regarding the density and acetylation level of nucleosomes. The topology/ supercoiling of DNA can be readily measured when it is in a circular form. To investigate the supercoiling of DNA of a particular region in the genome, one can use site-specific recombination to excise this region as a minichromosome circle in vivo which can be easily isolated and analyzed (43,44). Using this method, it was shown that DNA at HML or HMR is more negatively supercoiled when the locus is silenced than when it

6

Although heterochromatin is highly packed, it is not a rigid structure, but rather a dynamic structure. Using the aforementioned DNA topology-based assay, we found that the state of heterochromatin at HML fluctuates during the cell cycle; it is fully silenced at G0, G1, and early S phases of the cell cycle, but becomes partially disrupted in G2/M phase (43). This is consistent with an earlier report showing that a gene embedded in telomeric heterochromatin is subject to activation by an activator in G2/M, but not G0, G1, and early S phases of the cell cycle (46). Heterochromatin is dynamic even in G1 phase as free Sir proteins can be incorporated into existing heterochromatin (47). The dynamics of heterochromatin in larger cells can be visualized by using fluorescence recovery after photobleaching that measures the diffusion of proteins associated with cellular structures (48). It was found that HP1 molecules in mammalian heterochromatin are subject to dynamic exchange (48). This is believed to enable regulatory factors to access their targets in heterochromatin without affecting its stability (48).

Boundaries of Heterochromatin The fact that establishment of heterochromatin is mediated by the spreading of heterochromatin-specific complexes along chromatin poses the question of how a euchromatin locus is protected from adjacent heterochromatin or how the boundaries of a heterochromatin domain are defined. Early studies in Drosophila and vertebrates found chromatin boundaries to coincide with “special chromatin structures” that were hypersensitive to nucleases (49). Sequences that can prevent the propagation of transcriptional silencing were also identified in S. cerevisiae and named heterochromatin barriers (Fig. 2). These barriers are distinct in sequence and are composed of binding sites for transcription regulators or factors that associate with the nuclear pore complex (50–52). These findings suggest that a barrier may function by recruiting chromatin-modifying/remodeling complexes that promote active chromatin formation or by attaching to an “immobile” nuclear structure. We also found that nucleosome-excluding sequences such as long T-tracks are sufficient to block the spreading of heterochromatin (53). Consistently, the right boundary of HMR, a sequencing encompassing a tRNA gene, was found to be nucleosome-free (54). Therefore, a barrier element may actively promote nucleosome modification such as acetylation to counter heterochromatin-specific modifications such as histone deacetylation or act as a passive roadblock by excluding nucleosomes or attaching to an immobile nuclear structure. Consistently, studies in higher cells revealed similar

Structural Aspects of Heterochromatin in the Budding Yeast

mechanisms for the function of heterochromatin boundary elements (55).

Subnuclear Compartmentation of Heterochromatin Heterochromatin regions in the eukaryotic genome tend to associate with each other and attach to nuclear structures such as the nuclear envelope (56). In the budding yeast, it was found that the 32 telomeres interact to form about eight perinuclear foci of subtelomeric heterochromatin (57,58). Both the telomere-binding complex Yku70/Yku80 and SIR complex contribute to the tethering of telomeres to nuclear envelope (59). Clustering of telomeric heterochromatin loci increases the local concentration of Sir proteins, which is thought to reinforce transcriptional silencing of the loci within the compartment. As HML and HMR are located close to the left and right telomeres of chromosome III (12 and 23 kb, respectively), they are each also located near a cluster of telomeres. Moreover, there is evidence suggesting that HML specifically interacts with the left telomere of chromosome III and that the HML and HMR loci also interact with each other (60–62). These interactions seem to be mediated by silencer–telomere and silencer–silencer contacts. HML–HMR interaction can occur independently of telomere anchoring to the nuclear envelope, suggesting that it is not the side effect of potential interactions between the left and right telomeres of chromosome III (61,62). Interestingly, HML–HMR interaction was shown to be dependent on SMC proteins known to participate in chromosome organization, as well as factors involved in DNA doublestrand break (DSB) repair, in addition to Sir proteins (62). The biological implication of this finding remains to be defined.

