Liposome-Coated Arsenic–Manganese Complex for Magnetic Resonance Imaging-Guided Synergistic Therapy Against Carcinoma

Introduction

Primary carcinoma of the liver (PCL) is a type of cancer in the liver cells or intrahepatic bile duct cells, PCL is ranked as the fourth leading cause of cancer-associated deaths worldwide. Among them, hepatocellular carcinoma (HCC) accounts for the vast majority with a high incidence and mortality rate.^1,2 Even though various clinically adopted HCC treatment methods have shown to be significant to ameliorate the disease, the use of chemotherapy is often associated with adverse side effects and limited therapeutic efficacy.³ Thus, due to the limitation of chemotherapy, it is necessary to explore alternative strategies for the treatment of HCC.

Arsenic trioxide (ATO) is an active ingredient in traditional Chinese medicine and it has been approved by the United States Food and Drug Administration (FDA) as a first-line antitumor drug against acute promyelocytic leukemia.⁴ Furthermore, ATO has also demonstrated excellent antitumor effects on various solid tumors, eg, liver cancer, glioma, and breast cancer, by inhibiting their proliferation and inducing cell apoptosis.^5–10 Clinically, ATO is administered in the form of the arsenite ion (AsO₂⁻). Even though ATO can exhibit excellent antitumor effects, the narrow therapeutic window, the lack of specificity in the body distribution, liver toxicity, neurotoxicity, and skin and nail pigmentation have significantly hindered the mass adoption of ATO in the clinical setting. Thus, to utilize ATO as high efficacy antitumor drug, it is necessary to realize targeted ATO delivery to the tumors so that it can achieve high bioavailability and reduced side effects.^11–15

Manganese (Mn) is an essential element in the human body and it has been widely applied as a T1 contrast agent for cancer diagnosis due to its excellent biocompatibility.¹⁶ Moreover, Mn²⁺ can form insoluble complexes with multiple anionic groups, and their subsequent biodegradation can be triggered by the acidic environment of the tumor tissue.¹⁷ The released Mn²⁺ can then serve as a T1-weighted magnetic resonance imaging (MRI) contrast agent for real-time detection of the release and in vivo distribution of AsO₂⁻.¹⁸ Also, Mn²⁺ can catalyze the tumor-abundant hydrogen peroxide (H₂O₂) to generate toxic hydroxyl radicals (∙OH) via Fenton-like reactions (chemodynamic therapy, CDT) to kill the cancer cells.^19–22 Despite these advantages, arsenic–manganese complex exhibits limited circulation half-life and aggregation tendency, which can tremendously hinder its future application in the clinical setting.

The recent significant advances in nanotechnology have enabled more options in cancer diagnosis and treatment by leveraging the tumor microenvironment.^23–25 Among various drug carriers, liposome, which consists of an amphiphilic lipid bilayer, is of significant importance due to its unique flexible structure and excellent biocompatibility. As such, liposomes have been used for targeted delivery of both hydrophilic and hydrophobic drugs.^26,27 In particular, hybrid lipid nanoparticle with core-shell nanostructures, which consists of a functional inter-core and an outer lipid bilayer, has demonstrated numerous desirable features.^28–31 Within this core-shell structure, the lipid layer can enhance the colloidal stability of the inter-core for prolonged circulation time and provide a barrier to prevent the cargo from diffusion. Thus, such a core-shell structure can achieve improved drug-loading yields and controlled release profiles, which can ultimately enhance the therapeutic efficacy of the treatment.³² Several liposomes have been approved by FDA to utilize as drug carriers, such as liposomal doxorubicin (Doxil). However, the way to directly load ATO into liposomes was limited by low encapsulation efficiency and rapid leakage of ATO, thereby causing systemic toxicity.³³ Recently, metal-arsenic complex has been raised due to its high loading efficiency of ATO, but suffering from its low solubility in water. Previous attempts have shown that actively load ATO into liposome by utilizing metal ions to form a complex in the liposome was a successful strategy to improve the loading efficiency of ATO, enhance the stability of metal-arsenic complex, and prolong the in vivo circulation time in the meantime.³⁴

