Ученые создали клеточный тест для быстрого выявления галлюциногенного потенциала психоделиков

A team of scientists spanning Spain, the United Kingdom, and the United States has identified a precise molecular fingerprint — a coordinated pattern of cellular signaling events — that reliably separates hallucinogenic psychedelics from their chemically similar, non-hallucinogenic counterparts, according to a paper published today in Translational Psychiatry, a Nature Portfolio journal. The finding could give drug developers something they have never had before: a fast, cell-based test for hallucination potential — one that could be run in a dish, on hundreds of candidate compounds at once, before a single animal is dosed. The hallucination barrier is not a minor inconvenience in psychedelic medicine. It is the central obstacle preventing compounds like psilocybin and LSD from reaching the estimated 84 million people worldwide with treatment-resistant depression — conditions that have failed to respond to at least two standard antidepressants. Every psilocybin dosing session currently requires two licensed clinical monitors on-site for eight or more hours, a physician accessible within fifteen minutes, and a specialized facility prepared for extended supervised care, according to FDA psychedelic dosing requirements. That structure is medically necessary because the hallucinations are real, intense, and can be distressing — but it also means psychedelic-derived therapy will remain inaccessible, expensive, and logistically complex for the foreseeable future unless something changes at the molecular level. Today's paper represents the most specific account yet of what, exactly, needs to change. Why Two Similar Molecules Can Produce Such Different Experiences The puzzle at the center of psychedelic pharmacology is one of its most striking features: psilocin (the active form of psilocybin) and lisuride, for example, are structurally related compounds that bind to the same primary target in the brain — the serotonin 5-HT2A receptor, a G protein-coupled receptor (GPCR) concentrated in the prefrontal cortex — yet produce radically different subjective effects. One is among the most potent hallucinogens known to exist; the other is a non-hallucinogenic drug used for decades in Parkinson's disease and migraine treatment. The conventional explanation has focused on the biased agonism phenomenon, or functional selectivity: the idea that two drugs can bind the same receptor but favor different downstream signaling arms, producing different cellular outcomes. Research published in Nature Communications in 2023 identified Gq protein efficacy — not beta-arrestin2 recruitment — as the key predictor of a compound's hallucinogenic potential in mice. A February 2026 paper in Nature, from researchers at West China Hospital, extended this with cryo-electron microscopy structures of the 5-HT2A receptor bound to both hallucinogenic and non-hallucinogenic compounds, finding that Gi protein signaling — a non-canonical arm of the same receptor — is essential for hallucinogenic effect, and that non-hallucinogenic analogues make a specific structural contact with the receptor that biases signaling away from Gi. What neither approach could provide was a systems-level picture of what happens inside the neural cell as a whole when a hallucinogenic compound — versus a non-hallucinogenic one — hits the receptor. Studying one signaling arm at a time is like identifying which key opens a lock without examining what the door, the room, or the alarm system behind it does once the key turns. How to Photograph a Cell's Entire Response at Once The research group led by Sandra M. Martin-Guerrero and Juan F. Lopez-Gimenez at the Instituto de Parasitología y Biomedicina "López-Neyra" (IPBLN-CSIC) in Granada, Spain — working with Pedro R. Cutillas at Queen Mary University of London and Javier González-Maeso at Virginia Commonwealth University — used phosphoproteomics mass spectrometry technique to take that system-level photograph, as described in the Translational Psychiatry paper. Phosphoproteomics is a branch of proteomics that uses mass spectrometry to identify and quantify thousands of phosphorylation events across the cell simultaneously. Phosphorylation — the attachment of a phosphate group to a protein at a specific site — is the cell's primary language for signal transduction: kinases write messages by adding phosphate groups, phosphatases erase them, and the resulting pattern of phosphorylated proteins at any given moment is a snapshot of the cell's entire signaling state. Between 30 and 65 percent of all human proteins are estimated to be phosphorylated at some point, making the phosphoproteome a comprehensive readout of what any stimulus — including