Fumonisin B2 Induces Mitochondrial Stress and Mitophagy in Human Embryonic Kidney (Hek293) Cells—A Preliminary Study

Ubiquitous soil fungi parasitise agricultural commodities and produce mycotoxins. Fumonisin B2 (FB2), the structural analogue of the commonly studied Fumonisin B1 (FB1), is a neglected mycotoxin produced by several Fusarium species. Mycotoxins are known for inducing toxicity via mitochondrial stress alluding to mitochondrial degradation (mitophagy). These processes involve inter-related pathways that are regulated by proteins related to SIRT3 and Nrf2. This study aimed to investigate mitochondrial stress responses in human kidney (Hek293) cells exposed to FB2 for 24 h. Cell viability was assessed via the methylthiazol tetrazolium (MTT) assay, and the half-maximal inhibitory concentration (IC50 = 317.4 µmol/L) was estimated using statistical software. Reactive oxygen species (ROS; H2DCFDA), mitochondrial membrane depolarisation (JC1-mitoscreen) and adenosine triphosphate (ATP; luminometry) levels were evaluated to assess mitochondrial integrity. The relative expression of mitochondrial stress response proteins (SIRT3, pNrf2, LONP1, PINK1, p62 and HSP60) was determined by Western blot. Transcript levels of SIRT3, PINK1 and miR-27b were assessed using quantitative PCR (qPCR). FB2 reduced ATP production (p = 0.0040), increased mitochondrial stress marker HSP60 (p = 0.0140) and suppressed upregulation of mitochondrial stress response proteins SIRT3 (p = 0.0026) and LONP1 (p = 0.5934). FB2 promoted mitophagy via upregulation of pNrf2 (p = 0.0008), PINK1 (p = 0.0014) and p62 (p < 0.0001) protein expression. FB2 also suppressed miR-27b expression (p < 0.0001), further promoting the occurrence of mitophagy. Overall, the findings suggest that FB2 increases mitochondrial stress and promotes mitophagy in Hek293 cells.


Introduction
Favourable weather conditions cause ubiquitous soil fungi to parasitise agricultural commodities and produce mycotoxins [1]. Ingestion of mycotoxin-contaminated food can induce adverse effects on the health of the consumer. Mycotoxin consumption is associated with acute and chronic toxicity in both humans and animals [1,2].
Fumonisin B 2 (FB 2 ) is the structural analogue of the popularly studied fumonisin B 1 (FB 1 ). Fumonisins are highly prevalent toxins produced mainly by the Fusarium species,

FB2 Increased ROS Production and Mitochondrial Membrane Depolarisation in Hek293 Cells
Compromised mitochondrial function may result in exacerbated ROS production that further induces mitotoxicity. ROS production increased significantly (p = 0.0156) following exposure with FB2 ( Figure 2A). Mitochondrial functionality can be observed by quantifying mitochondrial membrane depolarisation. FB2 caused a significant increase in mitochondrial membrane depolarisation in Hek293 cells (p < 0.0001) ( Figure 2B).

FB2 Induces Mitochondrial Stress in Hek293 Cells
We next determined the effects of FB2 on mitochondrial stress. To determine the effect of FB2 on mitochondrial stress, ATP quantification was carried out (mitochondrial output and functionality), and HSP60 protein expression (a marker for mitochondrial stress) was assessed via Western blots [22]. A significant decrease (p = 0.0040) in ATP production compared to the control was observed ( Figure 3A). Furthermore, FB2 induced a considerable

FB 2 Increased ROS Production and Mitochondrial Membrane Depolarisation in Hek293 Cells
Compromised mitochondrial function may result in exacerbated ROS production that further induces mitotoxicity. ROS production increased significantly (p = 0.0156) following exposure with FB 2 (Figure 2A). Mitochondrial functionality can be observed by quantifying mitochondrial membrane depolarisation. FB2 caused a significant increase in mitochondrial membrane depolarisation in Hek293 cells (p < 0.0001) ( Figure 2B).

FB2 Increased ROS Production and Mitochondrial Membrane Depolarisation in Hek293 Cells
Compromised mitochondrial function may result in exacerbated ROS production that further induces mitotoxicity. ROS production increased significantly (p = 0.0156) following exposure with FB2 ( Figure 2A). Mitochondrial functionality can be observed by quantifying mitochondrial membrane depolarisation. FB2 caused a significant increase in mitochondrial membrane depolarisation in Hek293 cells (p < 0.0001) ( Figure 2B). FB2 increased ROS production and mitochondrial membrane depolarisation in Hek293 cells. ROS production was significantly increased by FB2 ((A); * p < 0.05) with a corresponding increase in mitochondrial membrane depolarisation ((B); *** p < 0.0001).

