Lipo-MGN nanoparticle hypoxia attenuation-mediated single-dose radiotherapy- and pH/ROS-responsive T1 contrast magnetic resonance imaging in hepatocellular carcinoma

Abstract

Tumor hypoxia is an important factor for developing resistance to radiation therapy (RT) and presents a bleak prognosis in cancer patients undergoing treatment for RT resistant hepatocellular carcinoma. Here, we present the synthesis of liposome-coated Mn₃O₄ (MGN) nanoparticles (Lipo-MGN) and investigation of their therapeutic potential with RT utilizing a HepG2 cancer model. According to in vitro research, Lipo-MGN effectively produced oxygen in the presence of H₂O₂ and significantly reduced the expression of HIF-1 in human HepG2 cells that were under hypoxic conditions. Lipo-MGN reversed the radio-resistance brought on by hypoxia and increased cell damage. When Lipo-MGN and RT were administered together in a HepG2 xenograft mice model, the tumor growth was delayed more than with RT alone. As determined by MR imaging, liposome-MGN also exhibited T1 contrast enhancement in tumor. According to these findings, Lipo-MGNs may increase the impact of RT by focusing tumor hypoxia. Hypoxic, radioresistant HepG2 cancer may be treated with Lipo-MGN in clinical studies.

Introduction

Hepatocellular carcinoma (HCC) is the fifth most prevalent type of malignant tumor in the world. Approximately 700,000 people are diagnosed annually with liver cancer in the United States alone (Xu et al. 2021). The curative treatment for HCC has been liver transplantation, surgical resection and radiofrequency ablation (Park et al. 2020; Sayan et al. 2019). The radiation generated from radioactive source or X-ray, Gamma ray generator have high energy and short wavelength which when directed to region affected by tumor will result in direct damage or nucleic acid degeneration leading to tumor shrinkage. The disadvantage of radiation therapy is reduction in efficiency due to hypoxia in HCC because tumor nodules consume a high amount of oxygen. Therefore, HCC is considered to be one of the most hypoxic tumor (Elbanna et al. 2021; Chen and Lou 2017). To overcome this issue therapeutic agents can be introduced to convert hypoxic microenvironment to normoxic microenvironment such as manganese oxide nanoparticle which has the ability to decompose in presence of reactive oxygen species and glutathione to produce oxygen (Yang et al. 2017; Scheme 1).

Scheme 1

Intravenous injection of Lipo-MGNs reduced hypoxia in tumours and promoted cell death in radiation therapy

Hypoxia can promote tumor progression and induce RT resistance. HCC is considered to be a hypoxic tumor (Chen and Lou 2017). The direct effect of hypoxia can be seen in the gene expression level of hypoxia inducible factor-1 (HIF-1) which will increase with hypoxia level and decrease with oxygen level due to hydroxylation. Angiogenesis, proliferation, invasion, metabolism, metastasis, and resistance to therapy are only a few of the ways that HIF-1 genes contribute to the development and spread of tumors and cancer. Radiation resistance due to hypoxia is a major issue facing the clinical outcome of RT (Sorensen and Horsman 2020).

MGN is the type most commonly modified for the generation of oxygen (Prasad et al. 2014). MnO₂ nanoparticles have been loaded in hyaluronic acid (HA)-based nanoparticles to target tumor-associated macrophages (TAMs) and reduce hypoxia in tumors to improve chemoresistance caused by hypoxia. Additionally, HA-MnO₂ nanoparticles showed T1 and T2 contrast ability in tumor MR imaging (Song et al. 2016). Albumin MnO₂ nanoparticles were able to reduce hypoxia and downregulate HIF1-alpha. Breast tumor growth was greatly reduced when albumin MnO₂ nanoparticles and radiation therapy were used in combination (Prasad et al. 2014). Fucoidan-coated MnO₂ nanoparticles Fuco-MnO₂-NPs effectively produced oxygen in the presence of H₂O₂ and significantly reduced HIF-1 expression under hypoxic conditions in human pancreatic cancer cells (Shin et al. 2018).

MGN has the advantages that it has over other manganese oxide types, such as MnO, MnO₂ and Mn₂O₃, for generating more molecular oxygen. Mn₃O₄ has a higher ROS-scavenging ability owing to the double oxidation states of Mn³⁺ and Mn²⁺, which can lead to the conversion of ROS (O₂^.– and ˙OH) to molecular oxygen released in inflammation (Choi et al. 2022; Yao et al. 2018). There are a few reports about Mn₃O₄ nanoparticles (MGNs) for cancer therapy. Doxorubicin-loaded, serotonin–stearic acid-based Mn₃O₄ nanocuboids (Jain et al. 2020) for HCC therapy and Au@Mn₃O₄ magneto-plasmonic nanoflowers (Ijaz Dar et al. 2020) for photothermal cancer therapy have been reported.

