Preparation of Neutrally-charged, pH-responsive Polymeric Nanoparticles for Cytosolic siRNA Delivery

Summary

Methods to prepare and characterize the physicochemical properties and bioactivity of neutrally-charged, pH-responsive siRNA nanoparticles are presented. Criteria for successful siRNA nanomedicines such as size, morphology, surface charge, siRNA loading, and gene silencing are discussed.

Abstract

The success of siRNA as a targeted molecular medicine is dependent upon its efficient cytosolic delivery to cells within the tissue of pathology. Clinical success for treating previously ‘undruggable’ hepatic disease targets with siRNA has been achieved. However, efficient tumor siRNA delivery necessitates additional pharmacokinetic design considerations, including long circulation time, evasion of clearance organs (e.g., liver and kidneys), and tumor penetration and retention. Here, we describe the preparation and in vitro physicochemical/biological characterization of polymeric nanoparticles designed for efficient siRNA delivery, particularly to non-hepatic tissues such as tumors. The siRNA nanoparticles are prepared by electrostatic complexation of siRNA and the diblock copolymer poly(ethylene glycol-b-[2-(dimethylamino)ethyl methacrylate-co-butyl methacrylate]) (PEG-DB) to form polyion complexes (polyplexes) where siRNA is sequestered within the polyplex core and PEG forms a hydrophilic, neutrally-charged corona. Moreover, the DB block becomes membrane-lytic as vesicles of the endolysosomal pathway acidify (< pH 6.8), triggering endosomal escape and cytosolic delivery of siRNA. Methods to characterize the physicochemical characteristics of siRNA nanoparticles such as size, surface charge, particle morphology, and siRNA loading are described. Bioactivity of siRNA nanoparticles is measured using luciferase as a model gene in a rapid and high-throughput gene silencing assay. Designs which pass these initial tests (such as PEG-DB-based polyplexes) are considered appropriate for translation to preclinical animal studies assessing the delivery of siRNA to tumors or other sites of pathology.

Introduction

Because siRNAs inhibit the translation of proteins from mRNA sequences, they can theoretically be used to drug all known pathologies^1,2,3,4,5. However, the use of siRNA in medicine is limited by the comprehensively poor pharmacokinetic profile of siRNA molecules^6,7. When injected intravenously, siRNAs are rapidly cleared through the kidneys and/or degraded by nucleases^8,9. Due to its large size and negative charge, siRNA cannot enter cells or escape the endolysosomal pathway to access the RNA-Induced Silencing Complex (RISC) that resides in the cytosol^10,11,12,13. Thus, extensive effort has focused on the design and implementation of siRNA delivery strategies¹⁴. This effort has largely focused on the development of lipid- and polymer-based nanoparticles which package siRNA, protect it from clearance and degradation in vivo, and initiate cellular uptake and endosomal escape through ionizable, cationic amine groups. Many pre-clinical successes have been reported and most recently, the first clinical success has been reported for nanoparticle-based hepatic siRNA delivery to treat hereditary transthyretin-mediated (hATTR) amyloidosis¹⁵.

There are many cancer-causing genes that are currently “undruggable” by conventional pharmacology (i.e., small molecule drugs), motivating the design of polymeric siRNA nanoparticles (si-NPs) to treat cancer¹⁶. However, there are a separate set of design parameters that must be considered for non-hepatic siRNA delivery. The delivery system must shield the cationic charge of the polyplex which causes agglutination within the systemic circulation^17,18,19. For tumor delivery, specifically, si-NP stability is essential to endow long circulation and thus increased accumulation within tumors via the enhanced permeability and retention (EPR) effect^20,21. Moreover, control over si-NP size is essential since only nanoparticles approximately 20 – 200 nm diameter in size leverage EPR²², and smaller si-NPs (~20 – 50 nm diameter) exhibit improved tumor penetration over larger sized nanoparticles and microparticles²³.

