Protein Acetylation: Much More than Histone Acetylation
By Tom Brock, Ph.D.
Just last decade, everyone was excited about the Human Genome Project, as well as all the other genome projects, and the gene was king. Today, epigenetics is reminding us of something that we already knew, that non-genetic factors are important in shaping gene expression and development. Similarly, where phosphorylation once seemed the primary way to modulate proteins, epigenetics has re-introduced us to acetylation as an important force in defining protein function. In particular, the acetylation of histones has moved to center stage, even though it was described over 45 years ago. Research on histone acetylation has led to a resurgence in the interest in enzymatically-mediated acetylation of other proteins. This article examines acetylation as a post-translational modification of proteins that impacts gene expression and plays a role in epigenetics.
Acetylation refers to the addition of an acetyl group (CH3CO) to organic compounds. Proteins can be acetylated by both enzymatic and non-enzymatic processes. One group of acetyltransferases commonly catalyze the transfer of an acetyl group from acetyl-CoA to the terminal amine on the side chain of lysine residues (Figure 1). These enzymes are commonly called HATs, because their best-known substrates have been histones. However, the nomenclature is being revised to lysine acetyltransferases (KATs), reflecting their ability to acetylate lysine (denoted ‘K’) on many proteins.1 The KATs are numerous, with many assigned, based on structural similarities, to either the GNAT (Gcn5-related N-acetyltransferases) superfamily or the MYST (MOZ, YBF2/Sas3, Sas2, Tip60) family. Other important KATs include p300 (E1A-associated protein 300 kDa), CBP (cAMP response element binding (CREB)-binding protein), and TAFII 250 (TATA-binding protein associated factor II 250). The conversion of the positively charged lysine to acetyl-lysine, like the addition of negative phosphates to uncharged amino acids during phosphorylation, alters protein structure and interactions with other biomolecules. For example, acetylation of histones typically promotes the recruitment of effector proteins, relaxation of chromatin conformation, and an increase in transcription.
Like phosphorylation, acetylation is reversible. Histone deacetylases (HDACs, a.k.a. KDACs) are a smaller group of evolutionarily conserved enzymes. The human class I HDACs are homologous to the yeast enzyme Rpd3 and include HDAC1, 2, 3, and 8. Class II HDACs are homologous to yeast HDA1 and are divided into class IIa (HDAC4, 5, 7, 9) and class IIb (HDAC6 and 10) based on structure. The human class III HDACs include the sirtuin family of NAD+-dependent protein deacetylases. The novel HDAC11 has a distinct structure and is a class IV HDAC. The HDACs often participate in the formation of transcriptional repressor complexes, inducing chromatin compaction through histone deacetylation, and silencing gene expression.
A Diversity of Partners
A great resource for the research scientist is the National Center for Biotechnology Information (NCBI), your tax dollars at work compiling information about everything molecular. This site should be your first stopping point when trying to learn authoritative information about a new protein or gene that you’re studying. Information at this site helps to underscore two points about KATs and deacetylases: they are social enzymes, always interacting with other proteins, and they are promiscuous, binding to an astounding array of partners. Take, for example, the KAT known commonly as p300. At the NCBI gene link, entering ‘human p300’ finds the gene EP300 (KAT3B), with a summary stating that it associates with the adenovirus protein E1A, acetylates histones, binds CREB, and is a co-activator of HIF-1α (hypoxia-inducible factor 1α). Further down, we find that it binds three different proteins produced by the lentivirus human immunodeficiency virus (HIV)-1. Then, impressively, is a list of over two hundred proteins that have been documented to directly interact with p300 (with links to references and other interactome datasets included). Similarly, the deacetylase HDAC1 is summarized as a histone deacetylase that also interacts with retinoblastoma tumor-suppressor to control cell growth and, together with metastasis-associated protein-2, deacetylates the tumor suppressor p53. Like p300, HDAC1 has an amazing list of partners: it interacts with some 300 proteins, with over 125 of these documented as direct binding partners.
The abundance of protein partners, for both KATs and HDACs, suggests that these enzymes tend to form multimeric complexes. In fact, such complexes serve the critical purpose of positioning the (de)acetylases at specific sites to perform their functions. Certainly, KATs can directly acetylate substrates in vitro. However, KAT activity in vivo is regulated, at least in part, by where it is positioned. For example, the classical model for activation of PPARs (peroxisome proliferator-activated receptors) posits that this receptor heterodimerizes at specific response elements with RXR (retinoid X receptor). In the absence of ligand, the unactivated heterodimer binds co-repressor proteins, such as nuclear receptor co-repressors (NCoR), G-protein pathways suppressor 2 (GPS2), and HDACs (Figure 2). The HDACs help prevent expression of PPAR-specific genes by keeping the neighboring histones deacetylated. The appearance of a ligand for PPAR causes dissociation of the co-repressor proteins followed by the recruitment of co-activators, including PPAR co-activator (PGC-1), CREB binding protein (CBP), and p300. Formation of the PPAR activation complex leads to histone acetylation by CBP and p300, giving rise to altered expression of genes involved in fatty acid metabolism, lipid homeostasis, and adipocyte differentiation. In this example, ligand binding to its receptor causes a large scale switch from a cluster of proteins serving various roles in preventing transcription to a different group designed to facilitate gene transcription.
