The combination of PF-429242 and chloroquine triggers pH-dependent cell death in hepatocellular carcinoma cells

TL;DR

A combination of PF-429242 and chloroquine causes pH-dependent cell death in liver cancer cells, offering a new treatment strategy.

Contribution

The study reveals a novel pH-dependent cell death mechanism in HCC cells not involving apoptosis or autophagy.

Findings

The combination of PF-429242 and chloroquine shows significant cytotoxicity against HCC cells.

The observed cell death is pH-dependent and distinct from alkaliptosis.

The treatment induces necrosis-like morphology and membrane rupture preventable by acidifying the medium.

Abstract

Hepatocellular carcinoma (HCC) remains a significant health challenge due to its resistance to conventional treatments and high recurrence rates. Developing novel therapeutic strategies is critical for improving outcomes for HCC patients. In this study, we identified the synergistic anticancer activity of the combination of PF-429242 and chloroquine against HCC cells. Combined treatment exhibited significant cytotoxicity against HCC cell lines, which was not observed with other therapeutic drugs. Notably, this synergistic effect was not mediated through apoptosis or autophagy. Further investigation revealed that the combination induced pH-dependent cell death, distinct from the previously described alkaliptosis. Unlike alkaliptosis, this cell death mechanism did not involve intracellular alkalinization or the IKKβ/NF-κB/CA9 signaling pathway. We also found that the ATP6V0D1/STAT3 axis, implicated in alkaliptosis, was not crucial for PF-429242/chloroquine-induced cell death. Additionally, site-1 protease inhibition by PF-429242 was not responsible for the observed synergistic effect. While the exact mechanism remains unclear, combined treatment induced a necrosis-like morphology and membrane rupture, which could be prevented by acidifying the culture medium. This research highlighted a novel pH-dependent cell death mechanism in HCC cells and suggests potential therapeutic implications for combining PF-429242 and chloroquine in cancer treatment.

Figures

Full text

Introduction

Liver cancer represents a significant and escalating global health challenge. Among primary liver cancers, hepatocellular carcinoma (HCC) is the most common, accounting for 75%-85% of cases. Key risk factors contributing to HCC development include chronic infections with hepatitis B virus (HBV) or hepatitis C virus (HCV), as well as chronic liver conditions such as nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) 1. Standard treatment options for HCC encompass surgical resection, ablation, liver transplantation, and systemic pharmacological therapies. However, these options are only suitable for 15%-25% of patients and recurrent HCC occurs in over 70% of individuals following resection 2,3. In addition, HCC tends to be resistant to chemotherapy 4,5. New therapeutic options for advanced HCC include kinase and immune checkpoint inhibitors, but they only produce a temporary improvement in median overall survival 6,7. Therefore, novel therapeutic strategies are still urgently needed.

A significant clinical barrier to effective cancer treatment is the resistance of cancer cells to chemotherapy. Apoptosis initially contributes to the response of tumors to therapy, but cancer cells might later survive and develop apoptosis resistance 8. Thus, developing effective cancer therapy to overcome apoptosis resistance is urgently needed. New anticancer strategies may be developed by targeting non-apoptotic cell death, such as necroptosis, autophagy-dependent cell death, ferroptosis, and other regulated cell death (RCD) pathways 9. In particular, a newly identified form of RCD, known as alkaliptosis, which is pH-dependent, has emerged as a promising therapeutic strategy for targeting various types of cancer 10.

Our previous study showed that PF-429242, a site-1 protease inhibitor 11, exerts anticancer activity against HCC cells by inducing autophagic cell death 12. In this study, interestingly, the blockade of autophagic flux with chloroquine, a lysosomotropic agent 13-16, further enhanced the anticancer activity of PF-429242 by triggering pH-dependent cell death. Our results provide an innovative strategy for HCC therapy.

