Optimizing Silk Nanoparticle Assembly with Potassium Ions: Effects on Physicochemical Properties and Encapsulation Efficiency

Abstract

Silk fibroin is a promising biomaterial for nanocarrier-based drug delivery due to its biocompatibility, biodegradability, and tunable mechanical properties. In addition, the silk protein is amenable to various processing strategies, offering flexibility for optimizing particle characteristics. Emerging evidence highlights that metal ions can modulate silk conformation and structure in the silk gland, as well as influencing self-assembly, potentially impacting silk nanoparticle fabrication. Our previous study highlighted the potential of Ca²⁺ in silk nanoparticle fabrication. However, other metal ions in the silk gland influence silk fibroin behavior too. Here, we investigate how potassium ions (K⁺), with similar abundance to Ca²⁺ in the silkworm gland, influence silk nanoparticle formation as modulators of self-assembly and material properties, aiming to produce nanoparticles with distinct physicochemical profiles. We show that K⁺ enhances silk assembly, increases nanoparticle size, alters surface charge (zeta potential), and boosts production yield, thereby minimizing silk wastage during silk nanoparticle preparation. Potassium ions also significantly improve payload encapsulation efficiency, making K⁺ inclusion valuable for a range of drug-loading applications. The resulting silk nanoparticles exhibit reduced toxicity and inflammatory response, highlighting their promise as safe and effective nanocarrier candidates for drug delivery. Our findings establish K⁺ as a fundamental yet powerful tool for tuning silk nanoparticle properties to meet pharmaceutical needs.

Keywords: Bombyx mori; antisolvent precipitation; metal ion; nanomedicine; silk fibroin.

1

Analysis of the structure and conformation of aqueous silk and solid nanoparticles. (a) The effect of K⁺ on aqueous silk fibroin (SF) structural alteration, as studied using Thioflavin T assays (ThT). The IPA-treated SF indicated the silk β-sheet structure and aggregation control, showing increased fluorescence intensity of bound ThT state (n = 3), (b) 1D ¹H NMR (n = 1), (c) and 2D [¹H, ¹⁵N] HSQC NMR spectra in the NH region (n = 1). The 1D 1H-NMR spectra are shown on the left side of the panel, with the difference spectra (red and blue) displayed on the right for comparison to the control spectrum (black). Black traces represent the control, red corresponds to 1.1 mg K⁺ (4.1% increase), and blue to 17.3 mg K⁺ (5.2% increase). The different spectra were adjusted by multiplying the vertical scale by a factor of 4. (d) The secondary structures of freeze-dried aqueous silk (17.3 mg K⁺) and silk nanoparticles (SNP) (0 and 17.3 mg K⁺) were studied with FTIR and compared to freeze-dried aqueous silk (0 mg K⁺) with low (silk I) and EtOH-treated air-dried silk film with high (silk II) β-sheet content; normalized amide I absorbance spectra, (e) secondary structure content, (f) and correlation coefficient (n = 3). The second-derivative amide I spectrum of an air-dried silk film (silk I) served as a reference for the correlation coefficient (R) calculations (mean ± SD, n = 3). The controls employed in all experiments were derived from the data set previously reported, carried out at the same time as this study. Statistical analyses: One-way ANOVA and Dunnett’s multiple comparisons test for the Thioflavin T assay; Two-way ANOVA and Dunnett’s multiple comparisons test for secondary structure content; p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***). Abbreviations: ANOVA: analysis of variance; IPA: isopropanol; SF: aqueous silk fibroin; SNP: solid silk nanoparticle; ThT: Thioflavin T; EtOH: ethanol

2

Silk nanoparticle physicochemical properties. (a) Analysis of particle size (using DLS and NTA) and zeta potential (ZP) using ELS (n = 3). (b) Analysis of particle size and morphology (using FE-SEM). The FE-SEM images were taken at 10, 20, and 60K magnifications. Size and circularity calculations were performed using 400–500 particles and 100–150 particles, respectively, derived from at least three different regions of interest. The controls employed in all experiments were derived from a previously reported data set, carried out at the same time as this study. One-way ANOVA and Dunnett’s multiple comparison test were used for statistical analysis: p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***). Abbreviations: ANOVA: analysis of variance; DLS: dynamic light scattering; ELS: electrostatic light scattering; FE-SEM: field emission scanning electron microscopy; NTA: nanoparticle tracking analysis; PDI: polydispersity index; SF: aqueous silk fibroin; ZP: zeta potential.

3

Model drug-loaded silk nanoparticles (SNPs) using Thioflavin T as a drug model. Thioflavin T-loaded SNPs were prepared at 1 mg/mL based on protein concentration and were characterized by NTA (n = 3): (a) size distribution profile measured by light scattering mode (black) and fluorescence mode (green); (b) the particle number; (c) Thioflavin T fluorescence intensity. (d) The production yield was calculated according to the SNP dried weight. The controls employed in all experiments were derived from the data set previously reported, carried out at the same time as this study. One-way ANOVA and Dunnett’s multiple comparison tests were used for statistical analysis: p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***). Abbreviations: ANOVA: analysis of variance; NTA: nanoparticle tracking analysis; SF: aqueous silk fibroin; SNP: silk nanoparticle; ThT: Thioflavin T.

4

Cellular responses of RAW 264.7 murine macrophages to silk nanoparticles (SNPs). (a) Cytotoxicity of macrophages exposed to SNP levels ranging from 31.25 to 500 μg/mL for 48 h (n = 3). Inflammatory responses: (b) release profile of cytokines and chemokines presented in log₂ fold-change (FC) of expression level compared with a basal control. Interpretation of log₂(FC): a value of zero indicates no change in expression, positive values indicate upregulation, and negative values indicate downregulation (n = 3); and (c) NO₂ ^– and TNF-α production in macrophages exposed to 500 μg/mL SNPs for 24 h (n = 3). Cells treated with 200 ng/mL LPS served as the positive inflammation-stimulated control. Cells exposed only to the complete medium served as a basal control. The controls employed in all experiments were derived from the data set previously reported, carried out at the same time as this study. One-way ANOVA and Dunnett’s multiple comparison test were used for statistical analysis: p < 0.05 (*), p < 0.01 (**), and p < 0.001 (***). Abbreviations: ANOVA: analysis of variance; LPS: lipopolysaccharide; SF: aqueous silk fibroin; SNP: silk nanoparticle.

5

Impact of calcium and potassium on silk nanoparticle performance. The correlation coefficient (r) between the measured variables in the K⁺ data set (this study) and the reference Ca²⁺ data set (previous study) (n = 3): positive linear correlation (r > 0), nonlinear correlation (r = 0), and negative linear correlation (r < 0). Abbreviations: SF: aqueous silk fibroin; SNP: silk nanoparticle; ThT: Thioflavin T; NMR: Nuclear magnetic resonance; DLS: Dynamic light scattering; NTA: Nanoparticle tracking analysis; FE-SEM: Field-emission scanning electron microscopy; PDI: Polydispersity index; ZP: zeta potential.