FB2 Induces Mitochondrial Stress in Hek293 Cells
We next determined the effects of FB2 on mitochondrial stress. To determine the effect of FB2 on mitochondrial stress, ATP quantification was carried out (mitochondrial output and functionality), and HSP60 protein expression (a marker for mitochondrial stress) was assessed via Western blots [22]. A significant decrease (p = 0.0040) in ATP production compared to the control was observed ( Figure 3A). Furthermore, FB2 induced a considerable . FB 2 increased ROS production and mitochondrial membrane depolarisation in Hek293 cells. ROS production was significantly increased by FB 2 ((A); * p < 0.05) with a corresponding increase in mitochondrial membrane depolarisation ((B); *** p < 0.0001).

FB 2 Induces Mitochondrial Stress in Hek293 Cells
We next determined the effects of FB 2 on mitochondrial stress. To determine the effect of FB 2 on mitochondrial stress, ATP quantification was carried out (mitochondrial output and functionality), and HSP60 protein expression (a marker for mitochondrial stress) was assessed via Western blots [22]. A significant decrease (p = 0.0040) in ATP production compared to the control was observed ( Figure 3A). Furthermore, FB 2 induced a considerable increase (p = 0.0140) in HSP60 protein expression, suggesting increased mitochondrial stress ( Figure 3B).

FB2 Suppresses Mitochondrial Stress Responses in Hek293 Cells
To confirm the induction of mitochondrial stress, protein (Western blots) and mRN (qPCR) levels of SIRT3 were analysed. Additionally, protein expression of LONP1 wa measured. FB2 induced a significant decrease in SIRT3 mRNA expression (p < 0.0001) (

FB2 Suppresses Mitochondrial Stress Responses in Hek293 Cells
To confirm the induction of mitochondrial stress, protein (Western blots) and mRNA (qPCR) levels of SIRT3 were analysed. Additionally, protein expression of LONP1 was measured. FB2 induced a significant decrease in SIRT3 mRNA expression (p < 0.0001) (Figure 4A) with coinciding decreases in SIRT3 protein expression (p = 0.0026) ( Figure 4B). No significant changes were observed for LONP1 protein expression (p = 0.5934) ( Figure 4C).

FB 2 Activates Nrf2 in Hek293 Cells
Phosphorylated Nrf2 (Ser40) (pNrf2) is the stable and activated form of Nrf2 that has dissociated from KEAP1, allowing it to translocate to the nucleus and transcribe for proteins [30,31]. Following FB 2 exposure, the expression of pNrf2 was significantly increased (p = 0.0008) ( Figure 5).

FB2 Activates Nrf2 in Hek293 Cells
Phosphorylated Nrf2 (Ser40) (pNrf2) is the stable and activated form of Nrf2 that has dissociated from KEAP1, allowing it to translocate to the nucleus and transcribe for proteins [30,31]. Following FB2 exposure, the expression of pNrf2 was significantly increased (p = 0.0008) ( Figure 5).

FB 2 induced Mitophagy in Hek293 Cells
Since Nrf2 transcriptionally regulates mitophagy, the effects of FB 2 on proteins and genes involved in the process were analysed. Post-transcriptional regulation of mitophagy proteins was analysed by measuring miR-27b expression. MiR-27b expression was suppressed following exposure to FB 2 (p < 0.0001). The miR-27b mimic showed no significant changes in miR-27b expression compared to the control, whereas the miR-27b inhibitor showed a significant decrease in miR-27b expression (p < 0.0001) ( Figure 6A). FB 2 increased PINK1 transcript (p < 0.0001) ( Figure 6B) and protein (p = 0.0014) ( Figure 6C) expression levels in Hek293 cells. FB 2 also increased expression of p62 ( Figure 6D) (p < 0.0001).

FB2 Activates Nrf2 in Hek293 Cells
Phosphorylated Nrf2 (Ser40) (pNrf2) is the stable and activated form of Nrf2 that has dissociated from KEAP1, allowing it to translocate to the nucleus and transcribe for proteins [30,31]. Following FB2 exposure, the expression of pNrf2 was significantly increased (p = 0.0008) ( Figure 5).

