On June 8th, members of Yale’s moderately sized yet exuberant structural biology community flocked to West Campus to witness the long-awaited unveiling of a multi-million-dollar Titan Krios cryo-electron microscope. The Krios purchase consummated West Campus’ tenth birthday. And because the microscope is as powerful as its name suggests, it will no doubt prove a bulwark for Yale scientists wishing to partake in the burgeoning “resolution revolution” that is taking structural biology by storm. However, a recent Nature Methods paper published by Professor Gabriel Lander and coworkers at The Scripps Research Institute suggests that maybe Yale could have gotten away with saving more than a few dollars on its recent purchase.
Typically, it is believed that in order to solve structures of biological macromolecules at atomic resolution with cryo-electron microscopy (cryo-EM), a technique for which the 2017 Nobel Prize in chemistry was awarded, a high power microscope like the Krios must be employed. However, the Scripps team showed that a microscope with two-thirds the power of the Krios is sufficient for obtaining atomic resolution data.
The germ of Lander and team’s paper lay in a serendipitous discovery. The team had been working on a project that aimed to solve the structure of a membrane protein. “We were simply using a lower energy instrument to screen the quality of the sample. We wanted to produce preliminary data to make sure that it was good enough to put into the Krios for high resolution structure determination,” Lander said. “But these preliminary data produced a near-atomic resolution structure. We were absolutely astonished because we never expected to achieve these types of resolutions on this microscope, based on what we’d been told.”
With this observation in mind, Lander and coworkers were convinced that lower energy microscopes could produce results similar to those obtained from the Krios – at least some of the time. So, they set out to prove it by fine-tuning certain steps during the cryo-EM protocol and standardizing their procedure.
To solve a structure via cryo-EM, one must first obtain a pure, stabilized sample of the macromolecule in question. Next, the sample must be spread onto a thin grid and vitrified (rapidly flash frozen) so that it remains in its liquid phase conformation, the form it normally takes in the aqueous environment of the cell. Finally, the frozen specimens are loaded onto an electron microscope and shot with intense electron beams. Each vitrified copy of the macromolecule blocks electrons from hitting the detector, creating a two-dimensional projection, similar to a shadow cast by an ordinary object under the sun. Like shadows, which are crisper and darker when the sun is brighter, the more intense the electron beam, the better the projection. However, unlike shadows, whose intensity does not depend on the thickness of the object, the “thicker” (i.e., the more electron-dense) the macromolecule, the darker the projection. Finally, an algorithm combines all of the projections to construct a three-dimensional map of the macromolecule.
For Lander and team, there was no single key to providing a proof of principle that lower energy microscopes can generate atomic resolution structures. Rather, they toiled around in the lab, optimizing many elements of the cryo-EM protocol, such as the thickness of the vitrified sample and the uniformity of the electron beam. If there is one lesson to glean from their work, it is that scientists should not always jump to more expensive, flashier equipment to generate good data. Sometimes, a cheaper, simpler approach may suffice. Hopefully, scientists at universities that cannot afford an expensive Krios microscope will take note of Lander and team’s discovery and invest in cheaper microscopes that may be capable of comparable results.