Electron Microscopy at the DCI
Electron microscopy in Life Sciences can be grouped into several categories:
This method is used to determine the high-resolution three-dimensional (3D) structure of isolated protein complexes. Samples are flash frozen in a thin layer of vitreous ice and imaged at cryogenic temperatures by transmission electron microscopy (TEM), e.g., using the TFS Titan Krios G4i microscope. The direct electron detectors (DEDs) of these instruments greatly increase the achievable image contrast, and allow ‘movie mode’ imaging. In this mode, a series of dose-fractionated images (i.e., movies that are containing up to 320 frames) of suitable sample regions are recorded. The movie mode data collection allows sample drift to be detected and corrected. After the required alignment, the individual frames are merged to increase image contrast, i.e., to improve the overall signal-to-noise ratio.
The Falcon4i detectors employed by the DCI alternatively allow recording image data as a series of X/Y electron impact coordinates, together with the time point of the arrival of the electron. Such « electron event recordings » can later be interpreted by software and transformed into movies of variable pixel resolution and a variable number of frames for a movie.
Each sample region imaged contains many protein complexes oriented in different ways. Their images, i.e., projections in 2 dimensional space (termed 2D projections) are matched, classified and reconstructed in a procedure called ‘single particle analysis’ to obtain the 3D structure of the complex. Alpha helices and beta sheets can be visualized by this method, and, since the introduction of DEDs, atomic resolution is frequently achieved.
- Sample Requirements for Single Particle or Helical Cryo-EM
- Cryo-EM workflow
- Cryo-EM workflow for helical samples
- Cryo-EM image processing
A single particle cryo-EM project requires about 30 microliters of protein solution at a concentration of around 1 mg/ml. For a 100 kDa protein that would correspond to 10 micromolar concentration. For a 200 kDa protein it would correspond to 5 micromolar concentration. Samples at lower concentration can sometimes be up-concentrated, but one has to pay attention to not also up-concentrate detergents if present.
The protein should be in a buffer that it is happy with. Salt concentrations should be below 200mM, otherwise the solution becomes rather dense for the electron beam and contrast is lowered in the electron microscope. Mercaptoethanol or higher concentrations of glycerol often make problems for cryo-EM investigations and should be avoided.
The protein should be pure (An SDS-PAGE shows that >50% of the particles are the protein of interest), homogeneous (a SEC profile shows a symmetric narrow peak), and stable (the SEC profile looks the same a few hours later).
As first characterization, negatively stained preparations on TEM grids should be prepared and investigated with a 120kV TEM. This can be done at the EMF at UNIL or at the CIME or the BioEM Lab at the EPFL. (For info on these partner facilities, please see on the front page of this website). Neg. stain TEM requires only 0.1 mg/ml protein concentration, and is often incompatible with phosphate buffers, as these tend to precipitate with the Uranyl Acetate stain.
If neg.stain TEM shows a suitable population of particles, cryo-EM grids have to be prepared by plunge freezing. For this, typically 3 microleters of sample solution at 1 mg/ml is given onto a glow-discharged fenestrated cryo-EM grid, blotted for a few seconds, and rapidly frozen to -196ºC by plunging into LN2-cooled liquid ethane. Grids are then inspected in a 120kV or 200kV cryo-EM instrument, which is available at the EMF@UNIL or the CIME/BioEMLab@EPFL or also at the DCI.
If cryo-EM grids show a suitable density and distribution of randomly oriented particles that are freely suspended in thin vitrified ice, a larger set of dose-fractionated cryo-EM images can be recorded with a 300kV Titan Krios instrument at the DCI.
A data collection session takes 24 hours or longer, depending on the biological question, target resolution, and the particle density and homogeneity. In some cases, a heterogeneous particle population is desired, to simultaneously capture various protein conformations within one data set.
Is my sample suitable (pure, homogeneous, stable)?
- Preparation of negatively stained grids
- TEM imaging @ 120kV
- Computer analysis for sample homogeneity control
Can my sample be frozen well?
- Preparation of frozen grids
- Cryo-EM imaging @ 200k/
- Computer analysis for sample quality control
Get the high-resolution structure
- Identification of the optimal, frozen grid
- Cryo-EM imaging @ 300k/
- Computer analysis for 3D structure reconstruction
Cryo-EM workflow for helical samples
Cryo-EM image processing
Recorded data will be computer processed during data collection on DCI computers, so that a first image processing results is available during data collection. Subsequent in-depth processing of recorded images, 3D structure reconstruction and atomic model buildingrequires GPU computing hardware, certain software systems, and expertise, all of which is available at the DCI. Please contact us for more info.
