Fig 2: Knocking down USP5 reduces the steady state of a model MLP.A) HeLa TRex Flp-In cells stably expressing OP91 after induction, were seeded at 50% confluence and transfected with 1 nM of siRNAi targeting the indicated DUBs. OP91 expression was induced 48 h later and cells grown for 20 hrs post induction. Total cell lysates were prepared, and products analysed by Western blotting with antibodies against OP91, SGTA, USP5, USP9X, USP14 and tubulin (loading control) using fluorescence- based detection (LICOR). The % knockdown efficiency is indicated for each DUB in lane 2, 3 and 4. B) OP91 signals were quantified using Odyssey 2.1 software and normalised to the tubulin loading control, values show standard errors for n = 3. Pairwise comparisons relative to the NT control, were performed using GraphPad Prism. Student t-test: P > 0.05 = ns, P ≤ 0.05 = *, P ≤ 0.01 = **, P ≤ 0.001 = *** C) HeLa TRex Flp-In cells stably expressing OP91 were transfected with non-targeting siRNAi or two independent siRNAi duplexes targeting USP5. Cells were treated as described in A, followed by blotting for OP91, USP5 and tubulin. D) OP91 signals were quantified using Odyssey ® Fc Imaging system and normalised to the tubulin loading control, values show standard errors for n = 3. Pairwise comparisons relative to the NT control, were performed using GraphPad Prism. Student t-test: P > 0.05 = ns, P ≤ 0.05 = *, P ≤ 0.01 = **, P ≤ 0.001 = *** and P ≤ 0.0001 = ****.

Fig 3: Overexpression of USP5 stabilises OP91.A) HeLa T-REx Flp-In cells stably expressing OP91 upon induction were transfected with V5-tagged Pex19 and SGTA or HA -tagged WT or C335A mutant of USP5. OP91 expression was induced 6 hours post-transfection. Cells were grown for a further 20 hours, post-induction. Total cell lysates were prepared and analysed by western blotting using antibodies against OP91, V5 tagged SGTA and Pex19, HA tagged USP5 and USP5C335A and tubulin (loading control). B) OP91 signals were quantified using Odyssey ® Fc Imaging System and normalised to the tubulin loading control, values show standard errors for n = 3. C) HeLa T-REx Flp-In cells stably expressing OP91 upon induction were transfected with plasmids encoding V5-tagged Pex19 and SGTA or HA -tagged WT USP5 or USP5C335A. OP91 expression was induced by addition of medium containing 1 μg/ml of tetracycline and grown for 20 hrs post induction. Prior to harvesting transfected cells were treated with 100 μg/ml cycloheximide (Sigma, Aldrich) to inhibit protein synthesis. Cells were lysed directly into SDS-PAGE sample buffer at indicated time-points followed by Western blotting with antibodies against opsin (OP91), HA (USP5 and USP5C335A), V5 (SGTA and Pex19) and tubulin (loading control) using fluorescence- based detection (LICOR). D) OP91 signal for each condition was quantified and plotted against the respective time points.

Fig 4: USP5 specifically enhances SGTA activity.A) HeLa T-REx Flp-In cells stably expressing OP91 after induction were seeded at 50% confluence and transfected with 1 nM of siRNAi targeting USP5 or a nontargeting control and incubated for 48 hours. After this, plasmids expressing V5 tagged SGTA or PEX19 were transfected followed by incubation for 6 hours before the medium was replaced with medium containing 1 mg/ml of tetracycline to induce OP91 expression. Cells were grown for 20 hrs post induction and total cell lysate were prepared and products analysed by western blotting using antibodies for opsin (OP91), V5 (SGTA and PEX19), and tubulin (loading control). B) OP91 signals were quantified using Odyssey ® Fc Imaging System and normalised to the tubulin loading control, values show standard errors for n = 3. Pairwise comparisons relative to Pex19, were performed using GraphPad Prism. Student t-test: P > 0.05 = ns, P ≤ 0.05 = *, P ≤ 0.01 = **, P ≤ 0.001 = *** and P ≤ 0.0001 = ****. C) HeLa T-REx Flp-In cells were treated with siRNAi targeting USP5, USP9X, USP14 and a nontargeting control and analysed as detailed in A. D) OP91 signals were quantified using Odyssey ® Fc Imaging System and normalised to the tubulin loading control, values show standard errors for n = 3. Pairwise comparisons relative to Pex19, were performed using GraphPad Prism. Student t-test: P > 0.05 = ns, P ≤ 0.05 = *, P ≤ 0.01 = **, P ≤ 0.001 = *** and P ≤ 0.0001 = ****.

Fig 5: Loss of USP5 does not affect other substrates.A) HeLa M cells were seeded at 50% confluence and transfected with 1 nM of siRNAi targeting the indicated DUBs. Ub-MGFP and Ub-RGFP were transfected 48 h later and cells grown for 20 hrs post-transfection. Total cell lysates were prepared and analysed by Western blotting using GFP antibodies and tubulin (loading control). B) Ub-RGFP signals were quantified using the Odyssey ® Fc imaging system and normalised to the tubulin loading control, values show standard errors for n = 3. Pairwise comparisons relative to the NT control, were performed using GraphPad Prism. Student t-test: P > 0.05 = ns, P ≤ 0.05 = *, P ≤ 0.01 = **, P ≤ 0.001 = ***. C) HeLa TRex Flp-In cells stably expressing OpD after induction were seeded at 50% confluence and transfected with 1 nM of siRNAi targeting the indicated DUBs. OpD expression was induced 48 h later, and cells grown for 20 hrs post induction. Total cell lysates were prepared and analysed by Western blotting using opsin antibodies to detect OpD, and tubulin (loading control). D) OP91 signals were quantified using Odyssey ® Fc imaging system and normalised to the tubulin loading control, values show standard errors for n = 3. Pairwise comparisons relative to the NT control, were performed using GraphPad Prism. Student t-test: P > 0.05 = ns, P ≤ 0.05 = *, P ≤ 0.01 = **, P ≤ 0.001 = *** and P ≤ 0.0001 = ****.