Material and methods

Chemicals and reagents

Minimum essential medium (MEM; #11095080), fetal bovine serum (FBS; #26140079), sodium pyruvate (#11360070), non-essential amino acids (NEAA; #11140050), alamarBlue cell viability reagent (#DAL1100), M-PER mammalian protein extraction reagent (#78501), pHrodo Green AM Intracellular pH Indicator (#P35373), Live Cell Imaging Solution (#A59688DJ), Intracellular pH Calibration Buffer Kit (#P35379), Lipofectamine 3000 (#L3000015), and RNAiMAX (#13778150) transfection reagents were purchased from Thermo Fisher Scientific. IκB (#GTX110521), phospho-Ser536-nuclear factor (NF)-κB/p65 (#GTX133899), ATP6V0D1 (GTX118317), STAT3 (#GTX104616), GFP (#GTX113617), and GAPDH (#GTX627408) antibodies were purchased form GeneTex. Poly ADP-ribose polymerase (PARP; #9542) and MAP1LC3B (LC3B; #2775) antibodies were purchased from Cell Signaling Technology. Caspase-3 (#ab13585) and carbonic anhydrase 9 (CA9; #ab184006) antibodies were purchased from Abcam. An anti-Flag antibody (#F3165) was purchased from Merck. Horseradish peroxidase (HRP)-labeled secondary antibodies were purchased from Jackson ImmunoResearch. Chloroquine (#C6628), N-acetylcysteine (NAC; #A7250), Trolox (#238813), ferrostatin-1 (#SML0583), deferoxamine (#D9533), cobalt chloride (#15862-1ML-F), and dimethyl sulfoxide (DMSO; #D5879) were purchased from Sigma. PF-429242 (#A11230), SC-514 (#A12746), IMD0354 (#A12837), MCC950 (#A13478), CA-074Me (#A13593), necrostatin-1 (#A11973), and Z-VAD-FMK (#A12373) were purchased from Adooq BioScience. Doxorubicin (#D-4000), sorafenib (#S-8502), regorafenib (#R-8024), and bafilomycin A1 (#B-1080) were purchased from LC Laboratory. JTC801 (#HY-13274) was purchased from MedChemExpress. A 7-aminoactinomycin D (7-AAD) viability staining solution (#402404) was purchased from BioLegend. A GENzol TriRNA kit (#GZX200) and RNase-free DNase I set (#DNS300) were purchased from Geneaid. IQ2 SYBR Green Fast qPCR System Master Mix (#DBU-006) was purchased from Bio-Genesis. An iScript cDNA synthesis kit (#1708891) was purchased from Bio-Rad Laboratories.

Cell culture

The PLC/PRF/5 (PLC5; #60223) and HepG2 (#RM60025) cell lines were purchased from the Bioresource Collection and Research Center (BCRC; Hsinchu City, Taiwan). Cells were maintained in MEM supplemented with 0.1 mM NEAA, 1.0 mM sodium pyruvate, and 10% FBS. Cultures were incubated at 37°C in a humidified atmosphere with 5% CO₂.

Cell viability assay

Assessment of cell viability was conducted using the alamarBlue cell viability reagent, a solution based on resazurin. Resazurin is a safe, cell-permeable compound initially blue in color that exhibits minimal fluorescence. Upon entry into living cells, it undergoes reduction to resorufin, thereby acquiring a red coloration and significant fluorescence. In this study, alterations in fluorescence, with excitation at 530-560 nm and emission at 590 nm, were employed to ascertain cell viability. The combined effect of drugs was calculated with Combenefit software 17.

7-AAD Staining Assay

The integrity of plasma membranes was assessed by determining the ability of cells to exclude 7-AAD, a membrane-impermeable, fluorescent DNA probe. 7-AAD/DNA complexes can be excited by a 488-nm laser and has an emission maximum of 647 nm. After treatment, cells were stained with 1 μg/ml 7-AAD for 15 min in the dark. Levels of 7-AAD staining were observed under a fluorescence microscope.

pH measurement

Extracellular pH values in culture media were measured with a pH meter. Intracellular pH values were assessed with the pHrodo Green AM Intracellular pH Indicator following the manufacturer's instructions. In brief, cells were plated onto a 96-well plate with a black wall and a clear bottom. After drug treatment, cells were rinsed with Live Cell Imaging Solution (LCIS) and subsequently loaded with the pH indicator within LCIS. Following a 30-min incubation at 37 °C, the fluorescence was measured by excitation at 509 nm and emission at 533 nm. Intracellular pH values were determined using the Intracellular pH Calibration Buffer Kit.

