a16z Podcast - Psychedelics: Striking a Balance Between Benefits and Side Effects

Episode Date: January 23, 2021

In recent years, there’s been a shift in how we think about psychedelics – from drugs of abuse and recreation, to powerful drugs for treating neuropsychiatric conditions such as depression, addiction, and PTSD.  But there’s still a lot we don't know about how they work, and how we can maximize their therapeutic benefits while minimizing their adverse side effects. So this episode of Journal Club discusses a method for striking that balance, from a paper published in Nature last month, “A non-hallucinogenic psychedelic analogue with therapeutic potential“... which could represent a major step forward in psychedelic medicine. This episode first appeared on Bio Eats World:https://a16z.com/2021/01/21/journal-club-safer-psychedelic/

Hi everyone, welcome to the A6 and Z podcast. I'm Sonal, and today I'm sharing our most recent Journal Club here. This particular episode is all about a safer psychedelic. Yes, you heard that right. Before I tell you more about the topic, as many of you may recall, I announced our new show Journal Club in this feed last year because we're all about bringing you new ideas and how to translate them from theory to practice. And it's now available as part of our show BioEats World. So be sure to subscribe to that wherever you get your podcast if you haven't already. So now let me quickly share more about this episode. In recent years, there's been a shift in how we think about psychedelics, from drugs of abuse and recreation to powerful drugs for treating neuropsychiatric conditions such as depression, addiction, and PTSD. But there's still a lot we don't know about how they work and how we can maximize their

therapeutic benefits while minimizing their adverse side effects. So this episode of Journal Club discusses a paper published in nature last month on a method for striking the balance between these benefits and side effects, which could represent a major step forward in psychedelic medicine. Hi, I'm Hannah. And I'm Lauren, and this is the BioEats World Journal Club, where every Thursday we discuss breakthrough scientific research, the new opportunities it presents, and how to take it from paper to practice. And today, we're talking about one possible path to take psychedelic drugs from the lab and into the clinic. So, Lauren, are we talking about using psychedelics like mushrooms, LSD, ayahuasca, as

treatments for some mental health conditions? Sort of, but with two key differences. First, we're talking about ibogaine, which is another hallucinogenic plant that has anecdotal evidence of treating depression and curbing drug-seeking behaviors. And second, we're not talking about using ibogaine itself. It's a dangerous drug. It induces strong, long-lasting hallucinations, and it can be a very drug. cause heart attacks and arrhythmias. So could we somehow maximize the potential therapeutic

benefits of Ibegain without these major downsides? Possibly, and that's where the work of my guest, David Olson of UC Davis, comes in. His lab's approach is to try and develop new drugs based off the structures of psychedelics that retain their therapeutic properties, but that have better safety profiles and that, importantly, are non-halucinogenic. In our conversation today, we cover his team's recent nature paper creating a non-hollucinogenic derivative of ibegain, the animal model evidence of its ability to treat depression and alcohol and heroin-seeking behaviors and the unexpected challenges facing the psychedelic medicine field. So what does it mean for a drug to be psychedelic? What are the key features of this class of drugs?

The word psychedelic means mind manifesting, and so this has different meanings to different people. But the one class that everyone agrees on are what we would call the classic hallucinogens or the classic serotonergic secondelics. And the defining feature of this class is that they bind to and turn on a receptor known as the serotonin 2A receptor. Serotonin is a super interesting and important neuromodulator. There are actually multiple serotonin receptors in the brain that are found in different neural circuits. and they all play different roles. And so serotonin itself has often been said to be important for everything but responsible for nothing because it seems to modulate everything from, you know, sleep wake cycles, mood,

memory, sexual behavior, reward. It's really critical for a whole bunch of things. But the psychedelics are very specific in that they activate this one key serotonin receptor called the 5HT2A result. So serotonin does all these different things in your brain. What mediates the effect of serotonin is the different receptors that it binds to. And psychedelics bind to this one specific receptor and turning on their subsequent pathways. What happens downstream of that binding?

