Cadmium overkill: autophagy, apoptosis and necrosis signalling in endothelial cells exposed to cadmium

Apoptosis, necrosis, or autophagy—it is the mode of cell demise that defines the response of surrounding cells and organs. In case of one of the most toxic substances known to date, cadmium (Cd), and despite a large number of studies, the mode of cell death induced is still unclear. As there exists conflicting data as to which cell death mode is induced by Cd both across various cell types and within a single one, we chose to analyse Cd-induced cell death in primary human endothelial cells by investigating all possibilities that a cell faces in undergoing cell death. Our results indicate that Cd-induced death signalling starts with the causation of DNA damage and a cytosolic calcium flux. These two events lead to an apoptosis signalling-related mitochondrial membrane depolarisation and a classical DNA damage response. Simultaneously, autophagy signalling such as ER stress and phagosome formation is initiated. Importantly, we also observed lysosomal membrane permeabilization. It is the integration of all signals that results in DNA degradation and a disruption of the plasma membrane. Our data thus suggest that Cd causes the activation of multiple death signals in parallel. The genotype (for example, p53 positive or negative) as well as other factors may determine the initiation and rate of individual death signals. Differences in the signal mix and speed may explain the differing results recorded as to the Cd-induced mode of cell death thus far. In human endothelial cells it is the sum of most if not all of these signals that determine the mode of Cd-induced cell death: programmed necrosis. Electronic supplementary material The online version of this article (doi:10.1007/s00018-015-2094-9) contains supplementary material, which is available to authorized users.


Quantification of cell death
For the analysis and quantification of Cd-induced cell death, Annexin V/Propidium iodide (PI) staining was used to determine the amount of dead cells (apoptotic and necrotic).Endothelial cells were seeded in 6-well plates and pre-and co-incubation with indicated Sucrose, Fructose, and Mannitol concentrations and treated with 15µM and 30µM Cd respectively, for the indicated times.The staining of the cells was performed as described in Bernhard et al. (Bernhard et al. 2003) and the quantification was performed using a flow cytometry device (FACS Canto II, BD Biosciences, Germany).Annexin V and PI positive cells were summarized and quantified compared to corresponding controls to display all dead cells.

Quantification of intracellular Ca2+ concentration
For the analysis and quantification of intracellular Ca 2+ concentrations, HUVECs were seeded in 6-well plates and incubated with 15µM and 30µM Cd respectively for the indicated times.After the treatment, cells were incubated with 1µM FLUO 3/AM, a cell-permeable Ca 2+ indicator that exhibits an increase in fluorescence upon Ca 2+ binding.Increase in fluorescence intensity after Cd-treatment was detected using flow cytometry analysis (FACS Canto II, BD Biosciences) and quantified as change in the intensity compared to the control.

ATP-Assay
Intracellular ATP-concentration was measured using the ApoSENSOR Viability Kit from Biovision, according to the manufacturer's instructions (Biovision,.ATP-concentration was analysed as ATP amount per number of cells.

Quantification of the number of viable cells
To analyse the number of cells, endothelial cells were treated with 15µM and 30µM Cd respectively over a time period of 96h.Quantification of the number of cells was performed using the XTT assay (Biomol, Hamburg, Germany) according to the manufacturer's instructions.As indicated, cells were pre-incubated with 50µM calpain inhibitor for 30 min before Cd treatment.

Cell cycle analyses
To analyse the influence of Cd on the cell cycle, endothelial cells were treated with 15µM and 30µM Cd respectively for 6h.After harvesting, the cells were fixed in 70% ethanol for 1h at 4°C.Afterwards, cells were washed three times with PBS and stained with a solution containing 0.1% Triton-X100, 50µg/ml PI and 100µg/ml RNase for 15min at 37°C.The cells were analysed using flow cytometry (FACS Canto II, BD Biosciences, Germany) and quantified as percent cells in the respective cell cycle phase.

