Computational protocol: Expression, Purification, and Structural Insights for the Human Uric Acid Transporter, GLUT9, Using the Xenopus laevis Oocytes System

Similar protocols

Protocol publication

[…] Negative-stain TEM and grid preparation was performed as described . In brief, solubilized hGLUT9b (15 µg/ml) was adsorbed for 10 seconds to parlodion carbon-coated hydrophobic copper grids. Grids were washed in ddH20 and negatively stained with 0.75% (w/v) uranyl acetate. Electron micrographs were recorded at a magnification of 110,000× on a Morada CCD camera from OLYMPUS where pixel size was 3.092 Å. The Philips CM-12 electron microscope operated at 80-kV acceleration voltage.Single Particle Reconstruction (SPR) was performed using EMAN2 (Electron Micrograph ANalysis) open-source suite programs . Digital TIFF images (2970×2100) were recorded in 16-bit using the software iTEM (OLYMPUS) under constant focus and astigmatism corrections. Standard EM parameters were used to capture micrographs. The images were obtained free of drift, vibration and astigmatism, with slight under-focus controlled by the iTEM software during live acquisition. A detailed process of the subsequent SPR is described below.i. Particle selection (e2boxer.py)Contrast level were adjusted by Photoshop CS suite for each micrographs and saved in 8-bit. Electron micrographs were import in EMAN2 suite. In EMAN2, all program are executed using the built-in workflow GUI (e2workflow.py). We apply a number of common filters to the data before importing such as Edge nom thought e2workflow.py program. All micrographs are saved as “MRC” (Medical Research Council) files. Particles were selected with box dimension 84×84 pixels as 7056-dimensional vector using semi-automatic picking function by Swarm mode (algorithm uses a trainable heuristic based approach) in e2boxer.py program. The results were manually verified, and false positives were eliminated, at this step of the process image quality weight can be addressed 0 to 4. When all particles are selected output can be written with box coordinates. Images were normalized by the normalize.edegemean option. Output images were saved in default format “BDB”files used for processing in the workflow interface.ii. CTF and phase flipping corrections (e2ctf.py)The aim of SPR is to generate the “true” 3D structure of a macromolecule based on its 2D projections. Inherent of contrast transfer function (CTF) and the envelope function of the electron microscope, the projections observed are not reflective of the real projections of the electron density of the specimen. CTF is a mathematical representation of the imaging process in the TEM, examined in reciprocal space. Begin with selecting the particles intended to generate the CTF parameters using the following steps: i/Autofit, ii/manually fine-tune parameters for a few sets at different defocuses iii/generate a structure factor using these sets iv/re-run autofit v/manually check the fitting results. In practice, 2D power spectra and 1D averaged power spectrum of the boxed out particles from each single image is used to characterize the CTF with three parameters: defocus, B-factor and %AC (Amplitude Contrast). The particle set is built from phase-flipped output. The Phase-flipping corrections simply consist of multiplying the Fourier transform of each particle image by -1 over the appropriate frequency ranges.iii. Reference-free class averages (e2refine2d.py)In this step, the strategy is to sort raw particles presenting the same orientations into different groups based solely on their 2D projection characteristics. A set of representative class-averages is generated and only used to assess the structural variability of the specimen and to create an initial 3D-model. The result is a reduction of the noise level to give a shape more detailed from specific or “class” angles of the specimen corresponding to the different views.iv. Initial model building (e2initialmodel.py)This program called e2initialmodel uses class-averages produced to create few initial models classified by quality for use in further refinement. Based on the crystal structure of XylE, hGLUT9 was assumed to be asymmetrical. Therefore, we utilized the Asymmetrical C1 parameter within the program to generate the initial model. The 3D-model generated from the class averages was used to accommodate the predicted homology-based model of hGLUT9 and for further refinement within SPR.v. Refinement and Resolution (e2refine.py, e2eotest.py, e2resolution.py)Based on the initial model, back projections in all obtained orientations are generated and computationally compared to the original 2D-projections to identify similarities among the individual classes. Sets of more similar projections are iteratively aligned and averaged. Class-averages, for which the orientation in known, as defined by the initial model, are used to build a new, more refined, 3D-model. Three iterations and the generated Fourier Shell Correlation curve at 0.5 were used to define the resolution of the reconstruction at 23 Å. […]

Pipeline specifications

Software tools EMAN, e2boxer.py
Application cryo-EM
Organisms Xenopus laevis, Homo sapiens
Chemicals Glucose, Uric Acid