Controlled Release of H2S from Biomimetic Silk Fibroin-PLGA Multilayer Electrospun Scaffolds

Abstract

The possibility of incorporating H₂S slow-release donors inside biomimetic scaffolds can pave the way to new approaches in the field of tissue regeneration and anti-inflammatory treatment. In the present work, GYY4137, an easy-to-handle commercially available Lawesson's reagent derivative, has been successfully incorporated inside biomimetic silk fibroin-based electrospun scaffolds. Due to the instability of GYY4137 in the solvent needed to prepare silk fibroin solutions (formic acid), the electrospinning of the donor together with the silk fibroin turned out to be impossible. Therefore, a multilayer structure was realized, consisting of a PLGA mat containing GYY4137 sandwiched between two silk fibroin nanofibrous layers. Before their use in the multilayer scaffold, the silk fibroin mats were treated in ethanol to induce crystalline phase formation, which conferred water-resistance and biomimetic properties. The morphological, thermal, and chemical properties of the obtained scaffolds were thoroughly characterized by SEM, TGA, DSC, FTIR, and WAXD. Multilayer devices showing two different concentrations of the H₂S donor, i.e., 2 and 5% w/w with respect to the weight of PLGA, were analyzed to study their H₂S release and biological properties, and the results were compared with those of the sample not containing GYY4137. The H₂S release analysis was carried out according to an "ad-hoc" designed procedure based on a validated high-performance liquid chromatography method. The proposed analytical approach demonstrated the slow-release kinetics of H₂S from the multilayer scaffolds and its tunability by acting on the donor's concentration inside the PLGA nanofibers. Finally, the devices were tested in biological assays using bone marrow-derived mesenchymal stromal cells showing the capacity to support cell spreading throughout the scaffold and prevent cytotoxicity effects in serum starvation conditions. The resulting devices can be exploited for applications in the tissue engineering field since they combine the advantages of controlled H₂S release kinetics and the biomimetic properties of silk fibroin nanofibers.

Figure 1

Schematic representation of the steps of the work: (a) production of fibrous biomimetic scaffolds of silk fibroin and PLGA by electrospinning, (b) preparation of multilayer bioactive scaffold releasing H₂S, composed of an inner layer of PLGA loaded with GYY sandwiched between silk fibroin mats, (c) H₂S release tests in vitro, and (d) h-MSC culture onto scaffolds.

Figure 2

SEM images and corresponding fiber diameter distribution of PLGA and PLGA/GYY electrospun mats: (a) PLGA, (b) PLGA-2GYY, (c) PLGA-5GYY (scale bar = 10 μm). The insets show the SEM images at higher magnification (scale bar = 2 μm).

Figure 3

SEM images and corresponding fiber diameter distribution of (a) the SF electrospun mat and (b) SF-mat15 electrospun mat. Scale bar = 2 μm. Insets: scale bar = 1 μm. (c) ATR-IR spectra, (d) WAXD, (e) DSC curves (heating scan after quenching), and (f) TGA curves of SF and SF-mat15 mats.

Figure 4

Schematic representation of the H₂S quantification method: (a) 4 mL of PB, sonicated for 30 min before use to remove dissolved gasses and/or entrained gas bubbles, and the multilayer sample was placed into a 20 mL plastic tube first; (b) afterward, the tube was wrapped with an aluminum film and incubated in the thermostatic shaking bath at 37 °C; (c) derivatization reaction of releasing H₂S with the fluorescent alkylating agent MBB; (d) after each withdrawal, the PB solution was refilled with the same amount of PB (30 μL); (e) quantification of hydrogen sulfide: derivatization of H₂S with MBB, forming a sulfide-dibimane (SDB) product via S-alkylation; the resultant fluorescent SDB is analyzed by HPLC-FLD; the H₂S concentration was then obtained after SDB area interpolation in a calibration curve (Supporting information). Steps 1, 2, and 3 were carried out in a hypoxic chamber (1% O₂) at RT.

Figure 5

H₂S-release from sterilized multilayer samples from 0 to 168 h. Three replicates for each kind of sample were assembled, and two derivatizations for each assembled sample were tested and analyzed by HPLC-FLD. Each point is the result of the measurement of six different concentrations. The values plotted represent mean ± SD.

Figure 6

SEM images of SF-mat15, PLGA, PLGA-2GYY, and PLGA-5GYY as produced (upper row) and after 7 days of H₂S-releasing tests (lower row). Scale bar = 5 μm.

Figure 7

(a) Cell viability evaluated by Calcein-AM/Ethidium homodimer staining. Human MSCs were seeded on the scaffolds and stained after 72 h in culture. Panels show a top-view, single-layer picture, and a picture representing the transversal section of each scaffold for a total depth of 78.4 μm across each scaffold. (b) Cytotoxicity assay: LDH released in the supernatants by h-MSC was measured after 72 h in culture. Data represent the % cytotoxicity using cells grown on plastic as a control (negative) sample and are expressed as mean ± SD of N = 3 independent experiments. (c) Histograms representing the enrichment of cytoplasmic oligonucleosomes in h-MSC after 72 h of starvation. Data are expressed as mean ± SD of N = 3 independent experiments. *p < 0.05 compared to control cells grown on α-MEM 15% FBS; CTRL+ represents a positive control provided by the manufacturer, constituted by lyophilized DNA–histone complex. ns = not significant.