When Olson was at Stanford, he learned from a mentor who had honed a new method for developing drugs: function-oriented synthesis. Within a given molecule, specific groups of atoms could be catalogued according to their individual effects on the body. If you determined which group did what, then you could potentially synthesize a compound that isolates what you wanted, leaving out the rest. “It’s a very reductionist approach,” Olson said. He compared a chemist using this method to a mechanic working on a car. “It’s got all these complex parts,” he said. But those parts can be grouped by function—axles go in this bin, spark plugs go in that bin—and you could change the car’s performance by adjusting its components. Olson’s theory was that one part of a psychedelic molecule caused a trip, while another stimulated dendritic growth. If he could remove some of the former but preserve a little of the latter, then he might have a recipe for a non-psychedelic psychedelic medicine.
Olson, who has a shaved head and piercing eyes, showed me how the scientists in his lab break psychedelic molecules into parts, as though they’re cars in a chop shop, and build new ones. There were beakers everywhere, full of chemical reagents such as sodium hydrosulfite and inorganic bases. We walked by liquid-chromatography machines and hulking specimen freezers. Graduate students in tie-dye shirts worked under fume hoods; on the glass that protected them from chemicals, synthesis reactions were scribbled in black marker. One researcher, Andrian Basargin, explained that he was making a substructure of LSD in an acetone and dry-ice bath. It would become a component in a new compound, which would then be taken for a test drive.
Olson got hints on where to begin from an odd pair of books by Alexander and Ann Shulgin: “PiHKAL,” short for “Phenethylamines I Have Known and Loved,” and “TiHKAL: The Continuation,” short for “Tryptamines I Have Known and Loved.” Alexander was a chemist who, beginning in the nineteen-sixties, created nearly two hundred novel chemical compounds, many of them psychedelic, and tested some of them on himself. The books contain extensive notes on the drugs’ synthesis and effects. Olson tasked a grad student with reading the books and cross-referencing drug forums on Reddit—“kind of a weird thing,” he admitted. He wanted to know which of Shulgin’s concoctions didn’t produce much of a trip. “Some of the first molecules we made were informed by procedures from those books,” he told me.
A long trial-and-error process gave Olson a sense of which molecular motifs seemed likely to cause mind-altering effects. “You make a change, you do a round of testing, then you see, Oh, this change takes us closer to where we want to be,” he said. He showed me a large black box about the size of an industrial printer. Inside were lab-grown cells studded with modified receptors. A new substance would be squirted over the cells, Olson explained. If the molecule was likely to have hallucinogenic properties, it would trigger a fluorescence reaction that sensors in the box would detect. This helped the team weed out trippy compounds. In this way, for example, the team discovered that flipping two atoms within the LSD molecule—Olson compared the change to a tire rotation—affected how hallucinogenic it was. To confirm these findings, this new compound was also tested on rodents.
If a molecule passed these tests, the next step was to determine whether it stimulated the growth of dendrites. In another lab across Davis’s leafy campus, we met John Gray, a neuroscientist who tested experimental drugs on living neurons. Slices of mouse brain floated in dishes of synthetic cerebrospinal fluid; on a monitor, I could see a single teardrop-shaped neuron. I watched through a microscope as a postdoc, Raghava Jagadeesh Salaka, broke through the neuron’s cell membrane with a micropipette. It looked like a needle pricking a dollop of translucent jelly.
The micropipette allowed the researchers to measure electrical activity in the neuron, Gray explained. Blue spikes, which suggested increased signalling and connectivity, soon appeared on another computer monitor. The team also looked for physical changes. After exposing neurons to a compound, they used a microscope to count any new dendritic spines that had popped up.
