Field of Science

A government center for drug discovery?

The New York Times has an article about a new center for drug discovery that the government is going to set up in the face of the declining pace of new drug discovery in the pharmaceutical industry. We all know what's wrong with the industry, with the numero uno factor being the pursuit of short-term profit goals at the expense of basic science and long-term benefit. The question is, can the government make up for this shortfall?

Perhaps, if it's done right. But the vision for the new center does not inspire me with much confidence. The center seems to mainly be a result of NIH director Francis Collins's conviction that gene-based drug discovery is the wave of the future. Collins is disappointed with the fact that Big Pharma has been unsuccessful with "translational genomics" in spite of spending millions of man hours and dollars. He thinks that if done right, this kind of translational approach will result in new drugs. As he makes it clear in his book "The Language of Life", Collins is a longstanding proponent of genomics-based medicine.

But isn't this what everyone has been saying ever since genome sequencing became possible? We all remember the hype about genomics enabling a new generation of 'rational' drugs based on intimate knowledge of genes and their protein products. The fact that this vision has not panned out could partly be ascribed to the lackluster efforts and the constant waves of lay-offs in industry, but maybe there's also a deeper reason why the optimism has hit a wall. Maybe, and experts have been saying this
for a while now, it's simply because taking a genomics-based approach to drug discovery ignores all the other complexities of biological systems like signal-transduction and epigenetics. Discovering the gene and protein is one thing, understanding the intricate interactions of the protein as part of a multi-layered cellular communication network is quite another. We are still struggling to understand the very basics of how proteins and genes link up in cells to enable complex physiological and behavioral responses, let alone rationally design drugs to block specific parts of those responses. In the absence of such understanding, simply pinning your hopes on the promise of 'translational genomics' seems to me to be another big sink for money and personnel.

So what else can a government center for drug discovery do which could be more concrete and fruitful? Ironically, the article contains part of the answer when it highlights the complexity of biological systems stated above.

Consider this remarkable fact; in the last century, only two breakthrough treatments for mental illness have been developed, lithium for bipolar illness and chlorpromazine (Thorazine) for schizophrenia. Many other successful antipsychotic drugs were spinoffs of thorazine. Also consider that even today we have little clue about how these drugs work, let alone how to rationally design them. Thorazine likely affects multiple neurotransmitter pathways involving serotonin, norepinephrine, dopamine etc. While we have made great advances in the last fifty years, brain chemistry remains as complex as ever. At a molecular level, the main problem is in understanding the basic mechanisms and specificity through which a molecule as simple as dopamine binds to only certain subtypes of a neurotransmitter receptor, stimulates multiple second-messenger pathways to different extents and elicits a complex behavioral response. In this case we know most of the genes and we know most of the protein products, yet we are light years away from understanding how Thorazine works. Lithium is an even more mysterious substance whose workings are almost akin to black magic. Instead of chasing the genes, thoroughly understand the action of a few of these drugs on a biochemical level and we can make significant inroads into understanding brain chemistry.

So if there's one thing a new government center for drug discovery can do, it should be to focus on these specific problems in the most general way rather than pool together resources for "translational genomics". Doing translational genomics will simply mean advancing work which industry is already involved in; it will largely be more of the same and there's good reasons why it may not work.

Here's what I think instead. If the center truly wants to do something productive that pharmaceutical companies cannot, let it put together a mini Manhattan Project type team focused on understanding a few specific problems, like how lithium works in the brain. The problems should be picked based both on their medical importance and their potential impact in enabling general understanding of the field. Just like the Manhattan Project did, get together the best people in the country from several disciplines who are experts at multidisciplinary thinking and problem solving. If you want to attack the lithium problem for instance, get together chemists, biologists, neuroscientists, pharmacologists, doctors and perhaps even a few physicists, mathematicians and computer scientists. Put them in a couple of large rooms (and maybe even seclude them in the majestic mountains of New Mexico) and give them enough funds. And then most importantly, give them almost complete freedom to brainstorm about specific problems. Let them consider every possible approach, from running basic biophysical experiments to the most advanced neural imaging techniques. Don't limit yourself to any one philosophy like translational genomics or any other currently fashionable mantra. Genomics can certainly be part of the mix but it should not be put on a pedestal. Combine the oldest tools of classical pharmacology with the newest tools of molecular biology. If the industry has been missing one thing, it's been the presence of bright young people who are given complete freedom to focus on their diverse ideas without strings attached and constant fear of unemployment. The government can give these men and women what industry has taken away from them in recent years.

The government has always been good at this kind of free-for-all interaction among talented scientists who are unencumbered by research funds and job insecurity. A new government center for drug discovery could be a great idea, but only if it provides the kind of freedom to operate that brings out the best in creative minds. Focusing on translational genomics to me seems to be another way to repeat what has been done and to waste more funds, time and personnel. Instead, do what the government does best; let them think, and let their minds soar.


  1. I would say that the center should focus on developing new antibiotics and vaccines. They are the most effective drugs and that's is precisely the reason the pharmaceutical industry is not that interested in them: if all it takes to cure someone is a two weeks treatment with antibiotics, there's not a lot of profit to be made.

  2. I'm far from sure that we know just what our drugs are doing, even when we think we do. For a totally unexpected mechanism of action of one of the tricyclic antidepressants - see



Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS