Various dihydropyridine effectively enhance Btz-mediated cell death in breast cancer

Multidrug resistance (MDR) is frequently appeared by the overexpression of P-glycoprotein (P-gp) pump in cancer cells (Zhou J et al., 2006). Therefore, we have explored the drug repositioning approach to identify candidate modulators of anticancer activities. Since some compounds of the 1,4-dihydropyridines (1,4-DHP), which are known to be anti-hypertensive agents with calcium channel blocker activities, have shown to have potential to kill cells by inhibiting P-gp pump (Viale M et al., 2011), we investigated whether 1,4-DHP could sensitize bortezomib (Btz) to breast cancer cells.

We found that when MDA-MB 435S (breast cancer) cells were treted with amlodipine (Amlo), niguldipine (Nigul), nicardipine (Nicar), and felodipine (Felo), and lercanidipine (Ler) up to 15 μM, cytotoxicity was not observed. But combined treatment with subtoxic doses of Btz and DHPs dose-dependently enhanced cell death in these cell lines (Figure 3). Isobologram analysis revealed that Btz and any DHPs synergistically induced cell death in these cells (Figure 4). Next, we tested whether combination of DHP and Btz affects the viability of normal cells. The combined treatment with Ler and Btz (Ler/Btz) did not affect the viability of MCF-10A cells, whereas Amlo/Btz, Nigul/Btz, Nicar/Btz, or Felo/Btz slightly reduced it (Figure 5). These results suggest that Ler may more effectively and safely sensitize cancer cells to Btz-mediated cell death than other DHPs.

２４

Fig. 3. 1,4-DHPs sensitize MDA-MB 435S cells to bortezomib-mediated cell death.

MDA-MB 435S cells were treated with the indicated concentrations of Btz and/or Amlo, Nigul, Nicar, Felo, or Ler for 24 h and cellular viability was assessed using IncuCyte.

The percentage of live cells was normalized to that of untreated control cells (100%).

Data represent the means ± S.D. (n=7). One way ANOVA and Bonferroni’s post hoc test. *p <0.001 vs. Btz-treated cells.

２５

Fig. 4. Synergistic induction of cancer cell death by DHPs and Btz.

MDA-MB 435S cells were treated with various concentrations of various DHPs and/or Btz for 24 h. Isobologram analysis was performed as described in MATERIALS AND METHODS. Isoboles for the combination of CCB and Btz, which were iso-effective (IC50) for inhibition of cell viability, are shown.

２６

Fig. 5. Effects of DHP and/or Btz on the viability of normal breast cells.

MCF-10A cells were treated with the indicated concentrations of DHP and/or Btz for 24 h and cellular viability was assessed using IncuCyte as described in MATERIAL AND METHODS. The percentage of live cells was normalized to that of untreated control cells (100%). Data represent the means ± S.D. (n=7).

One-way ANOVA and Bonferroni’s phost hoc test. *p <0.001 vs. Btz-treated cells.

２７

2. Lercanidipine effectively enhances PI-mediated cell death in various cancer cells

We investigated whether Ler can effectively sensitize other types of cancer cells to Btz-mediated cell death. Cytotoxicity of Ler up to 15 μM was not observed in SNU-668 (stomach cancer), NCI-H460 (lung cancer), BxPC-3 (pancreatic cancer), or SNU-475 (liver cancer) cells, but combined treatment with subtoxic doses of Btz and Ler dose-dependently enhanced cell death in these cancer cell lines (Figure 6). Isobologram analysis revealed that Btz and Ler synergistically induced cell death in these cells (Figure 7). Since Btz is widely used in the patients with multiple myeloma (MM), we investigated whether Ler could enhance the Btz-mediated cell death in the MM cells, RPMI 8226. We found that treatment with Ler dose-dependently increased Btz-mediated cell death in these MM cells (Figure 8A), showing synergistic effects (Figure 8B). We further tested whether Ler affects the viability of MDA-MB 435S cells treated with two other FDA-approved PIs, including calfilzomib (Cfz) and ixazomib (Ixz). We found that Ler effectively enhanced the cell death mediated by these proteasome inhibitors (PIs) too (Figure 9A), showing synergistic effects (Figure 9B). Interestingly, Ler did not increase the death of MCF-10A (normal breast) cells treated with Cfz or Ixa (Figure 10).

