A new study published in Translational Psychiatry suggests that miniature, lab-grown brain models can reveal distinct electrical activity patterns in different types of autism. By analyzing brain tissues grown from patient urine samples, scientists provide evidence that these models can accurately distinguish between neurotypical individuals and those with various autism profiles. These findings tend to offer a new way to understand the biological roots of autism and test personalized treatments.
Autism is a neurodevelopmental condition characterized by differences in social communication and restricted, repetitive behaviors. While some cases are linked to specific genetic mutations, known as syndromic autism, most cases have unknown origins and are classified as idiopathic.
Traditional animal models often struggle to replicate the complex features of the human brain. This makes it difficult to study how specific genetic changes affect human brain function. Patient-derived brain organoids offer a biological solution to this problem.
Brain organoids are tiny, three-dimensional clusters of brain cells grown in a laboratory that mimic early human brain development. Because they are grown from a patient’s own cells, they retain that individual’s unique genetic makeup. This allows scientists to study human brain networks in a highly personalized way.
“During my PhD in neuroscience, I studied new methods for drug development and delivery to neurological and psychiatric disorders, with a focus on autism spectrum disorders,” said study author Nisim Perets, the CEO and co-founder of Itay&Beyond.
“I published several papers on the use of nanoparticles (exosomes) as a therapeutic and delivery potential for autism. These papers became well known in this field, and one day, I was approached by a high-tech entrepreneur with a son who was diagnosed with low-functioning autism. Together, we established Itay&Beyond and decided to start with a ‘simulation’ of the patient’s brain to develop and test a new generation of drugs and compounds based on the patient’s biology, rather than animal models.”
To conduct the study, the researchers collected urine samples from 15 participants. This group included 11 individuals diagnosed with autism and four neurotypical controls. Among the participants with autism, ten had syndromic autism linked to five specific genetic mutations, and one had idiopathic autism.
The specific genetic conditions studied included variations in the SHANK3, PPP2R5D, SCN2A, GRIN2B, and STXBP1 genes. These genes are known to play distinct roles in how brain cells develop, send signals, and form neural connections. The researchers extracted epithelial cells, which are cells that line the urinary tract, from the urine.
They then reprogrammed these cells into induced pluripotent stem cells. These are a special type of cell that can be guided to become almost any cell type in the human body. The scientists directed these stem cells to develop into more than 400 individual brain organoids over the course of about 60 days.
Once the organoids matured, the researchers placed them on multi-electrode arrays. These are small microchips equipped with sensors that can record the electrical signals sent between neurons. Neurons are the primary cells in the brain responsible for sending and receiving information.
The scientists recorded the resting electrical activity of the organoids for five minutes. They then applied a brief electrical stimulation and recorded the activity for another five minutes. This procedure allowed them to measure 18 different electrical features.
These features included the firing rate of the neurons, the frequency of synchronized bursts of activity, and the overall connectivity of the neural network. To make sense of this complex data, the researchers used a mathematical technique called principal component analysis. This method compresses complex data into a simplified visual map, allowing researchers to group similar electrical patterns together.
The scientists found distinct differences in the electrical activity of organoids derived from autistic individuals compared to the control group. The organoids from the four neurotypical controls displayed highly consistent electrical patterns. They clustered closely together in the data analysis, showing low variability.
“Some of the results were better than expected,” Perets said. “Our first question was whether there was any difference in organoid activity between patients with autism and controls, but when we started analyzing the data, we saw interesting differentiation also between the sub-populations of the patients with autism spectrum disorder. ”
The organoids from the individual with idiopathic autism tended to exhibit reduced electrical activity. These samples showed significantly lower firing rates and fewer bursts of activity than the control samples.
Most organoids derived from patients with syndromic autism provided evidence of hyperactivity. For example, the samples linked to STXBP1, PPP2R5D, and GRIN2B mutations demonstrated significantly increased firing rates. The SCN2A samples showed mixed firing rates but had noticeably reduced electrical signal amplitudes compared to controls.
The researchers also observed differences in how the neural networks responded to electrical stimulation. This part of the experiment measured short-term synaptic plasticity. Synaptic plasticity is the ability of a neural network to adapt by strengthening or weakening the connections between neurons over short periods.
The scientists specifically looked at short-term depression and short-term potentiation. Short-term depression involves a temporary decrease in neural communication, while short-term potentiation involves a temporary increase. The data suggests that specific genetic mutations severely alter this natural balance.
For instance, organoids with STXBP1 and SHANK3 mutations showed significantly reduced short-term potentiation and increased short-term depression compared to controls. In the control organoids, the networks maintained a relatively stable functional size and density after being stimulated. Many of the autism-derived organoids displayed altered structural responses to this electrical input.
Organoids with STXBP1 mutations exhibited a pronounced and early collapse in network connectivity after stimulation. The samples linked to PPP2R5D mutations maintained high connectivity before stimulation but experienced a sharp drop immediately after. This suggests an underlying network fragility or an impaired ability to recover from external inputs.
The researchers noted that even organoids from patients with the exact same genetic mutation sometimes displayed different electrical profiles. For example, abnormal rhythmic bursting was seen in a sample from a patient with a history of seizures, but not in all samples from that genetic group. This observation aligns with the fact that individuals with the exact same genetic diagnosis can exhibit very different daily symptoms.
“We successfully showed that our technology is not only non-invasive but can also differentiate between lab-grown brain tissues from patients with autism spectrum disorder and control patients; moreover, the data show that the system can differentiate between different autism subpopulations based on the electrophysiological activity of the lab-grown brain tissues,” Perets told PsyPost.
While these models provide detailed insights into neural activity, the scientists caution against overstating the current capabilities of the technology. Brain organoids are still in the developmental stages of research. They do not simulate the complete structure or complex anatomy of a fully formed human brain.
The organoid networks are simplified models. This means they cannot capture every aspect of how a whole, mature brain functions or experiences the world. Future research aims to use this technology to develop and test new medications for various psychiatric and neurological conditions.
“Itay&Beyond aims to develop and test a new generation of drugs for neurological and psychiatric disorders, including autism, epilepsy, dementia, schizophrenia, and more, and to produce patient-based functional readout for the efficacy and safety of drugs,” Perets added.
“Our technology is already being used to test drugs and compounds for pharma companies and academic institutions, and to help them with their studies. We also test and develop our own proprietary drugs for the subpopulation of autism and other disorders.
“Excitingly, our technology is a platform for human-brain computer interaction and can extend far beyond drug development and testing. Some of our collaborations are not in the field of medicine and drug development, but in deep technologies and new models for brain-computer interfaces, such as biological neuronal networks and computational, energy-efficient models for AI.”
The study, “Patient-derived brain organoids reveal divergent neuronal activity across subpopulations of autism spectrum disorder,” was authored by Nisim Perets, Liya Kerem, Nir Waiskopf, Noa Horesh, Itay Goldman, Jasmine Avichzer, Doron Bril, William Tobelaim, Milcah Barashi, Liat David, and Ariel Tenenbaum.
