Effective therapy of advanced breast cancer through synergistic anticancer by paclitaxel and P-glycoprotein inhibitor

Abstract

Multi-drug resistance (MDR) in advanced breast cancer (ABC) is triggered by the high expression of P-glycoprotein (P-gp), which reduces intracellular concentration of anti-tumor drugs, in turn preventing oxidative stress damage to cytoplasmic and mitochondrial membranes. It is therefore of clinical relevance to develop P-gp-specific targeted nanocarriers for the treatment of drug resistant ABC. Herein, a drug carrier targeting CD44 and mitochondria was synthesised for the delivery of encequidar (ER, P-gp inhibitor) and paclitaxel (PTX). HT@ER/PTX nanoparticles (ER:PTX molar ratio 1:1) had excellent P-gp inhibition ability and targeted mitochondria to induce apoptosis in MCF-7/PTX cells in vitro. Furthermore, HT@ER/PTX nanocarriers showed more anti-tumor efficacy than PTX (Taxol®) in a xenograft mouse model of MCF-7/PTX cells; the tumor inhibitory rates of HT@ER/PTX nanoparticles and Taxol® were 72.64% ± 4.41% and 32.36% ± 4.09%, respectively. The survival of tumor-bearing mice administered HT@ER/PTX nanoparticles was prolonged compared to that of the mice treated with Taxol®. In addition, HT@ER/PTX not only inhibited P-gp-mediated removal of toxic lipid peroxidation byproducts resulting from anti-tumor drugs but also upregulated the expression of mitochondrial dynamics-related protein, fostering oxidative stress damage, which induced activation of the Caspase-3 apoptosis pathway. Our findings indicate that mitochondria targeted co-delivery of anti-tumor drugs and P-gp inhibitors could be a practical approach in treating multi-drug resistance in ABC.

Graphical abstract

1. Introduction

In the last ten years, the incidence and mortality rates of advanced breast cancer (ABC) have significantly increased worldwide. Breast cancer could spread to other parts of the body, such as bones, lungs, liver, or brain, and is the leading cause of cancer death in females [[1], [2], [3]]. Chemotherapy is still one of the most commonly used treatment options for ABC; however, due to the emergence of drug resistance, its effectiveness is low [3,4]. It has been reported that repeated treatments cause the development of drug resistance in tumor cells to all chemotherapeutic agents, including the specific agent [[5], [6], [7], [8]]. This phenomenon, called multi-drug resistance (MDR), is characterized by resistance to drugs with different mechanisms of action or chemical structures.

MDR in ABC is ascribed to the elevated expression of P-glycoprotein (P-gp) in the cytoplasm and mitochondrial membrane of tumor cells, in turn keeping cellular levels of chemotherapeutic drugs below the cell-killing threshold [6,7,[9], [10], [11], [12], [13], [14]]. Furthermore, Gergely Szak'acs’ group found that P-gp removes toxic lipid peroxidation byproducts from drug-tolerant persister cells, thus maintaining the stemness of tumor cells. Prolonged administration of P-gp inhibitors during drug holidays could likely prevent or delay therapy resistance [10]. Thus, realizing higher intratumoral accumulation of P-gp inhibitors and anti-tumor drugs may be practical for alleviating MDR and enhancing anti-tumor efficacy.

Encequidar (ER), a specific P-gp inhibitor, has been used to improve the bioavailability of P-gp substrate drugs through oral administration [[15], [16], [17]]. Herein, we synthesised ER using a low cost synthetic route that resulted in a high yield compared to previous findings (Scheme 2) [15]. In addition, we found that an equimolar ratio of ER and paclitaxel (PTX) (ER:PTX = 1:1) had excellent P-gp inhibition ability and significantly reduced the viability of the PTX-resistant human breast cancer cell line MCF-7/PTX in vitro. However, ER could not penetrate tumor tissues and specifically inhibit P-gp overexpression in cytoplasmic and mitochondrial membrane in vivo, limiting its efficacy against MDR [[18], [19], [20]]. Thus, it was cardinal to develop selective delivery systems targeting ER and PTX to the cytoplasm and mitochondrial membranes of tumor cells in an attempt to overcome MDR.

Scheme 2.

Chemical synthesis of ER. (a) 4-methylbenzenesulfonhydrazide, EtOH, 65 °C, 0.5 h; 90% for S2; (b) Thionyl chloride, DMF, DCM, 85 °C, 5 h; 98% for S9; (c) 1-(2-bromoethyl)-4-nitrobenzene, acetonitrile, 45 °C, 2 h, RT, 8 h; 80% for S4; (d) Acetic acid, Zn, DCM, RT, 3 h; 85% for S5; (e) S2, Sodium nitrite, H₂O, hydrochloric acid, ethanol, pyridine, 0 °C, 0.5 h, 25 °C, 3 h; 53% for S6; (f) Acetic acid, Zn, DCM, rt, 3 h; 90% for S7; (g) S9, Pyridine DCM, 0 °C, 6 h; 85% for ER.

CD44, an adhesion receptor highly expressed on the cell membranes of MDR ABCs, plays a crucial role in cell adhesion, signaling, and migration [21]. Clinical data has confirmed that a high level of CD44 contributes to a poor prognosis by maintaining tumor stem cell properties and heterogeneity in the microenvironment [[22], [23], [24]]. Therefore, CD44 can serve as a prospective target to attenuate chemoresistance, invasion, and metastasis of MDR ABCs. Notably, as a natural ligand of CD44 [25,26], hyaluronic acid (HA) can specifically bind to CD44 receptors and regulate a variety of biological behaviors in tumor cells, such as adhesion and migration [27]. Owing to this, HA can be used to modify drugs vectors, enhancing their targeting capabilities against MDR tumors.

Mitochondria are bioenergetic and signaling organelles that play a crucial role in regulating apoptosis, metabolism, and the progression of cancer cells [28,29]. Accumulating evidence has suggested that P-gp contributes to the reduction of mitochondrial damage by increasing the efflux of anti-tumor drugs from mitochondria, resulting in MDR at specific subcellular organelles [9,30]. Moreover, a high mitochondrial membrane potential also contributes to maintaining chemoresistance by promoting ATP generation and increasing P-gp activity [13,31]. The highly polarized mitochondrial membranes and mitochondrial membrane P-gp are significant targets for therapeutic interventions aimed at overcoming MDR. Therefore, selective delivery of P-gp inhibitors and chemotherapeutic drugs to mitochondria might be a promising therapeutic strategy to attenuate chemoresistance and enhance the efficiency of chemotherapy [32].