Furthermore, Ler had no cytotoxic effects on Chang (normal liver) cells treated with Btz (Figure 10). These results suggest that Ler can safely and effectively sensitize to various cancer cells to PIs.

２８

Fig. 6. Combination of Ler and Btz induces Btz-mediated cell death in various cancer cells.

SNU-668, NCI-H460, BxPC-3, and SNU-475 cells were treated with the indicated concentrations of Btz and/or Ler for 24 h and cellular viability was assessed using IncuCyte as described in MATERIALS AND METHODS. The percentage of live cells was normalized to that of untreated control cells (100%). Data represent the means ± S.D. (n=7). One-way ANOVA and Bonferroni’s post hoc test. *p<0.001 vs. Btz-treated cells.

２９

Fig. 7. Ler and Btz synergistically induce cell death in various cancer cells.

Cells were treated with various concentration of Ler and/or Btz for 24 h. Isobologram analysis was performed as described in MATERIALS AND METHODS. Isoboles for the concentration of Btz and Ler that proved iso-effective (IC50) for inhibiting cell viability.

３０

Fig. 8. Ler sensitizes to Btz-mediated cell death in multiple myeloma cells.

(A) RPMI 8226 cells were treated with the indicated concentrations of Btz and/or Ler for 24 h and cellular viability was assessed using MTT assay as described in MATERIALS AND METHODS. The percentage of live cells was normalized to that of untreated control cells (100%). Data represent the means ± S.D. (n=7). One-way ANOVA and Bonferroni’s post hoc test. *p<0.001 vs. Btz treated cells. (B) Isobologram was performed as described in MATERIALS AND METHODS. Isoboles for the concentration of Btz and Ler that proved iso-effective (IC50) for inhibiting cell viability.

３１

Fig. 9. Ler sensitizes to the proteasome inhibitor-mediated cell death in breast cancer cells.

(A) MDA-MB 435S cells were treated with the indicated concentrations of PIs and/or Ler for 24 h and cellular viability was assessed using IncuCyte as described in MATERIALS AND METHODS. The percentage of live cells was normalized to that of untreated control cells (100%). Data represent the means ± S.D. (n=7). One-way ANOVA and Bonferroni’s post hoc test. *p<0.001 vs. Btz-treated cells. (B) Isobologram was performed as described in MATERIALS AND METHODS. Isoboles for the concentration of PIs and Ler that proved iso-effective (IC₅₀) for inhibiting cell viability.

３２

Fig. 10. Combined treatment with Ler and PIs did not increase cell death in normal cells.

MCF-10A and Chang cells were treated with the indicated concentrations of PIs and/or Ler for 24 h and cellular viability was assessed using IncuCyte as described in MATERIALS AND METHODS. The percentage of live cells was normalized to that of untreated control cells (100%). Data represent the means ± S.D. (n=7).

３３

3. Combined treatment with Ler and Btz induces apoptotic, non-necroptotic, and non-autophagic cell death

To understand how Ler overcomes the resistance of the cancer cells to PIs, we observed cell morphology after treatment with DHP and/or Btz. We found that treatment with 4 nM Btz or 10 μM Amlo, Nicar, Nigul, Felo, and Ler for 16 h did not induce morphological change in MDA-MB 435S cells. Combinations of DHP and Btz induced the extensive vacuolation in MDA-MB 435S cells (Figure 11 and Figure 12). Btz-Ler did not induce any vacuolation or cell death in MCF-10A cells.

Btz/Amlo, Btz/Nicar, Btz/Nigul, or Btz/Felo induced cytoplasmic vacuolation in MCF-10A cells, but at a much lesser extent than that in MDA-MB 435S cells. In contrast, treatment with Ler/Btz, but not in single treatment, induced a dramatic cellular vacuolization in SNU-668, NCI-H460, BxPC-3, and SNU-475 cells (Figure 13). Moreover, not only Ler/Cfz or Ler/Ixa but also any tested DHP/Btz induced an extensive vacuolation and subsequent cell death in MDA-MB 435S cells, but not in MCF-10A cells (Figure 14). These results suggest that the combined treatment with DHP and PI may commonly induce vacuolation-mediated cell death in these cancer cells, while conserving normal breast cells. Next, we tested whether Ler/Btz induces apoptosis. Ler/Btz-induced cell death was not accompanied by the apoptotic morphologies, including cell blebbing, formation of apoptotic bodies (Figure 15A) and pretreatment with the pan-caspase inhibitor, z-VAD-fmk, did not affect Ler/Btz-induced cellular vacuolation and subsequent cell

３４

death in MDA-MB 435S cells (Figure 15A and 15B). In addition, while doxorubicin treatment induced the cleavage of p32 procaspase-3 into p20 (intermediate form) and p17 subunit of caspase-3, Ler/Btz treatment did not affect the proteolytic processing of procaspase-3 in MDA-MB 435S cells (Figure 15C).

Furrthermore, pretreatment with necroptosis inhibitor, necrostatin-1 (Nec-1), or autophagy inhibitors, 3-methyladenine (3-MA) and bafilomycin A1 (BafA1), did not inhibit Ler/Btz-induced cellular vacuolation (Figure 15A) and cell death (Figure 15B). Moreover, Ler did not notably affect Btz-meditaed upregulation of LC3 and p62, suggesting that the sensitizing effect of Ler on Btz-mediated cell death is not associated with autophagy (Figure 15D). Taken together, these results suggest that apoptosis, necroptosis, or autophagic cell death is not critically involved in the anticancer effects of Ler/Btz.

３５

Fig. 11. The morphological changes in MDA-MB 435S and MCF-10A cells treated with Amlo/Btz, Nicar/Btz, Nigul/Btz, or Felo/Btz.

Morphologies of MDA-MB 435S or MCF-10A cells treated with Amlo, Nicar, Nigul, Felo and/or Btz for 24 h were observed by phase-contrast microscopy. Bars, 20 μm.

３６

Fig. 12. Combined treatment with Ler and Btz induces cellular vacuolation in breast cancer cells, but not in normal breast cells.

Cells were treated with 10 μM Ler and/or 4 nM Btz for indicated time points and cellular morphologies were observed by phase-contrast microscopy. Bars, 20 μm.

３７

Fig. 13. Combined treatment with Ler and Btz induces cellular vacuolation in various cancer cells.

Cellular morphologies were observed by phase-contrast microscopy. Bars, 20 μm. Cells were treated with Ler and/or Btz (for SNU-668 cells, 4 nM Btz and/or 10 μM Ler; for NCI-H460 cells, 15 nM Btz and/or 15 μM Ler; for BxPC-3 and SNU-475 cells, 20 nM Btz and/or 10 μM Ler) for 24 h.

３８

Fig. 14. The morphological changes in breast cancer cells or normal breast cells treated with Ler/Cfz or Ler/Ixz.

Cells were treated with 10 μM Ler and/or PIs (20 nM Cfz or 100 nM Ixz) for 24 h and cellular morphologies were observed by phase-contrast microscopy. Bars, 20 μm.

３９

Fig. 15. Ler/Btz-induced vacuolization and cell death is not associated with apoptosis, necroptosis, or autophagic cell death.