PEG (polyethylene glycol) is hydrophilic and has also been shown to protect nanoparticles from phagocytosis by the reticuloendothelial system in the liver so that blood circulation increases, mediating more bioavailability to the tumor region. We used PEGylated liposomes to load MGN to increase the solubility of hydrophobic Mn₃O₄ and increase bioavailability. To our knowledge, no other work has reported PEGylated liposome-loaded Mn₃O₄. Here, we synthesized PEGylated liposome-loaded Mn₃O₄ to attenuate hypoxia in HepG2 cancer cells for radiosensitization and contrast-enhanced TI-weighted MR imaging.

Materials and methods

Synthesis of manganese oxide nanoparticles and Lipo-MGNs

1-Dodecanethiol-stabilized hydrophobic Mn₃O₄ nanoparticles were prepared as previously described (Choi et al. 2022). The film hydration process was used to make liposomes. Briefly, 1.5:1.5:1 (weight ratio) of DSPE-PEG, DPPC, and cholesterol were added to the appropriate solvents (methanol for DSPE-PEG, DPPC, and cholesterol) and combined in a vortex for 2 min in a glass vial. Thin film hydration was formed in a vacuum chamber at room temperature by evaporation of the aforementioned solvents. Then, the film was hydrated using 1 ml of distilled water and heated in a glass vial for 30 min at 60 °C to promote the formation of multivesicular liposomes, which are heterogeneous in nature. To make Mn₃O₄-loaded liposomes (Lipo-MGN), we prepared 1.5 times (1.5×) the concentration of lipids and added 1.5 mg of Mn₃O₄ during thin film formation. Finally, the Lipo-MGNs were probe-sonicated for 10 min. Unloaded Mn₃O₄ was removed by dialysis in 4 °C for 2 h.

Characterization of the Lipo-MGNs

Using a Zetasizer instrument, the size and Zeta potential of Lipo-MGN were examined to calculate the surface charge (Nano-Z590, Malvern Instruments, Worcestershire, UK). Using field emission transmission electron microscopy, the size and morphology were validated (FE-TEM). For liposome drug loading capabilities, Mn₃O₄ encapsulation and loading were studied using inductively coupled atomic emission spectroscopy (ICP-AES) and thermogravimetric analysis (TGA).

Cell viability of Lipo-MGNs

The MTS assay was used to determine the cell viability of Lipo-MGNs and MGNs in the NIH3T3 cell line. 96-well plates were seeded with a total of 10⁴ cells per well and cultured for one day at 37 °C in a CO₂ incubator. Lipo-MGN effect in HepG2 after radiation (10 Gy) was studied by MTS assay. HepG2 cells from the human liver cancer cell line were seeded in 35 mm petri dish from Lab-Tek2 (New York, USA) at a density of 10⁵ cells and incubated for one day at 37 °C in a humidified CO₂ atmosphere. The Lipo-MGNs and MGNs viability was examined in triplicate across the concentration range of 100–0.78 µg/ml. After the cells had been treated for 24 h, they were incubated for another 24 h in fresh media before 20 µl of the MTS reagent was added and incubated for 3–4 h in each well. In case of Lipo-MGN treatment in HepG2 cell line, cells were irradiate with 10 Gy of radiation before MTS reagent was added. With the aid of a microplate reader, the absorbance was calculated.

Cell uptake of liposomes by human liver cancer cell line|

HepG2 cells from the human liver cancer cell line were seeded in 8-well chamber slides from Lab-Tek2 (New York, USA) at a density of 5 × 10⁴ and incubated for one day at 37 °C in a humidified CO₂ atmosphere. HepG2 cells were cultured in Dulbecco's modified Eagle's medium (DMEM), which contains 10 vol% FBS and 1 vol% penicillin–streptavidin–amphotericin B (Gibco Anti-Anti (100×), USA). Following a 24-h incubation period, 100 µg/ml of DiD dye, a lipophilic carbocyanine dye-loaded liposome, was added to each well. The cells were then incubated after receiving these treatments. The cells were incubated for 24 h before being rinsed with PBS to remove any excess dye and fixed with a 4% formaldehyde solution. Using a confocal microscope, the cells were captured (Zeiss LSM 510, Oberkochen, Germany).

In vitro hypoxia assay

We used Image-iT™ Green Hypoxia Reagent (ThermoFisher, Massachusetts, USA) to check the conversion from hypoxia to normoxia in the HepG2 cell line. When the oxygen concentration is decreased (below 5%) in live HepG2 cells, the fluorogenic compound in the hypoxic reagent is activated and sustains the fluorescence. We measured the hypoxia level based on the intensity of fluorescence.

