The proteasome 19S cap and its ubiquitin receptors provide a versatile recognition platform for substrates

Abstract

Proteins are targeted to the proteasome by the attachment of ubiquitin chains, which are markedly varied in structure. Three proteasome subunits-Rpn10, Rpn13, and Rpn1-can recognize ubiquitin chains. Here we report that proteins with single chains of K48-linked ubiquitin are targeted for degradation almost exclusively through binding to Rpn10. Rpn1 can act as a co-receptor with Rpn10 for K63 chains and for certain other chain types. Differences in targeting do not correlate with chain affinity to receptors. Surprisingly, in steady-state assays Rpn13 retarded degradation of various single-chain substrates. Substrates with multiple short ubiquitin chains can be presented for degradation by any of the known receptors, whereas those targeted to the proteasome through a ubiquitin-like domain are degraded most efficiently when bound by Rpn13 or Rpn1. Thus, the proteasome provides an unexpectedly versatile binding platform that can recognize substrates targeted for degradation by ubiquitin chains differing greatly in length and topology.

Fig. 1. Model substrates and proteasome particles.

a Schematic representation of the substrate proteins analyzed. GFP shown in green, co-translated ubiquitin domains shown in light blue, enzymatically added ubiquitin moieties shown in dark blue, intrinsically unfolded regions of proteins shown as red lines. In the three-dimensional structure of GFP, the N- and C-termini are adjacent to each other so that ubiquitin chain or UBL and initiation region are next to each other in space. b Structure of the proteasome in the s1 state (PDB4CR2) in two orientations. The structure broadly includes one half of the 20S core particle (gray) and the 19S regulatory particle (multiple colors). The ATPase (Rpt) subunits are color-coded respect to domains rather than subunits, with the AAA+ domains in magenta, the OB ring in pink, and the coiled-coil domains in purple. The position of the pore-1 loop is indicated by a red circle in the right panel. The Ub/UBL receptors Rpn1, Rpn10, and Rpn13 are shown in green, and Rpn2 is shown in light blue. The DUB Rpn11 is shown in yellow, and the remaining components of the lid are shown in light yellow. Black lines indicate a direct path along the surface of the proteasome from the Ub/UBL-binding sites in Rpn1 and Rpn13 to the pore-1 loop at the top of the ATPase ring. The Ub/UBL-binding UIM domain of Rpn10 is not visualized in this structure; instead, the black line indicates the path from the last resolved residue of the VWA domain of Rpn10 to pore-1 loop. The distances are approximately 107 Å for Rpn13, 100 Å for Rpn1, and 95 Å for Rpn10. The UIM domain of Rpn10 is attached to the VWA domain by an unstructured linker of approximately 20 amino acids, which may reduce the distance from the ubiquitin-binding site to the pore-1 loop by 20–30 Å.

Fig. 2. Degradation of substrates with K48-linked polyubiquitin chains.

Degradation of substrate proteins with K48-linked ubiquitin chains by the indicated proteasome mutants was followed under single-turnover conditions (5 nM substrate, 25 nM proteasome) in the presence of 1 mM ATP at 30 °C. The graphs show substrate fluorescence as a percentage of the initial fluorescence as a function of time in minutes. Proteasome types are described in Supplementary Table 1 (TM triple mutant proteasome). Each panel shows the degradation of particular substrates, and as follows: a Ub₅(K48)-GFP-35; b Ub₉(K48)-GFP-35; c Ub₅(K48)-GFP-95, and d 95-GFP-Ub₅(K48). Source data are provided as a Source Data file.

Fig. 3. Degradation of substrates with K63-linked polyubiquitin chains.

Degradation of substrate proteins with K63-linked ubiquitin chains by the indicated proteasome mutants was followed under single-turnover conditions (5 nM substrate, 25 nM proteasome) in the presence of 1 mM ATP at 30 °C. The graphs show substrate fluorescence as a percentage of the initial fluorescence as a function of time in minutes. Proteasome types are described in Supplementary Table 1 (TM triple mutant proteasome). Each panel shows the degradation of particular substrates, and as follows: a Ub₅(K63)-GFP-35; b Ub₅(K63)-GFP-95; c Ub₉(K48)-GFP-95; d 35-GFP-Ub₅(K63); and e 95-GFP-Ub₅(K63). Source data are provided as a Source Data file.

