Mitochondria-Targeting Biquaternary Ammonium Compounds: Pancreatic Anticancer Activity and Synergistic Interaction with Metformin

Abstract

Challenges in pancreatic cancer treatment primarily arise from chemotherapy resistance, cancer cell metastasis, and frequent late-stage diagnoses. These issues significantly compromise the effectiveness of standard treatments and highlight the urgent need for targeted approaches. In this context, we explored the anticancer potential of bis-quaternary ammonium-based compounds (BQACs), which remains largely uncharted. This study examines the structure–activity relationship of amphiphilic bicationic compounds as anticancer agents, focusing on their selectivity against pancreatic cancer cells. Our analysis revealed a potent antiproliferative effect associated with mitochondrial accumulation and subsequent mitochondrial membrane depolarization. Furthermore, combination therapies involving BQACs and chemotherapeutic drugs were explored to enhance treatment efficacy. Consequently, we propose a novel combination of BQACs with metformin, resulting in enhanced cellular uptake of the latter. The synergistic effect of the combination enables a significantly lower effective dose of metformin when used alongside BQACs to achieve therapeutic outcomes.

Introduction

Pancreatic cancer is one of the most challenging forms of cancer, characterized by its late diagnosis, aggressive progression, and resistance to current treatment options. Despite representing only 3% of all cancers, it is the fourth leading cause of cancer-related mortality, with a five-year survival rate of just 12%. , This low prognosis stems from several factors. First, pancreatic cancer is frequently detected at advanced stages due to its clinically silent nature and lack of early detection methods. Nearly half of patients present with metastatic cancer at the time of diagnosis, excluding them from potentially curative surgical resection. Second, the tumor’s dense stromal structure and poor vascularization hinder drug delivery, reducing the effectiveness of conventional chemotherapy. Moreover, the rapid development of drug resistance further limits the efficacy of standard treatments, underscoring the need for innovative therapeutic approaches.

Targeting mitochondrial function represents a promising strategy in pancreatic cancer therapy due to the organelle’s critical role in tumor cell metabolism and survival. It was reported that pancreatic cancer cells are especially dependent on mitochondrial oxidative phosphorylation (OXPHOS) for energy production, and that mitochondrial metabolism represents a key metabolic vulnerability. − Disrupting OXPHOS can impair energy production, reduce cancer cell viability, and sensitize tumors to treatment. Furthermore, mitochondria play a central role in regulating apoptosis through the release of cytochrome c, which activates caspase-dependent cell death pathways. By inducing mitochondrial dysfunction, therapeutic agents can trigger intrinsic apoptosis via mitophagy, thereby enhancing cancer cell death. Exploiting mitochondrial vulnerabilities offers a strategy to overcome the inherent resistance of pancreatic tumors to conventional therapies.

Given the multifaceted resistance mechanisms of pancreatic cancer, combination therapy has emerged as an attractive approach to improve treatment efficacy. Combination strategies enable dose optimization, potentially reducing side effects while maintaining therapeutic efficacy. Additionally, combining drugs with complementary mechanisms of action can enhance cytotoxicity, reduce the likelihood of resistance, and target multiple cancer cell populations simultaneously. For instance, combining a conventional anticancer drug such as doxorubicin, which primarily targets nuclear DNA, with a mitochondria-targeting agent offers to address different aspects of cellular metabolism and survival pathways, potentially leading to improved therapeutic outcomes. Another compelling strategy is the combination of drugs that share the same target to exhibit synergistic effects. For example, pairing metformin, a metabolic modulator that disrupts mitochondrial function, with another mitochondria-targeting agent. Metformin has attracted significant attention for its ability to inhibit mitochondrial complex I, reduce OXPHOS, and induce metabolic stress in cancer cells. However, clinical trials involving metformin as an anticancer drug have encountered challenges, primarily due to its low bioavailability.

In this study, we started by exploring the pancreatic anticancer activity of new cationic amphiphilic drugs (CADs), known to exert their effects through membrane disruption, lysosomal accumulation, and mitochondrial targeting. Their cationic nature allows them to bind negatively charged cancer cell membranes, while their hydrophobic regions insert into lipid bilayers, leading to membrane permeabilization. These types of drugs are also known to effectively target mitochondria, which have higher negative membrane charges. Mitochondrial uptake causes oxidative stress and cytochrome c release, inducing apoptosis at lower doses and necrosis at higher doses, making them valuable candidates for cancer therapies. By interfering with lipid metabolism and bypassing drug resistance mechanisms, these compounds can enhance chemotherapy efficacy and offer promising strategies for cancer treatment. Herein, we introduce innovative biquaternary ammonium compounds (BQACs) as a strategy to overcome pancreatic cancer cells resistance. We investigated the structure–activity relationship of benzimidazolium and quinolinium derivatives. Benzimidazolium and quinolinium are valuable structural motifs for cationic drugs due to their unique structural and electronic properties. These advantages stem from their cationic nature, aromatic systems, stability and tunable functional groups, enabling precise mitochondrial targeting and enhanced membrane interaction. The benzimidazolium motif facilitates π-π stacking and hydrogen bonding with mitochondrial targets, enhancing mitochondrial accumulation and membrane disruption, minimizing off-target effects thus improving therapeutic precision. , Quinolinium analogues, characterized by their delocalized positive charge, exhibit excellent membrane permeability and exploit the mitochondrial membrane potential (ΔΨm ≈ −180 mV) to drive selective mitochondrial accumulation. This property is particularly advantageous in cancer cells, which often display higher ΔΨm than normal cells, allowing for enhanced selective targeting. Some quinolinium derivatives are known to bind mitochondrial RNA, inducing dysfunction through apoptosis and OXPHOS inhibition. Both motifs share a multifunctional scaffold with modifiable alkyl groups, and a planar aromatic system which facilitates insertion into lipid bilayers, destabilizing membranes and promoting permeabilization. Hence their cationic amphiphilicity, and structural adaptability, allows for drug development and optimization. To elucidate the underlying anticancer mechanisms of benzimidazolium and quinolinium based BQACs, we conducted a thorough investigation into their membrane-permeabilizing activity and mitochondrial-targeting effects. Additionally, we explored the combination of BQACs with two different chemotherapeutic drugs – doxorubicin and metforminto propose novel combination therapies approach.

