Mapping the brain cell by cell

October 23, 2017

Harvard and collaborators receive $64.7 million NIH grant to build a detailed map of brain cells in mice

The human brain contains many billions of cells, and scientists are still working to assemble a comprehensive catalog of all of the different types. Neuroscientists have dreamt for years of understanding the role that each one of these cells plays, and how the cells differ, connect and function. But the number of cells has been simply too massive to study or map all of them on an individual level.

Now, thanks to advances in technology, the neuroscientists’ dream is closer to being realized than ever before. In a project of unprecedented scale, the National Institutes of Health has awarded an international consortium of laboratories $64.7 million over the next five years to begin to build a detailed mouse brain atlas that will catalog and map the cells in a mouse brain. The project aims to identify all of the different cell types in the brain, determine what set of genes each type uses, and map their physical locations. In addition, the consortium will examine the forebrain, the part of the brain involved in most forms of cognition, emotion, and sensorimotor processing, in more detail, aiming to also determine the size and shape of each of the major types of neuron, and trace the connections they make to other regions of the brain. Though a mouse brain contains far fewer cells than the human brain, its complexity still makes cataloging its cellular constituents a huge undertaking. Part of the NIH Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative, this project complements efforts by other BRAIN Initiative consortia to map the brains of humans and non-human primates.

The mouse brain cell atlas project is led by Paola Arlotta, Harvard Department of Stem Cell and Regenerative Biology and the Eli & Edythe Broad Institute of MIT and Harvard, and Josh Huang, Cold Spring Harbor Laboratory,  and it brings together co-investigators Catherine Dulac (Harvard, Molecular and Cellular Biology), Xiaowei Zhuang (Harvard, Chemistry and Chemical Biology), Aviv Regev and Edward Boyden (the Broad Institute and MIT), Evan Macosko (the Broad Institute) and many distinguished investigators at Cold Spring Harbor Laboratory, University of Southern California, UC Berkley, John Hopkins, and Huazhong University of Science and Technology in China. “This is a highly collaborative project, with the institutions involved opening up their methods and data to each other in order to achieve this immense task that none could complete alone”, Arlotta says.

A comprehensive catalog of the different types of brain cells, mapped to their physical locations in the brain, will ultimately improve researchers’ understanding of how the brain develops, connects and works—and what can go wrong with it in disease.

In order to solve medical problems, you really need this foundational knowledge of what cell types there are, where in the brain they are, and where they connect to”, Arlotta says. “And much of this information has been missing in the field, because we were not able, technology-wise, to gather it across the whole brain.

To identify the different cell types and determine their “gene expression profile”, the set of genes the cell uses, the consortium will examine each cell’s RNA, the DNA-like strands that follow genetic instructions to build the cell parts that allow a cell to perform its functions. “Researchers can use these profiles to figure out “who” each cell is”, Arlotta says.

Sequencing RNA takes time, so to profile the many millions of cells in the mouse brain, the labs need to sequence many cells at once. They also need to keep the cells isolated to get their individual profiles. They will do this using a technique that relies on separating cells—or the cells’ nuclei—into tiny droplets. The separated cells are barcoded, and then analyzed in large batches, allowing the researchers to greatly speed up the cell profiling process without losing track of individual-cell-level detail.

After the labs catalog each brain cell’s gene expression profile, they will match these profiles to a “brain atlas”, a 3D virtual map of the brain, connecting each cell’s molecular level data to a physical location.

Location is important in neuroscience,” says Arlotta. “Knowing what kinds of neighbors a cell has can tell us a lot about what the cell is involved in.

To help map the identified cell types to their locations, the member labs will use a technique recently developed in the Zhuang lab at Harvard, MERFISH (multiplexed error-robust fluorescence in situ hybridization), which can detect hundreds of genes at a time in slices of intact tissue, without needing to separate the cells from each other, and so identify cells in place using their molecular-level data

A wide variety of research projects will be facilitated by the atlas, from basic neuroscience to disease modeling. “One day people will open a window, pull up the atlas, zoom in on brain region of interest, and find all the cell data they need to make their inquiries”, Arlotta says.

The grant reported in this publication was awarded by the National Institute Of Mental Health of the National Institutes of Health under Award Number U19MH114821. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Written by Greta Friar