HepG2 cells were cultured in 24-well plates in a humidified CO₂ chamber at 37 °C for 1 day. Dulbecco’s modified Eagle’s medium (DMEM) containing 10 vol% FBS and 1 vol% penicillin–streptavidin–amphotericin B (Gibco Anti-Anti (100×), USA) was used for HepG2 cells. After 24 h of incubation, the plate was placed in a hypoxia chamber (hypoxia chamber spec) for overnight incubation. Lipo-MGN was treated with 25–100 µg/ml [Mn] and incubated for another 6 h. Cells were washed with PBS, treated with hypoxia reagent at a 5 µM concentration and incubated at 5% CO₂ and 37 °C for 4 h. The cells were washed again with PBS and observed under a fluorescence microscope.

In vitro T1-weighted MR imaging of Lipo-MGNs

To evaluate the effects of pH and ROS on Lipo-MGN dissociation, we conducted T1-weighted MR imaging by 3 T clinical MRI instrument (Magnetom Tim Trio, Siemens Medical Solutions, Erlangen, Germany) at different pH and H₂O₂ values using Lipo-MGNs. Lipo-MGN at 0.72 mg Mn/mL was serially diluted to 0, 0.25, 0.5, 1, and 2 mM Mn concentrations, and T1-weighted MR imaging was conducted.

Animal tumor model

Male BALB/c nude mice (18–20 g, 5 weeks old) or female BALB/c nude mice (18–20 g, 5 weeks old) were purchased from Jungang Lab Animal Inc. in Korea for biodistribution research, in vivo tumor reduction investigations, and animal toxicity studies. The Chonnam National University Medical School Research Institutional Animal Care and Use Committee approved the use of animals in the study, and it was conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (CNU IACUC-H-2019-6). HepG2 tumors were induced in mice by subcutaneous injection of 106 HepG2 cells close to the hind limb. We utilized the formula L * B²/2 to determine the tumor's volume, where L stands for the tumor's length (major axis) and B for its width (minor axes).

In vivo biodistribution of Lipo-MGNs by ICP-MS

Lipo-MGN (in PBS, 200 ul) at a concentration of 0.15 mg [Mn₃O₄]/kg (7.5 mg/kg of Mn₃O₄) was injected into each mouse intravenously through the tail vein of HepG2 tumor-bearing mice after HepG2 tumors attained a size of around 5–7 mm (after around 15 days). At the 8- and 24-h time points, organs were extracted and added to 3 ml of aqua regia and mixed with 27 ml of distilled water to make a 10% aqua regia:water mix and submitted for ICP-MS. ICP-MS was performed in a Varian 820 ICP Mass Spectrometer (Varian Medical Systems, UK).

In vivo T1-weighted MR imaging

For in vivo MR imaging, male BALB/c nude mice (5–6 weeks old, 20 g) were used. Subcutaneous injections of HepG2 cells at a density of at least 1 × 10⁷ cells were made in DMEM-cultured cells at 37 °C and 5% CO₂. Mice were anesthetized using isoflurane (Forane soln. Abbot, USA). T1-weighted MR imaging (M7™ Compact MRI System (AspectImaging, Shoham, Israel) was performed as follows: TR/TE: 400/10 ms TI: 100 ms, echo train length: 10, flip angle: 150°, field of view: 4 cm, number of slices: 15, slice thickness: 1.5 mm, interslice gap: 0 mm, and matrix: 256 × 210. When the tumor size reached 100–150 mm³, MR imaging was obtained for Lipo-MGNs in PBS at pre-injection and 3, 6, 8, and 24 h post-injection.

Cytotoxicity analysis by radiation using FDA/PI assay

HepG2 cells (5 × 10⁴) were seeded in confocal petri dishes (Lab-Tek2, USA) and cultured in DMEM at 37 °C with 5% CO₂ in a humid atmosphere. After 24 h incubation, 100 µg/ml Lipo-MGN was added and incubated for 6 h. The wells were washed 3 times with warm PBS and then irradiated at 3 Gy and 10 Gy. After radiation, the Lipo-MGN sample was treated with 8 µl of FDA (4 μM) and PI (0.5 μM) dissolved in PBS for 5 min at room temperature. With Texas Red® dye filter for PI and a fluorescein bandpass filter for FDA, the wells were observed under a fluorescence microscope.