Results and Discussion

Design and Synthesis

Building on our recent success designing potent, broad-spectrum antimicrobial agents, we turned our attention to the development of novel anticancer therapeutics. Our prior work involved the development of a series of bis-benzimidazolium salts, characterized by their tunable degree of hydrophobicity and rigidity. These salts exhibited potent activity against multidrug-resistant bacteria by disrupting bacterial cell membranes and showed remarkable resilience to resistance development. Given the success of bis-benzimidazolium salts as antimicrobial agents, primarily due to their cationic amphiphilic nature facilitating selective interaction and disruption of negatively charged membranes, we hypothesized that these properties might be advantageous to target negatively charged components of cancer cell membranes and mitochondria. This approach aims to elicit an anticancer response in pancreatic cancer cells, notorious for their resistant nature and challenging treatment.

To explore this hypothesis, we repurposed our first-generation bis-ammonium compounds (Figure ). Our primary objective was to establish a ground understanding of the structure–activity relationship (SAR) that governs the anticancer activity of these bis-cationic-amphiphilic small molecules against pancreatic cancer cells (Table ). Preliminary results revealed that the most hydrophilic analoguescharacterized by a short hexyl side chain and hydrophilic ether, diamide or short alkyl linkersexhibited the highest selectivity. These compounds demonstrated significant antiproliferative effects against pancreatic cancer cells while maintaining minimal cytotoxicity toward normal cells (Table ).

1.

Structure of the rationally designed bis-ammonium compounds.

1. Cytotoxic Activity of BQACs (Series 1–4) against Pancreatic Cancer and Normal Cell Lines.

		IC50 (μM)
Series	Compound	KP4	PANC1	IMR90	hTerT-HPNE	S.I. IMR90/KP4	S.I. HPNE/KP4	S.I. IMR90/PANC1	S.I. HPNE/PANC1	C log P
Series 1	4a	15 ± 1	99 ± 9	278 ± 19	>285	19	> 19	2.8	> 2.9	1.4
	6a	6.8 ± 0.7	42 ± 5	105 ± 12	84 ± 11	15	12	2.5	2.0	2.1
	7a	3.6 ± 0.4	10.6 ± 0.4	70 ± 11	39 ± 8	19	11	6.6	3.7	2.0
	8a	2.2 ± 0.3	5 ± 1	48 ± 9	38 ± 1	22	17	9.6	7.6	2.6
	8b	0.7 ± 0.1	ND	2.4 ± 0.4	ND	3.4	ND	ND	ND	3.9
	8d	1.7 ± 0.1	ND	6 ± 1	ND	3.5	ND	ND	ND	4.6
Series 2	9a	5.0 ± 0.7	>285	443 ± 18	>285	89	57	ND	ND	2.4
	9b	1.5 ± 0.2	ND	5 ± 2	ND	3.3	ND	ND	ND	3.8
	9d	1.0 ± 0.2	4 ± 1	13 ± 2	>285	13	>285	3.3	>71	4.3
Series 3	10a	2.11 ± 0.04	53 ± 8	215 ± 12	>285	102	>135	4.1	>5.4	3.2
	10d	1.9 ± 0.5	9.4 ± 0.8	118 ± 5	70 ± 6	62	37	13	7.4	4.8
	11a	2.0 ± 0.3	8.3 ± 0.5	55 ± 3	110 ± 15	28	55	6.6	13	3.2
	11d	0.87 ± 0.04	ND	6 ± 2	ND	6.9	ND	ND	ND	5.5
	12a	2.0 ± 0.1	8 ± 2	28 ± 3	52 ± 8	14	26	3.5	6.5	3.5
	12d	0.54 ± 0.01	ND	3 ± 1	ND	5.6	ND	ND	ND	5.5
	14a	0.28 ± 0.05	ND	0.75 ± 0.05	ND	2.7	ND	ND	ND	5.8
	14b	0.30 ± 0.01	ND	1.2 ± 0.5	ND	4.0	ND	ND	ND	7.1
	14d	0.6 ± 0.2	ND	1.3 ± 0.4	ND	2.2	ND	ND	ND	7.7
Series 4	15a	0.41 ± 0.05	ND	3 ± 1	ND	7.3	ND	ND	ND	5.0
	15b	0.6 ± 0.2	ND	2.5 ± 0.9	ND	4.2	ND	ND	ND	6.1

^aIC₅₀ values are the mean of three independent triplicates.

