Magazine
Lowe Agonistes (or is that Antagonistes?)
Knowing more and understanding less
By Derek B. Lowe
I have to admit it: I used to know a lot more about my work than I do now. To be sure, back when I started out, I didn’t know much at all. But I picked up the fundamentals of medicinal chemistry on the job, like most people in this area, and after a while I thought I had a reasonable grasp of what was going on.
Looking back, I think that my peak estimate of my own knowledge was probably after about eight or nine years in the labs. I suppose I should clarify that statement, especially for the sake of my current employers: no, it’s not like I’ve stopped learning my trade. In fact, I know more things about it than I ever have, which is as it should be. That peak estimate, though, was for what I knew as a fraction of what I thought there was to know. Unfortunately, although I’ve continued to expand my range, my estimates of how large a field I work in have grown even faster.
Even back then, I knew that there were several areas (big, important ones) that we didn’t have a very good grasp of. We still don’t. I’m talking about little issues like the SAR of cell penetration, or how to make a drug more orally available, or how to predict human side effects. It was already clear to me that these things were pretty much black boxes, and were mostly attacked through hard work and hope. There are still billions of dollars waiting to be scooped up by the first people who get their hands around these issues, which is the best evidence anyone needs for how much of a job that is.
But there are two other things that have made my grasp of medicinal chemistry seem a bit less impressive than it used to be. To steal a line that’s been attributed to Artemus Ward, among others, those are the things I didn’t know, and the things that I knew (but just weren’t so). A good example of the first class were the nuclear receptors and their ligands. I was almost completely ignorant of them during my first few years in the industry, and when the field finally came to my attention, I thought at first, “Oh, receptors. I know how those work . . . well, pretty much. So these must be the same kind of thing, just down at the nuclear membrane.”
Hah! I couldn’t have been more wrong if they’d been paying me to be. As I got into the field, it became clear that the biology in this area was head-shakingly complex. The more I read, the worse it got (and as time went by, more discoveries poured in that only made it even nastier). It also became clear that whoever gave these things the name “receptor” should have been hit over the head at a very early stage, ideally before anyone else heard him use the term. Using it only led naïve people (like me) to think “Oh, agonists and antagonists: got it, sure thing.” Worse yet, the literature continues to use all those terms from the cell-membrane receptor world, even though they manifestly don’t apply. So an agonist turns on signaling and an antagonist blocks it, you say? How about a compound that turns on dozens of signals, blocks dozens of others, and can totally mix (or even invert) those in different cells in the same organism, or in the same cells at different times? What do you call one of those, huh?
So, no, there were plenty of areas that I knew nothing about, and several years of work turned out to do disturbingly little to make me more confident that I understood them. That was food for thought. But even more worrisome was that second category, the things that I thought I knew. Here’s an example: I started out doing G-protein coupled receptor drug discovery (which was the source of all my mistaken confidence about nuclear receptors). Agonists, antagonists: I was a bit hazy about exactly how things worked down there at the mechanical level (as was most everyone else), but the concepts made sense.
As time went on and my familiarity with the GPCR literature increased, though, I had to grapple with allosteric modulators. Then with constituitively active receptors, and their associated inverse agonists. Then there were receptor homo- and hetero-dimers, which could respond quite differently to the same compounds. And if all that wasn’t enough, there were the alternate signaling mechanisms inside the cells themselves. Good old G-proteins weren’t the complete story, I realized, since some of these things could signal through various kinase pathways, and then there was all that craziness with beta-arrestins, and on and on. No, I realized, if I ever thought I understood GPCRs, I had been quite mistaken. The only good part, if you could call it that, was that no one else really understood them either.
So, what do I think that I understand now? The only field that I feel moderately safe in is enzyme catalysis. I think I have a reasonable grasp of how a serine protease does its thing, but to tell you the truth, I’m a bit nervous about looking closely into that. It’s probably another jigsaw puzzle that, as the pieces start to fit together, reveals a new set of instructions that requires you to take them all apart again.
That said, does all this complexity make drug discovery harder or easier? That’s another question for which my earlier answers don’t seem to be working out. I would have guessed, earlier in my career, that the more we knew about the real workings of cell biology and biochemistry, the better off we’d be. It’s still reasonable to think that, I suppose, but clearly the process is taking rather longer than I thought it would. We still seem to be in the making-things-harder stage of learning, rather than the “so-that’s-how-it-works!” stage. When we cross over is anyone’s guess, but I have to hope that we see some more obvious signs of it soon. For now, we keep getting clues about clues and complications on top of the complications. The end, or even AN end, does not present itself.
What would an end look like, though? The end of drug discovery would be, I suppose, when we have good, solid, therapies for everything, and have the tools to swiftly deal with whatever new things might come up. None of us are going to live to see that, not unless the first thing we find a cure for is death — and if we’re ever to have that, it’ll likely be only after we’ve discovered everything else first. No, we’re all going to have plenty to do for the rest of our careers. The big question is whether we’ll discover enough to ensure that we’ll all have careers. To that end, speed the day when we start to get a handle on what we do, and when medicinal chemists can feel overconfident for better reasons.
Derek B. Lowe has been employed since 1989 in pharmaceutical drug discovery in several therapeutic areas. His blog, In the Pipeline, is an awfully good read.