In electron tomography, each object of interest in the sample is imaged horizontally and at various angles to the electron beam. The recorded series of images from different tilt angles is then called a « tilt series ». The increase in thickness to the electron beam of the tilted sample limits the possible angular range to ±60°. Computer image processing of the images gives a 3D reconstruction of the sample, yet a lower resolution, because individual movements of sub-sections of the sample during recording of the tilt series cannot be compensated at this step (it will be compensated later, see below). A first 3D reconstruction from a tilt series is therefore often limited to 2nm 3D resolution. Because of the limited tilt range, the whole 3D Fourier space cannot be sampled, so that the resolution in the z direction is lower than in the X and Y directions.
Recorded tilt series and their first 3D reconstruction can then be inspected manually or by computer software, to identify regions of interest. If for example the several ribosomes in a cell would be the objects of interest, then the X/Y/Y coordinates of each ribosome could be detected. If software then extracts from each two-dimensional image from the tilt series the image segment that belongs to a certain ribosome, then for each ribosome the computer would obtain a set of individual images at different tilt angles. These could then be further processed in a single-particle-like approach, which allows movie-mode drift correction and precise correction of electron microscopy artifacts (such as the CTF), so that a much higher resolution can eventually be achieved at the end of this process. In several cases, this single-particle processing of electron tomography tilt series led to side-chain-resolving resolution on proteins that were still within the cellular enviroment.
- Sample Requirements for Cryo-Electron Tomography
- Extension of Electron Tomography
- Cryo-ET image processing
Sample Requirements for Cryo-Electron Tomography
Cryo-ET requires the sample to be 300nm thin or thinner. Biological cells or tissue that is thicker needs to be vitrified (i.e., frozen to LN2 temperature without ice crystal formation), and then sectioned to thin lamellae. This can be done with a cryo-ultramicrotome or with a cryo-FIB-SEM, both of which are available at the EMF at UNIL.
The work flows are vast and very complex for sample handling, fixation or staining of fluorescent labeling, freezing, sectioning, and correlative imaging with a light microscope or other instruments prior to the cryo-ET investigation. Please contact us for discussing your project and identifying a suitable sample handling strategy.
Once suitably thin cryo-preparations of the biological sample are available, these cryo-EM grids can then be imaged in a 300kV Titan Krios at LN2 temperature in so-called tilt series. During this procedure, the sample is imaged as dose-fractionated movies in the Titan, while the sample is rotated to various different tilt angles. This results in a series of images, each recorded at a different sample tilt angle. This “tilt series” of images can then with computer image processing be transformed into a first 3D reconstruction of the sample.
Sub-volumes containing similar proteins from the cell (e.g., lots of ribozomes) can then be localized in this 3D reconstruction with computer help, and a larger number of such extracted sub-volumes can then be analyzed, classified, aligned and averaged, to yield a high-resolution representative structure of that molecule within the cellular context. Such sub-volume analysis (SVA) can in favorable cases sometimes reach very high resolution, so that even secondary structure elements or still higher resolution can be recognized from a sample that was never purifed.
Extension of Electron Tomography
Electron tomography can be used to :
- Extend the information obtained by Serial Block-face Imaging in the scanning electron microscope.
When areas of interest are located by room temperature serial block-face imaging, the resin-embedded sample (e.g., small region of tissue or a small region of a mouse brain) can be removed from the microscope (SEM), and mounted in a conventional ultramicrotome, where thicker (e.g. 100 to 400 nm) sections are cut. These sections are then sequentially mounted on electron microscopy grids and analyzed by electron tomography in a transmission EM, e.g., the FEI Titan. Together the images reveal the 3-D structure of the resin-embedded sample at a resolution of about 30 nm.
- Examine cryo-sections of samples that were embedded in resin at cryo temperatures. Sub-nanometer resolution can be achieved.
- Examine cryo-sections of samples embedded in vitreous ice (CEMOVIS). The 3D structure of the sample under close-to-native conditions is obtained at nanometer resolution.
Cryo-ET image processing
Recorded data will be computer processed during data collection on DCI computers, so that a first 3D reconstruction is available during data collection. Subsequent in-depth processing of recorded images requires computing hardware with GPUs that have generous RAM on each GPU card, certain software systems, and expertise, all of which is available at the DCI. Please contact us for more info.