Biomimetic Platelet-Cloaked Nanoparticles for the Delivery of Anti-Inflammatory Curcumin in the Treatment of Atherosclerosis

Abstract

Atherosclerosis still represents a major driver of cardiovascular diseases worldwide. Together with accumulation of lipids in the plaque, inflammation is recognized as one of the key players in the formation and development of atherosclerotic plaque. Systemic anti-inflammatory treatments are successful in reducing the disease burden, but are correlated with severe side effects, underlining the need for targeted formulations. In this work, curcumin is chosen as the anti-inflammatory payload model and further loaded in lignin-based nanoparticles (NPs). The NPs are then coated with a tannic acid (TA)- Fe (III) complex and further cloaked with fragments derived from platelet cell membrane, yielding NPs with homogenous size. The two coatings increase the interaction between the NPs and cells, both endothelial and macrophages, in steady state or inflamed status. Furthermore, NPs are cytocompatible toward endothelial, smooth muscle and immune cells, while not inducing immune activation. The anti-inflammatory efficacy is demonstrated in endothelial cells by real-time quantitative polymerase chain reaction and ELISA assay where curcumin-loaded NPs decrease the expression of Nf-κb, TGF-β1, IL-6, and IL-1β in lipopolysaccharide-inflamed cells. Overall, due to the increase in the cell-NP interactions and the anti-inflammatory efficacy, these NPs represent potential candidates for the targeted anti-inflammatory treatment of atherosclerosis.

Keywords: anti‐inflammatory; atherosclerosis; biohybrid; nanoparticles; platelets membrane.

Scheme 1

Schematic illustration of Curc@Lignin@TA@PL. Curcumin was encapsulated in lignin NPs. The particles were then coated with a TA−iron complex, followed by cloaking with fragments of platelet membrane. Image created with Biorender.com.

Figure 1

Physicochemical characterization of the formulation. A) Size (nm); B) PdI; and C) ζ‐Potential (mV). The results are presented as mean±s.d. (n = 3 independent replicates, each constituted by three technical replicates). D–F) Transmission Electron Microscope images of Lignin, Lignin@TA, and Lignin@TA@PL, respectively. Scale bars: 500 nm in D, 200 nm in E, 100 nm in F. G–I) Scanning Electron Microscope images of Lignin, Curc@Lignin, Curc@Lignin@TA, respectively. Scale bars: 1 µm for G and H, 2 µm for I.

Figure 2

Colloidal stability of Lignin, Lignin@TA, and Lignin@TA@PL in RPMI‐1640 medium with 10% of FBS. A) Size (nm) of the particles over time (0–120 min); and B) PdI of the particles over time (0–120). The results are presented as mean±s.d. (n = 3 independent replicates, each constituted by three technical replicates).

Figure 3

Cytocompatibility in human endothelial cells and in rat myoblasts after 24 h of incubation. A) Cell Viability (%) of HCAEC after 24 h incubation with the final system (Curc@Lignin@TA@PL) and the intermediates. B) Cell Viability (%) of H9C2 after 24 h incubation with the final system (Curc@Lignin@TA@PL) and the intermediate NPs. The results are presented as mean±s.d. (n = 3 independent biological replicates, each composed of ≥3 technical replicates). The results were analyzed with two‐way ANOVA, followed by Tukey's post‐test with no statistically significant difference between curcumin‐loaded and empty particles, or the relative intermediates, at all the concentrations assessed. Complete medium and Triton X‐100 (1%) represent the negative and positive control, respectively.

Figure 4

Cytocompatibility in human immune cells after 24 h of incubation. A) Cell viability (%) of KG‐1 after 24 h of incubation with the final system (Curc@Lignin@TA@PL) or the intermediate NPs; B) Cell viability (%) of BDCM after 24 h of incubation with the final system (Curc@Lignin@TA@PL) or the intermediate NPs; and C) Cell viability (%) of PBMC after 24 h incubation with the final system (Curc@Lignin@TA@PL) or the intermediate NPs. Complete medium and Triton X‐100 (1%) represent the negative and positive control, respectively. The results are presented as mean±s.d. (n = 3 independent biological replicates, each composed of ≥3 technical replicates). The results were analyzed with two‐way ANOVA, followed by Tukey's post‐test. The cell viability of the loaded final system (Curc@Lignin@TA@PL) was compared to the empty final system (Lignin@TA@PL), the cell viability of the intermediate NPs coated with the TA‐iron complex (Curc@Lignin@TA) was compared to the relative empty NPs control (Lignin@TA), and, finally, the viability of curcumin‐loaded lignin NPs (Curc@Lignin) was compared to empty lignin NPs (Lignin). The levels of significance were set at the probabilities of * p<0.05, ** p<0.01, *** p<0.001, and **** p<0.0001.

Figure 5

Percentage (%) of FITC⁺ A) HCAEC or B) KG‐1 cells after interaction with Curc@Lignin@TA@PL and intermediate samples for (A) 1 h and (B) 30 min at +37 °C. The NP interaction with the cells, at a concentration of 50 µg mL−1, was evaluated by flow cytometry. The fluorescence of curcumin (loaded within the particles) was used as fluorochrome in the FITC channel. The results are presented as mean±s.d. (n = 3 biological replicates, each constituted of 3 technical replicates). The results were analyzed with two‐way ANOVA followed by Tukey's post‐test. The percentage of FITC+ cells in each sample was first compared to the other samples within the same type of cells (healthy vs healthy, e.g., Curc@Lignin vs Curc@Lignin@TA), and then compared with the respective particle in the activated cells (e.g., Curc@Lignin in healthy cells vs Curc@Lignin in activated cells). The levels of probability were set at the probabilities of * p<0.05, *** p<0.001, and **** p<0.0001.

Figure 6

Mean Fluorescence Intensity of APC anti‐human CD86 antibody in A) KG‐1 cells and B) PBMC after 48 h incubation. The cells were incubated with the final empty NP (Lignin@TA@PL) or intermediates for 48 h before being stained for CD86 expression. We run the samples in BD Accuri equipped with a C6 automatic sampler. The results are presented as mean±s.d. (n = 3 biological replicates each constituted of three independent technical replicates). The data were analyzed with two‐way ANOVA, followed by Tukey's post‐test comparing all the samples to the control within the same group (healthy or LPS‐activated) or comparing the same sample between the different groups (lignin in healthy cells vs lignin in LPS‐activated cells). The level of significance was set at the probability of ****p<0.0001.

Figure 7

Normalized ΔΔCT of A) NFKB1 and B) TGFB1 in HCAEC cells after inflammation with LPS and incubation with curcumin‐loaded or empty NPs and relative intermediates. The cells were inflamed with LPS (10 ng mL−1) for 24 h, followed by 24 h incubation with the NPs (50 µg mL⁻¹). The ΔCCT values were normalized to the housekeeping gene 18s, and are further normalized by the negative control. The results are presented as mean±s.d. (n≥3 independent samples) and were analyzed by paired One way ANOVA versus the positive control LPS. The level of significance was set at the probability of * p<0.05, ** p<0.01, *** p<0.001, and **** p<0.0001.

Figure 8

IL‐6 (A) and IL‐1β (B) quantification in the media of HCAEC cells after 24 h incubation with the samples. The cells were pre‐treated for 24 h with LPS (10 ng mL ⁻¹) before adding the samples. The results are presented as mean±s.d. (n≥3 independent samples) and were analyzed by paired One way ANOVA versus the positive control LPS. The levels of significance were set at the probabilities of * p<0.05, ** p<0.01, *** p<0.001, and **** p<0.0001.