Overcoming chemotherapy resistance using pH-sensitive hollow MnO2 nanoshells that target the hypoxic tumor microenvironment of metastasized oral squamous cell carcinoma

Background Smart nanoscale drug delivery systems that target acidic tumor microenvironments (TME) could offer controlled release of drugs and modulate the hypoxic TME to enhance cancer therapy. The majority of previously reported MnO2 nanostructures are nanoparticles, nanosheets, or nanocomposites incorporated with other types of nanoparticles, which may not offer the most effective method for drug loading or for the controlled release of therapeutic payloads. Previous studies have designed MnO2 nanoshells that achieve tumor-specific and enhanced combination therapy for localized advanced cancer. However, the therapeutic effect of MnO2 nanoshells on metastatic cancer is still uncertain. Result Here, intelligent “theranostic” platforms were synthesized based on hollow mesoporous MnO2 (H-MnO2) nanoshells that were loaded with chemotherapy agents docetaxel and cisplatin (TP) to form H-MnO2-PEG/TP nanoshells, which were designed to alleviate tumor hypoxia, attenuate angiogenesis, trigger the dissolution of Mn2+, and synergize the efficacy of first-class anticancer chemotherapy. The obtained H-MnO2-PEG/TP nanoshells decomposed in the acidic TME, releasing the loaded drugs (TP) and simultaneously attenuated tumor hypoxia and hypoxia-inducible factor-1α (HIF-1α) expression by inducing endogenous tumor hydrogen peroxide (H2O2) decomposition. In vitro experiments showed that compared with the control group, the proliferation, colony formation and migration ability of CAL27 and SCC7 cells were significantly reduced in H-MnO2-PEG/TP group, while cell apoptosis was enhanced, and the expression of hypoxia-inducible factor-1α(HIF-1α) was down-regulated. In vivo experiments showed that tumor to normal organ uptake ratio (T/N ratio) of mice in H-MnO2-PEG/TP group was significantly higher than that in TP group alone (without the nanoparticle), and tumor growth was partially delayed. In the H-MnO2-PEG/TP treatment group, HE staining showed that most of the tumor cells were severely damaged, and TUNEL assay showed cell apoptosis was up-regulated. He staining of renal and liver sections showed no obvious fibrosis, necrosis or hypertrophy, indicating good biosafety. Fluorescence staining showed that HIF-1α expression was decreased, suggesting that the accumulation of MnO2 in the tumor caused the decomposition of H2O2 into O2 and alleviated the hypoxia of the tumor. Conclusion In conclusion, a remarkable in vivo and in vitro synergistic therapeutic effect is achieved through the combination of TP chemotherapy, which simultaneously triggered a series of antiangiogenic and oxidative antitumor reactions. Graphic abstract Supplementary Information The online version contains supplementary material available at 10.1186/s12951-021-00901-9.


Introduction
The tumor microenvironment (TME) is characterized by oxygen-deficient solid tumors. Hypoxic conditions significantly promote tumor heterogeneity and increase metastatic spread [1][2][3]. Furthermore, undesirable angiogenesis and immunosuppression are caused by disordered metabolism and unstable genome phenotype accelerated in hypoxic TMEs, which consistently contribute to tumor resistance to various oxygen-related therapeutics [4,5]. Many of the hallmarks of metastatic cancers are related to the hypoxic TME [6], which stimulates hypoxia-inducible factor (HIF)-driven transcriptional responses that upregulate the expression of hypoxia-inducible genes that facilitate invasion and metastasis [7,8]. These hypoxic effects play essential roles in the outcomes of various cancer therapies [5,9,10].
Nanoscale drug delivery systems (nano-DDSs) have been regarded as an ideal approach to overcome the hypoxic TME, as nano-DDSs are capable of responding to the inherent acidic and hypoxic features of the TME [11,12]. Recently, several studies have focused on the development of manganese dioxide (MnO 2 ) nanostructures that can decompose under the acidic TME [13][14][15][16], generating Mn 2+ ions that could enhance T1 images during magnetic resonance imaging (MRI) [17,18]. Accordingly, MnO 2 nanostructures could offer a safe DDS without long-term toxicity in in vivo therapy [13][14][15][16]. MnO 2 nanostructures could also relieve tumor hypoxia by triggering the decomposition of H 2 O 2 that is present in the TME [19,20]. Hollow nanostructures with mesoporous shells (such as hollow mesoporous silica) have large cavities that have been demonstrated to be excellent drug loading/delivery systems, loading high quantities of therapeutic agents, whose release may be precisely controlled by tuning the shell structure or coating [21,22].
The combination of docetaxel and cisplatin (TP) has become a first-line anticancer therapy in advanced OSCC, which provides good progression-free survival (PFS) and overall survival (OS) [23,24]. Previous research has also demonstrated the limitations caused by TME hypoxia, which contributes to metastasis and angiogenesis of OSCC [25][26][27][28]. TP can be co-loaded into the H-MnO 2 -PEG nano-platform (H-MnO 2 -PEG/ TP) with high loading capacities. Under acidic pH, the rapid decomposition of the MnO 2 nanoshells leads to the release of the loaded drugs (TP), while simultaneously resulting in significantly enhanced T1-images during MRI.
showed that tumor to normal organ uptake ratio (T/N ratio) of mice in H-MnO 2 -PEG/TP group was significantly higher than that in TP group alone (without the nanoparticle), and tumor growth was partially delayed. In the H-MnO 2 -PEG/ TP treatment group, HE staining showed that most of the tumor cells were severely damaged, and TUNEL assay showed cell apoptosis was up-regulated. He staining of renal and liver sections showed no obvious fibrosis, necrosis or hypertrophy, indicating good biosafety. Fluorescence staining showed that HIF-1α expression was decreased, suggesting that the accumulation of MnO 2 in the tumor caused the decomposition of H 2 O 2 into O 2 and alleviated the hypoxia of the tumor.

