Trim-Away ubiquitinates and degrades lysine-less and N-terminally acetylated substrates

Abstract

TRIM proteins are the largest family of E3 ligases in mammals. They include the intracellular antibody receptor TRIM21, which is responsible for mediating targeted protein degradation during Trim-Away. Despite their importance, the ubiquitination mechanism of TRIM ligases has remained elusive. Here we show that while Trim-Away activation results in ubiquitination of both ligase and substrate, ligase ubiquitination is not required for substrate degradation. N-terminal TRIM21 RING ubiquitination by the E2 Ube2W can be inhibited by N-terminal acetylation, but this doesn't prevent substrate ubiquitination nor degradation. Instead, uncoupling ligase and substrate degradation prevents ligase recycling and extends functional persistence in cells. Further, Trim-Away degrades substrates irrespective of whether they contain lysines or are N-terminally acetylated, which may explain the ability of TRIM21 to counteract fast-evolving pathogens and degrade diverse substrates.

Fig. 1. Dimeric Ube2W and RING clustering is required for ligase autoubiquitination.

a 2.25 Å X-ray structure of TRIM21 RING (blue) in complex with Ube2W (pink). b Close-up of the E2:E3 interface. c, d Structural models of a RING:Ube2W~Ub complex with Ub in an open (c) or closed (d) conformation respectively. Based on superposition with the Ub-RING:Ube2N~Ub:Ube2V2 structure 7BBD. In the closed conformation (d; boxed), Ub K11 makes a potential salt bridge with TRIM21 E13 and buries I44 at an interface with the E2. e Ube2W-mediated TRIM21 RING mono-ubiquitination assay using 2 µM R-NbGFP and 1 µM Ube2W with WT, I44A or K11E ubiquitin. Representative example from n = 2 independent experiments. f Schematic of antibody-induced recruitment of either two RINGs or two RING dimers. Only the latter satisfies the ‘two-plus-one’ model for RING autoubiquitination^,. g Ube2W-mediated mono-ubiquitination assay using 10 µM R-PS or R-R-PS and 0.25 µM Ube2W WT or monomeric V30K/D67K. Representative example from n = 2 independent experiments. h Ubiquitination of 100 nM R-PS or R-R-PS in the absence or presence of equimolar anti-GFP antibody. Ube2W was titrated (25, 50, 100, 200 nM). Representative example from n = 3 independent experiments. i Quantification of monoubiquitination from (h). Graph shows mean and s.e.m. from n = 3 independent experiments. Statistical significance based on two-tailed Student’s t test (two-tailed). Source data are provided as a Source Data file.

Fig. 2. Ligase N-terminal ubiquitination is not required for substrate degradation and antiviral activity.

a Schematic showing N-terminal acetylation of TRIM21 by AcCoA and NAT and then incubation of the acetylated or unmodified ligase with ubiquitin (Ub) and Ube2W in a ubiquitination reaction. b Protein-stained gel of ubiquitination reaction depicted in (a) using R-R-PS ligase. Monoubiquitination of R-R-PS that has been incubated with AcCoA and NAT is inhibited. Representative example from n = 3 independent experiments. c Schematic showing Trim-Away experiment in which antibody is electroporated into cells together with Ac-R-R-PS. Once inside cells, a ternary complex with the substrate is formed. If degradation is driven by ligase N-terminal ubiquitination, then Ac-R-R-PS activity will be inhibited. d Results of Trim-Away experiment described in (c). RPE-1 CAV1-mEGFP cells were electroporated with PBS or anti-GFP antibody ± R-R-PS proteins and CAV1-mEGFP fluorescence was quantified using the IncuCyte system. Time shows hours (h) post-electroporation. Values normalized to PBS control condition. Graphs shows mean and s.e.m. from n = 4 technical replicates. Representative example from n = 2 independent experiments. Note that there is CAV1-mEGFP degradation with anti-GFP alone due to the presence of endogenous cellular TRIM21. e Schematic showing electroporation of Ac-R-R-PS into cells followed by infection with Adv5 in the presence of anti-hexon antibody 9C12. If ligase N-terminal ubiquitination is necessary for TRIM21 antiviral function, neutralization of infection and immune signaling will be inhibited. f, g Neutralization of AdV5 infection by increasing 9C12 concentrations (f) and AdV5-9C12-induced NFκB activation (g) in HEK293T TRIM21 KO cells infected immediately after electroporation with PBS or R-R-PS ± acetylation. 9C12^H433A does not bind TRIM21 PRYSPRY. Graphs shows mean and s.e.m. from n = 3 independent experiments. Black dots in (g) show individual data points. Statistical significance between R-R-PS and Ac-R-R-PS is based on two-way ANOVA (f; significance is represented with labels ns (not significant, P > 0.05), *** (P ≤ 0.001)) and two-tailed Student’s t test (g). Source data are provided as a Source Data file.

