Phospholipid scramblase 3: a latent mediator connecting mitochondria and heavy metal apoptosis

Lead and mercury are the ubiquitous heavy metals triggering toxicity and initiating apoptosis in cells. Though the toxic effects of heavy metals on various organs are known, there is a paucity of information on the mechanisms that instigate the current study. A plausible role of phospholipid scramblase 3 (PLSCR3) in Pb2+ and Hg2+ induced apoptosis was investigated with human embryonic kidney (HEK 293) cells. After 12 h of exposure, ~30–40% of the cells were in the early stage of apoptosis with increased reactive oxygen species (ROS), decreased mitochondrial membrane potential, and increased intracellular calcium levels. Also, ~20% of the cardiolipin localized within the inner mitochondrial membrane was translocated to the outer mitochondrial membrane along with the mobilization of truncated Bid (t-Bid) to the mitochondria and cytochrome c from the mitochondria. The endogenous expression levels of PLSCR3, caspase 8, and caspase 3 were upregulated in Pb2+ and Hg2+ induced apoptosis. The activation and upregulation of PLSCR3 mediate CL translocation playing a potential role in initiating the heavy metal-induced apoptosis. Therefore, PLSCR3 could be the linker between mitochondria and heavy metal apoptosis.


Introduction
Heavy metals such as lead, mercury, cadmium, and arsenic are well-known environmental pollutants resulting in adverse health effects [1]. Heavy metals contaminate the environment through various environmental and anthropogenic activities causing various biochemical and physiological changes affecting different organ systems [2,3]. Lead and mercury exposure affects the hematopoietic, cardiovascular, nervous, reproductive, and renal systems. Although lead and mercury ions are distributed throughout the body upon exposure, the heavy metal ions mainly accumulate in the kidney and brain, especially elemental, inorganic, and organic forms of mercury [2,4]. The metal-induced toxicity mechanisms are still intriguing as the heavy metals are ubiquitously present [5,6]. Lead affects several processes within a cell by mimicking calcium or other divalent metal ions to bind with proteins or enzymes, inhibiting their functions or activity. Lead-mediated protein or enzyme inhibition affects the cellular processes, including calcium channel impairment, maturation of proteins, regulation of genes, vitamin D and energy metabolism, transportation of metal ions, ionic conduction, neurotransmitters' release, and signal transduction pathways [7][8][9][10]. Mercury inhibits antioxidant enzymes by binding with sulfhydryl groups resulting in increased reactive oxygen species (ROS) and peroxidation of lipids causing oxidative damage in mitochondria [2,11,12]. Heavy metals such as lead, mercury, cadmium, and arsenic in low doses either individually or as mixtures induced toxicity in liver, brain, and kidneys of young mice, causing neuronal degeneration and renal tubular necrosis affecting immature and developing organs. The study insisted on exploring the mechanisms associated with the low dose toxicity of the heavy metals as a pre requisite in protecting public health [13][14][15]. In a biological system, heavy metal exposure results in several biochemical changes as cellular response leading to apoptosis [6]. Apoptosis, a tightly regulated process where cell death is executed through specific signaling pathways, is classified as extrinsic and intrinsic apoptosis based on the initiating factors. Extrinsic apoptosis is mediated via receptors such as TNF or Fas, while intrinsic occur via disruption of cellular homeostasis [5]. Although many cellular organelles such as the Golgi apparatus, lysosomes, mitochondria, and endoplasmic reticulum play a crucial role in apoptosis, the mitochondrial role is prominent as both pathways converge at this organelle [5,16]. Mitochondria are the vital executioner of apoptosis, and the development of mitochondrial permeability transition pore (MPTP) in the intrinsic pathway is a crucial step for the release of several pro-apoptotic factors activating apoptotic pathway [17]. Environmental pollutants such as heavy metals target mitochondria leading to mitochondrial dysfunction, increased oxidative stress, and MPTP formation [18][19][20][21][22][23]. Mitochondria are essential in arbitrating metal-induced apoptosis through reactive oxygen species (ROS) generation [5,24]. Excessive ROS increased the permeability of mitochondrial membrane permeability and inhibited the respiratory chain favouring more ROS generation [24]. Membrane potential disruption mediated MPTP formation precedes the changes at the nuclear and plasma membrane and is putatively the initial step in apoptosis [25]. The permeability in the mitochondrial membrane favors cytochrome c release and eventually triggers the downstream effector caspases leading to apoptosis. Metal exposure is often associated with ROS generation and mitochondria, but apoptosis's underlying mechanism is unclear [5].
Cationic heavy metals, including lead and mercury, have shown preferential accumulation in mitochondria by two possible mechanisms: (1) entry by mimicking as calcium ions via calcium transporters and (2) by charge-based interactions and the mitochondrial pH. The mitochondrial matrix is negatively charged with a slightly alkaline pH ranging between~6.9 and 8.0 (between inter-membrane space and matrix) [4,23,26]. Both lead and mercury induce apoptosis in cells. The mitochondria play an essential role in lead-induced apoptosis. It was established that lead depolarizes mitochondria by MPTP formation, thereby releasing cytochrome c and activates caspases to initiate apoptosis. In lead-induced rod apoptosis, MPTP formation was due to translocation of Bax onto mitochondria and not because of oxidative stress [27,28]. It was established from an earlier study that liver and kidney mitochondria exhibited a strong binding with lead ions [4]. In macrophages, the mercury triggered apoptosis by increasing intracellular calcium levels, ROS generation, and caspase 3 activation [29]. In lymphoid cells, both organic and inorganic forms of Hg induced mitochondrial dysfunction, MPTP formation, and cytochrome c release, promoting apoptosis [11].
Previous reports suggest that human phospholipid scramblase 3 (PLSCR3), a mitochondrial membrane protein, is involved in maintaining mitochondrial morphology and function. PLSCR3 was shown to be involved in the translocation of cardiolipin (CL), a mitochondrial phospholipid from the inner mitochondrial membrane (IM) to the outer mitochondrial membrane (OM). Further, it was reported that PLSCR3 expression enhanced the mitochondrial CL transport in UV-induced apoptosis [30]. CL is essential for mitochondrial import of various proteins and their essential functions [31]. CL distribution in the IM plays a crucial role in ATP synthesis. CL redistribution in the mitochondrial membrane during early phase of apoptosis occurs in a time dependent manner and it happens before the appearance of apoptotic markers including phosphatidyl serine exposure in the plasma membrane and mitochondrial membrane potential changes [32]. PLSCR3 plays a regulatory role in TNF-α induced apoptosis. Calcium activation of PLSCR3 enhanced CL translocation to the OM, which enhanced t-Bid binding (truncated BH3 interacting domain) on the mitochondrial surface. Bax recruitment by the membrane-bound t-Bid enhanced Bax/ Bak oligomerization, triggering MPTP formation favouring cytochrome c release and caspase activation. It was also reported that t-Bid and PLSCR3 form a positive feedback loop in apoptosis initiation [33,34]. ROS generation enhanced upon overexpression of PLSCR3 in HeLa cells, implying a positive correlation between PLSCR3 expression, mitochondrial changes, and ROS production [33][34][35]. A recent biochemical study reported that hPLSCR3 binding to micromolar concentrations of Pb 2+ or Hg 2+ mediated phospholipids (PLs) translocation in synthetic proteoliposomes. Also, F258V-hPLSCR3, a calcium-binding mutant, displayed decreased affinity towards heavy metal ions and a~60% loss in the PL translocation ability [36]. Heavy metal-induced apoptosis is often associated with mitochondria, ROS generation, and cytochrome c release, but the underlying mechanism needs further investigation. Based on the available reports, we hypothesize that hPLSCR3 localized in mitochondrial membrane upon binding with heavy metals might result in CL translocation to the mitochondrial surface favouring t-Bid binding induced MPTP formation and cytochrome c release triggering the downstream effector caspases. Hence, the hypothesis put forth the following queries: (1) Do heavy metal-induced apoptotic mechanisms involve PLSCR3? (2) What is the effect of heavy metal ions on CL translocation in the mitochondrial membrane? (3) Do heavy metals, PLSCR3 expression levels, and CL translocation are related? We have observed from the results that endogenous expression levels of PLSCR3 are upregulated, favouring CL translocation onto the mitochondrial surface, recruiting more t-Bid and cytochrome c release from the mitochondria.