Tracking of lysosome stability
Labelling of lysosomes was performed as previously described in Messner et al. (Messner et al. 2012) Endothelial cells with or without inhibitor and p53 knock-down (KD) cells were incubated with 15µM and 30µM Cd respectively for the indicated times.As indicated, cells were pre-incubated with 50µM calpain inhibitor or 2mM 3MA for 30 min before Cd treatment.

Detection and quantification of cellular DNA content and PI staining.
After the treatment, endothelial cells were enzymatically detached.Following permeabilization of cell membranes with 1mg/ml saponin, the DNA was stained with PI and the DNA content was quantified by flow cytometry (FACS Canto II, BD Biosciences, Germany).Inhibitors used were: 50µM Aurintricarboxylic acid ammonium salt (ATA; A0885, Sigma-Aldrich) and 1mM EGTA (ethylene glycol tetra-acetic acid, Sigma Aldrich).As indicated, cells were pre-incubated with 50µM calpain inhibitor or 2mM 3MA for 30 min before Cd treatment.

Generation of stable Bcl-XL OE cells as well as p53 KD cells.
Generation of stable BCL-XL OE and p53 KD endothelial cells was performed as already described.(Messner et al. 2012)

Western Blot analyses
Whole protein extracts from Cd treated and control cells were obtained by incubation of detached cells in triple detergent lysis buffer (50mM Tris-Chloride, 150mM Sodium Chloride, 0.02% Sodium Azide, 0.1% SDS, 1% Nonident P-40, 0.5% Sodium Deoxycholate, 5µg/ml Aprotinin, 1µg/ml Leupeptin, 1µg/ml Pepstatin and 1mM ABESF).Cells were lysed by repeated freezing and thawing and subsequent sonication.After a centrifugation step, the protein concentration of the supernatant was determined using a BCA protein assay kit (Pierce) according to the manufacturer's instructions.Equal amount of proteins were separated on SDS-polyacrylamide gels and subsequently transferred onto a nitrocellulose membrane (Schleicher and Schuell, Germany).The primary antibodies used were: mouse anti-p53 (#554294, BD, Austria), rabbit anti-BCL-XL (#2764, Cell Signaling, Germany) as well as rabbit anti-GAPDH (ab9485; Abcam, UK, dilution 1:2500) and the secondary antibodies used were goat anti-mouse IgG antibody-HRP conjugate (#32430, Thermo Scientific, Austria) and goat anti-rabbit IgG antibody-HRP conjugate (#31460, Thermo Scientific, Austria).To control equal protein amount loading, membranes were stained with Ponceau S. Furthermore equal loading was controlled by evaluating GAPDH expression.

Cd induced increase in intracellular Ca 2+ concentration over time.
Ca 2+ is an important intracellular messenger molecule and is known to be involved in the induction of cell death.
Cd induces a significant increase in the intracellular Ca 2+ concentration, which already starts 1h after Cd treatment and steadily increases over 48h. Figure S1 shows that Ca 2+ concentration increased over time, whereby the increase depends on the Cd concentration.Treatment with a higher Cd concentration of 30µM provokes a stronger increase in intracellular Ca 2+ concentration in comparison to the incubation with 15µM Cd.

Cd treatment does not change intracellular ATP concentration.
As apoptotic cell death is known to be an energy consuming process (in contrast to necrosis), we analysed the intracellular ATP concentration in endothelial cells after Cd treatment.Detailed analyses revealed that treatment with 15µM Cd initially induces a significant increase in intracellular ATP concentration after 6h, but a nonsignificant drop in the concentration over 72h.In contrast, treatment of cells with 30µM Cd causes no increase in intracellular ATP concentration but a significant decrease after 72h (Figure S2).

Cd treatment inhibits cell proliferation and p53 KD is able to preserve cells from proliferation arrest.
Treatment of endothelial cells with 15µM Cd reduces the proliferation ability after 48h significantly, whereas incubation with 30µM Cd exerts its proliferation inhibiting effect already after 24h.Further on, untreated cells grow exponentially but Cd reduces the number of cells over 96h.Compared to wild type endothelial cells stable knock-down of p53 protects cells from Cd induced proliferation arrest significantly.However, this knock-down is not able to preserve the proliferation ability after Cd treatment entirely (Figure S3).In contrast, pre-treatment of endothelial cells with an inhibitor of calpain I and II is not as effective as a knockdown of p53 in influencing the Cd induced proliferation arrest, although the protective effect is significant after 96h of treatment.In summary, neither the knock down of p53 nor the inhibition of calpains is able to completely reverse the Cd induced proliferation arrest (Figure S4).

