The covalent attachment of a molecule to an enzyme or protein can modify its activity. In these instances, a donor molecule provides the functional moiety being attached. Most modifications are reversible. Phosphorylation and dephosphorylation are common means of covalent modification. The attachment of acetyl groups to lysine residues by acetyltransferases and their removal by deacetylases are another example. Histones—proteins that are packaged with DNA into chromosomes—are extensively acetylated and deacetylated in vivo on lysine residues (Section 31.3). More heavily acetylated histones are associated with genes that are being actively transcribed. Although protein acetylation was originally discovered as a modification of histones, we now know that it is a major means of regulation, with more than 2000 different proteins in mammalian cells regulated by acetylation. Protein acetylation appears to be especially important in the regulation of metabolism. The acetyltransferase and deacetylase enzymes are themselves regulated by phosphorylation, showing that the covalent modification of a protein can be controlled by the covalent modification of the modifying enzymes.

294

Modification is not readily reversible in some cases. The irreversible attachment of a lipid group causes some proteins in signal-transduction pathways, such as Ras (a GTPase) and Src (a protein tyrosine kinase), to become affixed to the cytoplasmic face of the plasma membrane. Fixed in this location, the proteins are better able to receive and transmit information that is being passed along their signaling pathways (Chapter 14). Mutations in both Ras and Src are seen in a wide array of cancers. The attachment of the small protein ubiquitin can signal that a protein is to be destroyed, the ultimate means of regulation (Chapter 23). The protein cyclin must be ubiquitinated and destroyed before a cell can enter anaphase and proceed through the cell cycle.

Virtually all the metabolic processes that we will examine are regulated in part by covalent modification. Indeed, the allosteric properties of many enzymes are altered by covalent modification. Table 10.1 lists some of the common covalent modifications.

We will see phosphorylation used as a regulatory mechanism in virtually every metabolic process in eukaryotic cells. Indeed, as much as 30% of eukaryotic proteins are phosphorylated. The enzymes catalyzing phosphorylation reactions are called protein kinases. These enzymes constitute one of the largest protein families known: there are more than 500 homologous kinases in human beings. This multiplicity of enzymes allows regulation to be fine-tuned according to a specific tissue, time, or substrate.

ATP is the most common donor of phosphoryl groups. The terminal (γ) phosphoryl group of ATP is transferred to a specific amino acid of the acceptor protein or enzyme. In eukaryotes, the acceptor residue is commonly one of the three containing a hydroxyl group in its side chain. Transfers to serine and threonine residues are handled by one class of protein kinases and to tyrosine residues by another. Tyrosine kinases, which are unique to multicellular organisms, play pivotal roles in growth regulation, and mutations in these enzymes are commonly observed in cancer cells.

295

Table 10.2 lists a few of the known serine and threonine protein kinases. The acceptors in protein-phosphorylation reactions are located inside cells, where the phosphoryl-group donor ATP is abundant. Proteins that are entirely extracellular are not regulated by reversible phosphorylation.

Protein kinases vary in their degree of specificity. Dedicated protein kinases phosphorylate a single protein or several closely related ones. Multifunctional protein kinases modify many different targets; they have a wide reach and can coordinate diverse processes. Comparisons of amino acid sequences of many phosphorylation sites show that a multifunctional kinase recognizes related sequences. For example, the consensus sequence recognized by protein kinase A is Arg-Arg-X-Ser-Z or Arg-Arg-X-Thr-Z, in which X is a small residue, Z is a large hydrophobic one, and Ser or Thr is the site of phosphorylation. However, this sequence is not absolutely required. Lysine, for example, can substitute for one of the arginine residues but with some loss of affinity. Thus, the primary determinant of specificity is the amino acid sequence surrounding the serine or threonine phosphorylation site. However, distant residues can contribute to specificity. For instance, a change in protein conformation can open or close access to a possible phosphorylation site.

Protein phosphatases reverse the effects of kinases by catalyzing the removal of phosphoryl groups attached to proteins. The enzyme hydrolyzes the bond attaching the phosphoryl group.

296

The unmodified hydroxyl-containing side chain is regenerated and orthophosphate (P_i) is produced. This family of enzymes, of which there are about 200 members in human beings, plays a vital role in cells because the enzymes turn off the signaling pathways that are activated by kinases. One class of highly conserved phosphatase called PP2A suppresses the cancer-promoting activity of certain kinases.