Cell viability assay
HEK 293 cells were cultured in a 25 cm 2 cell culture flask till 90% confluency at 37°C in a 5% CO 2 incubator. The cells were then seeded into each well of 96 wells microculture plate at a 104 cells/ml concentration and allowed to attach and grow for 24 h (~80% confluency). After the incubation period, the monolayer of cells was washed with phosphate buffer saline (PBS) pH-7.4, and fresh DMEM media were added to the wells. The cells were added with PbCl 2 or HgCl 2 dissolved in PBS such that their final concentration varied between 2.5 µM to 100 µM and incubated for different time intervals of 12 h and 24 h. Cells incubated with PBS were used as control. MTT assay was used to assess the cell viability after heavy metal exposure. MTT assay is based on the conversion of soluble yellow-colored tetrazolium salt, [3-(4, 5-dimethyldiazol (-2-yl)-2, 5-diphenyltetrazolium bromide] (MTT), into an insoluble purple-colored formazan by viable cells. MTT assay was performed as per the manufacturer's protocol (EZcount TM , HiMedia). Briefly, after the exposure period, MTT prepared in buffer was added to each well at a final concentration of 1 mg/ml and incubated for 4 h. The insoluble formazan crystals were dissolved using the solubilization buffer provided in the kit. The absorbance was measured at 570 nm using xMark TM microplate spectrophotometer (BIO-RAD). The results were represented as a percentage of control (untreated) cells.

Apoptosis assay
HEK 293 cells were grown in each well of 24 wells microculture plates at a concentration of 2 × 10 4 cells/ml and allowed to attach and grow for 24 h at 37°C, 5% CO 2 . After the incubation period, the monolayer of cells was washed with phosphate buffer saline (PBS) pH-7.4, and fresh DMEM media were added to the wells. Apoptosis was induced by adding PbCl 2 or HgCl 2 dissolved in PBS such that their final concentrations are 2.5 µM, 5 µM, 10 µM, and 20 µM and incubated further for 12 h, and untreated cells were used as control. Post heavy metal exposure, the cells were washed with phosphate buffer saline (PBS) and harvested by adding trypsin solution. The cells were collected by centrifugation at 1000 rpm for 10 min at 4°C. Harvested cells were then washed once with PBS and with 1x binding buffer (10 mM HEPES pH 7.4; 140 mM NaCl, 2.5 mM CaCl 2 ). The washed cells were then resuspended in binding buffer at a concentration of~1×10 6 cells/ml. Annexin V-FITC (Sigma, Missouri, USA) staining was done as per the manufacturer's instructions. Briefly, 5 µl of AnnexinV-FITC (50 μg/ml) was added to the cell suspension and incubated in the dark at room temperature for 10 minutes. After washing with 1x binding buffer, 5 µl of propidium iodide (PI) (100 μg/ml) was added and incubated in ice for a further 10 min in the dark. Cell sorting was performed using BD FACS CANTO II Flow cytometer (BD Biosciences) with FITC and PI dual staining. Results were analyzed using BD FACSuite software (BD Biosciences, USA).

