Structure of the human 26S proteasome: subunit radial displacements open the gate into the proteolytic core

Abstract

The 26S proteasome plays an essential role in regulating many cellular processes by the degradation of proteins targeted for breakdown by ubiquitin conjugation. The 26S complex is formed from the 20S core, which contains the proteolytic active sites, and 19S regulatory complexes, which bind to the 20S core to activate it and confer specificity for ubiquitinated protein substrates. We have determined the structure of the human 26S proteasome by electron microscopy and single particle analysis. In our reconstructions the crystallographic structure of each of the subunits of the 20S core can be unambiguously docked by direct recognition of each of their densities. Our results show for the first time that binding of the 19S regulatory particle results in the radial displacement of the adjacent subunits of the 20S core leading to opening of a wide channel into the proteolytic chamber. The analysis of a proteasome complex formed from one 20S core with a single 19S regulatory particle attached serve as control to our observations. We suggest locations for some of the 19S regulatory particle subunits.

FIGURE 1.

Electron micrograph of a negatively stained 26S proteasome sample. Examples of side views of double-capped and single-capped 26S proteasome images are identified with black and white circles, respectively. End views of 26S proteasome complexes are identified by black dashed circles.

FIGURE 2.

Selection from the initial class averages derived for side views of double-capped 26S proteasomes by automated alignment and classification procedures. A and B, class averages with 2-fold rotational symmetry characteristic of a projection along a C2 axis. C, class average with mirror symmetry characteristic of a projection normal to a C2 axis. D, example of an asymmetric class average, corresponding to intermediate projections to those along (A and B) and orthogonal (C) to the C2 symmetry axis.

FIGURE 3.

Three-dimensional analysis of the 26S proteasome. A, double-capped 26S proteasome. B, single-capped 26S proteasome. i, examples of class averages used for three-dimensional reconstruction; ii, corresponding surface views of the three-dimensional reconstructions; iii, corresponding forward projections of the three-dimensional reconstructions. C, resolution assessment by Fourier shell correlation.

FIGURE 4.

Three-dimensional reconstructions of the 26S proteasome. A, double-capped 26S proteasome. B, single-capped 26S proteasome. i-iii, side views, with the approximate location of proteasome components indicated (left) with viewing directions parallel to the C2 axis (i), normal to the C2 axis (ii), and parallel to the C2 axis (iii), in opposite direction to i. iv, views from capped ends of the proteasome complexes. v, end view from uncapped side of single-capped 26S proteasome.

FIGURE 5.

Subunit docking into the 20S-CP region of the double-capped 26S proteasome. A, the crystal structures of the 20S-CP subunits (schematic representation) were docked into the three-dimensional map of the 26S proteasome (transparent surface representation), viewed parallel (i and iii) and orthogonal (ii and iv) to the C2 axis shown. The subunits of the 20S-CP were docked as a single rigid body by aligning the common C2 axis (i and ii) or independently fitted (iii and iv). The common C2 symmetry axis is represented, and the approximate location of proteasome components is indicated (left). B and C, sections of the 26S proteasome three-dimensional map are shown as chicken wire, and crystal structures of docked subunits are represented as schematics. B, end view ofα ring (viewed from 19S-RP). i, 20S-CP coordinates docked as a single unit; ii, 20S-CP subunits docked independently; iii, comparison between i (gray shade) and ii (colored contours); iv, the section shown in i-iii is identified in side view of the 26S proteasome by the red box. C, end view of β ring (viewed from the adjacent α ring). Panels i-iv as for B. D, sections showing the α and β subunit rings, as for Bi and Ci, but where the subunits of the 20S core were fitted with a 180° rotation around the proteasome long axis, corresponding to the alternative orientation to that shown in A, B, and C, taking in account the common C2 symmetry axis between the 20S core and the double-capped 26S proteasome. Here the mismatch between the densities of the three-dimensional map and the 20S core crystal structure is apparent.

FIGURE 6.

Links between the 19S-RPs and the 20S-CP of the double-capped 26S proteasome. A-C, side views of three-dimensional map enlarged to show one 19S-RP and approximately half the 20S-CP, with the different links indicated. A, stereo representation of view parallel to C2 axis. B, view as in A, with the L-shaped link indicated. C, viewed normal to C2 axis, with links I, II, and III identified. D, end view of α ring viewed from 19S-RP with independently docked subunits, as shown in Fig. 5, Bii, identifying the attachment sites of the links to the α ring.

FIGURE 7.

Subunit docking within the 20S-CP of the single-capped 26S proteasome. Sections of the three-dimensional map are shown as chicken wire, and crystal structures of independently docked subunits are represented as schematics. A, side view, sectioned close to long axis of the proteasome, showing a clear aperture in the α ring adjacent to the 19S-RP. B, end view of α ring adjacent to 19S-RP (viewed from the capped end). C, end view of α ring at the uncapped end of the 20S-CP (viewed from the free end).