This is particularly true if one is pursuing lead compound discovery by high throughput
screening of millions of compounds.
More importantly, the biological model should accurately reflect the human disease
for which it is being used as a drug-screening tool. Just because an animal model pro-
duces a similar disease as is found in humans, there is no guarantee that it will truly
reflect the corresponding human disease. For example, occluding the middle cerebral
artery in a gerbil will produce a “gerbil stroke.” However, are strokes in gerbils identi-
cal to strokes in humans? A drug that successfully treats strokes in gerbils may not nec-
essarily treat strokes in humans. Alternatively, there may be diseases that are unique to
humans or primates, thereby making it difficult to develop meaningful biological assays
in species such as rodents. Alzheimer’s disease is a superb example of this dilemma. For
years, drug design in Alzheimer’s disease was delayed by the absence of a reasonable
animal model. Modern advanced molecular biology techniques are addressing this issue
by enabling the engineering of rats that over-express the β-amyloid protein that seems
to cause Alzheimer’s disease in humans.
Biological assays for compound evaluation may be broadly categorized as follows:
- In silico—theoretical evaluation achieved by computing simulation
- In vitro—“test-tube” evaluation done without a whole animal
a. Binding studies (e.g., competition studies using radioactive ligands)
b. Functional assay studies (e.g., functional assays using enzymes) - In vivo—evaluation done with a whole animal
Each of these models has its strengths and weaknesses.
Thein silicomethods are the most rapid and cost effective. They are also the ones
that are most divorced from reality. Consequently,in silicomethods are acceptable for
preliminary screens, but are completely unacceptable for the advanced assessment of a
candidate compound.
In vitromethods are a definite step up. Radiolabeled competitive binding studies can be
used to ascertain whether a drug binds to a receptor. Such studies simply give information
DESIGNING DRUG MOLECULES TO FIT RECEPTORS 131Figure 3.4 The synthesis of ibuprofen is initiated by a Friedel-Crafts acylation of an alkyl-
substituted benzene ring. The resulting ketone is then reduced to an alcohol with sodium boro-
hydride. The alcohol functionality then undergoes a functional group interchange by conversion
to a bromide. In turn, this permits the introduction of an additional carbon atom in the form of
a nitrile introduced via an SN2 nucleophilic displacement. This is then hydrolyzed to give the
target molecule.