A novel multifunctional microneedle patch for synergistic photothermal- gas therapy against maxillofacial malignant melanoma and associated skin defects | Journal of Nanobiotechnology

Materials and reagents

Na₂[Fe(CN)₅NO]·2H₂O and [CH₃CH(OH)COO]₂Fe·xH₂O were purchased from Sigma-Aldrich, USA. Carboxymethyl cellulose (SCMC; MW. ~ 90,000, #C104983), crystal violet, and rhodamine Bwere purchased from Aladdin, China. Phosphate buffer solution (PBS), minimum essential medium (MEM), Roswell Park Memorial Institute (RPMI) 1640 medium, trypsin, penicillin, streptomycin, and fetal bovine serum (FBS) were purchased from Gibco, USA. Cell Counting Kit-8 (CCK-8), antibody dilution buffer, radioimmunoprecipitation assay (RIPA) lysate, Total Nitric Oxide Assay Kit, and DAF-FM DA were purchased from Beyotime, China. A cyanine-3 dye-conjugated TdT-mediated dUTP nick end labeling kit (TUNEL), distilled RNase-free water, a bicinchoninic acid (BCA) protein assay kit, TRIzol RNA isolation reagents, RevertAid first strand cDNA synthesis kit and polyvinylidene difluoride membranes were purchased from Thermo Scientific, USA. 2X SYBR Green Pro Taq HS Premix qPCR kit was purchased from Accurate Biology, China. Calcein AM/PI Double Staining Kit, paraformaldehyde (PFA, 4%), and hematoxylin–eosin (H&E) Staining Kit were purchased from Solarbio, China. One light-curing resin component was purchased from Friedrichsdorf, Germany. A PDMS microneedle mold and negative pressure centrifugal pump were purchased from Engineering for Life, China. The SDS-PAGE kit was purchased from Bio-Rad, USA. Polyvinylidenedifluoride membrane (PVDF) was purchased from Millipore, Bedford, MA, USA. NcmBlot Rapid Transfer Buffer, NcmBlot Blocking Buffer, and Ncm Enhanced Chemiluminescent (ECL) High were purchased from NCM Biotech, China. The primary antibodies used were: hypoxia inducible factor-1 (HIF-1a), vascular endothelial growth factor (VEGF), AKT, p-AKT, heat shock protein-70 (HSP-70), heat shock protein-90 (HSP-90), Caspase-3, β-actin and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), all of which were purchased from Immunoway USA. The secondary antibodies were purchased from Immunoway in the USA. Other antibodies, Ki67 and CD31, were purchased from Abcam, USA. Matrigel was purchased from ABW, China. All reagents were used without further purification.

Preparation of SNP nanoparticles and fabrication of SNP-Fe@MN patches

SNP nanoparticles were firstly synthesized by dissolving in deionized water and ethanol totally, and then the dispersion of the particles was proceeded by ultrasonic vibration. After being kept at 80 °C for 24 h, the freeze-dried nanoparticles were obtained by lyophilization (Creatrust, China). SNP-Fe@MN patches were fabricated by a one-step negative pressure centrifugation casting method using polydimethylsiloxane (PDMS) molds (Engineering For Life, China). During the suction exhaust process, 250 μL 8% w/v sodium carboxymethylcellulose (SCMC) solution with a certain concentration of drugs was added into PDMS molds, centrifugated at − 0.08 MPa in the suction pump (Engineering For Life, China) for 3 min, then excess bubbles and solutions were discharged. The above steps were repeated 3 times, followed by 24 h of ambient drying at room temperature. Drug gradient concentration for MNs used in this study were 0.01, 0.03, 0.05, 0.10 and 0.15 g mL⁻¹ sodium nitroprusside (SNP) for each type of SNP-Fe@MNs respectively, mixture of SNP nanoparticles and [CH3CH(OH)COO]₂Fe∙xH₂O powders were under the molar ratio of 1:1 in deionized water, and non-drug for SCMC@MNs. Subsequently, MNs were obtained after discreetly demolding. All fabrication and storage processes were performed in dark or low light and dry conditions to avoid the degradation of MNs. The diameter of the MN mold was 17.5 mm in the bottom base with a 2-mm fillister. The microneedles were arrayed in a whole circularly arranged sequence, with a total number of 385.

