**3. Results**

#### *3.1. MRCC I Inhibition Led to Cytotoxicity in L02 Hepatocytes*

Glutamate/malate (Glu/Mal), known as the substrates of MRCC I, can directly initiate the mitochondrial main respiratory chain (NADH respiratory chain) by activating MRCC I. The inhibitor of MRCC I is rotenone (ROT). We first evaluated whether changes in MRCC I activity would a ffect cytotoxicity. The L02 cells were treated with di fferent concentrations of Glu/Mal (0, 5/5, 10/10 mM; 1 h) and ROT (0, 2.5, 5 μM; 24 h). As shown in Figure 1A, the application of substrates Glu/Mal slightly inhibited ROS generation, while the application of the inhibitor ROT significantly enhanced ROS generation in a concentration-dependent manner. In most cases, the degree of ALT and AST increase was consistent with the degree of hepatocyte damage, which is the most commonly used indicator of hepatocytes/liver function. The distribution of these two enzymes in hepatocytes is di fferent. ALT is mainly distributed in the cytoplasm, and the increase of ALT leakage indicates the membrane damage. AST is mainly distributed in both cytoplasm and the mitochondria, and the increase of ALT indicates that hepatocytes are damaged to organelle level. Compared with the control group, the treatment of Glu/Mal did not alter the leakage of ALT/AST to the culture medium, while the application of ROT increased the leakage of these two enzymes (especially ALT) in a concentration-dependent manner (Figure 1B). We further examined whether the alteration of MRCC I activity would a ffect cell growth and proliferation. As shown in Figure 1C, from day 2 and compared with control, Glu/Mal increased while ROT significantly decreased cell number, indicating that Glu/Mal stimulated while ROT suppressed cell growth and proliferation. Caspase-3 activity was inhibited by Glu/Mal and increased by ROT (Figure 1D). ROT treatment significantly increased the percentage (%) of apoptosis cells in a dose-dependent manner, while Glu/Mal exposure showed no e ffect on apoptosis (Figure 1E).

**Figure 1.** MRCC I inhibition led to cytotoxicity in L02 hepatocytes. (**A**) The L02 hepatocytes were either exposed to different concentrations of MRCC I substrates Glu/Mal (0, 5/5, 10/10 mM) for 1 h and then cultured for 23 h, or exposed to different concentrations of ROT (0, 2.5, 5 μM) for 24 h. ROS level was determined using the fluorescent probe DCFH-DA. (**B**) AST/ALT level was determined using the related kits. (**C**) The L02 hepatocytes were cultured for consecutive 5 days and treated with various concentrations of Glu/Mal (0, 5/5, 10/10 mM; 1 h) and ROT (0, 2.5, 5 μM; 24 h) in day 0 and day 3. The cell number, which indicated growth and proliferation of the hepatocytes, was recorded using a hemocytometer by trypan blue exclusion method. (**D**) Caspase-3 activity was examined using the commercial colorimetric assay kit. (**E**) The cell apoptosis was determined using commercial Annexin V-FITC Apoptosis Detection Kit. \* *p* < 0.05, compared with the control group.

#### *3.2. Cr(VI) Induced Mitochondria-Related Cytotoxicity*

It is well known that mPTP and MMP are important parameters reflecting mitochondrial function. The cells were exposed to various concentrations of Cr(VI) (0, 8, 16 μM) for 24 h. Treatment with Cr(VI) caused a marked increase in mPTP opening rate (Figure 2A) and a decrease in MMP (Figure 2B) in a concentration-dependent manner, suggesting the occurrence of mitochondrial damage. The mtDNA copy number and mitochondrial mass, which might be altered during mitochondrial damage, were also detected. mtDNA encodes 13 proteins which closely related to mitochondrial function. We designed the primers for two genes encoded by mtDNA, MRCC I subunit, ND1, and MRCC IV subunit, COX4I1. The expression amount of the two genes can reflect the copy number of mtDNA. As shown in Figure 2C, both ND1 and COX4I1 mRNA levels were decreased after Cr(VI) treatment, indicating the decline of mtDNA copy number. Cr(VI) exposure also decreased mitochondrial mass (Figure 2D). Mitochondria are energy providers, generating more than 95% of the ATP required for cells. Mitochondrial damage results in the decline of the intracellular ATP level, as confirmed in Figure 2E. The mPTP opening marks the emergence of irreversible point of apoptosis, which is accompanied by the release of apoptosis-inducing factors such as AIF and Cyt C from mitochondria to the cytoplasm. Cr(VI) increased the protein levels of AIF and Cyt C (Figure 2F). As a major member of cysteinyl aspartate-specific protease (caspase) family, caspase-3 plays a vital role in apoptosis. Both the protein expression (Figure 2F) and activity (Figure 2G) of caspase-3 were enhanced by Cr(VI) treatment. Cr(VI) also increased the % of apoptosis cells in a concentration-dependent manner (Figure 2H).

