and will constitute the pharmacophore. These drug-based functional groups are then
“clicked together” in three-dimensional space by being covalently attached to a rela-
tively rigid hydrocarbon frame. The number of functional groups determines the
number of contact points between the drug molecule and the receptor macromolecule.
A three-point pharmacophore will have three different intermolecular interactions
between the drug and the receptor. The number of points of contact is also an important
consideration. A large number of points of contact is favorable from a pharmacody-
namic perspective since it enables a more specific and unique drug–receptor interaction,
concomitantly decreasing the likelihood of toxicity. However, a large number of points
of contact is unfavorable from a pharmacokinetic perspective, since the resulting
increased polarity of the drug molecule tends to decrease the pharmacological half-life
and also to decrease the ability of the drug to diffuse across membranes during its dis-
tribution throughout the body. In general, most neuroactive drugs have 2–4 points of
contact, while most non-neuroactive drugs have 3–6 points of contact.
Once the pharmacophore has been designed, the remainder of the molecular frag-
ments (individually composed of metabophores or toxicophores or inert bioinactive
spacers, but collectively referred to as molecular baggage) are assembled. One of the
primary goals of the molecular baggage component is to hold the pharmacophore in a
desired conformation such that it can interact with its receptor. However, these addi-
tional molecular fragments also serve other functions. For instance, a metabophore can
be inserted into this portion of the molecule. If one is designing an intravenous drug
with a short half-life, one may want to include an ester moiety. This would constitute a
metabophore since the ester would be hydrolyzed, resulting in rapid inactivation of the
whole molecule.
Once the prototype drug molecule has been prepared and biologically evaluated, var-
ious toxicities may become apparent during preliminary testing in animals. Fragments
of the molecule that are responsible for these toxicities (i.e., toxicophores) can then be
deduced. If the toxicophore is separate and distinct from the pharmacophore, a new
toxicity-free molecule can be engineered. If there is too much overlap between the tox-
icophore and the pharmacophore, it may not be possible to “design out” the toxicity. It
is important to emphasize that a drug molecule may have many different toxicophores,
reflecting different toxicities. A toxicophore is merely a pharmacophore that permits an
undesirable interaction with an “untargeted” receptor.
The multiphore method is versatile and is not restricted to de novo drug design, as
the above discussion might imply. For example, if the drug molecule is discovered by
accident or in a random screening process, the multiphore conceptualization is still
applicable. Through structure–activity studies (discussed below) it is still possible to
discern fragments that constitute the pharmacophore and potential toxicophores, and
thus it is still possible to re-engineer the molecule for improved performance. The
strength of the multiphore method is its treatment of drug molecules as collections of
bioactive fragments. If one fragment is giving problems, it is possible to simply insert
another biologically similar fragment (bioisostere) that will hopefully overcome the
identified problem.
Clearly, the multiphore method of drug design is an iterative process. It takes
repeated rounds of re-evaluation and redesign before a final candidate drug molecule is
developed. This iterative process has the following five steps:
DESIGNING DRUG MOLECULES TO FIT RECEPTORS 107