^bS.I. Selectivity index = IC₅₀ value of normal cell line over IC₅₀ value of cancer cell line.

^cConsensus on the calculated log P from Swiss ADME.

^dND: not determined.

Building on these insights, new series of compounds were designed based on the essential structural elements that previously demonstrated the highest selectivity, including the critical linkers and hexyl side chain. Hence, to explore new compounds, the benzimidazolium building block was modified, and three additional series were developed (Figure and Scheme ). These new series focus on adding hydrophilicity to the benzimidazolium unit and explore alternative sources of cationic charge. These modifications are expected to enhance membrane permeability, thereby improving the therapeutic efficacy of BQACs. Series 5 featured guanidinium groups incorporated into 2-aminobenzimidazolium moieties (Scheme A). Series 6 included dimethoxybenzimidazolium groups, selected for their potential into enhancing both solubility and biological interactions (Scheme B). Series 7 incorporated a quinolinium moiety, characterized by a delocalized positive charge, that promotes membrane permeability and is known to exhibit diverse biological effects, especially targeted toward mitochondria (Scheme C). ,

1. Synthetic Procedures of the Second Generation BQAC Analogues.

Additionally, we synthesized a series of compounds featuring a photoactive diamide-stilbene linker (Series 8), to enable fluorescence imaging, which can provide valuable insights into the localization and dynamics of these compounds within cellular environments (Scheme D).

In Vitro Viability of Pancreatic Cancer and Normal Cells

In vitro viability assays were conducted to evaluate the selectivity of series 1 to 4 toward pancreatic cancer cells and normal cells. The compounds were tested on two pancreatic cancer cell lines, KP4 and PANC1, which exhibit notable resistance-related characteristics, including independence from the KRAS signaling pathway and high expression of drug transporters such as ABCB1, ABCG2, and ABCC1. , To evaluate the selectivity of the compounds, we conducted a comparative analysis of their anticancer efficacy in relation to their cytotoxic effects on two prominent models of normal cell lines. The first model utilized was IMR90, which consists of human lung fibroblasts, providing insights into the compounds’ effects on mesenchymal cells. The second model employed was hTERT-HPNE, representing immortalized human pancreatic epithelial cells, which serves as a benchmark for assessing the impact on epithelial tissues. This dual approach allows for a comprehensive understanding of the therapeutic potential of the compounds while simultaneously gauging their safety profile in normal cellular contexts. The selectivity index was calculated for each compound, providing insights into the compound’s therapeutic potential and biocompatibility. Several candidates exhibited high selectivity indices, indicating their ability to target diseased cells while minimizing toxicity to normal cells.

Overall, the results suggest that the designed BQACs exhibited stronger activity against the KP4 cancer cell line compared to the PANC1 cell line, which may be attributed to the higher expression of resistance markers in PANC1 or the tendency of PANC1 cells to form large aggregates in in vitro culture. The most selective compounds were generally the more hydrophilic BQACs within each series. Specifically, compounds 4a, 6a, 7a, and 8a from series 1, 9a and 9d from series 2, and 10a, 10d, 11a, and 12a from series 3 demonstrated excellent selectivity for KP4 cells and good affinity for PANC1 cells compared to IMR90 and hTERT-HPNE cells. Although these hydrophilic compounds required higher concentrations to achieve 50% inhibition of cancer cell growth, they showed exponentially reduced cytotoxicity toward normal cells (Table ).

These findings motivated the design of the second generation of series with enhanced hydrophilic properties, incorporating guanidinium, dimethoxybenzimidazolium, and quinolinium groups to improve membrane permeability and interactions with biological systems. The 2-aminobenzimidazolium derivatives (series 5) showed notable cytotoxicity, but exhibited very low selectivity (Table ). In contrast, the dimethoxybenzimidazolium-based compounds (series 6) and benzimidazolium-dimethoxyisoquinolinium-based BQACs (series 7) demonstrated a more favorable profile. Although these compounds showed reduced overall activity, they exhibited significantly improved selectivity. Notably, compounds 23, 26, and 31 achieved selectivity indices exceeding 10 for the KP4 cell line (Table ). Compound 31 was particularly noteworthy, as it demonstrated targeted activity against both KP4 and PANC1 cells, with high selectivity indices for both cancerous lines compared to both normal cell lines.

