Janelia Farms Research Campus, Ashburn VA
On Site Research Dates:
March 30th 2014 - April 24th 2014
Title of Research Project:
"Astrocyte-like glial cells in Drosophila respond to neuronal activity"
I spent the month of April as a visiting scientist in Dr. Albert Cardona’s laboratory at the Howard Hughes Medical Institute’s Janelia Farm Research Campus in Ashburn, Virginia, examining the anatomical relationship between glial cells and synapses using a one-of-a-kind three-dimensional electron microscope dataset. Glial cells are the co-inhabitants of neurons in brain tissues. They comprise half of the human brain and are also found in smaller brains, including the brain of the organism I have studied throughout my dissertation, Drosophila melanogaster (fruit fly). Glial cells come in a variety of flavors, the cells of interest in this study are termed “astrocyte-like” due to their observed anatomical similarity to a subset of glial cells called astrocytes that are found in vertebrate species. My dissertation work is the first effort to compare the physiology and function of these cells to vertebrate astrocytes, with the long-term goal of using fly glia to discover novel glial genes and expand upon our nascent understanding of how glial cells actively contribute to brain function.
Synapses are the sites of neuron-neuron connection where one neuron releases neurotransmitter to communicate with another. The experiments I have been conducting here at the University of Arizona indicate that astrocyte-like glia interact with synapses. Synapses are fine anatomical features that can only be visualized using very high-magnification imaging techniques like electron microscopy. For decades, scientists have used two-dimensional electron microscopic images to study these structures, but recent advances in automation have allowed EM scientists at Janelia Farm to assemble thousands of 2-D images into a 3-D dataset. The computer programmers in the Cardona research group have created a navigable user interface that allows browsing, annotation, measurement, and various calculations to be performed. This multi-user environment is called “CATmaid”, an abbreviation for Collaborative Annotation Tool for Massive Amounts of Image Data. Even for a fly brain, reconstructing the entire organ is an enormous undertaking – so enormous that scientists from over a dozen labs around the world are collaborating to complete this task.
My project was the first to examine glial cells in the 3-D dataset. I began by marking and naming the cell body of each astrocyte-like glial cell. These are readily identifiable because glycogen content is higher in glia than in neurons, causing glial cytoplasm to appear darker in EM preparations than neuronal cytoplasm and the glial nucleus is irregularly shaped. Next, I selected a single cell of interest and began tracing the trajectory of each process emanating from the cell body. The tools in CATmaid enable a “skeleton”, or “stick” reconstruction, and after spending nearly two weeks trying to faithfully represent the structure of one glial cell in this manner, we reached the conclusion that the sort of anatomical analysis we desired would require a space-filling reconstruction. Such a tool is currently in development for the CATmaid environment, but has not yet reached the software testing phase, so using the tools available at present, I sampled four three-dimensional volumes and assessed the synaptic density, average distance from a synapse to the nearest glial process, and quantified instances of direct contact between astrocyte-like glia and synapses. We expected direct synapse-glia contact to be a rare occurrence, but instead I found that nearly a quarter of synapses contact a glial process. This finding has significant implications for my physiology-based dissertation.
In addition to contributing to the annotation of the fly brain, I also used this visit to learn as much as possible from the anatomical experts who have been tracing neuronal circuits. This information will allow me to isolate small groups of neurons with shared features, such as neurotransmitter identity, activate them selectively, and examine resultant events in glial cells as well as any glia-mediated effects on the neurons that receive synapses from this population. Through discussions with Dr. Cardona, Dr. Jim Truman, and Dr. Casey Schneider-Mizell, I was able to identify an ideal group of glutamate-releasing neurons for this purpose and am now pursuing these experiments here in Arizona. Finally, it is important to note that the CATmaid environment is remotely accessible, allowing me to continue working on this project for the duration of my dissertation. Specifically, I am measuring the distances between glia and synapses made by the small group of neurons I described above and comparing the distance to synapses which release different neurotransmitters to test the hypothesis that glutamate-releasing synapses associate more closely with glial cells. In the future, as the space-filling tool becomes available, I plan to return to anatomical analyses of individual astrocyte-like glia.