Oxygen generation of the Lipo-MGNs

The O₂ generated by Lipo-MGNs was measured using an O₂-ADPT Oxygen Adapter (Microelectrodes Inc., USA). Briefly, 15 ml of a solution of 1 mg/ml Lipo-MGN was prepared, and H₂O₂ was added to a final concentration of 1 M. The oxygen level was measured in mg/l dissolved oxygen in a solution of Lipo-MGN + H₂O_2. Measurements were taken every 5 min and measured for up to 60 min.

mRNA isolation, cDNA conversion, and quantitative real-time PCR HIF alpha PCR study

HepG2 cells were seeded in 6-well plates and cultured in DMEM in a 5% CO₂ incubator at 37 °C in a humidified environment. Then, Lipo-MGN was added at a concentration of 100 µg/ml and incubated for 6 h under normoxic and hypoxic conditions. Cells were washed thoroughly with ice-cold PBS before mRNA extraction. Before extracting the mRNA from cells, cells were lysed using cell lysis buffer at 0 °C to maintain mRNA integrity. Total RNA was isolated using an RNeasy kit (QIAGEN, Germany).RT-PCR was done by SYBR Green PCR master mix (Applied Biosystem, UK).

Immunohistochemical detection of tumor hypoxia

HepG2 tumor BALB/c nude mice (n = 3) were intravenously injected with Lipo-MGNs at a dose of 7.5 mg/kg. Hypoxia detection assay (Hypoxyprobe-1 plus kit, Hypoxyprobe Inc., USA) was carried out using the kit's specified methodology for ex vivo tissue staining of hypoxia. One day after injection of Lipo-MGN, pimonidazole hydrochloride was injected intraperitoneally. Pimonidazole hydrochloride will accumulate in tumors within 60 min. After 60 min, tumor tissue was collected and preserved for histological examination in 10% neutral buffered formalin solution (Sigma Aldrich, USA).

In vivo radiation treatment

HepG2 xenograft liver cancer model in BALB/c mice was created. Four groups of mice (n = 5 each group) were used. (a) The non-radiation control group (NC), (b) the radiation only (RO) group, (c) the radiation high-concentration Lipo-MGN (RH) group, and (d) the radiation low-concentration Lipo-MGN (RL) group. When the tumors reached volume of about 100 mm³, treatment started. The mice were held in a restrainer in a low-dose-rate research radiator (Gammacell, 40 exactor, Theratronics, ON, Canada). For RH and RL groups i.v. administration of Lipo-MGNs were given at 7.5 mg/kg [MGN] and 3.75 mg/kg [MGN], respectively. The tumors were irradiated locally with 6 Gy, 10 min for RO, RH and RL group. Vernier callipers were used to assess the tumor size over time in two dimensions, and the tumor volume was computed using the formula V = [(length) × (width)²]/2. The animals were killed at the conclusion of the trial, and the tumor masses were removed and weighed. TUNEL assay was used to determine the level of apoptosis in tumor tissue after radiation. After fixing tumor tissue after 4% PFA, the tissue was deparaffinized. Using the ApopTag^® Peroxidase in Situ Apoptosis Detection Kit (Millipore, Burlington, MA, USA), TUNEL staining was carried out.

Results

Lipo-MGN characterization

Lipo-MGN was successfully synthesized. A final hydrodynamic size of 150 nm and a negative Zeta potential were noted. In EDS elemental analysis, the total manganese content was found to be 34.48%, and the oxygen content was 65.52%. The encapsulation and loading efficiencies were 25% and 12.10%, respectively, for the 300-kDa filtration sample compared to 30% and 10.96% for the 6–8 kDa sample. We used the 300-kDa dialysis method for purification of Lipo-MGN, as it has 50 nm pores that can effectively filter out unloaded MGN nanoparticles, which have an average size of 20–30 nm. Thus, only loaded lipo-MGNs could be collected for further experiments. This is the reason why 300 kDa had a better loading efficiency than 6–8 kDa and was selected (Fig. 1).

Fig. 1

Lipo-MGN characterizations. A FE-TEM imaging of the Lipo-MGNs. Manganese oxide nanoparticles are embedded in the hydrophobic region on liposomes. The inset image shows MGN loaded in liposome. B Elemental analysis of Lipo-MGNs by EDX showing the presence of manganese and oxygen. C DLS and Zeta potential analyses of the Lipo-MGNs. D Encapsulation and thermogravimetric analyses of Lipo-MGNs after purification at 300 kDa and 6–8 kDa

Lipo-MGN cellular uptake and viability

The viability of Mn₃O₄ and Lipo-MGNs was measured by MTS assay at concentrations up to 100 µg/ml in the NIH3T3 cell line (Fig. 2A).

