Ubc9 Attenuates Myocardial Ischemic Injury Through Accelerating Autophagic Flux

Aims SUMOylation is a post-translational modification that plays a crucial role in the cellular stress response. We aimed to demonstrate whether and how the SUMO E2 conjugation enzyme Ubc9 affects acute myocardial ischemic (MI) injury. Methods and Results Adenovirus expressing Ubc9 was administrated by multipoint injection in the border zone of heart immediately after MI in C57BL/6 mice. Neonatal rat cardiomyocytes (NRCMs) were also infected, followed by oxygen and glucose deprivation (OGD). In vivo, Ubc9 adenovirus-injected mice showed decreased cardiomyocyte apoptosis, reduced myocardial fibrosis, and improved cardiac function post-MI. In vitro, overexpression of Ubc9 decreased cardiomyocyte apoptosis, whereas silence of Ubc9 showed the opposite results during OGD. We next found that Ubc9 significantly decreased the accumulation of autophagy marker p62/SQSTM, while the LC3 II level hardly changed. When in the presence of bafilomycin A1 (BAF), the Ubc9 adenovirus plus OGD group presented a higher level of LC3 II and GFP-LC3 puncta than the OGD group. Moreover, the Ubc9 adenovirus group displayed increased numbers of yellow plus red puncta and a rising ratio of red to yellow puncta on the mRFP-GFP-LC3 fluorescence assay, indicating that Ubc9 induces an acceleration of autophagic flux from activation to degradation. Mechanistically, Ubc9 upregulated SUMOylation of the core proteins Vps34 and Beclin1 in the class III phosphatidylinositol 3-kinase (PI3K-III) complexes and boosted the protein assembly of PI3K-III complex I and II under OGD. Moreover, the colocalization of Vps34 with autophagosome marker LC3 or lysosome marker Lamp1 was augmented after Ubc9 overexpression, indicating a positive effect of Ubc9-boosted protein assembly of the PI3K-III complexes on autophagic flux enhancement. Conclusions We uncovered a novel role of Ubc9 in protecting cardiomyocytes from ischemic stress via Ubc9-induced SUMOylation, leading to increased PI3K-III complex assembly and autophagy-positioning. These findings may indicate a potential therapeutic target, Ubc9, for treatment of myocardial ischemia.


Figure S1
Figure S1. Transfer efficiency assessments of Ubc9 overexpression and silence (A) Ubc9 adenovirus carrying GFP was multiple injected in the heart (0.5×10 9 pfu per heart), the heart was removed at day 5 post-MI. Frozen sections were prepared. Slices position of heart pattern, and the fluorescence intensity of GFP was observed under a fluorescence microscope and photographed, the photographs were arranged from the ligation point to the apical region. (B) The hearts were multiple injected by Ubc9 adenovirus, and then removed at day 5, 7 and 14 post-MI. Western blotting analysis of Ubc9, n=4-5 per group. (C, D) NRCMs were infected with Adv-GFP 25 moi, and Adv-Ubc9 transfected with 10, 20, 50 and 100 moi for 48 h, respectively, and then maintained in hypoxic condition for 6 h. The fluorescence intensity of GFP was observed under fluorescence microscope and photographed; Western blotting analysis of Ubc9 and Cleaved caspase3, n=3-4 per group. (E) Divided Adv-Null or Adv-Ubc9 into three volumes: 0.05ul, 0.2ul, 0.5ul, and transferred them in NRCMs, culturing with 1 ml DMEM added 10% FBS for 48 h and then maintained in hypoxic condition for 6 h. Western blotting analysis of Ubc9 and Cleaved caspase3. (F) Ubc9-siRNA-1 and Ubc9-siRNA-2 were transfected in NRCMs 48 h. Western blotting analysis of Ubc9. Data were analyzed by one-way ANOVA, followed by a Bonferroni post-hoc test.   NRCMs were pre-treated with 3-MA (5 mM) or CQ (10 μM) for 12 h, or BAF (50 nM) for 2 h, and then underwent OGD for 6 h. Western blotting analysis of Ubc9, Cleaved caspase3 and autophagy marker LC3Ⅱ and p62, n=3 per group. Data were analyzed by one-way ANOVA, followed by a Bonferroni post-hoc test.  Adv-Ubc9 or Adv-GFP was multipoint injected around the ligature of heart in the C57BL/6 mice immediately after MI. Western blotting analysis of globally SUMO-1 and SUMO-2/3 conjugated protein in the border area at day 5 post-MI, n=5 per group. (E, F) Western blotting analysis of globally SUMO-1 and SUMO-2/3 conjugated protein of NRCMs at different time points post-OGD, n=5 per group. (G, H) Adv-Ubc9 was transfected in NRCMs 48 h, then maintained in hypoxic condition for 6 h. Western blotting analysis of globally SUMO-1 and SUMO-2/3 conjugated protein in NRCMs under OGD 6 h, n=3-5 per group. Data were analyzed by one-way ANOVA, followed by a Bonferroni post-hoc test.
Table S1. Applying the SUMOsp 2.0 software to predict SUMO-modified site numbers of the autophagy related-proteins Selecting the target proteins (*) by the characteristics of having large numbers of SUMO-modified sites or having small numbers of SUMO-modified sites while being classic in autophagic flux.