Conclusion:
In conclusion, a remarkable in vivo and in vitro synergistic therapeutic effect is achieved through the combination of TP chemotherapy, which simultaneously triggered a series of antiangiogenic and oxidative antitumor reactions. Lung metastasis is a common feature of advanced OSCC and associated with a poor prognosis (5-year OS < 30%), which is contributed to by hypoxia-induced resistance [26,27]-unlike pulmonary metastasis of HER2-positive breast cancer, which can be effectively treated through targeted therapies. Currently, there is a lack of effective targeted treatments for OSCC lung metastasis. Moreover, previous research has shown that HIF, which can be downregulated after hypoxia is relieved, is a promoting factor of angiogenesis. Considering MnO 2 are able to target the TME and generate O 2 , we hypothesized that using hollow MnO 2 nanostructures as smart DDSs may achieve an ideal therapeutic effect in the application of pulmonary metastasis of OSCC by reversing hypoxia-induced chemotherapy resistance.
Previous studies have designed similar smart DDSs to achieve tumor-specific enhanced combination therapy [29,30]. In a previous system, mesoporous MnO 2 shells combined with chlorine e6 (Ce6) and doxorubicin (DOX) were used to treat 4T1 cells in a subcutaneous model, which obtained an ideal effect. To the best of our knowledge, the therapeutic effect of MnO2 particles on OSCC pulmonary metastasis has not been assessed. In our research, we aimed to develop a system that could target OSCC lung metastasis by using pH-sensitive mesoporous MnO2 nanoshells co-loaded with first-line OSCC chemotherapeutic drugs, TP. We also assessed whether these TP-loaded MnO2 nanoshells were effective against TPresistant OSCC cells.

Synthesis of H-MnO 2 -PEG/TP
Solid silica nanoparticles (sSiO 2 ) were synthesized following the reported method [31]. Then, an aqueous solution of KMnO 4 (300 mg) was added dropwise to the sSiO 2 suspension (40 mg) under ultrasonication. After 6 h, the precipitate was obtained by centrifugating the suspension at 14,800 rpm. The as-prepared mesoporous MnO 2 -coated sSiO 2 were dissolved in 2 M Na 2 CO 3 aqueous solution at 60 °C for 12 h. The obtained hollow mesoporous MnO 2 (H-MnO 2 ) nanoshells were centrifuged and washed with water several times. Then, 5 ml of the H-MnO 2 solution (2 mg/ml) was added to 10 ml PAH solution (5 mg/ml) under ultrasonication. After stirring for 2 h, the above solution was centrifuged and washed with water. The obtained H-MnO 2 /PAH solution was added drop wise to PAA (10 ml, 5 mg/ml) under ultrasonication. After 2 h of stirring, the above solution was centrifuged and washed with water, before it was mixed with mPEG-5 K-NH 2 (50 mg) under ultrasonication for 30 min. After adding EDC (15 mg) and stirring for 12 h, the prepared H-MnO 2 -PEG was collected by centrifugation and washed with water three times. For docetaxel and cisplatin (TP) loading, the H-MnO 2 -PEG solution (0.2 mg/ml) was mixed with different concentrations of TP for 12 h. TP were co-loaded into H-MnO 2 -PEG with appropriate concentrations, yielding H-MnO 2 -PEG/TP which were used in further experiments.