Transient Transfection

Transient plasmid overexpression (forward transfection) and small interfering (si)RNA knockdown (reverse transfection) analyses were respectively performed using Lipofectamine 3000 and RNAiMAX transfection reagents according to the manufacturer's instructions. After transfection for 24-48 h, cells were used for further experiments. siRNA specific for the LC3B (#sc-43390), ATP6V0D1 (#sc-63207), MBTPS1 (#sc-36496), and MBTPS2 (#sc-41652) genes were purchased from Santa Cruz Technology. pEGFP-STAT3 (#111934), pcDNA3-SREBP1a (#26801), pcDNA3-SREBP1c (#26802), and pcDNA3-SREBP2 (#26807) plasmids were purchased from Addgene. The pCMV3-CA9 (#HG10107-UT) plasmid was purchased from Sino Biological.

Real-time Quantitative Polymerase Chain Reaction (qPCR)

Total RNA was extracted using a GENzol TriRNA kit, following the manufacturer's instructions. First-strand cDNA synthesis was performed with an iScript cDNA synthesis kit, and a qPCR was carried out using IQ2 SYBR Green Fast qPCR System Master Mix on a LightCycler 96 System (Roche, Indianapolis, IN, USA) in triplicate. Primer sequences for amplification included MBTPS1 (forward: 5′-TCCAATTGCTTGGATGACAG-3′; reverse: 5′-TCCAGAACCTTGGAGTACCG-3′), MBTPS2 (forward: 5′-ACCCGTCAATCAACTGACCT-3′; reverse: 5′-TGCCAGATACCTGCACAAAA-3′), and 18S ribosomal (r)RNA (forward: 5′-CGGCGACGACCCATTCGAAC-3′; reverse: 5′-GAATCGAACCCTGATTCCCCGTC-3′). No-reverse transcription controls were included to ensure the absence of contamination. Gene expression levels were quantified using the comparative CT (ΔΔCT) method, with 18S rRNA as the reference gene and untreated cells as the calibrator.

Western Blotting

Total cellular proteins were extracted using the M-PER Mammalian Protein Extraction Reagent, followed by a 30-min incubation on ice. Lysates were centrifuged at 13,000 ×g for 30 min at 4 °C, and the supernatant was collected. After determining protein concentrations using the Bio-Rad Protein Assay, equal amounts (50 μg) of protein were resolved on a 7%-12% sodium dodecyl sulfate (SDS)-polyacrylamide gel and transferred to a nitrocellulose membrane. The membrane was incubated with the appropriate primary antibody overnight at 4°C, washed thoroughly, and then treated with a HRP-conjugated secondary antibody for 30 min at room temperature. Immunoblots were visualized using an enhanced chemiluminescence (ECL) reagent.

Statistical analysis

All data are presented as the mean ± standard deviation (SD), calculated from at least three independent experiments. Statistical analysis was performed using a two-tailed paired Student's t-test, with p < 0.05 considered indicative of statistical significance.

Results

Combined treatment with PF-429242 and chloroquine induces synergistic anticancer activity

Chloroquine is a lysosomotropic agent that readily crosses the lysosomal membrane, becomes protonated, and accumulates within the lysosome, resulting in an increase in the lysosomal pH, which then inactivates acid hydrolase enzymes 18,19. Chloroquine is commonly used to block autophagic flux. Our previous study identified that PF-429242 induced autophagic cell death in HCC cells 12. We incidentally found that a combination of PF-429242 and chloroquine exhibited synergistic anticancer activity against two HCC cell lines: PLC/PRF/5 (hereafter denoted as PLC5) and HepG2 cells (Figure 1A) The synergism between PF-429242 and chloroquine was calculated with Combenefit software 17 using the Bliss method (Figure 1B). To confirm the effect of chloroquine, another lysosomotropic agent (bafilomycin A1) was used. Bafilomycin A1 can either inhibit vacuolar-type H+-ATPase (V-ATPase) that acidifies the lysosome or disrupt calcium-dependent autophagosome-lysosome fusion 20. Like chloroquine, bafilomycin A1 also enhanced PF-429242- induced cytotoxicity (Figure 1C). To investigate whether PF-429242 also synergizes with other therapeutic drugs to kill HCC cells, doxorubicin was used to treat PLC5 and HepG2 cells in combination with PF-429242. However, no synergistic anticancer effect was found (Figure 1D, E). In addition, sorafenib and regorafenib, two molecular-targeted agents approved for HCC treatment 21,22, did not exhibit synergistic anticancer activity when combined with PF-429242 (Figure 1F).

