Astrocytes are like exes (they need their space): discovering astrocytic domains

Can you imagine being the first person who ever looked through a telescope? Everyone is familiar with the night sky in their home town; you can look up and see a few twinkling lights, a favorite constellation or two. If you live in a dark, rural area, you may even be able to see the creamy band of the Milky Way meandering across the sky. You’re probably also familiar with famous images of galaxies and nebulas, the billions of other stars existing in our universe, even if we can’t see them with our naked eye. What must it have felt like, looking through a telescope that first time, and seeing stars that no other human being had ever laid eyes on? To suddenly realize that the night sky is a vaster, deeper, and more complex space than you ever could have imagined?  This analogy seems fitting to me when I consider how Dr. Eric Bushong and his colleagues must have felt when they first looked closely at the astrocyte, and understood how misunderstood this starry-shaped cell had been.

At the time, scientists knew that glial cells were important for neuronal growth and survival. They were interested in knowing how astrocytes were spaced throughout the brain, to better understand how they might be supporting and influencing neurons. Many advances were being made in microscopy, letting us image cells more clearly and closely than ever before. In particular, scientists were figuring out new ways to look at individual cells “in situ” – that is, looking at an individual cell within whole tissue, rather than plating the cells out in a single layer in a dish. Cells grown in a dish don’t always take the same shape or develop the same way as they would in a whole brain.

Up until this point, scientists mostly used a protein called glial fibrillary acidic protein, or GFAP, to mark astrocytes. This is done through a processed called immunohistochemical staining. A slice of brain can be fixed, or preserved, using paraformaldehyde. The slice is then submerged in a blocking solution containing detergents, which punch tiny holes in the cells. Next, a solution containing antibodies (just like the ones your body uses to fight off intruders!) is added. Antibodies fight diseases by targeting specific proteins that mark pathogens, binding tightly to the intruding cell and marking it for destruction by body’s immune system. Researchers have taken advantage of this system by creating antibodies with fluorescent markers attached to them, so when the antibody binds to its target, it will show up under a certain wavelength of light. The antibodies can get into the cell through the holes created by the detergent, where they bind their target. GFAP is a protein that is mostly expressed in astrocytes, so the GFAP antibody will only bind to the GFAP protein it finds inside those astrocytes – making it a (fairly) effective marker.

Studies using this marker seemed to indicate that astrocytes overlapped a great deal, which led scientists to believe that this intermingling was important for the structure of the brain and helping astrocytes space themselves properly during development. It was becoming increasingly apparent, however, that not all astrocytes were exactly the same – just like your brain is full of different kinds of neurons, so too is it full of many different kinds of astrocytes (even today, we’re still working on this problem of astrocyte heterogeneity!) Not all astrocytes showed the same pattern of staining with GFAP, and scientists knew that GFAP staining might not actually be labeling the entire astrocyte.

Here, the scientists decided to use a combination of staining techniques to examine astrocytes in more depth, hoping to better understand the limits of GFAP labeling and to get a better idea of what an astrocyte actually looks like in the brain. By injecting individual astrocytes directly with a fluorescent dye (rather than relying on antibodies binding to proteins), they were able to completely fill each cell out to the very tips of its processes. Then, they used a GFAP antibody with a different fluorescent signature to stain the same tissue, so the dye-filled astrocytes would show up as red while the GFAP-stained astrocytes would show up as green. What they ended up with was something like this:

Figure 1: GFAP-labeled (green) and dye-filled astrocytes (red), with yellow showing the overlap of labeling between the two methods. In C, we see a 3D projection through the tissue, demonstrating the density of the astrocyte’s processes and the defined boundaries of each cell.

Rather than the thin, star-like shapes shown by GFAP, the researchers found that the dye-filled astrocytes were actually much larger and very bushy, with many tiny processes, with an irregular shape but defined edges. All of the cells filled with dye were also marked by GFAP labeling, but as you can see, the GFAP only labeled a very small portion of the cell, around 13% of the total cell volume – the astrocytes were much bigger than the GFAP would have led you to believe!

By using two different colors of dye to inject the astrocytes, the scientists were also able to look at the overlap and interactions between adjacent cells. Unlike what earlier studies seemed to show, they actually found that the astrocytes don’t really overlap – the very fine processes at the edges intermingle, but the larger ones seemed to avoid crossing into the territory of another astrocyte. When the scientists looked at where red and green signals mingled, they found a distinct, thin line where the astrocytes met, but nothing beyond it (Fig. 2).

Figure 2: The green and the red astrocyte only overlap where the thin, yellow line is visible.

The overlap line is so narrow, and the processes are so thin, that normal light microscopy isn’t enough to examine the nature of the interactions between the two astrocytes, so the scientists decided to use a technique called electron microscopy to look even closer. Electron microscopes use beams of electrons instead of photons to create images, and allow for much greater magnification of the specimen than a light microscope. The same dye used to fill the astrocytes for light microscopy can be used to image them during electron microscopy through a process called oxidation, a reaction that converts the dye so it will appear with the electron microscope. At this high magnification, the scientists confirmed what the light microscopy had shown – that the astrocytes don’t overlap very much, only intermingling among their fine processes.

This study marked the first time that scientists were able to really visualize the astrocyte in its entirety, getting a good idea of the shape of the cell and looking closely at its interactions with its next-door neighbors. The discovery of the astrocytic “domain” – that is, the fact that each astrocyte has its own territory, without overlap from other astrocytes – led us to reconsider what we thought we knew about the astrocyte’s role in the brain. Visualizing the dense, fine astrocytic processes also helped scientists realize that each astrocyte must influence thousands of individual synapses, opening the door to further questions about how astrocytes can dynamically interact with each synapse.

To me, this is part of what makes science so very exciting; the opportunity to see something that no one else has really seen before. Dr. Bushong and his colleagues would have had some guesses as to what they might seen when they looked closely at these astrocytes, given the previous literature and their own research – but I bet it was a pretty amazing moment, to look through the microscope at those bushy, dense cells for the first time. And just like the first person who looked at those far away, never-before-seen stars changed the course of astronomy, Bushong et. al. changed the way we think about astrocytes.

1) Bushong EA, Martone ME, Jones YZ, Ellisman MH. Protoplasmic astrocytes in CA1 stratum radiatum occupy separate anatomical domains. J Neurosci. 2002 Jan 1;22(1):183-92. 


Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s