A Neuronal Hint: BRAIN Initiative introduces world’s first comprehensive brain cells atlas
by John Kevin Ceasar S. Palay
Hundreds of neuroscientists built a ‘parts list’ of the motor cortex, laying groundwork to map the whole brain and better understand how cellular networks are disrupted in mental and physical disorders.
From extensive investigations of mice, monkeys, and humans, the NIH Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative Cell Census Network (BICCN) has released an atlas of cell types and neuronal wiring map for the mammalian primary motor cortex.
This freely accessible resource is the result of a global collaboration involving more than 250 scientists from more than 45 universities on three continents. The atlas lays the groundwork for a more in-depth examination of cell types in the remainder of the mammalian brain, and it could aid scientists in interpreting findings from animal models to humans. The findings were presented in 17 related studies in a special issue of the journal Nature on October 6, 2021.
With over a hundred billion neurons, an equal number of non-neural cells, and trillions of neuronal connections, the brain is considered to be a hugely complicated organ. Even though people have the same basic brain anatomy, how brains develop and operate are influenced by a number of environmental and hereditary variables. To put it another way, each person’s brain reflects his/her individual history and life experiences. Understanding what makes a human a human requires a thorough understanding of all the cell types that make up the brain, their properties, and how they differ from person to person.
It is of great academic and clinical interest to list all sorts of brain cells and characterize their forms, chemical components, and input-output activities in detail. Defects and vulnerabilities in certain types of cells are the roots of many neurological disorders such as Retinal Blindness, Spinal Muscular Atrophy, Dravet Syndrome or sever myotonic Epilepsy of infancy, Frontotemporal Dementia, Alzheimer’s Disease, Amyotrophic Lateral Sclerosis (ALS), and Schizophrenia.
That was why the United States’ BRAIN effort, led by the National Institutes of Health, established the Brain Initiative Cell Census Network (BICCN) in 2017 to address the demand for such a cell census. Its goal is to find about all of the many cell types that make up the mammalian brain. BICCN is a large-scale collaboration between top researchers at universities and non-profit research institutes across the United States, funded by a number of large grants. The researchers focused on the main motor cortex, which orchestrates movements among mammalian species.
“There is an urgent need to develop therapies for brain-based disorders. Research conducted as part of BICCN will provide scientists with tools that enable precise targeting of cell types and neural networks by genomic therapy for disorders that affect thinking, memory, mood, and movement,” said Walter Koroshetz, a doctor and the director of the National Institute of Neurological Disorders and Stroke.
Scientists have already defined hundreds of cell kinds based on their form, size, electrical characteristics, and gene expression patterns. Although many are subtypes of well-known cell types, the latest research uncovered approximately five times more cell types. For humans, 127 types emerged. In that sense, cells that produce particular neurotransmitters, such as gamma-aminobutyric acid or glutamate, can be divided into more than a dozen subtypes based on gene expression and electrical firing patterns.
Through a powerful piece of molecular technology, the single-cell RNA sequencing or scRNA-seq, the BICCN team was able to identify distinct cell types in three specified mammals. This technique is referred to as transcriptomics, which involves identifying all the specific messenger RNA molecules and their levels in each cell. More than that, the study also involved various methods for measuring gene expression levels, the genome’s chromatin structure, and the DNA methylation status, also known as the epigenome (a series of chemical changes to a cell’s DNA that modifies how the cell’s genetic information is expressed). Classic electrophysiological patch clamp recordings were used to differentiate cells based on how they fire action potentials, as well as classifying cells by form, determining their connections, and looking at where the cells are physically situated within the brain. To differentiate cell types, some of these employed machine learning or artificial intelligence.
As a result of the series of studies, the neuroscientists discovered common cell types in mice, monkeys, and humans, as well as significant variances in gene expression that might explain discrepancies in how these three species receive brain information.
“Our results illuminate clearer paths for future studies to follow by illustrating how different brain structures organized into networks and communicate with one another,” said Hong-Wei Dong, a co-corresponding author of the flagship paper who co-leads the UCLA Brain Research & Artificial Intelligence Nexus and is bound to explore the subthalamic nucleus, which is a crucial objective for the treatment of Parkinson’s Disease using deep brain stimulation in the future.
These preliminary findings serve as a starting point for future studies into the structure and function of cells in the mammalian brain. These studies might involve a study into how the brain grows and develops as well as the functions that different cell types have in the development of sophisticated cognition and behavior. Moreover, scientists will be able to better grasp how malfunction in one tiny brain area might affect the operation of its broader neuronal circuit as a result of these discoveries.
According to Dirk Hockemeyer, a member of UC Berkeley’s BRAIN Initiative efforts, the paper’s conclusion states that these diverse approaches for classifying cell kinds have a lot of overlap and regularity.
Based on identified variations in expression and epigenetic patterns among these cells, a team of statisticians integrated data from all of these experimental approaches to discover how best to categorize or cluster cells into various kinds and, presumably, different activities. While there are many statistical algorithms for analyzing such data and identifying clusters, the challenge, according to Sandrine Dudoit, a UC Berkeley professor and chair of the Department of Statistics who was a major part of the statistical team and co-author of the flagship publication, was determining which clusters were truly different from one another — truly different cell types.
It was emphasized that the goal of this initiative is not to invent another clustering approach, but to figure out how to combine the strengths of multiple methods and analyze the stability of the finding, as well as the repeatability of the clusters being obtained. Essentially, the number of cell types found in the motor cortex may be overstated, but the present findings are a promising start in building a brain cell atlas.
