Biology Reference
In-Depth Information
mammalian cells, yet other approaches are available to
address this issue (see the Chemogenomics section).
therapies; however, 'designing' polypharmacology will
require screens that can accommodate drug effects on
multiple targets, and allow the measurement and character-
ization of the cellular response, to guide our understanding of
the net effect of multiple perturbations. More simply,
designing polypharmacology will require a much deeper
understanding of cellular biology and complex disease. The
approach must first be tackled by understanding the cellular
response to drug.
The development of chemogenomic technologies allows
the cellular response to drug to be surveyed across the
genome. These technologies are based on chemical
Drugs are Promiscuous; the Discovery
of Polypharmacology
Molecular biology has long relied on the reductionist
approach of intensive study of individual genes for a better
understanding the cell. This approach has been tremen-
dously successful, and is responsible for most of the major
biological breakthroughs. However, as previously dis-
cussed, the reductionist approach does not effectively allow
for the study of interaction-dependent gene function.
Modern drug discovery adopted the molecular biologist's
reductionist
genetic
strategies and allow comprehensive interrogation of
compound-target relationships in the context of the living
cell. Chemogenomic approaches have proved to be powerful
tools that provide a genome-wide view of both the cellular
targets and the biological networks or processes disrupted by
the inhibition of the drug target.
e
approach,
revising
the
one
gene
e
one
protein
e
one function concept only slightly, to one gene-
e
one drug. This 'forward' or
'target-oriented' drug discovery approach can be practically
described as the concept that a small molecule with
demonstrated ability to bind or inhibit a protein target with
high potency in vitro will translate into a beneficial medical
therapy in a human patient. As discussed above, this 'magic
bullet' philosophy has rarely delivered, indicating that the
effectiveness of a reductionist approach has reached its
limits in MDD as well.
One explanation for the collective failure of target-
oriented screening may lie in the recent 'discovery' that
many drugs exhibit 'polypharmacology', defined as inter-
acting with more than one protein target in the cell.
Paralleling the interaction-pervasive behavior of genes, the
discovery of 'promiscuous' drug behavior has proved with
time to be much more prevalent than had been expected.
Somewhat unexpectedly, polypharmacology is not limited
to interactions with proteins in the same family as the
intended target, but can involve proteins that seemingly
share no structural or any other similarity with the intended
target. Computational analyses that aim to predict poly-
pharmacology have met with some success: specific scaf-
folds have been demonstrated to be promiscuous, and
specific chemical properties are characteristic, though not
definitive, of drugs
one protein
e
one disease
e
Phenotypic Screening
Consistent with the accumulating failures of target-oriented
screening, current trends in drug discovery suggest a re-
emergence of phenotypic screens in drug discovery,
bolstered by improved robotics, better imaging and image
analysis algorithms. Phenotypic screens search for small
molecules in whole cells or organisms that reverse
a phenotype of interest, such as one that may be charac-
teristic of a particular disease. A recent analysis
[33]
reported that a surprisingly high percentage (37% cell-
based compared to 23% target-based) of the novel molec-
ular entities (NMEs) discovered in the last decade resulted
from cell-based phenotypic screening. This is despite the
heavy bias on target-oriented screening campaigns during
this period. Phenotypic screens are promising, but present
new challenges. For example, because cellular phenotypes
serve primarily as a cumulative readout of complex genetic
and protein processes that may be coordinated by several
pathways, mechanistic follow-up is required. However,
tracking the cellular phenotypic effects to a specific target
or mechanism is a challenge, and rapidly results in a new
research bottleneck.
that
exhibit polypharmacology
[28
30]
. Interestingly, a recent analysis concluded that
drugs exhibited a greater degree of polypharmacology than
bioactive molecules, suggesting that polypharmacology is
important for therapeutic efficacy
[28]
.
Polypharmacology is now not only well established but is
considered critical to the efficacy (and MOA) of approved
drugs. Once again following molecular biology, systems-
level approaches are required to increase our understanding of
drug behavior. This approach is often referred to as network
pharmacology, and although not yet fully accepted by
the pharmaceutical industry, the need for such an approach
has been highlighted in several recent studies
[31,32]
.
Polypharmacology may be one answer to discovering new
e
Chemical Probes; Like Drugs, but Designed
to Serve as Tools to Study Gene Function
With more academic
[34]
and non-pharmaceutical groups
(e.g.,
[35]
) using small molecules to study biology,
the term 'drug' as a catch-all to describe these molecules
is inadequate, and in most cases wrong. By way of
example, nocodazole is a microtubule depolymerizing
molecule that has been instrumental in understanding
many aspects of cytoskeletal and cellular biology.
Nocodazole had its chance to become a drug, but failed
