ScienceBeta

Menu

How Transcriptomics Is Changing How We Think About Brain Cells

human cortical spheroids astrocyte

Neurons, the fundamental component of the nervous system, exist in a multitude of forms and types. There may be as many as 10,000 specific types of neurons in the human brain.

There is even a type of cell discovered in 2005 known as the “Jennifer Aniston neuron” that specifically responds to images of well-known celebrities like JA or Barack Obama or to famous landmarks like the Golden Gate bridge. These neurons even fire if you just think about these images.

Up until recently, neuroscientists have sorted all these brain cells into different types mostly based on their function and cellular anatomy.

Now, as part of a broader effort to more deeply understand the brain biology of mental illnesses like anxiety and depression, some are turning to cutting-edge genetic sequencing technologies like transcriptomics.

transcriptomics
© 2017 PHG Foundation

The human cerebral cortex contains is comprised of around 16 billion neurons and 61 billion non-neuronal cells.

A mouse’s cerebral cortex is around 1000 times smaller in area and number of neurons. So it might not exactly seem like the best model to use for researching human brains.

But the genetic, biological and behavior characteristics of rodents do closely resemble those of humans. Many symptoms of human conditions can be replicated in mice and rats.

Those similarities have become even stronger over the past 2 decades, as scientists can now breed genetically-altered mice called “transgenic mice” that carry genes that are similar to those that cause human diseases.

Science’s current understanding of how the brain regulates our mood has come from experiments using human volunteers and patients, added to experiments performed in laboratory animals, mostly rats and mice. Rodent models of depression have been key in testing theories about the cause and treatment of depression, and have made possible the development of newer and more effective treatments like SSRI antidepressants.

Even with these modern improved medications,

    • Approximately one fifth of the population will suffer at least one episode of clinical depression resulting in a significant social and economic burden on society (Doris A, Ebmeier K, Shajahan P 1999).

Neuroscientists are working hard to find out why.

Human DNA

The next time somebody tells you that humans and chimpanzees share 96% of our DNA, try this come back:

Our brains express a much larger stock of genes than any other species. One study demonstrated over 2000 genes that are expressed differentially in the human brain compared to chimpanzee brains.

Not only that, but these genes are spread out across cellular processes ranging from synaptic transmission and signal transduction to neuronal differentiation and development. This is a strong suggestion that the cells have some fairly global differences in how they function.

One set of results I found particularly intriguing while researching this series were those from a 2017 Allen Institute of Brain Science study investigating gene expression differences between mouse and human brain cells. They found that gene expression patterns associated with various diseases in humans such as Alzheimer’s are not seen in the mouse.

The implication is that this is one reason why drugs that work for mouse brains often don’t work in humans. For example, once promising Alzheimer’s treatment candidates from Biogen, Merck & Co.Eli Lilly and Company, and others have all seen clinical trials halted and development shelved from lack of efficacy.

A more recent study from the Allen Institute for Brain Science utilised a transcriptomics analysis technique known as Single Nucleus RNA Sequencing (snRNA-seq) to revisit this comparison, and found some more important differences.

(Transcriptomics studies an organism’s transcriptome, the sum of all of its RNA transcripts. The information content of an organism is recorded in the DNA of its genome and expressed through transcription. Here, mRNAserves as a transient intermediary molecule in the information network, whilst non-coding RNAs perform additional diverse functions.)

In this latest study:

    • 15,298 brain cell nuclei were sequenced and analysed
    • 75 transcriptomically distinct cell types were identified
    • Comparision of human to mouse cell types showed a mix of similar and divergent gene expression in correlated cell types
    • Non-neuronal cell types had the most divergent expression
    • Serotonin receptors were found to have highly divergent gene expression between mouse and human cells

“They’re expressed in both mouse and human, but they’re not in the same types of cells,”

said study co-author Ed Lein, speaking to NPR.

“serotonin would have a very different function when released into the cortex of the two species.”

Other takeaways from this study?

First of all, it puts the debate over how different human cortexes are from other mammals into a quant framework.

In addition, it helps resolve the mystery of why mouse model results of pre-clinical trials are not translating better into human results for neurological drugs.

It shows the effectiveness of Single Nuclei RNA Sequencing[1] for accelerating our understanding of brain evolution and disease.

And it highlights the need to analyse the human brain in addition to those of current model organisms.

The biggest emerging story here, is that it turns out the biggest differences are not in neurons, the cells involved in bioelectrical/chemical signaling, but in glia.

Glial Cells

More than half of our brains are made up of glial cells, which wrap around nerve fibers and insulate them. In a manner similar to how the plastic casing of an electric cable insulates the copper wire within, these cells allow electrical and chemical impulses to travel faster.

These cells had long been regarded as passive support cells, but in fact are vital to nerve-cell development in the brain, a team of biologists revealed in 2017.

Sidenote: when some neuroscientists examined preserved samples of Albert Einstein’s cerebral cortex in 1985 they found that his neuronal: glial ratio was significantly smaller than the average; in other words, he had more glia than other people.

Astrocytes and oligodendrocytes
Astrocytes and oligodendrocytes from neural stem cells.
Credit: Yirui Sun. CC BY

A 2009 study found human astrocytes (a kind of glial cell) to be 2.6 times larger in diameter, reach 10 times more processes in humans and come in many more complex sub types than in mice.

Stanford University’s Ben Barres reported in a 2017 paper that human astrocytes respond far more readily to glutamate (an essential nutrient for the brain ) relative to mice.  So, glia may play a more prominent role in shaping the electrical flow of the human brain.

All of which might lead you to guess that the area that gene expression contrasts most between mouse and humans is in glia rather than neurons.

You would be right.

A large-scale Allen Institute study from 2011 revealed larger mouse-human differences in non-neuronal gene expression than in neurons.

In 2013, Pavladis, et al. found mouse neuron markers that showed gene expression patterns more correlated with the human oligodendrocyte (another type of glia cell) pattern and vice-versa. They were swapped, in other words!

These findings come at the same time there is growing evidence of the involvement of glia in the neuropathology of mood disorders.

For example, a 2016 study[2] found that people with major depressive disorder may have fewer astrocytes in some parts of the brain’s hippocampus.

“People thought depression had to do with neurons. No one thought about glia,. The decrease in glial density is much more dramatic than the changes in neurons,”

said co-author Dr. Grazyna Rajkowska of the University of Mississippi.

Beyond depression, a list of disorders involving glial cells would be a long one and include Amyotrophic lateral sclerosis, Alzheimer´s Disease, Attention deficit hyperactivity disorder, Bipolar Disorder, obsessive-compulsive disorder, Schizophrenia, and various substance use disorders.

Fortunately, the development of this next generation sequencing technology has made it possible to measure the molecular activity of specific cells as a routine part of research. Humans and other mammalian species are very different in several advanced behaviors, such as language, cognition and sleep. How to explain the differences of these behaviors at the molecular level remains a mystery.

[1] What is so great about snRNA-seq? The bigggie is it can be used with frozen samples, which for research on human brains, is huge, for obvious reasons (like most people don’t get too excited about donating live sample of their brain). There are other advantages too. https://www.ncbi.nlm.nih.gov/pubmed/30510133

[2] J.A.Cobb, et al. Density of GFAP-immunoreactive astrocytes is decreased in left hippocampi in major depressive disorder. Neuroscience
Volume 316, 1 March 2016, Pages 209-220