Using Stem Cells to Cure Autism, Epilepsy & Schizophrenia | Dr. Sergiu Pașca
Autism, Schizophrenia, Stem Cells, Organoids, and Assembloids with Dr. Sergiu Pasca
概览
This episode centers on how scientists can move beyond behavior-only labels in autism, schizophrenia, and other neurodevelopmental disorders. Dr. Sergiu Pasca argues that autism is not one disease, that profound autism should be distinguished from high-functioning autistic traits, and that genetics offers the clearest causal foothold for some severe cases.
The discussion then turns to stem cells, induced pluripotent stem cells, organoids, and assembloids. Pasca explains how patient-derived cells can be reprogrammed, grown into human neural cells, assembled into circuits, and used to observe developmental timing, disease mechanisms, and treatment responses outside the human brain.
The central conclusion is cautious but optimistic: unregulated stem cell injections and simplistic causal stories are not supported, but rigorous human stem-cell models are beginning to produce actionable therapies. Timothy syndrome is presented as the leading example, with a nucleic-acid therapeutic moving toward clinical testing.
分段落总结
[00:00] Episode frame and guest
[事实] Andrew Huberman introduces Dr. Sergiu Pasca as a Stanford professor of psychiatry and behavioral sciences and director of the Stanford Brain Organogenesis Program.
[事实] The episode covers autism, schizophrenia, human brain development from pregnancy through the third decade of life, and stem-cell-derived organoids and assembloids.
[事实] Huberman says the discussion includes rising autism prevalence, the role of genes, and treatments being developed for profound autism.
[02:12] Autism as a behavioral spectrum
[事实] Pasca describes autism and neurodevelopmental disorders as behaviorally defined conditions with no biomarker.
[事实] He says autism prevalence is now close to almost 3% of the general population, whereas it was considered rare when he was in medical school.
[事实] He distinguishes autistic traits in fully functional individuals from profound autism, which can involve lifelong care, intellectual disability, and epilepsy.
[事实] He compares autism to older medical use of fever: one outward category may contain many different biological causes requiring different approaches.
[07:17] Sex differences and early behavioral features
[事实] Pasca says autism has generally been reported at roughly a four-to-one ratio, with more males than females.
[事实] He notes that girls may sometimes be underdiagnosed because of masking, while also saying male and female nervous systems mature differently and may respond differently to injury.
[事实] Reduced eye-focused gaze is not a diagnostic criterion, but joint attention and related social-attention features are associated with autism.
[事实] Reports that fever temporarily improves autism symptoms are described as mostly anecdotal, with possible explanations involving neuromodulators, cytokines, ion channels, or altered circuit function.
[15:44] Microbiome, sleep, diet, and genes
[事实] Pasca says improving the microbiome may improve quality of life for some individuals, but there is no clear evidence that microbiome deficits cause autism.
[事实] He says 70-80% of profoundly impaired patients can have severe sleep disturbances, and improving sleep could meaningfully improve functioning.
[事实] He notes that picky eating and dietary issues can affect patients, but most evidence points toward a strong genetic component.
[事实] Autism-linked genes include synaptic proteins, ion channels, chromatin-related proteins, and genes also expressed outside the brain.
[18:27] Genetic diagnosis and current treatment limits
[事实] Pasca says about 20% of children with autism may leave clinic with a genetic diagnosis, including inherited or de novo mutations.
[事实] Timothy syndrome is described as a genetic form involving a calcium channel mutation, autism, epilepsy, and sometimes webbing of fingers.
[事实] A genetic diagnosis can help families join a disease community and prepare for future targeted therapies, but most forms still lack specific treatments today.
[推测] Current care therefore remains mostly supportive, behavioral, and focused on associated problems rather than a single autism treatment.
[21:47] Why autism prevalence appears to rise
[事实] Pasca says the rise in autism prevalence is still puzzling and is partly explained by changed diagnostic criteria, diagnostic migration from intellectual disability, and service availability.
[事实] He says autism has very high heritability among psychiatric disorders, but many rare genetic forms remain unmapped.
[事实] Environmental exposures such as thalidomide historically increased autism risk, though thalidomide is no longer prescribed for that pregnancy-related use.
[事实] Huberman and Pasca state that there is no epidemiological evidence that vaccines cause autism.
[29:30] Human brain inaccessibility and slow development
[事实] Pasca emphasizes that schizophrenia appears across populations at around 1%, and that human brain disorders are hard to test causally because scientists cannot switch variables on and off in living human brains.
[事实] He says human brain development is unusually slow: myelination continues into the third decade, and neurons are born and migrate into early postnatal years.