Notably, PEG_2k-PDLLA_2k nanomaterials extend in vivo drug circulation, lower systemic drug exposure, and enable drug-specific delivery [33]. The resulting nanomicelles enter cells via endocytosis, overcoming the drug-resistant effects of cell membrane P-gp, and passively accumulate in tumor tissues through the enhanced permeability and retention (EPR) effect [34]. Our preliminary studies affirmed that the triphenyl phosphine (TPP)-modified PEG_2k-PDLLA_2k (termed TPP-PEG_2k-PDLLA_2k, Scheme 1) cationic polymeric carrier displayed excellent mitochondrial targeting and prolonged circulation upon systemic administration [35].

Scheme 1.

Schematic representation of the assembly of HT@ER/PTX nanomicelles and mode of action. ER provided synergistic effect with PTX through upregulating the expression of DRP1 to promote oxidative stress damage, and lead to mitochondrial dysfunction.

Considering the aforementioned factors, we fabricated a novel delivery system (termed HT@ER/PTX, Scheme 1) loading PTX and ER which could stepwise target CD44 receptors and mitochondria [[36], [37], [38], [39]]. Mechanistically, the nanoparticles, with their HA surface, facilitated efficient drug internalisation through CD44 receptor-mediated endocytosis and exhibited enhanced accumulation in the mitochondria, attributed to the positively charged nano-nucleus. Encouragingly, the HT@ER/PTX nanoparticles simultaneously delivered ER and PTX to the mitochondria of cancerous cells with longer systemic circulation and superior anti-tumor efficacy than Taxol® (a clinically used anti-tumor drug) in tumor-bearing mice. We further evaluated the mechanisms for reversing MDR through the combination of P-gp inhibitors and anti-tumor drugs from a novel perspective. In a PTX-resistant human breast cancer model, PTX and ER were co-delivered through mitochondria-targeting, resulting in the upregulation of mitochondrial dynamics-related protein expression and increased oxidative stress damage. These findings corroborated those by Gergely Szak'acs’ group [10]. This strategy holds promise for improving the treatment outcomes in drug-resistant tumors and provides a potential avenue for overcoming MDR in ABC.

2. Materials and methods

2.1. Materials

The comprehensive information, including the reagents, cell line, and animal details, was described in the Supporting information.

2.2. P-gp inhibitor ER and TPP-PEG_2k-PDLLA_2k synthesis

The synthesis experimental method and characterization of ER and TPP-PEG_2k-PDLLA_2k polymers were exhaustively elucidated in the Supporting information.

2.3. Preparation and characterization of nanoparticles (NPs)

The CD44-targeted dual-drug co-loaded HT@ER/PTX NPs were prepared using a modified nanoprecipitation method. TPP-PEG_2k-PDLLA_2k (35.55 mg), PTX (1.89 mg) and ER (1.79 mg) were dissolved in dimethyl sulfoxide (0.4 mL), and the mixture was continuously sonicated for 10 min to ensure complete dissolution. Then, the solution was added dropwise into deionized water (5 mL) and continuous stirred for 1 h at room temperature. The mixture was centrifuged with ultrafiltration (MWCO 3.5 kDa) to remove the nonencapsulated drug and DMSO to afford cationic nanoparticles T@ER/PTX NPs. After that, T@ER/PTX NPs were added dropwise into deionized water (2 mL) which contained 3.5 mg of HA. Subsequently, the solution was stirred for 1 h at 25 °C to afford HT@ER/PTX NPs. For the purpose of making further comparisons in experiments, HT@PTX nanoparticles without ER loading were similarly prepared, excluding the addition of ER during the process. The morphology of HT@ER/PTX nanoparticles was determined using a Talos L120C transmission electron microscope (TEM, Thermofisher, USA). The size and zeta potentials of all nanoparticles were detected by ZetaSizer Nano ZS90 (Malvern Panalytical, England).

The drug encapsulation efficiencies (DEE) and drug loading capacity (DLC) of PTX and ER were detected by a LC1260 high-performance liquid chromatography system (HPLC, Agilent, US, UV 227 nm and 254 nm, mobile phase: acetonitrile/water = 10/90 to 100/0, with 0.1% formic acid, flow rate: 1.0 mL/min). After the HPLC quantitative analysis, the DEE and DLC of PTX and ER can be calculated by the following formulas:

$$\text{DEE}\%=\frac{\text{weight}\text{of}\text{ER}/\text{PTX}\text{in}\text{nanoparticles}}{\text{weight}\text{of}\text{ER}/\text{PTX}\text{fed}}\times 100\%$$

$$\text{DLC}\%=\frac{\text{weight}\text{of}\text{the}\text{ER}/\text{PTX}\text{in}\text{nanoparticles}}{\text{weight}\text{of}\text{feeding}\text{polymer}\text{and}\text{ER}/\text{PTX}}\times 100\%$$

Furthermore, we also detected the stability of the HT@PTX and HT@ER/PTX in different conditions in vitro according to the protocol described in Ref. [40].

2.4. Drug release

The drug release profile of T@ER/PTX and HT@ER/PTX nanoparticles were evaluated through the dynamic membrane dialysis method in Phosphate buffer saline containing 0.5% (w/v) Tween-80 (PBS, pH = 7.4 and 5.5). The T@ER/PTX nanoparticles (containing 0.355 mg PTX and 0.353 mg ER) and HT@ER/PTX nanoparticles (containing 0.324 mg PTX and 0.321 mg ER) solutions were transferred into the dialysis bags (MWCO = 3.5 kDa) and were immersed in release medium (50 mL, pH = 7.4 and 5.5). After that, they were incubated in the orbital shaker incubator at 37 °C with gentle shaking at 100 rpm. At the set time interval, 2 mL of solution was withdrawn, and 2 mL of release medium was replenished. After sample collection is completed at each time point, the drug concentration was detected via HPLC (According to the previous section). The accumulated drug release amount can be calculated by the following formulas:

$$\text{Accumulated}\text{drug}\text{release}\text{amount}(\%)=\frac{{\mathrm{M}}_t}{{\mathrm{M}}_{\text{total}}}\times 100\%$$

M_t means the amount of drug released at the different time interval, and M_total means the total amount of drug.

2.5. Cell uptake and mitochondria localization of micelles

The uptake behavior of HT@ER/PTX NPs by MCF-7/PTX cells was investigated using Coumarin 6 (Cou-6) as a model drug. Firstly, HA/TPP-PEG_2k-PDLLA_2k was mixed with Cou-6 and dialyzed to afford HT/Cou-6 NPs. The concentration of Cou-6 was determined by Fluorescence spectrophotometer (FP-6200, Ex = 456 nm, Em = 506 nm). MCF-7/PTX cells were seeded in a 6-well plate (2 × 10⁵ cells/well), incubated at 37 °C in 5% CO₂. After 12 h, cells were incubated with free Cou-6 and HT/Cou-6 for 8 h, and detected by flow cytometry (NovoCyte, Agilent, USA).

