Welcome back to The Drop In, DoubleBlind’s newsletter serving up news, culture, and independent journalism about psychedelics straight to your inbox.

Today’s lead story is about a new type of genetically-modified, “humanized” mouse that was engineered specifically to study psychedelics. It feels like something out of Mary Shelley’s Frankenstein — and in a lot of ways it is that — but the researcher behind it argues that it fixes an issue that’s held rodent studies back for a long time. You can find that piece immediately below!

If you keep scrolling, you’ll find a story written by a delegation of Amazonian yagé healers, dangerous Trump-shaped pressies, a profile on Laganja Estranja, and an interview with a member of the Red Lake Nation who argues that the clinical model for psychedelics is built on the same colonial erasure Indigenous peoples have survived for centuries.

Blessings from Girona,
Mary Carreón
Editor-in-Chief

Using lab rodents for psychedelic research notoriously has its limitations. But a research team just engineered a new “humanized” rodent to fix that.

A lot of the data we have on psychedelics has been derived from research conducted on animals — more specifically, rodents. It’s a harrowing reality for animal lovers, but beyond that, the standard rodent models are limited in what they can tell us about how psychedelics impact the human brain. So why would we use mice in research at all?

The short answer is because accessing and working with rodents is a lot easier — fiscally and in terms of red tape — than researching on humans via clinical trials. Plus, rodents allow for more invasive research to be done, such as slicing the brain and genetically tagging receptors, which can’t be conducted on humans at all. That’s the premise behind a recent study published in Nature Neuroscience this past August. A research team led by Bryan Roth, a pharmacologist and former psychiatrist at the University of North Carolina at Chapel Hill, used CRISPR technology to engineer a new type of mouse carrying a humanized version of the 5-HT2A receptor that can be used for psychedelic research to determine where the receptors sit in the brain and what part of the cell they occupy. By doing this, we can have a greater understanding of how psychedelics work in the brain.

The reason a new mouse is needed is because the 5-HT2A receptor, the binding site classical psychedelics — like LSD and psilocybin — act on, isn’t quite the same in rodents as it is in humans. It differs by a single amino acid at the exact spot where the drugs bind to the receptors. That presents a problematic gap in our scientific understanding of how psychedelics impact the human brain. Genetically tagging receptors — or editing an animal's DNA so the receptor carries a glowing marker that lights up under a microscope and shows where it sits — has always presented this issue. The tools researchers used to find the receptor in brain tissue were also shaky, according to Roth.

"The [prior models] out there were just useless," Roth tells DoubleBlind in an interview. "People used them and tried to infer results from them."

The project came out of a grant from the Defense Advanced Research Projects Agency (DARPA) that originally worked with Roth’s lab to design therapeutic drugs without a psychedelic trip. Building those compounds — which, according to Roth, are novel compounds that have never existed before — meant the lab first needed animals to study that could properly register the effects. Now, the mice are available to any lab in the world.

Roth, who’s studied the 5-HT2A receptor for more than 40 years, says the study is the first truly detailed anatomical view of where the receptors live in the brain. DoubleBlind spoke to him about building a mouse from scratch, what psychedelics actually do once they reach the cortex, and the twist waiting at the end of nearly three years of work.

*This interview has been edited for length and clarity.

Bryan Roth: The mouse project arose from a grant we had with DARPA, the Defense Advanced Research Projects Agency, and they charged us with ultimately creating drugs that were therapeutic without being psychedelic. Along the way, we had various milestones to make, and one of them we realized even before we started the project was that there weren't any suitable mouse models to study the 5-HT2A receptor. The ones out there were just useless. People used them and tried to infer results from them. But those of us who study the receptor knew that the distribution of the reporter and all that stuff just didn't match with what was known about the receptor.

Then the other issue is that the mouse receptor is different from the human receptor in a significant way. It turns out there's a single amino acid in the binding pocket, this serine 242, and what we recently published and have published over the years is that every known class of psychedelic drug interacts with that serine in the human receptor. The mouse does not have a serine there, and that's missing a very significant interaction.

We also wanted to humanize the mouse, and we wanted to make it versatile, so in some cases we put fluorescent proteins attached to the receptor, or Cre recombinase, inducible or non-inducible. It turned out that one of the things these mice are probably going to be really useful to others for are causal neuroscience studies.

We made all these mice to test some compounds that ultimately we made for DARPA, and now they're trying to commercialize them and get them to humans. We did a huge amount of work validating the mice. The reviews were extraordinarily rigorous, so I think this is going to be useful to a lot of people out there. We published the paper. Anybody can go to Jackson Labs and get the mice.