[事实] Reported autism prevalence in countries such as Korea and Scandinavian countries appears around one in 30 to one in 40, making a purely United States-specific explanation unlikely.
[事实] Pasca notes that the same mutations can appear in patients from different countries and continents.
[33:26] Gene therapy, CRISPR, and delivery problems
[事实] Pasca defines gene therapy broadly as adding a missing gene, providing a protein or enzyme, or directly correcting DNA with tools such as CRISPR.
[事实] He explains that CRISPR requires guide nucleic acids and an enzyme to target and alter DNA, but delivery into affected cells remains a major challenge.
[事实] Brain delivery can involve surgery or intrathecal routes, but cell-type diversity, limited efficiency, and viral cargo-size limits make treatment difficult.
[事实] He warns that some gene therapies can be irreversible, may provoke immune responses after viral delivery, and require careful risk control.
[43:45] Stem cells and induced pluripotency
[事实] Pasca defines stem cells by two properties: they can become other cell types and they can self-renew under proper conditions.
[事实] Human embryonic stem cells became maintainable in dishes around 1998, but raised ethical questions because of their embryonic source.
[事实] Shinya Yamanaka showed that skin cells could be reprogrammed into induced pluripotent stem cells using a small combination of factors now called Yamanaka factors.
[事实] Pasca saw induced pluripotent stem cells as a way to make patient-derived human neurons and study live cellular activity without brain biopsies.
[52:54] Umbilical cells and commercial stem cell injections
[事实] Pasca says umbilical-cord stem cells are already restricted and mainly relevant to blood disorders, not universal regenerative therapies.
[事实] He says unregulated stem cell injections may use patient cells, umbilical cells, donor cells, or unknown material, and are often not supported by systematic trials.
[事实] For autism, he says there is no clear biological rationale for injecting stem cells because autism is not understood as a missing-cell condition in the brain.
[事实] He attributes some reported improvements to placebo by proxy, ongoing child development, or possible inflammatory effects, while emphasizing real safety risks.
[60:33] Organoids and Timothy syndrome models
[事实] Pasca describes induced pluripotent stem cells as cells reprogrammed into pluripotency and then guided toward neurons using developmental biology cues.
[事实] In Timothy syndrome patient-derived neurons, his team observed calcium entering and remaining longer than in control cells after stimulation.
[事实] To keep developing neurons alive longer, his lab moved from flat cultures to floating three-dimensional aggregates, later called organoids.
[事实] These organoids are not whole brains; they are self-organizing representations of parts of the nervous system.
[69:30] The developmental timer in organoids
[事实] Pasca’s lab maintained organoid cultures for hundreds of days and even years.
[事实] Around nine months in a dish, organoids switched to a postnatal-like signature, including the known NMDA receptor subunit change from 2B toward 2A.
[事实] Pasca says this suggests an intrinsic timer in human neural cells, though the molecular mechanism remains unknown.
[事实] He says understanding that timer could help scientists make aged neurons faster or explore whether aged neurons can be rejuvenated.
[72:24] Assembloids and self-organizing circuits
[事实] Pasca says autism and schizophrenia likely involve connectivity defects more than simply missing cell types.
[事实] His lab combined organoids representing different brain regions so inhibitory neurons could migrate toward cortical tissue, creating the first assembloid around 2015.
[事实] The term “assembloid” came from assembling circuit parts and was encouraged through discussions with Ben Barres.
[事实] Pasca says that if the right parts are made and placed properly, cells carry enough developmental instructions to connect themselves.
[78:24] Motor and sensory pathway models
[事实] A motor assembloid combined cortex-like tissue, spinal-cord-like tissue, and human muscle so that stimulating cortex could make muscle contract.
[事实] Pasca says this showed cells could find partners and form a functional circuit despite low expected odds.
[事实] A later four-part somatosensory assembloid connected sensory neurons, spinal cord, thalamus, and cortex in the correct order.
[事实] The sensory model can be used to study genetic pain conditions and potentially test pain medications.
[97:17] Ethics, nomenclature, and science communication
[事实] Pasca says his Stanford center works with ethicists, law scholars, sociologists of religion, and others on neuro-organoid, assembloid, and transplantation ethics.
[事实] Ethical issues include consent for human cells, animal welfare after transplantation, pain management, and possible emergent properties in circuits.
[事实] He says current organoids and assembloids are not entire brains, and he discourages terms like “mini-brains” because they mislead the public.
[事实] Researchers in the field developed consensus nomenclature and reporting frameworks through consortium efforts.
[105:41] Transplantation and in vivo validation
[事实] Pasca says transplantation questions should be reviewed before experiments, and his Stanford experiments undergo ethical approval.