MCF-7/PTX cells were seeded in confocal dishes at a density of 1.5 × 10⁴ per dish and incubated in the same way described in the previous paragraph. After it were incubated with free Cou-6 and HT/Cou-6 for 8 h, nucleus and mitochondria were labeled with Hoechst33342 and MitoTracker Red respectively and imaged by laser scanning confocal microscope (CLSM, TCS SP8 STED, Leica, Germany).

2.6. In vitro anti-tumor activity

2.6.1. Cytotoxicity studies

A CCK-8 assay was performed to measure the cytotoxicity of Taxol®, ER + PTX, HT@PTX, HT@ER/PTX in MCF-7 and MCF-7/PTX cells. Firstly, MCF-7 and MCF-7/PTX cells were seeded in a 96-well plate (5 × 10³/well) cultured for 12 h at 37 °C in 5% CO₂. After incubation with various drug formulations at different PTX concentrations for 24 h or 48 h, the original medium was discarded. Subsequently, the cells were incubated with fresh medium containing a 10% CCK-8 solution at 37 °C. The OD absorbance values at 490 nm were measured in each well using a microplate reader (Multiskan FC, Thermofisher, USA), and the cell viability of the tumor was calculated using the following formulas:

$$\text{Cell}\text{viability}\%=\frac{\mathrm{A}\text{sample}‐\mathrm{A}\text{blank}}{\mathrm{A}\text{control}‐\mathrm{A}\text{blank}}\times 100\%$$

2.6.2. Cell apoptosis analysis

MCF-7 and MCF-7/PTX cells were seeded in 6-well plates (2 × 10⁵ cells/well) and incubated at 37 °C in 5% CO₂ for 12 h. After incubating with different PTX formulations (4 μg/mL) for 24 h, the original medium was discarded, and the cells were washed with cold PBS. After incubation with Annexin V-FITC apoptosis detection kit, the cells were collected and detected by flow cytometry.

2.6.3. Cleaved Caspase-3 immunofluorescence staining

MCF-7/PTX cells were seeded in confocal dishes at a density of 1.5 × 10⁴ per dish and incubated in the same way described in the previous paragraph. After incubating with different PTX formulations (4 μg/mL) for 24 h, the original medium was discarded, and the cells were fixed using 4% paraformaldehyde (PFA), followed by permeabilization with 0.1% Triton X-100, and subsequent incubation with 1% bovine serum albumin for 30 min. Finally, the cells were incubated with Cleaved Caspase-3 antibody (1：500, Abcam) for 24 h at 4 °C. After that, the cells were washed and then stained with fluorochrome-conjugated secondary antibody. Subsequently, DAPI was used to label the nucleus and observed by CLSM.

2.6.4. TUNEL fluorescence staining

MCF-7/PTX cells were seeded and cultivated using the method described in the previous section. After 24 h of incubation with different PTX formulations (4 μg/mL) for 24 h, the cells were washed with PBS and fixed with 4% PFA for 25 min. Then the cells were permeabilized with 0.2% Triton X-100 for 5 min at room temperature. Subsequently, the cells were washed with PBS and incubated with TUNEL apoptosis detection kit. Finally, the cells were washed, and the nuclei were labeled with DAPI and observed using CLSM.

2.6.5. In vitro chemotherapeutic efficacy

To evaluate the chemotherapeutic efficacy of different PTX formulations in vitro, MCF-7/PTX cells were seeded and cultivated using the method described in the previous section. Afterward, the cells were incubated with Taxol®, ER + PTX, HT@PTX, and HT@ER/PTX at a PTX concentration of 4 μg/mL for 24 h. The cell viabilities were assessed using a living/dead cell double staining kit and observed with CLSM.

2.7. Mechanism investigations of HT@ER/PTX in overcoming MDR

2.7.1. ATP level assay

MCF-7/PTX cells were seeded in 6-well plates (2 × 10⁵ cells/well) and incubated at 37 °C in 5% CO₂ for 12 h. After incubating with different PTX formulations (4 μg/mL) for 12 h, the original medium was discarded, and the cells were washed with cold PBS. Afterward, the cells were lysed using ATP lysis buffer and centrifuged to obtain the supernatant for further analysis. The ATP levels in the treatment groups with different PTX formulations were measured using an ATP chemiluminescence assay kit (Elabscience Biotechnology Co., Ltd, Wuhan, China) and a multifunctional microplate reader (Spark, Tecan, Switzerland).

2.7.2. Mitochondrial function test

MCF-7/PTX cells were seeded in confocal dishes (1.5 × 10⁴ cells/dish) and incubated at 37 °C in 5% CO₂ for 12 h. Afterward, the cells were incubated with Taxol®, ER + PTX, HT@PTX, and HT@ER/PTX at a PTX concentration of 4 μg/mL for 12 h. Finally, the cells were washed with cold PBS, and incubated with mitochondrial membrane potential assay kit (JC-1). After JC-1 incubation, the cells were imaged by CLSM.

2.7.3. Western blot analysis

MCF-7/PTX cells were seeded in 6-well plates (2 × 10⁵ cells/well) and incubated at 37 °C in 5% CO₂ for 12 h. After 24 h of incubation with different PTX (4 μg/mL) drug formulations, cells were lysed for 0.5 h. Subsequently, the total protein concentrations were analysed by a BCA protein assay kit quantitatively. After that, the proteins of treatment groups were separated by SDS-PAGE, and transferred onto PVDF membranes (0.45-μm, Labselect, Beijing, China) for blocking by 5% bovine serum albumin (BSA, Coolaber, Beijing, China) for 2 h. After incubating with primary and secondary antibodies for the specified durations, the immunoreactive complexes were visualized using the Syngene Gel Imaging System (Tanon, Shanghai, China), and detected by ImageJ software. All values were normalized to those of GAPDH.

2.7.4. Quantitative real-time PCR analysis

MCF-7/PTX cells were seeded in 6-well plates (2 × 10⁵ cells/well) and incubated at 37 °C in 5% CO₂ for 12 h. After 24 h of incubation with various PTX (4 μg/mL) drug formulations, the cells were washed with PBS, and total RNA was extracted using TRIzol. Afterward, a reverse transcription kit was used to generate the cDNA. Quantitative real-time PCR (qRT-PCR) was performed by fluorogenic quantitative PCR detection system (ABI 7500, Thermofisher, USA), and the expression levels of RNAs were calculated according to the comparative Ct method (Please see the details in Supporting information).

2.8. Living images

The biodistribution of HT@ER/PTX NPs in the MCF-7/PTX xenograft tumor model was investigated using DiR as a model drug. HA/TPP-PEG_2k-PDLLA_2k was mixed with DiR and dialyzed to afford DiR NPs. The concentration of DiR was determined by Fluorescence spectrophotometer (FP-6200, Ex = 748 nm, Em = 780 nm). When the tumor volume reached 100 mm³, the mice were randomly divided into three groups and received injections of different DiR formulations. According to the set time interval, the fluorescence signal of DiR in mice was observed by IVIS Lumins XR (Caliper Life Sciences, USA). (Please see the details in Supporting information).