Herein, liposomes were used to encapsulate the arsenic–manganese complex in this work. Based on the results, the liposome-encapsulated arsenic–manganese complex was able to exhibit significantly improved stability and distribution of arsenic in the tumor. The as-prepared liposomal nanoparticle-containing arsenic–manganese complex containing (LP@MnAs_x) was able to demonstrate robust stability and sustained-release profile under physiological environments. LP@MnAs_x was able to degrade with high sensitivity towards pH variation in the tumor microenvironment, and this led to a burst release profile of Mn²⁺ and arsenite for synergistic chemodynamic therapy and chemotherapy.¹⁸ Thus, this study on inorganic arsenic-based antitumor drugs may provide a novel approach for the treatment of HCC.

Materials and Methods

Materials

Sodium arsenite (97%) was purchased from West Asia Chemical Industry Co., Ltd. (Shanghai, China). The standard arsenic solution and standard manganese solution were purchased from Beijing Century Audiocodes Biological Technology Co., Ltd. (Beijing, China). Manganese acetate (98%) was bought from Alfa Aesar (Massachusetts, England). Hydrogenated soybean phospholipids (HSPC, 98%), cholesterol (Chol, 97%) and cholesterol-polyethylene glycol 2000 (Chol-PEG2000, 98%) were bought from Shanghai Dongsheng Biological Technology Co., Ltd. (Shanghai, China). 3,3ʹ,5,5ʹ-tetramethylbenzidine (TMB) was purchased from Solarbio Co., Ltd (Beijing, China). MTT, penicillin and streptomycin were bought from Gibco (New York, USA). PI kit and apoptosis kit were purchased from BD Biosciences (New Jersey, USA). Sodium azide (NaN₃, 1%) was bought from Amresco (Washington, USA). TUNEL Assay Kit and Ki67 Cell Proliferation Kit were bought from Laval (Quebec, Canada). Nissl dye blue skies was purchased from Shanghai biological technology Co., LTD. (Shanghai, China). LO2 and HepG2 cells were obtained from Cell Bank of Chinese Academy of Sciences (Shanghai, China). Male BALB/c nude mice (6–8 weeks) and male SD rats (weighting 200 ± 15 g) were obtained from Xiamen Medical College. All animal experiments were performed under the protocols approved by the Administrative Committee of Laboratory Animals of Xiamen Medical College (Approval number: 2019008). All animal tests were approved by the Ethics Committee of Xiamen Medical College.

Preparation of LP@MnAs_x

100 mg HSPC, 10 mg Chol, and 5 mg Chol-PEG2000 were dissolved in 10 mL ethanol:chloroform solution (1:1, v/v) in a round-bottom flask. The lipid film was prepared by evaporating the ethanol and chloroform, and then it was hydrated in 30 mg/mL manganese acetate solution (10 mL, pH 7.3) under sonication in a water bath using a probe-type sonicator for 15 min and 10 min, respectively. After which, the liposomes were then extruded through polycarbonate membranes, and then it was ultrafiltrated to remove the excess manganese acetate to obtain LP@Mn. LP@As was prepared using similar procedures, except that sodium arsenite solution (NaAsO₂, 2 mg/mL, 10 mL) was used instead of manganese acetate solution. To prepare LP@MnAs_x, NaAsO₂ was firstly dissolved in deionized water, and then it was added to LP@Mn. After which, the mixture was reacted at 55°C for 30 min, and then it was poured into an ultrafiltration system. The collected concentrated solution was denoted as LP@MnAs_x.

Preparation of Arsenic–Manganese Complex (AMO)

Chloroform was added to the as-obtained LP@MnAs_x, and then it was sonicated with using a probe-type sonicator. The upper layer of water evaporated the chloroform, and then the deionized water was ultrasonically dissolved. Subsequently, it was repeatedly washed 10 times. The liquid was then ultrafiltered, centrifuged, and the upper solid was collected and freeze-dried to obtain AMO.

Characterization

The hydrodynamic diameter and zeta potential of liposomes were characterized via dynamic light scattering (DLS). The morphologies of the liposomes were observed under transmission electron microscopy (TEM). Inductively coupled plasma spectrometry (ICP, 6300, Thermo Electron Corporation, Waltham, MA, USA) was used to measure the entrapment efficiency (EE%) and drug loading (DL%) of the as-prepared liposomes.

