The brain is the most complex organ in the body, and presents countless mysteries – including the very nature of all the cells contribute to its function. By using single-cell RNA sequencing to map the genetic activity in more than 3,000 mouse brain cells, one at a time, scientists recently reported that they were able to identify 7 hitherto unknown brain cell types. The findings mark the first time this method has been used on a large scale on such complex tissue, investigators say, thus demonstrating the power of using RNA patterns to identify new, discrete classes of neurons.
Brain function relies on an extraordinary variety of specialized cell types, and much remains uncertain about the workings of the brain because of this complexity. Knowing more about how each cell type works is critical to understanding the inner-workings of the brain as a whole.
Working with mouse neurons, neuroscientists at the Karolinska Institute in Stockholm, Sweden, and their colleagues employed a technique known as single-cell RNA sequencing, which involves identifying cell’s transcriptome, or all its RNA, in order to reveal what genes are active in those cells. Single-cell RNA sequencing has been used to classify cells in the mouse spleen, mouse lung and human embryonic brain cells. However, the adult nervous system has greater complexity and more cell types, the researchers say.
They carried out single-cell RNA sequencing on more than 3,000 cells, comparing which of 20,000 genes were active in each one. These mouse cells came from the somatosensory cortex, which is linked to the sense of touch, and the hippocampus, which plays a key role in memory.
“It took us five years to develop the tools to be able to do it, but once we had them it took 18 months from start to finish,” said study co-senior author Sten Linnarsson, a neuroscientist at the Karolinska Institute.
The researchers identified nine major classes of brain cells divided into 47 molecularly distinct subclasses of cells. For instance, neurons from the somatosensory cortex known as S1 pyramidal cells were marked by a transcription factor known as Tbr1, required for their final differentiation, and by Gm11549, a long noncoding RNA; in another case, hippocampal pyramidal cells were marked by Spink8, a serine protease inhibitor.
Their analysis yielded 7 cell types that hadn’t yet been identified — a neuron type in the outermost layer of the cortex, and six different types of oligodendrocyte, which are cells that produce the electrically insulating myelin sheath around neurons. These six types of oligodendrocyte likely represent different stages of maturation, they added. The scientists detailed their work in the March 6 issue of the journal Science.
These findings helped the scientists identify specific marker genes for each of the nine major classes of brain cells. This discovery could help researchers develop novel ways to target specific classes of brain cell, which could contribute to the development of cell-type-specific gene therapy, says study co-senior author Jens Hjerling-Leffler, a neuroscientist at the Karolinska Institute. For instance, a method to target oligodendrocytes could help result in new ways to fight diseases that affect myelin, such as multiple sclerosis.
“These studies are both technically and biologically important for understanding the brain,” says neuroscientist Jerold Chun at the Scripps Research Institute in La Jolla, Calif., who did not take part in this research. “Technically, they provide some of the first comprehensive single-neuron transcriptome data. Biologically, the data underscore the extensive diversity of individual neurons from even a single brain region at the level of whole transcriptome.”
Single-cell analysis is getting routine, and promises to shed even more light on complex biology in the future, Linnarsson said. “In a few years, there will be datasets with 10 or 100 times more cells,” Linnarsson says. Ultimately, he adds, he’d like to “see entire organisms sampled at high resolution — millions of cells.”
As for performing a similar analysis on human brain tissue, Hjerling-Leffler says his group is open to this possibility, but that “access to starting material is limited.”