Archives

  • 2018-07
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • 2022-02
  • 2022-03
  • 2022-04
  • 2022-05
  • 2022-06
  • 2022-07
  • 2022-08
  • 2022-09
  • 2022-10
  • 2022-11
  • 2022-12
  • 2023-01
  • 2023-02
  • 2023-03
  • 2023-04
  • 2023-05
  • 2023-06
  • 2023-08
  • 2023-09
  • 2023-10
  • 2023-11
  • 2023-12
  • 2024-01
  • 2024-02
  • 2024-03
  • 2024-04

    • Different from stable transfection transient expression was

    2021-09-14

    Different from stable transfection, transient expression was also found to yield γ-secretase with robust proteolytic activity []. Co-transfection of all four, but not any three, components of human γ-secretase yielded high expression levels. This observation strongly suggests the co-folding nature of the entire γ-secretase complex. The nuance of co-transfection of four plasmids was alleviated by the cloning of all four γ-secretase components into a single mammalian expression vector, named pMLink []. With a cassette design, pMLink allows convenient cloning and expression of multiple open reading frames (ORFs), each under an independent promoter (Figure 1a). The resulting plasmid was transfected into HEK293F Alrestatin for overexpression of human γ-secretase. A FLAG tag fused to the amino-terminus of PEN-2 greatly facilitated purification []. This design was the outcome of an exhaustive screening effort, in which several different tags in distinct positions for each of the four γ-secretase components were attempted []. Recombinant human γ-secretase was extracted from the membrane fractions by the zwitterionic detergent CHAPSO and purified using an anti-FLAG affinity column. A final step of size exclusion chromatography ensures sample homogeneity and removes high-molecular weight contaminants (Figure 1b). The purified γ-secretase showed excellent solution behavior in the presence of digitonin or amphipol (Figure 1c). Five polypeptides from the four components, with PS1 represented by the NTF and CTF, can be visualized by Coomassie blue staining (Figure 1d) []. Using this strategy, approximately 0.2–0.3mg homogeneous γ-secretase was obtained from each liter of HEK293F culture.
    Electron microscopy Procedures of electron microscopy reviewed here only pertain to the four structures of γ-secretase at near-atomic resolutions [26••, 30••, 31••, 32••]. Getting recombinant human γ-secretase into vitreous ice proved to be tricky. Among many detergents tested, only digitonin and amphipol work well. Preparation of the cryo-EM specimens using an FEI Vitrobot follows a standard protocol, with an aliquot of 3-μl concentrated γ-secretase on a glow-discharged holey carbon grid blotted for 4s and flash frozen in liquid ethane. Despite few differences in proteolytic activity, different batches of human γ-secretase appeared to behave quite differently in terms of sample quality for cryo-EM analysis. This likely reflects the inherent conformational heterogeneity in γ-secretase that can be inadvertently mitigated by unnoticed differences in manipulation and/or experimental conditions. The four published structures of γ-secretase relied on three batches of sample preparation, two in amphipol and one in digitonin (Table 1). All micrographs were recorded by a Gatan K2 Summit direct electron detector in superresolution counting mode on a Titan Krios microscope. The dose rates were either 2.5 or 4.5 electrons per square angstrom per second, the exposure time varied from 10.4 to 16s, and the defocus values ranged from 0.6 to 4.4μm (Table 1). The energy filter and the relatively low defocus value, together with the quality of the sample, likely helped improve the overall resolution for γ-secretase. Analysis of the micrographs was standard, with whole-image motion corrected by MOTIONCORR [29] and contrast transfer function parameters estimated by CTFFIND3 [36]. Particles were manually picked from a subset of the micrographs to calculate 2D references, which were used as the template for reference-based auto-picking. Reference-free 2D and 3D classifications were performed using RELION (Figure 1e) [27]. Multiple rounds of 3D classification proved essential for the identification of a subset of the particles that exhibited homogeneous conformation. Per-particle motion correction and radiation damage weighting (particle polishing) further improved the map quality using RELION. The highest resolution achieved for γ-secretase is 3.4Å (Table 1). Notably, the quality of the EM density map appears to be better than X-ray-based electron density map at a similar resolution, with most amino acid side chains in the 20 TMs clearly defined. In this regard, the cryo-EM structure of human γ-secretase genuinely reaches an atomic resolution. Nonetheless, atomic modeling was facilitated by the crystal structures of an archaeal presenilin homologue PSH [] and an eukaryotic NCT (DpNCT) []. One of the most important factors for achieving an atomic resolution is the number of particles. More than 1.8 million particles were used for 2D and 3D classifications, allowing selection of a small subset of 159549 particles to yield a final reconstruction at an average resolution of 3.4Å.