Fenton-Like Activity of LP@MnAs_x

TMB was used to monitor the generated ∙OH due to Mn²⁺ of LP@MnAs in the presence of H₂O₂. Briefly, 800 µL LP@MnAs_x solution of different pH values were mixed with 100 µL H₂O₂ (10 mM) and 100 µL TMB (1 mM). After 10 min, a microplate reader was used to determine the absorbance between 400 to 800 nm.

In vitro Release Evaluation

The drug release characteristics of four formulations were investigated by dialysis bag method. The release medium was a phosphate buffer with pH values of 7.4, 6.8, and 5.5. The in vitro release rates of LP@As, LP@MnAs_x, AMO, and sodium arsenite aqueous solution (NaAsO₂-Sol) were investigated. The released As content was tested using ICP according to formula (1), and the release curve was obtained.

(1)

where M_n represents the cumulative release at time n, and M₁ is the content in the sample.

Cellular Uptake Study

HepG2 cells were seeded in 6-well plates with a density of 5×10⁵ cells/well, and it was allowed to incubate. The culture medium was replaced with NaAsO₂-Sol, LP@As, and LP@MnAs_x (all containing about 5 μg/L As) after 24 h of incubation. ICP was used to determine the intracellular drug concentrations after 2, 4, 6, 8, 12, and 24 h. Each experiment was performed in triplicate.

HepG2 cells were seeded in 6-well plates with a density of 1×10⁶ cells/well, and then it was incubated for 24 h. It was pretreated with 5 μg/mL sodium azide (NaN₃) without serum for 30 min, suctioned, and washed twice with PBS. The other experimental methods were similar to those used for the cytodynamics research.

In vitro Cytotoxicity

Firstly, approximately 8×10³ HepG2 and LO2 cells were seeded into a 96-well plate. After 24 h of incubation, the medium was replaced with NaAsO₂-Sol, LP@As, and LP@MnAs_x contained fresh medium with different concentrations. After 24 h of incubation, the medium was replaced with 180 µL fresh medium and 20 µL MTT (0.5 mg/mL). After another 4 h of incubation, 150 µL DMSO was added and incubated for 10 min. The absorbance was subsequently determined by measuring the 490 nm absorption using a microplate reader (Synergy TM2, BIOTEK, Winooski, VT, USA). The cell viability was calculated according to equation (2).

(2)

Intracellular Reactive Oxygen Species (ROS) Detection

The cell-permeable dye DCFH-DA was used to detect ROS production, whereby the ROS could oxidize the non-fluorescent dye to green fluorescent of DCF. In a typical setup, 2×10⁵ HepG2 cells were seeded and incubated overnight. Subsequently, it was treated with LP@MnAs_x for 6 h. Then, the culture medium was then replaced with DCFH-DA (10 µM), and then incubated at 37°C for 30 min. Finally, the sample was observed under a confocal laser scanning microscope (CLSM, Olympus, Tokyo, Japan). And the quantitative analysis of DCF with different treatments was assessed via flow cytometry (Guava Easycyte, Merck Millipore, Darmstadt, Germany).

Cell Apoptosis Study

HepG2 cells were seeded with a density of 5×10⁵ cells/well. After 24 h of incubation, each plate was treated with the serum-free medium of NaAsO₂-Sol, LP@As, and LP@MnAs_x. Then, the cells were collected with Annexin V binding buffer for the Annexin V-FITC/PI staining and they were assessed via flow cytometry.

Cell Cycle Study

HepG2 cells were seeded with a density of 5×10⁵ cells/well, and then it was cultured with medium of NaAsO₂-Sol, LP@As and LP@MnAs_x for 24 h. After which, the cells were collected and fixed with 70% ethanol for 12 h. Subsequently, it was stained with PI solution and assessed via flow cytometry.

