Facile engineering of silk fibroin capped AuPt bimetallic nanozyme responsive to tumor microenvironmental factors for enhanced nanocatalytic therapy

Background: Reactive oxygen species (ROS), as a category of highly reactive molecules, are attractive for eliminating tumor cells in situ. However, the intrinsic tumor microenvironment (TME) always compromises treatment efficacy. In another aspect, silk fibroin (SF), as a category of natural biomacromolecules, is highly promising for synthesis of metallic nanocrystals via biomineralization. Methods: As a proof-of-concept study, AuPt bimetallic nanozyme derived from bioinspired crystallization of chloroauric acid and chloroplatinic acid was facilely developed in the presence of silk fibroin (SF). Antitumor effects caused by the as-synthesized AuPt@SF (APS) nanozyme were demonstrated in 4T1 tumor cells in vitro and xenograft tumor models in vivo. Results: APS nanozyme can decompose glucose to constantly supply H2O2 and deplete intracellular glutathione (GSH). APS nanozyme can simultaneously convert adsorbed O2 and endogenic H2O2 into superoxide radicals (•O2-) and hydroxyl radical (•OH), respectively, upon highly efficient catalytic reaction. Subsequently, these cytotoxic ROS cause irreversible damage to the cell membrane, nucleic acid and mitochondria of tumors. Upon fluorescence/photoacoustic (FL/PA)-imaging guidance, remarkable tumor damage based on the current nanoplatform was confirmed in vivo. Conclusion: The objective of our investigation is to supply more useful insights on the development of SF-based nanocatalysts, which are specifically responsive to TME for extremely efficient tumor theranostics.

3 30 min to eliminate the gum-like silk sericin. The cotton like precipitation were thoroughly washed with DI water and incubated in a drying oven at 37 o C overnight. Then, 1 g silk fiber was mixed with in 125 g Ajisawa's reagent containing CaCl2, ethanol, and DI water (molar ratio at 1:2:8) for 2 h at 90 o C. Impurities such as small molecules and salts were removed from the SF solution on the basis of traditional dialysis technique. Finally, the regenerated SF solution was stored at 4 °C prior to use.

Preparation of APS NPs
The synthesis of APS NPs was inspired according to previously reported biomineralization techniques. 4,5 HAuCl4 (3 mL, 6 mM) and H2PtCl6 (3 mL , 6 mM) were mixed to form a homogenous solution, which was further introduced into SF solution (6 mL, 4 mg/mL) under stirring at room temperature. Afterwards, pH value of the reaction system was adjusted to three by dropwise adding HNO3 solution (1.5 wt%). During another incubation for 24 h, the color of the mixture gradually changed from yellow to purple. At last, assynthesized APS NPs were purified through traditional dialysis technique by using Slidea-Lyzer dialysis cassette (MWCO 3500, Pierce) against 1000 mL of water for 12 h.
Brunauer-Emmett-Teller (BET) surface area and pore volume were quantified by recording the nitrogen adsorption-desorption isotherm through a Quantachrome Nova 4 1200e analyzer. The content of Au and Pt in APS NPs was determined by inductively coupled plasma mass spectrometry (ICP-MS; XSeriesII, Thermo Scientific).

Enzymatic activity of APS NPs
Glucose depletion activity: Glucose consumption was conducted by using DNS reagent.
Working solution of DNS was prepared by adding 200 mL DI water into 100 mL previous stock solution, and stored in dark. To perform the testing, various concentrations of APS dispersion (1 mL, 0~500 µg/mL) was added into 1 mL glucose solution (10 µg/mL), and reaction was allowed to proceed for 8 h at 37 o C. Thereafter, as-collected supernatant (1 mL) was diluted by two folds, followed by the addition of DNS working solution (3 mL).
At last, optical absorption intensity at 532 nm was measured using a microplate reader (Infinite M200 PRO, TECAN, Switzerland).
GSH deprivation activity: GSH consumption was measured by standard Ellman reagent.
Briefly, Ellman working reagent was prepared by dissolving DTNB (40 mg) into 10 mL PBS (0.1 M, pH = 8.0) containing 1 mM EDTA. To perform the detection, 1 mL APS solution (0~600 µg/mL) was mixed with 1 mL GSH (0.2 mg/mL), and the reaction system was under gent shaking at 37 o C for 4 h. Next, 400 µL Ellman reagent was added into the above mixture, and another incubation was continued for 30 min. After that, optical absorption intensity of supernatant was measured at 412 nm via a microplate reader. To evaluate intracellular GSH consumption, 4T1 tumor cells (1×10 4 per well) in a 96-well plate were exposed to APS NPs at different concentrations for 4 h. Then, the cells were treated by TCA solution at 4 o C for 30 min. Finally, intracellular GSH deprivation was spectrophotometrically determined based on abovementioned Ellman reagent reaction.
Fluorescence spectra (λex: 520 nm, λem: 610 nm) for DHE testing were monitored by using a microplate reader. To determine the generation of  OH, electron spin resonance (ESR) spectrum was measured by using a Bruker EMX Nano spectrometer, upon the addition of DMPO (10 mg/mL) as the trapping agent. In another aspect, H2O2 consumption by APS NPs was measured by a H2O2 quantitative assay kit (water-compatible). Briefly, 1 mL APS (40 µg/mL) was incubated with 10 mL H2O2 (10 mM) for 15 min, followed by the measurement of H2O2 level change following the manufacturer's protocol.