a drug — is doing to a living cell. The team applied a panel of drugs to a neural cell model consisting of neurons and glia differentiated from murine neural stem cells — an important choice because the model retains the signaling machinery of native nervous tissue, including cell-to-cell contacts that simpler overexpression systems lack. The drug panel included four hallucinogenic serotonergic psychedelics (psilocin, DMT, LSD, and DOI), three non-hallucinogenic structural analogues (Ariadne, 2BrLSD, and lisuride), ketamine as a non-serotonergic hallucinogenic control, and BDNF as a neurotrophin control. The result was a dataset of phosphorylation changes across hundreds of proteins in each condition — and within that dataset, a distinct signature that consistently separated the hallucinogenic compounds from the non-hallucinogenic ones, as documented in the open phosphoproteomic dataset released alongside the paper. Decoding the Fingerprint: Coordinated, Not Single-Protein The most important thing to understand about the fingerprint the team found is what it is not. It is not a single protein that goes up or down in hallucinogenic conditions. It is a coordinated, global pattern of phosphorylation changes — a kind of molecular chord rather than a single note — that consistently appears across chemically diverse hallucinogenic compounds and consistently fails to appear when non-hallucinogenic analogues are tested instead. This matters because the field has spent years looking for that single-note marker and has struggled to find one that holds across compound classes. Hallucinogenic tryptamines, ergolines, and phenethylamines are structurally distinct — the fact that psilocin and LSD and DOI all cause hallucinations while sharing so little in their molecular architecture suggests the hallucination-producing mechanism lives downstream of receptor binding, in the intracellular machinery the receptor activates. A coordinated phosphoproteomic pattern is exactly where such a downstream convergence would show up. What the Glycolysis Shift Reveals Among the individual findings within the fingerprint, one stands out for its unexpectedness: the differential phosphorylation of FOXK2, a transcription factor that regulates the cell's glucose-processing machinery. FOXK2 belongs to the forkhead family of transcription factors and acts as a master regulator of aerobic glycolysis — the conversion of glucose to lactate as an energy source, even when oxygen is plentiful, as established by the FOXK2 glycolysis regulating role described in a 2019 Nature paper. FOXK2 promotes glycolysis by increasing the activity of key enzymes: hexokinase-2, phosphofructokinase, pyruvate kinase, and lactate dehydrogenase. When FOXK2 is active, cells produce more lactate and less of their energy through mitochondrial oxidation. When it is suppressed, the reverse happens. The Martin-Guerrero team found that hallucinogenic compounds, but not their non-hallucinogenic analogues, phosphorylate FOXK2 in a way that enhances lactate accumulation in neural cells, producing a 2- to 4.5-fold increase over vehicle controls, as detailed in the open phosphoproteomic dataset released publicly alongside the paper. Work from the same research group, described in a companion preprint, confirmed that this lactate response depends on both Gq/11 and Gi/o protein coupling at the 5-HT2A receptor — meaning it requires both major G-protein arms to be engaged. What does lactate production have to do with hallucinations? The honest answer is that no one yet knows the full chain of causation. What the finding does demonstrate is that hallucinogenic compounds push neural cell metabolism along a different trajectory than non-hallucinogenic ones — a difference that was invisible to all prior receptor-centric studies, which measured receptor binding and immediate downstream signaling events but did not track what the cell was doing with its energy as a consequence. It also adds a new candidate biomarker for any future hallucination screen: lactate accumulation, measurable by standard biochemical assay, now appears alongside the broader phosphoproteomic pattern as a functional readout. From Fingerprint to Drug Screen: What This Actually Enables The practical consequence of this work is what separates it from a technically elegant but narrowly academic finding. Currently, the standard method for predicting whether a new psychedelic-like compound will cause hallucinations in humans is the head-twitch response test in mice: a specific, involuntary rapid head movement that correlates reliably with 5-HT2A receptor activation and hallucinogenic potential in preclinical models. The test is well-validated, but it is an in vivo behavioral assay — it requires dosing animals, observing