2.9. In vivo tumor growth inhibition

MCF-7/PTX xenograft tumor model was constructed to investigate the anti-tumor efficacy of different PTX formulations in vivo. When the tumor volume reached 100 mm³, the mice were randomly assigned to five groups (saline, Taxol®, ER (5 mg/kg) + PTX, HT@PTX, HT@ER/PTX) with twelve animals per group. Four groups of mice were administered different PTX formulations (5 mg/kg), the control group received saline, and the treatments were administered five times (once every two days). After treatment, the main organs (heart, liver, lung, kidney, spleen) and tumors were embedded in paraffin, making pathological section, and H&E stain to evaluate safety. In order to calculate the tumor inhibition rate of different drug formulations, tumor weight and tumor volume were measured precisely. The tumor inhibition rate could be calculated by the following formula:

$$\text{Tumor}\text{inhibition}\text{rate}\%=\frac{\text{tumor}\text{weight}(\text{Saline})‐\text{tumor}\text{weight}(\text{treatment})}{\text{tumor}\text{weight}(\text{Saline})}\times 100\%$$

Furthermore, the tumor tissues were performed histopathological and analysis by immunofluorescence (TUNEL, Ki67 and DRP1) and immunohistochemical (P-gp, XIAP and TUBB3) to evaluate the mechanisms of HT@ER/PTX NPs for overcoming drug resistance in MCF-7/PTX xenograft tumor model.

3. Results and discussion

3.1. P-gp inhibitor ER and TPP-PEG_2k-PDLLA_2k synthesis

A recent study reported that co-administration of the P-gp inhibitor ER increased the oral bioavailability and efficacy of PTX [18]. Herein, we improved the synthesis process to obtain high-purity ER (purity 99.0%) in a simple and convenient manner. Compounds S2/S4/S9 were synthesised from commercially available compounds S1/S3 via a sequence of linear steps (Scheme 2). Compound S5 was obtained via a reduction reaction catalysed by Zn powder and acetic acid. Compound S6 was obtained via a diazo reaction between compounds S2 and S5 under sodium nitrite-catalysed conditions. Compound S7 was obtained via the reduction of S6. Finally, ER was obtained via the pyridine-catalysed acylation reaction between compounds S9 and S7. The structure of ER was characterised using mass spectrometry (MS) and nuclear magnetic resonance spectroscopy (NMR). The purity and yield of ER were determined using high performance liquid chromatography (HPLC, see in Supplementary Fig. 18).

Conjugating therapeutic agents with cationic TPP is a highly promising strategy for mitochondrial targeting to enhance the pharmacokinetic properties and bioavailability of drugs [41,42]. TPP-PEG_2k-PDLLA_2k was synthesised by a condensation reaction using commercially available NH₂-PEG_2k-PDLLA_2k and compound I in a mixture of 3-ethylcarbodiimide hydrochloride and 1-hydroxybenzotriazole solution at 25 °C for 3 h with a yield of 60 % (Scheme 3). The structure of TPP-PEG_2k-PDLLA_2k was characterised using ¹H NMR spectroscopy (Supplementary Fig. 19).

Scheme 3.

Chemical synthesis of TPP-PEG_2k-PDLLA_2k. (a) 3-ethylcarbodiimide hydrochloride，1-hydroxybenzotriazole, 25 °C, 3 h; 60% for TPP-PEG_2k-PDLLA_2k.

3.2. Cytotoxicity and P-gp inhibitory effect of ER

Initially, the effect of ER on cytotoxicity was evaluated in fibroblast L929 and MCF-7/PTX tumor cells using the Cell Counting Kit-8 (CCK-8) assay. The viabilities of L929 cells and MCF-7/PTX cells were higher than 80% after incubation with ER at different concentrations for 24 h (Fig. 1A and B). This indicated that ER has little cytotoxicity to normal and tumor cells, with an IC₅₀ greater than 100 μM. In addition, all cells exhibited almost no growth inhibitory effect with ER treatment at 5 μM, indicating that this concentration could be the maximum concentration used for MDR reversal experiments.

Fig. 1.

The toxicity and P-gp inhibitory effect of ER in vitro. Viabilities of L929 (A) and MCF-7/PTX (B) cells incubated with different concentrations of ER for 24 h. The MCF-7/PTX cells was incubated with PBS (Control), Verapamil (10 μM), and ER at the indicated concentrations (0.1, 0.5, 2.5, 5.0 μM) for 4 h. Flow cytometry analysis (C), bar chart quantifying the accumulation (D) and cellular uptake images (E) of the free Rho 123 dye in the MCF-7/PTX cells. Scale bar = 10 μm. All data are presented as mean ± SD (n = 3). *P < 0.05, **P < 0.01 (one-way ANOVA).

As mentioned above, the efflux function of P-gp plays an important role in MDR. Thus, Rho123 (fluorescent substrate of P-gp) was selected as a probe to investigate the inhibition of P-gp by ER. Briefly, MCF-7/PTX cells were incubated with ER and Rho123 at different concentrations, and intracellular fluorescence intensity was measured to indirectly characterise P-gp inhibition by ER. Rho123 intensity significantly increased compared to that in the control group when verapamil or ER was added to the cells (Fig. 1C). Rho123 intensity in the ER (5.0 μM) group showed a two-fold increase compared to that in the verapamil-treated group (Fig. 1D). This was confirmed by cellular uptake images from the Rho123 accumulation assay (Fig. 1E). Thus, we concluded that ER could block the outflow of Rho123 from MCF-7/PTX cells and increase their uptake by inhibiting P-gp function (See the supporting information for experimental procedures).

3.3. Preparation and characterization of nanoparticles (NPs)

In this study, we designed a novel therapeutic system (HT@ER/PTX) to overcome MDR in ABC by combining PTX with ER. The Chou–Talalay method [43] was employed to calculate the Combination Index (CI) at various molar ratios of ER and PTX to determine the optimal ratio to achieve synergistic effects between the two. Specifically, the IC₅₀ and CI values were measured for different molar ratios (ER: PTX = 8:1, 5:1, 2:1, 1:1, 1:2, 1:5, 1:8) using MCF-7/PTX cells within a 24-h time frame. The most pronounced synergistic effect between ER and PTX was observed at a 1:1 M ratio, as evidenced by the lowest CI value when the fraction affected (Fa) was equal to 0.5 (Supplementary Fig. 20). The HT@ER/PTX NPs were composed of a hydrophilic HA shell and a hydrophobic cationic T@ER/PTX nanocore, co-loaded with PTX and ER in a 1:1 M ratio (Scheme 1).