2. Cytotoxic Activity of BQACs (Series 5–7) against Pancreatic Cancer and Normal Cell Lines.

		IC50 (μM)
Series	Compound	KP4	PANC1	IMR90	hTerT-HPNE	S.I. IMR90/KP4	S.I. HPNE/KP4	S.I. IMR90/PANC1	S.I. HPNE/PANC1	Clog P
Series 5	17	9 ± 1	ND	26 ± 7	ND	2.9	ND	ND	ND	1.0
	18	4.5 ± 0.4	ND	16 ± 5	ND	3.6	ND	ND	ND	1.3
	19	1.16 ± 0.07	ND	7 ± 1	ND	6.0	ND	ND	ND	1.7
	20	2.3 ± 0.2	ND	6 ± 2	ND	2.6	ND	ND	ND	1.5
	21	1.1 ± 0.3	ND	2.90 ± 0.01	ND	2.6	ND	ND	ND	2.7
Series 6	22	>285	ND	>285	ND	ND	ND	ND	ND	0.99
	23	24 ± 4	145 ± 10	>285	>285	>12	>12	>2.0	>2.0	1.6
	24	11 ± 1	ND	99 ± 19	ND	9.0	ND	ND	ND	2.4
	25	202 ± 54	ND	>285	ND	1.4	ND	ND	ND	2.0
	26	6.6 ± 0.9	121 ± 13	>285	>285	>43	>43	>2.4	>2.4	3.2
Series 7	27	78 ± 5	ND	244 ± 26	ND	3.1	ND	ND	ND	–0.03
	28	56 ± 9	ND	276 ± 52	ND	4.9	ND	ND	ND	0.79
	29	19 ± 3	ND	156 ± 49	ND	8.2	ND	ND	ND	1.3
	30	100 ± 9	ND	829 ± 190	ND	8.3	ND	ND	ND	1.2
	31	1.8 ± 0.2	17 ± 2	172 ± 11	>285	96	>158	10	>17	2.3

^aIC₅₀ values are the mean of three independent triplicates.

^bS.I. Selectivity index = IC₅₀ value of normal cell line over IC₅₀ value of cancer cell line.

^cConsensus on the calculated log P from Swiss ADME.

^dND: not determined.

Selectivity

A large therapeutic window is a significant advantage in pharmacology, as it allows for dosage adjustments to optimize therapeutic effects while minimizing potential toxicity. To validate the selective antiproliferative activity of the promising anticancer compounds, we conducted a comparative cell viability analysis between the KP4 cancer cell line and the IMR90 normal fibroblast line. Both cell lines were treated with concentrations exceeding the established anticancer IC₅₀ values for each compound. Our findings revealed that, at a dosage set at 10-fold the respective IC₅₀ for each BQAC against KP4 cells, these compounds successfully inhibited cancer cells proliferation by over 80%, with little to no effect on the growth of IMR90 cells (Figure ). This result highlights the selective nature of these compounds, as reflected by a selectivity index (S.I.) greater than 10 in vitro (Tables and ).

2.

Relative viability levels of KP4 (red) and IMR90 (green) cells treated with BQAC derivatives at ten times their respective IC₅₀ value against KP4 cells.

Cell Membrane Permeabilization

To investigate the anticancer mechanisms of our most selective compounds, we focused on the cell membrane as a potential target. To assess membrane integrity, we used the membrane-impermeable fluorescent dye propidium iodide (PI), which only binds to nucleic acids when the cell membrane is significantly compromised. KP4 cells were treated with each compound at their respective IC₅₀ concentrations and cultured for 48 h before adding PI. Control experiments included KP4 cells treated with DMSO and Triton X-100. Observations made using confocal microscopy revealed that KP4 cells treated with compounds 8a, 9a, 11a, 26, and 31 exhibited minimal intracellular PI fluorescence compared to the Triton positive control (Figure A). This result suggests that these compounds induce only slight permeabilization of the cell membrane to PI. Therefore, our findings indicate that the anticancer mechanisms of these specific compounds likely do not primarily involve the permeabilization of the cell membrane.

3.

(A) Cellular uptake of propidium iodide (PI) in KP4 cells treated with compounds 8a, 9a, 11a, 26, and 31 at their respective IC₅₀ value for 48 h prior to PI addition (red). DMSO-treated cells served as negative control, while Triton X-100-treated cells served as positive control. (B) Membrane potential analysis of KP4 cells treated with 8a, 9a, 11a, 26, and 31 at half, once and twice their respective IC₅₀ value, following the fluorescence intensity of DiSC₂(5). DMSO-treated cells served as negative control, while Triton X-100-treated cells served as positive control. Measurements were performed in triplicate.

Cell Membrane Depolarization

Given the compounds showed minimal anticancer effects through cell membrane lysis, we shifted our focus to their impact on cell membrane potential. KP4 cells were incubated with the cationic lipophilic dye DiSC₂(5) and then treated with BQACs 8a, 9a, 11a, 26, and 31 at concentrations corresponding to half, once and twice their respective IC₅₀ values. We monitored the fluorescence intensity of DiSC₂(5) over a three-hour period following treatment. DiSC₂(5) accumulates in polarized membranes, leading to self-quenching of the dye’s fluorescence. When membrane depolarization occurs, the dye is released, resulting in an increase in fluorescence intensity.