Fig. 3. N-terminal ubiquitination regulates ligase turnover in cells.

a Schematic showing electroporation of R-R-PS ± N-terminal acetylation and proteasome-dependent degradation. b Western blot of experiment depicted in (a) 1 h post-electroporation. R-R-PS protein levels are rescued by addition of 10 µM proteasome inhibitor epoxomycin. Ac-R-R-PS protein persists in cells irrespective of proteasome inhibition. Representative example from n = 2 independent experiments. c Schematic showing electroporation of R-R-PS ± acetylation into cells, followed by delayed Trim-Away or Adv5 neutralization assays. d For the delayed Trim-Away assay, mRNA encoding the antibody construct (NbGFP-Fc) responsible for recruiting R-R-PS to substrate (CAV1-mEGFP) was co-electroporated into NIH3T3 CAV1-mEGFP cells with PBS or R-R-PS ± acetylation; Trim-away is delayed for ~2 h until NbGFP-Fc protein is translated. Graph shows mean and s.e.m. from n = 4 technical replicates of CAV1-mEGFP fluorescence quantified using the IncuCyte system. Values normalized to PBS control condition (no NbGFP-Fc). Time shows hours (h) post-electroporation. Representative example from n = 2 independent experiments. Note that NIH3T3 cells do not contain endogenous TRIM21 and expression of NbGFP-Fc in the absence of TRIM21 activity leads to CAV1-mGFP stabilization. e For the delayed Adv5 neutralization assay, HEK293T TRIM21 KO cells were infected with AdV5 ± 9C12 2 h post-electroporation of PBS or R-R-PS ± acetylation. Graph shows mean and s.e.m. from n = 3 independent experiments. Statistical significance between R-R-PS and Ac- R-R-PS is based on two-way ANOVA and represented with labels ns (not significant, P > 0.05), * (P ≤ 0.05), ** (P ≤ 0.01), *** (P ≤ 0.001). See also Supplementary Fig. 3. Source data are provided as a Source Data file.

Fig. 4. In vitro Trim-Away reveals direct substrate ubiquitination.

a Schematic showing an in vitro ubiquitination reaction in which ligase (R-R-PS), substrate (GFP) and anti-GFP antibody are incubated together with E2 enzymes Ube2W and Ube2N/Ube2V2 to promote either mono- or polyubiquitination. b Western blot of experiment described in (a). Top panel is blotted for GFP, middle panel for IgG and lower panel for TRIM21. Depending on the E2s present, monoubiquitinated species or a higher molecular weight smear indicative of polyubiquitin are observed. Representative example from n = 2 independent experiments. c Western blot of experiment similar to (b) but comparing R-R-PS to Ac-R-R-PS. Note that while R-R-PS ubiquitinates itself, antibody heavy chain and substrate, Ac-R-R-PS only ubiquitinates antibody and substrate. Representative example from n = 2 independent experiments. Source data are provided as a Source Data file. See also Supplementary Fig. 4.

Fig. 5. Substrate ubiquitination parallels substrate degradation during Trim-Away in cells.

a Schematic showing Trim-Away experiment in which antibodies are electroporated into cells expressing endogenous TRIM21. Ubiquitination and degradation are then monitored in the presence or absence of proteasome inhibitor MG132. b, c RPE-1 cells were electroporated with PBS or anti-ERK1 antibody (b) or anti-IKKα antibody (c) and whole cell lysates harvested at the indicated times after electroporation for immunoblotting. Short exposures show degradation of substrates. Long exposures reveal substrate ubiquitination followed by degradation of ubiquitinated species. d RPE-1 cells were electroporated with PBS or anti-IKKα antibody ± MG132 and whole cell lysates harvested 1 h post-electroporation for immunoblotting. e RPE-1 TRIM21 KO cells reconstituted with TRIM21-HA (T21-HA) or empty vector (EV) were electroporated with PBS or anti-ERK1 antibody ± MG132 and whole cell lysates harvested 1 h post-electroporation for immunoblotting. Representative examples (b–e) from n = 3 independent experiments. Source data are provided as a Source Data file. See also Supplementary Fig. 5.