Importantly, the phosphorylation and dephosphorylation reactions are not the reverse of one another; each is essentially irreversible under physiological conditions. Furthermore, both reactions take place at negligible rates in the absence of enzymes. Thus, phosphorylation of a protein substrate will take place only through the action of a specific protein kinase and at the expense of ATP cleavage, and dephosphorylation will take place only through the action of a phosphatase. The result is that target proteins cycle unidirectionally between unphosphorylated and phosphorylated forms. The rate of cycling between the phosphorylated and the dephosphorylated states depends on the relative activities of kinases and phosphatases.

Phosphorylation is a common covalent modification of proteins in all forms of life, which leads to the question, What makes protein phosphorylation so valuable in regulating protein function that its use is ubiquitous? Phosphorylation is a highly effective means of controlling the activity of proteins for several reasons:

297

Let us examine a specific protein kinase that helps animals cope with stressful situations. The “flight or fight” response is common to many animals presented with a dangerous or exciting situation. Muscle becomes primed for action. This priming is the result of the activity of a particular protein kinase. In this case, the hormone epinephrine (adrenaline) triggers the formation of cyclic AMP (cAMP), an intracellular messenger formed by the cyclization of ATP. Cyclic AMP subsequently activates a key enzyme: protein kinase A (PKA). The kinase alters the activities of target proteins by phosphorylating specific serine or threonine residues. The striking finding is that most effects of cAMP in eukaryotic cells are achieved through the activation of PKA by cAMP.

PKA provides a clear example of the integration of allosteric regulation and phosphorylation. PKA is activated by cAMP concentrations near 10 nM. The quaternary structure is reminiscent of that of ATCase. Like that enzyme, PKA in muscle consists of two kinds of subunits: a 49-kDa regulatory (R) subunit and a 38-kDa catalytic (C) subunit. In the absence of cAMP, the regulatory and catalytic subunits form an R₂C₂ complex that is enzymatically inactive (Figure 10.16). The binding of two molecules of cAMP to each of the regulatory subunits leads to the dissociation of R₂C₂ into an R₂ subunit and two C subunits. These free catalytic subunits are enzymatically active. Thus, the binding of cAMP to the regulatory subunit relieves its inhibition of the catalytic subunit. PKA, like most other kinases, exists in isozymic forms for fine-tuning regulation to meet the needs of a specific cell or developmental stage. In mammals, four isoforms of the R subunit and three of the C subunit are encoded in the genome.

How does the binding of cAMP activate the kinase? Each R chain contains the sequence Arg-Arg-Gly-Ala-Ile, which matches the consensus sequence for phosphorylation except for the presence of alanine in place of serine. In the R₂C₂ complex, this pseudosubstrate sequence of R occupies the catalytic site of C, thereby preventing the entry of protein substrates (Figure 10.16). The binding of cAMP to the R chains allosterically moves the pseudosubstrate sequences out of the catalytic sites. The released C chains are then free to bind and phosphorylate substrate proteins. Interestingly, the cAMP-binding domain of the R subunit is highly conserved and found in all organisms.

298

X-ray crystallography reveals the three-dimensional structure of the catalytic subunit of PKA bound to ATP and a 20-residue peptide inhibitor. The 350-residue catalytic subunit of PKA has two lobes, an amino terminal N-lobe and a carboxyl terminal C-lobe (Figure 10.17). ATP and part of the inhibitor fill a deep cleft between the lobes. The N-lobe makes many contacts with ATP–Mn²⁺, whereas the C-lobe binds the peptide and contributes the key catalytic residues. As with other kinases, the two lobes move closer to one another on substrate binding; mechanisms that restrict this domain closure provide a means of regulating protein kinase activity. The PKA structure has broad significance because residues 40 to 280 constitute a conserved catalytic core, called the kinase fold, that is common to essentially all known protein kinases. We see here an example of a successful biochemical solution to a problem (in this case, protein phosphorylation) being employed many times in the course of evolution.

The bound peptide in this crystal occupies the active site because it contains the pseudosubstrate sequence Arg-Arg-Asn-Ala-Ile (Figure 10.18). The structure of the complex reveals the interactions by which the enzyme recognizes the consensus sequence. The guanidinium group of the first arginine residue forms an ion pair with the carboxylate side chain of a glutamate residue (Glu 127) of the enzyme. The second arginine likewise interacts with two other carboxylate groups. The nonpolar side chain of isoleucine, which matches Z in the consensus sequence, fits snugly in a hydrophobic groove formed by two leucine residues of the enzyme.

299