Silk fibroin was obtained by degumming Bombyx mori silkworm cocoons as previously described (28). Briefly, raw cocoons were cut into small sections and washed thoroughly in nanopure water (resistivity 18.2 MΩ-cm). Degumming was performed by boiling washed cocoon pieces in a solution of Na₂CO₃ in nanopure water (0.005 g/mL, 0.5% w/v) for 30 minutes with slow stirring. Degummed material was washed three times in nanopure water for 15-20 minutes in each wash, pulled into thin sections, and dried overnight at room temperature. The dried silk fibroin material was dissolved in 9.4M lithium bromide at 60°C (2 g silk fibroin per 10 mL solution) with stirring by adding the silk piece wise. Once dissolved, the solution was continuously stirred for an additional 4 hours at 60°C. Dissolved silk solution was clarified through cotton filter paper, transferred to 3.5 kDa dialysis membranes, and dialyzed against nanopure water for 72 hours at 4°C. Concentration and weight percentage of resultant silk fibroin solution was determined by back-calculation of the mass of dried, known volume of silk in a pre-weighed weigh boat. Purified silk composition was characterized by FT-IR on a Thermo Nicolet 6700 spectrometer.

Films were cast in a 100 mm petri dish by dispensing 4 or 8 mL of an 8% w/v solution followed by overnight solvent evaporation at room temperature. Rheometry of 10x15-mm films was performed by Dynamic Mechanical Analysis on a Discovery HR 30 rheometer (TA Instruments) with a frequency sweep from 1 Hz to 10 Hz and a strain of 0.1% to measure storage and loss moduli. Round films were produced by using a 6-mm hollow punch tool. Films were made insoluble by placing in an autoclavable pouch and autoclaving in a 20-minute steam/heat cycle. Films were measured using a micrometer-type caliper (Rexbeti) to confirm thickness prior to use.

Silk films were sputter coated with 1.5-2nm gold coating on a Cressington 108 sputter coater (exposure 120s with an argon gas environment). High-resolution field emission SEM was performed on an Auriga (Zeiss) scanning electron microscope equipped with a high-resolution Gemini Field Emission-SEM column and Schottky thermal field emitter operating at 5.00 kV HV and a monopole magnetic immersion final lens. Images were collected en face and in cross-section after cutting the film with a scalpel.

Autoclaved films were placed in 1X PBS and incubated for up to 40 days at room temperature. Film thickness was measured with a micrometer-type caliper (Rexbeti) and degradation was determined by protein quantity in solution at 40 days using a BCA protein assay (Pierce).

A438079 was docked to the published structure for P2X7 in the closed, apo state (PDB 5U1L) (29) using Webina, a webserver for Autodock Vina (30). Docking results were visualized in Chimera X version 1.3 (31). Results were validated against prior docking reported by Allsopp et al. (32).

A438079 was purchased from Santa Cruz Biotechnology and dissolved in a solution of 50% DMSO in nanopure water to produce a 25 mM stock solution. Aliquots were prepared and stored at -20°C. Two-fold serial dilutions from 2.5 mM to 0.04 mM were generated in 1X PBS and dispensed in 50 µL volumes in a UV-transparent 96-well plate. Absorbance spectroscopy was performed on a Biotek Synergy H2 plate reader from 200-300 nm with a 2 nm resolution. Diagnostic peaks were observed at 226 nm and 258 nm.

Insolubilized silk films were incubated in solutions of 1.25 mM A438079 for 24 hours. The amount of the small molecule inhibitor drug loaded was evaluated by absorbance spectroscopy of the post-loaded solution versus solution that was not incubated in the presence of a film. Release studies were performed by incubating a loaded film in 1X PBS at room temperature with shaking. At regular intervals, the PBS soaking solution was removed and analyzed by absorbance spectroscopy. The films were placed in fresh solutions of PBS to generate continuous sink conditions. Cumulative release was calculated and fit to a two phase burst release model in GraphPad Prism with the “two phase association” equation (33):

J774DUAL cells (Invivogen) were cultured in DMEM with 10% FBS containing 1% penicillin/streptomycin and additional supplementation of selective antibiotics according to manufacturer’s procedure (5 µg/mL Blasticidin, 100 µg/mL Zeocin^®). Cells were fed twice a week and passaged by cell scraping. Cell viability during passaging was evaluated by Trypan Blue staining using an EVE™ Plus Automated Cell Counter (NanoEnTek) to ensure 95% or greater viability prior to use.

