"HAT" indicates the catalytic HAT subunit; "complex" indicates the purified HAT complex. The substrates refer to the in vitro specificity of the HAT subunit. Known functions of each modification in transcription are listed in the last column. ND: not determined member possesses a bromodomain of about 110 residues for specific interaction with lysine-acetylated histone tails (Sterner and Berger, 2000; Roth et al., 2001). GCN5L and PCAF are the best characterized human HATs among the GNAT family as transcriptional coactivators. Although PCAF and GCN5L share significant homology, only homozygous GCN5L null mutations exhibit mouse embryonic lethality (Xu et al., 2000), suggesting different physiological roles of different GNAT family members. Members of the MYST family are grouped together on the basis of their possession of a particular highly conserved 370 residue MYST domain, which has a catalytic mechanism different from that shared by other families of HATs. These proteins are involved in a much broader range of biological processes in various organisms. TIP60 was the first identified human MYST HAT and was recently reported as a key regulator for DNA repair and apoptosis (Ikura et al., 2000). Another group of HATs is p160 coactivators which interact with nuclear hormone receptors to upregulate receptor-dependent gene transcription in a ligand dependent manner (Sterner and Berger, 2000; Roth et al., 2001). In addition to these three major groups of HATs, several other proteins such as p300/CBP and TAFII250 have also been shown to possess HAT activity to regulate transcription of a wide range of genes (Bannister and Kouzarides, 1996; Mizzen et al., 1996; Ogryzko et al., 1996). The essential role of p300/CBP for development is also confirmed by embryonic lethality of CBP and p300 mutated mice (Goodman and Smolik, 2000).

HATs are recruited to certain gene promoters through interactions with DNA-binding regulatory factors, which leads to the targeted acetylation and subsequent activation of transcription (Kundu et al., 2000; An et al., 2002). The importance of histone acetylation in gene regulation began to be intensively investigated when several known transcription coactivators such as p300/CBP and GCN5 were found to have intrinsic HAT activities. As listed in Table 1, subsequent studies identified many additional proteins to possess HAT activity. Most HAT enzymes are assembled into multiprotein complexes for their proper recruitment by sequence-specific transcription activators to target genes and for their substrate specificity on nucleosomes. These HAT complexes have recently been purified from human and yeast cells and functionally characterized as positive regulators of transcription (Table 1). Similar to HATs, the HDACs are also part of large multisubunit complexes (Yang and Seto, 2003). Therefore, the level of acetylation in a specific gene is maintained by a dynamic competition between the activities of HAT and HDAC after their recruitments, and aberrant regulation of this competition has been shown to cause inappropriate gene expression.

Although histone acetylation status of promoters and regulatory elements is generally correlated with HAT-mediated transcription (Roth et al., 2001; Liang et al., 2004; Roh et al., 2005), many studies have proven that some of the HATs such as p300 and pCAF/GCN5 can also acetylate non-histone proteins (Sterner and Berger, 2000; Glozak et al., 2005). Thus, caution has been emphasized when interpreting a requirement for a HAT to conclude that a histone acetylation event is responsible for transcriptional activation. In most cases, the evidence for the effect of histone acetylation by HAT proteins in transcription is provided by the inactivation of their HAT activities. This approach may assist in determining the role of histone acetylation in transcription; however, it is likely that a great deal of functional redundancy exists between histone and non-histone modifying activities of HAT proteins. Thus care should be taken when interpreting HAT mutation/inactivation data with regards to its role in transcription. The best approach for proving the contribution of histone acetylation per se in the regulation of gene activation would be to create and test mutations of histone genes within the nucleus of multicel-lular eukaryotes. However, since several clusters of replication-dependent histone genes are present in all metazoans (Mosammaparast et al., 2001, 2002), it has been impossible to use specific mutations in histone genes to modify cellular histones. To clarify this uncertainty, a reconstituted chromatin transcription system has recently been developed by utilizing recombinant chromatin templates that are assembled from recombinant histone proteins (Loyola et al., 2001; Agalioti et al., 2002; An et al., 2002; Georges et al., 2002; Levenstein and Kadonaga, 2002). The power of the recombinant chromatin template assay is that it allows one to preclude the histone modification of interest by making amino acid substitutions at the physiological modification sites. Therefore a failure to promote transcription upon mutation of modification sites from histones will directly verify that the histone modifications are essential for transcription activation. Using this experimental system, a direct connection between targeted acetylations of H3-H4 and p300-mediated transcription has been established based on the transcription-inhibiting effects of substitution of major lysine substrates within H3-H4 (An et al., 2002). Although a specific combination of HAT and activator has been used in these experiments, the results clearly highlight the requirement of acetylated histone tails to activate transcription.