Compounds 8a, 9a, 11a, 26, and 31 exhibited a transient effect on membrane potential, inducing a brief depolarization followed by a plateau (Figure B). This suggests a temporary interaction with the cell membrane, likely due to their low affinity for the lipid bilayer. We hypothesize that these compounds briefly permeabilize the membrane, allowing ion flux, before diffusing into either the extracellular environment or the cytoplasm. Importantly, the observed membrane depolarization is not sustained, indicating that the anticancer activity of these compounds is unlikely due to irreversible membrane disruption like permanent depolarization or lysis. While these compounds do demonstrate transient membrane permeabilization, their primary mechanism of action likely involves other cellular targets. This behavior is consistent with cationic amphiphilic drugs (e.g., ACPs, TPP-functionalized molecules), which often traverse the lipid bilayer to engage with intracellular components and exert their anticancer effects.

Subcellular Localization

To determine the subcellular localization and potential targets of the synthesized BQACs, we designed fluorescent probes, compounds 16a and 16b (Figure A). These probes exhibit excitation and emission spectra within the short-nanometer range, comparable to the DNA-binding dye DAPI (Figure B, full spectrum in Figure S2). We utilized confocal microscopy with the DAPI filter to visualize the intracellular distribution of 16a and 16b in KP4 cells (Figure C). Given the cationic amphiphilic nature of these compounds and the known tendency of such molecules to localize in mitochondria, we investigated their colocalization with the commercially available mitochondrial probe, MitoTracker Deep Red (MTDR). Confocal microscopy revealed a high degree of colocalization between both 16a and 16b with MTDR (Figure C). Quantitative analysis confirmed this observation, yielding Pearson correlation coefficients of 0.87 and 0.93 for 16a and 16b, respectively (Figure D). These findings suggest that the bis-cationic amphiphilic structure of these compounds allows them to readily cross the cell membrane and specifically target mitochondria.

4.

Subcellular localization study. (A) Fluorescent bis-benzimidazolium analogues 16a and 16b and their respective IC₅₀ against KP4 cells. (B) Comparison of the fluorescent properties of 16a and 16b with those of DAPI. (C) Colocalization of 16a and 16b (blue, revealed with the DAPI filter) with the mitochondria-targeting fluorescent dye MTDR (red) in KP4 cells. (D) Corresponding Pearson’s correlation coefficients indicating the degree of colocalization between 16a and 16b with MTDR, demonstrating their accumulation in mitochondrial regions.

Mitochondrial Membrane Depolarization

Mitochondrial membrane disruption, driven by the interaction of cationic moieties with anionic phospholipids like phosphatidylglycerol and cardiolipin, is a well-established mechanism of action for cationic amphiphilic compounds. To investigate whether our BQACs function similarly, we assessed their impact on mitochondrial membrane potential using the fluorescent probe JC-1. JC-1 emits red-orange fluorescence when aggregated in healthy mitochondria with high membrane potential, but shifts to green fluorescence upon depolarization-induced monomerization. KP4 cells were treated with compounds 8a, 9a, 11a, 26, and 31 at their respective IC₅₀ value for 48 h. Controls included DMSO and FCCP, a known mitochondrial uncoupler. Microscopic analysis revealed significant green fluorescence in BQAC-treated cells, indicating mitochondrial membrane depolarization (Figure ). In contrast, DMSO-treated cells exhibited predominantly red-orange fluorescence, signifying intact mitochondrial membrane potential (Figure ).

5.

Assessment of mitochondrial membrane depolarization in KP4 cells treated with BQACs using the JC-1 probe. Orange fluorescence indicates healthy mitochondria, while green fluorescence signifies damaged mitochondria.

These findings strongly suggest that the anticancer activity of these BQACs involves disruption of mitochondrial membrane integrity, leading to depolarization. Such mitochondrial permeabilization is a hallmark of early apoptosis, triggering cytochrome c release and subsequent activation of the apoptotic cascade. Therefore, these results highlight the potential of BQACs to selectively induce programmed cell death in cancer cells.

In Vitro Synergistic Activity of BQAC Derivatives with Chemotherapeutics

The complex and divers resistance mechanisms of pancreatic cancer have necessitated the exploration of innovative treatment strategies. The implementation of combination therapy has emerged as an attractive approach to enable dose optimization, potentially reducing side effects while maintaining therapeutic efficacy, and overcoming drug resistance by targeting multiple metabolic pathways.

To assess the potential benefit of combination therapy in the context of pancreatic cancer, we conducted in vitro assays to evaluate the interactions between BQACs and two different chemotherapeutic agents: doxorubicin (DOX) and metformin (MET) (Figure ). Doxorubicin is a well-established conventional antineoplastic drug that exerts its effects primarily by targeting nuclear DNA. Despite its widespread use in cancer treatment, doxorubicin is notorious for its severe side effects, which can significantly impact a patient’s quality of life. On the other hand, metformin, a drug traditionally used for type 2 diabetes, has been repurposed in oncology due to its ability to target complex I of the mitochondrial respiratory chain. This mechanism induces metabolic stress in cancer cells, making them more susceptible to treatment. However, metformin’s high hydrophilicity poses challenges for its clinical application as an anticancer agent, primarily due to its limited membrane permeability and consequently low bioavailability.

6.

Structure of metformin hydrochloride.