behavior, and it can only screen compounds one at a time in relatively small batches. The phosphoproteomic fingerprint, if it can be developed into a robust cellular assay, offers a fundamentally different screening approach: apply a candidate compound to a neural cell culture, measure the resulting phosphorylation pattern (or, more cheaply, measure lactate accumulation as a surrogate), and ask whether the hallucinogenic signature is present. This could be done in a high-throughput format, on hundreds of compounds simultaneously, before any animal testing begins. The authors were explicit about this implication: they released the open phosphoproteomic dataset released as an open community resource, specifically to enable hypothesis generation and targeted mechanistic validation by other research groups. In a field moving as fast as psychedelic pharmacology, releasing raw data openly is both a scientific norm and a competitive signal — it says the authors believe the findings are robust enough to invite scrutiny. Where Does This Land in the Race to Build Non-Hallucinogenic Psychedelics? The effort to develop therapeutically effective, non-hallucinogenic compounds is now a defined and competitive subfield of psychedelic pharmacology. Delix Therapeutics, in October 2025, published peer-reviewed data from Delix Therapeutics zalsupindole results in ACS Chemical Neuroscience showing its non-hallucinogenic neuroplastogen zalsupindole promotes neuroplasticity comparable to ketamine and psilocybin in preclinical models. UC Davis researchers reported in March 2026 that modified psilocin derivatives retained UC Davis psilocin derivatives study antidepressant potential in animal models without hallucinogenic effects. PNAS published an LSD-derived compound JRT study in April 2025 that produced therapeutic effects in animal behavioral assays without hallucinogenic behavioral markers. What all of these programs share is the absence of a validated, standardized cellular assay for hallucination potential. Each team has relied on the head-twitch response or closely related behavioral models to establish non-hallucinogenic status, which means each compound required its own animal study before human work could begin. A validated phosphoproteomic fingerprint assay would give the entire field a common cellular benchmark — the equivalent of a standardized safety screen that precedes other testing — and would allow medicinal chemists to generate and eliminate candidate structures far faster at the molecular design stage. The Central Open Question: Can the Therapeutic Effect Survive? Today's paper advances the field in precisely the way the field needs: it maps the cellular territory more completely and provides practical tools for the next step of drug design. What it cannot yet answer is the debate that underlies the entire enterprise of non-hallucinogenic psychedelic development. The question is whether the hallucinogenic experience — the subjective "trip" — is necessary for the therapeutic effect, or whether it is an epiphenomenon that can be stripped away without losing efficacy. On one side of this debate: researchers who argue that the transformative altered state is what reorganizes neural circuits and produces lasting psychological change, meaning that removing it might also remove the benefit. On the other: researchers who argue that neuroplastic and anti-inflammatory mechanisms can be fully decoupled from perception, meaning the drug's molecular effects on circuits are the therapy, and the trip is a side effect. The FDA, in its FDA July 2026 psychedelic guidance, made the stakes of this question explicit: the contribution of the psychotherapy component — and by extension, the subjective psychedelic experience — to any observed efficacy "has not been characterized." Compass Pathways, which holds the most advanced NDA-track psilocybin program — its Compass Pathways COMP360 NDA targeting treatment-resistant depression has received rolling NDA review and a Commissioner's National Priority Voucher, with a complete submission expected in Q4 2026 — built its clinical program around the traditional supervised-session model, which includes the hallucinogenic experience. A non-hallucinogenic compound that retains efficacy would change that clinical model entirely. It could be prescribed in a standard outpatient setting, dispensed at a pharmacy, and taken without eight hours of supervised monitoring — making it accessible to millions of patients who currently cannot access psychedelic-assisted therapy because of cost, geography, or logistical barriers. But it would first need to prove efficacy in clinical trials, and those trials would not exist without a reliable early-stage screen to identify which compounds are worth advancing. The