Characterization
Scanning electron microscopy (SEM; JSM-2100F, JEOL, Tokyo, Japan) was applied to characterize the nanoparticle morphology. Ultraviolet-visible (UV-Vis) spectra were measured with a PerkinElmer Lambda 750 UV/ Vis/NIR spectrophotometer. Nanoparticle size and zeta potential were determined by a Malvern Zetasizer (ZEN3690, Malvern, UK) and Nano ZS90 (Malvern, UK). Surface area and pore size were measured by Surface Area and Porosity Analyzer (Micromeritics Instrument Corp. ASAP2050). The functional groups and chemical structure of the nanofibers were performed by Fourier transform infrared (FT-IR) spectroscopy (Nicolet iS50) in the wavenumber range of 4000-400 cm -1 .

Degradation and drug release studies
H-MnO 2 -PEG was incubated with PBS at different pH values (4.5, 5.5, 6.5, and 7.4) for different time periods (0-36 h). At a given time point, the solution was measured by SEM and UV-Vis spectrometry for characterization. To study the TP release, a solution of H-MnO 2 -PEG/ TP was dialyzed against PBS with different pH values (4.5, 5.5, 6.5, and 7.4) at room temperature. The amount of TP released at different time points was measured by UV-Vis spectrometry.

In vitro cell experiments
The tongue squamous cell carcinoma cell line CAL27 was purchased from ATCC (Manassas, VA, USA). Mice oral squamous cell carcinoma cell line SCC7 was provided as a gift from the Nanjing Medical University (China). For cell toxicity assays, cells were seeded into 96-well plates (1 × 10 4 per well) for 24 h and incubated with a series of concentrations of TP. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) solution (0.5 mg/ml) was added to the wells to measure the cell viability of the treated cells relative to the untreated cells. For confocal fluorescence microscopy, CAL27 cells were seeded onto coverslips at the bottom of a dish containing H-MnO 2 -PEG/TP (docetaxel: 3.1 nM, cisplatin: 18 nM) for different incubation times (1, 4, 8, and 12 h). After washing with PBS three times, the cells were labeled with 4′,6-diamidino-2-phenylindole (DAPI) and imaged using a laser scanning confocal microscope (Leica SP5). Lentivirus-transduced stable cells were seeded into 6-well plates at a density of 1000 cells per well and incubated for 10-14 days. The colonies were fixed and stained, and those with more than 50 cells were counted under a dissecting microscope.

Western blot analysis
Proteins were extracted from OSCC cells treated with H-MnO 2 -PEG/TP (docetaxel: 3.1 nM, cisplatin: 18 nM) for 0, 2, 4, 8, 12, and 16 h. The membranes were blocked by adding QuickBlock ™ Blocking Buffer at room temperature for 20 min and then incubating with primary antibodies (β-actin, 1:1000; HIF-1α, 1:1000) overnight at 4 °C. The membranes were then incubated with fluorescently-labelled anti-rabbit IgG secondary antibodies (7704, Cell Signaling Technology, USA) at a 1:10000 dilution for 1 h at room temperature. Immunoreactive bands were detected using enhanced chemiluminescence. The observation and analysis of immunoreactive bands were performed using the Odyssey Infrared Imaging System (LI-COR Biosciences, USA).

Animal models
SPF BALB/c nude mice (nu/nu, 4 weeks old, weighing approximately 20 g) were purchased from Shanghai experimental animal center (Shanghai, China) and placed in the SPF facility of the Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine. All laboratory procedures were approved by the Laboratory Animal Care and Use Committee at the hospital. High metastatic oral squamous cell carcinoma was established. After luciferase lentivirus transfection, CAL27 cells (1 × 10 6 ) were suspended in 50 μl of PBS and injected by intravenous injection (IV) or subcutaneous injection (SC). The mice bearing CAL27 tumors were treated 10 days after injection.

In vivo imaging
In vivo fluorescence imaging was performed using the Maestro In-Vivo Fluorescence Imaging System (CRi Inc., USA). MRI was conducted under a BioSpec 70/20 USR (Bruker, USA) with a special coil for small animal imaging.

