Q&A: High-throughput brain mapping – a barcode for every synapse

Image
Photo credit: Avery Krieger

By Gordy Slack | Part of our Next-Generation Neuroscience series

Wu Tsai Neurosciences Institute Interdisciplinary Scholar Boxuan Zhao grew up in small-town China, far from the glow of research institutes or universities, but he inherited a love of the natural world and exploration from his parents, both of whom were science teachers. On weekends he loved to peer at botanical specimens in the microscope at the school where his father taught.

Always drawn to complexity, biology was a natural choice for Zhao, especially given his fascination with living things and the systems that underlie them. And brains, possibly the ultimate complex systems, were an early preoccupation, too. But he ended up studying chemistry as an undergraduate at Peking University, where he learned, for example, to apply protein engineering to make biosensors for small molecules in living cells.

In grad school, at the University of Chicago, he turned his attention to DNA and RNA epigenetics and their regulatory roles in biological systems. The field was growing rapidly and full of excitement, but perhaps his favorite part was the amazing array of different biologists he got to work with. Zhao realized that chemistry could open doors to whole new areas of the living world.

The Wu Tsai Neurosciences Institute Interdisciplinary Scholar Awards provide funding to extraordinary postdoctoral scientists at Stanford University engaging in highly interdisciplinary research in the neurosciences.

Learn more >>

It wasn’t until Zhao arrived at Stanford for his postdoc that he was able to study the human brain full time under the dual mentorships of chemical biologist Alice Ting and neuroscientist Liqun Luo. The new team was interested in finding new ways to create high-resolution neural wiring diagrams known as a brain connectome. Putting Zhao’s Chicago experience with DNA and RNA dynamics to work, the researchers engineered an RNA barcoding tool and harnessed powerful high-throughput sequencing technology to uniquely identify which neurons were communicating with each other at every synapse in a given brain region.

The Neuro-Omics initiative is dedicated to creating tools to bridge the chasm between neuroscientists’ understanding of the brain’s genes and proteins on the one hand, and its circuits and behavioral systems on the other.

Learn more>> 

Their invention, dubbed connectome-seq, is at the heart of the Neuro-Omics Initiative, a Wu Tsai Neuro Big Ideas in Neuroscience project led by Ting, Luo, and Stephen Quake of Stanford and the CZBiohub. The team is currently writing up the first fruit of this work: a remarkably detailed connectome map showing how a pair of mouse brain regions are connected by different kinds of neurons. Next up, the team plans to apply connectome-seq to map more pairs of brain regions, ultimately piecing together a complete reference map of the mouse brain, enabling new questions about how this map changes with learning in the healthy brain as well as in neuropathological conditions, such as Alzheimer’s disease.

 

We spoke with Zhao to learn more about his time as a Wu Tsai Neuro interdisciplinary scholar, his current research, and his plans for the future:

You were originally trained as a chemical biologist. What drives you to do neuroscience, and why connectome studies in particular?

When I was a boy, I wondered a lot about sleep. To someone who loved to explore, it seemed like such a waste of time! So, the human brain was my very first fascination in science. Of course, now I know sleep is crucial for memory consolidation and other essential things, but we still haven’t really figured out how and why we sleep as much as we do.

For example, let’s say during the day you learn say 100 things, but only 10 of them matter. So, during sleep cycles you consolidate and strengthen the new brain connections related to those 10 things and the other 90 are pruned away. It would be great to know what drives that intricate process.  But to grasp what’s going on you'd first need to have a detailed blueprint of the brain, to know which parts are connected to which other parts and how they communicate during different kinds of behavior. Only then could you track how those brain-wide circuits change when you learn and, conversely, when you forget, which involves pruning specific connections in your sleep. Without a blueprint, it’s also hard to know what’s going on during ageing, neurodegenerative diseases, or genetic diseases, and it’s hard to know how best to treat them. You need the connectome, the basic neuronal blueprint. Our tool, connectome-seq, will help us figure out who’s connected to whom, neuron-wise.

Revealing the blueprint has been hard because of its physical structure and because of the scale. There are more than 80 billion neurons and each one has on average a thousand synapses. Until now, the only tractable way to figure out who is synaptically connected to whom is to take a small piece of brain tissue and slice it into super thin pieces, examine them individually under an electron microscope, and then reconstruct synapses back into three dimensions.  Mapping one cubic centimeter of brain tissue this way would take more than 1,000 person-years. It’s not feasible at scale.

We approach the problem from a completely different direction. We’re using high-throughput sequencing technology to speed the mapping process up by many orders of magnitude.

Zhao studies fluorescent images of the mouse brain on a large monitor connected to a high-resolution microscope
Zhao studies fluorescent images of the mouse brain captured by a high-resolution confocal microscope, highlighting connections between cortex and striatum barcoded using the connectome-seq technique. 
Photo credit: Avery Krieger
 

How does connectome-seq work?

How do we translate something physiological like a synapse—which is just proteins and lipids stuck together and buried in the brain—into something that can be identified by a machine?