Mechanism of HeterochromatinMediated Transcriptional Silencing Heterochromatin generally silences the expression of genes embedded in it, whereas euchromatin is permissive to gene expression. Gene silencing is conceptually different from gene repression as it is region/locus-specific but gene-nonspecific, whereas gene repression is a gene-specific phenomenon. Heterochromatin has long been linked to gene silencing (1). Given the condensed nature of heterochromatin structure and its association with silencing complexes, it has been generally believed that heterochromatin silences the expression of genes embedded in it by limiting the access of DNA to transcription factors. However, this notion is challenged by increasing evidence. Sekinger and Gross (63) found heterochromatin to be actually permissive to activators, components of the preinitiation complex, TATA-binding protein, and RNA polymerase II. They further obtained data suggesting that it is the transition between RNA polymerase II initiation and elongation that is blocked by heterochromatin (64). They, therefore, concluded that heterochromatin prevents the recruitment of transcription elongation factors to the promoter of the gene in it (64). However, another study found that heterochromatin allows activator binding but strongly reduces the occupancy of basal

Bi

transcription factors TFIIB and TFIIE as well as RNA polymerase II (65). In a recent in vitro reconstitution experiment, chromatin associated with SIR complex was shown to allow activator binding but hinders the recruitment of coactivators (66). The above results demonstrate that transcriptional silencing is not achieved by preventing activators from binding their targets in heterochromatin. However, there is still debate regarding the exact step of the transcription process that is blocked by heterochromatin.

Epigenetic Inheritance of Heterochromatin Once established, heterochromatin is stably maintained during the cell cycle, withstanding the disruptive process of DNA replication, which exemplifies the phenomenon of epigenetic inheritance of chromatin states. DNA replication is a semiconservative process with each of the two single strands of DNA duplex serving as a template for the synthesis of a new complementary stand, yielding two copies of double-stranded DNA. DNA replication is intimately linked to nucleosome disassembly in front of the replication fork and nucleosome assembly on nascent DNA strands (67). The “old” histones from preexisting nucleosomes on the template chromatin are believed to be evenly distributed to the two nascent strands of DNA, constituting half of the histones incorporated in new chromatin. The other half of the histones is newly synthesized. Therefore, chromatin on newly synthesized DNA strands consists both old and new histones. After a heterochromatin region is replicated, the old histones bearing heterochromatin-specific modifications may remain associated with, or recruit, heterochromatin binding complex, which would modify the new nucleosomes, thereby restoring the state of heterochromatin. In short, heterochromatin may template its own duplication during DNA replication. Recent studies have begun to reveal the molecular details of the mechanism of epigenetic inheritance of chromatin states (68).

Heterochromatin in the Fission Yeast and Metazoans Historically, transcriptional silencing was studied in parallel in both the budding yeast S. cerevisiae and fission yeast Schizosaccharomyces pombe. Schizosaccharomyces pombe is similar to metazoans, but distinct from budding yeast with respect to heterochromatin, and research in this organism has yielded deep molecular insights into the mechanism of heterochromatin formation (3). Heterochromatin in fission yeast exists at centromeres, the mat locus, and subtelomeric regions. Heterochromatin is high in H3-K9 methylation and low in H3-K4 methylation (69). HP1 (also known as swi6 in fission yeast) is an essential component of heterochromatin that specifically binds methylated H3-K9. HP1 also associates with a H3-K9 histone methyltransferase (HMT) and an HDAC to form a

7

IUBMB LIFE

complex that is proposed to promote the repeated cycles of H3-K9 deacetylation, H3-K9 methylation, and HP1 recruitment, leading to the propagation of HP1 complex along chromatin. Two distinct mechanisms are responsible for the initiation of heterochromatin assembly. One is mediated by special sequences (also called silencers) containing recognition sites for trans-acting factors that recruit HDAC and HMT, which is similar to the initiation process found in the budding yeast. The other mechanism uses the RNA interference (RNAi) machinery that targets repetitive DNA elements to help recruit HDAC and HMT. The centromeric, mat, and subtelomeric loci share centromeric repeats that are the target of RNAi machinery. Small RNAs homologous to the centromeric repeats associate with the siRNA-binding complex RITS and target it to silent loci through pairing with nascent RNA transcripts still associated with the elongating RNA polymerases in these regions. How the RITS complex subsequently helps to recruit HDAC and HMT has yet to be elucidated. In multicellular organisms, heterochromatin can be divided into two categories: one is constitutive heterochromatin that remains condensed throughout the cell cycle, and the other one is called facultative heterochromatin that is development-dependent and subject to change in response to cellular signals (70). Although constitutive heterochromatin is found at centromeres, telomeres, and other loci bearing a high density of repetitive DNA elements, facultative heterochromatin is found at certain developmentally regulated loci. Constitutive and facultative heterochromatins are similar but not identical in structure, function, and assembly mechanism. Constitutive heterochromatin best represented by pericentric heterochromatin is structurally similar to fission yeast heterochromatin in that it is generally hypoacetylated, hypomethylated at H3-K4 and H3-K9, and is associated with HP1 (3). There is also evidence implicating RNAi in heterochromatin formation in Drosophila and mammals; however, the exact mechanisms remain unclear (71). Recently, noncoding RNAs in mammalian cells have also been shown to target silencing complexes to specific genes, thereby regulating transcription or pre-mRNA splicing (72–74). Note that S. cerevisiae does not use RNAi in the formation of heterochromatin as it happens to lack the RNAi system altogether.