What's the mechanism of action for psychedelics? Yeah, that is a debate among neuropharmacologists. So we had hypothesized that maybe psychedelics were doing. something similar to what ketamine is known to do and to regrow these neurons in the prefrontal cortex. It seems that activation of the 5HT2A receptor turns on a growth pathway critical for this change in neuronal structure. The prefrontal cortex is super important because it projects the other brain regions that control fear and motivation and reward. If you look at stress-related neuropsychiatric diseases like depression, post-traumatic stress disorder, and substance use

disorder, all of these diseases are characterized by a loss of synapses and atrophy of neurons in a prefrontal cortex. So we had hypothesized that maybe drugs that could regrow these critical neurons in the prefrontal cortex could have broad-ranging therapeutic effects. In 2018, my lab showed that this is, in fact, the case that psychedelics can be. regrow these neurons, suggesting that maybe their long-lasting effects stem from their ability to kind of rewire pathological neurosurcuitary. So the activation of the serotonin receptor leads to new neuronal growth, and it's through new neuronal growth and the strengthening of the synapses in this brain region. That's how psychedelics induce hallucinations and have these other potentially

therapeutic effects. Not quite. The activation of the 5HT2A receptor, which leads to neuronal growth, is what we're hypothesizing is what results in their long-lasting therapeutic properties. But it is not directly related to their hallucinogenic effects. The hallucinogenic effects are still not entirely clear what's going on, but it seems that they cause very short-term massive excitation of cortex. This makes it difficult for the brain to filter out what is noise and what is real signal. And that's why you kind of have all these perceptual. disturbances. Some of my colleagues at places like Johns Hopkins believe that the hallucinogenic effects are really, really critical to the therapeutic effects of the drugs, whereas I'm not entirely

convinced that they are necessary for some of the therapeutic effects. Okay. That's really interesting. So there's a debate in the field whether the hallucinations are necessary for the therapeutic benefits. And you're hypothesizing that they're not, that the hallucinations are distinct and separable from the therapeutic benefits, which are mediated through this new neuronal growth. So what are the key downsides to using psychedelics that induce hallucinations? The issue with using psychedelics for treating neuropsychiatric disorders is, well, there's a couple. One of the most important ones is the amount of time and the cost associated with it. So before somebody goes in for a silozybin session, they have to be prepared for that session.

they then have to be under the supervision of a health care worker for up to eight hours, and then they have to have some sessions after the fact to integrate their experience and to make sure that everything is okay. And that, you know, from a throughput perspective, it's going to be hard on our health care system to treat a large number of patients like that. And that's critical because about one in five people will suffer from a neuropsychiatric disease at some point in their lifetime. Another potential issue with using psychedelics is that they're contraindicated for people that have

things like psychotic disorders, things like schizophrenia, and a lot of neuropsychiatric diseases are comorbid with other disorders. And so being able to separate the hallucinogenic effects from the therapeutic effects might extend the use of these drugs for other patient populations. So let's talk about the psychedelic that's the focus of this article, which is called Ibogaine. What's its history and what do we know about its therapeutic potential versus its adverse effects? Ibegene is a natural product that was primarily isolated from a plant in West Africa, but it's actually being found in lots of plants all over the world. It was used many, many years ago in France as an antidepressant,

but it was pulled from the market because of safety issues. In more recent years, it was anecdotally shown that Ibegain might have, some anti-addictive properties that seem to be pretty profound. There are no double-blind placebo-control clinical trials with Ivagane. So all of this information has really been through anecdotal reports and open label trials. And what people have seen is that a single administration of Ibegain has the potential to keep heroin addicts drug-free for up to six months. And then with a second administration, they can be drug-free for up to three years. But the huge issue, with ibegaine is its safety profile and not only does it cause hallucinations but the more

