Painting Neural Circuitry With a Viral Brush: Part 1

We often describe neural connections in the brain as if they were a cellular version of Michelangelo’s famous painting “Hand of God Giving Life to Adam,” on the ceiling of the Sistine Chapel. Like the painting, the neural explanations usually invoke 2 outstretched limbs nearly touching each other-one presynaptic, one postsynaptic-that are separated by their 20-nanometer synaptic cleft.

Such descriptions hardly depict the neurological reality of the brain, of course. A better metaphor for 2 neurons might be 2 trees that have been uprooted and turned 90 degrees so their root systems face each other. Then, some Paul Bunyan–type character jams both ends together. Thousands of connections from 1 tree now face thousands of connections from the other. Multiply those 2 neurons by thousands while all their root systems are still jammed together with the same Bunyanesque enthusiasm and you can visualize an approximation of the real world of brain wiring. Not as elegant as 2 limbs-and not as simple either.

How can we understand the way all those myriad connections work together to produce the various neurological abilities of the brain? Given its complexity, the task is enormous. Since we are only in the initial stages, it will take a long time before we will be able to map structure to function.

Like any exploration in its initial stages, we first need a good map-a schematic that shows how each slender dendritic branch interacts with a specific nerve cell. From Nissl and Weigert stains to the canonical Golgi stain, we have traditionally used dye technology to help us visualize these interactions. However, there are severe limitations to most of these staining technologies that center around their inability to make visible all the connections that actual neurons possess. We need something with far greater resolution and, perhaps, with a bit more elegance.

The topic of this column and the next is a technology that promises to deliver just such circuit diagrams at a very high level of resolution. The technology involves the exploration of viruses, which in their native form cause some pretty tough diseases (eg, rabies, cancer). Thoughtful genetic engineering has transformed these viruses’ job description from fearful disease inducer to doughty cartographer. We are going to follow how this transformation has occurred.

I give you fair warning that it is a complex manipulation. I encourage you to closely follow the Figure that accompanies this column as well as the figure in Part 2. Both diagrams describe the story of hypothetical “Neuron A” and our attempts to understand all its many intricacies and connections. To explain Neuron A’s interactions, I need to review not only 2 pathogenic viruses but also several manipulated proteins, genetically engineered neural cells, and protein-based fluorescent dyes. In Part 1, I will discuss some background, which I have divided into 3 parts. In Part 2, I will explain the data.

The actual experiment that took place asks and answers the simple question: Are the neighbors green yet?

Why rabies virus?

The rabies virus has 2 characteristics that make it an ideal “truck” for the delivery of molecular products to mammalian neural interiors. The first has to do with its legendary life cycle. Once inside a nerve cell, the virus sets up a manufacturing site to create more viruses, like any typical virus. At maturity, these progeny can jump to other neurons, spreading the virus along specific neural routes. If you could find a way to follow the virus, you could identify the routes.

The second characteristic has to do with access to the CNS, which is a topic of great concern to researchers who are interested in drug delivery, for example. The rabies virus is not a respecter of neural borders. It can infect neurons in the peripheral nervous system and then jump the neurological border to enter the CNS. That is why a bite anywhere on the body can result in a catastrophic brain infection. By investigating how this transfer occurs at the molecular level, biologists can exploit the viral life cycle to understand neural interactions.

I will now discuss how a modified rabies virus could confront our Neuron A described previously. We will “genetically engineer” a rabies virus and, at the same time, genetically engineer its neural target to make this happen.

The gene we are going to engineer is one of the proteins that the rabies virus needs to set up a productive infection. It is a glycoprotein, which is carried in gene form by the virus that infects Neuron A. If the virus can manufacture this glycoprotein after it enters Neuron A, it can make progeny and spread the infection to an adjacent nerve cell. If the virus cannot make the glycoprotein, it will not be able to jump to an adjacent nerve cell, even if Neuron A is otherwise infected. The virus will simply stay inside its host.

Today, researchers can be quite clever about forcing the hand of the rabies virus. For example, they can gut the gene that encodes the glycoprotein from the virus, rendering it impotent to spread infection. The virus can get into Neuron A but it cannot get out.

While we are working on the rabies virus, we are also going to genetically modify Neuron A. We will engineer the nerve cell to make the glycoprotein by extracting its gene from the virus and inserting it into the nucleus of Neuron A. This process is called transfection, and there are many ways to perform the procedure. Soon, Neuron A will be flooded with the glycoprotein.

Why do we go through these steps? This double manipulation allows the virus to set up a productive infection in just 1 cell. If the hobbled rabies virus is allowed to infect the modified Neuron A, the cell will allow the virus to make new progeny, which can then escape Neuron A and infect all its neighbors. But that is where everything stops. The neighbors will be able to create a new virus just like Neuron A; however, these progeny are still reproductively incompetent (remember, the viral genetic information is gutted from the glycoprotein). Neuron A will serve as the birth mother, and the virus will spread to neighboring nerve cells. If the neighbors do not have the viral glycoprotein already working in the background, the virus will not be able to spread outward from them. The buck, as it were, stops there.

This finding could be of great value to a neurobiologist, especially if Neuron A is embedded in a thicket of normal neurons that are not genetically engineered. It is the neurobiologist’s job to find out how they are connected. This concept will become very important as I further discuss this technology.

ASLV-A

The second virus in our discussion goes by the tongue-twisting name avian sarcoma and leukosis virus subgroup A (ASLV-A). The virus attacks via a surface grappling hook-a protein called EnvA. If EnvA finds a receptor protein called TVA, ASLV-A will set up an infection. This will occur even if the TVA protein is found on a cell other than an avian cell.

Why am I bringing this up? It is possible to mix and match viral proteins and to alter their host ranges. For example, a researcher could remove an EnvA gene from ASLV-A and stick it to the genetic background of a rabies virus. The rabies virus now has a grappling hook for avian cells. This means that the rabies virus is now fully capable of binding to an avian cell or any cell that carries the TVA receptor protein. The virus could be directed to go anywhere the researcher wanted it to, as long as the EnvA-TVA matching conditions were met. This, too, turns out to be a critical part of this story.

Two fluorescent proteins

The fluorescent proteins are probably the least complex piece of molecular biology in our discussion. There are a number of proteins that glow in the dark if given the proper substrates at different wavelengths when inserted into cells. There are genes whose proteins glow red and genes whose proteins glow green (green fluorescent protein, GFP). These genes can be stitched into the genetic machinery of viruses, which are then allowed to infect cells. Researchers can trace not only the success of the infection but also the routes of infection just by looking for the presence of the fluorescent color in the cell. For example, if the GFP gene is inserted into the rabies genetic information, every cell in which the rabies virus sets up an infection will turn green. As you might suspect, this phenomenon provided researchers with everything they needed to trace neural interactions.

With this information in mind, we are prepped to discuss the data, which will occur in Part 2. We will see that the clever use of these viruses, genes, and manipulated cellular backgrounds has allowed researchers to create their own Lilliputian Sistine masterpiece.

They did it all by answering that simple question, “Are the neighbors green yet?”