**3. Results**

#### *3.1. ANT1 KD Increases Cellular Proliferation Without A*ff*ecting Cell Viability*

Transfection with *ANT1* siRNA significantly reduced ANT1 expression by 37% (*P* < 0.001) 48 h after transfection (Figure 1A). Interestingly, we found that *ANT1* KD increases cell number by 32% (*P* < 0.001, Figure 1B) and the number of alive cells by 22% (*P* < 0.05, Figure 1C) without affecting cell viability (Figure 1D). Altogether, these results sugges<sup>t</sup> that *ANT1* KD does not affect cell viability, but it increases cellular proliferation, possibly as an adaptive response to ANT1 downregulation.

**Figure 1.** Cell viability is not affected by *ANT* KD in H9c2 cells. (**A**) Protein levels of ANT1 in negative control (NC) and *ANT1* KD cells. *Top panel:* representative immunoblots. *Bottom panel:* quantitative data of ANT1 protein expression normalized to ATP5a (a mitochondrial housekeeping protein); (**B**) Total number of cells 48 h after transfection. *Top panel:* representative images of cells. *Bottom panel:* quantitative data of cells; (**C**) total number of live cells 48 h after transfection; (**D**) cell viability 48 h after transfection calculated as (alive cells/total cells) × 100. \* *P* < 0.05 and \*\* *P* < 0.001 vs. NC. Data represent 4–7 independent experiments.

#### *3.2. ANT1 KD Increased Total ATP and ROS Levels with no E*ff*ect on the ETC Activity and mtROS Production*

Total ATP levels were elevated by 36% (*P* < 0.01) in *ANT1* KD cells (Figure 2A). Total ATP levels were normalized to μg of total cellular protein to account for the observed increase in cell number (Figure 1B). However, it should be noted that this method does not distinguish glycolytic from mitochondrial ATP. Although these cells appear to have higher levels of total ATP, a decrease in ΔΨm could hint towards a higher glycolytic ATP production. Results showed that *ANT1* KD cells had significantly lower ΔΨm compared to NC cells (Figure 2B), suggesting that the elevated ATP levels might result from increased glycolysis but not OXPHOS.

Next, we examined total ROS production using H2DCFDA fluorescent dye. Results demonstrate that *ANT1* KD cells have a 38% increase (*P* < 0.001) in total ROS generation when compared to NC (Figure 2C). Analysis of mtROS production by MitoSOX shows that *ANT1* KD has no effect on mtROS (Figure 2D). The lack of a difference in mtROS levels can be explained with no significant electron leakage due to low ETC flow in *ANT1* KD cells. In favor of this, analysis of the enzymatic activity of the ETC complexes I, II, III, and IV demonstrated no difference between *ANT1* KD and control cells (Figure 2E–H). Our data are consistent with previous studies where the activity of complexes I, III, and IV were unaffected by ANT expression in HEK293 cells [8]. Altogether, these data demonstrate that ANT1 deficiency has no effect on the enzymatic activity of individual ETC complexes and mtROS production.

#### *3.3. Mitochondrial Oxygen Consumption Rate and OXPHOS is not A*ff*ected by ANT1 Downregulation*

We measured mitochondrial oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) in H9c2 cells treated with scrambled and *ANT1* siRNA using the Seahorse XFe24 analyzer. Results demonstrate that *ANT1* silencing does not affect the OCR and ECAR in these cells (Figure 3A,B). Likewise, basal and FCCP-induced maximal respiration rates were found unchanged in *ANT1* KD cells (Figure 3C,D). As expected, mitochondrial ATP production was not affected by ANT1 downregulation (Figure 3E). Altogether, these results demonstrate that *ANT1* silencing does not affect OXPHOS in H9c2 cells.

#### *3.4. ANT1 KD in H9c2 Cells Induce RCS Dissociation: the Role of Acetylation*

Our recent studies [12] demonstrated that pharmacological inhibition of ANT by atractyloside induces RCS dissociation in isolated cardiac mitochondria. Analysis of RCS in scrambled (NC) or *ANT1* siRNA-treated H9c2 cells demonstrated that ANT1 deficiency induces disassembly of respirasome by 9% (*P* < 0.01) compared to control cells (Figure 4A,B), suggesting that ANT is involved in RCS integrity and stabilization. In order to validate that the decrease in RCS was due to ANT1 downregulation but not ΔΨm loss (Figure 2B), mitochondria isolated from rat hearts were treated with 1 μM FCCP, an uncoupler, for 45 min [12], prior to RCS analysis. Results demonstrate that loss of ΔΨm does not affect RCS integrity (*P* < 0.1473, Figure 4C,D). Taken together, these results sugges<sup>t</sup> that the loss of RCS in *ANT1* KD cells is not due to the loss of ΔΨm and may result from the downregulation of ANT1.