Apoptosis and autophagy are not involved in the synergistic anticancer activity of PF-429242 and chloroquine

To investigate whether apoptosis was involved in the synergistic anticancer activity of PF-429242 and chloroquine, Western blotting was performed to analyze the level of apoptosis by the cleavages of PARP and caspase-3. Unlike doxorubicin (a positive control for inducing apoptosis), treatment with PF-429242, chloroquine, and their combinations did not induce apoptosis (Figure 2A), suggesting that other forms of cell death were responsible for the synergistic anticancer activity of PF-429242 and chloroquine. We also confirmed the induction of autophagy by PF-429242 as indicated by the accumulation of LC3B-II (Figure 2A). Chloroquine enhanced PF-429242-induced LC3B-II accumulation (Figure 2A), validating the blockade of autophagic flux. To ascertain the role of autophagy in the synergistic anticancer activity of PF-429242 and chloroquine, endogenous LC3B expression and PF-429242-induced LC3B-II accumulation were knocked down by siRNA (Figure 2B). However, LC3B-knockdown did not affect the cell viability in HCC cells treated with PF-429242 and chloroquine (Figure 2C). Therefore, neither apoptosis nor autophagy was involved in the synergistic anticancer activity of PF-429242 and chloroquine.

Combined treatment with PF-429242 and chloroquine induces pH-dependent cell death

To systematically investigate the type of cell death responsible for PF-429242/chloroquine-induced synergistic anticancer activity, we employed various inhibitors of apoptosis (Z-VAD-FMK), necroptosis (necrostatin-1), ferroptosis (deferoxamine, ferrostatin-1, Trolox which is a water-soluble analog of vitamin E, and NAC), pyroptosis (MCC950), lysosome-dependent cell death (CA-074Me, deferoxamine, and NAC), alkaliptosis (NAC, IMD0354, and SC-514), and oxeiptosis (NAC) (as reviewed in 23). Among these inhibitors, NAC is a synthetic precursor of the endogenous cellular antioxidant glutathione (GSH). Because NAC is a weak acid, the cell culture medium became acidic (pH 6.7) when we prepared the working solution. We adjusted the pH to 7.2 and NAC in both neutral and acidic conditions was used to treat cells. As shown in Figure 3A, PLC5 and HepG2 cells were only rescued from PF-429242/chloroquine-induced cytotoxicity by acidic and neutral NAC. Other antioxidants, such as ferrostatin-1 and Trolox, did not exhibit a rescue effect, suggesting that the antioxidant activity of NAC was not involved in its cytoprotective effect. It was reported that the low pKa (3.252) of NAC results in time-dependent acidification of NAC-containing culture media 24. Therefore, it is possible that the combination of PF-429242 and chloroquine induces intracellular alkalinization and then cell death (alkaliptosis), which can be reversed by the pH-lowering effect of NAC. Surprisingly, IMD0354 and SC-514, two I-kappa-B-kinase beta (IKKβ) inhibitors that inhibit JTC801-induced alkaliptosis 25, failed to rescue cells from PF-429242/chloroquine-induced cytotoxicity (Figure 3A), indicating that one or more alternative mechanisms may be involved. Similar to JTC801-induced alkaliptosis 25, a combination of PF-429242 and chloroquine induced a necrosis-like morphology (plasma membrane rupture) as detected by the staining of 7-AAD, a plasma membrane-impermeable fluorescent DNA-binding dye (Figure 3B). To confirm that a combination of PF-429242 and chloroquine induced pH-dependent cell death, PLC5 and HepG2 cells were cultured in media adjusted to different pH values. As shown in Figure 3C and 3D, the adjustment of the culture media's pH values from pH 7.4 to 6.0 almost fully prevented PF-429242/chloroquine-induced cytotoxicity and plasma membrane rupture. Furthermore, pH values of drug-containing media were not changed during preparation (Figure 3E) and drug treatment for 24 h (Figure 3F), indicating that intracellular, but not extracellular, pH fluctuations may have contributed to the pH-dependent cell death induced by combined PF-429242/chloroquine treatment. Therefore, the combination of PF-429242 and chloroquine triggered pH-dependent cell death.

