Here is a picture that shows the reason for the selectivity.

Chymotrypsin and Synthetic Substrate

The substrate must fit against the enzyme with the benzene ring of the phenylalanine in the large hole at the upper left, so that the peptide bond is near the serine side chain that it will react with. If we switch the configuration of the phenylalanine, the benzene ring would project outward, and the substrate would not fit tightly into the enzyme catalytic site.

The classic demonstration of stereospecificity in an enzymatic reaction is the oxidation of ethanol by alcohol dehydrogenases.

• These are zinc metalloenzymes, that use NAD⁺ as a coenzyme; they oxidize a wide range of alcohols, and in fact are among the least substrate specific enzymes.

• However, they are very stereospecific, as was first established by Westheimer [J. Am. Chem. Soc., 1951, 73, 2043; J. Biol. Chem., 1953, 202, 687] for the oxidation of ethanol in a classic series of experiments.

Horse liver alcohol dehydrogenase, one of the most studied (along with the yeast enzyme), is a symmetrical dimer. Each chain is about 40 kDaltons, contains two Zn⁺⁺ (orange), only one of which is catalytically active, and each binds a single NAD⁺ (green).

Horse Liver Alcohol Dehydrogenase (2ohx)

Carbon-1 of ethanol is a prochiral center, bearing enantiotopic hydrogens.

Similarly, acetaldehyde, the product of oxidation is also a prochiral species: delivery of a ligand from one face of the carbonyl produces the enantiomer of delivery from the other face.

• Such prochiral faces are designated re and si (rectus and sinister) for the stereochemistry produced when the new ligand is delivered.

The reagent that accepts H from ethanol, NAD⁺, likewise has prochiral faces; when the new ligand is attached, carbon-4 of the NADH is a prochiral center.

Since the R group is chiral, the faces and ligands are diastereotopic.

Of course, when the redox transformation involves only hydrogens, one cannot distinguish the stereochemical course. However, if one uses deuterium labeling, one discovers that:

• The pro-R hydrogen of ethanol is removed

• The hydrogen is transferred to the re face of NAD⁺

• If the reaction is run in reverse, the pro-R H of NADH is transferred to the re face of acetaldehyde

The implication of this stereospecificity is that when ethanol and NAD⁺ are bound to the enzyme, their binding sites must orient them so that the pro-R H of ethanol is directed toward the re face of the NAD⁺, as in this model:

Therefore, the enzyme must bind at least two of the groups attached to the prochiral center, leaving the orientation of the NAD⁺ ring to distinguish between the two hydrogens.

Here are a couple of views of the active site of the horse enzyme:

Horse LADH with NAD+ Bound (2ohx)
Horse LADH with NAD+ and Trifluoroethanol (1a71)

• The zinc (orange) is coordinated to His-67, Cys-174, Ser-48, and a water molecule (not visible)

• Just above the Zn is the hydride acceptor ring of NAD⁺

• The alcohol binds to the Zn replacing the water molecule, and thus must lie between the Zn and the pyridinium ring.

• Leu-57, Phe-93 (mutated to Trp in this case), and His-51 form the rest of the binding pocket