**Figure 2.** Cr(VI) induced mitochondrial-related cytotoxicity. L02 hepatocytes were treated with different concentrations of Cr(VI) (0, 8, 16 μM) for 24 h. (**A**) The mPTP opening was examined using the commercial kit. (**B**) The MMP was detected using JC-1. (**C**) The mRNA levels of ND1 and COX4I1, which could reflect the copy number of mtDNA were detected using quantitative real-time polymerase chain reaction (qRT-PCR). (**D**) Mitochondria mass was examined using Mito-Tracker Green by flow cytometer. (**E**) The ATP level was examined using a luciferase-based luminescence enhanced ATP assay kit. (**F**) AIF, Cyt C, and caspase-3 protein expressions were detected using Western blotting analysis. The protein bands were quantitated using Image J software. (**G**) Caspase-3 activity was determined using the commercial colorimetric assay kit. (**H**) The cell apoptosis was determined using commercial Annexin V-FITC Apoptosis Detection Kit. \* *p* < 0.05, compared with the control group.

#### *3.3. Cr(VI) Targeted MRCC I to Induce Cytotoxicity*

We have confirmed that MRCC I inhibition led to cytotoxicity in L02 hepatocytes. Next, we verified whether Cr(VI) targeted MRCC I to induce cytotoxicity. L02 hepatocytes were treated with different concentrations of Cr(VI) (0, 8, 16 μM) for 24 h. As shown in Figure 3A, among the five complexes, the activity of MRCC I was significantly decreased while the activity of MRCC II was slightly decreased after Cr(VI) exposure. The protein expression of MRCC I subunit ND1 was also decreased by Cr(VI) in a concentration-dependent manner (Figure 3B). For the combination treatments, L02 hepatocytes were exposed to Cr(VI) (16 μM) or PBS for 24 h with or without the pretreatment of Glu/Mal (10/10 mM) for 1 h; the cells were exposed to Cr(VI) (16 μM) or PBS with or without the cotreatment of ROT (5 μM) for 24 h. The pretreatment of Glu/Mal alleviated Cr(VI)-induced increase of ALT/AST leakage, while the cotreatment of ROT significantly aggravated Cr(VI)-induced increase of ALT/AST leakage (Figure 3C). We then detected the effect of MRCC I on Cr(VI)-induced caspase-3 activation and apoptosis induction. As shown in Figure 3D,E, the pretreatment of Glu/Mal significantly alleviated Cr(VI)-induced caspase-3

activation and apoptosis induction, while the cotreatment of ROT obviously aggravated Cr(VI)-induced caspase-3 activation and apoptosis induction. The inhibiting effect of ROT on MRCC I was more obvious than the enhancing effect of Glu/Mal.

**Figure 3.** Cr(VI) targeted MRCC I to induce cytotoxicity. (**A**) L02 hepatocytes were treated with different concentrations of Cr(VI) (0, 8, 16 μM) for 24 h. The activities of MRCC I-V were determined spectrophotometrically using the commercial kits. (**B**) The protein expression of ND1 was determined using Western blotting. (**C**) L02 hepatocytes were exposed to Cr(VI) (16 μM) or PBS for 24 h with or without the pretreatment of Glu/Mal (10/10 mM) for 1 h; or the cells were exposed to Cr(VI) (16 μM) or PBS with or without the cotreatment of ROT (5 μM) for 24 h. AST/ALT level was determined using the commercial kits. (**D**) Caspase-3 activity was determined using the commercial colorimetric assay kit. (**E**) The cell apoptosis was determined using commercial Annexin V-FITC Apoptosis Detection Kit. \* *p* < 0.05, compared with the control group. # *p* < 0.05, compared with the Cr(VI) alone treatment group.