Prior studies demonstrated that TPP-PEG_2k-PDLLA_2k possesses notable amphiphilic properties, enabling self-assembly into NPs for effective loading of anticancer drugs and targeted delivery to the mitochondria [35]. Therefore, first, TPP-PEG_2k-PDLLA_2k was synthesised and characterised by ¹H NMR and subsequently used to prepare the T@ER/PTX nanocore with equal PTX and ER loading using the nanoprecipitation method. The average diameter of T@ER/PTX NPs was 18.8 ± 1.22 nm, and the zeta potential was approximately 9.14 ± 3.21 mV (Fig. 2A and B).

Fig. 2.

Preparation and characterization of micelles. (A, B, C, D) Size distribution, PDI and z-potential of T@ER/PTX and HT@ER/PTX micelles measured by DLS machine respectively; (E) Morphology of HT@ER/PTX micelles measured by transmission electron microscope (TEM, scale bar: 100 nm); (F) PTX's release from HT@ER/PTX and T@ER/PTX nanomicells at different pH. Data are presented as the mean ± SD (n = 3).

HA is a natural ligand for the CD44 receptor in tumor cells and is widely used in drug delivery owing to its good biocompatibility and degradability [44]. “Core–shell” structured HT@ER/PTX NPs were formed by electrostatic modification of the surface of T@ER/PTX NPs, and the zeta potential was approximately −15.1 ± 3.45 mV, which decreased the surface potential and dramatically improved stability (Fig. 2C and D). Furthermore, the size of HT@ER/PTX NPs gradually increased with a relatively wide distribution as shown using dynamic light scattering. Transmission electron microscopy (TEM) images showed that HT@ER/PTX NPs had a regular spherical shape and compact structure (Fig. 2E). These characteristics were in agreement with previous results and confirmed that HA was successfully modified on the surface of the T@ER/PTX nanocore [35].

Both the drug encapsulation efficiency (DEE) and drug-loading capacity (DLC) were measured using HPLC. As shown in Table 1, the total loading content of both PTX and ER in T@ER/PTX nanocores could reach up to 9.27% ± 1.60% with drug encapsulation efficiencies more than 75.00%. When the feed amounts of PTX and ER were 1.89 mg and 1.79 mg, their drug loading contents were 5.02% ± 0.95% and 4.25% ± 1.01%, respectively for T@ER/PTX NPs, and 4.23% ± 1.34% and 4.03% ± 1.25%, respectively for HT@ER/PTX NPs. Following HA coating, there was almost no leakage of PTX or ER from the T@ER/PTX nanocore. These results indicated that the T@ER/PTX nanocores and HT@ER/PTX NPs could efficiently co-deliver PTX and ER, providing strong evidence for achieving synergistic anticancer effects. To uphold the overall integrity of the study, we formulated and characterized HT@PTX NPs as reserve samples for subsequent experiments (Supplementary Fig. 21).

The stability of HT@PTX and HT@ER/PTX NPs was investigated by adding the carriers to phosphate buffered saline (PBS) or cell culture media containing 10% FBS at 37 °C for 24 h. There were no significant changes in particle size (Supplementary Fig. 22). This demonstrated that the negatively charged HA shell protected ER/PTX from interference with plasma proteins, enhancing its blood circulation stability. However, after induced with HAase, the zeta potential of HT@ER/PTX NPs exhibited obvious change with pH dependence (Supplementary Fig. 23). The values increased from −15.3 mV to +8.9 mV and −3.2 mV at pH 5.5 and 7.4, respectively, following 6 h of HAase induction. This suggested that HT@ER/PTX NPs successfully escaped from lysosomes and entered the cytoplasm through the proton sponge effect, in accordance with literature reports [45] (Supplementary Fig. 24).

The release behaviours of ER and PTX from HT@ER/PTX were investigated at different pH conditions. In a medium of pH 7.4, the release of PTX from HT@ER/PTX nanomicelles at 24 h was approximately 35% (Fig. 2F). However, at pH 5.5, it rapidly increased to 81%. This indicated that the release of PTX in plasma (pH = 7.4) was low, while it rapidly released in the lysosome (pH 5.0–6.5), achieving precise drug release within tumor cells. The rate of PTX release from T@ER/PTX nanomicelles, akin to their HT@ER/PTX counterparts, remained approximately constant for 72 h, indicating a prolonged release.

3.4. HT@ER/PTX NPs are taken up by cells via clathrin-mediated endocytosis and specifically accumulate in mitochondria

We initially investigated CD44 expression in MCF-7 and MCF-7/PTX cells to determine the cellular uptake and mitochondrial localization of HT@ER/PTX NPs (Fig. 3A). CD44 expression in MCF-7/PTX cells showed a 50-fold increase compared to MCF-7 cells (Fig. 3B and C). Therefore, we selected the MCF-7/PTX cancer cell line (which overexpresses CD44) as our model system to explore the ability of HT@ER/PTX NPs to deliver PTX to drug-resistant tumor cells.

Fig. 3.

Cell uptake behavior and mitochondrial colocalization of micelles. (A) CD44 expression on the tested MCF-7 and MCF-7/PTX cells by immunofluorescence. MCF-7 and MCF-7/PTX cells were treated as indicated, stained with CD44 antibodies (green) and DAPI for nuclear staining (blue), and visualized by confocal microscopy at a z-plane near to the center of the cell. Scale bar = 50 μm. (B) and (C) Mean fluorescence intensity of CD44 on the tested MCF-7 and MCF-7/PTX cells detected by flow cytometry. (D) and (E) Examination of cellular uptake of the nanotaxanes upon pretreatment with inhibitors of specific endocytosis pathways. Chlorpromazine (28 μM), Colchicine (5 μM), Nystatin (20 μM) were used to block clathrin-mediated endocytosis, macropinocytosis, and caveolin-mediated endocytosis, respectively. (F) and (G) fluorescence intensity of MCF-7/PTX cells after 8 h of incubation with Free Cou-6 and HT/Cou-6 (targeted) detected by flow cytometry. (H) Co-localization of the HT/Cou-6 NPs into mitochondria at MCF-7/PTX cells at 8 h observed by CLSM. The mitochondria were stained by MitoTracker Red. Cou-6 emits green fluorescence itself. Yellow fluorescence indicates the overlay between Cou-6 and MitoTracker Red. Scale bar = 10 μm. All data are presented as mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001 (one-way ANOVA). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Subsequently, DiI was used as a model drug to investigate the transport of HT@ER/PTX NPs across cell membranes. Warming (37 °C) completely increased DiI/HT NPs uptake compared with the ineffective NP uptake at 4 °C, implying an energy-dependent endocytosis pathway (Fig. 3D). Furthermore, the cells were pre-incubated with different internalisation inhibitors to investigate uptake pathway at 37 °C (Fig. 3E). The fluorescence intensity of the chlorpromazine-treated group was the lowest among all the internalisation inhibitor groups. Hence, the pathway by which HT@ER/PTX NPs enter cells was postulated as clathrin-mediated endocytosis.

