Mechanistic insight into the allosteric activation of a ubiquitin-conjugating enzyme by RING-type ubiquitin ligases

Abstract

Ubiquitin-conjugating enzymes (E2s) collaborate with the ubiquitin-activating enzyme (E1) and ubiquitin ligases (E3s) to attach ubiquitin to target proteins. RING-containing E3s simultaneously bind to E2s and substrates, bringing them into close proximity and thus facilitating ubiquitination. We show herein that, although the E3-binding site on the human E2 UbcH5b is distant from its active site, two RING-type minimal E3 modules lacking substrate-binding functions greatly stimulate the rate of ubiquitin release from the UbcH5b-ubiquitin thioester. Using statistical coupling analysis and mutagenesis, we identify and characterize clusters of coevolving and functionally linked residues within UbcH5b that span its E3-binding and active sites. Several UbcH5b mutants are defective in their stimulation by E3s despite their abilities to bind to these E3s, to form ubiquitin thioesters, and to release ubiquitin at a basal rate. One such mutation, I37A, is distant from both the active site and the E3-binding site of UbcH5b. Our studies reveal structural determinants for communication between distal functional sites of E2s and suggest that RING-type E3s activate E2s allosterically.

Fig. 1.

SCA of E2s. (A) Hierarchical clustering of a submatrix of pairwise statistical coupling energies of E2 residues, which are plotted as a color gradient, with blue and red representing the lowest (0 kT*, with kT* being an arbitrary energy-like unit) and highest (2 kT*) energies, respectively. Both columns and rows of the matrix represent residue positions in UbcH7 and UbcH5b (in parentheses). The two clusters of residues that exhibit similar coupling patterns are boxed with dashed lines. (B) The two clusters of residues in A are shown as space-filling models and mapped onto the structure of UbcH5b. Cluster I and II residues are colored green and cyan, respectively. E3-binding residues are shown in red. The active-site cysteine and asparagine are labeled.

Fig. 2.

E3-stimulated release of ubiquitin from the UbcH5b–ubiquitin thioester. (A) WT and 31 mutants of UbcH5b were tested for their ability to release ubiquitin from their thioesters in the presence of buffer (–) or Apc2/11 (+) with continuous E1-catalyzed ubiquitin charging. The bands corresponding to free UbcH5b, UbcH5b–ubiquitin thioester, and monoubiquitinated UbcH5b (with isopeptide-linked ubiquitin) are indicated. (B) Same as in A except that CNOT4N was used as the E3. (C) UbcH5b I88A was defective in supporting APC/C^Cdh1-mediated ubiquitination of cyclin B1. Xenopus egg APC/C was incubated with buffer or Cdh1 and assayed for its ability to ubiquitinate an N-terminal fragment (residues 1–102) of human cyclin B1 in the presence of the varying concentrations of WT or mutant UbcH5b. Cyclin B–ubiquitin conjugates are indicated.

Fig. 3.

Kinetics of ubiquitin charging and release from the ubiquitin thioester of UbcH5b WT, N77A, I88A, and S94G. (A) UbcH5b was incubated with E1, ubiquitin, and ATP for the indicated times, separated on nonreducing SDS/PAGE, and blotted with anti-UbcH5b. Bands for UbcH5b and UbcH5b–ubiquitin thioester are indicated. (B) After ubiquitin charging, NEM was added to the reactions to inactivate E1. Samples were taken at the indicated time points, separated on nonreducing SDS/PAGE, and blotted with anti-UbcH5b. (C) Quantitation of reactions in B. The average and standard errors are shown. Filled circles, WT; open circles, I88A; open triangles, N77A; filled triangles, S94G. (D) After ubiquitin charging, apyrase was added to deplete ATP and thus to inactivate E1-mediated ubiquitin charging of UbcH5b. CNOT4N was also added to the reactions after 3 min. Samples were taken at the indicated time points, separated on nonreducing SDS/PAGE, and blotted with anti-UbcH5b. (E) Quantitation of reactions in D. The average and standard errors are shown. Symbols are the same as in C.

Fig. 4.

Allosteric determinants of UbcH5b. (A) UbcH5b residues involved in the long-range communication between the E3-binding and active sites of UbcH5b are shown in space-filling models. The E3-binding L1 and L2 loops, I37, and C85 are labeled. (B) A subset of allosteric UbcH5b residues is shown to illustrate the position of I88 relative to the E3-binding and active sites of UbcH5b. (C) The chemical-shift changes of UbcH5b upon binding to CNOT4N are mapped onto the structure of UbcH5b. The color schemes of the chemical-shift changes are as follows: red, >0.2 ppm; pink, 0.15–0.2 ppm; orange, 0.1–0.15 ppm; yellow, 0.075–0.1 ppm; green, 0.05–0.075 ppm. (D) Schematic drawing of UbcH5b to illustrate the relative positions of key residues. The inter-residue distances are between Cα atoms.