Elucidating siRNA Cellular Delivery Mechanism Mediated by Quaternized Starch Nanoparticles

Abstract

Starch-based nanoparticles are highly utilized in the realm of drug delivery taking advantage of their biocompatibility and biodegradability. Studies have utilized Quaternized starch (Q-starch) for small interfering RNA (siRNA) delivery, in which quaternary amines enable interaction with negatively charged siRNA, resulting in self-assembly complexation. Although reports present numerous applications, the demonstrated efficacy is nonetheless limited due to undiscovered cellular mechanistic delivery. In this study, a deep dive into Q-starch/siRNA complexes' cellular mechanism and kinetics at the cellular level is revealed using single-particle tracking and cell population level using imaging flow cytometry. Uptake studies depict the efficient cellular internalization via endocytosis while a significant fraction of complexes' intracellular fate is lysosome. Utilizing single-particle tracking, it is found that an average of 15% of cellular detected complexes escape the endosome which holds the potential for the integration in the cytoplasmatic gene silencing mechanism. Additional experimental manipulations (overcoming endosomal escape) demonstrate that the complex's disassembly is the rate-limiting step, correlating Q-starch's structure-function properties as siRNA carrier. Structure-function properties accentuating the high affinity of the interaction between Q-starch's quaternary groups and siRNA's phosphate groups that results in low release efficiency. However, low-frequency ultrasound (20 kHz) application may have induced siRNA release resulting in faster gene silencing kinetics.

Keywords: delivery mechanism; endocytosis; particle tracking; polysaccharide; siRNA; starch.

Figure 1

Q‐starch/siRNA complexes at N/P 2 efficiently internalize into cells via the endocytosis uptake mechanism. a) Confocal microscopy images of NAR cell treated with free siRNA^Cy5 and Q‐starch/siRNA^Cy5 complexes for 24 h (Nucleus in blue, cell membrane in yellow, and siRNA in red). (b–d) Image stream flow cytometry quantitative study: (b) Cells’ images presenting fluorescently labeled Q‐starch/siRNA^Cy5 complexes (red) uptake into NAR cells along the study course, (c) Quantitative analysis of complexes uptake study demonstrating percentage of cells presenting uptake, and (d) mean cell fluorescent intensity following continuous incubation with Q‐starch/siRNA^Cy5 complexes in the presence (+ Serum) or absence (‐ Serum) of serum in the cell medium. Quantitative results are normalized to control untreated cells (n = 3, mean ± STDEV). e) Q‐starch/siRNA complexes cellular uptake route. NAR cells were pre‐treated by Dynamine inhibitor (Dynasore, 80 µM), incubated for 4 h with fluorescently labeled complexes and the uptake inhibition was visualized and quantitatively analyzed using imaging flow cytometry (results normalized to no inhibition, n = 3, mean ± STDEV, ***p < 0.001 between no inhibition to the inhibition treatment). Scale bar = 10 µm, siRNA concentration is 50 nM.

Figure 2

Co‐localization of labeled Q‐starch/siRNA complexes with intracellular compartments obtained by live cell fluorescent microscopy presenting representative images of cells (the image represents the cells in the well and n = 3 biological replicates, 3 separate wells) and flow cytometry. a) Illustration indicating the tracked compartments which complexes might co‐localize with (early endosome, late endosome, and lysosome) and reference for each stage's supporting results. b) Tracking Q‐starch^5‐DTAF/siRNA complexes’ (green) co‐localization with early endosome (red) at an early time point (4 h) with and without Bafilomycin A1 (0.1 µM), the H+ ATPase inhibitor. c) Tracking complexes’ co‐localization at late time point (24 h) studied by Q‐starch^5‐DTAF/siRNA complexes (green) co‐localization with early endosome or late endosome (red) and Q‐starch^Cy3.5/siRNA complexes (red) with Lysotracker (green). d) Q‐starch/siRNA^Cy5 (red) co‐localization quantitative analysis with the lysosome (green) under the “pulse and chase” experimental setup. Imagestream flow cytometry cell images over a 48 h study. e) Quantitative co‐localization analysis of flow cytometry data by bright detail similarity (BDS) feature. The graph presents means BDS change along the experimental kinetic, (n = 3, mean ± STDEV) *p < 0.05, **p < 0.01.

Figure 3

Q‐starch/siRNA complexes tracking analysis. NAR cells were incubated with 5‐DTAF‐labeled (green) complexes (siRNA concentration was 50 nM) for the indicated time and imaged over time using a live cell fluorescent microscope. a) Representative images taken from live cell imaging and particle cellular tracks following 15 min, 30 min, and 4 h of incubation with the complexes. b) Tracking analysis and mean square displacement (MSD) of Q‐starch/siRNA complexes using the Fiji Trackmate plugin. MSD of particle population following 4 h of incubation demonstrating 3 distinct types of motion (flow, free diffusion, and obstructed motion). Each line represents an average particle population with the same type of motion. c) Cellular particle tracking analysis at varying times of incubation with complexes demonstrating the fraction of particles presenting each type of motion.

Figure 4

Effect of Imidazole grafting on Q‐starch complexes gene silencing efficiency. a) Synthesis and chemical structure of Q‐starch‐Imidazole (Q‐starch‐Im) using 4‐Imidazolecarboxylic acid. b) In vitro gene silencing transfecting NAR cells with Q‐starch‐Im ⁻¹siP‐gp complexes at N/P 4–8 with and without chloroquine (Chlorq.). Noncoding siRNA (NC control) was utilized as off‐targeting control at N/P = 8. Gene silencing was also examined for non‐grafted Q‐starch/siRNA complexes in the presence of chloroquine. Gene silencing was quantified at the mRNA level using RT‐PCR and the results were normalized to the untreated group (n = 3, mean ± STDEV).

Figure 5

Effect of LFUS on Q‐starch/siRNA delivery barriers and gene silencing efficiency. a) Gene silencing efficiency of NAR cells treated by Q‐starch/siPgp complexes for 1 h followed by application of LFUS (2 W cm⁻², 10 and 20 s). Gene silencing was quantified at the protein level 24 h post‐treatment (siRNA concentration 50 nM, n = 3, mean ± STDEV). b) Effect of LFUS on Q‐starch/siRNA^Cy5 complexes uptake into NAR cells visualized by confocal microscopy, images present representative cells from 3 biological replicates, n = 3 (Cell membrane labeled in green, Complexes labeled in red).