[事实] Adult brains are less permissive to new large-scale connections, so early developmental timing is important for integration.
[事实] His lab transplanted cortical organoids into newborn rat somatosensory cortex, where grafts became vascularized, received microglia, and formed connections.
[事实] Human neurons in the graft could respond when the rat’s opposite whiskers were moved, and transplanted neurons grew six to eight times larger than dish-grown ones.
[113:34] Testing Timothy syndrome therapy in living systems
[事实] Transplantation revealed a Timothy syndrome phenotype not visible in the dish: patient-derived neurons were much smaller than control neurons after in vivo growth.
[事实] The model allowed researchers to inject a therapeutic into the animal nervous system and observe effects on human neurons in an in vivo context.
[事实] Pasca says cross-species integration is imperfect because rat cortex development is far faster than human cortical development.
[推测] The approach is intended as a disease-modeling and therapeutic-testing platform, not as a route to perfect human-animal neural integration.
[118:50] Genetic testing, penetrance, and 22q11.2 deletion
[事实] Huberman raises embryo sequencing, parental testing, and eugenics concerns in relation to autism-linked genes.
[事实] Pasca says some conditions are deterministic, but many mutations have variable penetrance, meaning the same mutation can cause severe, mild, or minimal phenotypes.
[事实] 22q11.2 deletion syndrome is described as the most common human microdeletion at about one in 3,000 births, involving around 60 genes in the classic deletion.
[事实] Pasca says it carries about 30% risk of schizophrenia and about 30% risk of autism, but some carriers have minimal visible effects.
[122:41] From models to therapies
[事实] Timothy syndrome became the first major example because Pasca’s team studied it across 2D neurons, 3D organoids, assembloids, and transplantation.
[事实] His team developed a tiny nucleic-acid therapeutic that changes how the calcium channel is processed and rescues all defects they had described in their models.
[事实] Pasca says they have found patients worldwide, built a cohort, started producing the drug, and are preparing for a first clinical trial.
[事实] He frames this as potentially the first psychiatric-disease therapeutic developed exclusively through human stem-cell models.
[126:37] Other neurological and psychiatric targets
[事实] Pasca’s lab is also modeling severe genetic epilepsies, including conditions with up to 60 seizures per day.
[事实] Other targets include intellectual disability, schizophrenia, and 22q11.2 deletion syndrome as a high-risk entry point for schizophrenia.
[事实] He says conditions strongly driven by social stress, such as some anxiety or depression forms, are harder to model because the social environment cannot be reproduced in a dish.
[推测] The platform is best suited first to disorders with clear biological or genetic entry points.
[128:14] Dystonia and the meaning of profound autism
[事实] Huberman asks Pasca’s lab to tackle dystonia after hearing about a child with a devastating movement disorder.
[事实] Pasca says his lab is working on dystonias and dyskinesias caused by genetic mutations that lead to severe uncontrollable movements.
[事实] The lab has built loop assembloids involving cortex, striatum, midbrain, and thalamus to study basal-ganglia-related circuits.
[事实] Pasca clarifies that his therapeutic goal in autism is focused on profound forms, not high-functioning individuals for whom autism may be part of identity.
[134:34] Scientific motivation and closing
[事实] Pasca says he thinks about the work all the time and does not see it as work.
[事实] His lab and trainees have helped more than 350 labs worldwide implement these technologies through Stanford courses.
[事实] He says clinical exposure to devastating psychiatric disorders motivated him to enter neuroscience.
[事实] In closing, Huberman emphasizes the path from human disease to models, circuits, therapies, and patients, and Pasca mentions walking 12,000 to 15,000 steps per day, art, and typically eating one meal daily.
播客点评/总结
[推测] This episode’s strongest value is translation: it explains organoids and assembloids in accessible language while repeatedly warning against hype, misleading names, and unsupported stem-cell interventions.
[推测] The highlight is the concrete arc from a single calcium-channel mutation in Timothy syndrome to patient-derived neurons, organoids, transplantation models, and a therapeutic candidate. That makes the technology feel less like a laboratory novelty and more like a practical route toward treatments.
[推测] The main limitation is that much of the therapeutic promise is still early-stage. The discussion is strongest for genetic and profound neurodevelopmental disorders, while broader autism, schizophrenia, anxiety, and depression remain harder to explain or model.
[推测] The episode is especially suited for listeners interested in neuroscience, families affected by profound autism or rare genetic disorders, and anyone trying to distinguish evidence-based brain research from oversimplified claims about vaccines, microbiomes, CRISPR, or stem-cell tourism.