Cold-catalytic antitumor immunity with pyroelectric black phosphorus nanosheets

Catalytic nanomedicine with the innate features of catalysts brings incomparable properties to biomedicine over traditional drugs. The temperature-dependent activity of catalysts provides catalytic nanomedicines with a facile strategy to control their therapeutic performance. Tuning catalytic nanomedicine by cold treatment (4–37 °C) is safe and desired for practical applications, but there is a lack of cold-catalytic platforms. Herein, with black phosphorus (BP) as a model pyroelectric nanocatalyst, we explored the potential of cold-catalysts for antitumor therapy. BP nanosheets with pyro-catalytic activity catalyze the generation of oxidative stress to activate antitumor immunity under cold treatment. Due to the cold-catalytic immunomodulation, immune memory was successfully achieved to prevent tumor metastasis and recurrence. Considering the safety and conductive depth (>10 mm) of cold in the body, pyroelectric nanocatalysts open up exciting opportunities for the development of cold-catalytic nanomedicine.


Pyro-electrochemical measurements
BP nanosheets (0.5 mg) were dispersed in 10 μL of ethanol and then mixed with 10 μL of 5% Nafion perfluorinated resin solution. The mixture solution was dropped on a glassy carbon electrode. After drying, the electrode was immersed in 0.5 M Na2SO4 solution. The pyroelectric current and voltage were recorded on an electrochemical workstation (Kerui RST5200, China).
The measurement was performed with Ag/AgCl as a reference electrode and Pt wire as a counter electrode. The cold temperature was controlled by a thermoelectric cooler (TES1-04903, HongDaEr, China), and powered by a direct current power supply (VC3003A, VICTOR, China). The temperature was decreased from 37 o C to 4 o C for 5 min and then transferred to a 37 o C water bath for another 5 min to recover temperature. Both the pyroelectric currents and pyroelectric potentials were recorded during 4 cycles of temperature variations. The pyroelectric current was measured at 0.02 V vs Ag/AgCl, and pyroelectric potential was measured at 5 mA.

ROS detection
The radicals in cold-catalysis of BP nanosheets were measured through an electron spin resonance (ESR) spectrometer (Bruker model A300, Bruker Biospin GmbH, Germany). Briefly, BP nanosheets were treated with temperature variations between 4 o C and 37 o C for 20 cycles in water or DMSO. ESR spectra of the mixture solution were collected with DMPO as a spin trap agent.
•OH was quantitatively detected with terephthalic acid (TA) as a fluorescent probe. TA (0.83 mg) was dissolved in 10 mL of NaOH aqueous solution (2 mM), and then 100 μL of TA solution was mixed with 1 mL of BP nanosheets (50 μg mL −1 ) in water. The solution was treated with a thermoelectric cooler (4 o C) for 5 min. Afterwards, the samples were transferred into a 37 o C water bath and incubated for 5 min to recover temperature. The temperature was varied for different cycles (0, 5, 10, 15, 20 cycles), and the fluorescence spectra were measured on a fluorescence spectrometer.
To quantify •O2 − production during the pyro-catalytic process, 100 μL of NBT solution (0.01 mM) in DMSO was mixed with 1 mL of BP nanosheets (50 μg mL −1 ) in water. Similar as above mentioned procedure, the temperature was varied between 37 o C and 4 o C for different cycles (0, 5, 10, 15, 20 cycles), and the absorption at 680 nm was measured on a UV−2450 spectrometer.
DCFH-DA was utilized to detect the total production of ROS in the solution. Typically, 0.5 μL of DCFH-DA (10 mM) solution was mixed with 20 μL of NaOH solution (10 mM) for hydrolysis into DCFH. Afterwards, the mixture was transferred to 10 mL of BP nanosheets (50 μg mL −1 ) in PBS (10 mM, pH 7.4). Both a thermoelectric cooler (0.6 W ~ 7.5 W) and a 37 o C water bath were used for temperature variation between 4 o C and 37 o C. The fluorescence intensity of the mixture was recorded on a fluorescence spectrometer.