Cd treatment does not impair endothelial cell cycle
To further analyse Cd induced signalling pathways leading to endothelial cell death, we incubated endothelial cells with 15µM and 30µM Cd respectively for 6h.After treatment the cells were subjected to flow cytometrybased cell cycle analyses.As shown in Figure S5, Cd treatment does not influence the endothelial cell cycle.Protein extracts of CTRL cells (wild type endothelial cells) and BCL-XL OE cells were prepared and equal protein amounts were loaded.Western blot based detection of the expression of BCL-XL was performed.Figure S11 shows that the expression BCL-XL is strongly over expressed in the lentiviral transfected cells compared to wild type endothelial cells.Detection of GAPDH loading control revealed a strongly increased loading of proteins in the BCL-XL OE lane compared to the CTRL lane.Nevertheless, calculation of expression ratios between BCL-XL and GAPDH showed the obvious OE of BCL-XL (ratio 0.019 in CTRL compared to 0.72 in BCL-XL OE cells). 12

Figure S1 :
Figure S1: Treatment of endothelial cells with different Cd concentrations increases intracellular Ca 2+ concentration over time.Figure S1 shows the flow cytometry-based quantification of intracellular Ca 2+ concentration in wild type endothelial cells after treatment with 15µM Cd and 30µM Cd respectively for 48h.All experiments were performed in triplicates and were repeated at least three times.Results depict the mean ± standard deviation.Asterisks indicate significant differences compared with the corresponding control ( p<0.05; p<0.01, p<0.001).

Figure S2 :
Figure S2: Intracellular ATP concentration of Cd treated endothelial cells.Endothelial cells were treated with 15µM and 30µM Cd respectively over a time period of 72h.Intracellular amount of ATP was quantified and analysed as percent per number of cells.All experiments were performed in quadruplicates and were repeated at least three times.Results depict the mean ± standard deviation.Asterisks indicate significant differences ( p<0.05; p<0.001) compared with the controls; ns = not significant.

Figure
Figure S3 A and B: Analysis of the effect of Cd treatment on the proliferation of wild type endothelial cells and cells with stable p53 KD.Endothelial cells were treated with 15µM and 30µM Cd respectively over a time period of 96h.The number of viable cells relating to the proliferation ability of endothelial cells was analysed using an XTT-assay.Figure S3 A depicts the proliferation of wild type endothelial cells under Cd treatment over 96h, whereas Figure S3 B shows the proliferation of p53 KD cells under Cd treatment over 96h.All experiments were performed in quadruplicates and were repeated at least three times.Results depict the mean ± standard deviation.Asterisks indicate significant differences compared to the corresponding controls ( p<0.05; p<0.001; ns = not significant) and hash signs between CTRLs (HUVECs) and p53 KD cells ( # # # p<0.001) ns = not significant.

Figure
Figure S4 A and B: Analysis of the effect of Cd treatment on the proliferation of wild type endothelial cells and cells incubated with an inhibitor of calpain I and II.Endothelial cells were treated with 15µM and 30µM Cd respectively over a time period of 96h.The number of viable cells relating to the proliferation ability of endothelial cells was analysed using an XTTassay.Figure S4 A shows the proliferation of CTRL cells treated with 15µM and 30µM Cd over 96h. Figure S4 B shows the proliferation of Cd treated cells pre-incubated with the calpain inhibitor over 96h.All experiments were performed in quadruplicates and were repeated at least three times.Results depict the mean ± standard deviation.Asterisks indicate significant differences compared to the corresponding controls ( p<0.05; p<0.001; ns = not significant) and hash signs between HUVECs (CTRL) and HUVECS pre-treated with an inhibitor of calpain I and II ( # # # p<0.001) ns = not significant.