Characterization of SNP-Fe@MNs Patches

Firstly, fluorescence stereomicroscopy was performed to characterize the surface morphology and overall view of drug-loaded and drug-free MNs, respectively (AxioZoom.V16, Zeiss, Germany). Next, in order to evaluate the puncture ability of MNs towards tissues and the melting and absorption rate of MNs, scanning electron microscopy (SEM) was conducted to determine the MNs surface topography before and after the MNs were penetrated into ex vivo mouse skin in 1, 5, 15, and 30 min (Flexsem 1000, HITACHI, Japan). Meanwhile, observations on the puncture ability of the MNs after pressing for 1 min in vivo were implemented at 0, 2, 5, 10, 15, and 20 min. Then, for the sake of simulating of transdermal drug release process, rhodamine B (MW. = 479.01) was used via fluorescent dye to simulate SNP (MW. = 297.95) due to their rather similar molecular weights. Rhodamine B-loaded MNs (RB-MNs) were prepared by the same casting step as previous with only 0.03 g mL⁻¹ Rhodamine B-loaded and SCMC solution to form MNs. The morphology of R-MNs patch was characterized by fluorescence stereomicroscopy (AxioZoom.V16, Zeiss, Germany). After that, RB-MN’s patches were pressed on mouse skin for 1 min, the overall top vertical view of the skin treated by R-MNs was captured, and the treated skin was sliced along the row of holes for cross sections that were captured by a fluorescence stereomicroscope to show the radial diffusion of rhodamine B at every hole and reveal the penetrability generated by MN tips. In addition, after mixing the nanoparticles of equal molar concentrations of SNP and ferrous lactate, one group was activated by ultraviolet light, and the other group was stored in the dark. After freeze-drying (Creatrust, China), the two groups of mixed nanoparticles were obtained, and the corresponding characteristic peaks were detected by FTIR (Thermo Scientific Nicolet iS50, USA) and XRD (Bruker D8 ADVANCE, Germany), respectively to ascertain the successful preparation of Prussian blue analogues. Additionally, 0.1 M SNP solution and 0.1 M FeCl₂ solution were mixed evenly in a 1.5 mL EP tube at a ratio of 1:1. The obtained solution was vibrated by ultrasonic for 30 min, and the absorption spectrums of the solution were recorded by a UV–Vis spectrophotometer before and after 10-min UV irradiation.

The photothermal properties and NO releasing behavior of SNP-Fe@MNs

SNP-Fe@MNs loading with 0.01, 0.03, 0.05, 0.10 and 0.15 g mL⁻¹ SNP were irradiated by ultraviolet (UV) light for 1 min then put in 48-well plates. Next, all MNs were irradiated by an 808 nm NIR laser (808-5W, BOT, China) at a power density of 1.0 W cm⁻² and 1.5 W cm⁻² separately for 10 min. Furthermore, 0.03 mL⁻¹ SNP-Fe@MNs were put in a 48-well plate irradiated by an 808 nm NIR laser at various power densities of 1.0, 1.5 and 2.0 W cm⁻² for 10 min. Then, the 0.03 mL⁻¹ SNP-Fe@MNs were cyclically heated—cooled at 1.5 W cm⁻² by regularly turning the laser device on/off. All the photothermal evaluation experiments in the study were conducted under the monitor of an infrared camera (628c, Fortric, China).

To measure actual rate of NO releasing, SNP-Fe@MNs loading at 0.01, 0.03, 0.05, 0.10, and 0.15 g mL⁻¹ SNP were immersed in 1 mL of PBS with or without 1 min of irradiation by UV light. The solutions were then extracted to 96-well plates, and levels of NO at 1, 5, 15, 30, 60, 90 min and 12 h were measured by the Griess reaction using the microplate reader at 540 nm (Biotek, USA). To determine the influences of NIR on the releasing rate of NO, the obtained SNP-Fe@MNs loading 0.03 g mL⁻¹ SNP and Fe²⁺ in our work were immerged in 1 mL PBS and separated into two groups named as NIR(−) and NIR(+) respectively. Group NIR(+) were irradiated with NIR (808 nm) for 1 min. Then, the NO concentrations two groups were detected with Griess Reagent (Beyotime, China) at 1-, 15-, 30-, 60-, 90-min and 12 h under dark environment. To assess the intracellular NO level in B16 and HUVEC, cells were washed three times with PBS. DAF-FM DA fluorescence probes were added and incubated for 45 min at 37 °C. The fluorescence at excitation and emission wave lengths of 495 nm and 520 nm was then measured, accordingly. Finally, each fluorescence intensity was analyzed by image J software.