Immunohistochemistry
CAL27 tumor-bearing mice were injected intravenously with PBS or H-MnO 2 -PEG/TP. Liver, kidney, and lungs bearing tumors were surgically excised 20-or 120-min post injection. Tissue sections (4 mm) were stained with hematoxylin and eosin (HE). Terminal deoxynucleotide transferase dUTP notch end labeling (TUNEL) was used to detect apoptotic cells. To detect oxidation, the tumor sections were treated with mouse anti-HIF1α primary antibody (dilution 1:200, Abcam Inc. USA) and Alexa Flour ® 488-conjugated goat anti-rabbit secondary antibody (dilution 1:200, CST Inc. USA) following the instructions. Tumor blood vessels were stained by

Synthesis and characterization of H-MnO 2 -PEG
The H-MnO 2 -PEG/TP synthetic process is illustrated in Fig. 1a. Monodispersed silica nanoparticles were synthesized by hydrolyzation of tetraethyl orthosilicate (TEOS) and then used immediately as the hard template. A uniform layer of mesoporous MnO 2 was grown on the surface of the as-made silica nanoparticles by mixing them with KMnO 4 , which was reduced by unreacted organosilica existing on the prepared silica nanoparticles. The H-MnO 2 nanoshells were obtained after incubating MnO 2 @SiO 2 nanoparticles with a Na 2 CO 3 solution to dissolve silica. To enhance their water solubility and physiological stability, H-MnO 2 nanoshells were modified with PEG through a layer-by-layer (LBL) polymer-coating method. In this process, negativelycharged H-MnO 2 nanoshells were coated with a cationic polymer PAH and then an anionic polymer PAA through electrostatic interaction. Amino-terminated PEG (NH 2 -PEG) was then conjugated to the surface of PAAcoated H-MnO 2 nanoshells via amide formation, producing H-MnO 2 -PEG nanoshells. TP were simultaneously loaded into the hollow structure of the H-MnO 2 -PEG nanoshells, yielding H-MnO 2 -PEG/TP. SEM and TEM images revealed the spherical morphology and the hollow structure of the H-MnO 2 -PEG nanoshells (Fig. 1b,   c). The hollow structure of the H-MnO 2 -PEG nanoshells was further confirmed by SEM-EDS (Fig. 1d, e). The change in zeta potential for the nanoparticles obtained at different steps of synthesis are shown in Fig. 1f. In the process of surface functionalization, the step wise altered spectrogram indicated successful LBL coating of polymers on the nanoparticles (Fig. 1g). No diffraction phenomena was observed according to XRD result (Fig. 1h), which indicates a low crystallinity structure of synthesized nanoparticles. Because of the structural defects, oxygen vacancies and low manganese of the low crystallinity structure synthesized at low temperature, it is easier to degradate in vivo [32,33].

pH-dependent nanoparticle decomposition and drug behavior
It is well known that MnO 2 is stable at neutral and alkaline pH but can decompose into Mn 2+ at reduced pH [34]. Therefore, TEM images of H-MnO 2 -PEG incubated in PBS with different pH values (5.5 and 7.4) at different processing times were recorded (Fig. 2a). The morphology of H-MnO 2 -PEG nanocrystals showed no significant change at pH 7.4 after eight hours, indicating that H-MnO 2 -PEG nanocrystals were stable in neutral environments. However, due to the decomposition of MnO 2 into Mn 2+ ions, H-MnO 2 -PEG showed timedependent degradation behavior in acidic solutions. The degradation rate was determined by decreasing the MnO 2 -characteristic absorption band (Fig. 2b), which appears to be stable at pH 7.4, but rapidly decreases at pH 6.5, 5.5, and 4.5, further demonstrating the ultra-sensitive pH-responsive degradation behavior of H-MnO 2 -PEG. The H-MnO 2 -PEG with mesoporous shells were expected to have an efficient drug-loading ability. H-MnO2-PEG nanoshells were loaded with TP. Under ultrasonication, H-MnO 2 -PEG nanoshells were incubated and stirred with different concentrations of free DOC and DDP. Drug-loading capacities was evaluated by UV-Vis spectroscopy. At the feeding weight ratio of 1:1 (DOC:DDP), the drug loading capacity of the nanoshells was high: 75.53% (DOC:MnO 2 ) and 71.75% (DDP:MnO 2 ; Fig. 2c). DOC and DDP could also be simultaneously loaded into the hollow structure of H-MnO 2 -PEG nanoshells, obtaining dual drug co-loaded H-MnO 2 -PEG/TP nanoparticles (Fig. 2d) and enhanced release with increasing concentration (Fig. 2e). Drugrelease behaviors of DOC and DDP from H-MnO 2 -PEG/ TP were studied in solutions with different pH values (Fig. 2f ). Compared to the slow drug-release profiles of H-MnO 2 -PEG/TP at pH 7.4, the release speeds of both DOC and DDP were found to be much faster in mild acidic solutions at pH 6.5, 5.5, and 4.5, owing to the acidic-triggered decomposition of H-MnO 2 nanocarriers into Mn 2+ ions.