Outlook Heterochromatin plays important roles in genome organization and the programming of gene expression and has been a topic of intense interest for many decades. Formation of S. cerevisiae heterochromatin is accompanied by alterations in the primary chromatin structure; however, the underlying mechanism remains unclear. It will be crucial to delineate how the Fun30 chromatin remodeler contributes to these alterations and to examine if additional chromatin remodeling factors are also involved. Whether the SIR complex directly modulates nucleosome positioning during the formation of heterochromatin should also be addressed. Another outstanding issue about

8

heterochromatin structure is the nature of the putative higher order structure. It would be useful to determine what directly mediates silencer–silencer and silencer–telomere “interactions” and how SMC proteins and DSB repair factors contribute to long-range HML–HMR interaction. Moreover, the proposed “superstructure” at the HMR locus awaits confirmation and further characterization. Exactly how heterochromatin silences gene expression has not been resolved. Future studies should address the discrepancy between different studies regarding the step(s) of transcription blocked by heterochromatin. Several attempts have been made to assemble SIR-dependent heterochromatin in vitro; however, current reconstituted structures likely only partially reflect the properties of native heterochromatin. A better understanding of the constituents of heterochromatin may lead to improved methodology for assembling SIR– chromatin complex that is structurally and functionally closer to heterochromatin in vivo.

Acknowledgements This work was supported by NSF grant MCB-1158008.

REFERENCES [1] Dillon, N. (2004) Heterochromatin structure and function. Biol. Cell. 96, 631– 637. [2] Kueng, S., Oppikofer, M., and Gasser, S. M. (2013) SIR proteins and the assembly of silent chromatin in budding yeast. Annu. Rev. Genet. 47, 275– 306. [3] Grewal, S. I., and Jia, S. (2007) Heterochromatin revisited. Nat. Rev. Genet. 8, 35–46. [4] Kayne, P. S., Kim, U. J., Han, M., Mullen, J. R., Yoshizaki, F., et al. (1988) Extremely conserved histone H4 N terminus is dispensable for growth but essential for repressing the silent mating loci in yeast. Cell 55, 27–39. [5] Laurenson, P., and Rine, J. (1992) Silencers, silencing, and heritable transcriptional states. Microbiol. Rev. 56, 543–560. [6] Braunstein, M., Rose, A. B., Holmes, S. G., Allis, C. D., and Broach J. R. (1993) Transcriptional silencing in yeast is associated with reduced nucleosome acetylation. Genes Dev. 7, 592–604. [7] Brownell, J. E., Zhou, J., Ranalli, T., Kobayashi, R., Edmondson, D. G., et al. (1996) Tetrahymena histone acetyltransferase A: a homolog to yeast Gcn5p linking histone acetylation to gene activation. Cell 84, 843–851. [8] Taunton, J., Hassig, C. A., and Schreiber, S. L. (1996) A mammalian histone deacetylase related to the yeast transcriptional regulator Rpd3p. Science 272, 408–411. [9] Imai, S., Armstrong, C. M., Kaeberlein, M., and Guarente, L. (2000) Transcriptional silencing and longevity protein Sir2 is an NAD-dependent histone deacetylase. Nature 403, 795–800. [10] Lynch, P. J., and Rusche, L. N. (2009) A silencer promotes the assembly of silenced chromatin independently of recruitment. Mol. Cell. Biol. 29, 43–56. [11] Strahl, B. D., and Allis, C. D. (2000) The language of covalent histone modifications. Nature 403, 41–45. [12] Shogren-Knaak, M., Ishii, H., Sun, J. M., Pazin, M. J., Davie, J. R., et al. (2006) Histone H4-K16 acetylation controls chromatin structure and protein interactions. Science 311, 844–847. [13] Hecht, A., Laroche, T., Strahl-Bolsinger, S., Gasser, S. M., and Grunstein, M. (1995) Histone H3 and H4 N-termini interact with SIR3 and SIR4 proteins: a molecular model for the formation of heterochromatin in yeast. Cell 80, 583–592. [14] Carmen, A. A., Milne, L., and Grunstein, M. (2002) Acetylation of the yeast histone H4 N terminus regulates its binding to heterochromatin protein SIR3. J. Biol. Chem. 277, 4778–4781.

