Photothermal Controlled‐Release Immunomodulatory Nanoplatform for Restoring Nerve Structure and Mechanical Nociception in Infectious Diabetic Ulcers

Abstract Infectious diabetic ulcers (IDU) require anti‐infection, angiogenesis, and nerve regeneration therapy; however, the latter has received comparatively less research attention than the former two. In particular, there have been few reports on the recovery of mechanical nociception. In this study, a photothermal controlled‐release immunomodulatory hydrogel nanoplatform is tailored for the treatment of IDU. Due to a thermal‐sensitive interaction between polydopamine‐reduced graphene oxide (pGO) and the antibiotic mupirocin, excellent antibacterial efficacy is achieved through customized release kinetics. In addition, Trem2+ macrophages recruited by pGO regulate collagen remodeling and restore skin adnexal structures to alter the fate of scar formation, promote angiogenesis, accompanied by the regeneration of neural networks, which ensures the recovery of mechanical nociception and may prevent the recurrence of IDU at the source. In all, a full‐stage strategy from antibacterial, immune regulation, angiogenesis, and neurogenesis to the recovery of mechanical nociception, an indispensable neural function of skin, is introduced to IDU treatment, which opens up an effective and comprehensive therapy for refractory IDU.

Peak at 535 eV in figure S1(B) is the sodium Auger peak, and this is also confirmed by the wide XPS spectrum, (0~1200 eV, not shown here), where peaks at 1071 eV and 497 eV are displayed. The ratio between C-O and C=O calculated by O 1s spectra is roughly consistent with that calculated by C 1s for GO and pGO, while for MpGO they are not consistent, potentially because the concentration of mupirocin is relatively low, therefore, when X-ray irradiates the area containing mupirocin the C-O proportion is higher, otherwise C=O proportion higher.

Characterization of pGO
The morphology of nanomaterials was obtained by a JEM-2100F Field emission transmission electron microscope (TEM, 200 kV) and atomic force microscope (AFM). X-ray diffraction (XRD) was measured on a D/max-2550 XRD system (parameters: Cu Kα, λ = 1.54 Å, 40 mA, and 40 kV). Hydrated particle size was obtained by Malvern nanosize. X-ray photoelectron spectroscopy (XPS) spectrum was conducted on an 250XI (Thermal Scientific, US). Raman spectrum was recorded by LabRAM HR Evolution (Horiba, France).

Characterization of hydrogels
The morphology of hydrogels was obtained by high resolution Zeiss field emission scanning electron microscopy. The chemical structure of hydrogel is synthesized by

NIR light-triggered Mup release from MpGel and pGel/M
Release of Mup from MpGel and pGel/M was determined in DI water. 1 mL of DI water (above) was placed in a 24-well cell culture plate containing 0.5 mL of MpGel or pGel/M (below). MpGel or pGel/M was irradiated using an 808 nm NIR laser (0.5 W/cm 2 ) for 10 min followed by an interval of 30 min and the procedure was repeated three times. Then, a liquid chromatography-mass spectrometry system (AB Qtrap 6500) was used to analyze the above solution to determine the release amount of Mup. In comparison, MpGel or pGel/M without NIR light irradiation was set as a control.

CCK8 assay
Divide the experiment into four groups：Gel, pGel, MpGel, MpGel-NIR(3min). L929, a Mouse fibroblast cell line, with an initial density of 2 × 10 5 /mL were seeded on hydrogels and cultured in a 5% CO2 incubator at 37 °C. The cell culture medium was replaced every other day. Cell proliferation on the hydrogels was evaluated by cell counting kit-8 reagent (CCK8, Dojindo Molecular Technologies) after 1, 4, 7 days of culture.

Live-death test on Schwann Cells cultured on pGel.
After 3 days of culture in the growth medium, the Schwann Cells were rinsed twice with 1×PBS pre-warmed at 37°C.The cytotoxic effect of the hydrogel was evaluated with Calcein-AM/PI (Dojindo Molecular Technologies) followed by observation with Confocal laser microscope (CLSM).

