My laboratory applies an array of in silico methods such as
virtual screening and
to decipher how life works at atomic level and, in particular, to spur drug discovery for the treatment of human disease.
In partricular, we employ structure-based rational approach to screen novel drug candidates from a virtual library comprised of close
to one million natural products, nearly half of which are derived from traditional Chinese and Indian medicines. Additionally,
our goal is to identify novel allosteric modulators of proteins, which bear the potential to not only lead to the
development of less toxic drugs but could also be exploited to prevent rather than cure disease. For example, novel drugs that could prevent
amyloid formation would serve as invaluable therapeutic tools in the treatment of neurodegenerative disorders such as Alzheimer's disease.
Natural products are small organic compounds derived from plants, microbes and marine organisms. Notably, natural products
must interact with at least two proteins: one that is involved in their biosynthesis and the other being the target which they must
recognize to elicit their chemical effects. Because natural products have undergone an evolutionary design and optimization over
millions of years to a pristine level of molecular specificity and recognition, their superior medicinal power compared to their
synthetic counterparts can be easily appreciated. For the same reason, drugs derived from natural
products tend to be far less toxic with minimal side effects compared to synthetic compounds. Indeed, natural products have been used as a
medicine for millenia and, even today, they constitute close to 50% of therapeutic drugs available on the market.
Importantly, the probability of finding a potential drug candidate against a target protein from natural products (~10-3)
is nearly three orders of magnitude greater than the synthetic compound library (~10-6). It is noteworthy that the
need for synthetic drugs only arises due to the scarcity of natural products. The chemical complexity of natural products only
adds to our woes as it presents a nightmare for the medicinal chemist to reconstruct them in vitro.
Nonetheless, natural products represent a glaring opportunity to discover new
drugs that are more compatible with human physiology. To say that natural-product-based medicine is not only head and shoulders but also
chest above synthetic drugs would be something of an understatement. What is so remarkable about natural products is that less than 1% have been
hitherto tested for their biological activity. Accordingly, the vast wealth of therapeutic potential hidden in our environment
must be unearthed to improve the quality of human health. Given the unprecedented availability of high-throughput technology to
probe life at atomic level, it is time to bridge not mind the the gap between traditional and contemporary medicine.
In our laboratory, we have also recently developed a novel technique to determine the specificity and selectivity of putative or therapeutic
drugs against their target proteins and/or macromolecules. In this procedure, dubbed retro (or reverse) screening, one proceeds in the opposite
direction to the so-called virtual screening discussed above. In virtual screening, the goal is to use a protein target as a bait to identify a
high-affinity ligand from a search library typically comprised of hundreds of thousands of small molecules. In contrast, retro screening employs
a known drug molecule (clinically-proven, pre-clinical or a lead compound) to screen a protein library comprised of hundreds of thousands
of individual structures (obtained from both experimental and modeling techniques). In other words, a known or potential drug is used as a bait
to fish out proteins as potential binding partners. Accordingly, the extent to which this drug molecule
cross-reacts with the human proteome provides a measure of its efficacy and the potential long-term side-effects. Of particular note is
the realization that retro screening should also hold a great deal of promise to test the protein landscape of thousands of environmental
toxins, cosmetic agents and food additives whose growing influence is expected to cause havoc to human health and well-being in the
twenty-first century. Given that our knowledge largely remains elusive vis-a-vis the ability of these ubiquitous chemicals to target the
human proteome, retro screening should shed new light on their mechanisms of toxicity and thereby offer new insights to combat their
Simply put, retro screening bears the potential to deliver medicine along three major fronts in a timely and cost-effective manner:
To assess the potential side effects of pre-clinical drugs, thereby not only eliminating poor candidates for clinical
trials but also saving time and money for delivering new drugs from bench to bedside.
To lend structural insights into the side effects of clinical drugs, thereby empowering clinicians and patients-alike
to more effectively weigh the medicinal benefits of a clinical drug.
To discover novel protein targets of clinically-poven drugs, thereby allowing the use of a clinical drug for the treatment of a new
disease for which there are limited alternative therapies.
To shed light on the mechanisms of toxicity of ubiquitous environmental toxins, cosmetic agents and food additives,
thereby allowing clinicians to not only better understand their impact on human health but also foster the development of
new strategies and social trends to combat their undesirable effects.
In sum, we remain deeply devoted to the advancement of biomedical knowledge as evidenced by close to 100 publications of wide international acclaim
over the past two decades or so. Through our scientific efforts, we are determined to accelerate rational development of natural-product-based
therapies harboring greater efficacy coupled with low toxicity to alleviate human suffering due to an ever increasing myriad of pathological