On day 1, J774DUAL cells were seeded at 10⁵ cells/well in a tissue culture-treated 96-well plate and incubated overnight at 37°C/5% CO₂ in a humidified incubator in complete DMEM without selective antibiotics. On the same day, an ELISA plate (R&D systems, DY008) was coated with capture antibody to IL-1β (R&D Systems, DY401) according to manufacturer’s instructions, sealed with plate film, and incubated at room temperature overnight. On day 2, cells were either left untreated or primed with 1 µg/mL E. coli LPS (O111:B4; “first signal”) in 100 µL final volume fresh DMEM (no selective antibiotics) and returned to the incubator for 4 hours. Concurrently, A438079-loaded films were incubated in 220 µL fresh DMEM (10% FBS, 1% pen/strep, no selective antibiotics) at room temperature in a 1.5 mL microcentrifuge tube. ELISA plates were washed and blocked during the priming period according to manufacturer’s protocol. During the final 30 minutes of priming (3.5 hours after initial treatment with LPS), 50% (i.e., 50 µL) of the incubation medium was replaced with either fresh DMEM or film-conditioned medium. The plate was returned to the incubator for the final 30 minutes of priming. To stimulate inflammasome activation (“second signal”), cells were treated with 3 mM final concentration of extracellular ATP (3 µL of a 100 mM stock solution prepared in fresh DMEM mixed into 100 µL final culture volume) and returned to the incubator for 45 minutes. After 45 minutes, complete (100 µL) cell supernatants were collected, transferred to the blocked ELISA plate along with recombinant IL-1β standards, sealed with plate film, and incubated overnight at 4°C. On day 3, the ELISA plate was washed, probed with detection antibodies, and developed according to manufacturer’s procedure. The developed ELISA plate was read on a Biotek Synergy H2 plate reader at 450 nm with background subtraction at 540 nm. Data analysis using 4-parameter logarithmic regression was performed in GraphPad Prism.

Cells were treated as in Section 2.8 above for ELISA analyses. After 45 minutes of ATP treatment, the culture medium was aspirated, and cells were fixed in 2% formaldehyde (freshly prepared from paraformaldehyde in 1X PBS) for 20 minutes at room temperature. Cells were permeabilized with 1X TBS containing 0.2% Tween-20 (TBST) for 20 minutes at room temperature and blocked in 5% BSA prepared in 1X TBST for 1 hour at room temperature. Cells were incubated overnight at 4°C with rabbit polyclonal antibody to ASC (AL177, 1:200 dilution, Adipogen) in 5% BSA/TBST. The following day, cells were washed with TBST and incubated with CF568-conjugated donkey anti-rabbit secondary antibody (1:500 dilution, Biotium) and iFluor 488-conjugated Phalloidin (1:1000 dilution, AAT Bioquest) for 2 hours at room temperature protected from light. Cells were washed with TBST and incubated for 30 minutes with 10 µg/mL DAPI (Abcam) in 1X PBS at room temperature protected from light. Cells were washed with PBS and submerged in 250 µL Fluoromount G (Thermo-Fisher) and stored at 4°C protected from light until imaging. Cells were imaged on a Nikon AXR confocal (22.1 µm pinhole) mounted to a Ti2 base with a Plan Apo 60× Oil NA 1.42 objective lens using 408, 488, and 561 nm laser lines paired to DAPI (429-474 nm), AF488 (503-541 nm), and AF568 (571-625 nm) emission windows, respectively, with at 1024x1024 resolution setting with 2X frame averaging. Images were collected with NIS-Elements AR software (ver. 5.41.01 build 1709) and analyzed using FIJI/ImageJ (ver. 2.14.0/1.54f).

All animal procedures were approved by the Institutional Animal Care and Use Committee of Arizona State University under protocol #21-1830R. Mice were purchased from the Jackson Laboratory and kept on a standard 12/12 light-dark cycle in specific pathogen-free housing conditions and given food and water ad libitum. Full-thickness wound healing was performed in male and female 12-week old obese, diabetic “db/db” mice (BKS.Cg-Dock7^m +/+ Lepr^db /J; JAX strain code 000642) as previously described (34, 35). Briefly, non-fasting blood glucose was measured within one week of surgery by saphenous vein collection and confirmed to be >400 mg/dL using a glucometer. Mice were anesthetized and a 1x1-inch midline intrascapular area between the base of the neck and apex of the spine was shaved and sterilized with successive washes with alcohol and chlorhexidine gluconate solution. A 6-mm full-thickness biopsy punch was performed and wounds were either treated directly with 20 µL saline, 100 µM A438079 in 20 µL saline, empty silk film, or silk film loaded with A438079. Treatments were applied a single time during the study. A silicone splint (14mm OD x 7mm ID x 0.5mm thick, Grace Biolabs) covered with Tegaderm occlusive dressings (3M) were affixed using cyanoacrylate glue (Krazy Glue) and six interrupted sutures (4-0 black Ethilon monofilament with a FS-2 reverse cutting needle; Ethicon, Inc.). Mice were returned to single housed cages to prevent removal of splints by cage mates and monitored daily until euthanasia at 10 days post-wounding. Wound area (planimetry) after splint removal was documented by digital photography using a mobile phone and analyzed with calibration in ImageJ/FIJI.