In vitro evaluation on the antitumor performance of SNP-Fe@MNs

To verify the biosafety of the MN materials, murine fibroblast (L929, 2 × 10⁴) (Procell, China) cells were cultured in 24-well plates and cocultured on SNP-Fe@MNs loading 0.01, 0.03, 0.05, 0.10 and 0.15 g mL⁻¹ SNP and SCMC@MNs for 24 h and 48 h in dim, respectively, then were washed three times by PBS. Next, 600 mL MEM with 60 mL of cell counting kit-8 (CCK-8) was added to each well then cultured in the cell incubator for 2 h. Finally, 100 mL of the solution was transferred to a 96-well plate and read at OD490 by the microplate reader and measured by the following relative cell viability equation:

For further biocompatibility confirmation, human umbilical vein endothelial cells (HUVECs, 2 × 10⁴) (Procell, China) were seeded for 24 h with SCMC@MNs, SNP-Fe@MNs-NO(+), and SNP-Fe@MNs-NO(−), with the latter one pre-freeze-dried NO after UV irradiation during production. Besides, group of SNP-Fe@MNs-NO(+) was set to simulate the process of releasing NO in the later stage of wound restoration. MNs were applied 60 min after UV irradiation according to the previous gas releasing profile to get rid of burst release of NO. Then each group was stained with the Calcein-AM/PI Kit and measured by image J.

To explore whether ferroptosis was involved in the therapeutic process, Murine melanoma B16 cell line (2 × 10⁴) (Procell, China) were incubated for 24 h with SCMC@MNs, 0.03 g mL⁻¹ SNP-Fe@MNs-UV(−) and the individuality of the same amount as former 0.03 g mL⁻¹ SNP-Fe@MNs-UV(−), which were labelled as Fe@MNs together with SNP@MNs. Then the detection of relative cell viability by CCK-8 were conducted.

Furthermore, B16 (2 × 10⁴) were seeded in 24-well plates individually and cocultured for 24 h in groups that were divided as follows: control (processless), SCMC@MNs, UV-cell (simulate UV irradiation for 1 min when no material existed), UV-SNP-Fe@MNs-NIR(−)-NO(+), UV-SNP-Fe@MNs-NIR(+)-NO(−), and UV-SNP-Fe@MNs-NIR(+)-NO(+). SNP-Fe@MNs indicated for the 0.03 g mL⁻¹ group, as the following represents the same meaning. Specifically, groups of NO(−) were pre-freeze-drying to remove NO, moreover, groups of NIR(+) were exposed to an 808 nm NIR laser at 1.5 W cm⁻² while being monitored by the infrared camera until the temperature reached around 47 °C and was held for 15 min by reducing the output power of the laser device gradually. Then, all of the groups were cultured in the cell incubator (37 °C, 5% CO₂) (Thermo Scientific, USA) for another 30 min in order for final CCK-8 detection. Next, B16 (2 × 10⁴) were cocultured for 24 h with SNP-Fe@MNs-NIR(−)-NO(+), SNP-Fe@MNs-NIR(+)-NO(−) and SNP-Fe@MNs-NIR(+)-NO(+), then were stained with the Calcein-AM/PI Kit and counted by image J. Moreover, the B16 migration condition using medium extracts of different MNs was presented by scratch testing and stained with crystal violet at day 2.

Then, Western blot analyses were conducted with the same group and the same condition as stated above, except for adjusting the photothermal time to 8 min and 15 min so as to explore the optimal time and related mechanisms. After extracting total proteins with RIPA buffer, the concentration was determined by the BCA protein assay kit. Samples each with 15 μg of protein were separated in SDS-PAGE gels, as each group was loaded together with representative front lysates from the Control group for standardization, then rapidly transferred to PVDF membranes. Next, the membranes were blocked with NcmBlot blocking buffer for 10 min in the ambient, with further trimming into as narrow strips according to the specific molecular weight of the protein of interest on the basis of the given markers. Later, the strips were incubated overnight at 4 °C with the appropriate primary antibodies: Caspase-3, HSP-70, HSP-90, AKT, and GAPDH, diluted to 1:2000. After being washed in Tris-buffered saline and Tween (TBS-T) for 3 times, certain stripes were incubated with secondary horseradish peroxidase (HRP)-labeled goat anti-rabbit IgG diluted 1:5000 for 1 h in the ambient. Protein bands were determined by transilluminator and imaging analysis system (Bio-Rad, USA) with NcmECL imaging while the densities of which were calculated by Quantity One software (Bio-Rad, USA).

In vivo evaluation on the antitumor therapeutic performance of the SNP-Fe@MNs

All animal experiments were approved by the Laboratory Animal Welfare and Ethical Committee of Xian Jiaotong University (No. 2022-1606). The B16-bearing Balb/c nude mouse (6-week-old female) model was utilized for research on the in vivo antitumor efficiency of MNs. 80 × 10⁴ B16 cells in 100 µL PBS were injected on the midback of mice subcutaneously. On day 6, when the volume of the tumor reached around 100 mm³, full thickness wounds slightly wider to adhere to tumor body were surgically produced, with no tumor tissue being excised during the whole procedure. After surgical exposure, the wound was then penetrated and covered with different types of MNs. During the therapy, the patches were fixed with a transparent membrane dressing on the surface.

