Carbon dioxide is a major end product of aerobic metabolism. In complex organisms, this carbon dioxide is released into the blood and transported to the lungs for exhalation. While in the blood, carbon dioxide reacts with water. The product of this reaction is a moderately strong acid, carbonic acid (pK_a = 3.5), which becomes bicarbonate ion on the loss of a proton.

Even in the absence of a catalyst, this hydration reaction proceeds at a moderate pace. At 37°C near neutral pH, the second-order rate constant k₁ is 0.0027 M^-1 s^-1. This corresponds to an effective first-order rate constant of 0.15 s^-1 in water ([H₂O] = 55.5 M). Similarly, the reverse reaction, the dehydration of bicarbonate, is relatively rapid, with a rate constant of k_-1 = 50 s^-1. These rate constants correspond to an equilibrium constant of K₁ = 5.4 × 10^-5 and a ratio of [CO₂] to [H₂CO₃] of 340:1.

Despite the fact that CO₂ hydration and HCO₃^-dehydration occur spontaneously at reasonable rates in the absence of catalysts, almost all organisms contain enzymes, referred to as carbonic anhydrases, that catalyze these processes. Such enzymes are required because CO₂ hydration and HCO₃^- dehydration are often coupled to rapid processes, particularly transport processes. For example, HCO₃^- in the blood must be dehydrated to form CO₂ for exhalation as the blood passes through the lungs. Conversely, CO₂ must be converted into HCO₃^- for the generation of the aqueous humor of the eye and other secretions. Furthermore, both CO₂ and HCO₃^- are substrates and products for a variety of enzymes, and the rapid interconversion of these species may be necessary to ensure appropriate substrate levels. So important are these enzymes in human beings that mutations in some carbonic anhydrases have been found to cause osteopetrosis (excessive formation of dense bones accompanied by anemia) and mental retardation.

Carbonic anhydrases accelerate CO₂ hydration dramatically. The most active enzymes, typified by human carbonic anhydrase II, hydrate CO₂ at rates as high as k_cat = 10⁶ s^-1, or a million times a second. Fundamental physical processes such as diffusion and proton transfer ordinarily limit the rate of hydration, and so special strategies are required to attain such prodigious rates.

Less than 10 years after the discovery of carbonic anhydrase in 1932, this enzyme was found to contain bound zinc, associated with catalytic activity. This discovery, remarkable at the time, made carbonic anhydrase the first known zinc-containing enzyme. At present, hundreds of enzymes are known to contain zinc. In fact, more than one-third of all enzymes either contain bound metal ions or require the addition of such ions for activity. The chemical reactivity of metal ionsassociated with their positive charges, with their ability to form relatively strong yet kinetically labile bonds, and, in some cases, with their capacity to be stable in more than one oxidation stateexplains why catalytic strategies that employ metal ions have been adopted throughout evolution.

The results of x-ray crystallographic studies have supplied the most detailed and direct information about the zinc site in carbonic anhydrase. At least seven carbonic anhydrases, each with its own gene, are present in human beings. They are all clearly homologous, as revealed by substantial levels of sequence identity. Carbonic anhydrase II, present in relatively high concentrations in red blood cells, has been the most extensively studied (Figure 9.22).

Zinc is found only in the + 2 state in biological systems; so we need consider only this oxidation level as we examine the mechanism of carbonic anhydrase. A zinc atom is essentially always bound to four or more ligands; in carbonic anhydrase, three coordination sites are occupied by the imidazole rings of three histidine residues and an additional coordination site is occupied by a water molecule (or hydroxide ion, depending on pH). Because all of the molecules occupying the coordination sites are neutral, the overall charge on the Zn(His)₃ unit remains +2.

How does this zinc complex facilitate carbon dioxide hydration? A major clue comes from the pH profile of enzymatically catalyzed carbon dioxide hydration (Figure 9.23). At pH 8, the reaction proceeds near its maximal rate. As the pH decreases, the rate of the reaction drops. The midpoint of this transition is near pH 7, suggesting that a group with pK_a = 7 plays an important role in the activity of carbonic anhydrase and that the deprotonated (high pH) form of this group participates more effectively in catalysis. Although some amino acids, notably histidine, have pK_a values near 7, a variety of evidence suggests that the group responsible for this transition is not an amino acid but is the zinc-bound water molecule. Thus, the binding of a water molecule to the positively charged zinc center reduces the pK_a of the water molecule from 15.7 to 7 (Figure 9.24). With the lowered pK_a, a substantial concentration of hydroxide ion (bound to zinc) is generated at neutral pH. A zinc-bound hydroxide ion is sufficiently nucleophilic to attack carbon dioxide much more readily than water does. The importance of the zinc-bound hydroxide ion suggests a simple mechanism for carbon dioxide hydration (Figure 9.25).