Fig. 2

Cell viability by MTS assay and cell uptake analysis. A MTS assay showing the viability of Lipo-MGNs (black bar) and MGN (grey bar) in NIH3T3 cell line. B Lipo-DiD uptake in Hep-G2 cells. C MTS assay showing viability of Lipo-MGNs after radiation in HepG2 cell line. D Log-dose versus response curve of Lipo-MGNs after radiation in HepG2 cell line. MTS assay data show the mean cell viability of quadruplicate samples ± SD. A two-tailed probability value (P) < 0.05 was considered statistically significant. *P < 0.05 relative to Lipo-MGN. **P < 0.01 and ***P < 0.001 relative to the control

We can see the toxicity of MGN starting from the 6.25 µg/ml concentration, decreasing below 80% cell viability and reaching up to 55% cell viability at 100 µg/ml compared to Lipo-MGN, which showed 80% cell viability at 100 µg/ml. This result can be attributed to liposomes, which are biocompatible and protect cells from the direct interaction of manganese oxide nanoparticles, which induce toxicity from oxidative stress. Fluorescence analysis of DiD-loaded liposomes showed high uptake of Lipo-DiD at 100 µg/ml (Fig. 2B). Lipo-MGN effect in HepG2 after radiation (10 Gy) was studied by MTS assay. Data show 50% cell toxicity after radiation compared to the control owing to the conversion of hypoxia to normoxia by lipo-MGN (Fig. 2C). LogIC50 was calculated to be 2.284 based on log-dose versus response curve. IC50 value calculated was 192.3 ug/ml of Lipo-MGN after radiation (Fig. 2D).

In vitro hypoxia study

Our study aimed to convert the hypoxic state of the tumor microenvironment to a normoxic state to promote the RT effect. We used the HepG2 cell line incubated in a hypoxia chamber (CO₂ 5% and O₂ 1%) and treated them with Lipo-MGN at different concentrations to produce oxygen and reinstate normoxia. We used Image-IT™ green hypoxia reagent to check the conversion from hypoxia to normoxia. We observed that 100 µg/ml Lipo-MGNs in hypoxia maximally suppressed fluorescence compared to the control. In FACS analysis, we can see almost 20% suppression of fluorescence for 100 µg/ml Lipo-MGN compared to the hypoxia group with no nanoparticles. At decreasing concentrations up to 25 µg/ml, the fluorescence increased (40% for 50 µg/ml, 68% for 25 µg/ml). The hypoxia group had 90% fluorescence. These data clearly showed the role of Lipo-MGNs in generating oxygen that interacts with fluorogenic compounds in the Image-iT™ Green Hypoxia Reagent (Fig. 3).

Fig. 3

In vitro hypoxia assay in HepG2 cells using Lipo-MGNs by the Image-iT™ Green Hypoxia Reagent. A FACS analysis showing the fluorescence level of the HepG2 cell line after treatment with the Image-IT probe for the control, hypoxia, and 25, 50, 75, and 100 µg/ml groups. B Quantification of FACS data. C Qualitative fluorescent image of HepG2 cells. The data are presented as the mean ± standard deviation with number of samples (n) = 3. The mean % of fluorescence level showed significant differences between the hypoxia and Lipo-MGN treated groups (25, 50 and 100 µg/ml) as analyzed by the Student’s unpaired t-test. A two-tailed probability value (P) less than 0.05 was considered statistically significant. *indicates P < 0.05, ** indicates P < 0.0001

In vitro hypoxia cell toxicity study

As per the in vitro hypoxia assay, 100 µg/ml Lipo-MGN generated sufficient oxygen to bring hypoxic HepG2 cells to normoxia. This oxygenation is very important in the RT of HCC, as oxygen prevents the repair of double-strand breaks in DNA, leading to cell death. To validate this theory, we performed a live/dead assay after RT of HepG2 cells treated with Lipo-MGNs at 3 and 10 Gy. Live/dead assay is a method to evaluate the cell viability using green dye (FDA) representing live cells and red dye (PI) to represent dead cells. We observed that at lower radiation (3 Gy), propidium iodide fluorescence started to appear to be caused by cell death in the hypoxia group, whereas normoxic cells were relatively viable. At 10 Gy radiation, almost 80% of the cells were killed even in the normoxia group. Compared to the normoxia group, hypoxia resulted in more cell death in the 3 and 10 Gy groups, which may be due to toxicity related to the hypoxic environment (Fig. 4).

