Neuroscience research has been revolutionized by technologies such as optogenetics and calcium imaging to manipulate and image neuronal activity. However, we still lack a detailed blueprint of how neural circuits connect in vivo. The only technology that allows for the imaging of neurons and their connections at resolutions high enough to view the components of synapses and cellular organelles is electron microscopy (EM).
The laboratories of David Tank and Sebastian Seung are debuting a brand new facility within the Princeton Neuroscience Institute (PNI) that will utilize high-throughput transmission electron microscope (TEM) to describe the brain’s connectome. Due to new technological advancements, this new facility will allow the collection of connectomic data at unprecedented speeds.
Why would researchers use TEM as opposed to traditional light microscopy? The small size of structures like axons (a couple hundred nanometers wide) and synaptic vesicles (40-50 nanometers in diameter) can pose problems for traditional light imaging, which cannot resolve structures this small. Thus, researchers must resort to using electrons to image these structures. When a section of tissue is being imaged, particle beams of electrons are passed through the section. Cellular structures stained with heavy metals scatter electrons more than the space around the stained structures, which creates contrast. How many electrons pass through a section depends on the thickness of the sample and membranes. Ultimately this information is captured by a high signal-to-noise camera and processed to create high resolution images.
Why is this new PNI facility such an improvement compared to existing technology? As usual, tissue samples will be cut down to a one millimeter cube and fixed. These fixed cubes will then be stained with heavy metals and embedded in resin for thin sectioning. This is where advancements in the new PNI facility comes into play. The 40 nanometer sections will be placed on a film tape and collected in a reel1. Previously, TEM required the manual imaging of each section, creating a labor-intensive workflow. The reel-to-reel automated imaging of these sections dramatically reduces the time spent imaging. Additionally, the stages on which the sections are placed are also able to move faster, lessening the time required to scan each section.
Previous transmission electron microscopes have an acquisition rate of ten to one-hundred megahertz. Building on previous technological advances2, Princeton’s new TEM facility has an unprecedented four-hundred megahertz raw acquisition rate. Combined with the four microscopes that can run simultaneously, the new TEM facility can acquire 4 terabytes of data per hour. Additionally, the sections can be re-imaged, if necessary, increasing the reliability of the data. In summation, the main two advantages of the new TEM facility are the automation and speed of imaging.
The facility finished construction in January of 2021 and the last microscope was installed in March. Researchers are now building the data acquisition software and the first volumes of data are expected to be collected in the following months. According to one researcher involved with the facility, “the only limiting factor now is collecting enough samples”. The researchers hope that the image acquisition pipeline used by the new TEM facility can be inserted at the end of many studies conducted at PNI, thereby connecting studies of functionality with electron microscopy to describe the circuits at high resolution.
by Chris Suriano
Phelps JS, Hildebrand DGC, Graham BJ, Kuan AT, Thomas LA, Nguyen TM, Buhmann J, Azevedo AW, Sustar A, Agrawal S, Liu M, Shanny BL, Funke J, Tuthill JC, Lee WA. Reconstruction of motor control circuits in adult Drosophila using automated transmission electron microscopy. Cell. 2021 Feb 4;184(3):759-774.e18. doi: 10.1016/j.cell.2020.12.013. Epub 2021 Jan 4. PMID: 33400916.
Yin W, Brittain D, Borseth J, Scott ME, Williams D, Perkins J, Own CS, Murfitt M, Torres RM, Kapner D, Mahalingam G, Bleckert A, Castelli D, Reid D, Lee WA, Graham BJ, Takeno M, Bumbarger DJ, Farrell C, Reid RC, da Costa NM. A petascale automated imaging pipeline for mapping neuronal circuits with high-throughput transmission electron microscopy. Nat Commun. 2020 Oct 2;11(1):4949. doi: 10.1038/s41467-020-18659-3. PMID: 33009388; PMCID: PMC7532165.