Intracellular ROS measurement
Intracellular ROS content was measured using 2′,7′-Dichlorofluorescin diacetate (DCFDA). DCFDA, a cell-permeable non-fluorescent probe upon intracellular de-esterification and oxidation, converts into a highly fluorescent 2′,7′-dichlorofluorescein (DCF). The fluorescence of DCF is either visualized or measured to evaluate the intracellular ROS content. Briefly, 1×10 4 cells were seeded onto a 96-well plate and exposed to either Pb 2+ or Hg 2+ of different concentrations for 12 h. Untreated cells served as the control for the experiment. 10 mM DCFDA solution was prepared by dissolving in DMSO and from which 10 µM working solution of DCFDA was prepared in DMEM. Cells were washed once with DMEM, and to each well, 200 µl of 10 µM DCFDA solution was added and incubated for 30 minutes at 37°C. After the incubation period, the DCFDA solution was removed, and the cells were washed with DMEM and PBS. Finally, 200 µl of PBS was added to each well, and the images were captured using the fluorescence microscope (Zeiss AX10 microscope). The fluorescence intensity values of the cells were immediately measured using a multi-mode microplate reader (Enspire, PerkinElmer, UK) at Excitation/Emission wavelengths of 485 nm/535 nm.
Mitochondrial membrane potential measurement JC-1 (5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolylcarbocyanine iodide) staining was used to understand the changes in the mitochondrial membrane potential. JC-1 (structure in Fig. 1) exhibits potential-dependent accumulation in mitochondria, dictated by a shift in fluorescence emission wavelength from~535 nm (green) to~600 nm (red). Mitochondrial depolarization was identified by a decrease in the red/green fluorescence intensity ratio. Briefly, 1×10 4 cells were seeded onto a 96-well plate and are exposed to either Pb 2+ or Hg 2+ of different concentrations for 12 h. The untreated cells served as the control for the experiment. A stock solution of 10 mM JC-1 dye was prepared by dissolving in DMSO and from which 10 µM working solution of JC-1 was prepared in PBS. The cells were washed with PBS, followed by the addition of 200 µl 10 µM JC-1 solution. After the incubation period (30 minutes at 37°C), the JC-1 dye was removed, and 200 µl PBS was added to the wells, and the images were captured using the fluorescence microscope (Zeiss AX10 microscope). Mitochondrial depolarization was quantified by measuring the ratio of red/green fluorescence intensity ratio. The fluorescent intensities were measured at excitation/emission wavelengths of 550 nm/600 nm (red) and 485 nm/535 nm (green) using a multi-mode microplate plate reader (Enspire, PerkinElmer, UK).

Intracellular calcium measurement
Fura-2AM (AcetoxyMethylester) (Thermo Fisher Scientific), a cell-permeable calcium indicator, was used for intracellular calcium measurement. A stock solution of Fura-2, AM dye was prepared by dissolving in DMSO and from which 5 µM working solution was prepared in 1XPBS. HEK 293 cells were treated with PbCl 2 or HgCl 2 (2.5 μM, 5 μM, 10 μM, and 20 μM) for 12 h, washed with PBS and then the cells were loaded with 5 μM Fura-2AM in the PBS for 30 min at 37°C in the dark. Excess Fura-2AM was removed by washing twice with PBS. The cells were added with PBS and incubated for another 30 min before measuring the fluorescence using a multi-mode microplate reader (Enspire, PerkinElmer, UK). Fluorescence intensity was measured at an emission wavelength of 510 nm by using a pair of excitation wavelengths at 340 nm and 380 nm and subsequently expressed as the ratio of light excited at 340 nm to that at 380. Changes in the mitochondrial calcium levels were visualized by probing for mitochondria and calcium subsequently. The cells were stained with 0.1 μM Mito Tracker TM red in the PBS for 30 min at 37°C in the dark. Excess stain was removed by washing twice with PBS, followed by staining with Fura-2AM as described above. The images were captured using the fluorescence microscope (Zeiss AX10 microscope).
Nonyl Acridine Orange (NAO) staining CL distribution within mitochondrial membrane was estimated using NAO dye. NAO interacts with diacidic phospholipids and forms dimers. Therefore the fluorescence emission shifts from 535 nm (green fluorescence-monomeric NAO) to 640 nm (red fluorescence-dimeric NAO). The heavy metal exposed and control cells were fixed using 1% formaldehyde (in PBS) and incubated with the various concentrations of NAO for 15 min at room temperature as described earlier [32,34,37]. The cells were washed twice with cold PBS and resuspended in PBS at a final concentration of 1 × 10 4 . The fluorescence intensity of the cells was measured using a multi-mode microplate reader (Enspire, Perkin Elmer, UK) with excitation at 490 nm and emission at 640 nm.