In the following set of experiments, we examined whether ANT1 acetylation is involved in RCS formation. First, we analyzed liver mitochondria isolated from WT and *SIRT3* KO mice to determine the acetylation of total mitochondrial proteins. SIRT3 is the main mitochondrial isoform of sirtuins. Hyperacetylation of mitochondrial proteins due to SIRT3 deficiency has been shown to associate with cardiovascular, neurodegenerative diseases, diabetes, and aging [17–19]. We have previously demonstrated that SIRT3 ablation enhances lysine acetylation (Ac-K) of mitochondrial proteins [20]. However, immunoprecipitation analysis revealed no changes in ANT1 acetylation in the mitochondria of *SIRT3* KO mice (Figure 5A). Interestingly, the mitochondria of *SIRT3* KO mice demonstrated lower RCS levels compared to the WT group (Figure 5B,C). These results sugges<sup>t</sup> that acetylation of mitochondrial proteins, but not ANT1, can stimulate RCS disassembly in mitochondria.

**Figure 2.** *ANT1* KD disturbs mitochondrial membrane potential (ΔΨm) without affecting enzymatic activity of ETC complexes. (**A**) Total cellular ATP levels normalized to the total amount of protein. (**B**) Mitochondrial membrane potential determined with JC-1 after transfection and calculated as the ratio of J-aggregates to JC-1 monomers. (**C**) Total cellular ROS assessed with H2DCFDA; (**D**) MtROS assessed using MitoSOX red. Data on the fluorescence activity in the cells (**B**–**D**) are presented as percent change of negative control (NC). (**E**–**H**) The enzymatic activity of complexes I, II, III and IV in mitochondria isolated from NC and *ANT1* KD cells. Data were normalized to mitochondrial protein levels. \* *P* < 0.05, and \*\* *P* < 0.01 vs. control (NC). Data represent 3 independent experiments.

**Figure 3.** Mitochondrial oxygen consumption and ATP production is not affected by ANT1 downregulation. (**A**) oxygen consumption rate (OCR); (**B**) extracellular acidification rate (ECAR); (**C**) basal respiration; (**D**) maximal respiration; (**E**) mitochondrial ATP production. All parameters were determined using the Seahorse XFe24 analyzer (Agilent) after the addition of (in μM): 0.5 oligomycin (Oligo), 4 FCCP, and 0.5 rotenone/antimycin A (Rot/AntA). The data was extracted using the Seahorse XFe24 report generator and normalized to total protein levels. Data represent 3 independent experiments.

**Figure 4.** *ANT1* KD stimulates mitochondrial respirasome disintegration in H9c2 cells. (**A**) representative blue native (BN) gel of mitochondria isolated from *ANT1* KD cells and subjected to BN-PAGE; (**B**) quantitative data of respirasome levels; (**C**) representative BN-PAGE gel of mitochondria isolated from rat heart and treated with vehicle (Veh, 0.01% DMSO), 500 nM rotenone (Rot), 500 nM antimycin A (Ant A), or 1 μM FCCP; (**D**) quantitative data of respirasome levels in the groups shown in C; (**E**) representative two-dimensional BN-PAGE of ETC complexes in mitochondria isolated from the rat heart. RCS were analyzed in mitochondria where membrane proteins were solubilized using digitonin and separated by BN-PAGE. ETC complexes were visualized using specific antibodies against the subunits for complexes I (NDUFB8), II (SDHB), III (UQCRC2), IV (MTCO1), and V (ATP5A). Respirasome is shown as I+III+IV. Data in B and D were normalized to mitochondrial protein levels and presented as percent change of negative control (NC) or control (Con). \* *P* < 0.05 and \*\* *P* < 0.01 vs. NC or Con. Data represent 3–4 independent experiments.

**Figure 5.** ANT1 is not acetylated but acetylation of mitochondrial proteins stimulates RCS disassembly in *SIRT3* KO mitochondria. (**A**) Immunoprecipitation (IP) of liver mitochondrial proteins with acetylated lysine (Ac-K) antibodies followed by immunoblotting (IB) against ANT1. *Input:* the sample before IP; *Supernatant:* sample that did not bind to Ac-K antibodies; (**B**) BN-PAGE gel of liver mitochondria isolated from WT and *SIRT3* KO− mice. (**C**) Quantitative data of respirasome levels. The data were normalized to total protein levels and presented as percent change from the WT group. \* *P* < 0.01 vs. WT; n = 6–7 animals per group.
