“I wrote the book because…all drug discovery scientists have probably had a similar experience…people turn to you and they say “What do you do?”. And I say “Oh, well I’m in drug discovery”. And then people start asking you all these questions that you really can’t answer in an elevator conversation. So, that was really why I wrote the book, to say the things that I wanted to say at a cocktail party,”
Donald R. Kirsch
Author, The Drug Hunters
Donald R. Kirsch has been a drug hunter for thirty five years, holds twenty-four drug-related patents, has written more than fifty papers, has been a reviewer for prestigious journals, a director, research group leader, and chief science officer at Wyeth, Cyanamid, Squibb, and Cambria Pharmaceuticals, currently teaches drug discovery at Harvard Extension School and recently wrote a book explaining what we do titled “The Drug Hunters: The Improbable Quest to Discover New Medicines”.
Interviewed by Barry Bunin, Collaborative Drug Discovery, Inc., February 15, 2017.
Barry Bunin: I’m joined by Donald Kirsch, who is the author of the book The Drug Hunters, and has a long career working as a biologist in the drug discovery field. Maybe just starting at the beginning: Donald, if could you give us a little bit about your background in biology and how that played out across a couple different big pharma companies as well as entrepreneurial companies.
Donald Kirsch: Sure. As a kid, I always liked science; I always wanted to be a scientist. I went to college and I majored in biochemistry, and I still had the yearning to learn more science, so I did a PhD at Princeton. I just followed my nose, I actually was a Drosophila geneticist of all things. But by the time I was finishing my PhD I realized I had more interest in applied science than in basic science, so I did a postdoctoral fellowship at the Robert Wood Johnson Medical School, which then turned into a junior faculty position at the Department of Pharmacology. My objective was to learn pharmacology, and reinvent myself as someone who could work in industry. And I guess it worked, because after I finished that stint, I got a job at Squibb, which is one of the parent companies of Bristol-Myers Squibb. I worked there for 8 or 9 years, in anti-infectives mainly, but did some other things as well, such as receptor-ligand projects.
Anyone who’s been in this business for more than 6 months knows that mergers and acquisitions are common, so facing the Bristol merger I left, and I went to Cyanamid – their pharmaceutical division was called Lederle. And I was there for, goodness, through several mergers and acquisitions for almost 20 years. I did a lot of different things there: antifungals, I ran a functional genomics group where we focused on GPCRs and ion channels, did some anticancer drug work. And then, in sort of the classic big pharma move, I was laid off when I was 54, just a year prior to fully vesting in the pension plan.
In any case, I took it as an opportunity. I had never done a start-up, so I said “Okay, this is my chance to do a start-up”. So, I went to work for Cambria Pharmaceuticals in Kendall Square in Massachusetts. Pretty raw start-up. They were a contract research company and hired me to be the CSO, and bring them into neurological drug discovery. When I got there, we settled on working on ALS, amyotrophic lateral sclerosis, Lou Gehrig’s disease. And that’s what I did there for about the better part of a decade. We came up with some interesting compounds which are proteasome activators, the idea being that they would clear out toxic protein aggregates from motor neurons and by so doing rescue them from toxicity. And that project is still going on at Northwestern University.
As you said I just had a book published, Drug Hunters, which is really a book that I wrote to explain to the laity to answer three questions: why are drugs so expensive, why do drugs have side effects, and why isn’t there a drug for a disease that you may be very, very concerned about?
BB: Alright. Well, why don’t we start with those three questions. Why do you think these things are true?
DK: Drugs are discovered products. They are not engineered products. Paul Ehrlich was a 1908 Nobel Laureate and the discoverer of Salvarsan, the first drug that was discovered from scratch. Ehrlich was German, and he was fond of saying that in order to be successful in drug discovery you need four G’s: Geschick which means smarts, intelligence, know how, Geduld which means patience, Geld, money, and Glück, luck. And he said that 110 years ago, and it’s still true today. You need to be smart, but just being smart’s not enough. These things take a long, long time. They’re very, very expensive and most projects fail. And you need luck. Even the smartest idea and the best laid plans more commonly than not come to naught. And that’s a recipe for drugs being both expensive and having side effects, because drugs are not a perfected thing, they’re a discovered thing. You don’t plan to make some precise thing. You’re happy if what comes out of it is useful.
BB: You mentioned earlier working with Northwestern. And I know Professor Richard Silverman has a hall named after him and a neuropthic pain drug. How did you come across Professor Silverman, and then a related question, maybe just talk about an anecdote another brilliant scientist that you’ve interacted with in either industry or academia?