worrisome problem is the fact that it binds to an ion channel on the heart that causes cardiac arrhythmias several people have died during ibegain treatments of heart attacks and so i think that is the thing that has really limited its clinical potential the other thing that's really interesting about ibogaine is that it seems to work across addictive disorders. So alcohol use disorder, opioid use disorder, addiction to psychostimulants and nicotine. And that is something that is very uncommon for an anti-addictive treatment. The things that are shared among all of these disorders is kind of an atrophy of those neurons in the prefrontal cortex, which controls drug-seeking behavior. Again, we thought that if we could actually rectify the circuits, we might be able to have

beneficial effects across a variety of addictive disorders, and also comorbid neuropsychiatric diseases like depression. Addiction is correlated with the atrophy, so the kind of shriveling up of the neurons in the prefrontal cortex, and psychedelics encourage neurons to grow in this very same region. So that makes sense that this could possibly overcome these deformations almost in the brain and therefore block this drug-seeking behavior. Now that we have this background on what psychedelic are, and specifically what ibogane is, it's therapeutic potential, but it's very serious adverse side effects. Let's get into the meat of your study. You wanted to create a derivative of ibogaine that retained this ability to induce neuronal growth, and particularly you wanted to see if you

could separate the hallucinations from the therapeutic effect. So what was your strategy for doing this? how did you set up your study? Our strategy for this paper is a technique that is known as function-oriented synthesis. And so simply what we did is we started lopping off pieces of ibigene. We deleted certain bonds and deleted certain chemical motifs. And then we tested all of those compounds in our cellular neuroplasticity assays to identify the substructures of ibigaine that were most effective. We found that one substructure in particular was very interesting, and then we made one other small tweak.

The reason we thought that that structural change might be effective is from some really nice work by Richard Glennon done in the 1980s, suggesting that a similar structure was non-halucinogenic. Okay, let me break this down. You know the chemical structure of Ibogaine, and so you systematically removed different parts of the structure, creating a suite of derivatives and then you tested those derivatives in an in vitro assay to see do they induce neuroplasticity, which means like did they make the neurons that are in the dish grow? And that's how you determined like first off, do they have the baseline function that we want? Correct. So you found a simplified version of Ibogaine that did this and then you modified

it one step further by moving one functional group to a different part of the structure. So we should say that this compound that you've created, you call TBG. TBG, it stands for tavernanthalog because it's an analog of a natural product called tabernanthene. So you next needed to test if it was hallucinogenic. How do you do that in an animal model? How can you tell if a rodent is hallucinating or not? So there's a couple of ways to do this, but our favorite is a test called the

a mouse head twitch response assay. And it's pretty simple. If you administer a serotonergic hallucinogen to a mouse, they will increase the number of times they twist their head during a 20-minute period. Some really beautiful work by Adam Halberstadt that came out just this year correlated the head twitch response potency with human hallucinogenic potency across 30 or more structurally distinct psychedelic compounds and he sees a really, really strong correlation. So you found that TBG was able in this in vitro assay to inspire neuroplasticity and it doesn't inspire this head twitch response. It doesn't appear that it's having a hallucinogenic effect on the mice. In the paper, you then compared the safety profile of TBG to that of ibogaine.

So the first thing we did for safety was we looked at inhibition of herd channels. Those are the ion channels in the heart. And then we found that TBG was significantly less potent than Ibegain, suggesting that it could be safer. The next experiment that we did was in zebrafish and did some toxicity assays. And we found that TBG was much less toxic than Ibegain. And then eventually we went into our in vivo studies to confirm that we saw the same effects in vivo that we saw with the neurons in the dish. Now that you have TBG, you know that it's safer and less toxic than Ibogaine, that it does induce neuroplasticity in the brain. You wanted to look at its impact on behavior. So the first behavioral test is called a forced swim test.

What does this tell you about mouse behavior? So the forced swim test is a very, very good predictor of antidepressant efficacy in humans. On the first swim test, when you place the mouse into the water, the mouse will either swim around or will simply float. If you increase the amount of time the animal is swimming, that is typically what an antidepressant compound does. We found that TBG had very similar effects, at least after 24 hours, as something like ketamine. And ketamine is kind of the state-of-the-art, fast-acting antidepressant.