In its simplest form acetylation is merely another form of post-translational modification of proteins. A good example is the acetylation of tubulin, which can be deacetylated by HDAC6 or SIRT2. Acetylation of this key microtubule component appears to alter its affinity for kinesin-1 and redirect motor-based trafficking of vesicles.2,3 In short, acetylation changes protein function by adjusting protein-protein interactions. The net ‘global’ acetylation, in this case, may be determined by the balance of overall KAT and HDAC activities.
More commonly, acetylation is targeted to specific proteins and, possibly, specific lysine residues on those protein targets. One way that this can be achieved is by the formation of protein complexes containing either KATs or HDACs, as in the PPAR case described above. The assembly of the complex serves to place the KATs/HDACs near histones, transcription factors, or other targets. Histones, assembled as an octamer core surrounded by DNA, have amino termini that are freely exposed (Figure 3). Positively-charged lysine residues on these tails interact electrostatically with negatively-charged phosphate groups along the DNA backbone. Acetylation reduces these interactions and loosens the DNA, facilitating transcription. Bear in mind that, while it is generally true that histone acetylation increases transcriptional activation, there are exceptions. For example, acetylation of estrogen receptor-α suppresses ligand sensitivity and reduces ligand-induced transcriptional activity.4,5
In some cases, acetylation competes with other modifications.6,7 For example, the tumor suppressor p53 contains a lysine-rich basic domain near its carboxy terminus. Six different lysine residues, spanning sites 370-386 on human p53, can be modified by acetylation, methylation, ubiquitination, neddylation, or sumoylation. In addition, serines that can be targeted for phosphorylation are interspersed amongst the lysines. It is clear that acetylation facilitates p53 activation, leading to gene expression that is relevant to p53’s roles in responding to DNA damage and driving tumor suppression. At the other end of the response spectrum, ubiquitination of p53 targets it for degradation, preventing p53-mediated transcription and down-stream effects. Certain changes may predominate in the cytoplasm or in the nucleus, when p53 is associated with its negative regulator Mdm2, or when p53 monomers are forming homotetramers on gene-specific p53 response elements. For p53, acetylation may serve multiple roles, including stabilizing the protein, altering association with other proteins including other p53 monomers, enhancing its binding to DNA, and regulating transcription. By preventing ubiquitination, acetylation prevents the export and degradation of p53.
Acetylation and Epigenetics
While, strictly speaking, any mechanism for modifying gene expression (other than altering DNA sequence) constitutes an epigenetic change, the most interesting mechanisms are those that are long lasting. While acetylation marks can be readily removed by deacetylases, there are many ways to prolong acetylation. For example, positioning KATs in protein complexes next to specific targets may enable repeated acetylation of crucial residues, even if marks are spuriously removed. Levels of certain HDACs decline as cell differentiation progresses, reducing the rate of mark removal.8,9 Certain HDACs may be directly inhibited, as DBC-1 (deleted in breast cancer-1) does to SIRT1,10 preserving SIRT1-sensitive acetylation marks. Interestingly, acetylation of tubulin occurs on a site that is concealed following microtubule assembly,2 physically preventing HDACs from access until the tubulin becomes exposed. As yet, it is not known if there are proteins that physically protect certain acetylated targets, as 14-3-3 isomers shield specific phosphorylated proteins. Certainly, these and other ways to extend (or decrease) acetylation half times remain to be discovered.
1. Glozak, M.A., Sengubpta, N., Zhang, X., et al. Gene 363, 15-23 (2005).
2. Hammond, J.W., Cai, D., and Verhey, K.J. Curr. Opin. Cell Biol. 20, 71-76 (2008).
3. Gao, Y., Hubber, C.C., and Yao, T.P. J. Biol. Chem. epub ahead of print (2010).
4. Wang, C., Fu, M., Angeletti, R.H., et al. J. Biol. Chem. 276, 18375-18383 (2001).
5. Popov, V.M., Wang, C., Shirley, L.A., et al. Steroids 72, 221-230 (2007).
6. Mellert, H.S. and McMahon, S.B. Trends Biochem. Sci. 34, 571-578 (2009).
7. Yang, X.J. and Seto, E. Mol. Cell 31, 49-461 (2008).
8. Wilson, A.J., Byun, D.S., Popova, N., et al. J. Biol. Chem. 281, 13548-13558 (2006).
9. Vincent, A. and Van Seuningen, I. Differentiation 78, 99-107 (2009).
10. Li, Z., Chen, L., Kabra, N., et al. J. Biol. Chem. 284, 10361-10366 (2009).
Figure 1. The enzymatic interconversion of lysine and acetyl-lysine
Figure 2. Binding of a PPAR ligand to the PPAR ligand binding domain (LBD)
results in the release of co-repressor proteins, including NCoR, G-protein pathway suppressor 2 (GPS2), and histone deacetylase (HDAC), followed by the recruitment of PPAR co-activator (PGC-1), histone acetyltransferase p300, and CREB binding protein (CBP). Acetylation of histones by CBP and p300 relaxes chromatin, allowing transcription.
Figure 3. Nucleosomal structure, highlighting amino termini of histones projecting outward
Most figures are available in high-resolution formats for use in other works at a nominal fee.