The rationale behind combining doxorubicin with BQACs lies in their ability to target different cellular components, thereby creating a complementary therapeutic effect. The combination of doxorubicin and BQACs was tested at a molar ratio of 1:40, with doxorubicin concentrations ranging from 0.00078 to 0.2 μM and BQACs from 0.031 to 8.0 μM, based on their respective IC₅₀ values. The results indicated that there was no synergistic effect between doxorubicin and the tested BQACs at this ratio against KP4 pancreatic cancer cells (Figure ). The combination indexes close to 1 suggest an additive interaction and the possibility to use these drugs in combination to target multiple cellular pathways and overcome resistance mechanism, with no antagonist effect (Table ).

7.

Activity of doxorubicin against KP4 cells in combination with BQACs at a molar ratio of 1:40 DOX/BQAC. *p ≤ 0.05 (ANOVA).

3. Properties of Doxorubicin Combined with BQACs.

Compound	IC50 (μM)	IC50 of doxorubicin in combination (DOX/BQAC 1:40) (μM)	Combination index
Doxorubicin	0.066 ± 0.006
8a	2.2	0.035 ± 0.004	1.15
9a	5.0	0.042 ± 0.007	0.97
11a	2.0	0.027 ± 0.006	0.95
26	6.6	0.0630 ± 0.0008	1.33
31	1.8	0.030 ± 0.005	1.13

On the other hand, combining two mitochondria-targeting agents, such as metformin and BQACs, aims at improving therapeutic efficacy at lower concentrations. The combination of metformin and BQACs was examined at a molar ratio of 50:1, with metformin concentrations spanning from 1.56 to 400 μM and BQACs from 0.031 to 8.0 μM. The results demonstrated a significant reduction in the IC₅₀ of metformin against KP4 pancreatic cancer cells when used in conjunction with BQACs. The combination indexes for this pairing fell significantly below 1, indicating a synergistic effect, which suggests that this strategic combination may increase metformin uptake and efficacy at lower concentrations (Table and Figure ).

4. Synergistic Activity of Metformin Combined with BQACs.

Compound	IC50 (μM)	IC50 of metformin in combination (MET/BQAC 50:1) (μM)	Combination index
Metformin	987± 10
8a	2.2	59 ± 9	0.59
9a	5.0	106 ± 5	0.53
11a	2.0	40 ± 5	0.44
26	6.6	105 ± 10	0.43
31	1.8	32 ± 5	0.39

8.

Activity of metformin against KP4 cells in combination with BQACs at a molar ratio of 50:1 MET/BQAC. ****p ≤ 0.0001 (ANOVA).

Mechanism of Action of Metformin in Synergism

The consequences of metformin’s mechanism of action, particularly its interaction with complex I of the mitochondrial respiratory chain, lead to the activation of AMP-activated protein kinase (AMPK) through phosphorylation. To assess the synergistic effects of the metformin/BQAC combination, we quantified the levels of phosphorylated AMPK (P-AMPK) relative to the total level of AMPK in the KP4 cell line (Figure ).

9.

Effect of metformin and BQACs alone and in combination on P-AMPK protein level. Immunoblot and quantification of P-AMPK levels relative to AMPK levels. *p ≤ 0.05 (ANOVA).

In this study, KP4 cells were treated with metformin at a concentration of 100 μM and BQACs at 2 μM for 48 h. This was compared to KP4 cells treated with a 50:1 ratio of metformin/BQAC (100 μM and 2 μM, respectively). The observed increase in phosphorylated AMPK levels following treatment with the various combinations, compared to KP4 cells treated with metformin alone, indicates improved uptake of metformin and effective action at sub-IC₅₀ concentrations when administered in conjunction with BQACs. Specifically, the P-AMPK/AMPK ratio was significantly enhanced in KP4 cells treated with compounds 9a and 11a, whereas treatment with 100 μM metformin and BQACs separately showed no significant increase in P-AMPK levels.

Conclusion

In summary, we designed and synthesized a series of BQAC based on benzimidazolium and quinolinium, followed by a comprehensive pancreatic anticancer activity study. Our results revealed that these bicationic compounds exhibited significant cytotoxicity against KP4 and PANC1 pancreatic cancer cells. Analysis of the IC₅₀ values for the initial series indicated that compounds with lower hydrophobicity displayed minimal toxicity toward normal pancreatic and lung fibroblast cell lines. This observation led us to investigate the design of more hydrophilic biquaternary ammonium compounds by incorporating functional groups such as 2-aminobenzimidazolium, dimethoxybenzimidazolium, and benzimidazolium-dimethoxyisoquinolinium to enhance their therapeutic potential.

Among the compounds studied, 8a, 9a, 11a, 26, and 31 emerged as the most selective agents, demonstrating favorable selectivity for the KP4 and PANC1 cell lines with minimal antiproliferative effects on IMR90 lung fibroblasts, even at concentrations up to ten times their IC₅₀ values against KP4 cells. Their selectivity was attributed to interactions between the cationic moieties of these compounds and the anionic residues on cancer cell membranes; however, their mechanisms of action did not involve significant disruption of the cell membrane.

