Van Andel Institute Trains Next Generation of Cryo-EM Structural Biologists
September 20, 2018 § 2 Comments
By George Van Den Driessche
Cryo-electron microscopy, or cryo-EM, is structural biology’s golden ticket these days. Since the game-changing technology won the 2017 chemistry Nobel Prize, it feels like there is no problem that cannot be solved by cryo-EM.
Do you have a large protein that won’t crystallize? Are you studying transporter channels or membrane signaling proteins and have no idea what the protein looks like? Or are you designing new antiviral agents and need the viral capsid structure? Then, at least the current trend suggests, you should consider using cryo-EM, a technique that flash freezes proteins in liquid nitrogen and analyzes electron diffraction patterns to determine the molecular structure of proteins.
The Van Andel Institute, a nonprofit biomedical research facility in Grand Rapids, Michigan, invested more than $10 million dollars to install the David Van Andel Advanced Cryo-EM Suite in 2016. Last year, VAI celebrated the opening of its new facility with a conference about cryo-EM’s advancement of structural biology. This year, VAI hosted a special cryo-EM training workshop for graduate students, postdocs and new faculty. (Full disclosure: VAI awarded travel scholarships to workshop attendees, and I was one of them.)
Huilin Li, the VAI cryo-EM core director, designed the workshop so attendees could learn about cryo-EM theory and research advances on the first day and then spend the remaining time working with the microscopes, discussing protein-preparation techniques and reviewing image processing software. “This is a privately endowed research institute. We do research, collaboration, outreach and training. This workshop is part of the training,” Li said.
VAI assistant professor Juan Du, one of the presenters, studies TRPM4 channel proteins. TRPM4 is a calcium-activated nonselective (CAN) cation channel that monitors cellular charge by measuring calcium ion concentration. When the cell’s charge becomes too positive, calcium ions bind to TRPM4, and single-charged cations, such as sodium or potassium ions, flow through the channel. This process decreases the cell’s overall positive charge.
Du’s team determined the structure of TRPM4 and identified key calcium and modulator binding sites using single-particle cryo-EM. Their work appears in Nature. She advised attendees, “We are structural biologists, so we use all kinds of tools to solve problems, and very soon cryo-EM will be equal to X-ray crystallography, and, when that day comes, success will depend on who has a sample in hand. So, remember that biochemistry is the key to success.”
Keynote speaker Steve Ludtke, a professor of biochemistry and molecular biology at Baylor College of Medicine, presented his lab’s research streamlining cryo-electron tomography (cryo-ET) image annotation.
Cryo-ET allows researchers to see the dynamic interactions of frozen macromolecules and offers insight into the native environments of cells. Traditional cryo-EM requires protein purification, and this raises the possibility of disrupting a protein from its natural orientation.
However, a major hurdle facing cryo-ET is the time required for image annotation. Image annotation is a task where scientists identify all the individual macromolecules by hand in a series images. A fully annotated cryo-ET image takes about a week to complete.
Ludkte is training deep learning models to overcome the image annotation time barrier. These models are capable of learning a wide-range of macromolecule features (such as distinguishing double-membrane mitochondria from single-membrane organelles) and require minimal training. Each model needs only 10 human annotated training images before the algorithm can begin recognizing patterns in the cryo-ET image.
Georgia Institute of Technology graduate student Kasahun Neselu said he enjoyed the hands-on training and was leaving feeling excited and filled with ideas to test on his own research projects.
Fatemeh Abbasi Yeganeh, a graduate student at Florida State University, said she enjoyed learning more about cryo-EM, meeting new people who share her passion for structural biology, and gaining confidence in the field.
“Currently, EM is the main focus of my Ph.D. project, and I take any opportunity to learn more and meet new people in the field,” Yeganeh said. “So, what can be better for a graduate student than attending a workshop that pays all your expenses, teaches you about your favorite field of study, and, also, gives you the opportunity to meet and talk to experts in the field?”
George Van Den Driessche (firstname.lastname@example.org) is a graduate research assistant in the Fourches lab at North Carolina State University.