One major point still to be clarified concerns the molecular mechanism underlying the effects exerted by histone acetylation in regulating gene transcription. There are at least two possible mechanisms by which histone acetylation can facilitate transcription activation; charge neutralization effects versus protein recognition/recruitment effects. The effect of charge neutralization of histone tails by acetylation is consistent with the hypothesis that the chromatin fiber is stabilized by charge dependent interactions between basic histone tails and acidic patches on adjacent nucleosomes (Wade et al., 1997; Hansen et al., 1998). Thus histone acetylation neutralizes the positive charges of lysine side chains to affect the interaction between the lysine residues and the negatively charged DNA backbone. As a consequence, compacted chromatin would be destabilized to allow efficient binding of transcription machinery to the transcription initiation site. In this case, the contribution of acetylation to transcription should be rather cumulative (Ren and Gorovsky, 2001). Indeed, evidence of a direct effect of histone acetylation on the stability of nucleosomal arrays has been provided from in vitro studies by several groups (Garcia-Ramirez et al., 1995; Tse et al., 1998). A recent microarray study also showed the interchangeability of all acetylated lysines (K5, K8, K12) except K16 in H4 tails for transcription in yeast, providing support for a rather simple mechanism for the effect of histone acetylation in gene regulation (Dion et al., 2005).

In contrast to simple charge neutralization effects, the effects on protein recognition/recruitment are collectively referred to as the "histone code". This hypothesis predicts that specific patterns of histone tail acetylations and other modifications serve as epigenetic marks for distinct sets of regulatory proteins to differentially modulate chromatin structure and function (Strahl and Allis, 2000; Turner, 2000; Jenuwein and Allis, 2001). Indeed, several recent findings have demonstrated that histone acetylation creates a signal for the binding of a bromodomain which has been found to be present in many chromatin and transcription regulators such as GCN5, PCAF, p300/CBP, TAF250 and others (Zeng and Zhou, 2002). The most prominent example is that acetylation of specific lysines of H3-K9, K14 and H4-K8 functions as a cognate mark for the bromodomain-containing proteins BRG1 and TAFII250 to recruit SWI/SNF and TFIID complexes during in vitro activation of the IFN-6 gene (Agalioti et al., 2002). Selective recognition of acetylated histones by bromodomains has also been established in the intact nuclei of living cells by FRET analysis. This in vivo study showed that the bromodomain of transcriptional regulator BRD2 specifically recognizes acetylated H4-K12 whereas those of TAFII250 and PCAF recognize acetylated H3 and H4 with broader acetyl-lysine specificity (Kanno et al., 2004). Another bromodomain-containing protein GCN5 has also been found to be required for the yeast SAGA complex to associate with acetylated nucleosomes during chromatin remodeling (Hassan et al., 2002). All these results indicate that the physical interaction of acetylated histones with different bromodomain proteins is rather specific, supporting highly selective recognition/retention of a subpopulation of bromodomains by a certain acetylation mark in chromatin. However it is currently unknown whether all known bromodomains are able to recognize acetylated lysine.

Another important question to be addressed is whether different genes share a similar pattern of acetylation to get expressed or a unique pattern of acety-lation is associated with a distinct group of genes. A variety of HAT activities exhibit unique substrate specificity, discriminating between individual lysine residues of histones (Sterner and Berger, 2000; Roth et al., 2001); thus the interaction of a HAT with a gene specific activator that itself binds to a distinct promoter could explain the phenomenon of specificity of HAT and lysine acety-lation in certain transcription processes (Deckert and Struhl, 2001). The result that mutations of the acetylated residues of individual histones in yeast influence the expression of particular genes supports the specificity of HAT as a component of the transcription process (Grunstein et al., 1995). The most prominent example of such gene specificity includes the acetylations of H3-K9, K14 and H4-K8 by GCN5/PCAF, shown to be important for activation of the human interferon-^ gene upon viral infection (Agalioti et al., 2002). Additionally, it appears that acety-lation of histone tails can also be important for the recruitment of proteins without any bromodomain motif as in the case for interaction of acetylated H3 tails with CARM1 (Daujat et al., 2002). Therefore, it will be important to determine the specificity of histone acetylation in differential gene activation and how these different patterns of acetylation impart their selectivity. Considering that most recent studies utilized antibodies against di-acetylated H3 (on lysines 9 and 14) and tetra-acetylated H4 (on lysines 5, 8, 12 and 16) to identify histone acetylation at different genes (e.g., (-globin, PHO8, CATD, INO1 and HO), it is our challenge to investigate genome-wide acetylation profiles of all lysine substrates of the four core histones in human cells, which will contribute to a full understanding of the role of histone acetylation in transcription as well as transcriptional regulation per se.

Was this article helpful?

0 0
Single Parenting

Single Parenting

Finally! You Can Put All Your Worries To Rest! You Can Now Instantly Learn Some Little-Known But Highly Effective Tips For Successful Single Parenting! Understand Your Role As A Single Motherfather, And Learn How To Give Your Child The Love Of Both Parents Single Handedly.

Get My Free Ebook

Post a comment