Discussion
The kidney is a primary target for fumonisin toxicity due to the accumulation and excretion of toxins via this organ [32,33]. However, limited biochemical studies exist demonstrating the effects of fumonisins in the human kidney . Unlike FB 2 , FB 1 (a structural analogue) has well-established mechanisms, including induction of mitochondrial toxicity [10,12]. Apart from the canonical mechanism of sphingolipid metabolism disruption [16], little is known about FB 2 -induced toxicity . To date, the effects of FB 2 on kidney cells and mitochondrial function have not been established. This study provides evidence that FB 2 can induce mitophagy by preventing mitochondrial stress responses from occurring in kidney cells. This is the first study, to our knowledge, that illustrates mitochondrial toxicity induced by FB 2. Furthermore, this study introduces a novel concept of FB 2 -induced post-transcriptional regulation (via miRNAs) of genes involved in mitophagy.
The MTT assay determined the cytotoxic potential of FB 2 ; FB 2 decreased cell viability in Hek293 cells ( Figure 1) and induced cell death (Appendix A). This suggests that FB 2 decreased NADH availability in cells; this altered the NAD + ratio in cells, further compromising cell function and metabolism [34]. More importantly, studies have shown that a decline in the availability of NAD + compromises mitochondrial function [35,36]. Therefore, the decrease in the NADH/NAD + ratio promotes mitochondrial dysfunction.
A common consequence of mitochondrial dysfunction is depolarisation of the mitochondrial membrane [10]. An increase in ROS production contributes significantly to both depolarisation and stress of the mitochondria (as the mitochondrial respiratory chain has various sites for ROS production) [37]. Fumonisins are known to induce toxicity through increased ROS production via electron transport chain (ETC) inhibition and depolarisation of the mitochondrial membrane [10,12].
We show that FB 2 can increase ROS production and cause depolarisation of the mitochondrial membrane in Hek293 cells (Figure 2). Elevations in ROS production are commonly attributed to aberrations in the mitochondrial ETC. Due to similarity in structure, FB 2 , like other fumonisins [10,12], possibly disrupts the ETC, resulting in excessive ROS production and ultimately depolarisation of the mitochondrial membranes ( Figure 2).
A crucial marker for compromised mitochondrial activity is reduced ATP synthesis and increased mitochondrial stress markers [38]. The significant reduction in ATP production ( Figure 3A) occurred due to increased mitochondrial stress and disturbances in the mitochondrial respiratory chain (mitochondrial dysfunction) post FB 2 exposure. Previous studies have shown a direct correlation in the upregulation of HSP60 expression in response to mitochondrial stress and ROS, making it a suitable biomarker for the phenomenon [22]. FB 2 significantly increased HSP60 protein expression ( Figure 3B). Additionally, the decrease in ATP promotes stress in the mitochondria, causing the upregulation of HSP60.
Increased mitochondrial stress induces an increase in the expression of mitochondrial sirtuins [39]. SIRT3 expression has been shown to increase following the induction of mitochondrial stress significantly [40]. Inhibition of SIRT3 has led to mitotoxicity and cell death due to an inadequate stress response. Fluctuations in SIRT3 expression can be considered a suitable biomarker for mitochondrial stress [41]. FB 2 significantly decreased both SIRT3 mRNA and protein expressions ( Figure 4A,B). SIRT3 (a class of NAD + -dependent deacetylases) is native to the mitochondria. SIRT3 expression and activity are dependent on the NADH/NAD + ratio in the cell [42]. A decrease in the NADH/NAD + ratio causes an increase in nicotinamide, an inhibitor of sirtuin function [43].
LONP1 reduces the effects of stress via the degradation of oxidatively damaged and misfolded proteins [20,44]. SIRT3 post-translationally regulates LONP1 via deacetylation. This causes a decrease in LONP1 protein expression when SIRT3 is elevated [45,46]. FB 2 did not alter LONP1 expression despite the downregulation of SIRT3 ( Figure 4C) and upregulation of ROS (Figure 2A). LONP1 (an ATP-dependent protease) contains a highly conserved ATPase domain with an AAA + module and a proteolytic domain with an Nterminal domain [20,44]. FB 2 depleted cellular ATP levels and consequently may have inhibited LONP1 catalytic activity and the degradation of oxidised proteins ( Figure 3A). These findings suggest that FB 2 prevents the upregulation of critical mitochondrial stress proteins necessary for the amelioration of stress.
Toxins 2022, 14, 171 7 of 14 FB 2 induced mitochondrial stress and aberrations in mitochondrial function. Next, the effects of FB 2 on Hek293 cell mitophagy were investigated. Mitophagy is a quality control process that allows for the degradation of damaged mitochondria, thus promoting homeostasis. PINK1 and p62 are critical proteins involved in the mitophagy process [19,25]. PINK1 is activated on the depolarised mitochondrial membrane allowing for the recruitment of other mitophagy proteins such as p62 [24,25,47]. Several promoters and inhibitors regulate the expression of PINK1 and p62.
A positive regulator of both PINK1 and p62 is the transcription factor Nrf2 [26,27]. Furthermore, p62 has been shown to increase the expression of Nrf2 [28]. Excessive ROS production activates phosphorylation pathways, triggering the phosphorylation of Nrf2 and promoting the translocation of pNrf2 to the nucleus, wherein PINK1 and p62 are transcribed [48]. The Nrf2-PINK1-p62 axis occurs as a means of cell survival [26,27]; however, excessive stimulation may result in cell death [49]. Figure 5 shows a significant increase in pNrf2, indicating mitophagy promotion. The finding is in agreement with excess ROS production ( Figure 2A) and increased stress.
Conversely, miRNAs have been implicated in the negative regulation of mitophagy. MiR-27b is a negative regulator of mitophagy as it can directly inhibit PINK1 expression via binding to the 3 -UTR of PINK1 mRNA [29]. FB 2 significantly reduced miR-27b expression in Hek293 cells ( Figure 6A). This coincided with the result obtained for the inhibitor of miR-27b, illustrating the toxins' ability to act as an inhibitor for miRNAs expression ( Figure 6A), thereby promoting mitophagy. Fumonisins have been shown to repress miR-27b expression [11]. However, FB 2 induced inhibition of the miRNA and promoted mitophagy in Hek293 cells. FB 2 induced a significant increase in the gene and protein expressions of PINK1 ( Figure 6B,C). This was accompanied by correlating increases in p62 protein expression ( Figure 6D). The data agree with data obtained for pNrf2 and miR-27b, suggesting that FB 2 promotes mitophagy at a transcriptional and translational level.
It can be deduced that the Nrf2-PINK1-p62 axis was positively regulated by FB 2 . The inhibition of miR-27b further supports that FB 2 promotes mitophagy in Hek293 cells; this is the first study to our knowledge to report the promotion of mitophagy by a fumonisin in kidney cells.
The occurrence of mitophagy post FB 2 exposure is in agreement with the suppressed mitochondrial stress responses, increased ROS, increased mitochondrial membrane depolarisation and compromised function of the organelle. However, mitophagy did not act as a cell survival mechanism as a significant reduction in cell viability was observed due to overstimulation of mitophagy, resulting in kidney cell death.
This study provides insight into the role of FB 2 -induced mitotoxicity in Hek293 cells. FB 2 increased ROS production, which increased mitochondrial stress and mitochondrial membrane depolarisation, dampened SIRT3 and LONP1 mitochondrial stress responses and promoted mitophagy. Furthermore, this study provides evidence of posttranscriptional regulation of PINK1 by miR-27b.