(A) MDA-MB 435S cells were pretreated with 20 μM z-VAD, 20 μM Nec-1, 0.5 mM 3-MA, or 20 nM Bafilo and further treated with 10 μM and 4 nM Btz for 12 h. Cells were observed by phase-contrast microscopy. Bars, 20 μm. (B) MDA-MB 435S cell were pretreated with the indicated concentrations of z-VAD-fmk (z-VAD), necrostatin-1 (Nec-1), 3-methyladenine (3-MA), or bafilomycin A1 (Baf), and further treated with Ler/Btz for 24 h. Cellular viability was assessed using IncuCyte. Data represent the means ± S.D. (n=7). One-way ANOVA and Bonferroni’s post hoc test. *p <0.001 vs.

untreated cells, control. (C and D) Cells were treated with 10 μM Ler and/or 4 nM Btz, or 5 μg/ml doxorubicin (Doxo.) for 24 h. Western blotting of the indicated proteins was performed with β-actin used as a loading control.

４０

4. Combination of Ler and Btz induces paraptosis

We investigated whether Ler/Btz-induced vacuolation is derived from the organelles, including the endoplasmic reticulum (ER) and/or mitochondria. To observe these structures, we performed confocal microscopy in YFP-ER cells (MDA-MB 435S sublines stably transfected with the YFP-ER plasmid) after stained Mitotracker-Red (MTR). In untreated cells, we observed filamentous mitochondria and reticular morphology of the ER (Figure 16), and in treated with 4 nM Btz did slightly reduce the mitochondrial length (Figure 17). Although in treated with 10 μM Ler, we did not observe morphological change in the ER, mitochondrial length was shortened at 8 h and mitochondrial dilation was slightly increased at 16 h. But, at 24 h, filamentous morphology of mitochondria was restored (Figure 17). Interestingly, combined treatment with Ler/Btz-treated cells exhibited a slight dilation of mitochondria at 8 h, and gradually increased mitochondrial swelling and the size of enlarged mitochondria at 16 h (Figure 16). Since the fluorescence of MTR was weakened by Ler/Btz treatment at 16 h, we tested whether combined treatment with Ler/Btz induces the loss of mitochondrial membrane potential (MMP). After Ler/Btz treatment for 24 h, we observed the initiation of death-mediated cellular detachment. To further study the relationship between mitochondrial morphology and MMP after Ler/Btz treatment, we performed confocal microscopy in YFP-Mito cells treated with Ler and/or Btz then stained with tetramethylrhodamine methyl ester (TMRM) (Figure 18 and Figure 19).

We found that mitochondrial dilation showed a peak at 16 h of Ler/Btz treatment

４１

(Figure 19). Interestingly, many expanded mitochondria lost MMPs (green mitochondria, blue arrows), although some of them still retained MMPs (yellow mitochondria, white arrow heads). At 24 h of Ler/Btz treatment, most mitochondria were showed irregularly fragmented morphology and weaker TMRM fluorescence, indicating the MMP loss. In adddtion, dilation of the ER was observed in Ler/Btz-treated cells at 12 h (Figure 16). The size of extended the ER was increased from 16 h and it was maintained 24 h.

Since the vacuolation of the ER and mitochondria is the morphological feature of paraptosis (Sperandio S et al., 2000; Sperandio S et al., 2004; Lee D et al., 2016), and paraptosis is known to require de novo protein syntyesis (Sperandio S et al., 2000;

Sperandio S et al., 2004), we examined whether pretreatment with protein synthesis blocker, cycloheximide (CHX), affects the Ler/Btz-induced vacuolation and cell death in these cells. We found that CHX pretreatment significantly blocked the cell death (Figure 20A) and vacuolization (Figure 20B) induced by Ler/Btz in MDA-MB 435S cells. In addition, pretreatment with CHX effectively suppressed the expansion of mitochondria and the ER in YFP-ER and YFP-Mito cells treated with Ler/Btz for 12 h (Figure 20C). Collectively, these results suggest that combined treatment with Ler and Btz eliminates various cancer cells by induction of paraptosis.

４２

Figure 16. Combined treatment with Ler/Btz induces the ER and mitochondrial dilation in cancer cells.

