Figure 8-18 The structure of chymotrypsin. (a) A representation of primary structure, showing disulfide bonds and the location of key amino acids. Note that the protein consists of three polypeptide chains. The active-site amino acids are found grouped together in the three-dimensional structure. (b) A space-filling model of chymotrypsin. The pocket in which the aromatic amino acid side chain is bound is shown in green. The amide nitrogens of G1y¹⁹³ and Ser¹⁹⁵ in the polypeptide backbone make up the oxyanion hole (see Fig. 8-19), and are shown in orange. The side chains of other key active site residues, including Ser^l95, His⁵⁷, and Asp¹⁰² are shown in red, and are explained in Fig.(d) 8-19. (c) The polypeptide backbone of chymotrypsin shown as a ribbon structure. Disulfide bonds are shown in yellow; the A, B, and C chains are shown in dark blue, light blue, and white, respectively.(d) A close-up of the chymotrypsin active site with a substrate bound. Ser¹⁹⁵ attacks the carbonyl group of the substrate (shown in purple); the developing negative charge on the oxygen is stabilized by the oxyanion hole (amide nitrogens shown in orange), as explained in Fig. 8-19. In the substrate, the aromatic amino acid side chain and the amide nitrogen of the peptide bond to be cleaved are shown in light blue.

The transition state of a reaction is difficult to study, because by definition it has no fmite lifetime. To understand enzymatic catalysis, however, we must dissect the interaction between the enzyme and this ephemeral moment in the course of a reaction. The idea that an enzyme is complementary to the transition state is virtually a requirement for catalysis, because the energy hill upon which the transition state sits is what the enzyme must lower if catalysis is to occur. How can we obtain evidence that the idea of enzyme-transition state complementarity is really correct? Fortunately, there are a variety of approaches, old and new, to this problem. Each has provided compelling evidence in support of this general principle of enzyme action.

If enzymes are complementary to reaction transition states, then some functional groups in the substrate and in the enzyme must interact preferentially with the transition state rather than the ES complex. Altering these groups should have little effect on formation of the ES complex, and hence should not affect kinetic parameters (K_s, or sometimes K_m if K_s = K_m) that reflect the E + S ES equilibrium. Changing the same groups, however, should have a large effect on the overall rate (k_cat or k_cat/K_m) of the reaction, because the bound substrate lacks potential binding interactions needed to lower the activation energy.

Figure 1 Effects of small structural changes in the substrate on kinetic parameters for chymotrypsin-catalyzed amide hydrolysis.

An excellent example is seen in a series of substrates for the enzyme chymotrypsin (Fig. 1). Chymotrypsin normally catalyzes the hydrolysis of peptide bonds next to aromatic amino acids, and the substrates shown in Figure 1 are convenient smaller models for the natural substrates (long polypeptides and proteins; see Chapter 6). The additional chemical groups added in going from A to B to C are shaded in red. Note that the interaction between the enzyme and these added functional groups has a minimal effect on K_m (which is taken here as a reflection of KS), but a large, positive effect on k_cat and k_cat/K_m. This is what we would expect if the interaction occurred only in the transition state. Chymotrypsin is described in more detail beginning on page 223.

A complementary experimental approach to this problem is to modify the enzyme, eliminating certain enzyme-substrate interactions, by replacing specific amino acids through site-directed mutagenesis (Chapters 7 and 28). A good example is found in tyrosyl-tRNA synthetase (p. 227).

Even though transition states cannot be observed directly, chemists can often predict the approximate structure of a transition state based on accumulated knowledge about reaction mechanisms. The transition state by definition is transient and so unstable that direct measurement of the binding interaction between this species and the enzyme is impossible. In some cases, however, stable molecules can be designed that resemble transition states. These are called transition-state analogs. In principle, they should bind to an enzyme more tightly than the substrate binds in the ES complex, because they should fit in the active site better (i.e., form more weak interactions) than the substrate itself. The idea of transition-state analogs was suggested by Pauling in the 1940s, and it has been used for a number of enzymes. These experiments have the limitation that a transitionstate analog can never mimic a transition state perfectly. Analogs have been found, however, that bind an enzyme 10² to 10⁶ times more tightly than the normal substrate, providing good evidence that enzyme active sites are indeed complementary to transition states.