Structural Aspects of Heterochromatin in the Budding Yeast

[15] Liou, G. G., Tanny, J. C., Kruger, R. G., Walz, T., and Moazed, D. (2005) Assembly of the SIR complex and its regulation by O-acetyl-ADP-ribose, a product of NAD-dependent histone deacetylation. Cell 121, 515–527.

[37] Valenzuela, L., Dhillon, N., Dubey, R. N., Gartenberg, M. R., and Kamakaka, R. T. (2008) Long-range communication between the silencers of HMR. Mol. Cell. Biol. 28, 1924–1935.

[16] Martino, F., Kueng, S., Robinson, P., Tsai-Pflugfelder, M., van Leeuwen, F., et al. (2009) Reconstitution of yeast silent chromatin: multiple contact sites and O-AADPR binding load SIR complexes onto nucleosomes in vitro. Mol. Cell 33, 323–334.

[38] Sperling, A. S., and Grunstein, M. (2009) Histone H3 N-terminus regulates higher order structure of yeast heterochromatin. Proc. Natl. Acad. Sci. USA 106, 13153–13159.

[17] Armache, K. J., Garlick, J. D., Canzio, D., Narlikar, G. J., and Kingston R. E. (2011) Structural basis of silencing: Sir3 BAH domain in complex with a nucleosome at 3.0 Å resolution. Science 334, 977–982. [18] Ehrentraut, S., Hassler, M., Oppikofer, M., Kueng, S., Weber, J. M., et al. (2011) Structural basis for the role of the Sir3 AAA1 domain in silencing: interaction with Sir4 and unmethylated histone H3K79. Genes Dev. 25, 1835–1846.

[39] Johnson, A., Li, G., Sikorski, T. W., Buratowski, S., Woodcock, C. L., et al. (2009) Reconstitution of heterochromatin-dependent transcriptional gene silencing. Mol. Cell 35, 769–781. [40] Thurtle, D. M., and Rine, J. (2014) The molecular topography of silenced chromatin in Saccharomyces cerevisiae. Genes Dev. 28, 245–258. [41] Simpson, R. T., Thoma, F., and Brubaker, J. M. (1985) Chromatin reconstituted from tandemly repeated cloned DNA fragments and core histones: a model system for study of higher order structure. Cell 42, 799–808.

[19] Oppikofer, M., Kueng, S., Keusch, J. J., Hassler, M., Ladurner, A. G., et al. (2013) Dimerization of Sir3 via its C-terminal winged helix domain is essential for yeast heterochromatin formation. EMBO J. 32, 437–449.

[42] Norton, V. G., Imai, B. S., Yau, P., and Bradbury, E. M. (1989) Histone acetylation reduces nucleosome core particle linking number change. Cell 57, 449–457.

[20] Hsu, H. C., Wang, C. L., Wang, M., Yang, N., Chen, Z., et al. (2013) Structural basis for allosteric stimulation of Sir2 activity by Sir4 binding. Genes Dev. 27, 64–73.

[43] Bi, X., and Broach, J. R. (1997) DNA in transcriptionally silent chromatin assumes a distinct topology that is sensitive to cell cycle progression. Mol. Cell. Biol. 17, 7077–7087.

[21] Loo, S., and Rine, J. (1994) Silencers and domains of generalized repression. Science 264, 1768–1771.

[44] Cheng, T. H., Li, Y. C., and Gartenberg, M. R. (1998) Persistence of an alternate chromatin structure at silenced loci in the absence of silencers. Proc. Natl. Acad. Sci. USA 95, 5521–5526.