We start by introducing a bioengineered complex made of a pair of proteins and a pair of RNA molecules into parts of the brain. The proteins anchor the complex to the pre- and post-synaptic sides of the synaptic connections between pairs of neurons and mark each side with a unique RNA barcode. That “barcoded” synapse now forms a stable complex that can later--when the brain section is ground up--be identified by a flow cytometer, revealing the identities of the neurons on both sides of every synaptic connection. That synapse-barcoding complex, called a SynBar, is the key component of our whole technology. It lets us tell which neurons share synapses with other neurons, and eventually will tell us which kinds of neurons and which parts of the brain are connecting during specific behaviors.

The next iteration of this technology will be combined with in-situ sequencing using technologies such as Karl Deisseroth’s STARmap technique. The idea would be to use a specialized probe that can read the barcodes from individual neurons while they’re still in place, eventually allowing us to identify the region where the presynaptic and postsynaptic neurons are located without having to grind up a brain. This process is much slower, but it can be powerful when combined with the high-throughput technology.

It seems like one theme of your work is fitting disparate things together in new ways.  Is connectome-seq an example of that?

Zhao works at a microscope with his mentors Liqun Luo and Alice Ting
Zhao works at a microscope with his mentors Liqun Luo and Alice Ting in Ting's lab at Stanford's James H. Clark Center.
Photo credit: Avery Krieger

Yes. And Wu Tsai Neuro’s role in this is crucial because it supports a dual mentorship type of training experience.  In my Stanford postdoc I’m being trained by two amazing scientists from different fields: Alice Ting is a renowned chemical biologist in the department of genetics. She has invented a lot of different tools--especially in the proteomics field--for biologist to examine things that were just not accessible without them. Liqun Luo is a distinguished neuroscientist who made numerous key discoveries in neural wiring and created new methods to study them. He asks the deep neuro questions and Alice figures out how to make the tools to answer them. Their complementary skill sets and mindsets make them a great team perfectly suited for connectome-seq.  The three of us connecting is how the idea for connectome-seq was born. Their insight, their mentorship, and their support has been crucial.

Let’s go even further back to before you arrived at the University of Chicago.

I grew up in China where both of my parents were biology teachers; they were responsible for my love of the natural and scientific worlds.  I first hoped to study biology, but at that time in China biology was by far the most popular and competitive major, so I ended up in chemistry instead. Initially, that was a huge bummer. Until I met my college mentor, chemical biologist Peng Chen, I’d had no idea that chemists could address intriguing biological questions or come up with new tools that biologists can use. Once I realized that, chembio was all I wanted to do! It requires constant innovation and I’m really steeped in that radically innovative mindset. It set my career trajectory. So, I’m very grateful to Peng. He showed me the way.

I went to the University of Chicago because that’s where Peng’s own mentor, Chuan He, was a professor. When I got to Chuan’s lab, he had just discovered the first reversible RNA modification, m6A, which signified a new layer of control over gene expression and key functions in various biological processes. Suddenly there were endless things to do, we were all super excited, and everything we discovered was basically a first. So, we had all kinds of collaborations; everyone wanted to explore m6A function in their own biological system. So, I was able to work with biologists from completely different backgrounds: developmental biologists studying zebra fish, cancer biologists studying glioblastoma, and virologists working with influenza, zika virus, and HIV. If I’d been a biologist, I never would have been able to dip my toe in all these different fields.

Speaking of toe dipping, I understand you’re also a triathlete and a SCUBA diver. Is there a relationship between those activities and the way you approach your research?

I think so. In college back in China I was a competitive swimmer. Yet when I went to Chicago, where swimming was a serious varsity sport, I couldn’t join the team as a busy graduate student. But the triathlon team was new and much more casual, so I joined it instead and fell in love with this “interdisciplinary sport.” I liked that you had to find ways to improve all three disciplines with each exercise. I was trying to do something similar in chembio; I was stitching together different skill sets, but each project advanced and deepened my understanding of the others.

Again, like chemical biology, Scuba diving allows you to see things that would be invisible otherwise. I went diving in the Philippines in waters that are some of the highest-density-biodiversity places on earth. Amazing, and again, not unlike my scientific work:  Mastering new technologies lets you go deeper underwater and see whole different worlds!

 

Zhao recieved a Wu Tsai Neurosciences Institute Interdisciplinary Postdoctoral Scholar award in 2017. He is now LSRF Postdoctoral Research Fellow in the Department of Genetics. Luo is Ann and Bill Swindells Professor in the Department of Biology and a member of Stanford Bio-X, the Stanford Cancer Institute, Sarafan ChEM-H, and the Wu Tsai Neurosciences Institute. Ting is professor in the departments of biology and genetics and a member of Stanford Bio-X, the Maternal & Child Health Research Institute (MCHRI), the Stanford Cancer Institute, Sarafan ChEM-H, and the Wu Tsai Neurosciences Institute.