BR: There are new compounds that we made. They're new to the universe. In terms of technology, there's a very nice paper that Brian Shoichet, myself, and others published in Science in 2024, where we use this technology called ultra-large-scale docking. It's an approach where you take compounds that don't exist in the physical universe but could, in theory, be synthesized by a chemist, and then you have the binding pocket, and you ping them with a billion molecules, and each molecule binds at a million different orientations, so the data sets are humongous. But then you rank them, and the goal was to get compounds that are absolutely new to the universe, optimize them, et cetera. So we did that, and along the way we needed these mice. Hence the mice. And now they're available for anybody to use.

BR: These days it's actually very simple. We go to the mouse store, the mouse genetic store, and we say we would like to make this mouse, and then they send us back a design, and then we get the mice. But really, it's basically like that. It's very simple these days using CRISPR technology. They take a wild-type mouse egg, a fertilized egg, and then just CRISPR in the DNA, and then they slip it in, and that's basically it. We start with the wild-type receptor, and then we'll stitch things in. We didn't actually build a mouse from the ground up.

BR: This paper was in review for almost three years. During that time, we shipped the mice to various other labs, and in the meantime, there's a lab somewhere else in another part of the world that made a similar mouse, a very similar mouse. They made it humanized, like we did, but they didn't add a lot of the other stuff. We've actually been in communication, and without saying too much, there was this one really strange result that they got with their mouse with a certain psychedelic drug, and so we tested it in our mouse, and we got the exact same weird response. So that's two humanized mice made independently, with a different strategy in different parts of the world, getting the same result. As I say, it's at least a reproducible artifact.

BR: No, it's part of his student's dissertation. I'll respect his privacy, but it is a very striking result.

BR: No.

BR: It's a funny thing. We sort of knew. I would say there was general consensus that this receptor, which is the main receptor for LSD and psilocybin, drugs like that, isn't a single band in the cortex. There was general agreement on that, but I would say there never was really convincing anatomical evidence for it because the antibodies were never any good. Normally you would use a technology called immunofluorescence histology, or immunohistochemistry, and you can really identify, no question, this is where those receptors are. They're on this particular neuron. They're on the cell body. They're inside the cell. You know where they are. There's all this controversy about where the heck they are in the cell, because we didn't have any way of detecting them.

And believe it or not, it astounds me. I've been studying this receptor literally for more than 40 years. Literally. I was there at the very beginning of the discovery of this receptor. People spent decades trying to make antibodies against this damn thing without success. Along the way, we learned how you can put a little protein tag on it to make sure you don't muck it up. So we did that, and the results we got were entirely consistent with the vast amount of prior literature, although consistently at higher resolution. It is kind of funny that this is probably the very first truly detailed anatomical view of where those receptors are in the brain, and what part of the cell they're on. Unbelievable.

BR: I would say our data are really consistent with hypotheses myself and others have had, that these receptors are localized on this part of the neuron called the apical dendrite, and they're in a particular type of cortical neuron called layer five A. Those neurons are responsible for interactions with other parts of the cortex. So you could think of them as maybe integrating information from every sensory and internal stimulus. And what psychedelics do is cause those neurons to start firing in this very asynchronous way.

So it puts noise into the system that is interpreting reality for you. And then because there's no reliable external stimulus, internally derived stimuli start to become attended to by the awareness. So the part of the brain that's responsible for telling you what reality is, all of a sudden, there's noise put in there. And then the brain starts making crap up. That makes sense as what happens, which I think is consistent with this entropic brain idea, to a certain extent. But I'm not an expert on that. That's what we're thinking.

BR: What we saw were variable responses depending on the psychedelic drug. With some psychedelics, the humanized mouse showed a greater response. I thought it was a lesser response. And then, as I said, we got this communication that if you look at a different psychedelic, you get this bizarre response. So what we're doing now, there are three labs that have experience with many different psychedelic drugs and two versions of this mouse. And what we're going to do is collate the data and then report it. But I think what it shows is that because this particular amino acid is essential for the action of psychedelic drugs, of course, if it's not there, the response is going to be different. And that's what we're finding.

BR: I don't think these animal models are that useful for ultimately explaining how it is. I'll give you an example. I have a friend, a fellow scientist, who was depressed. Seriously depressed. And he lived in a country where psilocybin-assisted therapy was legal, and he availed himself of that. A few months later, I didn't know about this, but a few months later I saw him at a meeting, and he was obviously changed. I'm a psychiatrist, but even somebody who's not a psychiatrist could tell there was something. His face was brighter, and he wasn't depressed anymore. So I went up to him, and he said, psilocybin-assisted therapy, and he said it was miraculous.

We're not going to get that from a mouse. But what we hope with my work, at the molecular level that I study, is that to a certain extent some of those processes are probably going to be conserved. So we're going to study those things until we can get a better model system. But it's difficult.