Wound tissues were collected at day 10 post-wounding and fixed in 10% neutral-buffered formalin. Tissues were dehydrated through graded alcohol into xylene followed by paraffin perfusion. Tissues were embedded into paraffin blocks and 5-6 µm sections were captured onto positively charged glass slides. Sections were rehydrated and epitope retrieval was performed by incubation in sodium citrate buffer (10 mM, pH 6.0 with 0.05% Tween-20) at 60°C overnight. Sections were blocked with 5% bovine serum albumin (BSA) in TBS/0.1% Tween-20 for 1 hour at room temperature and incubated overnight with goat anti-mouse IL-1β primary antibody (R&D Systems, AF-401-NA, 1:100) in 5% BSA in TBS/0.1% Tween-20 at 4°C overnight. Sections were washed with TBS/0.1% Tween-20 and peroxide quenching was performed with 3% hydrogen peroxide in PBS for 15 minutes at room temperature followed by thorough washing. Sections were incubated in horseradish peroxidase (HRP)-conjugated donkey anti-goat secondary antibody (Jackson Immunoresearch #705-035-147, 1:500) for 2 hours at room temperature followed by thorough washing. Sections were developed with ImmPACT DAB substrate (Vector Labs #SK-4105) for 5 minutes, counterstained with hematoxylin (Gill No. 2, #GHS232, Sigma Aldrich), dehydrated, and mounted with CytoSeal XYL (Thermo Fisher). Slides were scanned at 40X magnification on an Olympus VS200 Slide Scanner and quantified using QuPath software v0.4.3 (36).

We sought to generate a tunable thickness, insoluble silk fibroin film as a long-lasting wound dressing matrix. We enriched silk fibroin from Bombyx mori silkworm cocoons and characterized the polypeptide by FT-IR spectroscopy ( Figure 1A ) with expected amide I peak at 1650, amide II peak at 1520, and amide III peak at 1240 cm^-1 (37). We cast 100-mm diameter sheets of silk fibroin in petri dishes and generated films by solvent evaporation at room temperature overnight. Using a hollow punch tool, we fabricated 6-mm diameter wound dressings ( Figure 1B ), which we made insoluble by autoclaving (38). Autoclaving is thought to increase the beta sheet content in silk fibroin, which makes them resistant to dissolution in aqueous solutions (39). We performed scanning electron microscopy (SEM; Figure 1C ) on insolubilized films and observed that while the exposed surfaces of the films were relatively smooth and without notable features, the cross-section of the films indicated a highly complex network structure. Dynamic mechanical analysis of the films ( Figure 1D ) indicated that the films exhibited durable elastic properties with high storage and loss moduli of ~3300 MPa and ~200 MPa, respectively, with a low tan(δ) of ~0.06, which indicated elastic nature of these films. Film thickness was tuned by modulating the amount of silk solution used to cast the initial film sheet, with 4 mL of 8 w/v% silk solution in a 100-mm petri dish resulting in films of ~60-µm in thickness, and 8 mL of 8 w/v% silk solution resulting in films of ~120-µm thickness ( Figure 1E ). We evaluated the passive degradation of the films in PBS over a period for 40 days at room temperature and found that 60-µm films cast from 4 mL solutions did not exhibit any swelling behavior (-2% swelling ratio), while 120-µm films cast from 8 mL solutions exhibited minor swelling behavior to ~140-µm (16.7% swelling ratio). In both cases, BCA protein assay of the incubation solution indicated no presence of dissolved protein (data not shown) and thus no passive degradation over 40 days. Taken together, we generated insoluble and durable, tunable thickness silk fibroin films appropriate for wound dressing applications (40, 41).