Energy band analysis
The mixed solution of BP nanosheets was dropped on the glassy carbon electrode according to method 1.4. After drying, the electrode was immersed in 0.5 M Na2SO4 solution. The Mott-Schottky curve was measured by an electrochemical workstation (Kerui RST5200,China) with Ag/AgCl reference electrode and Pt wire counter electrode placed in an electrolytic cell. The Tauc plot was calculated on UV-vis diffuse reflectance spectrum data as following equation hν, A and Eg represent the values of photon energy, proportionality constant and band gap respectively. The direct transition of n = 2 or the indirect transition of n = 1/2 can be defined by the Kubelka-Munk function F(R∞) 1-2 : where α, S and R are absorption, scattering coefficient, and diffuse reflection, respectively.
The XPS valence band spectrum were recorded on an X-ray photoelectron spectrometer (ThermoFischer, ESCALAB 250XI, USA) with an Al Kα ray (hv = 1486.6 eV) as the excitation source.

Intracellular oxidative stress and oxidative damage
4T1 cells were incubated with BP nanosheets (50 μg mL −1 ) for 12 h, and then subjected to temperature variations between 4 o C and 37 o C for 20 cycles as mentioned above. Intracellular oxidative stress was then measured with DCFH-DA (50 μg mL −1 ) as a probe. In order to simulate a hypoxic environment, 4T1 cells were cultured in a medium bubbled with N2, and sealed with paraffin oil. The other operations were the same as above. To detect mitochondrial integrity, cells were stained with JC-1 (10 μg mL −1 ) and Hoechst 33342 (10 μg mL −1 ) for 30 min. Cells were imaged under an inverted fluorescence microscope (Olympus IX-83, Japan), and the fluorescence intensity was analyzed by CellSens software (Olympus, Japan).

Immunogenic cell death
To detect immunogenic cell death, 4T1 cells were incubated with BP nanosheets (50 μg mL −1 ) and then subjected to temperature variations between 4 o C and 37 o C for 20 cycles. After incubated for another 12h, ATP level in the supernatant of each well was measured by the ATP assay kit following the manufacturer's protocol. For calreticulin immunostaining, cells were stained with anti-calreticulin at 4 °C for 12 hours, and incubated with AF647-conjugated secondary antibody at 37 °C for another 2 h. For HMGB-1 immunostaining, cells were stained with anti-HMGB-1 antibody at 4 °C for 12 hours, and incubated with AF647-conjugated secondary antibody at 37 °C for 2 h. After nucleus staining with DAPI at 37 o C for 30 min, cells were imaged under a confocal laser scanning microscope (ZEISS LSM880, Germany).
4T1 cells were incubated with BP nanosheets (50 μg mL −1 ) and then subjected to temperature variations between 4 o C and 37 o C for 20 cycles. After incubation for 12h, cells were collected for cytoplasmic/membrane protein extraction with an extraction kit. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was conducted with 10% arylamide gels for protein separation. The gels were then transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, USA) at a current of 290 mA for 2 h. To prevent non-specific binding of proteins to antibodies, PVDF membranes were blocked with skimmed milk (5 wt%) for 3 h.
Afterwards, the PVDF membranes were stained with the primary antibodies at 4 °C overnight and with the secondary antibodies at room temperature for 2 h. Protein bands were detected with a BeyoECL Plus kit and imaged with a chemiluminescence gel imaging system (Universal Hood II, BIO-RAD, USA).

Live and dead staining
4T1 cells were incubated with BP nanosheets (50 μg mL −1 ) for 12 h and then subjected to temperature variations between 4 o C and 37 o C for 20 cyles. After incubated for another 12h, all the cells were washed with PBS and stained with calcein-AM/PI at 37 °C for 20 min. Then, cells were washed with PBS twice and imaged under an inverted fluorescence microscope.