DK: I have been collaborating with a Northwestern professor named Rick Morimoto, an expert on protein aggregation and the systems for disaggregating and getting rid of aggregated proteins, and while we were working together it got to the point where we really needed a medicinal chemist to work with us. And he said “Oh, I’ve got this colleague, Rick Silverman. The guy is absolutely, super brilliant and he’s easy to work with,” and every good thing he’s said about Silverman is true. One of the best collaborations I’ve ever had. We got him involved and that led us to these molecules which are proteasome activators. I did the biology, and Silverman did the chemistry. Other people in my career? Richard Sykes was the CEO of GlaxoSmithKline, then later he was the Rector at Imperial College. He actually was my first boss. When I went to work for Squibb, he was no CEO back then, he was the Associate Department Director. He hired this young punk—me—as a senior scientist. And he was really a very inspiring leader, a very, very strong leader. I really liked him and learned a lot from him about how to do drug discovery. He was, by the way, in addition to being the CEO of Glaxo, when I was working for him, he had just discovered aztreonam , the first monocyclic beta-lactam. So, really a special person.
BB: That’s really a good segue to talking a little bit about what you’ve found to be the most interesting discoveries, targets, techniques that you’ve seen over the years that have impacted or changed the way you think about drug discovery and the way we focus on this hard challenge.
DK: Yeah so in my lifetime, and I’ll reveal that I’m 66 almost 67 years old, I started learning pharmacology in 1978. It was interesting for me as a molecular biologist to start learning pharmacology. I was astonished that although the receptor theory of pharmacology was well established, for most of the drugs out there nobody knew the molecular definition of what the receptor for the drug was. So it worked on something but in most cases, say it was an adrenergic receptor, maybe they knew it was a GPCR, but not much more than that. So then going from there to now, where every receptor that one could possibly think of has been cloned and there’s probably an assay for it. Today, you can profile drugs. Prior to that, drugs really in general weren’t discovered to act on a specific receptor. Take my earlier example – Richard Sykes discovered Aztreonam which is a PBP inhibitor of bacteria, but PBPs are called PBPs because they’re penicillin binding proteins. Which is to say that people didn’t know what the heck they did initially, other than the fact they bound to penicillin—since they bound to penicillin they must be involved in the action of penicillin, and it was only years later that people realized they were enzymes that catalyzed the synthesis of the bacterial cell wall.
So, going from this pretty astounding lack of knowledge, it’s almost amazing to me that drugs could be found at all in that previous state, to coming to the point where the whole human genome has been sequenced. We know pretty much what most every receptor could be, and it’s just an astounding journey in my lifetime.
BB: Let’s talk a little bit more about anti-infectives and why they are challenging for drug discovery and how those challenges are maybe greater or less today than what they used to be for anti-infectives.
DK: So, I think these days the main challenge is an economic one. I’m just about one of the youngest people ever to have participated in a full blown antibiotic discovery program. I’ll tell you a factoid that I find astonishing: last year more people died in the United States from bacterial infections that had previously been easily treated by antibiotics than died from AIDS. So, we’re really running out of antibiotics, and it’s not because people are too stupid to discover them, it’s because people stopped doing it at the same level of effort, and turned instead to more lucrative therapeutic areas. If you and I were to get together and discover a new class of antibiotics that were not cross resistant with current clinical antibiotics, what’s going to happen is it’s going to go to the hospitals and it’s going to be put on reserve formulary because everyone’s going to worry about resistance developing to it and that means that sales are going to be paltry. It’s ironic that the more useful your new antibiotic is, the wiser it is not to use it widely and hold it in reserve. And that’s compounding the problem that antibiotics themselves, even in the golden age, were too good. You know, it’s not like anti-hypertensive it’s not like an anti-psychotic, it’s not like an anti-depressant where it’s a chronic drug. You’re on a statin for high cholesterol, you’re going to take that drug every single day for the rest of your life. Antibiotic, you take a course of 5, 6, 7 days and most often you’re cured. No more sales. So I think that what has to happen is somebody powerful has to wake up to the fact that these previously easily treatable infections are becoming more and more difficult to treat, and someone’s going to have to cut the Gordian economic knot, and put more effort into it. When I first started at Squibb in 1981, pretty much every single major pharma had an anti-infective group. I’m going to guess off the top of my head some 80-90% of the companies have now exited that area of research. Hopefully we’ll get back into it.