So you were able to modify the structure of iBagane to create. this derivative that is, you know, safer and that has this strong, fast-acting antidepressive effect. Other than ketamine, most of the antidepressants on the market, like SSRIs, act really slowly. That's correct. I think that's a really exciting result, but maybe even more exciting is the next set of results, which demonstrate a therapeutic potential that's distinct from anything that's on the market today, at least. And that's an impact on alcohol and heroin-seeking behaviors. So can you tell me how you design these experiments? Sure.

So for the alcohol study, basically what you do is you give mice access to two different bottles. One bottle has water in it and the other bottle has some alcohol. And you give them access to both bottles on one day, the next day, they just have access to water. And then you keep flipping back and forth for about seven weeks. And so if you do this, the mice will start to drink the alcohol. alcohol to kind of comparable extents as people with alcohol use disorder. So their blood alcohol content will be very high, and they drink it kind of in this binge drinking fashion.

And then basically the next day when you take the alcohol away, it's kind of like you're inducing, you know, a hangover. And so if you do this for seven weeks, the mice start to really prefer the alcohol. What we did was we administered TBG. we waited three hours. So the acute effects of the drug had kind of worn off. And we had hoped that only kind of the longer lasting effects on neural circuitry would remain. And what you see is a pretty drastic reduction in alcohol consumption. And so we ran roughly 20 animals in this test. And I think every single one of them had a reduction in their alcohol consumption, but not necessarily a

reduction in their water consumption, which suggests that TBG is pretty selective for alcohol. So you had trained these mice, basically, to binge drink. You gave TBG and you saw that all of a sudden now they're not drinking as much anymore, which is a really incredible result for a behavior that has been programmed over seven weeks of training. Yeah, we agree. And again, this comes back to our initial hypothesis is that, in our opinion, the best way to treat these disorders is to fix the pathological neural circuitry that's associated with them. Yeah, which is the atrophying of the neurons in the,

prefrontal cortex. So that's the alcohol result. Let's talk about the heroin seeking behavior. So for this behavior, you need to train the rats to self-administer heroin. So you give them access to a couple of levers, and one of the levers will give them an infusion of heroin. And when they receive the infusion, they also are exposed to a sound and a light. And these serve as cues that's helped the rat associate the light and the sound with the rewarding experience of the heroin infusion. And this is very similar to what happens in people. You know, when a heroin addict is using in a particular location or around a certain group of friends, or they see drug paraphernalia, those things can trigger relapse because there's memories associated with those drug

cues. We first train the animal to self-administer heroin, and then at that point, we take the heroin away. And so then over time, they learn that they're not getting any reward from pressing the lever, so they stop pressing the lever. And then we play those drug cues, that light and that tone that was associated with the heroin previously. And that induces relapse and causes them to want to press the lever because they think heroin's available. And so what was really interesting about this experiment is that we gave TBG at three different time points. We gave it during self-administration. So this would be like if a human was actively using heroin. We gave it at the early phases of heroin extinction. So this would be like the

early stages of withdrawal for a human. And then we gave it right before this queued reinstatement, which would be somebody who had been drug-free for a little while and they were exposed to drug use. And we found that no matter when we gave it, there was a reduction in heroin-seeking behavior. And so what was really impressive was when we gave it during heroin administration or during early withdrawal, that was about one to two weeks before we did this cued reinstatement test, which suggests that Tavern Anthelag has some really long-lasting protective effect. against heroin relapse. The fact that it is able to block that trigger, desire almost,

that's one of the hardest things about recovering from drug abuse and all of this while not inducing hallucinations, which is just a really incredible result. I think this is a good segue into the broader context of this paper. Thinking about how this work goes from lab to clinic, paper to practice, do you think the work going forward is going to continue? to be an academic pursuit, or do you think that this is the type of project that would be better suited for a startup? So I would say both. In my academic lab, we're really interested in