Pharmacokinetic Study

Sprague Dawley (SD) rats were randomly divided into three groups (n = 5), and they were intravenously injected with NaAsO₂-Sol, LP@As, and LP@MnAs_x (As: 0.6 mg/kg). At predetermined time intervals, blood samples were obtained from the orbital vein of the SD rats. The As content in the plasma was tested using inductively coupled plasma mass spectrometer (ICP-MS), and pharmacokinetic parameters were analyzed using the PK Solver pharmacokinetic software.

Biodistribution

A total of 5×10⁶ HepG2 cells were injected into the back of nude mice. When the tumor volume increased to 50–100 mm³, 15 tumor-bearing nude mice were randomly divided into three groups, ie, NaAsO₂-Sol group, LP@As group, and LP@MnAs_x group (n = 5). The mice in each group were sacrificed at 24 h after a single injection of drugs. The heart, left lobe of the liver, spleen, left lung, left kidney, brain, abdominal skin, left leg bone, and tumor tissues were collected. ICP was used to determine the drug concentration. The tissue distribution map was plotted with the percentage of tissue per gram of injection volume (% ID/g) as the ordinate.

In vitro/vivo T1-Weighted MRI Imaging Evaluation

LP@MnAs_x with various Mn concentrations, ie, 0, 0.05, 0.1, 0.2, 0.4, and 0.5 mM, were dispersed in buffer solutions with various pH values for 24 h. The MRI images were then obtained via 1.5-T MRI system.

For in vivo MRI imaging, 0.5 mL LP@MnAs_x was intravenously administrated at Mn concentration of 3 mg/kg. Subsequently, the MRI images were recorded at 1, 2, 4, 8, and 12 h post-injection using the 1.5-T unit MRI system.

Pharmacodynamics Evaluation

Twenty tumor-bearing nude mice were randomly divided into four groups, ie, saline, NaAsO₂-Sol, LP@As, and LP@MnAs_x groups (n = 5). Normal saline or medicine (approximately 0.6 As mg/kg) was injected intravenously every day. The tumor volume and body weight of nude mice were recorded every other day. The tumor volume was calculated using Formula (3). On the 15th day, the tumors were collected to weigh.

(3)

Pathological Analysis

To further prove the ROS generation in vivo, 12 tumor-bearing nude mice were intravenously injected with saline, NaAsO₂-Sol, LP@As, and LP@MnAs_x, respectively (n = 3). At 24 h post injection, the tumor tissues were collected and co-stained with DCFH-DA and DAPI. The generation of ROS within the tumor cells was observed via CLSM.

On Day 15, the nude mice were sacrificed, and the heart, liver, spleen, lung, kidney, hippocampus, and striatum tissue were collected and fixed with 10% formalin to prepare the histological sections. These tissues were then stained with H&E, and histological images were obtained using an optical microscope (AXIO SCOPE.A1, Zeiss, Jena, Germany).

The tumor tissues of nude mice from each group were subjected to TUNEL and Ki67 staining, and the tissue morphology was observed under a microscope.

The brain tissue of nude mice was fixed with 10% paraformaldehyde, stained with Nissl, and then observed under a microscope and imaged.

Statistical Analysis

Data were presented as mean or mean ± standard deviation (SD). Statistical significance was evaluated by SPSS 22. P < 0.05 indicated a statistical difference, while P < 0.01 was indicated as a significant difference.

Results and Discussion

Characterization of LP@MnAs_x

As depicted in Figure 1, the LP@MnAs_x was designed to entrap arsenic by interacting with Mn²⁺ to realize an MRI-guided synergistic therapy of chemodynamic therapy and chemotherapy. The LP@MnAs_x was prepared via a two-step method: (1) preparation of manganese acetate-containing liposome using the filming-rehydration method and (2) the formation of the arsenic–manganese complex within the interior hydrophilic space drived by ion gradient. The interactions among the arsenic–manganese complex, including the electrostatic interaction and dipole–dipole interaction, enable the sustained release of the entrapped cargo, and the liposome improves the stability of the as-prepared nanoparticles and it can also facilitate their accumulation at the tumor site for enhanced therapeutic efficacy.¹⁶

Figure 1 Schematic illustration of the synthesis of LP@MnAsx and the subsequent cellular uptake and drug release.

