The Histone Code Is Based on Reversible Covalent Modifications Histones and other DNA-binding proteins in chromatin are subject to extensive modification by acetylation, methylation, ADP-ribosylation, as well as phosphorylation and the addition of a small ubiquitin-like modifier (SUMO) protein, a process subbed sumoylation. These modifications modulate the manner in which the proteins within chromatin inter act with each other as well as the DNA itself. The resulting changes in chromatin structure within the region affected can render genes more accessible to the proteins responsible for their transcription, thereby enhancing gene expression or, on a larger scale, facilitating replication of the entire genome (see Chapter 38). On the other hand, changes in chromatin structure that restrict the accessibility of genes to transcription factors, DNA-dependent RNA polymerases, etc., thereby inhibiting transcription, are said to silence gene expression.

The combination of covalent modifications that determine gene accessibility in chromatin has been termed the “histone code.” This code represents a classic example of epigenetics, the hereditary transmission of information by the means other than the sequence of nucleotides that comprise the genome. In this instance, the pattern of gene expression within a newly formed “daughter” cell will be determined, in part, by the particular set of histone covalent modifications embodied in the chromatin proteins inherited from the “parental” cell.

Thousands of Mammalian Proteins Are Modified by Covalent Phosphorylation

Protein kinases phosphorylate proteins by catalyzing trans fer of the terminal phosphoryl group of ATP to the hydroxyl groups of select seryl, threonyl, or tyrosyl residues in proteins, formingO-phosphoseryl, O-phosphothreonyl, or O-phosphotyrosyl residues, respectively (Figure 1). The unmodified form of the protein can be regenerated by hydrolytic removal of phosphoryl groups, a thermodynamically favorable reaction catalyzed by protein phosphatases.

Fig1. Covalent modification of a regulated enzyme by phosphorylation–dephosphorylation of a seryl residue. Shown is the protease cascade responsible for activating pancreatic zymogens (Red) by partial proteolysis into their enzymatically active (Green) forms. The cascade is triggered by the brush border enzyme entero peptidase (Yellow), which converts trypsinogen to trypsin. Once activated, trypsin catalyzes the targeted proteolytic clips (Blue arrows) that transform chymotrypsinogen into chymotrypsin, proelastase into elastase, the procarboxypeptidases into carboxypeptidases, prophospholipase into phospholipase, and pancreatic prolipase into pancreatic lipase.

A typical mammalian cell possesses thousands of phosphorylated proteins and several hundred protein kinases and protein phosphatases that catalyze their interconversion. The ease of interconversion of enzymes between their phospho and dephospho forms accounts, in part, for the frequency with which phosphorylation–dephosphorylation is utilized as a mechanism for regulatory control. Unlike structural modifications, covalent phosphorylation persists only as long as the covalently modified form of the protein serves a specific need. Once the need has passed, the enzyme can be converted back to its original form, poised to respond to the next stimulatory event. A second factor underlying the widespread use of protein phosphorylation–dephosphorylation lies in the chemical properties of the phosphoryl group itself. In order to alter an enzyme’s functional properties, modifications to its chemical structure must influence the protein’s three-dimensional con figuration. The high-charge density of protein-bound phosphoryl groups, −1 or −2 at physiologic pH, their propensity to form strong salt bridges with arginyl and lysyl residues, and their high exceptionally high hydrogen-bonding capacity renders them potent agents for modifying protein structure and function. Phosphorylation generally influences an enzyme’s intrinsic catalytic efficiency or other properties by inducing changes in its conformation. Consequently, the amino acids modified by phosphorylation can be and typically are relatively distant from the catalytic site itself.

Protein Acetylation: A Ubiquitous Modification of Metabolic Enzymes

As is the case with covalent phosphorylation, covalent acetylation possesses the twin virtues of employing the conformation altering potential of changing the charge character of the side chain which they target, in this case from the +1 of the protonated ε-amino group of lysine to neutral acetylated form, and being reversible in vivo. It is thus not surprising that the number of proteins subject to and regulated by covalent acetylation–deacetylation now numbers in the thousands. These include histones and other nuclear proteins as well as nearly every enzyme in such core metabolic pathways as glycolysis, glycogen synthesis, gluconeogenesis, the tricarboxylic acid cycle, β-oxidation of fatty acids, and the urea cycle. The potential regulatory impact of acetylation–deacetylation has been established for only a handful of these proteins. However, the latter include metabolically important enzymes such as acetyl-CoA synthetase, long-chain acyl-CoA dehydrogenase, malate dehydrogenase, isocitrate dehydrogenase, glutamate dehydrogenase, carbamoyl phosphate synthetase, phosphoenol-pyruvate carboxykinase, aconitase, and ornithine trans carbamoylase. Perturbations in the acetylation–deacetylation of lysine residues in proteins is thought to be associated with aging and neurodegeneration.

In the mitochondria, it is believed that many proteins react with acetyl-CoA directly, without the intervention of an enzyme catalyst. Their degree of acetylation thus is thought to respond to and reflect changes in the concentration of this central metabolic intermediate. The acetylation of other proteins, particularly those residing outside the mitochondria, requires the intervention of a lysine acetyltransferase. These enzymes catalyze the transfer of the acetyl group of acetyl-CoA to the ε-amino groups of lysyl residues, forming N-acetyl lysine.

All protein deacetylation is believed to be enzyme catalyzed. Two classes of protein deacetylases have been identified: histone deacetylases and sirtuins. Histone deacetylases catalyze the removal by hydrolysis of acetyl groups, regenerating the unmodified form of the protein and acetate as products. Sirtuins, on the other hand, use NAD+ as substrate, which yields O-acetyl ADP-ribose and nicotinamide as products in addition to the unmodified protein.

Covalent Modifications Regulate Metabolite Flow

In many respects, the sites of protein phosphorylation, acetylation, and other covalent modifications can be considered another form of allosteric site. However, in this case, the “allo steric ligand” binds covalently to the protein. Phosphorylation dephosphorylation, acetylation–deacetylation, and feedback inhibition provide short-term, readily reversible regulation of metabolite flow in response to specific physiologic signals. All three act independently of changes in gene expression. As with feedback inhibition, protein phosphorylation–dephosphorylation generally targets an early enzyme in a protracted metabolic pathway. Feedback inhibition involves a single protein that is influenced indirectly, if at all, by hormonal or neural signals. By contrast, regulation of mammalian enzymes by phosphorylation–dephosphorylation involves one or more protein kinases and protein phosphatases, and is generally under direct neural and hormonal control.

Acetylation–deacetylation, on the other hand, targets multiple proteins in a pathway. It has been hypothesized that the degree of acetylation of metabolic enzymes is modulated to a large degree by the energy status of the cell. Under this model, the high levels of acetyl-CoA (the substrate for lysine acetyltransferases and the reactant in nonenzymatic lysine acetylation) present in a well-nourished cell would promote lysine acetylation. When nutrients are lacking, acetyl-CoA levels drop and the ratio of NAD+/NADH rises, favoring protein deacetylation.