Sub-cellular fractionation
Sub-cellular fractionation was done as described earlier [38]. Cells were harvested at 1000 rpm in STM Buffer. The harvested cells were vortexed at maximum speed for 15 s. After a brief incubation period on ice for 30 min, the cell suspension was subjected to centrifugation at 800 × g for 15 min. The supernatant fraction was collected and spun again for additional 10 min at 800 × g to remove any debris. The collected supernatant was spun at 11,000 × g for 10 min, and the supernatant (S1) and the pellet fractions (P1) were separated. The collected supernatant (S1) was precipitated in 100% cold acetone at −20°C for an hour and centrifuged at 12,000 × g for 5 min. The collected pellet fraction upon resuspension in STM buffer comprised the cytosolic fraction. The P1 pellet fraction was resuspended in STM buffer and spun at 11,000 × g for 10 min. The pellet (P2) fraction collected was resuspended in SOL buffer and sonicated at 30% amplitude on ice for three cycles of 10 s pulse with 30 s pause. The sonicated pellet is comprised of the mitochondrial fraction.

Western blotting
The protein samples (30 µg) were separated by a 10% SDS-PAGE and transferred onto a nitrocellulose membrane (Sigma). The membrane was probed using primary antibodies for PLSCR3, caspase 8, caspase 3, VDAC-1, β-actin, Bid, and cytochrome c. The probed proteins were then visualized by adding specific secondary HRP-conjugated antibodies followed by Western ECL substrate (Clarity TM , Bio-Rad Laboratories). The images were visualized using the VersaDoc TM imaging system (Bio-Rad Laboratories). The band intensities were calculated using Image Lab 5.0 software (Bio-Rad Laboratories).

Statistical analysis
Statistical analysis of the data was done by One-way ANOVA followed by Dunnett's comparison test using Prism 5.0 Graphpad software (San Diego, CA, USA) unless otherwise mentioned. Significance was acknowledged when probability values p < 0.05 were observed. The analyses corresponding to individual experiments are presented in the respective figure legends.

Results
HEK cells were used in this study to comprehend the role of PLSCR3 in heavy metal-induced cytotoxicity. The cells were exposed to either Pb 2+ or Hg 2+ , and its effect on the viability of the cells, ROS generation, and apoptosis induction was studied. The molecular mechanism involving PLSCR3 during heavy metal-induced apoptosis and its implication on the mitochondrial membrane were studied.

Cell viability
Human embryonic kidney cells were exposed to heavy metals, Pb 2+ or Hg 2+ , over a broad range of concentrations ranging from 5 µM to 100 µM. The level of cytotoxicity induced by the heavy metals on HEK cells was estimated by MTT assay after 12 h and 24 h of exposure. From Fig. 1, it was evident that both Pb 2+ and Hg 2+ exhibited dose and duration-dependent cytotoxicity. Hg 2+ was more toxic than Pb 2+ after 24 h of exposure, significantly beyond 50 µM. After 12 h of exposure, both the heavy metals were mildly cytotoxic against the cells. Both lead and mercury resulted in a 3-10% loss of viability in the lower concentration range from 5 µM to 20 µM. The cells exhibited morphological changes even at a lower concentration of 2.5 µM (Supplementary information Fig. S1) at 12 h of exposure. To further investigate the molecular basis of the toxicity, an apoptosis assay was performed by dual staining of the cells exposed to lead or mercury.

Apoptosis assay by FAC
Annexin-V/PI staining analyzed by fluorescence-assisted cell sorting differentiated the viable, early apoptotic, late apoptotic, and necrotic cells upon exposure to different concentrations of heavy metals. From Fig. 2, it is evident that Q3 represents viable cells, Q4 represents Annexin-V positive cells in the early phase of apoptosis, and Q2 represents Annexin-V and PI-positive late-apoptotic cells, and Q1 represents PI-positive necrotic cells. When compared to untreated (control) cells ( Fig. 2A), 2.5 µM Pb 2+ exposed cells had~30% of the cells in early apoptotic (Fig. 2B) stage and~40% with 2.5 µM Hg 2+ exposure for 12 h (Fig. 2F). With the increase in heavy metal exposure, i.e., with increased concentration of heavy metals, the % of cells entering the late apoptotic phase increased, which is evident from Fig. 2. Almost 40% of cells reached the early apoptotic phase while another 45-50% of the cells enter the late apoptotic phase when exposed to 20 µM of Pb 2+ or Hg 2+ . The results suggested that phosphatidylserine (PS) exposure had occurred upon heavy metal exposure, i.e., the early stage of apoptosis was initiated with minimal amount of heavy metals exposure (2.5 µM). The plasma membrane changes generally follow the mitochondrial membrane potential disruption. Therefore, we sequentially explored the other molecular events leading to apoptosis in the mitochondria by exposing the cells to a minimal amount of heavy metal in the further studies.