As shown in the TEM image (Figure 2A), LP@As, LP@Mn, and LP@MnAs_x possessed spherical morphology and they were uniformly dispersed with average diameters of around 120 nm. Such a result is consistent with the results obtained by DLS (Figure 2B). When the liposome was removed from LP@MnAs_x by subjecting it to chloroform and ultrasonication, the interior core of the nanoparticle was observed by TEM, suggesting the successful formation of AMO within the liposome. According to the ICP results, the As drug-loading rate and encapsulation efficiency in the LP@MnAs_x were remarkably high at 4.68 ± 0.82% and 92.96 ± 5.35%, respectively. Furthermore, LP@MnAs_x showed excellent colloidal stability over 30 days with negligible changes in the hydrodynamic diameter and zeta potential (Figure 2C), further illustrating the excellent stability of LP@MnAs_x.

Figure 2 Characterization of LP@MnAsx. Notes: (A) TEM images and (B) hydrodynamic diameter, PDI, and Zeta potential of LP@Mn, LP@As, LP@MnAsx, and AMO. (C) Hydrodynamic diameter and Zeta potential stability of LP@MnAsx. (D) UV-Vis spectra and the color change of aqueous TMB of LP@MnAsx at varying pH values. Cumulative releases of NaAsO2-Sol, LP@As, LP@MnAsx and AMO at pH 7.4 (E), 6.8 (F), and 5.5 (G).

Fenton-Like Activity of LP@MnAs_x

To investigate the Fenton-like activity of Mn²⁺, TMB was selected as a signal for ∙OH. As shown in Figure 2D, the absorbance of TMB in LP@Mn and LP@MnAs_x reaction system increased significantly as compared to that in LP@As. Such a result indicates the generation of ∙OH by the reaction between Mn²⁺ and H₂O₂. Especially, a greater extent increase of TMB was induced of LP@Mn and LP@MnAs_x under pH 5.5 environment, which demonstrated that the Fenton-like reaction was more evident with the increasing acidity of PBS, displaying the pH-dependent Fenton-like reaction activity. Also, the same results could be clearly observed according to the corresponding digital images of the insert photograph in Figure 2D.

In vitro Release Evaluation

The release profiles of the four formulations, ie, NaAsO₂-Sol, LP@As, LP@MnAs_x, and AMO, were investigated under various pH values, ie, 7.4, 6.8, or 5.5, based on a dialysis method. As shown in Figure 2E-G, more than 95% of As was released from NaAsO₂-Sol within 1 h under all pH conditions. Notably, after liposome-coating, there was a slow release of As without pH responsiveness, whereby approximately 80% As was released from LP@As after 24 h under all pH conditions. However, the LP@MnAs_x demonstrated a sustained-release profile under physiological conditions (pH = 7.4) with a cumulative release of 22.69 ± 0.76% within 24 h. Such a sustained-release profile could be attributed to the synergistic effects of the interactions among the arsenic–manganese complex and the diffusional barrier provided by the liposome (Figure 2E). Upon incubation in an acidic environment, LP@MnAs_x exhibited an obvious pH-dependent drug release behavior, whereby cumulative release amounts of 33.47 ± 3.43% and 63.15 ± 1.90% of As were attained at pH of 6.8 and 5.5, respectively (Figure 2F-G). The biodegradation of arsenic–manganese complex could be triggered by the acidic environment of the tumor tissue,¹⁷ which could be attributed to the dissociation of the AMO triggered by decreased pH, and the coordination between manganese and arsenic would be weakened in the meantime.

Cellular Uptake Study

HepG2 cells were incubated with NaAsO₂, LP@As, or LP@MnAs_x with or without NaN₃ for 24 h. After which, the intracellular As concentration was determined at predetermined time intervals to investigate the internalization dynamics and pathway of various formulations. NaN₃ was used as an endocytosis inhibitor to evaluate the depletion of cellular ATP.³⁵ As shown in Figure 3A-C, NaAsO₂ demonstrated a rapid fast entry into HepG2 cells, which was free from the influence of NaN₃ (P > 0.05). LP@As and LP@MnAs_x showed a time-dependent cellular uptake, and obviously enhanced the cellular uptake of As (P < 0.05). While with the addition of NaN₃, the cellular uptake of LP@As and LP@MnAs_x was significantly reduced that by more than a half. This result indicates the energy-consuming endocytosis-mediated internalization pathway.^36,37 In conclusion, the cellular uptake of LP@As and LP@MnAs_x would be significantly inhibited by the addition of NaN₃, suggesting that their endocytosis way was ATP-dependent (P < 0.05).