To address these additional design constraints for systemic tumor delivery of siRNA following intravenous administration, neutrally-charged, pH-responsive si-NPs have been developed (Figure 1)²⁴. These si-NPs are PEGylated, or most recently, Zwitterionated²⁵, for neutral surface charge and resistance to protein adsorption and opsonization in circulation. Since they cannot rely solely on cationic character to drive intracellular delivery, extremely efficient endosomal escape is imperative for achieving potent gene silencing. Accordingly, the core of these si-NPs is composed of a highly endosomolytic core which is inert at extracellular pH (7.4), but which is triggered in a switch-like manner in the acidified conditions of the endolysosomal pathway [pH 6.8 (early endosomes) – 5.0 (lysosomes)]. Lastly, a mixture of cationic and hydrophobic content within the core of si-NPs provide both electrostatic and van der Waals stabilization forces, improving stability of the si-NPs in blood compared to merely cationic systems.

The integration of many functions into a relatively simple design is possible using Reversible Addition-Fragmentation chain Transfer (RAFT) controlled polymerization to produce polymers with complex architecture and precise composition. To produce si-NPs with neutral surface charge, pH-responsiveness, and NP stability, RAFT is used to synthesize poly(ethylene glycol-b-[2-(dimethylamino)ethyl methacrylate-co-butyl methacrylate]) (PEG-DB; Figure 1A). PEG-DB is electrostatically complexed with siRNA, forming si-NPs with a PEG corona and DB/siRNA core (Figure 1B). PEG forms an inert, neutrally-charged hydrophilic layer on the si-NP corona. The DB block consists of a 50:50 molar ratio of 2-(dimethylamino)ethyl methacrylate (DMAEMA) and butyl methacrylate (BMA). Cationic DMAEMA electrostatically complexes negatively-charged siRNA. BMA self-associates within the NP core by van der Waals interactions, increasing NP stability. Together, DMAEMA and BMA impart pH-dependent lipid bilayer-lytic behavior to the DB polymer block. At extracellular pH, the DB block is sequestered to the si-NP core and is inert to lipid bilayers. Under acidic conditions, such as those within the endolysosomal pathway, ionizable DMAEMA within the DB block facilitates the proton sponge effect, where endosomal buffering leads to osmotic swelling and rupture²⁶. Additionally, hydrophobic BMA moieties within the DB block actively integrate into and lyse lipid bilayers, resulting in potent endosomolysis. Thus, siRNA is complexed with PEG-DB to form si-NPs that are neutrally-charged and highly stable at extracellular pH but which disrupt lipid bilayers at acidic pH, ensuring cytosolic delivery of the siRNA payload.

Herein are described the experimental procedures to produce si-NPs from PEG-DB. Methods to characterize the physicochemical parameters and bioactivity of si-NPs are presented and discussed. In order to rapidly assess si-NP bioactivity, luciferase is used as a model gene for knockdown studies. Firefly Luciferase is the protein responsible for the ‘glow’ of fireflies²⁷. Accordingly, mammalian cells transfected with the firefly luciferase gene produce a bioluminescent ‘glow’ that can be captured using a luminometer to quantify levels of Luciferase expression. Here, we use Luciferase to assess bioactivity of si-NPs by delivering siRNA against Luciferase and quantifying the corresponding reduction in bioluminescence in Luciferase-expressing cells compared to cells that receive a scrambled siRNA.

Protocol

1. Preparation and characterization of si-NPs

1. si-NP preparation
   1. Dissolve polymer in 10 mM citric acid buffer (pH 4.0) at 3.33 mg/mL. Polymer can first be dissolved at 10x concentration in ethanol to ensure dissolution.
      NOTE: Polymer can be dissolved at lower concentrations, but use at concentrations above 3.33 mg/mL can prevent homogenous NP formation.
   2. Add siRNA (50 μM in diH₂O) to result in N⁺:P^- ratio of 10. Mix polymer and siRNA solutions thoroughly by pipetting and let incubate for 30 min. The N⁺:P^- ratio represents the number of positively-charged amine groups on the polymer to the number of negatively-charged phosphate groups on the siRNA and is calculated by the formula below:
      
