World-First Human Brain Atlas Reveals New Cell Types

A research consortium has published a flurry of papers detailing a “major step forward” in our knowledge of the human brain. The project includes a draft genomic atlas of the brain that authors say could boost neuroscience much as the human genome project advanced genomics.

Big science vs. the brain

The history of neuroscience is littered with stories of researchers and entrepreneurs underestimating the brain’s complexity. The recently completed Human Brain Project originally set a goal of simulating the brain – with a target date of 2019. While the project helped advance technologies like brain implants and created digital maps of pockets of the brain, it never came close to achieving its original goal.

Now a similarly ambitious project to map the human brain at a genetic level has delivered on its objectives. The Brain Initiative Cell Census Network (BICCN), a subdivision of the NIH’s multi-billion-dollar BRAIN Initiative, has shown off its rich results in a glut of 21 papers published across three journals: Science, Science Advances and Science Translational Medicine. Some of the research’s highlights include:

• The identification of over 3000 cell types spread across the brain
• The discovery of a new type of brain cell – the splatter neuron
• Detailed maps of how our genes are regulated in brain cells – and how that regulation links to 19 different brain traits and diseases
• Features of the human brain that separate us from our nearest relatives – gorillas and chimpanzees

The papers widen the scope of what neuroscientists can study, say experts in the field.

Ed Lein, a senior investigator at the Allen Institute for Brain Science – which played a significant role in eight of the papers – is well aware of the challenge that faced his team at the project’s outset. “The brain is by far the most complex organ. By an order of magnitude, at least, more than other organs. It's really like 1000 organs. Each part of the brain is its own complex thing and looking at one part of the brain only gives you a very small answer about what the function and structure of the whole brain is,” said Lein in an interview with Technology Networks. The human brain’s structural details are well known to science, but with roughly 170 billion cells packed into a 3-pound lump, the variety and complexity within the brain has remained a mystery.

While every cell in our body carries the same genome, passed down from our parents, the information in our DNA represents a blueprint of what the cell could become. Converting these blueprints into the proteins and structures that make up our cells involves a two-stage process – transcription and translation. Together, these processes are called gene expression. BICCN explored transcription, a step in which the basic genomic blueprint is converted into RNA. This is a complex process – the base material of an origami crane is a sheet of paper, but what you do with that paper is what gives complexity to the final product. What Lein and the rest of the consortium, consisting of a global team from Seattle to Stockholm, wanted to explore was how different these transcription patterns were between cells in the brain.

A dive into the deep brain

The 21 studies in the block of papers can be roughly divided into five categories:

• Cell atlas studies mapping the adult human brain at the level of single cells using a technique called single-nucleus RNA sequencing, which isolates and reads RNA in cells’ genetic control centers
• Similar maps of the adult non-human primate (NHP) brain
• Comparative studies exploring the differences between humans and NHPs
• Brain development atlases, exploring how the human brain changes at the cellular level during development
• Functional studies that investigate how different brain cells behave

The papers feature a litany of major findings.

Viral genetic labeling of neocortical GABAergic neurons (green) in a human ex vivo brain slice. Additional cells are counterstained with DAPI in blue. Credit: Lee et al/Allen Institute for Brain Science.

In one study, samples were collected from 75 people with incurable epilepsy, who had part of their brains removed as a last-ditch solution to curb their seizures. Over 400,000 cells were analyzed in total. This study showed the importance of examining gene expression – while most people in the study had exactly the same cell types present in their brains, the abundance of each cell type and the gene expressed by those cells varied significantly between donors. The study concluded that factors like age, sex and disease state all feed into this variation.

Another paper detailed the discovery of a new type of neuron – the splatter neuron. One unique feature of the team’s analysis was its dive into the deep regions of the brain that sit below the cortex – the outer layer of the brain. The cortex is neatly organized into layers, but below, Lein said, the brain is a “mess of complexity”.

RNA sequencing separates cells into clusters depending on which genes have been expressed. Usually, this lines up well with the cell’s physical location in the brain. But splatter neurons break that rule – rather than separating into a discrete blob on a map of the brain, splatter neurons look like a “Rorschach test”, said Lein, and are found across multiple brain regions. Techniques like spatial transcriptomics, which links physical location to gene expression data, could be useful in future studies of these neuron populations, Lein explained.

A draft atlas of the brain

Lein struggled to highlight a particularly important paper from the package – “It’s really difficult to pick your favorite children!” – but mentioned the work that has produced the draft atlas of the human brain, spearheaded by Sten Linnarsson of the Karolinska Institute. This was a deeper analysis than that conducted on epilepsy patients and looked at just three post-mortem brain samples. It took in data from three million cells from every region of the brain – including those that are often overlooked in favor of the cortex. “For many years, we've really thought that complexity must be in the neocortex of the brain, which is responsible for most of our higher cognitive functions. It turns out that's not the case,” said Lein. “Actually, the greatest diversity is in subcortical regions of the brain, where much less focus has been put in.”

The team’s hope is that these findings will aid other neuroscientists in accelerating their own projects. Lein pointed to the Seattle Alzheimer’s Disease Brain Cell Atlas – which takes brain samples from donors who have died from this incurable dementia and uses a transcriptomic map to intimately detail what has happened to their brain at the genetic level. “We can now ask at this super-fine resolution what kinds of cells are lost in disease … it turns out that these are very specific kinds of cells. We never got to ask that question before because we didn’t have the resolution,” explained Lein.  

A wealth of information

Lein’s hopes for the project are mirrored by Sandra Jurado, a principal investigator at the Institute for Neuroscience UMH-CSIC in Alicante, Spain, who was not involved in the project.

Jurado studies how circuits in the brain change through processes like neuroplasticity. Her research has already benefited from the availability of mouse brain atlases. She told Technology Networks that the atlas was a “major step forward in our question to understand the intricacies of the human brain.”

“The emergence of a cell atlas of the human brain has the potential to significantly change the way neuroscientists work by providing them with a wealth of valuable information about the cellular and molecular composition of the brain,” said Jurado.

BICCN tried to avoid the overpromise of previous projects by focusing their efforts on developing technologies that could shine a light on the brain’s complexity. The work started with a predecessor project – the BRAIN Initiative Cell Census Consortium (BICCC) – which looked at the less complex mouse brain. After making sure the technologies that powered the project could walk the walk in the mouse brain, they were then put to work on the NHP and human brains explored in the BICCN. These technologies, said Lein, are benefiting from increased competition among suppliers and have “really hit primetime.”

First steps

The BICCN project only represents the first steps towards making use of this technology, said Lein. The successor project – the BRAIN Initiative Cell Atlas Network (BICAN) aims to systematize and expand the draft atlas. Professor Tara Spires-Jones, president of the British Neuroscience Association and deputy director of the Centre for Discovery Brain Sciences at the University of Edinburgh, who was not involved in the project, said that it was “fantastic to start seeing so much data come through characterizing the remarkable diversity of cells in the human brain.” But she pointed out that the brain maps created during the project were stitched together from a small number of people. “We still have a long way to go for a complete brain map,” she added.

Major projects in neuroscience have tried to provide definitive answers on the brain’s complexity and mystery. By diving into the labyrinth, rather than trying to find the exit, BICCN might have not provide definitive answers, but does give researchers a fighting chance to try and solve the brain’s big questions. 

Read more: https://www.science.org/collections/brain-cell-census