To investigate the intracellular dynamics of these BQAC derivatives, the photoactive compounds 16a and 16b were utilized, confirming their ability to enter the cytoplasm and specifically localize within mitochondria. Our findings indicate that the structural designs of compounds 8a, 9a, 11a, 26, and 31, which balance cationic moieties and amphiphilic surfaces, facilitate their selective targeting of cancer cell membranes, and transmembrane passage without causing irreversible membrane permeabilization. In the cytoplasm, these compounds are drawn to mitochondria due to electrostatic interactions, where their accumulation disrupts mitochondrial integrity.

Additionally, our exploration of the combined anticancer effects of these compounds with two chemotherapeutic drugs: doxorubicin and metformin, revealed an additive effect with doxorubicin and a synergistic interaction with metformin, which significantly enhanced metformin uptake and efficacy in mitochondria. This synergy is likely due to a BQAC-induced increased drug uptake, amplifying metformin’s efficacy. It is important to note that in vitro selectivity data have limitations, as they do not fully replicate the tumor microenvironment, metabolic processes, or resistance mechanisms found in vivo. Cell line models may not capture patient tumor heterogeneity, and drug concentrations tested in vitro may not reflect achievable systemic levels. To address these limitations, we plan to validate BQACs in pancreatic cancer xenograft and orthotopic mouse models, assessing their pharmacokinetics, toxicity, and interactions with the tumor microenvironment. Additionally, biomarker analysis and comparative studies with standard therapies will help determine their clinical potential and optimize combination strategies with metformin.

The strategy proposed here highlights the advantages of combination therapy in enhancing the intracellular uptake of poorly bioavailable anticancer drugs, thereby addressing the challenge of drug resistance in cancer treatment. Through these findings, we aspire to pave the way for more effective cancer therapies and optimize BQACs as therapeutic components.

Methods

Determination of the IC₅₀

Cancer cells (KP4, PANC1) were seeded in 96-well plates at a density of 1 × 10³ cells per well in regular DMEM. Normal cells (IMR90, hTERT-HPNE) were seeded in 96-well plates at a density of 4 × 10³ cells per well in DMEM supplemented with 10% Fetal Bovine Serum (FBS). After 24 h of incubation at 37 °C (5% CO₂ atm.), cells were treated with 100 μL of prepared aliquots of the test compounds diluted in DMEM. Controls included wells without cells (blank, only DMEM), and wells with cells treated with the maximum volume of DMSO used in the compound dilutions. Each condition was performed in triplicate wells. After 72 h, the culture medium was removed, and the cells were washed twice with PBS 1×. Then, 100 μL of 1% glutaraldehyde solution was added to each well for 10 min for fixation. The supernatant was removed, and cells were washed twice with water. Next, 200 μL of 2% crystal violet (CV) solution was added to each well for staining for 30 min under gentle agitation. After staining, the wells were washed multiple times with water. The plates were left to air-dry overnight, then the adherent CV was solubilized using 200 μL of 10% acetic acid solution. The absorbance was measured at 590 nm using a plate reader. Cell viability was calculated using the formula: Viability (%) = (AbsX – AbsBlank)/(AbsDMSO – AbsBlank) × 100.

Propidium Iodide Uptake Assay

KP4 cells were seeded at a density of 8 × 10⁴ cells per well in regular DMEM on top of coverslips placed in 6-well plates. After 24 h of incubation at 37 °C (5% CO₂ atm.), cells were treated with 100 μL of prepared aliquots of the test compounds diluted in DMEM, at a final concentration equal to the IC₅₀ of each compound against KP4 cells. Cells treated with DMSO served as negative control and Triton X-100 treated cells as positive control. After 48 h of treatment, the medium was removed, and cells were washed twice with PBS 1×. Then, 1 mL of PBS containing 1 μM of propidium iodide (PI) was added to the wells. Cells were stained for 30 min in the dark under gentle rocking. After staining, the medium was removed, and cells were washed twice with PBS 1×. Fixation was performed by adding 2 mL of 4% paraformaldehyde solution and incubating the cells at 4 °C for 10 min. Postfixation, cells were washed three times for 10 min with 0.1 M glycine solution in PBS 1× under gentle rocking. The coverslips were then mounted onto microscope slides using one drop of fluorescence mounting medium containing DAPI.

Membrane Depolarization Assay

KP4 cells were seeded in a 96-well black plate at a density of 1 × 10³ cells per well in regular DMEM. After 24 h of incubation at 37 °C (5% CO₂ atm.), the cells were washed once with DMEM. Then, 100 μL of DMEM containing 2 μM 3,3′-diethylthiadicarbocyanine iodide (DiSC₂(5)) was added to each well. The dye was allowed to be incorporated for 60 min at 37 °C. Following the incubation period, the cells were treated with 100 μL of prepared aliquots of the test compounds diluted in DMEM. Fluorescence intensity was measured using a microplate reader at Ex 622 nm/Em 670 nm. Controls included blank (untreated cells), cells treated with DMSO (negative control), and cells treated with Triton X-100 (positive control).