In vitro experiments with H-MnO 2 -PEG/TP
As described in previous research, the efficacy of TP is limited by the hypoxic TME of a solid tumor [35][36][37][38][39].  (Fig. 3b). Then, the in vitro efficacy of H-MnO 2 -PEG nanoshells as a multifunctional DDS was assessed by using CAL27 cells. As expected, no significant difference was observed in OSCC cells that were treated in different concentrations of H-MnO 2 -PEG (Additional file 1: Figure S1). Then, SCC7 and CAL27 cells were used to determine the IC50 of DOC and DDP. Cells were then treated with H-MnO 2 -PEG/TP in either an N 2 or O 2 environment, and the cell viabilities were determined by an MTT assay after incubation for 24 h (Fig. 3c). We used H-MnO 2 -PEG/TP for in vitro combination treatment. CAL27 cells incubated with H-MnO 2 -PEG/TP for different periods of time were then imaged by a confocal fluorescence microscope (Fig. 3d). Both DDP-coumarin (GFP) and DOC-rhodamine (RFP) fluorescence inside cells was significantly enhanced with increased incubation time. The colony-formation and -migration abilities were significantly decreased in CAL27 and SCC7 cells in the H-MnO 2 -PEG/TP group compared to the control group (p < 0.01) (Fig. 4a, b).
Flow cytometry assays also confirmed that apoptosis was induced by H-MnO 2 -PEG/TP (Fig. 4c). Hypoxia is a key concern during the treatment of non-small cell lung cancer (NSCLC) [39], as with pulmonary metastasis of OSCC, and hypoxia-inducible factor 1 alpha (HIF-1α) has been associated with increased tumor resistance to therapeutic agents such as cisplatin. To further evaluate, the downregulation of HIF-1α induced by H-MnO 2 -PEG/TP was confirmed using western blot. (Fig. 4d).

In vivo and ex vivo imaging with H-MnO 2 -PEG/TP
After confirmation of in vitro efficacy of H-MnO 2 -PEG/ TP nanoshells, the effect of H-MnO 2 -PEG/TP in an OSCC subcutaneous bearing or pulmonary metastasis model was assessed. Mn 2+ ions with five unpaired 3d electrons could decompose from MnO 2 under the acidic conditions of the TME and is known as a T1-shortening agent in MRI [41]. In solutions at different pH (5.5 and 7.4), the H-MnO 2 -PEG/TP incubated in solutions at pH 5.5 showed a brighter T1-shortening image compared with the image at pH 7.4 (Fig. 5a). To demonstrate the use of H-MnO 2 nanoshells for tumor-specific imaging, H-MnO 2 -PEG/TP nanoshells were injected into tumor, and into muscle on the opposite side of the tumor in tumor-bearing mice for MRI (Fig. 5b). Caused by the acidic TME, the tumor showed significantly enhanced images in T1 signaling after injection of H-MnO 2 -PEG/ TP nanoshells, whereas the muscle area with the same concentration of injected nanoparticles had reduced T1-signal enhancement (Fig. 5b). This phenomenon provides direct evidence that H-MnO 2 has ultra-sensitive pH-responsive T1-MR contrast performance, which is particularly useful for tumor-specific imaging. After intravenous injection of H-MnO 2 -PEG/TP nanoshells (dose of MnO 2 : 10 mg/kg, docetaxel: 10 mg/kg, cisplatin: 2.5 mg/kg), in vivo fluorescence imaging was used to track the nanoparticles in CAL27 pulmonary metastasis Balb/c mice (Additional file 2: Figure S2). Semi-quantitative biodistribution data based on ex vivo imaging of major organs and tumors was collected two hours post injection, indicating a high tumor uptake of H-MnO 2 -PEG/TP (Fig. 5c). Notably, strong fluorescence was found in the kidneys of mice after H-MnO2-PEG/TP injection, illustrating a rapid renal clearance of the decomposed nanoshells. Previous research [42] used the tumor to