As noted earlier, some carbonic anhydrases can hydrate carbon dioxide at rates as high as a million times a second (10⁶ s^-1). The magnitude of this rate can be understood from the following observations. At the conclusion of a carbon dioxide hydration reaction, the zinc-bound water molecule must lose a proton to regenerate the active form of the enzyme (Figure 9.27). The rate of the reverse reaction, the protonation of the zinc-bound hydroxide ion, is limited by the rate of proton diffusion. Protons diffuse very rapidly with second-order rate constants near 10^-11 M^-1 s^-1. Thus, the backward rate constant k_-1 must be less than 10¹¹ M^-1 s^-1. Because the equilibrium constant K is equal to k₁/k_-1, the forward rate constant is given by k₁ = K · k_-1. Thus, if k_-1 10¹¹ M^-1 s^-1 and K = 10^-7 M (because pK_a = 7), then k₁ must be less than or equal to 10⁴ s^-1. In other words, the rate of proton diffusion limits the rate of proton release to less than 10⁴ s^-1 for a group with pK_a= 7. However, if carbon dioxide is hydrated at a rate of 10⁶ s^-1, then every step in the mechanism (see Figure 9.25) must take place at least this fast. How can this apparent paradox be resolved?

The answer became clear with the realization that the highest rates of carbon dioxide hydration require the presence of buffer, suggesting that the buffer components participate in the reaction. The buffer can bind or release protons. The advantage is that, whereas the concentrations of protons and hydroxide ions are limited to 10^-7 M at neutral pH, the concentration of buffer components can be much higher, on the order of several millimolar. If the buffer component BH⁺ has a pK_a of 7 (matching that for the zinc-bound water), then the equilibrium constant for the reaction in Figure 9.28 is 1. The rate of proton abstraction is given by k₁ · [B]. The second-order rate constants k₁ and k_-1 will be limited by buffer diffusion to values less than approximately 10⁹ M^-1 s^-1. Thus, buffer concentrations greater than [B] = 10^-3 M (1 mM) may be high enough to support carbon dioxide hydration rates of 10⁶ M^-1 s^-1 because k₁ · [B] = (10⁹ M^-1 s^-1) · (10^-3 M) = 10⁶ s^-1. This prediction is confirmed experimentally (Figure 9.29).

The molecular components of many buffers are too large to reach the active site of carbonic anhydrase. Carbonic anhydrase II has evolved a proton shuttle to allow buffer components to participate in the reaction from solution. The primary component of this shuttle is histidine 64. This residue transfers protons from the zinc-bound water molecule to the protein surface and then to the buffer (Figure 9.30). Thus, catalytic function has been enhanced through the evolution of an apparatus for controlling proton transfer from and to the active site. Because protons participate in many biochemical reactions, the manipulation of the proton inventory within active sites is crucial to the function of many enzymes and explains the prominence of acid-base catalysis.

Carbonic anhydrases homologous to the human enzymes, referred to as a-carbonic anhydrases, are common in animals and in some bacteria and algae. In addition, two other families of carbonic anhydrases have been discovered. The b-carbonic anhydrases are found in higher plants and in many bacterial species, including E. coli. These proteins contain the zinc required for catalytic activity but are not significantly similar in sequence to the a-carbonic anhydrases. Furthermore, the b-carbonic anhydrases have only one conserved histidine residue, whereas the a- carbonic anhydrases have three. No three-dimensional structure is yet available, but spectroscopic studies suggest that the zinc is bound by one histidine residue, two cysteine residues (conserved among b-carbonic anhydrases), and a water molecule. A third family, the g-carbonic anhydrases, also has been identified, initially in the archaeon Methanosarcina thermophila. The crystal structure of this enzyme reveals three zinc sites extremely similar to those in the a-carbonic anhydrases. In this case, however, the three zinc sites lie at the interfaces between the three subunits of a trimeric enzyme (Figure 9.31). The very striking left-handed b-helix (a b strand twisted into a left-handed helix) structure present in this enzyme has also been found in enzymes that catalyze reactions unrelated to those of carbonic anhydrase. Thus, convergent evolution has generated carbonic anhydrases that rely on coordinated zinc ions at least three times. In each case, the catalytic activity appears to be associated with zinc-bound water molecules.