Figure 3 The cellular uptake, cytotoxicity, CDT activity, cell apoptosis and cell cycle of LP@MnAsx. Notes: As concentration in HepG2 cells after treatment with NaAsO2, LP@As, or LP@MnAsx in the absence (A) and presence (B) of NaN3 pre-treatment. (C) As concentration after treatment with NaAsO2-Sol, LP@As, or LP@MnAsx for 24 h for the normal group and NaN3 pre-treated groups. Cell viability of LO2 (D) and HepG2 (E) cells and IC50 value of LO2 and HepG2 (F) after treated with NaAsO2-Sol, LP@As, or LP@MnAsx. (G) ROS detection of HepG2 cells after incubating with LP@Mn, LP@As or LP@MnAsx. (H) The quantitative analysis by flow cytometry about DCF after different treatments. Cell apoptosis (I, J) and cell cycle (K, L) of HepG2 cells treated with blank medium, NaAsO2-Sol, LP@As, or LP@MnAsx. Data are expressed as means ± SD (n = 3). *P < 0.05; **P < 0.01. Abbreviations: CDT, chemodynamic therapy.

In vitro Cytotoxicity

The cytotoxicity of these formulations on HepG2 and normal liver cells (LO2) cells was assessed via MTT method. As shown in Figure 3D-E, the viabilities of HepG2 and LO2 cells decreased in a concentration-dependent manner. Based on the IC50 value (Figure 3F), HepG2 cells were more sensitive to each of the formulation as compared to LO2 cells. Thus, the damage to LO2 could be avoided, and the as-prepared formulations could facilitate the subsequent treatment against tumors. Among all samples, LP@MnAs_x exhibited the best tumor-inhibiting effect on HepG2 cells with an IC50 of 1.37 ± 0.47 μg/mL, which was remarkably lower than that of LP@As (3.15 ± 2.29 μg/mL) and NaAsO₂-Sol (3.83 ± 0.85 μg/mL), indicating the synergistic therapy of chemotherapy by As and CDT by Mn²⁺.

Intracellular ROS Detection

DCFH-DA probes were used to detect intracellular ROS production. As shown in Figure 3G, the green fluorescence of DCF was observed for the groups with LP@Mn and LP@MnAs_x due to the existence of Mn²⁺ when compared to the control group or LP@As. Moreover, after incubating with LP@MnAs_x, HepG2 cells exhibited morphological alteration, which further illustrated the cytotoxicity of As towards tumor cells. And the results of quantitative analysis via flow cytometry was in line with CLSM (Figure 3H).

Cell Apoptosis and Cell Cycle Study

The results of the HepG2 apoptotic rates after being treated with each formulation are shown in Figure 3I-J. Compared with the control group, each treatment exerted significant effects on the apoptosis rate (P < 0.01). The group treated with LP@MnAs_x exhibited the highest apoptotic rate of 42.49 ± 3.36%, which could be attributed to the combined effect of the toxic hydroxyl radicals generated by Mn²⁺ and the apoptosis-inducing effect mediated by ATO. According to the cell cycle analysis result (Figure 3K-L), the Control, NaAsO₂-Sol, LP@As, and LP@MnAs_x groups arrested 19.60 ± 1.58%, 46.43 ± 0.94%, 47.73 ± 1.31%, and 69.00 ± 0.90% of HepG2 cells at the 4N phase, respectively. Such results obtained from these groups were significantly different from that obtained from the blank group (P < 0.01). Thus, this can reasonably explain the achievement of apoptosis mainly by shortening the 2N phase and prolonging the 4N phase.