YFP-ER cells were treated with 10 μM Ler and 4 nM Btz for indicated time durations and then stained with MTR. Cells were observed by confocal microscopy. Bars, 20 μm.

４３

Figure 17. Morphological changes in cells treated with Btz or Ler.

YFP-ER cells treated with 4 nM Btz or 10 μM Ler for the indicated time durations and then stained with MTR. Cells were observed by confocal microscopy. Bars, 20 μm.

４４

Figure 18. Morphologies and membrane potential of mitochondria in cells treated with Btz or Ler alone.

YFP-Mito cells treated with 4 nM Btz or 10 μM Ler for the indicated time durations and then stained with TMRM. Cells were observed by confocal microscopy. Bars, 20 μm.

４５

Figure 19. Combined treatment with Ler and Btz induces mitochondrial vacuolization and MMP loss in cancer cells.

YFP-Mito cells were treated with 10 μM Ler and 4 nM Btz for the indicated time durations and then stained with TMRM. Cells were observed by confocal microscopy.

Bars, 20 μm.

４６

Figure 20. Pretreatment with CHX inhibits Ler/Btz-induced cytoplasmic vacuolation and cell death in cancer cells.

(A) MDA-MB 435S cells were pretreated with CHX and then further treated with Ler and/or Btz at the indicated concentrations for 24 h. Cellular viability was assessed using IncuCyte system. Data represent the means ± S.D. (n=7). One-way ANOVA and Bonferroni’s post hoc test. *p <0.001 vs. untreated cells; #p <0.05 vs. Ler/Btz-treated cells. (B) Cellular morphologies were observed by phase-contrast microscopy. Bars, 20 μm. (C) YFP-ER and YFP-Mito cells were pretreated with 2 μM CHX and further treated with 10 μM Ler plus 4 nM Btz for 12 h. Cells were observed by confocal microscopy. Bars, 20 μm.

４７

5. Combined treatment with Ler and Btz enhances ubiquitinated protein accumulation and ER stress

It is well known that major mechanism of PI-mediated cell death involves the accumulation of toxic poly-ubiquitinated proteins and misfolded protein aggregates, which induce ER stress (Ding WX et al., 2007). Therefore, we investigated whether the underlying mechanism of Ler/Btz-mediated anticancer effect was also related to ER stress. Western blotting showed that Ler enhances the Btz-mediated accumulation of poly-ubiquitinated protein (Figure 21A). Immunocytochemistry of ubiquitin revealed that while treatment with Btz formed aggresome-like structures in MDA-MB 435S cells, combined treatment with Ler and Btz induced more scattered expression of ubiquitinated protein aggregates for 16 h (Figure 21B). In addition, Ler increased the Btz-induced phosphorylation of PERK and eIF2α, and up-regulation of ATF4 and CHOP protein levels (Figure 21). Interestingly, levels of the ER stress marker proteins was much higher in breast cancer cells treated with Ler/Btz, compared to those in normal breast cells (Figure 22). These results suggest that Ler can sensitize cancer cells to Btz by enhancing Btz-mediated ER stress.

４８

Figure 21. Ler enhances Btz-mediated ubiquitinated protein and ER stress.

(A) MDA-MB 435S cells were treated with 10 μM Ler and/or 4 nM Btz for indicated time durations. Western blotting of the indicated proteins was performed with β-actin used as a loading control. (B) MDA-MB 435S cells were treated with 10 μM Ler and/or 4 nM Btz for 16 h. Immunocytochemistry of Ub was performed. Cells were observed by confocal microscopy. Bars, 20 μm.

４９

Figure 22. The expression levels of ER stress marker proteins in Ler/Btz-treated breast cancer cells or normal breast cells.

MDA-MB 435S or MCF-10A cells were treated with 10 μM Ler and/or 4 nM Btz for 8 h. Western blotting of the indicated proteins was performed with β-actin used as a loading control.