#### *3.4. Cr(VI) Induced Mitochondrial Hyper-Fission via Interfering with Drp1*

Drp1, a major protein regulating mitochondrial fission in mammals, acts as an important intrinsic factor involved in mitochondria-dependent apoptosis. We then examined whether Drp1 and Drp1-related mitochondrial fission were involved in Cr(VI)-induced apoptosis. L02 hepatocytes were treated with different concentrations of Cr(VI) (0, 8, 16 μM) for 24 h. As shown in Figure 4A, the mitochondria of the control group revealed the significant large, tubular network structure, while the mitochondria of the Cr(VI)-exposed group showed the short-shaped, divided, and segmented structure, indicating that Cr(VI) induced excessive mitochondrial fragmentation. The % of cells without elongated mitochondria was also shown. Drp1 mRNA expression levels were also increased after Cr(VI) exposure in a concentration-dependent manner, indicating that Cr(VI) caused the transcriptional alteration of Drp1 (Figure 4B). Cr(VI) up-regulated the protein expressions of both the total and mitochondrial Drp1, suggesting that in addition to inducing the translational change of Drp1, Cr(VI) also triggered the translocation of Drp1 from cytoplasm to mitochondria (Figure 4C). The mitochondrial Drp1 translocation was then observed with laser confocal microscopy. As shown in Figure 4D, Drp1 protein with a green dot-like distribution was dramatically increased after Cr(VI) treatment, and the translocation of the increased Drp1 protein to mitochondria was clearly observed.

**Figure 4.** Cr(VI) caused mitochondrial fragmentation. L02 hepatocytes were treated with different concentrations of Cr(VI) (0, 8, 16 μM) for 24 h. (**A**) Mitochondrial morphology was determined using Mitotracker Red by confocal microscope. The % of cells without elongated mitochondria was calculated. (**B**) Drp1 mRNA expression was determined using qRT-PCR. (**C**) Total and mitochondrial Drp1 protein expressions were detected using Western blotting analysis. Voltage-dependent anion channel 1 (VDAC1) served as the loading control of mitochondrial protein. (**D**) Drp1 mitochondrial translocation was observed under a confocal microscope. \* *p* < 0.05, compared with the control group.

We then explored the effect of Drp1-siRNA on Cr(VI)-induced fragmentation. We constructed Drp1-siRNA plasmid and confirmed its efficiency using qRT-PCR (Figure 5A) and Western blotting (Figure 5B). The L-02 hepatocytes transfected with Drp1-siRNA and its control (Con-si) were treated with Cr(VI) (16 μM) or PBS for 24 h. While the significantly short-shaped, divided, and segmented mitochondria were observed in the Cr(VI) exposure group, Drp1-siRNA alleviated the excessive mitochondrial fragmentation and partially restored the large and tubular network structure of mitochondria (Figure 5C).

**Figure 5.** Cr(VI) induced mitochondrial hyper-fission via interfering with Drp1. Drp1 siRNA plasmid was constructed and verified using qRT-PCR (**A**) and Western blotting (**B**). (**C**) The L-02 hepatocytes transfected with Drp1-siRNA and its control (Con-si) were treated with Cr(VI) (16 μM) or PBS for 24 h. Mitochondrial morphology was determined using Mitotracker Red by confocal microscope. The % of cells without elongated mitochondria was calculated. \* *p* < 0.05, compared with the control group. # *p* < 0.05, compared with the Con-si plus Cr(VI) treatment group.

#### *3.5. Drp1-Mediated Mitochondrial Fragmentation Contributes to Cr(VI)-Induced MRCC I-Dependent Cytotoxicity*

The effect of Drp1-siRNA on Cr(VI)-induced MRCC I-dependent cytotoxicity was further explored. The L-02 hepatocytes transfected with Drp1-siRNA and Con-si were treated with Cr(VI) (16 μM) or PBS for 24 h. The application of Drp1-siRNA alleviated Cr(VI)-induced ALT/AST leakage (Figure 6A), caspase-3 activation (Figure 6B), and apoptosis induction (Figure 6C). For the combination treatments, the hepatocytes transfected with Drp1-siRNA and Con-si were exposed to Cr(VI) (16 μM) for 24 h with or without the pretreatment of Glu/Mal (10/10 mM) for 1 h; the hepatocytes transfected with Drp1-siRNA and Con-si were exposed to Cr(VI) (16 μM) with or without the cotreatment of ROT (5 μM) for 24 h. As shown in Figure 6D,F, the ALT/AST leakage, caspase-3 activation, and apoptosis induction in Drp1-si plus Cr(VI) treatment group was alleviated by the application of Glu/Mal, and aggravated by the application of ROT. Drp1 siRNA can further promote the inhibition of Glu/Mal on Cr(VI)-induced cytotoxicity, and also can further alleviate the aggravation of ROT on Cr(VI)-induced cytotoxicity.