understanding the mechanism of action of these drugs. And so we are continuing to do all of the basic science around these non-holutionogenic neuroplasticity promoting compounds. We call these things psychoplastogens. Now, to get something into the clinic for it to be useful for patients, There are a lot of other things that need to be done in the drug discovery process. And that is really more appropriate for a private company to handle. And so I started a company just recently with my colleague Nick Haft. That's called Delix Therapeutics. And Delix is licensed a lot of the technology that has come out of my lab at UC Davis

in the hopes of bringing these non-holucinogenic cycloplastogens to the clinic for treating a variety of brain disorders. Are there other psychedelics that are amenable to the same manipulations that you did in this paper? And are there other indications that they could possibly treat? Yeah, both my lab and delics has been doing this on all of the major psychedelic scaffolds. We do believe that these scaffolds might be better for certain indications than others, because they do have polypharmacology associated with them. They do hit some other receptors or their metabolism properties might be a little different.

that might make them more useful for one indication over another. Do the regulations, the fact that most psychedelics are Schedule I, drugs, does this impact your research or the path to the clinic for these drugs? I would say, I've got a lot to say about this. So I would say that from the company's standpoint, it's not really an issue because the company doesn't really work with psychedelic compounds. But from an academic standpoint, it has. been challenging. We like to compare all of the non-holucinogenic compounds that we make with their

hallucinogenic counterparts. That means that we need a special license to work with all of the Schedule 1 hallucinogenic compounds. In order to do that, you have to go through the DEA, and you have to also go through your state regulatory agencies to get all of that set up. It took me over two years, and it costs probably $50,000. But I think that the biggest barrier to basic science in psychedelic medicine is an inability to collaborate. Because a lot of other people don't want to go through the hassle just for one experiment, for instance.

And I think that this paper highlights it beautifully. If you look at our neuroplasticity assays in vivo, this was done in the lab of our collaborator, He's Woe at UC Santa Cruz. And he is a world expert in measuring neural activity in awake behaving animals. So that microscope, of course, is super specialized and super expensive.

And I'm not going to buy the $2 to $3 million microscope and get it set up just to try an experiment that would take one day. And Yi is not going to spend a ton of money and time and effort just to test one compound. And so in that experiment, you'll notice that the control compound that we use to compare the TBG, it was a hallucinogenic compound, but it's not a scheduled compound. But ideally, we would have compared TBG to five methoxy DMT because structurally they're more similar.

We're not able to do that because of the federal regulations. Yeah, I was aware of how difficult it was to obtain a license to work with these compounds and how difficult it is to set up your lab to do this and just the bureaucracy and paperwork. But I hadn't considered that aspect of it actually preventing collaboration because that's so true. That ability to create the highest caliber of work

is completely limited by the fact that you can't take the compounds to their lab and they have this hardware that they can't bring to you. Now, I'm not arguing to deregulate psychedelics or completely take them off the scheduled list that I don't believe that at all. they can be dangerous drugs, but they have a lot of scientific value.

20 years ago, science was done very differently than it is today. And so this one investigator, one license model might have been fine. But today's science is really done in the, you know, an interdisciplinary, cross labs, cross countries even. And that's because science has become so specialized for specific techniques. And they're so difficult to learn. takes your entire training to get really good at one thing, but then to answer a question, you need to take that one thing and juxtapose it with a whole bunch of other specialized

techniques. And people have been able to do this through collaborations, but that's something that the psychedelic science field has been really hampered by, with the regulations. David, thank you so much for joining me on Journal Club today. I really enjoyed our discussion. Thanks for having me. And that's it for Journal Club this week. If you enjoyed this episode, please subscribe, rate, and review wherever you listen to podcast. And to learn more about how biology is technology, subscribe to our newsletter at A16Z.com forward slash newsletters.