      where, mol Pol is the molar amount of polymer, RU amine is the number of repeating units of positively-charged amines per polymer, mol siRNA is the molar amount of siRNA, and bp siRNA is the number of base pairs per siRNA molecule.
   3. Add a 5-fold excess of 10 mM phosphate buffer (pH 8.0) and mix gently either by pipetting or inverting the tube. To confirm that the final pH is neutral (~7.2-7.5), pipette 10 μL of si-NP solution onto pH test strips.
      NOTE: Citric acid and phosphate buffers are prepared according to the Millipore Sigma Buffer Reference Center Charts.
2. Physicochemical characterization of si-NPs
3. Record the size and surface charge of resulting si-NPs using dynamic light scattering (DLS). Prepare a DLS sample by filtering 1 mL of si-NPs (0.1 – 1.0 mg/mL) through 0.45 μm pore-size syringe filters into a square quartz or polystyrene cuvette. Record size and surface charge measurements using a DLS instrument according to the manufacturer’s specifications.
   1. Confirm the size and morphology of si-NPs by imaging analysis using transmission electron microscopy (TEM).
      1. Add 5 μL of si-NP solution at 1 mg/mL to TEM grids and incubate for 60 s. Blot dry for 3 s.
      2. Add 5 μL of 3% uranyl acetate solution and incubate for 20 s. Blot dry for 3 s. Dry grids overnight under desiccation.
      3. Image grids according to the protocol established for the specific microscope to be used.
   2. Characterize the loading of siRNA in si-NPs at various N⁺:P^- ratios using agarose gel retardation.
      1. To produce 2% agarose gel, add 2 g of electrophoresis grade agarose powder to 100 mL of 1x TAE (Tris-acetate-EDTA) buffer at pH 8.0. Stir to suspend agarose. Heat uncovered in microwave until all agarose is dissolved (1-3 min).
      2. Once cooled, add 5 μL of ethidium bromide (10 mg/mL in H₂O), and mix well. Pour agarose into a gel tray and place comb to produce wells, letting dry for 30 min. Carefully remove comb to leave behind loading wells, and fill the gel tray to the max fill line with 1x TAE buffer.
      3. Generate si-NPs (according to the procedure above) at 0, 1, 2, 5, 7, 10, 20, and 40 N⁺:P^- ratios. Place 2 μL aliquots of loading dye (no SDS and reducing agents) on paraffin film for each si-NP formulation. Mix 10 μL of si-NP solution with loading dye on paraffin film by pipette.
      4. Add si-NP/loading dye solutions to agarose gel wells. Run voltage source at 100 V for 35 min (or until samples have traversed 80% of gel length).
      5. Visualize siRNA bands on a UV transilluminator according to the manufacturer’s specifications.

2. Determining in vitro bioactivity of si-NPs

1. Knockdown of the model gene luciferase
   1. Generate luciferase si-NPs (according to the procedure above) using luciferase siRNA and scrambled si-NPs using a scrambled siRNA sequence as a control. Formualte both si-NPs at the same final N⁺:P^- ratio and at the optimum ratio identified by agarose gel retardation studies. Example siRNA sequences are included in the Table of Materials.
   2. Seed luciferase-expressing cells [MDA-MB-231/Luciferase (Bsd) stable cells] in 96-well black-walled plates at a density of 2,000 cells per well. Allow to adhere overnight in full media (DMEM, 10% FBS) in an incubator (37 °C, 5% CO₂, 95% humidity).
   3. Dilute si-NPs into full serum media for a final volume of 100 μL per well and siRNA concentration of 100 nM. Treat cells for 24 h with si-NPs.
   4. After 24 h, remove treatments and replace media with full serum media containing 150 μg/mL D-luciferin. Incubate cells for 5 min before measuring luminescence on a plate reader or in vivo optical imaging system according to the manufacturer’s specifications.
   5. Replace luciferin-containing media with fresh, full serum media, and incubate 24 h more. Repeat the step above, removing media and replacing with full serum media containing 150 μg/mL D-luciferin, followed by a 5 min incubation prior to measuring luminescence at the 48 h timepoint.
   6. For longitudinal studies, maintain cells under sterile conditions while measuring luminescence. Continue to culture in fresh, full media between measurements after replacing luciferin-containing.
      NOTE: The appropriate siRNA concentration will vary with different si-NPs and siRNA molecules. When using neutrally-charged polyplexes with an endosomolytic core (e.g., PEG-DB), 100 nM is typically well-tolerated by the cells and produces >75% luciferase knockdown. The mass ratio of PEG-DB to siRNA at 10 N⁺:P^- ratio and 100 nM siRNA treatments (assuming 26 bp siRNA) is 23.3, i.e., add 23.3 ng of PEG-DB for every 1.0 ng of siRNA. For example, add 1.16 μL of 3.33 mg/mL polymer for 166.5 ng of siRNA to treat one well at 100 nM in a 96-well plate (100 mL media volume per well).

