UBE2L3, a Partner of MuRF1/TRIM63, Is Involved in the Degradation of Myofibrillar Actin and Myosin

Figure 1. E2L3 knockdown preserved a-actin and MHCII in myofibrillar fraction in catabolic C2C12 myotubes. (a) A mixture of 11 recombinant E2 enzymes (E2 mix w/o E2L3) was used to address the selectivity of E2L3 antibody. A total of 1 μg of each E2 enzyme was used. Only E2L3 was recognized by the antibody used. (b) E2L3 is mainly present in the soluble fraction of C2C12 myotubes. C2C12-soluble and myofibrillar-enriched fraction were assayed for E2L3, GAPDH and alpha-tubulin (a-tubulin) by immunoblotting (IB) (20 and 40 μg of total proteins were loaded). GAPDH was used as marker of soluble-enriched fraction and a-tubulin as loading control. (c) Myofibrillar proteins were assayed for a-actin and MHCIIa levels by immunoblotting (15 μg of total proteins were loaded). C2C12 myotubes were cultured in 24-well plates, treated with Dex 1 μM and electroporated, as explained in Section 2, with two shRNA directed against E2L3 or MuRF1. (d) Densitometry analysis was performed on data from (c). Values were expressed after normalization with the loading control PCNA. Values are means SEM for n = 4 to 6 per group; p-values are indicated within the graph.

Figure 2. E2L3 KD and overexpression modify myofiber cross-sectional area (CSA). Mice treated or not with Dexamethasone (Dex 5 mg/kg/day) were transfected with either a scramble shRNA (scr-shRNA) targeting no known coding sequence, a mixture of two shRNAs (siE2L3, knockdown) directed against E2L3 or a plasmid encoding for E2L3 (E2L3, overexpression). Sh-encoding plasmids also encoded for emGFP, which allowed us to estimate the efficiency of transfection (typically around 50%; see Supplementary Materials Figure S2). Tibialis anterior muscle cross-sections were labeled with anti-laminin-α1 for determining CSA (see Section 2 for details). Four different zones and a minimum of 380 fibers were used for each muscle. (a) Representative cross-sections of T. anterior from each group are shown. Scale bar, 50 µm. (b) CSA analysis from n = 6 mice and a total of 17,945 fibers were analyzed, using a one-way ANOVA. Fibers distribution was analyzed within each muscle by grouping CSAs by 500 µm2 steps. A global shift towards lower CSA was observed when E2L3 was overexpressed in muscles. By contrast, larger CSA were observed in E2L3 KD muscles when compared to controls; n = 6 mice per group. * Statistically…

Figure 3. MuRF1–E2L3 affinity determination, using photobleaching rate. (a) Principle of the Monolith (adapted from Reference [50]). A fluorescent protein in solution in capillaries is excited with a UV LED; the fluorescence is recorded by a photomultiplier (PM) and plotted against time to get an MST (MicroScale-Thermophoresis) trace chart. (b) The fluorescent NT647–MuRF1 was mixed with the putative partner subjected to a 16-sample serial dilution. Variation of MuRF1 fluorescence was followed over time, in the presence of E2s; E2L3 (red), an interacting E2, and E2E1 (orange), a non-interacting E2 in this condition. When ligand binding changes the photobleaching properties of the fluorescent target, which is visible by the decrease in the initial fluorescence (IF) zone of the graph (t = −5 to 0 sec, before the Infra-Red (IR) laser was switched on), photobleaching rate has to be used for calculation of the affinity, instead of MST. (c) Dose–response curves are expressed in bleaching rate and plotted against E2s concentrations. (d) Transformation of dose–response curves from (c), in fraction bound for easier comparison between curves. (e) Dataset overview; n, number of replicates; NM, not…

Figure 4. MuRF1 has a high affinity for filamentous actin and heavy meromyosin. (a) Scheme of actin poly- and depolymerization. Actin can be present either as a free monomer (G-actin) or as a linear polymer microfilament called F-actin (filamentous). Monomeric actin polymerizes at >2 mM K+ or Na+, and at >0.05 mM Mg2+. The (+) end of the actin filament tends to capture ATP-actin, thus promoting polarization, while, at the (−) end, the actin filament is mostly composed of ADP-actin that promotes depolymerization. A stationary equilibrium is achieved when the G-actin monomers are exchanged at both ends of the microfilament without any change to its total length. (b) Interaction of MuRF1 with F-a-actin during F-a-actin depolymerization. Capillaries containing MuRF1 and F-actin in G buffer were read immediately (T = 0 min) or stored for 1 h 30 or 24 h before reading, leading to the depolymerization of actin). (c) Dose–response curves for MuRF1 vs. F-actin in F buffer and vs. G-actin in G buffer. (d) Dose–response curves for MuRF 1 vs. F-actin in F buffer, vs. G-actin in G buffer and vs. heavy meromyosin (HMM, HMM includes the first 800 amino acids of the MHC, plus the two pairs of light…

Figure 5. Impact of substrates on MuRF1–E2 interaction. (a) Dose–response curves for MuRF1–E2L3 interaction in presence or absence of substrates, a-actin or HMM (myosin). (b) Dataset overview of the photobleaching experiments performed with E2L3 in (a). (c) Dose–response curves for MuRF1–E2E1 interaction in presence or absence of substrates. (d) Dataset overview of the photobleaching experiment performed with E2E1 in (c). (e) Affinity change based on data presented in (a–d). (f) Impact of MHCIIa (residues 1–255) on the interaction between MuRF1 and E2L3, E2E1 or E2D2, determined by yeast-three-hybrid (Y3H) experiments. Diploid yeast colonies were replicated on selection plates (medium lacking leucine, tryptophan and histidine and supplemented with 20 mM Aureobasidin A and 2.5 mM 3-Amino-1,2,4-triazole), allowing the readout of interaction occurrence. Four independent transformations were performed, and 10 to 16 colonies were analyzed for each interaction studied. The strong interactions MuRF1–MuRF3 and p53–Large-T served as positive controls. Representative results are shown; two different “E2L3 + MuRF1/MHC” clones are presented to illustrate that MHC impacted the growth of 50% of the…