Future Recommendations and Limitations
The present study was performed to establish the toxicity of FB 2 at the estimated IC 50 as per other numerous fumonisin toxicology studies. In doing so, preliminary data surrounding toxicity were provided. However, future studies need to include varied concentrations of FB 2 and time periods to better understand FB2-induced toxicity. Furthermore, more cell lines that FB2 targets should be incorporated into studies. Additionally, future research can include comparative work between FB 1 and FB 2 . culture media and supplements were purchased from Lonza (Basel, Switzerland). Luminometry kits were purchased from Promega (Madison, WI, USA). Western blot reagents were purchased from Bio-Rad (Hercules, CA, USA). All other reagents were purchased from Merck (Darmstadt, Germany) unless otherwise stated.
A stock solution of 20 mM FB 2 was prepared in 0.1 M phosphate-buffered saline (PBS) and diluted using DMEM to achieve the concentrations for the MTT assay (0-500 µmol/L) and thereafter, for further experiments. All assays were performed three independent times and in triplicate.

Methyl Thiazol Tetrazolium (MTT) Assay
The cytotoxicity of FB 2 in Hek293 cells was determined using the MTT assay. Briefly, 20,000 cells/well were seeded and allowed to adhere overnight in a 96-well microtitre plate (37 • C, 5% CO 2 ). Thereafter, cells were incubated for 24 h with varying concentrations (0-500 µmol/L) of FB 2 . Control wells contained DMEM only. Following incubation, treatments were removed, cells were washed using 0.1 M PBS and incubated with MTT salt (20 µL; 5 mg/mL in 0.1 M PBS) and DMEM (100 µL) for 4 h. The MTT salt solution was then removed, and 100 µL of dimethyl sulphoxide (DMSO) was aliquoted per well and incubated at 37 • C for 1 h. Optical density was measured using a spectrophotometer (Bio-Tek uQuant Universal Microplate Spectrophotometer, Winoosiki, VT, USA) at 570 nm and a reference wavelength of 690 nm. Results were expressed as log concentration versus percentage cell viability.
The IC 50 was determined and used as the treatment concentration for all subsequent experiments. Non-linear regression analysis was used to estimate the IC 50 (GraphPad Prism v5.0). This was in accordance with other toxicology studies using fumonisins that only used the IC 50 for further testing [11,12]. All controls remained untreated (DMEM only) without the addition of FB 2.