Cell apoptosis and necrosis assay
4T1 cells were incubated with BP nanosheets (50 μg·ml −1 ) overnight, and subjected to temperature variations between 4 o C and 37 o C for 20 cycles, then cells were incubated for 12 h.
To detect cell apoptosis, cells were collected and stained with Annexin V-FITC/PI for 20 min according to the manufacturer's protocol. The percentage of apoptotic cells was analyzed with a flow cytometer (NovoCyte 2040, Agilent, USA).

In vitro macrophage polarization
J774A.1 macrophages were polarized with LPS (1 μg mL −1 ) or IL-4 (25 ng mL −1 ) overnight to induce M1 or M2 polarization. M2 macrophages were treated with BP nanosheets (50 μg mL −1 ), and subjected to temperature variations between 4 o C and 37 o C for 20 cycles, and then incubated for 4 h. After immunostaining with APC-labeled CD86 and FITC-labeled CD206 antibodies at 4 o C overnight, cells were detected with a flow cytometer for polarization analysis.
IL-6, TNF-α and IL-10 in the supernatants were collected for ELISA measurements following the manufacturer's protocols.
Since tumor cells were subjected to cold-catalytic oxidative stress prior to macrophages. In other words, tumor cells were exposed to cold-catalytic oxidative stress for a longer time than macrophages. To mimic this process in vitro, tumor cells were subjected to cold-therapy of BP nanosheets for 12 h; whereas macrophages were subjected to cold-therapy of BP nanosheets for 4 h. Afterwards, BP nanosheets were removed and macrophages were further incubated for another 8 h. The cell viability of macrophages was determined with LDH assay kit.

Immunoblotting analysis
Macrophages were treated with BP nanosheets (50 μg mL −1 ) for 4 h, and then subjected to temperature variations between 4 o C and 37 o C for 20 cycles. Afterwards, macrophages were incubated for 4 h and collected for nuclear/cytoplasmic protein extraction with an extraction kit.
Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was conducted with 10% arylamide gels for protein separation. The gels were then transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, USA) at a current of 290 mA for 2 h. To prevent non-specific binding of proteins to antibodies, PVDF membranes were blocked with skimmed milk (5 wt%) for 3 h. Afterwards, the PVDF membranes were stained with the primary antibodies at 4 °C overnight and with the secondary antibodies at room temperature for 2 h. Protein bands were detected with a BeyoECL Plus kit and imaged with a chemiluminescence gel imaging system (Universal Hood II, BIO-RAD, USA).

Cold conduction
The experiment was carried out in accordance with the regulations approved by Medical Ethics Committee of Xiangya Hospital, Central South University. Written informed consent was obtained from the volunteer and documented prior to the experiments. The thermoelectric cooler was fixed on a volunteer hand and the cooler temperature was set at 4 o C. After exposure to the thermoelectric cooler for 5 min, human hand was imaged under a thermal imaging camera to record the temperature variation (FLIR C2, Teledyne FLIR, USA). A pork skin (10 mm thickness) was exposed to the thermoelectric cooler for 5 min, and temperature on the other side of the skin was monitored under the thermal imaging camera.
A 4 °C cold source was applied outside the skin, and the cold distribution within skin tissue was simulated by COMSOL Multiphysics. The geometric model was built as a cuboid structure with length, width and height of 60mm, 60mm and 10mm, respectively. The structure parameters were set as following: density 1100 kg m -3 , thermal conductivity 0.9 W m -1 K -1 , specific heat capacity 3300 J Kg -1 K -1 . 3

Cold-induced injury
To investigate the cold-induced injury, mice (n = 4) were treated with the thermoelectric cooler (4 o C, 5 min) for 20 cycles. Mice without cold treatment were taken as control. Following mouse euthanasia, the skin was collected and histologically examined through H&E staining.