Furthermore, MCF-7/PTX cells were treated with fluorescent Cou-6-labeled HT@ER/PTX NPs to detect cellular uptake of the nanomicelles using flow cytometry. The intracellular accumulation of Cou-6 in the HT/Cou-6 group increased 3.14-fold compared to that in the free Cou-6 group (Fig. 3F and G). These results suggest that endocytosis of HT/Cou-6 nanomicelles by MCF-7/PTX cells is greatly facilitated by CD44-mediated internalisation.

We further investigated whether these endocytosed nanomedicines could localise to the mitochondrial compartments. MCF-7/PTX cells were incubated with HT/Cou-6 nanomicelles for 8 h, mitochondria staining was performed by MitoTracker red dye, and mitochondrial localization was observed by confocal laser scanning microscopy (CLSM). The merged yellow signal, arising from the overlay of green and red fluorescence, signifies the co-localization of nanomicelles with mitochondria. The merged light-yellow signal from HT/Cou-6 nanomicelles was significant compared to that of the free Cou-6 group, indicating that HT/Cou-6 nanomicelles localised in the mitochondria when incubated with MCF-7/PTX cells for 8 h (Fig. 3H).

3.5. In vitro anti-tumor efficiency

The CCK-8 assay was employed to assess MCF-7 and MCF-7/PTX cell growth inhibitory effects of different drug formulations, as well as the corresponding IC₅₀ values at 24 and 48 h (Fig. 4A and Table 2). All the experimental groups showed similar growth inhibition in MCF-7 cells. Contrarily, the HT@ER/PTX treatment group exhibited marginally increased toxicity and a slightly reduced IC₅₀, owing to the synergistic effects of PTX and ER. However, divergent experimental results were observed in drug-resistant MCF-7/PTX cells. Whether induced for 24 or 48 h, the inhibitory effect of PTX on cell proliferation was weak due to MDR, while ER + PTX and HT@PTX exhibited distinct effects in reversing MDR, leading to significant cell-killing activity. Interestingly, the highest cytotoxicity (IC₅₀ = 6.27 ± 2.07 μg/mL) was observed in the HT@ER/PTX treatment group, particularly at 24 h, and demonstrated a dose-dependent effect. It was confirmed that the combination of a P-gp inhibitor with PTX, as well as a nanodelivery strategy, was effective in inhibiting MDR.

Fig. 4.

The anti-tumor effect of micelles in vitro. (A) Cell viability value of Taxol®, ER + PTX, HT@PTX and HT@ER/PTX against MCF-7 and MCF-7/PTX cells after treatment for 24, 48 h. (B) Apoptosis analysis of MCF-7 and MCF-7/PTX cells at 24 h after various treatments (Q1: dead cells; Q2: late apoptotic cells; Q3: early apoptotic cells; Q4: living cells). All data are presented as mean ± SD (n = 3). *P < 0.05, **P < 0.01 (one-way ANOVA).

In addition, the drug resistant index (RI, the ratio of the IC₅₀ of MCF-7/PTX cells to MCF-7 cells in the same formulation at 48 h) of all formulations was determined. As shown in Table 2, the RI of Taxol®, ER + PTX, HT@PTX, and HT@ER/PTX was 355.8, 1.8, 3.9, and 1.6, respectively. HT@ER/PTX could sustain intracellular PTX concentration for an extended period, resulting in significant cell-killing activity, which is crucial for overcoming MDR in breast tumors.

Using flow cytometry, it was determined that MCF-7 cells treated with Taxol®, ER + PTX, HT@PTX, and HT@ER/PTX, respectively, showed 92.1%, 79.7%, 90.8%, and 88.8% viable cells, and 5.3%, 15.1%, 5.6%, and 5.33% late apoptotic cells (Fig. 4B). MCF-7/PTX cells treated with Taxol®, ER + PTX, HT@PTX, and HT@ER/PTX showed 96.3%, 87.3%, 95.4%, and 81.6% viable cells, and 2.52%, 9.69%, 3.08%, and 15.5% late apoptotic cells, respectively. Co-administration with ER or constructing a mitochondrial-targeted nanodelivery system (HT@ER/PTX) improved apoptosis and cytotoxicity compared with PTX alone.

Furthermore, tumor cell death was observed following treatment with different formulations of PTX (4 μg/mL), and some adherent cells transitioned to a suspended state as observed in calcein-AM/PI staining, wherein green/red meant live/dead cells, respectively. MCF-7/PTX cells treated with Taxol® maintained their fusiform shapes, and presented a slight red signal owing to MDR (Fig. 5A). However, an obvious red signal was observed in the ER + PTX- and HT@PTX-treated groups, which was attributed to the addition of the P-gp inhibitor ER and mitochondrial targeting, respectively. In contrast, the HT@ER/PTX treated group showed the strongest red signal, and all cells were almost suspended, indicating severe cell death.

Fig. 5.

The anti-tumor effect of micelles in MCF-7/PTX cells. (A) Living and dead cell staining of MCF-7/PTX cells after treatment for 24 h (green, living; red, dead; scale bar, 50 μm). (B) Cleaved Caspase-3 immunofluorescence of MCF-7/PTX cells after treatment for 24 h (green, Cleave Caspase-3; blue, DAPI; scale bar, 50 μm). (C) TUNEL assay of MCF-7/PTX cells after treatment for 24 h, as determined by CLSM (blue, DAPI; red, TUNEL; scale bar, 50 μm). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

To further study the anti-tumor effects of HT@ER/PTX in MCF-7/PTX cells, the expression of Cleaved Caspase-3 was evaluated using immunocytochemical staining. Consistent with previous research, the Taxol® treatment group showed a weak green signal, but ER + PTX- and HT@PTX-treated group showed an obvious green signal, demonstrating an increase in apoptosis (Fig. 5B). The HT@ER/PTX treatment group exhibited the strongest green signal, suggesting severe apoptosis, which was attributed to the mitochondria-targeted delivery of ER and PTX. In addition, these conclusions were confirmed by terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) analysis (Fig. 5C).

3.6. Mechanism investigations of HT@ER/PTX in overcoming MDR

Accumulating evidence has demonstrated that mitochondria play a key role in cancer progression, especially pronounced mitochondrial hyperpolarisation, causing accumulation of ATP and P-gp levels and MDR. Various experiments were designed to evaluate the efficacy of HT@ER/PTX NPs in PTX-resistant MCF-7/PTX cells to investigate the mechanism by which mitochondria-targeted paclitaxel overcomes MDR.