Cellular uptake and internalization
To analyze the endocytosis of APS NPs, Cy5.5 as a fluorescence dye was covalently labeled based on a typical strategy at first. 6 Briefly, 5 mL Cy5.5-NHS (5 mg/mL) was mixed with 10 mL APS (400 µg/mL), and the mixture was under stirring for 24 h in dark.
At last, cell uptake efficiency was calculated based on flow cytometry.
To perform dark-field imaging, 0.5 mL 4T1 cells (1×10 4 per mL) were seeded on a micro slide. After incubation overnight, the cells were treated by 0.5 mL APS NPs (200 µg/mL) for 4 h, followed by gently rinsing to remove loosely bound particles. The cells were fixed by formalin solution (neutral buffered 10%) for 2 h, and examined under a BX51 optical microscope (Olympus, Japan). 6

Biosafety of APS NPs
MTT assay: Biocompatibility of APS NPs was firstly evaluated by using humanized umbilical vein endothelial cells (HUVECs) and murine L929 fibroblasts (L929 cells).
Briefly, cells were seeded in a 96-well culture plate at an initial seeding density of 1×10 4 per well. After 12 h incubation, cells were exposed to the medium containing APS NPs at various concentrations (25~700 µg/mL). Subsequent to another incubation for 24 h, 200 µL MTT solution (250 µg/mL) was introduced into each well to replace the old medium. Blood routine test: KM mice (4~6 weeks, 25 g) were intravenously injected with saline buffer or APS NP dispersion (100 µL, 10 mg/mL). Next, the blood specimen was collected from the orbital venous plexus at day 1, 3, 7, and 14. Key blood index were measured by using an automatic hematology analyzer (NC-2600Vet, Mindary, China).

Cytotoxicity study in vitro
Cytotoxicity of APS towards 4T1 cells was firstly evaluated by using MTT assay. 4T1 cells in a 96-well plate (1×10 4 per well) were incubated overnight, and then exposed to APS NPs (25~700 µg/mL) for 24 h. Cell viability was measured on the basis of standard MTT assay as abovementioned. In another aspect, LIVE/DEAD cell staining was further conducted to directly visualize cell survival status. Briefly, 4T1 cells in a 12-well plate (1×10 5 per well) were incubated overnight, and treated by APS NPs (0~400 µg/mL) for 12 h. After costained by calcein AM and PI (following the manufacturer's protocol), the cells were examined through LSCM. To analyze cell apoptotic status after administration, 4T1 cells in a 12-well plate (1×10 5 per well) were cultured overnight, and exposed to APS NPs (400 µg/mL) for 1, 2, 4, 6 and 8 h. Then, the cells were digested by trypsin and re-suspended in PBS containing Ca 2+ and Mg 2+ . After being stained by annexin V-FITC and PI, intracellular fluorescence emission was analyzed by using flow cytometry.

Intracellular ROS generation
To MDA test: Cell membrane damage was evaluated through MDA test. 8 Briefly, 4T1 cells in a 6-well plate (3×10 5 per well) were cultured at 37 o C overnight. After undergoing various treatments, the cells were digested and incubated with TBA in boiling water for 10 min. Where "a" represents the longest dimension and "b" donates the shortest dimension of the tumor.

Biodistribution and pharmacokinetics in vivo
To understand the biodistribution of APS NPs in vivo, Cy7 To investigate the pharmacokinetics in vivo, fresh blood was withdrawn from retroorbital venous plexus, followed by homogenization. At last, the element content in blood was similarly quantified by using ICP-AES.

Photoacoustic (PA) imaging capacity
Before exploring the PA response of APS, photothermal conversion ability was first assessed in vitro. Briefly, APS NPs (3 mL) at various concentrations in a quartz vial were irradiated by a fibre-coupled semiconductor diode NIR laser (808 nm, 1W/cm 2 ) for 10 min.
A thermal imaging camera (Fluke, TiS55) was utilized to monitor the temperature elevation.
PA images of APS NP dispersion at different concentrations (15.6~600 µg/mL) were scanned by using a Vevo LAZR-X multimodal imaging system (VisualSonics Inc., Canada) in vitro. For in vivo PA imaging, tumor-bearing mice were intravenously injected with APS NPs (100 µL, 1 mg/mL in saline), and PA images of tumor region was captured at 0, 24 and 48 h post-injection. All the images were processed and quantitatively analyzed by using the same imaging system.

Antitumor effect in vivo
Upon the tumor volume reached 250 mm 3 , all the mice were allocated into 4 groups (n = 7 each group) at random, and intravenously injected with saline (100 µL), APS NPs (100 µL, 1 mg/mL in saline) or APS NPs (100 µL, 2 mg/mL in saline). Afterwards, tumor volume and body weight were dynamically recorded for 14 days. Tumor growth inhibition (TGI) was indexed according to equation (4).
where VC represents the tumor volume of saline group, and VT is donated as the tumor volume in treatment groups.
On day 14, all the mice were euthanized to harvest solid tumor and vital organs, which were further processed for histopathological analysis by hematoxylin-eosin (H&E), terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling (TUNEL), DHE and Ki67 staining.

Statistical analysis
All the results were displayed as mean ± standard (SD), as indicated. One-way analysis of variance (ANOVA) was carried out for the comparisons of multiple groups, while student t test was performed for two-group comparisons. The default thresholds for statistical significance were * p<0.05, ** p<0.01 and *** p<0.005.