Echocardiography
Transthoracic echocardiography was manipulated with a Visual Sonics (Vevo 2100; Visual Sonics Inc., Ontario, Canada) equipped with a 25-MHz imaging transducer on 14 days before sacrificing the mice. The anesthesia of the mice was kept with 2 % isoflurane gas with an inflow rate of 0.5-1.5 ml/min during the echocardiographic examination. The mice body temperature was monitored at 36-38 ℃ using a rectal thermometer, and the heart rate was maintained at 350-450 beats/min.

The left ventricle (LV) was analyzed under parasternal long-axis and
short-axis views, and then the output values of mice heart such as EF (ejection fraction), FS (fractional shortening), LVIDd (LV internal diameter at end-diastole) and LVIDs (LV internal diameter at end-systolic) were calculated according to the guidelines of the Vevo 2100.

TUNEL staining
At 5 days post-MI, adult male C57/B6 mice were euthanized by an overdose of sodium pentobarbital (100 mg/kg). Cell survival in ischemic area was determined by TUNEL staining on frozen sections (5 μm) or in vitro neonatal rat cardiomyocytes (NRCMs) as per manufacturer's instructions (Cell death detection assay, Roche, Indianapolis, IN).
Cardiomyocytes were stained by α-actinin, and the total number of nuclei were stained by DAPI. In vivo, apoptotic cells were collected under fluorescence microscope (Nikon T1); In vitro, five fields were captured randomly per sample under confocal microscope (Nikon American Inc., Melville, NY).

Masson's trichrome staining
On day 14 after MI, adult male C57/B6 mice were euthanized by an overdose of sodium pentobarbital (100 mg/kg). The fibrosis of myocardial tissue was determined by Masson's trichrome staining on frozen sections (5 μm). The percentage of fibrosis was analyzed by Image Pro-Plus software (Media Cybernetics Inc., Bethesda, MD, USA).

Cell culture and treatment
NRCMs were isolated as described previously. 19 Briefly, hearts from 1 to 2-day-old Sprague-Dawley rats were dissected and digested with 0.05% trypsin overnight at 4℃, then digested with Collagenase Type II for 45 minutes. cells were plated in 30mm Petri dishes and cultured in DMEM supplemented with 15% fetal bovine serum and maintained at 37℃ with 5% CO 2 in a humidified chamber for 24 h, then cells changed into DMEM supplemented with 10% fetal bovine serum and cultured for 6-8 h. Subsequently, adenovirus or siRNA were delivered for 6-12 h, and then the media was replaced by fresh DMEM including 10% fetal bovine Finally, the cardiomyocytes were harvested and analyzed.

Ubc9 gene transfer in cardiomyocytes
In order to overexpress Ubc9, adenovirus harboring Ubc9 were infected in NRCMs as follows: Adv-Ubc9 carrying GFP: Cells were infected with Adv-GFP at 25 multiplicity of infection (moi), or Adv-Ubc9 carrying GFP were transfected with 10, 20, 50 and 100 moi for 12 h individually. The culture medium was changed into fresh one. The cells were cultured for another 42 h before OGD treatment. Transfer efficiency of adenovirus were detected by confocal microscopy and Western blotting ( Figure S1B and   1C). Adv-Ubc9 was successfully transfected at 20, 50 and 100 moi, and the efficiency was almost the same at 50 and 100 moi, so we chose 50 moi to do the subsequent experiments.
Adv-Ubc9 not carrying GFP: In order to avoid the interreference of the green fluorescence, Adv-Ubc9 not carrying GFP was used. Adv-Null (1.53×10 13 vp/ml) or Adv-Ubc9 (1.53×10 13 vp/ml) was delivered to cardiomyocytes as three volumes respectively: 0.05ul, 0.2ul, 0.5ul per 1ml cultural medium. The details of infection and OGD treatment were same as that used in Adv-Ubc9 carrying GFP. The transfer efficiency was confirmed by protein levels of Ubc9 and Cleaved-caspase3. Western blotting assay showed that Ubc9 was over-expressed successfully at these three volumes respectively, but Cleaved-caspase3 was down-regulated after transfected Adv-Ubc9 at the volume of 0.05 ul. In addition, cells appeared in the same status and had none obvious damage after transfecting Adv-Null or Adv-Ubc9 at the volume of 0.05 ul. Therefore, we used the concentration of 0.05 ul/ml in the subsequent experiments ( Figure S1D). anti-RAB7), the binding state were detected by using 20×LumiGLO®

Western blot analysis
Reagent and 20×Peroxide , molecular band intensity was determined by ImageJ.

Flow cytometry
Flow cytometry was applied to detect Annexin V-APC positive apoptotic cell. The apoptotic cell was detected by the Annexin V-APC/7AAD Apoptosis Detection Kit, according to the instruction of manufacture. Briefly, the cells were stained by Annexin V-APC and 7AAD. Both early and late apoptotic cells were sorted by BD AccuriC6.
Cell apoptosis was reflected by the percentage of Annexin V-APC positive cell.