In this study, mitochondrial function including membrane potential (ΔΨm), reactive oxygen species (ROS), and ATP generation were evaluated by flow cytometry. Both the control and Taxol® treatment group cells showed slight green signals, suggesting high mitochondrial membrane potential (Fig. 6A). Exposure of the cells to ER + PTX and HT@ER/PTX NPs showed an apparent increase in the green fluorescence, resulting in red/green ratios of approximately 60% and 40%, respectively (Fig. 6B). Further, the ΔΨm of MCF-7//PTX was decreased, implying compromised mitochondrial function and early apoptosis. In contrast, HT@PTX NPs treatment showed no such effect, suggesting that ER played an important role in the drop of ΔΨm. Furthermore, dramatic ATP production suppression and ROS elevation were observed in the HT@ER/PTX NPs treatment group, corresponding to prominent apoptosis and the ΔΨm results (Fig. 6C and D).

Fig. 6.

Analysis of JC-1, ATP and P-gp expression in MCF-7/PTX cells. (A) and (B) JC-1 assay of MCF-7/PTX cells after treatment for 12 h, as determined by CLSM (red, J-aggregate; green, J-monomer; scale bar, 50 μm). (C) ATP assay of MCF-7/PTX cells after treatment for 12 h. (D) Measurements of ROS production after treatment for 12 h. (E) Analysis of Cyt-C and DRP1 by Western blot after treatment for 24 h in MCF-7/PTX cells. Scale bar = 50 μm. All data are presented as mean ± SD (n = 3). *P < 0.05, **P < 0.01 (one-way ANOVA). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Cytochrome C (Cyt-C) is an important component of the mitochondrial respiratory chain, and its release from the mitochondria is obvious evidence of mitochondrial dysfunction [46]. In the intact cell, apoptosis is intricately linked to the release of Cyt-C into the cytoplasm, a process mediated by dynamin-related protein 1 (DRP-1). Therefore, we used western blotting to investigate the expression of the proteins involved in this pathway (Fig. 6E). Cyt-C expression in the HT@ER/PTX NPs treatment group significantly increased (Supplementary Fig. 25A); no such effect was observed in the other groups. Notably, DRP1 expression significantly increased in cells treated with HT@ER/PTX (Supplementary Fig. 25B). Further, ER, in combination with PTX, was effective in enhancing apoptosis in MDR tumors, which is consistent with the above conclusions (Fig. 5B).

3.7. Expression of MDR related protein

Proteins related to PTX resistance were analysed by western blotting in MCF-7/PTX cells, including P-gp, XIAP (X-linked inhibitor of apoptosis), and TUBB3 (Beta-tubulin III) to further explore the underlying mechanism of HT@ER/PTX NPs in reversing MDR. The P-gp specific bands in the Taxol® treatment group were more obvious than those in the control group, implying that chemotherapy stimulated the expression of P-gp, thereby weakening the anti-tumor effect of PTX (Fig. 7A and B). Similar results were observed in the other treatment groups and were confirmed by quantitative real-time polymerase chain reaction (RT-qPCR) for P-gp (Fig. 7E). Collectively, these observations suggest that the mechanism by which HT@ER/PTX NPs reverse MDR may be related to the direct inhibition of the function of P-gp, but not its expression.

Fig. 7.

Proteins related to PTX-resistance were analysed in MCF-7/PTX cells. Analysis of P-gp, XIAP and TUBB3 expressions by Western blot (A), protein level of P-gp (B), XIAP (C), TUBB3 (D), qRT-PCR of MDR (E) and XIAP (F) after treatment for 24 h in MCF-7/PTX cells. All data are presented as mean ± SD (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001 (one-way ANOVA).

XIAP is an apoptotic protein that plays an important role in MDR by blocking the activation of the caspase cascade and promoting anti-apoptotic abilities [47]. Western blotting showed an obvious specific band in the Taxol® treated group, owing to chemotherapeutic stimulation (Fig. 7A and C). In contrast, the HT@ER/PTX NPs treated group showed significant band weakening, which was attributed to the mitochondria-targeting drug delivery strategy, consistent with the RT-qPCR results (Fig. 7F).

TUBB3 is the main component of spindle tubulin that reduces the anti-microtubule depolymerisation effect of paclitaxel and induces paclitaxel-resistance [48]. The TUBB3 expression level significantly decreased in the HT@ER/PTX NPs treatment group compared with that in the Taxol® treatment group (Fig. 7A and D, ***P < 0.001 versus Taxol® treatment). This result was interesting owing to the limitations of the scientific conditions, thus this mechanism of action was not studied in depth.

3.8. In vivo biodistribution

A DiR probe was loaded onto HA/TPP-PEG_2k-PDLLA_2k and TPP-PEG_2k-PDLLA_2k to generate DiR NP-TH NPs and DiR NP-T NPs, respectively, to investigate the biodistribution of HT@ER/PTX in vivo. The nude mice were anaesthetised and observed using IVIS Lumina XR (Caliper Life Sciences, USA) at set time points after administration of the different DiR formulations. The free DiR treatment group showed high liver aggregation fluorescence signals that were rapidly cleared from the body within 6 h. Interestingly, long-term retention was observed in the DiR NP-TH and DiR NP-T treatment groups, with accumulation gradually occurring at the tumor site within 6 h (Fig. 8A and C). To further investigate the biodistribution of NPs in vivo, major organs and tumors of tumor-bearing mice were excised for ex vivo imaging at 2, 6, 12 and 24 h post-injection. The fluorescence intensity of the DiR NP-T treatment group was mainly observed in the liver and slightly in tumor tissue, and disappeared within 12 h. After DiR NP-TH injection, a strong fluorescence signal was observed in tumor tissue, and reached a maximum at 12 h, owing to HA/CD44-mediated tumor active targeting (Fig. 8B). In addition, following ex vivo imaging at 12 h, the DiR NP-TH group exhibited the highest fluorescence intensity in tumor tissue, being 8.71-fold and 3.23-fold higher than that of the free DiR and DiR NP-T groups, respectively (Fig. 8D). Based on these findings, we concluded that the stepwise targeted delivery of tumor cell membranes and mitochondria could realise the long-term circulation of drugs in the body and preferential accumulation in tumors.

Fig. 8.