Figure S5 :
Figure S5: Influence of Cd treatment on endothelial cell cycle.Endothelial cells were treated with 15µM and 30µM Cd respectively.After 6h of treatment, endothelial cells were subjected to flow cytometry-based cell cycle analyses.Figure S5 shows, that Cd incubation does not inhibit endothelial cell cycle.All experiments were performed in quadruplicates and were repeated at least three times.Results depict the mean ± standard deviation (ns = not significant).

Figure
Figure S6 A and B: Neither the inhibition of calpain activity, nor the inhibition of autophagy prevents Cd induced lysosomal breakdown.Endothelial cells were incubated with 50µM

Figure
Figure S7 A and B: Influence of calpain inhibition and 3MA incubation on Cd induced DNA degradation.Endothelial cells were pre-incubated with the two indicated inhibitors for 30 min before Cd treatment.Figure S7 A shows inhibition of calpain activity was not able to prevent DNA degradation, on the contrary a significant acceleration of DNA degradation could be observed.Likewise, pre-incubation of Cd treated cells with 3MA showed no preventive effect, but rather an enhancement of Cd induced DNA degradation (Figure S7 B).All experiments were performed in quadruplicates and were repeated at least three times.Results depict the mean ± standard deviation.Asterisks indicate significant differences compared to the corresponding controls ( p<0.05; p<0.01; p<0.001) and hash signs indicate significant differences between the groups ( # p<0.05; # # p<0.01; # # # p<0.001).

Figure
Figure S8 A and B: Influence of EGTA and ATA pre-incubation on the DNA content of Cd treated endothelial cells.Cells were pre-incubated either with 50µM ATA or 1mM EGTA for 30min and then subjected to Cd treatment.After 96h of treatment, DNA content was quantified by Flow cytometry-based analyses.Figure S8 A shows that ATA pre-incubation was not able to prevent Cd-induced DNA degradation, whereas EGTA pre-incubation completely protects Cd treated endothelial cells from DNA degradation (Figure S8 B). Results depict the mean ± standard deviation.Asterisks indicate significant differences compared to the corresponding controls ( p<0.05; p<0.01; p<0.001; ns = not significant) and hash signs between endothelial cells without and with inhibitor incubation ( # # p<0.01; # # # p<0.001); ns= not significant.

Figure S9 A
Figure S9 A-C: Rescue of Cd treated endothelial cells from cell death by incubation withFructose, Sucrose and Mannitol.Endothelial cells were treated with 15µM and 30µM Cd respectively and simultaneously incubated with increasing concentrations of Fructose, Sucrose, and Mannitol for 96h.FigureS9A shows that Fructose was not able to inhibit Cd-induced cell death significantly.In contrast, incubation of Cd-treated cells with Sucrose or Mannitol significantly reduces the amount of dead cells (FigureS9 B and C).Results depict the mean ± standard deviation.Asterisks indicate significant differences compared to the corresponding controls ( p<0.05; p<0.01; p<0.001; ns = not significant) and hash signs between CTRL and CTRL pre-treated with an inhibitor of calpain I and II ( # p<0.05) ns = not significant.

Figure S10 :
Figure S10: Western Blot based analyses of the expression level of p53 in CTRL and p53 KD cells.Equal amounts of protein extract form CTRL as well as p53 KD cells were separated, blotted onto a nitrocellulose membrane, and p53 protein was detected.Figure S10 shows that p53 was knocked down in the lentiviral transfected cells compared to the wild type endothelial cells.

Figure S11 :
Figure S11: Western Blot based analyses of the expression level of BCL-XL in CTRL and BCL-XL OE cells.Equal amounts of protein extract form CTRL as well as BCL-XL OE cellswere separated, blotted onto a nitrocellulose membrane, and BCL-XL protein was detected.FigureS11shows that BCL-XL was over-expressed in the lentiviral transfected cells compared to the wild type endothelial cells.