Fig 2: BNIP3 expression in 786-O, A498, and ACHN cells following treatment with TSA. (A) Expression of BNIP3 mRNA in 786-O, ACHN and A498 cells was quantified by RT-qPCR after treatment with TSA for 24 h. The untreated controls (blank) for each of the three cell lines were assigned a relative value of 1. *P<0.05 compared with treatment groups (aP<0.05 compared with TSA 1.0 and 2.0 µmol/l treatment groups for ACHN cells; bP<0.05 compared with TSA 1.0 and 2.0 µmol/l treatment groups for A498 cells). (B) BNIP3 protein levels in 786-O, ACHN, and A498 cells following TSA treatment for 48 h were examined by western blotting. GAPDH was used as an internal control, and cells cultured without TSA were considered as untreated controls (Blank). *P<0.05 compared with treatment groups (aP<0.05 compared with the TSA 1.0 µmol/l treatment group). Representative bands are presented under the histograms. TSA, trichostatin A; RT-qPCR, reverse transcription-quantitative polymerase chain reaction.

Fig 3: The BNIP3 C-terminus can be phosphorylated.(A) Western blot detection of immunoprecipitated BNIP3 using an a-PKA substrate antibody specific to the phosphorylated RRXS/T sequence, located at the BNIP3 C-terminus and adjacent to the transmembrane (TM) domain. Results shown for 4 cell types: (left to right) HEK 293 cells expressing exogenous BNIP3 (dimer, 60 kD), and A549, MDA-MD-231, and AU565 cells expressing endogenous BNIP3 (monomer, 30kD). Lane 1 of each Western blot contains the whole cell lysate (WCL). (B) LC-MS/MS analysis of BNIP3 phosphorylation in HEK 293 cells with normal or elevated cAMP (8-Bromo-cAMP), showing peptide coverage (gray) and phosphorylation sites (red). The TM domain is underlined. (C) Table of BNIP3 phosphopeptides identified by LC-MS/MS, showing the percent probability, ion charge, actual and observed masses, and mass error (Da and ppm) for each peptide. Peptides shown are from analysis of BNIP3 purified from HEK 293 cells with elevated cAMP levels. (D) Schematic of the BNIP3 protein sequence, showing each C-terminal mutation representing phosphomimetic or nonphosphorylated BNIP3. (E) Expression of BNIP3 phosphomutants from stable doxycycline-inducible HEK 293 Tet On cells, treated with doxycycline (Dox) for 48 hr. (F) Subcellular localization of BNIP3 phosphomutants, showing Western blot of cytosolic and mitochondrial fractions. (G) Alkaline extraction of mitochondria-associated proteins, showing alkaline extract of the mitochondrial pellet and the alkaline-resistant mitochondrial pellet from cells expressing each BNIP3 phosphomutant. All blots are representative of at least 3 independent experiments.

Fig 4: Opposing effects of Mcl-1 on nonselective autophagy and selective mitochondrial autophagy (mitophagy). Mcl-1 at the mitochondria suppresses formation of autophagosomes during nutrient deprivation possibly to protect healthy mitochondria from degradation. In contrast, Mcl-1 promotes autophagy of dysfunctional mitochondria by functioning as a mitophagy receptor or by coordinating with Bnip3. Created with BioRender.com (accessed on 21 March 2022).

Fig 5: ROS are the upstream mediators of the pseudo-hypoxic response occurring after ATGL overexpression. HeLa cells were transfected with ATGL plasmid for 48 h and treated with 5 mM NAC 24 h before the end of the experiment. (a) HIF1a and BNIP3 levels were analyzed by Western blot analysis. Band intensity is indicated below the corresponding band and expressed as fold-change relative to CTRL. The images are representative of three independent experiments that gave similar results. ß-Actin and ATGL were used as loading and transfection control, respectively. (b) RT-qPCR analysis of BNIP3 levels. ACTB was used as reference control. Data are shown as fold change (n = 3; * p < 0.05 as indicated). (c) The evaluation of extracellular lactate content was performed on cell culture media collected after 48 h of ATGL over-expression by spectrophotometric analysis. Data are expressed as mean ± SD (n = 3; * p < 0.05 as indicated). (d) The proliferation was assayed by Trypan blue direct cell counting procedure. Data are expressed as mean ± SD (n = 3; * p < 0.05 vs. ATGL + NAC). (e) Cells were treated with Cycloheximide (CHX, 10 µg/mL) for 2, 4, 6 h and cobalt chloride (CoCl2, 150 µM) for 6 h before the end of experiment and HIF1a protein levels were evaluated by Western blot analysis. Band intensity is indicated below the corresponding band and expressed as fold-change relative to CTRL for each condition. The images are representative of three independent experiments that gave similar results. ß-Actin and ATGL were used as loading and transfection control, respectively. (f) HeLa cells were treated with MG132 (2 µM) for 8 h before the end of the experiment and Western blot analysis of HIF1a and ubiquitinated proteins was determined. Band intensity is indicated below the corresponding band and expressed as fold-change relative to CTRL. The images are representative of three independent experiments that gave similar results. ß-Actin and ATGL were used as loading and transfection control, respectively. HeLa cells were transfected as previously described and were treated with 5 mM NAC 24 h before the end of the experiment and (g) RT-qPCR analysis of HIF1a levels was performed. ACTB was used as reference control. Data are shown as fold change (n = 3; * p < 0.05 as indicated)