Final Report Summary - EM-FRAME (DNA-origami scaffolds for structure determination by single particle analysis)
We aimed to improve cryo-EM structure determination by single-particle analysis, in particular for relatively small complexes (100-300 kDa). In that context, we highlighted three fundamental problems in cryo-EM structure determination that arise from the intrinsically low signal-to-noise ratios in the experimental images. Our proposed solutions to these problems were centered on the use of 3D DNA-origami objects (EM-frames). However, the advent of direct-electron detectors resulted in images of unprecedented quality. This had the unforeseen side effect that for a wide range of samples, the use of 3D DNA-origami frames was no longer necessary. We still addressed (and solved!) the 3 problems, but we did this in unanticipated ways.
Problem 1): Selecting particles is hampered by very noisy backgrounds. Solution: better contrast of particles in the images from the new detectors greatly facilitated particle selection.
Problem 2): Beam-induced motions affect the attainable resolution. Solution: we developed an image-processing procedure that uses movies from the new detectors to correct for sample motions. Using this procedure, we obtained near-atomic resolution reconstructions from only 30,000 particles (Bai et al. eLife 2013).
Problem 3): The accuracy with which the orientations of individual particles can be determined is a limiting factor for structure solution. Solution: also in this case did the improved quality of the images lead to a huge improvement, such that we can now align particles in a wide size range.
We then went on to demonstrate the potential of the developed procedures by solving near-atomic resolution cryo-EM structures for the mitochondrial ribosome from yeast (Amunts 2014, Science) and human (Brown 2015, Science), the yeast cytoplasmic ribosome in complex with initiation factor 5B (Fernandez, 2013, Science), the cytoplasmic ribosome from P. falciparum (Wong, 2014, eLIFE), human gamma-secretase (Lu, 2014, Nature, Bai 2015, Nature), the DARK apotosome (Pang, 2015, Gene & Development), and the rabbit ryanodine receptor RyR1 (Yan, 2015,Nature).
Problem 1): Selecting particles is hampered by very noisy backgrounds. Solution: better contrast of particles in the images from the new detectors greatly facilitated particle selection.
Problem 2): Beam-induced motions affect the attainable resolution. Solution: we developed an image-processing procedure that uses movies from the new detectors to correct for sample motions. Using this procedure, we obtained near-atomic resolution reconstructions from only 30,000 particles (Bai et al. eLife 2013).
Problem 3): The accuracy with which the orientations of individual particles can be determined is a limiting factor for structure solution. Solution: also in this case did the improved quality of the images lead to a huge improvement, such that we can now align particles in a wide size range.
We then went on to demonstrate the potential of the developed procedures by solving near-atomic resolution cryo-EM structures for the mitochondrial ribosome from yeast (Amunts 2014, Science) and human (Brown 2015, Science), the yeast cytoplasmic ribosome in complex with initiation factor 5B (Fernandez, 2013, Science), the cytoplasmic ribosome from P. falciparum (Wong, 2014, eLIFE), human gamma-secretase (Lu, 2014, Nature, Bai 2015, Nature), the DARK apotosome (Pang, 2015, Gene & Development), and the rabbit ryanodine receptor RyR1 (Yan, 2015,Nature).