In vivo macrophage polarization
Tumor-bearing mice were received various treatments as mentioned above. On day 3 post-injection, the tumor tissue was collected and homogenized to obtain a single-cell suspension. After immunostaining with PerCP/Cy5.5-labeled F4/80, APC-labeled CD86 and FITC-labeled CD206 antibodies, macrophage polarization was analyzed with a flow cytometer.

Dendritic cell maturation and T cell infiltration within tumors
To construct a bilateral tumor-bearing mice model, 4T1 cells were firstly subcutaneously injected into the right flank and the left flank of mice to build primary and distant tumor models, respectively. When the size of primary tumors reached ~50 mm 3 , mice were received the following different treatments (n = 5 in each group): (1) blank control, (2) ΔT, (3) BP, (4) BP + ΔT.
Mice were received intratumoral injection of BP nanosheets (5 mg kg −1 ). At 24 h post injection, mice in group (2) and (4) were treated with the thermoelectric cooler treatment (4 o C, 5 min) for 20 cycles on the primary tumors. On day 3 post injection, tumors, spleens and blood were collected following mouse euthanization. After homogenization and hemolysis of erythrocytes, cell suspensions of tumors and spleens were obtained for immunostaining. Tumor cell suspensions were stained with APC-labeled CD3, PE-labeled CD4 and FITC-labeled CD8a antibodies. The splenic cell suspensions were stained with PerCP/Cy5.5-labeld CD11c, PE-labeled CD80 and FITC-labeled CD86 antibodies. All the cells were analyzed on a flow cytometer. Mice blood was centrifuged (1000 g, 5 min), and the serum IFN-γ level was measured with an ELISA kit.

Stability and biosafety study
To determine the degradability of BP nanosheets, the nanosheets (50 μg·ml −1 ) were dispersed in distilled water and DMEM cell culture medium, respectively. The solution was taken every 24 h to record UV-Vis absorption spectrum. Meanwhile, BP nanosheets were centrifuged at 12000 g for 10 min, and the concentration of phosphate ion in the supernatants was determined with the phosphate assay kit.
After in vivo cold-catalytic immunotherapy, all the mice were euthanized and the major organs were collected for histological hematoxylin & eosin (H&E) examination.
To study cytotoxicity of BP nanosheets, 4T1, L929 and A549 cells were incubated with BP nanosheets (50 μg·ml −1 ) overnight. The supernatants of medium were collected for LDH detection with LDH assay kit. Cell viability was calculated with the following equation (7).
Mouse erythrocytes were separated from the whole blood and washed with PBS three times.
Erythrocytes (2 μL) were incubated with BP nanosheets (0, 25, 50, 100, 200 μg mL −1 ) in PBS at 37 o C for 8 h. Then, the mixture solution was centrifuged at 1000 g for 5 min, and the supernatants were collected for absorption measured ( = 540nm). The hemolysis rate was calculated with the following equation: where, A is the absorption of supernatant after incubation with BP nanosheets, and A0 is the absorption of hemolytic erythrocytes in pure water.
The fluorescence of Cy5.5-labeled BP nanosheets in mice was monitored on an in vivo imaging system (IVIS Lumina III, PerkinElmer, USA).

Statistical analysis
Data were expressed as mean ±SD. Experiments were repeated three times in this work unless otherwise noted. A one-way analysis of variance was employed to analyzed the significance of the difference, and the statistical significance was defined as *p < 0.05 and **p < 0.01. Figure S1. AFM image of BP nanosheets.     Fluorescence spectra of DCFH with a thermoelectric cooler as the cold source (ex = 480 nm). The input power of the cooler was varied from 0.6 W to 7.5W. Figure S7. The fluorescence spectra of DCFH for ROS detection at separate cooling process, heating process and cycle process. **p < 0.01.                     ΔT, (4) ΔT + αPD-1, (5) BP nanosheets, (6) BP nanosheets + αPD-1, (7) BP nanosheets + ΔT, (8) BP nanosheets + ΔT + αPD-1. Figure S30. Main organs were histologically examined through H&E staining. All groups: (1) PBS,