ATP Assay
ATP concentration was measured using the CellTiter-Glo ® Luminescent Cell Viability Assay (Promega, #G7570). Cells were treated in 6-well plates for 24 h (37 • C; 5% CO 2 ). Following treatment, 20,000 cells/well in 0.1 M PBS were seeded in an opaque 96-well microtitre plate in triplicate. As per the manufacturer's instructions, the CellTiter-Glo ® Reagent was reconstituted, and 25 µL of reagent was added to each well. Plates were incubated in the dark for 20 min at room temperature, and luminescence was measured using a Modulus™ Microplate Reader (Turner Biosystems, Sunnyvale, CA, USA). Results were expressed as relative light units.

2 ,7 -Dichlorodihydrofluorescein Diacetate (H 2 DCFDA) Assay
ROS concentration was quantified using the DCF assay. Cells were treated in 6-well plates at 80% confluency for 24 h (37 • C, 5% CO 2 ). Thereafter, 50,000 cells were aliquoted in four separate micro-centrifuge tubes. A stock solution of 80 mM H 2 DCF-DA (Thermo-Fisher, Waltham, MA, USA) was diluted using PBS to produce a 5 µmol/L working solution; 100 µL of the working solution was added to each micro-centrifuge tube (37 • C; 30 min). Cells were washed with PBS and subsequently centrifuged (400× g; 10 min). PBS was removed, and the process was repeated. Cells were re-suspended in PBS (200 µL) and transferred to an opaque 96-well microtitre plate. Fluorescence was measured using a Modulus™ Microplate Reader (Turner Biosystems, Sunnyvale, CA, USA) with an excitation wavelength of 503 nm and an emission wavelength of 509 nm as per Arumugam  [12]. Results were expressed as relative fluorescent units (RFU) compared to the control.
Membranes were quenched using 5% hydrogen peroxide for 30 min at 37 • C, blocked using 5% BSA and incubated in HRP-conjugated antibody for β-actin (A3854, Sigma-Aldrich) as a house-keeping protein. Results were analysed using Image Lab™ Software v6.0 (Bio-Rad, Hercules, CA, USA). Results were presented as relative band density of protein of interest divided by relative band density of the respective β-actin.

RNA Isolation and Quantification
Following treatment for 24 h, cells were incubated with 500 µL Trizol and 500 µL PBS (5 min, RT). Samples were mechanically lysed, transferred to 2 mL micro-centrifuge tubes and stored (24 h, −80 • C). The samples were thawed at RT, followed by the addition of 100 µL chloroform and centrifugation (12,000× g, 10 min, 4 • C). The supernatant was transferred to fresh 2 mL micro-centrifuge tubes, and 250 µL isopropanol was added, followed by overnight storage at −80 • C. Thawed samples were then centrifuged (12,000× g, 20 min, 4 • C). The supernatant was discarded, and the pellet was washed in 500 µL cold ethanol (75%). Samples were centrifuged (7400× g, 15 min, 4 • C). The RNA pellets were air-dried (30 min, 24 • C) and re-suspended in 15 µL nuclease-free water. Quantification of the crude RNA was carried out using the Nanodrop2000 spectrophotometer (Thermo-Fisher Scientific, Waltham, MA, USA). RNA quality was determined using the A260/A280 ratio. All RNA samples were standardised to 500 ng/µL.

Quantification of mRNA Expression
The cDNA was synthesised from the crude RNA samples using the iScript™ cDNA Synthesis kit (Bio-Rad, 107-8890, Hercules, CA, USA).

Statistical Analysis
GraphPad Prism version 5.0 (GraphPad Prism Software Inc.) was used to perform all statistical analyses. Data were analysed using an unpaired t-test (data with 2 groups) and one-way analysis of variance (ANOVA) followed by a Bonferroni test for multiple group comparison. Data were considered significant at p < 0.05.  ( Figure A1B), suggesting the occurrence of cell death via the mitochondrial intrinsic apoptotic pathway. Figure A1. FB2-induced apoptosis in Hek293 cells. FB2 significantly increased the activity of caspase 9 ((A); *** p < 0.0001). A significant increase in executioner caspases 3/7 were observed following FB2 exposure ((B); *** p < 0.0001).