[22] Gottschling, D. E. (1992) Telomere-proximal DNA in Saccharomyces cerevisiae is refractory to methyltransferase activity in vivo. Proc. Natl. Acad. Sci. USA 89, 4062–4065. [23] Gaffney, D. J., McVicker, G., Pai, A. A., Fondufe-Mittendorf, Y. N., Lewellen, N., et al. (2012) Controls of nucleosome positioning in the human genome. PLoS Genet. 8, e1003036. [24] Kunkel, G. R., and Martinson, H. G. (1981) Nucleosomes will not form on double-stranded RNA or over poly(dA).poly(dT) tracts in recombinant DNA. Nucleic Acids Res. 9, 6869–6888. [25] Zentner, G. E., and Henikoff, S. (2013) Regulation of nucleosome dynamics by histone modifications. Nat. Struct. Mol. Biol. 20, 259–266. [26] Nasmyth, K. A. (1982) The regulation of yeast mating-type chromatin structure by SIR: an action at a distance affecting both transcription and transposition. Cell 30, 567–578. [27] Weiss, K., and Simpson, R. T. (1998) High-resolution structural analysis of chromatin at specific loci: Saccharomyces cerevisiae silent mating type locus HMLa. Mol. Cell. Biol. 18, 5392–5403. [28] Ravindra, A., Weiss, K., and Simpson, R. T. (1999) High-resolution structural analysis of chromatin at specific loci: Saccharomyces cerevisiae silent mating-type locus HMRa. Mol. Cell. Biol. 19, 7944–7950. [29] Yu, Q., Elizondo, S., and Bi, X. (2006) Structural analyses of Sum1-1pdependent transcriptionally silent chromatin in Saccharomyces cerevisiae. J. Mol. Biol. 356, 1082–1092. [30] Sun, F. L., Cuaycong, M. H., and Elgin, S. C. (2001) Long-range nucleosome ordering is associated with gene silencing in Drosophila melanogaster pericentric heterochromatin. Mol. Cell. Biol. 21, 2867–2879. [31] Kirchmaier, A. L., and Rine, J. (2001) DNA replication-independent silencing in S. cerevisiae. Science 291, 646–650. [32] Li, Y. C., Cheng, T. H., and Gartenberg, M. R. (2001) Establishment of transcriptional silencing in the absence of DNA replication. Science 291, 650– 653. [33] Bi, X. (2012) Functions of chromatin remodeling factors in heterochromatin formation and maintenance. Sci. China Life Sci. 55, 89–96. [34] Yu, Q., Zhang, X., and Bi, X. (2011) Roles of chromatin remodeling factors in the formation and maintenance of heterochromatin structure. J. Biol. Chem. 286, 14659–14669.

[45] Xu, E. Y., Bi, X., Holland, M. J., Gottschling, D. E., and Broach, J. R. (2005) Mutations in the nucleosome core enhance transcriptional silencing. Mol. Cell. Biol. 25, 1846–1859. [46] Aparicio, O. M., and Gottschling, D. E. (1994) Overcoming telomeric silencing: a trans-activator competes to establish gene expression in a cell cycledependent way. Genes Dev. 8, 1133–1146. [47] Cheng, T. H., and Gartenberg, M. R. (2000) Yeast heterochromatin is a dynamic structure that requires silencers continuously. Genes Dev. 14, 452– 463. [48] Cheutin, T., McNairn, A. J., Jenuwein, T., Gilbert, D. M., Singh, P. B., et al. (2003) Maintenance of stable heterochromatin domains by dynamic HP1 binding. Science 299, 721–725. [49] Chung, J. H., Whiteley, M., and Felsenfeld, G. (1993) A 50 element of the chicken b-globin domain serves as an insulator in human erythroid cells and protects against position effect in Drosophila. Cell 74, 505–514. [50] Bi, X., and Broach, J. R. (1999) UASrpg can function as a heterochromatin boundary element in yeast. Genes Dev. 13, 1089–1101. [51] Donze, D., and Kamakaka, R. T. (2001) RNA polymerase III and RNA polymerase II promoter complexes are heterochromatin barriers in Saccharomyces cerevisiae. EMBO J. 20, 520–531. [52] Ishii, K., Arib, G., Lin, C., Van Houwe, G., and Laemmli, U.K. (2002) Chromatin boundaries in budding yeast: the nuclear pore connection. Cell 109, 551– 562. [53] Bi, X., Yu, Q., Sandmeier, J. J., and Zou, Y. (2004) Formation of boundaries of transcriptionally silent chromatin by nucleosome-excluding structures. Mol. Cell. Biol. 24, 2118–2131. [54] Dhillon, N., Raab, J., Guzzo, J., Szyjka, S. J., Gangadharan, S., et al. (2009) DNA polymerase epsilon, acetylases and remodellers cooperate to form a specialized chromatin structure at a tRNA insulator. EMBO J. 28, 2583–2600. [55] West, A. G., Huang, S., Gaszner, M., Litt, M. D., and Felsenfeld, G. (2004) Recruitment of histone modifications by USF proteins at a vertebrate barrier element. Mol. Cell 16, 453–463. [56] Lieberman-Aiden, E., van Berkum, N. L., Williams, L., Imakaev, M., Ragoczy, T., et al. (2009) Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326, 289–293.