Mitochondrial membrane potential changes
To emphasize the mechanism involving heavy metal apoptosis, we explored the effect of Pb 2+ or Hg 2+ on the mitochondrial membrane potential (ΔΨ mito ). The changes in ΔΨ mito were  measured by staining the control and heavy metal exposed cells with JC-1 dye. The cationic dye is cell-permeant and, upon entering healthy mitochondria, accumulates and forms J aggregates shifting the emission spectrum from green to red. Therefore in depolarized mitochondria, green fluorescence will be higher. The red/green fluorescence intensity ratio is a direct measure of the changes in the ΔΨ mito . The higher ratio suggests a higher ΔΨ mito, and a decrease in the ratio implies that the mitochondrial membrane is depolarized. From Fig. 3A, it was observed that, compared to control cells, 2.5 µM Pb 2+ or Hg 2+ exposed cells exhibited a reduction in the J aggregate formation and increased green fluorescence. The red/green fluorescent intensity ratio data demonstrated a significant decrease in the ΔΨ mito in the heavy metal exposed cells as compared against the control (Fig. 3B). The results implied that the mitochondria are depolarized upon exposure to 2.5 µM Pb 2+ or Hg 2+ . There is a dose-dependent decrease in the ΔΨ mito and depolarization saturating at 10 µM of both lead and mercury.

Intracellular ROS measurement
ROS formation and perturbed mitochondrial membrane potential are critical processes involved in heavy metal-induced apoptosis. Therefore to investigate the formation of ROS, we have used DCFDA based ROS measurement. From the results, it was evident that heavy metal exposure significantly enhanced the production of reactive oxygen species (Fig. 4). Compared to the control cells, there was a two-fold increase in the relative fluorescent intensity in the cells exposed to either 2.5 µM Pb 2+ or Hg 2+ for 12 h. It appeared that both Pb 2+ and Hg 2+ influenced the ROS generation in HEK cells, and the heavy metal cations were capable of producing a stimulating effect (higher ROS production) with low concentrations at 12 h of incubation (Fig. 4B).

Intracellular calcium measurement
Intracellular calcium homeostasis is perturbed in heavy metal toxicity. Also, PLSCR3 is a calcium-dependent protein involved in PLs exposure. Therefore to investigate the changes in the intracellular calcium levels, additional experiments were performed. The fluorescence intensity ratio of Fura-2AM was measured in HEK cells following the incubation of the cells in the presence or absence of Pb 2+ and Hg 2+ . The cells exposed to either 2.5 µM Pb 2+ or Hg 2+ displayed a significant increase in the green fluorescence when compared against the control The fluorescence images were observed using a fluorescent microscope with green and red channels. The red fluorescence (JC-1 aggregates) represents intact mitochondria, and the green fluorescence (JC-1 monomers) represents a decrease in the ΔΨ mito, and yellow fluorescence (merged) represents damaged mitochondria. The fluorescent image analysis was performed using Image J software. B Bar graph depicting the changes in the red/green fluorescence intensity ratio between control and Pb 2+ or Hg 2+ exposed cells. Fluorescent intensity values were measured at 535 nm (green) and 600 nm (red) in a multi-mode plate reader. The ratio of red to green fluorescent intensity values was calculated. Changes in the ratio of red to green fluorescence intensity represent the mitochondrial membrane polarization. The results represented here are the mean values of three independent experiments (mean ± S.D). Statistical analysis was performed by One-way ANOVA followed by Dunnett's comparison test. *** represents p value < 0.05 and are considered significant. Compared to control cells, HEK 293 cells treated with 2.5, 5.0, 10.0, 20.0 µM Pb 2+ or Hg 2+ displayed a significant decrease in the ΔΨ mito cells (Fig. 5A), suggesting a possible rise in the intracellular calcium levels of the heavy metal exposed cells at such low concentrations. The fluorescence intensity of Fura-2AM treated cells was measured at 510 nm by exciting them at two different wavelengths, i.e., 340 nm and 380 nm, to illustrate the free intracellular calcium concentration changes. The intracellular calcium levels increased among the cells exposed to Pb 2+ and Hg 2+, which was evident from a significant increase in the fluorescence intensity ratio (340/380) (Fig. 5B), suggesting the possibility of scramblase activation. The increase in the calcium levels in apoptotic cells could be due to the release of calcium from the intracellular reserves. The calcium levels within the mitochondria of heavy metal exposed cells were marginally increased compared to the control cells signifying the activation of scramblase localized in mitochondria (PLSCR3) (Fig. S2).

CL distribution in the mitochondrial membrane
It was established from previous studies that in the TNF-α induced apoptosis, PLSCR3 activation modulates the CL distribution on the mitochondrial surface [34]. We utilized NAO to determine the CL distribution in the mitochondrial membrane as NAO binding CL results in an emission shift from green (535 nm) to red (640 nm). Increasing NAO concentrations result in an eventual fluorescence emission shift towards 640 nm because NAO forms dimers (red), and the monomers decrease (green). NAO binds CL at a ratio of 2:1, and the fluorescence intensity ratio at 640 nm correlates not only with the amount of CL but also with its distribution in the membrane. The NAO dimers upon saturating all of the CL molecules on the outer mitochondrial membrane result in a plateau in the fluorescence emission. Then, the NAO-CL interaction induces a modification in the inner membrane permeability, favouring NAO entry into the inner membrane. Therefore, CL in the outer mitochondrial membrane (OM) is saturated by NAO before the inner mitochondrial membrane (IM), and the outer leaflet of the bilayer is saturated earlier than the inner leaflet [32,34,37]. Hence, in theory, four plateaus would be formed upon NAO staining, but we have optimized the study by utilizing very low concentrations of NAO (0.25 µM to 5 µM) to visualize three plateaus so that the CL distribution from the IM to the OM could be investigated. Using the maximal fluorescence intensity at 5 µM NAO The plateau dips marked by 1, 2, and 3 in Fig. 6 represent the percentage of CL in the outer and inner leaflets of OM and IM, respectively. The dips were formed at 0.75 µM, 1.5 µM, and 2.5 µM NAO concentrations in control cells, and the corresponding percentages of CL were~10.5%,~30% in the outer and inner leaflet of OM, and~60.4% in the IM respectively. In 2.5 µM Pb 2+ exposed cells, the percentages of CL were~13%,~39% in the outer and inner leaflet of OM, and~46.7% in the IM, respectively. Similarly, in the cells exposed to 2.5 µM Hg 2+ , the percentages of CL were~17%, 39%, and 43%, respectively (Fig. 6D). The results showed that CL% increased from~10.5% to almost~13-17% at the first dip and from~30% to~39% at the second dip. There was a decrease in CL% from~60% to~40% at the third dip in heavy metal exposed cells. The observed results indicated that CL has translocated from IM to OM in heavy metal exposed cells. The observed results were similar to earlier reports wherein PLSCR3 overexpression resulted in the translocation of CL from IM to OM. Therefore we further studied the expression levels of PLSCR3 in the control and metal exposed cells.