phosphoproteomic fingerprint is, in that sense, a tool that could accelerate every other step that follows. The study's authors note that their findings are based on in vitro conditions — cultured neural cells, not living human brains — with the in vitro limitation acknowledged explicitly in the paper: whether the same signature reliably predicts hallucinogenic effects across brain regions, species, and ultimately in human clinical settings remains to be established. But the clearest cellular-level map yet of the divide between hallucinogenic and non-hallucinogenic psychedelics is now public, open, and available to every lab in the world trying to cross it. Frequently Asked Questions What is a hallucination "fingerprint," and how was it found? The fingerprint is a coordinated pattern of protein phosphorylation events — chemical tags added to hundreds of proteins inside neural cells — that consistently appears when hallucinogenic psychedelics like psilocybin, LSD, and DMT are applied, and consistently fails to appear when non-hallucinogenic structural analogues are applied instead. Scientists at IPBLN-CSIC, Queen Mary University of London, and Virginia Commonwealth University found it using phosphoproteomics mass spectrometry technique, which catalogs the cell's entire phosphorylation state simultaneously rather than measuring one signaling protein at a time. The power of the approach is that it captures the cell's coordinated response to a drug as a whole — which is where convergent features across structurally diverse hallucinogens were most likely to emerge. Why does it matter that hallucinogenic psychedelics shift neural cell metabolism toward lactate production? The finding that hallucinogenic compounds — but not non-hallucinogenic ones — specifically phosphorylate the glycolysis-regulating transcription factor FOXK2, boosting lactate accumulation in neural cells, is unexpected and significant for two reasons. First, it reveals that hallucinogenic compounds alter not just receptor signaling but the cell's energy metabolism — a downstream consequence invisible to receptor-centric studies. Second, lactate accumulation is a simple, measurable biochemical output that could serve as a practical readout in a future drug-screening assay. If a new compound triggers the FOXK2/lactate response in neural cells, that is a red flag for hallucinogenic potential, detectable before any animal testing or human exposure. The biological reason lactate specifically links to hallucination — what the metabolic shift does to neural circuits — remains an open research question, touching on what the FOXK2 glycolysis regulating role means for neural energy states. How does this change the process of developing non-hallucinogenic psychiatric medicines? Currently, chemists who want to know whether a new compound will cause hallucinations must test it in animal behavioral models — typically the head-twitch response in mice — which requires in vivo experiments and screens only one compound at a time. The phosphoproteomic fingerprint (and the simpler lactate surrogate) could enable a cell-based, high-throughput screening step: apply a panel of candidate compounds to neural cells, measure the phosphoproteomic or lactate response, and flag any compound that produces the hallucinogenic signature for elimination before resources are spent on animal studies. The authors released the full dataset as an open phosphoproteomic dataset released specifically to enable this kind of downstream application. Multiple companies — including Delix Therapeutics and UC Davis spinoffs — are racing to develop non-hallucinogenic psychedelic-class drugs, and a validated cellular hallucination assay would accelerate all of them. Are hallucinations actually necessary for psychedelic therapy to work? This is the central unresolved question in the field, and today's paper does not settle it. Some researchers argue that the transformative altered state is integral to how psychedelics reorganize neural circuits and produce lasting psychological change — in which case removing the hallucinations might also remove the benefit. Others argue that neuroplastic and anti-inflammatory molecular effects can be completely decoupled from the subjective experience. The FDA stated in its FDA July 2026 psychedelic guidance that the contribution of the psychotherapy component — and by extension, the subjective experience — to any observed therapeutic efficacy has not yet been characterized. What today's paper provides is the molecular toolkit to design and quickly screen compounds that lack the hallucinogenic signal — so that the efficacy question can finally be tested head-to-head in clinical trials. ⓒ 2026 TECHTIMES.com All rights reserved. Do not reproduce without permission.