Results

Some essential characteristics of effective si-NPs for in vivo siRNA delivery are the proper size (~20 – 200 nm diameter), siRNA packaging, and gene silencing bioactivity. While this is not an exhaustive list (as addressed in the Discussion), these basic characteristics should be confirmed before considering further testing of a formulation.

Figure 2 illustrates the characterization of si-NP size an...

Discussion

The si-NPs described here are formed by electrostatic association of anionic siRNA and cationic polymers into polyion complexes (polyplexes). Electrostatic complexing of siRNA and the cationic DB block of PEG-DB polymers is facilitated by mixing at low pH (4.0). At pH 4.0, DMAEMA is highly protonated, and consequently the DB block is highly charged. This ensures that the polymers dissolve as unimers in solution as opposed to forming micelles and that DB complexes efficiently with siRNA. Subsequently, the pH of solution i...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
0.45 μm pore-size syringe filters	Thermo Fisher Scientific	F25133	17 mm diameter, PTFE membrane
0-14 pH test strips	Millipore Sigma	P4786
10x TAE buffer	Thermo Fisher Scientific/Invitrogen	AM9869
6-7.7 pH test strips	Millipore Sigma	P3536
96-well black walled plates	Corning	3603	Tissue-culture treated
Agarose Powder	Thermo Fisher Scientific/Invitrogen	16500
Citric acid monohydrate	Millipore Sigma	C1909
dibasic sodium phosphate dihydrate	Millipore Sigma	71643
D-luciferin	Thermo Fisher Scientific	88294	Monopotassium Salt
DMEM	Gibco	11995065	High glucose and pyruvate
Ethanol	Millipore Sigma	459836
ethidium bromide	Thermo Fisher Scientific/Invitrogen	15585011
FBS	Gibco	26140079
loading dye	Thermo Fisher Scientific/Invitrogen	R0611
Luciferase siRNA	IDT	N/A	Antisense Strand Sequense: GAGGAGUUCAUUAUCAGUGC AAUUGUU Sense Strand Sequense: CAAUUGCACU GAUAAUGAACUCCT*C* *DNA bases
MDA-MB-231 / Luciferase (Bsd) stable cells	GenTarget Inc	SC059-Bsd	Luciferase-expressing cells sued to assess si-NP bioactivity
monobasic sodium phosphate monohydrate	Millipore Sigma	S9638
Scarmbled siRNA	IDT	N/A	Antisense Strand Sequense: AUACGCGUAUU AUACGCGAUUAACGAC Sense Strand Sequense: CGUUAAUCGCGUAUAAUAC GCGUA*T* *DNA bases
square polystyrene cuvettes	Fisher Scientific	14-955-129	4.5 mL capacity
TEM grids	Ted Pella, Inc.	1GC50	PELCO Center-Marked Grids, 50 mesh, 3.0mm O.D., Copper
Trisodium citrate dihydrate	Millipore Sigma	S1804
uranyl acetate	Polysciences, Inc.	21447-25