Pharmacokinetic Study

A plasma pharmacokinetic study was performed to investigate the pharmacokinetic behavior of various formulations. NaAsO₂-Sol, LP@As, and LP@MnAs_x were intravenously administered at the As dosage of 0.6 mg/kg, and the blood was collected at pre-determined time intervals for ICP-MS analyses. The As concentration–time curve and main parameters are shown in Figure 4A and Table 1, respectively. It could be observed that all the as-prepared formulations conformed to the two-compartment model. The NaAsO₂-Sol was rapidly removed from the body with a short half circulation time of 14.82 ± 2.34 h, and this was shorter than those of LP@As (16.72 ± 6.59 h) and LP@MnAs_x (19.54 ± 4.54 h). When compared to the NaAsO₂-Sol, the area under curve (AUC) of LP@As and LP@MnAs_x was increased by 1.69 and 2.06 times, and the mean residence time (MRT) increased by 23.4% and 49.0%, respectively. Such a result indicates that the liposome encapsulation could improve the plasma circulation behavior. When compared to LP@As, LP@MnAs_x further improved the t_1/2β and AUC_0-t, and it decreased the MRT, which could be attributed to the formation of AMO for the sustained drug release. Thus, based on the collective results, LP@MnAs_x displayed superior pharmacokinetic behavior of prolonged circulation time and decreased elimination from the body, benefiting from the liposome-coating.

Table 1 Main Parameters of the SD Rats After Injected with of NaAsO2, LP@as, or LP@MnAsx Intravenously (n = 5)

Figure 4 In vivo distribution and T1-weighted MR imaging of tumor-bearing nude mice. Notes: (A) The concentration-time curve and (B) tissue distribution of As after intravenous injection of NaAsO2-Sol, LP@As, or LP@MnAsx (n = 5). *P < 0.05. (C) In vitro T1-weighted MR imaging and (D) T1 relaxation rates of LP@MnAsx at various concentrations after incubating at different pH conditions. (E) In vivo T1-weighted MR imaging of tumor-bearing nude mice after being injected with LP@MnAsx intravenously. The tumor was marked by red dotted circles.

Biodistribution

To evaluate the biodistribution of various formulations, HepG2 xenograft tumor model was constructed. NaAsO₂-Sol, LP@As, and LP@MnAs_x were intravenously administered, and the tissues of the heart, liver, spleen, lung, kidney, brain, abdominal skin, bone, and tumor were harvested for ICP-MS examination at 24 h post-administration. As shown in Figure 4B, NaAsO₂-Sol was mainly distributed in the liver, spleen, lung, and skin, with only 0.54 ± 0.12%ID/g at the tumor site. After being encapsulated by liposomes, the liver, spleen, lung, and skin showed significantly reduced arsenic distribution, while there was increased arsenic distribution in the tumor, which indicated an improved arsenic distribution in various tissues. The amount LP@MnAs_x distribution could be arranged in the following decreasing order: liver > spleen > tumor > lung > heart > skin > bone > brain. Meanwhile, when compared to other groups, the LP@MnAs_x exhibited the highest accumulation at the tumor site and a decreased distribution in other major organs, which indicated superior tumor targeting ability and improved biodistribution. And the distribution in liver, spleen and lung was due to the reticuloendothelial system (RES) effect.

In vitro/vivo T1-Weighted MRI Imaging Evaluation

Due to the dissociation of LP@MnAs_x triggered by an acidic environment, the released Mn²⁺ can serve as a contrast agent for T1-weighted MRI contrast agent. With the increased concentration, the T1-weighted MR images of LP@MnAs_x at pH 5.0 revealed a greater contrasting effect as compared to those at pH 7.4 (Figure 4C). Furthermore, the r1 relaxivity of LP@MnAs_x significantly increased from 2.55 mM⁻¹s⁻¹ due to the accelerated release of Mn²⁺ (Figure 4D). Subsequently, T1-weighted MR imaging was monitored at selected times after being intravenously injected with LP@MnAs_x. An obvious high contrasting effect could be observed at the tumor site, whereby it continuously increased and peaked at 2 h post administration, and then gradually decreased. This result suggests that LP@MnAs_x was able to respond sensitively to the tumor microenvironment (Figure 4E).