BB: I hear that and I agree that there are economic challenges, but I also want to hear your thoughts just on the scientific side. To change gears, maybe you could compare going through the dual cell membrane to get past the periplasm into the cytoplasm for gram-negative bacteria and the resistance that emerges to anti-bacterials with the different challenges, for example, in the CNS area, working with getting through the blood-brain barrier and the sparsity of predictive animal models for a lot of diseases involving the brain. And just on a scientific side to compare and contrast those two areas you worked in.
DK: That sounds like a question that would be asked by a medicinal chemist, so no surprise here. Yeah, I remember in the early days when people started doing receptor-driven pharmacology, everyone would drive for potency and efficacy. The most potent molecule, the most efficacious molecule—that was what you wanted. But then after you had it, you wake up to the fact that the ADME properties of the molecule were absolutely no good, it wasn’t soluble in water etc., and then you have to go back and reengineer it to have a decent half-life, a decent tissue penetration, and in the case of a CNS drug, decent brain penetration, and in the case of an antibacterial to be able to get through the outer membrane of gram-negative bacteria. It is still a big challenge. And in my mind this is something that we still do stochastically, unfortunately. The molecule that gets through the blood-brain barrier effectively is the molecule that gets through the blood-brain barrier effectively. But you know philosophically and strategically, I think it’s kind of the same issue and we should over time get better at it. You get a molecule that hits your receptor of interest and while you’re elaborating that and refining it, you work simultaneously to make sure the molecule gets through the blood-brain barrier, if that’s what your challenge is, or the outer membrane of the gram-negative bacteria.
BB: So maybe talking both at the micro-level and macro-level, how do the different neuro transmitters work? We have GABA, serotonin, dopamine, histamine, and others and how can we understand what’s similar vs what’s different from a therapeutic perspective about these signaling compounds when thinking about neuro-diseases?
DK: Actually you might find this entertaining if I answer this question, knowing that you work with many medicinal chemists. In the 1940s chemists started making compounds, analogues, based on an ethylamine core structure. This is a structure that they derived because all of the biogenic amine neurotransmitters could be expressed as a simple Markush structure around this ethylamine structure. And they started elaborating it, and again no one knew what and where the receptors were, they just knew there were a fistful of biogenic amines. Diphenhydramine was the first drug that came from that exploration. And it really was the first drug of it’s kind because it was the first molecule that had enough selectivity for the H1 histamine receptor so that it was tolerable in a clinical setting. So, from there people saw diphenhydramine was a success, it was a huge success, a dramatic success of its time, and started making further derivatives of it from a chemist-centric point of view. Which is to say: “I’m going to make some cool drug-like compounds, let’s see what they’re good for”. And it turned out these compounds were fabulous. This line of research both established the neuroleptic anti-psychotics like thorazine which all came from a modification of the basic ethylamine core. Actually, thorazine is a great story – it was a molecule looking for a disease. It was tried for the silliest of reasons in schizophrenic patients and it worked, and of course then it took decades to figure out why it worked, because it’s a very dirty drug. Now we know it worked on dopaminergic receptors, but with also some additional serotonergic action was necessary to prevent severe tardive dyskinesia. Ethylamine compound based discoveries also led to the tricyclic anti-depressants. Imipramine, the first tricyclic, was actually a very, very subtle derivative of thorazine, and it failed as an anti-psychotic drug in the clinic, and, for reasons which I don’t think anybody completely understands, the clinician said “Ok, this thing isn’t working for the psychotic patients, I’ll give it to some of my depressed patients and see if it will work for them.” And then it was further worked out that that this was a drug that worked on biogenic amine reuptake pumps. Very, very small chemical change, big change in selectivity, so kind of fascinating.
BB: How do you think about different CNS compounds like benzodiazepines hitting GABA receptors vs serotonin uptake inhibitors at a target level, at a tissue level? You were looking at three different diseases, I believe, at Cambria Biosciences. Maybe talk a little bit about your approaches and how you think about target vs phenotypic assays in the CNS arena?
DK: Let’s focus on the benzodiazepines. To me the real challenge there is that GABAA receptors are made up of multiple subunits in an almost combinatorial fashion. There’s probably a hundred different GABA receptors. We know that benzodiazepines work on one subset of them. The non-benzo GABA inhibitors like zaleplon and zolpidem work on their own subset. The barbiturates work on a different subset. So, that’s the idea – to get the right molecule for the right subset. You got the barbiturates, you got the benzodiazepines, you got the non-benzodiazepine-types and I think the Holy Grail here is to find the perfect molecule that works on the perfect subset, so that you have a pure sedative, a pure anxiolytic, a pure antiepileptic, easier said than done. Of course the addictive properties these drugs is a whole other issue.