Fig. 4. Degradation of substrates with K11-linked, M1-linked, and linear polyubiquitin chains.

Degradation of substrate proteins with the shown ubiquitin chains by the indicated proteasome mutants was followed under single-turnover conditions (5 nM substrate, 25 nM proteasome) in the presence of 1 mM ATP at 30 °C. The graphs show substrate fluorescence as a percentage of the initial fluorescence as a function of time in minutes. Proteasome types are described in Supplementary Table 1. Each panel shows the degradation of particular substrates, and as follows: a Ub₅(K11)-GFP-35; b Ub₅(M1)-GFP-35, and c Ub₄(lin)-GFP-35 by mutant proteasome as indicated. For panels a and b, the positive control (Pos) is degradation of Ub₅(K48)-GFP-35 by wild-type proteasome. Source data are provided as a Source Data file.

Fig. 5. Inhibition of substrate degradation by free ubiquitin chains.

Degradation of substrate proteins with three different ubiquitin chain modifications by the indicated proteasome types was followed under single-turnover conditions (5 nM substrate, 25 nM) in the presence of 1 mM ATP at 30 °C by measuring GFP fluorescence. Increasing concentrations of ubiquitin chains were added and the initial rates of the reactions determined by curve fitting. Graphs show initial rates of degradation as a percentage of the rate in the absence of competitor. a Ub₅(K48)-GFP-35 on WT proteasome, b Ub₅(K48)-Ub-GFP-35 on Rpn10 proteasome, c Ub₄(lin)-GFP-35 on Rpn13 proteasome inhibited by free ubiquitin chains. Error bars show standard errors derived from at least three replicate experiments. Source data are provided as a Source Data file.

Fig. 6. Michaelis–Menten analysis of proteasomal degradation.

Degradation of substrate proteins with three different ubiquitin chain modifications by the indicated proteasome types was measured at 30 °C in the presence of 1 mM ATP and monitored by measuring GFP fluorescence. The indicated concentrations of substrate were incubated with 25 nM proteasome, fluorescence was recorded every minute for 120 min, and initial degradation rates were determined by curve fitting. The graphs show the initial rates as a function of substrate concentration. a Ub₅(K48)-GFP-35; b Ub₉(K48)-GFP-35; c Ub₅(K63)-GFP-35 by mutant proteasome. Error bars show standard errors derived from at least three replicate experiments. Source data are provided as a Source Data file.

Fig. 7. Degradation of multi-ubiquitinated substrates.

Degradation of substrate proteins modified with two K48-linked ubiquitin chains or bearing two co-translated single ubiquitin domains by the indicated proteasome types was followed under single-turnover conditions (5 nM substrate, 25 nM proteasome) in the presence of 1 mM ATP at 30 °C. The graphs show substrate fluorescence as a percentage of the initial fluorescence as a function of time in minutes. Proteasome types are described in Supplementary Table 1 (TM triple mutant proteasome). Each panel shows the degradation of particular substrates, and as follows: a Ub₃(K48)-35-Ub₃(K48)-GFP-35 and b Ub-35-Ub-GFP-35. Source data are provided as a Source Data file.

Fig. 8. Degradation of UBL-targeted substrates.

Degradation of substrate proteins with ubiquitin-like domains derived from Rad23 (UBL) by the indicated proteasome types was followed under single-turnover conditions (5 nM substrate, 25 nM proteasome) in the presence of 1 mM ATP at 30 °C. The graphs show substrate fluorescence as a percentage of the initial fluorescence as a function of time in minutes. Proteasome types are described in Supplementary Table 1 (TM triple mutant proteasome). Each panel shows the degradation of particular substrates, and as follows: a UBL-GFP-35 and b UBL-GFP-95 on different proteasome mutants. Source data are provided as a Source Data file.