[35] Li, G., and Reinberg, D. (2011) Chromatin higher-order structures and gene regulation. Curr. Opin. Genet. Dev. 21, 175–186.

[57] Gotta, M., Laroche, T., Formenton, A., Maillet, L., Scherthan, H., et al. (1996) The clustering of telomeres and colocalization with Rap1, Sir3, and Sir4 proteins in wild-type Saccharomyces cerevisiae. J. Cell Biol. 134, 1349–1363.

[36] Strahl-Bolsinger, S., Hecht, A., Luo, K., and Grunstein, M. (1997) SIR2 and SIR4 interactions differ in core and extended telomeric heterochromatin in yeast. Genes Dev. 11, 83–93.

[58] Hediger, F., Neumann, F. R., Van Houwe, G., Dubrana, K., and Gasser, S. M. (2002) Live imaging of telomeres: yKu and Sir proteins define redundant telomere anchoring pathways in yeast. Curr. Biol. 12, 2076–2089.

Bi

9

IUBMB LIFE

[59] Taddei, A., and Gasser, S. M. (2004) Multiple pathways for telomere tethering: functional implications of subnuclear position for heterochromatin formation. Biochim. Biophys. Acta 1677, 120–128. [60] Lebrun, E., Fourel, G., Defossez, P. A., and Gilson, E. (2003) A methyltransferase targeting assay reveals silencer-telomere interactions in budding yeast. Mol. Cell. Biol. 23, 1498–1508. [61] Miele, A., Bystricky, K., and Dekker, J. (2009) Yeast silent mating type loci form heterochromatic clusters through silencer protein-dependent longrange interactions. PLoS Genet. 5, e1000478. [62] Kirkland, J. G., and Kamakaka, R. T. (2013) Long-range heterochromatin association is mediated by silencing and double-strand DNA break repair proteins. J. Cell Biol. 201, 809–826. [63] Sekinger, E. A., and Gross, D. S. (2001) Silenced chromatin is permissive to activator binding and PIC recruitment. Cell 105, 403–414. [64] Gao, L., and Gross, D. S. (2008) Sir2 silences gene transcription by targeting the transition between RNA polymerase II initiation and elongation. Mol. Cell. Biol. 28, 3979–3994. [65] Chen, L., and Widom, J. (2005) Mechanism of transcriptional silencing in yeast. Cell 120, 37–48. [66] Johnson, A., Wu, R., Peetz, M., Gygi, S. P., and Moazed, D. (2013) Heterochromatic gene silencing by activator interference and a transcription elongation barrier. J. Biol. Chem. 288, 28771–28782.

10

[67] Li, Q., Burgess, R., and Zhang, Z. (2013) All roads lead to chromatin: multiple pathways for histone deposition. Biochim. Biophys. Acta 1819, 238–246. [68] Xu, M., Long, C., Chen, X., Huang, C., Chen, S., et al. (2010) Partitioning of histone H3-H4 tetramers during DNA replication-dependent chromatin assembly. Science 328, 94–98. [69] Noma, K., Allis, C. D., and Grewal, S. I. (2001) Transitions in distinct histone H3 methylation patterns at the heterochromatin domain boundaries. Science 293, 1150–1155. [70] Trojer, P., and Reinberg, D. (2007) Facultative heterochromatin: is there a distinctive molecular signature? Mol. Cell 28, 1–13. [71] Riddle, N. C., and Elgin, S. C. (2008) A role for RNAi in heterochromatin formation in Drosophila. Curr. Top. Microbiol. Immunol. 320, 185–209. [72] Morris, K. V. (2011) The emerging role of RNA in the regulation of gene transcription in human cells. Semin. Cell Dev. Biol. 22, 351–358.  , M., Buggiano, V., Fededa, J. P., Petrillo, E., Schor, I., et al. (2009) Con[73] Allo trol of alternative splicing through siRNA-mediated transcriptional gene silencing. Nat. Struct. Mol. Biol. 16, 717–724. [74] Ameyar-Zazoua, M., Rachez, C., Souidi, M., Robin, P., Fritsch, L., et al. (2012) Argonaute proteins couple chromatin silencing to alternative splicing. Nat. Struct. Mol. Biol. 19, 998–1004.

Structural Aspects of Heterochromatin in the Budding Yeast