PF-429242/chloroquine-induced pH-dependent cell death is distinct from JTC801-induced alkaliptosis

Intracellular alkalinization by activation of IKKβ/NF-κB and the subsequent reduction of CA9 expression is the major route for JTC801-induced alkaliptosis 25. Because IKKβ inhibitors failed to rescue PF-429242/chloroquine-induced cell death, we further measured intracellular pH levels using a pHrodo™ Green AM Intracellular pH Indicator. Surprisingly, PF-429242 alone or in combination with chloroquine induced mild intracellular acidification, although it did not reach statistical significance, except for the PF-429242/chloroquine combination sample in HepG2 cells (Figure 4A). This suggests that intracellular alkalization was not implicated in PF-429242/chloroquine-induced pH-dependent cell death. We also explored whether the NF-κB/CA9 signaling pathway was altered by the combined treatment with PF-429242 and chloroquine. As shown in Figure 4B, PF-429242/chloroquine combined treatment did not activate NF-κB, as indicated by the levels of p65 phosphorylation and IκB expression. In addition, endogenous CA9 expression was very low in PLC5 and HepG2 cells, and PF-429242/chloroquine combined treatment slightly induced CA9 expression (Figure 4B). CA9 is known to be transcriptionally induced by hypoxia-inducible factor (HIF)-1 during hypoxia 26. As a positive control, chemical hypoxia by cobalt chloride (CoCl_2_) dramatically induced CA9 expression (Figure 4B). These results implied that PLC5 and HepG2 cells might not rely on endogenous CA9 for survival and thus were less susceptible to alkaliptosis. Indeed, JTC801-induced cytotoxicity was not reversed by SC-514 in PLC5 or HepG2 cells (Figure 4C). To further clarify CA9's role, PLC5 and HepG2 cells were transfected with a CA9-encoding plasmid. CA9 overexpression slightly attenuated PF-429242-induced autophagy (Figure 4D), but did not alter PF-429242/chloroquine-induced cytotoxicity (Figure 4E).

Recently, JTC801 was found to directly bind and stabilize ATPase H^+^ transporting V0 subunit D1 (ATP6V0D1), a V-ATPase member. ATP6V0D1 then interacts and inhibits signal transducer and activator of transcription 3 (STAT3), leading to lysosomal pH dysfunction and alkaliptosis 27. To investigate the role of ATP6V0D1/STAT3 in PF-4294242/chloroquine-induced cell death, their protein levels were determined. As shown in Figure 5A, PF-429242 and chloroquine alone and/or additively inhibited STAT3 and induced ATP6V0D1 expression. However, ATP6V0D1-knockdown did not rescue, but enhanced, PF-429242- and PF-429242/chloroquine-induced cell death (Figure 5B, C). Furthermore, STAT3 overexpression had no effect on cell death induced by PF-429242 and chloroquine, alone or in combination (Figure 5D, E). Therefore, PF-429242/chloroquine combined treatment induced a form of pH-dependent cell death distinct from JTC801-induced alkaliptosis.

Site-1 protease inhibition does not synergize with chloroquine to kill HCC cells

Because PF-429242 is a site-1 protease inhibitor, we further investigated whether site-1 protease inhibition by PF-429242 was involved in PF-429242/chloroquine-induced cell death. MBTPS1 (which encodes a site-1 protease) and MBTPS2 (which encodes a site-2 protease and was used for comparison) messenger (m)RNA expressions in PLC5 and HepG2 cells were knocked down by siRNAs (Figure 6A). However, MBTPS1- and MBTPS2-knockdown did not synergize with chloroquine to kill HCC cells or cause plasma membrane rupture (Figure 6B, C). Site-1 and site-2 proteases cleave the pro-forms of SREBP1/2. Mature SREBP1/2 enter into nuclei and regulate genes related to sterol biosynthesis 28. PLC5 and HepG2 cells were transfected with plasmids encoding SREBP1a/c and SREBP2 (Figure 6D). Consistently, SREBP1/2 overexpression did not rescue PF-429242/chloroquine-induced cytotoxicity (Figure 6E). Therefore, site-1 protease inhibition is not involved in PF-429242/chloroquine-induced cell death.