Fig. 4

In vitro cell toxicity after radiation by live/dead assay. Lipo-MGN with radiation (Lipo-MGN + radiation) at 3 and 10 gy radiation dose treated in normoxic and hypoxic cell lines showed a higher level of red fluorescence (PI) indicating cell death compared to Lipo-MGN without radiation (Lipo-MGN) and control

In vitro T1-weighted MR imaging of Lipo-MGNs

For checking the possibility of MR imaging using Lipo-MGN, we tested T1 contrast at different pH (7.4 and 6.8) and H₂O₂ (0.1 mM and 1 mM) values. T1 contrast signal intensity incrementally increased as the pH value decreased and as the H₂O₂ concentration increased (Fig. 5A). The mean signal intensity significant differences between three groups at different Mn concentrations as analyzed by repeated measure ANOVA test are shown in Fig. 5B. T1 relaxivity (r1) was found highest for pH 6.8 + 1 mM H₂O₂ at 2 mM [Mn] concentration of Lipo-MGN (r1 = 4.02 mM⁻¹ s⁻¹) compared to pH 7.4 + 0.1 mM H₂O₂ at 2 mM [Mn] concentration of Lipo-MGN (r1 = 0.18 mM⁻¹ s⁻¹). This data validates that both pH and ROS played important roles in the dissociation of MGN to Mn ions, contributing to the T1 signal intensity shown in the data. Table comparing relaxation time (T1), relaxation rate (R1) and relaxivity (r1) of each group is included in Additional file 1: S2.

Fig. 5

In vitro MRI study of Lipo-MGN at different H₂O₂ concentrations and pH values. A T1-weighted MR contrast images of Lipo-MGNs from 0 to 2 mM concentrations at pH 6.8–7.4 and at 0.1 mM and 1 mM H₂O₂ concentrations. B Graph plotting the contrast signal intensity of Lipo-MGNs from 0 to 2 mM concentrations at pH 6.8–7.4 and at 0.1 mM and 1 mM H₂O₂ concentrations. The data are presented as the mean ± standard error of the mean with number of samples (n) = 3. The mean signal intensity significant differences between three groups at different Mn concentration as analyzed by repeated measure ANOVA test and in the post hoc analysis by Tukeys test (pH 7.4 + 0.1 mM H₂O₂ vs. pH 6.8 + 1 mM H₂O₂: *P < 0.05)

Ex vivo biodistribution by ICP-MS

When HepG2 tumors reached approximately 5–7 mm in size (after approximately 15 days), Lipo-MGN (in PBS, 200 μl) at a concentration of 0.15 mg [Mn₃O₄]/kg (7.5 mg/kg of Mn₃O₄) was intravenously injected into each mouse through the tail veins of HepG2 tumor-bearing mice. At 8- and 24-h time points, the organs were extracted and added to 3 ml of aqua regia and mixed with 27 ml of distilled water to make a 10% aqua regia:water mixture and submitted for ICP-MS.

To measure the manganese oxide content in organs and tumors after i.v. injection of Lipo-MGNs, ex vivo ICP-MS analysis was performed at different time points. Tumors were extracted at 6, 12, 24, 72 and 192 h, and the highest accumulation was found at the 24-h time point and had decreased by the 72-h time point, indicating elimination of Lipo-MGNs from the tumor region (Fig. 6A). The MR imaging data also validate this result. The highest accumulation levels of manganese oxide were found in the liver and kidney at the 8-h time point (Fig. 6B). After 24 h, there was a decrease in manganese oxide, which can be attributed to increased accumulation in tumors and natural elimination by the reticuloendothelial system (RES) in the liver and kidney.

Fig. 6

Ex vivo biodistribution of Lipo-MGNs by ICP-MS analysis in A HepG2 tumors at different time points and B organs at 8- and 24-h time points. The data are presented as the mean ± standard deviation with number of samples (n) = 5

In vivo MR imaging

For in vivo MR imaging, mice bearing HepG2 tumors were injected intravenously with Lipo-MGNs (7.5 mg/kg), and the mice were then scanned using MR technique. After Lipo-MGN injection, the MRI signal intensity of HepG2 tumors (n = 4) steadily decreased over time. At one day after injection, the contrast enhancement (signal decrease) was seen to be at its peak (Fig. 7).

Fig. 7

In vivo MR imaging of nu BALB/C mice injected with Lipo-MGNs at different time points. A T1-weighted MR contrast images of mice 1, 2 and 3 (m1, m2, m3) at different time points. B T1 contrast intensity ratios pre-injection and post-injection of Lipo-MGNs. The data are presented as the mean ± standard deviation with number of samples (n) = 3. The mean signal intensity showed significant differences between the pre-injection and post-injection time points (3, 6, 8 and 24 h) as analyzed by the Student’s unpaired t-test. A two-tailed probability value (p) < 0.05 was considered statistically significant. *Indicates P < 0.05, ** indicates P < 0.01, *** indicates P < 0.001