In vivo biodistribution of NPs in MCF-7/PTX tumor-bearing mice. (A) Living images of mice after administration of different DiR formulations (free DiR, DiR NP-TH and DiR NP-T) at 2, 6, 12 and 24 h. Tumors were marked by green circles. (B) Ex vivo fluorescence of tumors and organs isolated from mice after administration of different DiR formulations at 2, 6, 12 and 24 h. (H: heart, L: liver, Lu:l ung, K: kidney, S: spleen T: tumor). (C) Fluorescence intensity of living image at 2, 6, 12 and 24 h. (D) Fluorescence intensity of Ex vivo tumors and organs isolated from mice after administration of different DiR formulations at 12 h. Data are presented as the mean ± SD (n = 6). *P < 0.05, **P < 0.01, ***P < 0.001 (one-way ANOVA). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

3.9. In vivo efficiency of HT@ER/PTX NPs in overcoming MDR

Finally, we tested the ability of different PTX (5 mg/kg) formulations in inhibiting tumor growth in a MCF-7/PTX-xenograft Balb/c nude mouse tumor model. When the tumor volume reached 100 mm³, the mice were randomly divided into five groups and injected with PBS or different PTX formulations every other day for five consecutive doses (Fig. 9A). Images of the tumors removed from the mice after various treatments are shown in Fig. 9B. The tumor size in the Taxol® treatment group barely showed shrinkage due to PTX-resistance in vivo. In contrast, the tumor size in the ER + PTX treatment groups significantly decreased, demonstrating that ER promoted Taxol® anti-tumor activity. HT@PTX NPs also showed better anti-tumor efficacy than Taxol®, which could be ascribed to the efficient PTX tumor-accumulation achieved through CD44/TPP-mediated tumor active targeting. Furthermore, the HT@ER/PTX NPs were the smallest in size, confirming their superior anti-tumor efficacy in vivo. The tumor inhibition rates of Taxol®, ER + PTX, HT@PTX, and HT@ER/PTX were 32.36% ± 4.09%, 52.85% ± 2.81%, 58.53% ± 3.02%, and 72.64% ± 4.41%, respectively (Fig. 9C). Thus, stepwise targeted delivery of ER and PTX to tumor cell membranes and mitochondria suppressed MDR development in tumors. Tumor growth and animal survival curves confirmed this conclusion (Fig. 9D to G). The body weight of mice in the ER + PTX treatment group significantly decreased compared with that of mice in the HT@ER/PTX nanomicelle group (Fig. 9F). This was attributed to the inhibition of P-gp in normal tissues by ER, which increased systemic exposure to PTX and the occurrence of gastrointestinal side effects. However, the HT@ER/PTX treatment group showed almost no body weight change, which was attributed to the precise delivery of drugs to tumor tissue and the significantly reduced side effects. The HT@ER/PTX NPs treatment group showed the longest median survival time of 42 days, which was 1.68-fold and 1.55-fold longer than that of the saline and Taxol® groups, respectively (Fig. 9G). Taken together, these results show that the stepwise targeted delivery of tumor cell membranes and mitochondria, and P-gp inhibitory effect of ER are key factors in the high efficacy of HT@ER/PTX NPs against drug-resistant ABC in vivo.

Fig. 9.

In vivo anti-tumor efficiency of different formulations against MDR tumors. (A) Schematic illustration of tumor inoculation and timing of the treatment. (B) Ex vivo graphs of tumors showing after treatment with different formulations. (C) Tumor inhibition, (D) tumor weight, (E) tumor volume, (F) body weight, (G) animal survival. Data are presented as the mean ± SD (n = 12). *P < 0.05. **P < 0.01, ***P < 0.001 (one-way ANOVA).

Safety evaluation of the different drug formulations revealed no evident pathological alterations to the heart, spleen, lungs, or kidneys in each group based on haematoxylin and eosin (H&E) staining (Fig. 10A). However, the ER + PTX treatment group showed obvious pathological alterations of the liver, including hydropic degeneration and lymphocytic infiltration, which suggested that free ER may be hepatotoxic. In contrast, liver tissue was normal in the HT@ER/PTX group, confirming that the carrier was safe.

Fig. 10.

Histopathological analysis. (A) H&E staining of major organs harvested from MCF-7/PTX tumor-bearing mice treated with different NPs. (B) Microscopical images of H&E, TUNEL stained and Ki67 expression after treatment with different NPs in MCF-7/PTX xenograft tumor model. Quantitative data of the proliferating rate and apoptotic cell percentages are indicated in (C) and (D), respectively. Scale bar = 200 μm. The data are presented as the mean ± SD (n = 6). *P < 0.05, **P < 0.01, ***P < 0.001 (one-way ANOVA).

Tumor histological analysis was performed (including immunohistochemistry and immunofluorescence) on tumor tissues to further investigate the anti-tumor mechanism of HT@ER/PTX NPs in a drug-resistant tumor model. The saline- and PTX-treated groups displayed classical tumor tissue characteristics such as abnormal cell morphology, enlarged nuclei, and nuclear staining (Fig. 10B). However, after H&E staining, we observed some nuclear shrinkage in the ER + PTX- and HT@PTX NPs treatment groups, indicating slight cell necrosis. In addition, we observed prominent cell death, characterised by severe oedema, cytoplasmic shrinkage, and nuclear pyknosis after treatment with HT@ER/PTX NPs. The TUNEL assay (Fig. 10B and C) and Ki 67 staining (Fig. 10B and D, **P < 0.01 versus Taxol®) also confirmed that HT@ER/PTX NPs exhibited the highest efficacy in the drug-resistant tumor model.

Immunofluorescence and immunohistochemical staining were used to analyse the expression of DRP1, P-gp, XIAP, and TUBB3 (Supplementary Figs. 26A and 26B). HT@ER/PTX simultaneously downregulated the expression of XIAP and TUBB3, and upregulated DRP1 protein levels. This tendency was consistent with in vitro western blotting and mRNA results. These results indicated that HT@ER/PTX NPs had significant activity in reversing MDR, which holds promise for treating MDR in ABC by combination of paclitaxel with P-gp inhibitors.

4. Conclusions

Here, we prepared CD44 and mitochondria dual-targeted nanoparticles composed of a negatively charged HA shell and positively charged TPP-PEG_2k-PDLLA_2k core for co-delivery of ER and PTX, and elucidated their synergistic effects in MDR ABC. The delivery system showed potent activity in a PTX-resistant tumor model by inhibiting P-gp, inducing mitochondrial dysfunction, and extending the drug half-life. Various experiments were designed in vitro to elucidate the mechanisms of HT@ER/PTX nanoparticles in reversing MDR. At the protein level, HT@ER/PTX nanoparticles up-regulated DRP1 expression and induced mitochondrial metabolic disorders, thereby promoting oxidative stress reactions, including increased release of ROS and Cyt-C. Furthermore, a low level of TUBB3 contributed to the activation of the Caspase-3 apoptosis pathway by increasing the inhibitory effects of PTX on mitochondrial tubulin, as well as decreasing the expression level of XIAP. In vivo, HT@ER/PTX nanoparticles exhibited higher anti-tumor efficacy than Taxol® and prolonged the survival of mice with drug-resistant breast tumors. In summary, HT@ER/PTX nanoparticles were effective in steadily targeting CD44 receptors and mitochondria, imparting synergism between a P-gp inhibitor and PTX. We believe that our research holds significant value as a reference for the development of therapeutic drugs against MDR ABC.

Funding

This work was supported by the National Natural Science Foundation of China (NSFC31872754).

CRediT authorship contribution statement

Sifeng Zhu: Investigation. Chao Sun: Investigation. Zimin Cai: Data curation. Yunyan Li: Formal analysis. Wendian Liu: Data curation. Yun Luan: Investigation. Cheng Wang: Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Data availability

The data that has been used is confidential.