５０

6. Combination of Ler and Btz disrupts Ca²⁺ homeostasis via mitochondrial Ca²⁺ overload

We previously reported that Ca²⁺ homeostasis, particularly mitochondrial Ca²⁺

overload, plays an important role in the paraptosis induced by curcumin (Yoon MJ et al., 2012) or celastrol (Yoon MJ et al., 2014). In addition, Ler is known to inhibit

L-type Ca²⁺ channels (Burnier M et al., 2009; Bang LM et al., 2003), and Therefore, we tested whether Ler and/or Btz altered intracellular Ca²⁺ levels.

Fluorescence microscopy (Figure 23A) and flow cytometry (Figure 23B) using Fluo-3 showed that combination of Ler and Btz progressively and markedly increased cytosolic Ca²⁺ levels. In contrast, Btz or Ler alone triggered very slight increase of intracellular Ca²⁺ at 16 h (Figure 23A and 23B). Next, when we assessed the changes in mitochondrial Ca²⁺ levels using Rhod-2, Ler/Btz markedly increased mitochondrial Ca²⁺ levels from 4 h with a peak at 12 h (Figure 24). Thus, these results indicate that the increase in mitochondrial Ca²⁺ preceded the increase in cytoplasmic Ca²⁺ levels in response to Ler/Btz.

We next investigated the functional importance of increased cytosolic Ca²⁺ in Ler/Btz-induced paraptosis using BAPTA-AM, a scavenger of intracellular Ca²⁺. Since Ca²⁺ is mainly taken up into mitochondria through the mitochondrial Ca²⁺

uniporter (MCU) (De Stefani D and Rizzuto R, 2014; De Stefani D et al., 2015), we examined the effect of MCU inhibitors, such as Ru360 or ruthenium red (RR), in Ler/Btz-induced paraptosis. We found that pretreatment with Ru360, RR, or

５１

BAPTA-AM dose-dependently inhibited Ler/Btz-induced cell death, although the blocking effect of Ru360 or RR was greater than that of BAPTA-AM (Figure 25A).

We also investigated the effects of Ru360 or BAPTA-AM on the Ler/Btz-induced vacuolization of mitochondria and the ER. Interestingly, pretreatment with Ru360 effectively blocked the dilation of both mitochondria and the ER by Ler/Btz treatment. In contrast, BAPTA-AM partially inhibited mitochondrial dilation, whereas it potently inhibited the expansion of the ER (Figure 25B). Next, since Ca²⁺ was shown to be imported across the outer mitochondrial membrane (OMM) through VDAC (Rizzuto R et al., 2009), we investigated whether pretreatment with DIDS, a VDAC inhibitor (Benítez-Rangel E et al., 2015), blocks the Ler/Btz-induced cell death. In contrast to the effect of Ru360, DIDS did not affect this cell death (Figure 26A). Also, we investigated whether increased Ca²⁺ originates from the ER or extracellular environment following treatment with Ler/Btz.

Pretreatment with 2-APB, an IP₃ receptor inhibitor (Maruyama T et al., 1997), did not block the Ler/Btz-induced cell death. Moreover, high concentration of 2-APB (about 80 μM), which is shown to inhibit full name (SOCE) (DeHaven WI et al., 2008), did not block the Ler/Btz-induced cell death. Taken together, these results suggest that initially the increase in mitochondrial Ca²⁺ may critically contribute to Ler/Btz-induced paraptosis, while later increase in cytosolic Ca²⁺ may be important for the ER dilation triggered by Ler/Btz.

５２

Figure 23. Combined treatment with Ler and Btz markedly increases cytosolic Ca²⁺ levels.

MDA-MB 435S cells were treated with 10 μM Ler and/or 4 nM Btz for the indicated time points. Treated cells were stained with Fluo-3 and subjected to fluorescence microscopy (A) and flow cytometry (B). Bars, 20 μm.

５３

Figure 24. Combined treatment with Ler and Btz significantly increases

Figure 24. Combined treatment with Ler and Btz significantly increases