If a transition-state analog can be designed for the reaction S → P, then an antibody that binds tightly to the transition-state analog might catalyze S → P. Antibodies (immunoglobulins; see Fig. 6-8) are key components of the immune response. A molecule or chemical group that is bound tightly and specifically by a given antibody is referred to as an antigen. When a transition-state analog is used as an antigen to stimulate the production of antibodies, the antibodies that bind it are potential catalysts of the corresponding reaction. This approach, first suggested by William P. Jencks in 1969, has become practical with the development of laboratory techniques to produce antibodies that are all identical and bind one specific antigen (these are known as monoclonal antibodies; see Chapter 6).

Pioneering work in the laboratories of Richard Lerner and Peter Schultz has resulted in the isolation of a number of monoclonal antibodies that catalyze the hydrolysis of esters or carbonates (Fig. 2).

Figure 2 The expected transition states for ester or carbonate hydrolysis reactions. Phosphonate and phosphate compounds, respectively, make good transition-state analogs for these reactions.

In these reactions, the attack by water (OH^-) on the carbonyl carbon produces a tetrahedral transition state in which a partial negative charge has developed on the carbonyl oxygen. Phosphonate compounds mimic the structure and charge distribution of this transition state in ester hydrolysis, making them good transition-state analogs; phosphate compounds are used for carbonate reactions.Antibodies that bind the phosphonate or phosphate tightly have been found to catalyze the corresponding ester or carbonate hydrolysis reaction by factors of 10³ to 10⁴.Structural analyses of a few of these catalytic antibodies have shown that the catalytic amino acid side chains are arranged where they could interact with the substrate only in the transition state. These studies provide additional evidence for enzyme-transition state complementarity and suggest that new classes of antibody catalysts might be developed for research and industry.

Hexokinase This is a bisubstrate enzyme (M_r 100,000), catalyzing the interconversion of glucose and ATP with glucose-6-phosphate and ADP. The hydroxyl at position 6 of the glucose molecule (to which the γ-phosphate of ATP is transferred) is similar in chemical reactivity to water, and water freely enters the enzyme active site. Yet hexokinase discriminates between glucose and water, with glucose favored by a factor of 10⁶.

Figure 8-20 Observed kinetics of the hydrolysis of p-nitrophenylacetate (p-NPA) by chymotrypsin as measured by release of p-nitrophenol (a colored product). A rapid release (burst) of an amount of p-nitrophenol nearly stoichiometric with the amount of enzyme present is observed. This reflects the fast acylation phase of the reaction. The subsequent rate is slower because enzyme turnover is limited by the rate of the slower deacylation phase.

Hexokinase can discriminate between glucose and water because of a conformational change in the enzyme that occurs when the correct substrates are bound (Fig. 8-21). The enzyme thus provides a good example of induced fit. When glucose is not present, the enzyme is in an inactive conformation with the active-site amino acid side chains out of position for reaction. When glucose (but not water) and ATP bind, the binding energy derived from this interaction induces a change to the catalytically active enzyme conformation.

Figure 8-21 The conformational change induced in hexokinase by the binding of a substrate (D-glucose, shown in blue).

Hydrogen bonds formed in the ES complex affect K_s, which can be directly measured for this enzyme. Hydrogen bonds formed only in the transition state affect k_cat.In several cases, substitution of a nonhydrogen-bonding amino acid at key positions in the active site affectskcat but does not affect K_s. For example, substitution of an alanine for Thr⁴⁰and a glycine for His⁴⁵ has little effect on the observed K_s for ATP.

Fignre 8-22 The structure of tyrosyl-tRNA synthetase. At left, the enzyme is shown without substrate bound. Active site residues that hydrogenbond to the tyrosyl-AMP intermediate (Fig. 8-23a) are shown in red, and two residues (Thr⁴⁰and His⁴⁵) that contribute hydrogen bonds in the reaction transition state (Fig. 8-23b) are shown in orange. The amino acid side chains of Lys⁸² and Arg⁸⁶, which cover part of the active site, are not shown in order to expose more of the active site residues. At right, the same view is shown with the tyrosyl-AMP intermediate (shown in blue, with a phosphorus atom in yellow) bound.