A438079 is a water-insoluble, selective antagonist of the P2X7 purinergic receptor that allosterically binds in a region of the ion-permeating channel in the left flipper of the extracellular domain in close proximity to the previously described residues F88, F92, T94, F95, and F103 ( Figure 2A ) (32, 42). This allosteric site is ideal to facilitate druggability of the P2X7 receptor, as allosteric sites are commonly more selective, allow lower target-based toxicity, fewer side effects, and accessible physicochemical properties (43, 44). We previously investigated the inhibitory function of A438079 and reported a potent ability to ameliorate extracellular ATP-evoked potassium and calcium flux and subsequent inflammasome activation in mouse macrophages (26). We performed UV-VIS spectrophotometric analysis of A438079 in PBS ( Figure 2B ) and identified two diagnostic absorption peaks at 226 nm and 258 nm which exhibit linear response from 40 µM to 2500 µM ( Figure 2C ). We passively loaded 60-µm and 120-µm films with A438079 overnight in a 1.25 mM loading solution in PBS and measured release over 72 hours into PBS. Kinetics analysis indicated a release profile fitting to a two-phase burst release model (Equation 1), with release amount proportional to film thickness ( Figure 2D ). A concentration of approximately 100 µM and 200 µM in 100 µL PBS was measured in the first 6 hours from 60-µm and 120-µm films, respectively. Release continued at ~40 µM and ~60 µM per day for the first 2 days for 60-µm and 120-µm films, slowing to ~20 µM and ~30 µM by the third day, respectively. We observed that prolonging the drug loading time from 1 days to 3 days increased the amount of released drug, indicating a one-day load does not saturate loading capacity of 60-µm films ( Supplementary Figure S1 ). Films were stored dry in plastic bags after loading and kept at room temperature for up to 1 week prior to use. Thus, insoluble silk fibroin wound dressings were loaded with A438079 in a film-tunable and loading-tunable manner with differential release kinetics into aqueous solution.

We next sought to determine whether A438079-loaded films would exhibit therapeutic properties in an in vitro system of inflammation. We used the mouse macrophage cell line, J774DUAL (derivative of J774A.1), to investigate inflammasome signaling. We first performed ELISA to measure production and release of IL-1β, a key inflammatory cytokine and secreted inflammasome mediator ( Figure 3A ). J774DUAL cells primed with LPS for 4 hours with E. coli LPS and stimulated for 45 minutes with 3 mM extracellular ATP to activate P2X7 receptor signaling robustly secreted IL-1β. When cells were treated with conditioned medium exposed to A438079-loaded silk films, near-complete inhibition of IL-1β secretion was observed. We next evaluated whether this inhibition of IL-1β secretion was due to inhibition of inflammasome activity. We performed immunofluorescence of J774DUAL cells primed with LPS and stimulated with 3 mM extracellular ATP to evaluate the formation of perinuclear ASC specks, a canonical indicator of inflammasome assembly ( Figures 3B, C ). Without inhibitor treatment, approximately 6% of the population exhibited perinuclear ASC specks. We note these levels of ASC speck positivity are in agreement with prior work by us (26) and others (45, 46), but are likely an underestimation due to repeated wash steps during immunofluorescence processing. When treated with conditioned medium exposed to A438079-loaded silk films, <0.5% of the population exhibited ASC specks. Thus, A438079-loaded silk films robustly inhibit inflammasome assembly and activity when evaluated with mouse macrophages in vitro.

We generated full-thickness 6-mm biopsy punch wounds in the dorsum of genetically diabetic db/db mice ( Figure 4 ). Wounds were treated at the time of wounding either with saline or saline solution containing 100 µM A438079 without silk film (“direct A438079;” dosage selected according to release dynamics in vitro – Figure 2D ) or had empty silk films or silk films loaded with A438079 applied. Planimetric analysis was performed at 10 days post-wounding, similar to prior reports (19). We found that silk films alone and a topical aqueous solution of A438079 partially, but non-significantly, enhanced wound closure. By comparison, silk films loaded with A438079 resulted in a 40% wound closure which was statistically significant versus saline treated, empty film-treated, and direct drug-treated wounds ( Figures 4A, B ). We performed immunohistochemical staining for IL-1β as the most direct downstream inflammatory mediator of inflammasome activation ( Figures 4C, D ). Quantification of IL-1β positive cells at the wound margin indicated a statistically significant reduction in wounds treated with silk films loaded with A438079 compared to saline or empty silk film treated wounds. Comparison to direct A438079 treated wounds approached significance (p=0.21), suggesting even a day 0 treatment with A438079 exerted some anti-inflammasome activity over the course of diabetic healing. Thus, A438079-loaded films, acting as a depot, promote enhanced and sustained anti-inflammasome activity resulting in improved wound closure in diabetic wounds.