**Figure 6.** Drp1-mediated mitochondrial fragmentation contributes to Cr(VI)-induced MRCC I-dependent cytotoxicity. (**A**) The L-02 hepatocytes transfected with Drp1-siRNA and Con-si were treated with Cr(VI) (16 μM) or PBS for 24 h. AST/ALT level was determined using the commercial kits. (**B**) Caspase-3 activity was determined using the commercial colorimetric assay kit. (**C**) The cell apoptosis was determined using commercial Annexin V-FITC Apoptosis Detection Kit. (**D**) For the combination treatments, the hepatocytes transfected with Drp1-siRNA and Con-si were exposed to Cr(VI) (16 μM) for 24 h with or without the pretreatment of Glu/Mal (10/10 mM) for 1 h; the hepatocytes transfected with Drp1-siRNA and Con-si were exposed to Cr(VI) (16 μM) with or without the cotreatment of ROT (5 μM) for 24 h. AST/ALT level was determined using the commercial kits. (**E**) Caspase-3 activity was determined using the commercial colorimetric assay kit. (**F**) The cell apoptosis was determined using commercial Annexin V-FITC Apoptosis Detection Kit. \* *p* < 0.05, compared with its relative control. # *p* < 0.05, compared with the Drp1-si plus Cr(VI) treatment group.

#### *3.6. Cr(VI)-Induced Drp1 Modulation was Depend on MRCC I Inhibition-Mediated ROS Production*

Evidence suggested that oxidative stress could activate Drp1 and promote mitochondrial fragmentation, thus we also explored whether Cr(VI)-induced accumulation of ROS could regulate Drp1. L02 hepatocytes were treated with different concentrations of Cr(VI) (0, 8, 16 μM) for 24 h. As shown in Figure 7A, Cr(VI) caused a significant increase of intracellular ROS levels in a concentration-dependent manner. The hepatocytes were exposed to Cr(VI) or PBS for 24 h with or without the pretreatment of Glu/Mal; or the cells were exposed to Cr(VI) or PBS with or without the cotreatment of ROT. Glu/Mal slightly decreased ROS level compared with the control. Cr(VI)-induced over-production of ROS was alleviated by the application of Glu/Mal, and aggravated by the application of ROT (Figure 7B), indicating that MRCC I was involved in ROS accumulation induced by Cr(VI). The hepatocytes were exposed to Cr(VI) or PBS for 24 h with or without the pretreatment of NAC for 1 h. NAC partially decreased Cr(VI)-induced increase of the mRNA level (Figure 7C), and both the total and

mitochondrial protein level (Figure 7D) of Drp1, confirming that Drp1 can be regulated by ROS in both transcription and translation levels. The utilization of Glu/Mal also alleviated the increase of both total and mitochondrial Drp1 expression (Figure 7E). The above results together suggested that Cr(VI)-induced Drp1 modulation was dependent on MRCC I inhibition-mediated ROS production.

**Figure 7.** Cr(VI)-induced Drp1 modulation was dependent on MRCC I inhibition-mediated ROS production. (**A**) L02 hepatocytes were treated with different concentrations of Cr(VI) (0, 8, 16 μM) for 24 h. ROS level was determined using the fluorescent probe DCFH-DA. (**B**) L02 hepatocytes were exposed to Cr(VI) (16 μM) or PBS for 24 h with or without the pretreatment of Glu/Mal (10/10 mM) for 1 h; or the cells were exposed to Cr(VI) (16 μM) or PBS with or without the cotreatment of ROT (5 μM) for 24 h. ROS level was determined using the fluorescent probe DCFH-DA. (**C**) The hepatocytes were exposed to Cr(VI) (16 μM) or PBS for 24 h with or without the pretreatment of NAC (5 mM) for 1 h. Drp1 mRNA expression was determined using qRT-PCR. (**D**,**E**) Total and mitochondrial Drp1 protein expressions were detected using Western blotting analysis. \* *p* < 0.05, compared with the control group. # *p* < 0.05, compared with the Cr(VI) alone treatment group.