Pharmacodynamics Evaluation

To assess the in vivo therapeutic efficacy, 20 HepG2 tumor-bearing mice were randomly divided into four groups, ie, treated with saline, NaAsO₂-Sol, LP@As, and LP@MnAs_x. During the 14 days of treatment, tumor volume and body weight of mice were recorded each day. Based on the results, the average tumor volumes of the groups treated with saline, NaAsO₂-Sol, LP@As, and LP@MnAs_x were 1192.34 ± 482.21 mm³, 648.53 ± 198.62 mm³, 499.67 ± 116.72 mm³, and 316.20 ± 110.45 mm³, respectively, showing that NaAsO₂-Sol, LP@As and LP@MnAs_x have a difference degrees in the inhibition of the growth of tumor (Figure 5A-C). Compared with the group of NaAsO₂-Sol, LP@As and LP@MnAs_x could significantly inhibit the tumor, which might be owing to the tumor-targeting ability of liposome and the synergistic therapy of chemotherapy and CDT (P < 0.01). In addition, the body weight of the mice treated with NaAsO₂-Sol group was significantly lower than that of the saline group, indicating systemic cytotoxicity caused by the non-specific distribution of NaAsO₂. In contrast, the body weight of the mice treated with LP@As and LP@MnAs_x kept increased continuously, which suggests the attenuation of adverse side effects, showing the good biosafety of LP@As and LP@MnAs_x (Figure 5D).

Figure 5 In vivo anti-tumor evaluation. Notes: Tumor volume (A), tumor weight (B), digital photos of tumors (C) and body weight (D) of tumor-bearing nude mice treated with saline, NaAsO2-Sol, LP@As, or LP@MnAsx (n = 5). (E) ROS generation in tumor tissues collected from different groups at 24 h post injection and co-stained with DCFH-DA (green) and DAPI (blue). (F) Tunel and Ki67 staining of the tumor, and Nissl images of the hippocampus after saline, NaAsO2-Sol, LP@As, or LP@MnAsx treatment. (G) IOD values of Tunel and Ki67, and the number of Nissl bodies (n = 3). *P < 0.05; **P < 0.01.

Pathological Analysis

To further display the in vivo ROS generation, the immuno-histochemical analysis of tumor tissues were investigated by using DCFH-DA as the ROS probe. As shown in Figure 5E, the obvious green fluorescence could be seen in the tumor tissues after LP@MnAs_x treatment, suggesting a large amount of ROS generation in vivo.

Ki67 and TUNEL staining were used to perform the histological evaluation and statistical analysis so as to determine the tumor cell apoptosis caused by treatment, whereby saline group was used as the control (Figure 5F-G). When compared to other groups, the group treated with LP@MnAs_x resulted in the largest area of apoptotic cells, and therefore it achieved the best proliferation inhibiting effect. When considering the side effects of nerve cytotoxicity caused by arsenic treatment, Nissl staining was used to observe the damage inflicted onto the nerve cells. Based on the results, the number of Nissl bodies in the hippocampus of the LP@As and LP@MnAs_x groups were significantly less than that in NaAsO₂-Sol groups (P < 0.05). Furthermore, morphological alterations could be detected, which were indicative of severe nerve cytotoxicity. Meanwhile, LP@MnAs_x showed attenuated damage to the nerve cells with a comparable number of Nissl bodies (P < 0.01), and this result is consistent with the biodistribution results. According to the H&E staining result, administrating of NaAsO₂-Sol, LP@As, and LP@MnAs_x to the mice did not lead to obvious damage to the major organs relative to the control group (Figure 6).

Figure 6 H&E staining image of major organs collected from nude mice after treated with NaAsO2-Sol, LP@As, and LP@MnAsx.

Conclusion

In summary, the LP@MnAs_x was successfully developed for the carcinoma-targeted delivery of the arsenic–manganese complex via magnetic resonance imaging-guided synergistic therapy of chemodynamic therapy and chemotherapy. The enhanced therapeutic efficacy exhibited by the LP@MnAs_x was realized by the pH-sensitive drug release, prolonged circulation time, and improved tumor accumulation. Hence, based on the collective results, the use of the LP@MnAs_x is a promising strategy for carcinoma treatment, overcoming the shortage of ATO, and propelling the clinical translation process of arsenic trioxide based formulation.