PLSCR3 in heavy metal apoptosis
In TNF-α induced apoptosis, PLSCR3 enhanced CL exposure on the mitochondrial surface, favouring t-Bid binding on the mitochondrial membrane. Since the CL translocation was evident from Fig. 6D, the endogenous expression levels of PLSCR3 were analyzed by western blotting. From  Fig. 7A, it was evident that the endogenous expression levels of PLSCR3 were higher in the heavy metal-induced apoptotic cells. The results suggest that PLSCR3 was upregulated among the HEK cells exposed to either 2.5 µM Pb 2+ or Hg 2+ in 12 h, especially in lead-induced apoptosis. The expression levels were relatively higher than the control (Fig. 7C). Pb 2+ or Hg 2+ induced apoptotic response triggered up-regulation of caspase 8 and activation of caspase 3 apparent from western blot analyses of the whole-cell lysates (Fig. 7A). Sub-cellular fractions revealed the release of cytochrome c from mitochondria into the cytosol after Pb 2+ or Hg 2+ exposure. The t-Bid movement to mitochondria A Representative bright field and fluorescent images (green) of control (untreated) and 2.5 µM Pb 2+ or 2.5 µM Hg 2+ treated HEK293 cells. The green fluorescence represents the qualitative changes in the intracellular free calcium levels in the Pb 2+ or Hg 2+ exposed cells. B Bar graph representing a measure of free intracellular calcium. A comparison of the fluorescent intensity ratio (340/380) between control and Pb 2+ or Hg 2+ exposed cells. The ratio of the fluorescent intensity values measured at 510 nm by exciting at two different wavelengths of 340 nm and 380 nm was calculated. Compared to control, HEK 293 cells treated with 2.5, 5.0, 10.0, 20.0 µM Pb 2+ or Hg 2+ displayed a significant increase in the intracellular calcium levels. The results represented here are the mean values of three independent experiments (mean ± S.D). Statistical analysis was performed by One-way ANOVA followed by Dunnett's comparison test. *** represents p value < 0.05 and are considered significant  Fig. 7 Western Blot analysis. Upregulation of endogenous PLSCR3 expression enhances cytochrome c release in heavy metal-induced apoptosis. A PLSCR3 expression was upregulated in 2.5 µM Pb 2+ or Hg 2+ treated HEK 293 cells. Whole-cell lysate from untreated (control) and 2.5 µM Pb 2+ or Hg 2+ exposed HEK 293 cells after 12 h of treatment. Endogenous caspase 8 expression levels were upregulated, and caspase 3 was activated in Pb 2+ or Hg 2+ induced apoptotic cells. β-Actin was used as the loading control for whole-cell lysate. B Upregulation of PLSCR3 induced mitochondrial targeting of t-Bid and the release of cytochrome c into the cytosol. HEK 293 cells were treated with 2.5 µM Pb 2+ or Hg 2+ for 12 h followed by subcellular fractionation. The first lane depicts the release of cytochrome c from the mitochondria into the cytosol. The second lane depicts the movement of t-Bid to the mitochondrial surface in Pb 2+ or Hg 2+ induced apoptotic cells. β-Actin was used as the loading control for cytoplasmic fractions, and VDAC was used as the loading control for the mitochondrial fractions. The whole-cell lysates and the mitochondrial and cytoplasmic fractions were analyzed by western blotting with antibodies against PLSCR3, caspase 8, caspase 3, cytochrome c, and Bid for the expression levels and their distribution, while the marker proteins such as β-Actin and VDAC were used as the loading control. C Data analysis of western blot. The protein expression levels of PLSCR3, caspase 3 and caspase 8 were calculated by normalizing the band intensities of corresponding proteins with respect to β-Actin (loading control). The results represented here are the mean values of three independent experiments (mean ± S.D). Statistical analysis was performed by One-way ANOVA followed by Dunnett's comparison test. *** represents p value < 0.0001, ** and * represents p value < 0.001 and p < 0.05 are considered significant happened after exposure to the heavy metals and was more evident in the case of lead-induced apoptosis (Fig. 7B).