Colocalization Assay

KP4 cells were seeded at a density of 8 × 10⁴ cells per well in regular DMEM on top of coverslips in 6- well plates. After 24 h of incubation at 37 °C (5% CO₂ atm.), the cells were treated with 100 μL of prepared aliquots of the test compounds diluted in DMEM. The control included a coverslip treated with DMSO. After 48 h the medium was removed, and 1 mL of DMEM was added to each well with 10 μL of 100 μM MitoTracker Deep Red solution in PBS 1×. The dye was allowed to be incorporated for 30 min at 37 °C. The medium was removed, and the cells were washed twice with PBS. The cells were fixed with 2 mL of a 4% paraformaldehyde solution for 10 min at 4 °C. Postfixation, the cells were washed three times for 10 min each with 0.1 M glycine solution in PBS 1× under gentle rocking. The coverslips were mounted onto microscope slides using one drop of fluorescence mounting medium without DAPI.

Mitochondrial Membrane Potential Detection Assay

KP4 cells were seeded at a density of 8 × 10⁴ cells per well in regular DMEM on top of coverslips in 6-well plates. After 24 h of incubation at 37 °C (5% CO₂ atm.), the cells were treated with 100 μL of prepared aliquots of the test compounds diluted in DMEM. Controls included cells treated with DMSO (negative control) and cells treated with FCCP (positive control). After 48 h, the medium was removed, and 1 mL of DMEM containing 2 μM JC-1 was added to the wells. The dye was allowed to be incorporated for 60 min at 37 °C. The medium was then removed, and the cells were washed twice with PBS 1×. The cells were fixed with 2 mL of 4% paraformaldehyde solution for 10 min at 4 °C. Postfixation, the cells were washed three times for 10 min each with 0.1 M glycine solution in PBS 1× under gentle rocking. The coverslips were mounted onto microscope slides using one drop of fluorescence mounting medium without DAPI.

Combination Assays

KP4 cells were seeded in 96-well plates at a density of 1 × 10³ cells per well in regular DMEM. After 24 h of incubation at 37 °C (5% CO₂ atm.), the cells were treated with 50 μL of prepared aliquots of doxorubicin and metformin in DMEM, then 50 μL of prepared aliquots of the test compounds in DMEM. Controls included wells treated with gradients of doxorubicin, metformin and BQACs alone. Each condition was performed in triplicates.

Immunoblot Analysis

For the protein degradation assay, cells were seeded at a density of 1 × 10⁵ cells per well in 6-well plates overnight then treated with aliquots of metformin (100 μM) and BQACs (2 μM), alone and in combination, for 48 h. The culture medium was removed. Cells were washed twice with cold PBS and lysed with LAEMMLI 2× (4% SDS, 20% Glycerol, 0.125 M Tris-HCl (pH 6.8)). Protein concentrations were determined using a Nanodrop 2000c spectrophotometer. Equal amounts of protein were separated via sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to nitrocellulose membranes. Proteins were briefly revealed by Ponceau Red dye and membranes were washed with PBS-Tween 0.05% (PBS-T), before blocking with 5% milk proteins for 1 h. Membranes were washed again with PBS-T and incubated with the primary antibody overnight at 4 °C. After washing with PBS-T three times, the membranes were incubated with the secondary antibody for 1 h. After being washed with PBS-T three times, membranes were visualized by chemiluminescence using a Western Lightning Plus Chemiluminescence Reagent.

Glossary

Abbreviations

ACPs

Anticancer Peptides

Atm

Atmosphere

BQAC

Biquaternary Ammonium Compounds

CDC

Centers for Disease Control and Prevention

CI

Combination Index

Clog P

Calculated partition coefficient

DAPI

4′,6-Diamidino-2-phenylindole

DiSC₂(5)

3,3′-Diethylthiadicarbocyanine iodide

DOX

Doxorubicin

FCCP

Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone

HRMS

High-Resolution Mass Spectrometry

HPLC

High-Performance Liquid Chromatography

IC₅₀

Concentration required to inhibit 50% of cell growth

KRAS

Kirsten Rat Sarcoma Virus

MET

Metformin

MTDR

MitoTracker Deep Red

NMR

Nuclear Magnetic Resonance

PEB

4-(Phenylethynyl)­benzyl

PI

Propidium Iodide

S.I.

Selectivity Index

TPP

Triphenylphosphine

WHO

World Health Organization

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsbiomedchemau.4c00130.

• Materials and measurements; dose–response curves for the IC₅₀ determination (Figure S1); absorbance and emission spectrum of 16a and 16b (Figure S2); dose–response curves for metformin’s IC₅₀ determination in combination with BQAC derivatives (Figure S3 and S4); determination of the CI values (Tables S1 and S2; Figure S5); synthesis and characterization of the second-generation compounds by ¹H NMR, ¹³C NMR, HRMS exact mass, and HPLC purity (PDF)

A.R.S. and M.P. for conceptualization of the project. M.P. and E.D.P. for investigation and synthesis of the compounds. M.P. for biological studies and for the original draft preparation. M.P. and A.R.S. for writing, review, and editing. All authors have given approval to the final version of the manuscript.

This research was funded by the Natural Sciences and Engineering Research Council, grant number RGPIN-2021-03128. Access to the NMR via the Regional Centre for Magnetic Resonance (UdeM – Chemistry) was possible due to funding from the Canada Foundation for Innovation and the Institute Courtois.

The authors declare no competing financial interest.