Fig. 6. Trim-Away ubiquitinates and degrades a lysine-less substrate.

a Schematic showing R-NbGFP construct and substitutions to remove all lysines. b, c RPE-1 CAV1-mEGFP-Halo cells were electroporated with PBS or R-NbGFP ± lysines ± N-terminal acetylation. b CAV1-mEGFP-Halo fluorescence was quantified using the IncuCyte system. Time shows hours (h) post-electroporation. Values normalized to PBS control condition. Graphs shows mean and s.e.m. from n = 4 technical replicates. c Whole cell lysates were harvested 3 h post-electroporation for immunoblotting. Representative examples (b, c) from n = 3 independent experiments. d Schematic showing a model lysine-less substrate consisting of NbGFP (with lysine substitutions) fused to 12 copies of the naturally lysine-less ALFAtag epitope. e Scheme showing Trim-Away experiment in which the anti-ALFAtag antibody (NbALFA-Fc) is electroporated into cells expressing endogenous TRIM21 and the lysine-less substrate (12xALFAtag-NbGFP^0K). f, g RPE-1 WT or TRIM21 KO cells expressing either substrate with lysines (12xALFAtag-NbGFP) or without lysines (12xALFAtag-NbGFP^0K) were electroporated with PBS or NbALFA-Fc and whole cell lysates harvested 8 h post-electroporation for capillary-based immunoblotting. Lane view (f) and quantification (g) of substrate and TRIM21 protein levels normalized to PBS condition. Graphs show mean from n = 2 independent experiments (black dots). Note that binding of NbALFA-Fc in the absence of TRIM21 causes stabilization of substrate. h RPE-1 cells expressing either substrate with lysines (12xALFAtag-NbGFP) or without lysines (12xALFAtag-NbGFP^0K) were electroporated with PBS or NbALFA-Fc and whole cell lysates harvested at the indicated times after electroporation for immunoblotting. Short exposures show degradation of substrates. Long exposures reveal substrate ubiquitination followed by degradation of ubiquitinated species. Representative example from n = 2 independent experiments. Single stars (*) show 8x- and 4x-ALFAtag-NbGFP species. Source data are provided as a Source Data file.

Fig. 7. Trim-Away independently of lysine and N-terminal ubiquitination.

a Schematic of completely lysine-less Trim-Away assay. The substrate constitutes NbGFP with 12 copies of the ALFAtag epitope. The ligase constitutes the TRIM21 RING fused to NbALFA (R-NbALFA). The substitutions necessary to remove lysines from each domain in the assay are shown. Note that the ALFAtag and HA epitopes are naturally lysine-less. The 12xALFAtag allows clustering of multiple R-NbALFA molecules, triggering RING activation. b–e RPE-1 TRIM21 KO cells expressing either substrate with lysines (12xALFAtag-NbGFP) or without lysines (12xALFAtag-NbGFP^0K) were electroporated with water (control) or mRNA encoding the indicated constructs (b, c) or PBS (control) or R^0K-NbALFA^0K-HA protein ± acetylation (d, e) and whole cell lysates harvested 8 h (b, c) or 3 h (d,e) post-electroporation for capillary-based immunoblotting. Lane view (b, d) and quantification (c, e) of substrate protein levels normalized to control condition. Graphs shows mean (c) or mean and s.e.m. (e) from n = 2 (c) or n = 3 (e) independent experiments (black dots). Statistical significance from control condition is based on two-way ANOVA and represented with label **** (P ≤ 0.0001). Note that binding of NbALFA alone causes stabilization of substrate. f RPE-1 TRIM21 KO cells expressing either substrate with lysines (12xALFAtag-NbGFP) or without lysines (12xALFAtag-NbGFP^0K) were electroporated with ligase R-NbALFA ± lysines in the presence of MG132 and whole cell lysates harvested 30 min post-electroporation for immunoblotting. Single stars (*) show 8x- and 4x-ALFAtag-NbGFP species. Representative example from n = 2 independent experiments. g Lysine-less substrate 8xALFAtag-NbALFA^0K with (Ac) or without (ctrl) N-terminal acetylation was incubated with ligase R-NbALFA and Ube2W in a ubiquitination reaction. Representative example from n = 2 independent experiments. h RPE-1 TRIM21 KO cells were electroporated with substrate 8xALFAtag-NbGFP^0K ± acetylation and 4 h later electroporated with PBS or ligase R-NbALFA protein and whole cell lysates harvested at the indicated times after electroporation for immunoblotting. Short exposures show degradation of substrates. Long exposures reveal substrate ubiquitination followed by degradation of ubiquitinated species. Representative example from n = 2 independent experiments. Source data are provided as a Source Data file.