Discussion
Several reports on heavy metal-induced toxicity demonstrate the involvement of mitochondria, ROS generation, and cytochrome c release in different cell lines. The underlying mechanism coupling the three was unexplored, and the current study is an attempt to abridge the knowledge gap.
Several reports have established the involvement of mitochondria in Pb 2+ and Hg 2+ induced apoptotic cell death, suggesting mitochondria as the initial target site in heavy metal-induced apoptosis [18,21,28]. Divalent cations such as lead and mercury have shown preferential accumulation in mitochondria either by entering through calcium uniporters or through charge-based interactions at the mitochondrial matrix [4,23,28]. The convergence of pro-apoptotic signal transduction pathways in mitochondrial membranes favored its permeabilization. Also, in different models of apoptotic induction, numerous proteins are either translocated to or are released from mitochondria [16]. The production of reactive oxygen species in heavy metalinduced toxicity was reported to deplete glutathione and through protein-sulfhydryl group inhibitions by metal ions. As a result of ROS generation, the calcium homeostasis within the cells was perturbed. The metal ions were shown to influence the production of tumor necrosis factor α and enhance the protein kinase c activity [39]. Our results demonstrated that the human embryonic kidney cells displayed a two-fold increase in ROS production on exposure to either lead or mercury for 12 h (Fig. 4). The previous reports suggested that ingestion of mercuric chloride to rats resulted in nephrotoxicity and acute renal failure. The mercury induced ROS generation perturbed the membrane integrity and altered the antioxidant defense mechanism [40]. As early as 12 h post mercury administration, the lipid peroxidation occurred in kidneys favouring the ROS generation, which corroborates our study [41]. Lead toxicity in liver, kidney, and erythrocytes resulted in oxidative stress and increased ROS generation [39,42]. The intracellular calcium levels rise in the heavy metal-induced apoptosis, and in our report, the kidney cells displayed a~1.2 fold increase in the intracellular calcium levels. Pb 2+ acts as a calcium agonist in rod apoptosis and has been shown to increase intracellular Ca 2+ levels by~1.5-2 fold in erythrocytes [28,43]. The organic form of mercury generally altered the intracellular calcium levels through an influx of calcium ions from the external medium and mobilized intracellular calcium stores, whereas HgCl 2 predominantly influenced the influx of external Ca 2+ [44]. HgCl 2 induced a four-fold increase in intracellular Ca 2+ within rat hepatocytes in the presence of a high calcium medium while a very low increase in the low calcium medium [18]. One of the mechanisms by which metals interfere with calcium homeostasis is entry via the transport proteins or calcium channels into the intracellular space and interaction with the calcium store sites favouring the release into the cytosol. The other known mechanism is calcium reuptake by the calcium stores via the calcium extrusion mechanisms involving ATPases [45]. Therefore, the marginal~1.2 fold increase in the intracellular and mitochondrial calcium in our study could be the result of the calcium homeostasis maintenance by the cells in the absence of extracellular calcium. Pb 2+ induced a decrease in the ΔΨ mito and cytochrome c release in rod cell apoptosis, but the mechanism behind the MPTP formation is unclear. However, results were indicative that the Pb 2+ binding to the metal-binding site in the mitochondrial matrix opens the MPTP [28]. Lead by mimicking calcium, depolarized mitochondria, increased calcium overload, released cytochrome c through MPTP triggering apoptosis [46]. Our results agreed with the established reports of a decrease in the ΔΨ mito followed by cytochrome c release from mitochondria. Cardiolipin, a unique phospholipid synthesized and predominantly localized in the inner mitochondrial membrane, plays a vital role in membrane structure, stability, and functions, especially in the different stages of apoptosis involving mitochondria [47,48]. CL exposure to OM resulting from mitochondrial stress acts as a binding site in signaling events and promotes ROS generation. The exposed CL acts as a binding platform for t-Bid and Bax [31]. Our present study reported a translocation of~20% CL content from the IM to OM upon Pb 2+ or Hg 2+ exposure suggesting that the exposed CL could be a potent binding site for t-Bid and Bax triggering the MPTP formation.
PLSCR3, a calcium-dependent phospholipid scramblase, acts as a novel regulator of CL de novo biosynthesis and resynthesis and involves in the CL's translocation from the IM to the OM [49]. PLSCR3 plays a vital role in maintaining mitochondrial structure, function, and apoptotic response to t-Bid and UV-induced apoptosis through CL translocation [30]. Also, a previous report stated that PLSCR3 was able to bind Pb 2+ and Hg 2+ with much higher affinity (in the micromolar range) compared to Ca 2+ and induce PL translocation [36]. Therefore, we analyzed the endogenous expression levels of PLSCR3 in the heavy metal induce apoptotic cells. The western blot analysis established that similar to the t-Bid induced apoptosis and TNF-α induced apoptosis, the PLSCR3 levels were upregulated in both lead and mercury-induced apoptosis in kidney cells.
To further affirm the role of PLSCR3 in heavy metalinduced CL translocation, we screened for cell lines with relatively (relative to HEK 293) low expression of endogenous PLSCR3. Previous studies revealed that mRNA expression levels of PLSCR3 were relatively low or negligible among liver and brain tissue and monocyte cell lines [50][51][52]. A total protein of 30 µg was obtained from the whole-cell lysates of monocytes (THP-1) cell lines were analyzed by western blot probing for PLSCR3. From Fig.  S3A, it was evident that the relative expression of PLSCR3 was either low or negligible in the case of monocytes (THP-1), corroborating with the reports cited.
Therefore, we compared the changes in the cardiolipin distribution and the expression levels of PLSCR3 among THP-1 and HEK 293 cells upon 12 h exposure to 2.5 µM Pb 2+ or Hg 2+ . PLSCR3 expression was uninfluenced and remained negligible in heavy metal exposed THP-1 cells, unlike the HEK 293 cells. We have also analyzed the CL with THP-1 cells and found from the Fig. S3B that the CL content remained unchanged in the control and 2.5 µM Pb 2+ or Hg 2+ exposed THP-1 cells at~59% in the IM. Also, the redistribution from IM to OM did not happen as the CL content was~29% in the inner leaflet of the OM in the control and heavy metal exposed cells. The results corroborate the earlier report where the alkaloid treatment of cancer stem cells induced apoptosis, while in THP-1 cells along with the alkaloid, the presence of caspase inhibitor modulated scramblase activity to trigger apoptosis. Hence the differential kinetics exhibited by the THP-1 cells upon induction of apoptosis in the presence of caspase inhibitor could be the reason for PLSCR3 inactivation in heavy metal exposed THP-1 cells [53].
In lead-induced rod apoptosis, caspase-3 is the executioner caspase [28]. To explore the involvement of executioner caspase, the expression levels of caspase-3 were analyzed by western blot in both lead and mercury-induced apoptotic cells. Our data revealed that caspase-3 was upregulated in both the heavy metal-induced apoptosis. CL remodeling compromised the oxidative phosphorylation in PLSCR3 overexpressed cell lines resulting in ROS generation. Apart from CL being the binding platform for t-Bid, PLSCR3 expression modulates the t-Bid binding capacity on the mitochondrial surface. These two proteins form a positive feedback loop in amplifying the TNF-α induced apoptotic response. Also, cytochrome c release was attributed by two mechanisms: (1) Bax/Bak oligomerization mediated by t-Bid targeting on the mitochondrial surface and (2) CL peroxidation because of increased ROS [34]. Figure 7 clearly demonstrated that the cytochrome c release from the mitochondria and t-Bid targeting the mitochondrial surface in the apoptotic cells substantiated the CL translocation data and the previous reports [33,34].
Mitochondrial calcium uniporter specifically required CL for its stability and activity, and a loss in CL impaired the calcium transport [54]. The appearance of CL on the mitochondrial surface precedes the phosphatidylserine exposure on the plasma membrane, changes in the ΔΨ mito but occurs after the generation of ROS in TNF-α induced apoptosis suggesting the CL distribution as a significant and one of the primary events in apoptosis [32]. The levels of TNF-α significantly increased upon lead exposure, but this alone did not trigger apoptosis [55]. The heavy metal ions regulate TNF-α production and enhance the protein kinase c (PKC) activity [29,39]. PKC-δinduced apoptosis targets PLSCR3 in mitochondria [35]. Phosphorylation of PLSCR3 at threonine 21by PKC-δ increased the PL translocation rate [56,57]. Activation of PKC-δ enhances cell death through caspase 8 activation [58]. TNF-α related apoptosis-inducing ligand (TRAIL) induced PKC-δ activation, enhanced PLSCR3 induced CL translocation, and caspase 8 activation [33]. Activation of caspase 8 increases Bid processing, and in TRAILinduced apoptosis, PLSCR3 enhanced Bid cleavage into t-Bid [33]. Therefore, PKC-δ activation reported in heavy metal apoptosis could also enhance PLSCR3 phosphorylation and activation to induce CL translocation and further involve in the apoptotic process.