BB: Maybe just talk about one of the “Aha” moments you had in science, either basic science or applied science? It could be inside your research or outside your research.
DK: I guess consistent of what I was saying earlier, I was trained on the cusp between people testing things as a primary screen in animal models, and people screening things against receptors. So, Paul Janssen, who founded Janssen Pharmaceutica which later merged with J & J, totally believed in animal model research. Every product he found was based upon direct screening in small animals. And he very much disliked the kind of research where you look for a specifc receptor agonist or antagonist. No, no, no – you put the drugs in some disease model in the animals, and then the animals are going to tell you what the best compound is. He was extraordinary successful. If you look at Paul Janssen’s bio and all the important drugs he discovered, you can’t say he was wrong. So, I guess maybe that’s a contrarian “Aha Moment”.
For me, it was always more appealing to do more receptor work. I enjoy the highly biochemically selective studies to try to understand what is happening at a molecular level. Although, maybe I’m a little bit of a hypocrite here because I am a big fan of Paul Janssen, so I see both sides. The last project I worked on, which was the ALS project, was based on a phenotypic screen, where we found our lead molecules that we’re still pursuing to this day. Basically, we set up a cell-based assay in which aggregated SOD1, which is one of the insults in ALS, killed cells, and then we looked for compounds that were protective. And in the course of elaborating and optimizing those molecules, we discovered that they were stimulators of the proteosome. We pretty much know which regulatory subunit of proteosome they activated on. I guess that was a very Paul Janssen like project. Not in animals but in cell culture, but the same kind of idea.
BB: Great yeah. I’ve worked on proteasome inhibitor projects as well. So, it’s an interesting area. I’m curious about the reception to your Drug Hunters book. I got introduced to you by my colleague Dr. Janice Kranz who mentioned the book and then I read this funny interview you did for the mainstream media and figured we need to get the more scientific take. I’m interested both from the lay public as well as from scientists for feedback.
DK: Yeah, this is all real, real, new. The book just came out about four weeks ago, so I’m still trying to figure out how it’s doing. I wrote the book because—you’ve probably had a similar experience and as a matter of fact all drug discovery scientists have probably had a similar experience—you go out Saturday night with your significant other to a cocktail party, and the lawyer says she’s a lawyer, and the doctor says he’s a doctor, and the accountant says she’s an accountant, and people turn to you and they say “What do you do?”. And I say “Oh, well I’m in drug discovery”. And then people start asking you all these questions that you really can’t answer in an elevator conversation. So, that was really why I wrote the book, to say the things that I wanted to say at a cocktail party, but after about 30 seconds you just scratch the surface of the real story before folks are on to their next drink.
BB: And how do you feel with the result, I know you coauthored the book, and the experience for you. How satisfying is it to have finished the book, and how is your thinking either solidified or change in the process?
DK: So, I’m really happy to have had the chance to say what I wanted to say. That’s the happiest and most satisfying part of it—that I did get a chance to say what I wanted to say, even though people didn’t have time to listen at the cocktail party. But I also learned that writing is really, really hard. When I was writing the book, Philip Roth, who is one of my favorite authors, retired. The New York Times interviewed him and they said “Oh my goodness, winner of the Pulitzer Prize, your name is always brought up as a possible contender for the Nobel Prize in literature, you’re doing really well, people really like your books: why did you retire?” And Roth didn’t bat an eye, he said “The work is just too damn hard and too damn frustrating. I’ve had it”. And, you know, don’t think it’s so glamorous to write a book. It’s really hard, hard work—much harder, and much more frustrating than I ever expected. You have to really respect authors who can do this their whole life.
BB: And doubly I want to congratulate you for getting out the book Drug Hunters, and it’s available online, I appreciate you taking the time to talk.
DK: Take care. Thanks.
This blog is authored by members of the CDD Vault community. CDD Vault is a hosted drug discovery informatics platform that securely manages both private and external biological and chemical data. It provides core functionality including chemical registration, structure activity relationship, chemical inventory, and electronic lab notebook capabilities.
CDD Vault: Drug Discovery Informatics your whole project team will embrace
Translated with Google Translate