The basic tenets of molecular recognition may already be found in the "lock and key" concept, developed by Emil Fischer (1894) in his work with enzymes. The substrate—the "key"—must fit geometrically into a cleft on the receptor—the "lock". It was later recognized that such complementarity extends over energetic features as well as over the geometrical ones. A high level of recognition of a receptor molecule by a particular substrate σ indicates there exists a large difference between the binding free energies of σ and that of other substrates, and thus a selective binding. One may also distinguish between positive and negative recognition, depending on whether the discrimination among different substrates by a given receptor is dominated by attractive or repulsive interactions.
Factors that lead to a high degree of molecular recognition are:
Steric (shape and size) complementarity. The presence of convex and concave domains must be in the correct locations on the receptor molecule and on the substrate.
Interactional complementarity. Presence of complementary binding sites (positive/ negative charges, charge/dipole, dipole/dipole, hydrogen bond donor/acceptor, etc.).
Large areas of contact between receptor and substrate. This allows for:
Multiple interaction sites. Meaning more locations on the receptor and the substrate to bind together, leading to:
Strong overall binding. Although high stability does not in principle imply high selectivity, this is usually the case. Once the molecule is bound, it does not drift away again or does so very slowly.
Inclusion and Dynamics. In addition, requiring the receptor ρ and substrate σ to be in contact over a large area will be satisfied if ρ is able to wrap around σ so as to establish numerous noncovalent binding interactions and to sense its molecular size, shape, and architecture. This is the case for receptor molecules that contain intramolecular cavities, clefts, or pockets into which the substrate may fit. In such concave receptors, termed endoreceptors, the cavity is lined with binding sites directed toward the bound species. In addition to maximizing contact area, inclusion also leads to more or less completely excluding any solvent from the receptor site, thus minimizing the number of solvent molecules that need to be displaced by the substrate on binding.
The balance between rigidity and flexibility is of particular importance for the binding and the dynamic properties of ρ and σ. Rigid, "lock and key"–type receptors are expected to present very efficient recognition with both high stability and high selectivity. More flexible receptors that bind to the substrate through an "induced fit" process may display high selectivity but have lower stability, because part of the binding energy is used up in the change of conformation of the receptor. The trade-off here is that ρ can adapt and respond to changes. Flexibility is of great importance in biological receptor-substrate interactions, where adaptation is often required. Processes of exchange, regulation, cooperativity, and allostery all require a built-in flexibility.
Medium effects on molecular recognition. It must be remembered that none of this occurs in isolation. Because effects of the surrounding medium also come into play through the interaction of solvent molecules with the receptor and with the substrate, both should present geometrically matched hydrophobic or hydrophilic domains.
Molecular recognition is markedly affected by the surrounding medium. The medium can either enhance or decrease recognition, through enthalpies and entropies of the formation of a supermolecule. Examples are the negative entropies of the formation of alkali cryptates when occurring in an aqueous solution, (Kauffmann, Lehn, and Sauvage 1976) and the increasing exothermicity of binding to cyclophane receptors as the polarity of the solvent increases (Diederich et al. 1992). The medium can also have an effect on the shape of the receptor molecules themselves—for example, hydrophobic effects may deform the receptor. Enzymes can lose all of their activity through a relatively small change in pH of the medium, which causes the enzyme to lose its tertiary protein structure—that is, it uncoils. Shape modifications can strongly influence binding properties. Such medium effects can be illustrated by the two different crystalline forms of the water-soluble salt of a macrobicyclic cyclophane, which can be crystallized in two different shapes depending on the medium: an inflated cage structure building up cylinders in a hexagonal array; and a flattened structure stacked in molecular layers separated by aqueous layers in a lamellar arrangement (Cesario et al. 1993). Such medium effects may be expected to come into play when functional molecules are incorporated into membrane phases as well as in the determination of the form and function of biomolecules.