Conclusion
In conclusion, based on our results on lead and mercury induced toxicity, we propose a model interlinking ROS, mitochondria, and heavy metal apoptosis through upregulation of PLSCR3, a mitochondrial membrane protein. Pb 2+ and Hg 2+ up on entering mitochondria bind PLSCR3 to induce the CL translocation onto the mitochondrial surface. TNF-α activation in heavy metal apoptosis further enhances caspase 8 activation truncating Bid to t-Bid. The surface-exposed CL recruits t-Bid on to mitochondrial membrane, which further triggers Bax/Bak oligomerization resulting in MPTP formation releasing cytochrome c from the mitochondria into the cytosol. The released cytochrome c activates the executioner caspase 3 in heavy metalinduced apoptosis. Metal-binding mediated ROS generation and increased intracellular calcium form a positive feedback loop in upregulating and activating PLSCR3, favouring CL translocation mediated MPTP formation and cytochrome c release. It was also reported earlier that t-Bid further enhances CL translocation through a positive feedback loop with PLSCR3. Hence, PLSCR3 could be the potent membrane protein involved in mitochondriamediated heavy metal apoptosis (Fig. 8). Up-regulation and activation of PLSCR3 in the mitochondrial membrane is an immediate stress response. Examining the mechanisms associated with heavy metal toxicity is a need of the hour regarding public health safety. Mitochondrial response to heavy metal toxicity precedes other cellular changes, and therefore PLSCR3 could be utilized as a potential biomarker in the earlier assessment of risk associated with heavy metal exposure. The current study is a preliminary report revealing the plausible role of PLSCR3 in heavy metal induced apoptosis. Further investigation with the PLSCR3 knock down cells, other heavy metals and the mitochondrial changes mediated by PLSCR3 might provide profound evidence on the function of the protein in heavy metal apoptosis.

Data availability
The data underlying this article will be shared on reasonable request to the corresponding author. Fig. 8 An overview of plausible mechanism. The mechanism of action describing the possible linkage between PLSCR3 and mitochondria in heavy metal induced apoptosis