Oxygen generation and HIF alpha PCR studies

Oxygen generated by Lipo-MGNs was measured using an O₂-ADPT Oxygen Adapter (Microelectrodes Inc., USA). Briefly, 15 ml of a solution of 1 mg/ml Lipo-MGN was prepared, and H₂O₂ was added to a final concentration of 1 M. The oxygen level was measured in mg/l dissolved oxygen in a solution of Lipo-MGN + H₂O_2. Measurements were taken every 5 min and measured for up to 60 min. The oxygen release in mg/L reached 10 mg/L immediately after mixing with H₂O₂ and reached up to 38 mg/L in 60 min. Then, the release reached the threshold most likely when the Lipo-MGN reaction with H₂O₂ released oxygen and the oxygen escaping from solution reached equilibrium. In response to oxygen release in the HepG2 cancer cell line, hypoxia inducible factor-1 alpha is degraded by oxygen (Fig. 8A). To prove this effect, we conducted HIF-1α mRNA quantification in the presence of Lipo-MGNs by PCR. Compared to the control, HepG2 cells under hypoxia showed a fivefold increase in mRNA levels, and due to the oxygenation capability of Lipo-MGN, the HIF-1α mRNA level in the Lipo-MGN group was comparable to that of the control group. This prove the oxygenation capability of Lipo-MGN and therefore better radiation sensitization in hepatocellular carcinoma (Fig. 8B).

Fig. 8

In vitro oxygen release study of Lipo-MGNs. A Reaction scheme showing the reactivity of Lipo-MGNs towards H₂O₂ for the production of oxygen. B RT‒PCR analysis of HIF-1α mRNA expression after Lipo-MGN treatment in the HepG2 cell line. HIF-1α mRNA is degraded by the oxygen released by Lipo-MGN. The data are presented as the mean ± standard deviation with number of samples (n) = 3. The mean fold change mRNA expression showed significant differences between the hypoxia and Lipo-MGN (100 µg/ml) treated group as analyzed by the Student’s unpaired t-test. A two-tailed probability value (p) less than 0.05 was considered statistically significant. *Indicates p < 0.05

In vivo tumor reduction by radiotherapy

HepG2 tumors were grown in BALB/c mice. As shown in the graph of T1 contrast signal intensity, the peak intensity was reached at the 24-h time point, which we took as the time point to use for radiotherapy after the Lipo-MGN dosage. The tumor size after RT was measured and plotted, as shown in Fig. 9A. The tumor size of the RH group was significantly lower than that of the RO group. This can be attributed to the toxic effect caused by DNA damage caused by radiation and oxygen-mediated DNA repair prevention. The effect of oxygen was confirmed by the results of the RL group, which had a lower concentration of Lipo-MGN and therefore lower oxygen release. A survival study of mice after radiation therapy was conducted, with the last mice in the RH group surviving for more than 85 days compared with 38 days in the control (Additional file 1: S4).

Fig. 9

In vivo tumor reduction by radiotherapy. A Tumor volume plotted against time. B H&E staining of tumor tissue for the NC, RO, RL, RH groups after radiation. C TUNEL assay of the NC, RO, RL, RH groups after radiation. D Mean cell count of TUNEL-positive cells (green fluorescence) is represented as graph for the NC, RO, RL, RH groups after radiation. The data are presented as the mean ± standard deviation with number of samples (n) = 5 for tumor volume against time and n = 3 for mean cell count of TUNEL-positive cells. The mean tumor volume showed significant differences between the RO and Lipo-MGN-treated group (RL and RH) at day 21, as analyzed by the Student’s unpaired t-test. The mean cell count of TUNEL-positive cells showed significant difference between the RH and RL, as analyzed by the Student’s unpaired t-test. A two-tailed probability value (p) less than 0.05 was considered statistically significant. *Indicates p < 0.05, ** indicates p < 0.0001

Immunohistochemical detection of tumor hypoxia was done by hypoxia detection assay. DAB staining was performed to view the anti-HRP secondary antibody binding. It is seen from the data that compared to control, Lipo-MGN low and high concentration is having lesser staining for both DAB and FITC, which indicate lesser pimonidazole adducts. This experiment show that Lipo-MGN will increase oxygen and prevent formation on pimonidazole adduct in tumor (As shown in Additional file 1: Figure S5).

Ex vivo toxicity analysis was performed using H&E staining and TUNEL staining to check the effect of radiation with and without the presence of Lipo-MGNs. The RH group showed more TUNEL-positive cells than the RL and RO groups. H&E staining also clearly showed cells that were damaged due to radiation in the RH and RL groups compared to the RO and NC groups. These results suggest that Lipo-MGNs at higher concentrations can be re-oxygenated in tumors both in vitro and in vivo (Fig. 9B, C).