For the NIR (+) group, an 808 nm laser (1.50 W cm⁻², 15 min) was applied after the utilization of the MN patches from days 6 to 8. During the three consecutive treatments, the ranges of temperature in the wound area were recorded by the infrared camera. The volume of melanoma is measured by Caliper daily and calculated as \({\text{V}}=\frac{{{\text{ab}}}^{2}}{2}\), where a and b represent the longest and shortest diameters. Meanwhile photographs were taken on day 6, 9, 14, 17 and 20. The weight of mice was recorded every other day, and the mortality remained 0% during the treatment. At the end of the therapy on day 20, the mice were photographed and then sacrificed. The tumor tissues were carefully dissected, weighed, and photographed, then sliced and evaluated by H&E, Ki67 immunohistochemical, and TUNEL staining assays. Besides, the main organs (skin, heart, lung, liver, spleen and kidney) were analyzed by H&E staining assays.

In vitro angiogenic differentiation and proliferation of SNP-Fe@MNs

Angiogenic assays were carried out according to the manufacturer’s instructions, as Matrigel was diluted in 1640 medium with SCMC@MNs, SNP-Fe@MNs-NO(−), and SNP-Fe@MNs-NO(+), then added to a 96-well plate, and incubated at 37 °C for 4 h for solidification. Significantly, during the whole regeneration experimental period, a group of microneedles that released NO were set up to simulate the process of releasing NO in the later stage for wound repair, thus, MNs were applied 60 min after UV irradiation according to the previous NO release curve to avoid the burst of gas. HUVECs, which were previously cocultured with different groups of MNs for 5 days, were then detached by digestion with 0.25% trypsin. HUVECs (1.5 × 10⁴) were later seeded on the mixture of Matrigel, and the 96-well plate was then incubated at 37 °C with 5% CO₂ and tube formation was observed at 30 min and optimum 4 h. Furthermore, HUVECs were cocultured with different MNs, and the CCK-8 assay was performed at the end of days 1 and 2. Subsequently, the migration ability of HUVECs were measured through scratch assay, as the cells were incubated with the medium extracts of different types of MNs, and captured at 18 h and 24 h. The migration ratio was analyzed by Image J and measured by the following equation:

To further explore the mechanism of proliferation, a real-time quantitative reverse transcription polymerase chain reaction (RT-qPCR) assay was utilized to analyze the expression of angiogenic genes in HUVECs, which were VEGF, HIF-1a, and endothelial nitric oxides (eNOs). After being incubated with medium extracts of three types of MNs as described above in the 6-well plates for 5 days, HUVECs were harvested with the TRIzol reagent on day 6 after the medium was removed. Then 1 μg total RNA was extracted according to the instructions of the RevertAid first-strand cDNA synthesis kit for reverse transcription, and a 2× SYBR Green Pro Taq HS Premix qPCR kit was used for the reactants’ preparation. The related primer sequences were listed in Additional file 1: Table S1, which were synthesized by AUGCT (Beijing, China). RT-PCR analysis was performed using a Multicolor Real-Time PCR Detection System (iQ5, Bio-Rad, USA). Relative gene expressions of interest were determined using β-actin as the housekeeping gene for normalization by using the 2^−ΔΔCt method. Meanwhile, Western Blot analysis were conducted as the processes stated above. The primary antibodies were HIF-1a, VEGF, AKT, p-AKT and β-actin (all diluted 1:2000).

In vivo skin tissue regeneration capability of SNP-Fe@MNs

All animal experiments were approved by the Laboratory Animal Welfare and Ethical Committee of Xian Jiaotong University (No. 2022-1606). The Sprague–Dawley (SD) rat (7–8 weeks old, female) was anesthetized with isoflurane, depilated, disinfected, and drilled a round 17-mm-diameter hole-shaped full-thickness cutaneous back defect. For avoiding the wound from stretching or shrinking, an annular silicone ring was sewn adhesive to the defect, then groups of 17.5-mm-diameter MN patches were therefore attached to cover the wound surface. Subsequently, these wounds were fixed by transparent membrane dressings that were wrapped with sterile gauze to prevent gnawing and photographed at days 0, 4, 8, 12, and 16. The certain degree of wound closure was fitted into composite image, analyzed by image J and measured by the following equation:

By the end of the research at day 16, the wound tissues were sliced for further wound healing analysis via H&E, Masson, and CD31 immunohistochemistry staining assays.

Statistical analysis

All experiments involved in this paper were implemented at least twice with 3–8 replicates and the data were presented as the mean ± standard deviation (SD). Error bars were considered as the SD of the mean of each independent samples among one certain experiment. The values of *P < 0.05, **P < 0.01 and ***P < 0.001 were statistically significant. Statistical significance were evaluated by the unpaired two-tailed t-test by GraphPad Prism Software and Origin Software.