In vivo animal experience
The C57BL/6 mice were obtained from Beijing Hufukang Biotechnology Co., LTD.
After one week of adaptive feeding, the mice were administered STZ (100 mg/kg) with 0.01M citrate buffer as a vehicle once intraperitoneally. The random blood glucose of each mouse was monitored every two days and when it reached 11.1 mmol/l, the type I diabetes model was considered to be successfully established.
Following anesthetization, one full-thickness skin wound with a diameter of 10mm was created on the dorsum and 10 μl of bacteria liquid (Staphylococcus aureus, 1× 10 7 CFU) was applied to the wound. After that, the mice were randomly divided into ten groups (n=20) and

ELISA assay
To evaluate the inflammation on the 3rd day of treatment, we detected the level of IL-6 and TNF-α in the wound. Wound tissue was homogenized and the supernatant was collected for ELISA assay with commercial ELISA kits (Mouse TNF-α and IL-6 ELISA KIT, Solarbio). The experiments were carried out according to the manufacturers' instructions.

Antibacterial activity in vivo
To evaluate the antibiotic activity in vivo, the wound tissue was sampled for bacterial culture on the 3 rd , 6 th , 9 th , and 12 th day after the treatment. After overnight culture, 10 μl of the diluted bacterial solution was spread on the solid medium and incubated for instructions.

Histological and Immunofluorescence staining and analysis
Histological and immunofluorescence staining was conducted to evaluate the tissue regeneration of the wound. On day 12, the tissue of the wounds was embedded and sectioned. Hematoxylin and eosin were used for H&E staining and the Trichrome Stain Kit (Sigma-Aldrich) was used for Masson staining. The procedures were following the manufacturers' instructions. The skin thickness, hair follicles, blood vessels, and collagen deposition were observed and recorded under Optical Microscope (OM). In addition, the major organs of each mouse on the 28 th day were collected for H&E staining and observed to exclude the biotoxicity of the materials.

Western Blot
Western blot was performed as described previously 4 . On day 28, the tissue of wounds was sampled and lysed with Tissue Protein Extraction Reagent (Thermo Fisher). The expression of nerve growth factor (NGF) protein was quantified by the western blotting assay performed in the standard fashion. Anti-NGF antibody (Abcam) was used. Western blot bands were quantified with NIH ImageJ.

Toxicology analysis
To evaluate the biotoxicity of the materials, the major organs of each mouse on the 28th day were collected for H&E staining and observed under OM. In addition, the blood samples of each mouse were obtained for the hematology analysis and blood biochemistry assay.

Animal behavioral tests
Pain threshold (von Frey sensitivity) was tested with a calibrated set of von Frey filaments (Stoelting, Wood Dale, IL56). The animals were acclimated to the environment before the von Frey test for 15 minutes. Each von Frey hair was applied at the wound healing site for 3 s and tested 3 times at an interstimulus interval of 20s.
The pain threshold was determined by the value of strength when mice flee quickly.
The test was performed in a blinded manner, where the investigator did not know the identity of the animal and the study group.

Statistical Analysis
Data within this study were presented as mean ± SD. The sample size was n ≥ 3. No data pre-processing method was utilized. All the statistical analysis of significance test were performed among three or more groups, which were analyzed by using one-way ANOVA or nonparametric test. If data showed normal distribution and variance homogeneity, Tukey's post hoc test would be performed; if data was normally distributed but variance was not homogeneity, the Welch's correction would be performed followed by Games-Howell post hoc test; if the data was not normally distributed or had significant variance inhomogeneity, Kruskal-Wallis nonparametric test would be conducted instead of one-way ANOVA. A rough Pearson correlation analysis was made in Figure S15, where a point represents a pair of data (the average number of blood vessels and the average intensity of TREM2 fluorescence) from the same experiment group (n = 4, here, take one group of mice as the statistical individual). The statistical analysis was conducted by IBM SPSS Statistics 22.0.