Discussion

We prepared PEGylated Lipo-MGNs to generate oxygen and convert the hypoxic HCC microenvironment to a normoxic microenvironment. We also used manganese oxide in Lipo-MGNs to obtain T1 contrast MR images. Mn ions are biocompatible materials and provide bright T1-contrast MR imaging. Following incubation in acidic media, r₁ increased due to the evolution of free Mn²⁺ (Gale et al. 2015; Cai et al. 2019). The MRI data in our study also clearly show how manganese oxide dissociates in the presence of ROS and low pH and forms manganese ions, which are responsible for the TI contrast shown in MRI.

An in vitro cytotoxicity study showed that Lipo-MGNs are more biocompatible than bare manganese oxide nanoparticles, which is good if the circulation time to reach peak bioavailability in tumors is 24 h. The in vitro MRI results showed the relationship between pH and ROS levels in the MR T1 contrast level. When the nanoparticles are taken up by cell, they are transported to lysosomes and endosomes, where the pH is low. Inorganic D-Mn₃O₄ is insoluble at neutral pH. After cell internalization of Lipo-MGN, Mn²⁺ ions are released in endosomes where pH is low and lead to generating strong T1 MRI contrast (Chen et al. 2012; Bennewitz et al. 2011; Wang et al. 2015; Kim et al. 2011; Duan et al. 2018; Shapiro and Koretsky 2008).

The main functionality of manganese oxide in liposomes is to generate oxygen and convert hypoxic cancer cells to normoxic cells to make them susceptible to radiotherapy. To prove this hypothesis, a hypoxia assay was performed at different concentrations. The results showed decreasing fluorescent intensity corresponding to increasing oxygen concentration in cells, which is the result of the bio-reductive process that occurs with the hypoxia-sensitive agent in the assay (Takahashi et al. 2012). The effect of normoxia in cancer cells is the inability of cancer cells to repair themselves in the event of radiation, which makes them more susceptible to cell apoptosis (Zhang et al. 2019).

To show the anticancer effect of Lipo-MGNs in cancer cells after radiation, we conducted live/dead assays using the FDA/PI staining method. As predicted, the cells irradiated with 10 Gy showed almost 100% cell death compared to the 3 Gy, control and Lipo-MGN-only groups. As we confirmed HepG2 cell death with more than 3 Gy irradiation, we continued in vivo radiotherapy at a dose of 6 Gy single-dose after injecting Lipo-MGN. After 21 days, we observed a significant tumor reduction compared to the control. Single-dosage radiotherapy can benefit patients with fewer hospital visits and more effective recovery from cancer (Greco et al. 2021).

Our in vivo RT result using 6 Gy radiation is comparable to polymer lipid hydrophilic terpolymer–protein–MnO₂ nanoparticles, where 10 Gy of single-dose radiation was applied to achieve a 40% tumor recovery rate (Abbasi et al. 2016). Similar to our study, Mn₃O₄ was used for cancer therapy in the form of Mn₃O₄@Au-dsDNA/DOX, where dsDNA is used to activate the interferon gene pathway and doxorubicin (DOX) is chemotherapeutic. When combined with chemotherapy, immunotherapy exhibits strong anticancer efficacy by preventing tumor growth and extending in vivo survival times (Zhou et al. 2020). Tumor microenvironment modulation by DMn₃O₄-loaded nanoparticles can also improve therapeutic efficacy by triggering CTL-mediated antitumor immunity, which was seen in the case of hollow-MnO₂-PEG nanoparticles and doxorubicin-loaded alginate nanogel containing Mn²⁺ ions (Yang et al. 2017; Su et al. 2022; Zhou et al. 2022). To improve the tumor reduction effect of hypoxia-attenuating nanoparticle stimuli-responsive drug or gene release system can be introduced like in the case of siRNA-loaded hypoxia-sensitive micelleplex which is made from oligoethylenimine which is crosslinked with hypoxia breakable azobenzene (Li et al. 2022). Nitroimidazole is also a hypoxia stimuli-responsive linker which was used to release camptothecin from polymer-based micelle to treat HeLa cells (Zhou et al. 2021).

Overall manganese oxide nanoparticle-based nanosystems have high potential for therapy owing to catalytic reactivity with H₂O₂, increasing O₂ level in tumor environment. But in order to bring MGN-based nanosystem in clinical scenario, there must be a deeper study to understand the chemical reactivity(Bonet-Aleta et al. 2022) and long-term exposure study to analyze oxidative stress effect in different organ system such as neuronal cells (Alarifi et al. 2017).

In conclusion, we have developed biocompatible Lipo-MGN nanoparticles capable of attenuating hypoxia and treatment of hypoxic, radioresistant HepG2 cancer, along with enhanced MRI T1 contrast imaging capability.