Discussion

Results of this study unveil a novel approach for treating HCC by demonstrating the synergistic anticancer effects of PF-429242 and chloroquine. The combination of these drugs led to significant cytotoxicity against HCC cell lines, a phenomenon not observed with other therapeutic drugs such as doxorubicin, sorafenib, or regorafenib. This indicates a unique mechanism of action for the PF-429242/chloroquine combination, differentiating it from standard treatments. Importantly, the synergy was confirmed using another lysosomotropic agent, bafilomycin A1, further supporting the potential of this drug combination to induce potent anticancer effects by targeting lysosomal functions.

Chloroquine is widely used as an autophagy inhibitor and is often employed in clinical settings to counteract drug resistance resulting from autophagy induction in cancer cells 29. In our study, we specifically examined whether autophagy played a role in the observed cell death mechanism. Our findings indicate that the synergistic cytotoxicity of PF-429242 and chloroquine was not dependent on inhibition of autophagy. Instead, the combined treatment induced a distinct pH-dependent cell death mechanism. This suggests that our study effectively bypassed the conventional autophagy-mediated drug resistance pathway and instead leveraged a novel, alternative cell death mechanism that could be exploited for therapeutic benefit.

Our findings suggest that the synergistic effect of PF-429242 and chloroquine was not mediated through apoptosis or autophagy. This divergence from traditional cell death pathways prompted an investigation into alternative mechanisms. The observed pH-dependent cell death mechanism was notably distinct from known forms like alkaliptosis, as it did not involve intracellular alkalinization or the IKKβ/NF-κB/CA9 or ATP6V0D1/STAT3 pathways. Additionally, the role of site-1 protease inhibition by PF-429242 was ruled out, as neither knockdown of site-1 protease nor overexpression of SREBP1/2 affected the cytotoxicity induced by the drug combination. Further investigations are needed to elucidate the exact molecular mechanisms.

Lysosomal function is regulated by its internal pH, which is maintained in the range of pH 4.5-5 by V-ATPases and counter-ion transporters. V-ATPases are ATP-driven, pH-regulated proton pumps that acidify the lysosome. Due to the electrogenic nature of V-ATPases, parallel ion conductance is required for significant lysosomal acidification 30. An imbalanced lysosomal pH and lysosomal alkalization may inhibit lysosomal enzymes and impair the degradation capacity of lysosomes 31. Chloroquine and bafilomycin A1 can cause lysosomal alkalinization 32, which may contribute to their ability to synergize with PF-429242. Supporting this, ATP6V0D1-knockdown also enhanced the anticancer activity of PF-429242.

The discovery of a novel pH-dependent cell death mechanism presents significant implications for HCC treatment. Given the challenges associated with chemoresistance in HCC, targeting alternative forms of cell death represents a promising therapeutic strategy. Our findings suggest that lysosomal alkalinization, modulated by chloroquine, may play a key role in sensitizing HCC cells to PF-429242. This could open up avenues for repurposing chloroquine as a chemosensitizing agent in combination therapies for HCC. Additionally, pH modulation strategies could be further explored in conjunction with other drugs that disrupt lysosomal homeostasis.

While our study focused on HCC cells, we acknowledge that a comparison with normal hepatocytes would strengthen the precision and translational relevance of our findings. Future studies should incorporate non-cancerous hepatocyte models to assess whether the pH-dependent cell death mechanism is selectively cytotoxic to cancer cells. This could provide insights into the therapeutic window of PF-429242 and chloroquine combination treatment.

In conclusion, our study reveals a novel pH-dependent cell death mechanism induced by the combination of PF-429242 and chloroquine in HCC cells. This synergistic anticancer effect is distinct from traditional apoptosis, autophagy, and other known forms of cell death such as alkaliptosis. The findings suggest potential therapeutic implications for this drug combination, although further research is necessary to elucidate the precise molecular mechanisms involved. The unique pathway identified in this study offers a promising direction for developing targeted treatments for HCC.