Supramolecular assembly of polycation/mRNA nanoparticles and in vivo monocyte programming

Abstract

Size-dependent phagocytosis is a well-characterized phenomenon in monocytes and macrophages. However, this size effect for preferential gene delivery to these important cell targets has not been fully exploited because commonly adopted stabilization methods for electrostatically complexed nucleic acid nanoparticles, such as PEGylation and charge repulsion, typically arrest the vehicle size below 200 nm. Here, we bridge the technical gap in scalable synthesis of larger submicron gene delivery vehicles by electrostatic self-assembly of charged nanoparticles, facilitated by a polymer structurally designed to modulate internanoparticle Coulombic and van der Waals forces. Specifically, our strategy permits controlled assembly of small poly(β-amino ester)/messenger ribonucleic acid (mRNA) nanoparticles into particles with a size that is kinetically tunable between 200 and 1,000 nm with high colloidal stability in physiological media. We found that assembled particles with an average size of 400 nm safely and most efficiently transfect monocytes following intravenous administration and mediate their differentiation into macrophages in the periphery. When a CpG adjuvant is co-loaded into the particles with an antigen mRNA, the monocytes differentiate into inflammatory dendritic cells and prime adaptive anticancer immunity in the tumor-draining lymph node. This platform technology offers a unique ligand-independent, particle-size-mediated strategy for preferential mRNA delivery and enables therapeutic paradigms via monocyte programming.

Keywords: mRNA delivery; monocyte transfection; nanoparticle assembly; poly(beta-amino ester); vehicle size-mediated passive targeting.

Fig. 1.

Supramolecular assembly of PBAE/mRNA particles. (A) PBAE structure used in this study. Orange shading: hydrophobic structures predicted; Red shading: positively charged amines at pH 5.0. (B) TEM images of PBAE/mRNA nanoparticles at sub-100 nm size immediately upon mixing PBAE-containing alcohol solution with aqueous mRNA solution at pH = 5.0 and ionic strength I = 0.060 M. (C) Particle size change induced by altering solution conditions, monitored by consecutive DLS measurements. (D) Proposed supramolecular assembly mechanism. (E) Colloidal stability of particles by quenching by PEGylated lipid. Inset: bulk appearances of 100- or 400-nm particles. (F) Zeta potential and further size growth over 30 min upon addition of DMG-PEG at different weight ratios. (G) TEM images of stabilized particles at controlled sizes determined by z-average (D_z) from DLS. (H) Schematic overview of the cylindrical illumination confocal spectroscopy technique used to characterize the single-particle mRNA payload. (I) The fluorescent intensity distributions for particles with D_z below 200 nm (carrying 100% Cy5-mRNA). (J) The fluorescent intensity distributions for particles with D_z above 200 nm (carrying 5% w/w Cy5-mRNA). (K) The relationship between the weight average single-particle mRNA payload and D_z. In (C), n = 3 independent experiments for condition 4, and n = 1 experiment for all other conditions; in (E and F), n = 3 independent experiments; In (C and F), data are presented as mean ± SD.

Fig. 2.

Molecular dynamics simulation of supramolecular assembly of PBAE/mRNA nanoparticles. (A) Nanoparticle composition obtained in molecular dynamics simulations, as a function of the radial distance from the center of mass of mRNA at pH 5.0 and low ionic strength. (B) The state of protonation at different solution conditions. (C) Nanoparticle composition as a function of the radial distance after switching to pH 7.4 and high ionic strength. (D and E) Cryo-TEM images showing nanoparticle merging events after snap freezing particle samples at either 90 min following solution replacement into pH 7.4 and 0.125 M buffer (D) or 5 min following solution replacement into pH 7.4 and 0.205 M buffer (E), without adding DMG-PEG. (F) CLSM illuminating particles carrying Cy3-siRNA and Cy5-mRNA following uptake by HeLa cells. (G) Co-localization degree, indicative of co-encapsulation efficiency, quantified using the ImageJ Colocalization Finder plugin. The Inset of (A and C) shows the relationship between protonated and unprotonated amines from the center to the periphery of the nanoparticle. The schematics in the Inset depict a representative nanoparticle at either condition. Blue bead: negative charge on mRNA; Red bead: positive charge on PBAE; White bead: uncharged chain component. In (B), the data are presented as mean ± SEM derived from 57 replica of simulation with a single particle.

Fig. 3.

400 to 900-nm particles but not smaller nanoparticles enabled efficient transfection of primary murine macrophages in vitro. (A) Transfection efficiency of PBAE/mGFP particles at different particle size D_z evaluated in either HeLa (Upper) or in murine BMDMs (Lower). (B) Scatter plots showing GFP expression in BMDMs following transfection at 0.4 μg mRNA/mL. (C and D) Percent GFP⁺ cells (C) and single-cell expression level quantified by the GFP mean fluorescent intensity (MFI) in GFP⁺ cells (D) in BMDMs. (E and F) Effect of D_z on cellular uptake efficiency by dosing BMDMs with particles carrying Cy5-mRNA, showing proportion of Cy5⁺ cells (E) and single-cell uptake level indicated by Cy5 MFI in Cy5⁺ cells (F). (G) Viability of BMDMs after 4 h of transfection. (H) CLSM images of BMDMs after 8 h of transfection. (Scale bar, 10 μm.) The arrows point to colocalization events. (I) Fold change of the number of BMDMs positive for expression of various proinflammatory markers at 24 h after different treatments, in reference to nontreated cells. mOVA and mGFP denote mRNA encoding ovalbumin and GFP, respectively; ODN denotes CpG oligonucleotide 1826. (C–F) share the same legend, and n = 3 technical replicates; Data are presented as mean ± SD.

Fig. 4.

Biocompatibility, biodistribution, and delivery profiles of supramolecularly assembled PBAE/mRNA particles upon intravenous administration. (A) Representative histology images of lungs (Green arrow: pulmonary vein; Yellow arrow: pulmonary capillary) and kidneys (Green arrow: glomerulus; Yellow arrow: peritubular capillaries) from CD-1 mice treated with 400-nm particles or PBS, analyzed at 72 h following a single injection of particles at 0.5 mg mRNA/kg. (B) In vivo biodistribution of Cy7-labeled PBAE/mRNA particles at 2 h after injection, measured by IVIS imaging of tissues ex vivo. (C) Gating strategy for identification of 1. Ly6C^high monocytes, 2. Ly6C^low monocytes, and 3. Ly6C^–CD11b⁺ myeloid cells isolated from the blood. (D) Cellular uptake of Cy5-mRNA-encapsulated 100 or 400-nm particles by Ly6C^high monocytes collected at 1 min postinjection. (E) Relative cell abundance changes in the blood at 1 h compared to 1 min. (F and G) Relative cell counts of neutrophils (F) and subtypes of monocytes (G) following i.v. administration of 400-nm particles over 72 h. (H) Scatter plot showing inflammatory monocytes (Population A: CX3CR1^-Ly6C^high) and patrolling monocytes (B: CX3CR1⁺Ly6C^low/–) in the liver 2 h postinjection of 400-nm particles. (I and J) Uptake levels of 400-nm or 100-nm particles by various cell types, showing uptake rate (I) and single-cell uptake level (J). (K) Relative cell counts of mature macrophages (CD11b⁺F4/80⁺) in the liver (Population C) and spleen (Population D) measured at 72 h post injection. (L) Ly6C expression profile among the enriched macrophage population in the liver. (M) Relative cell counts of myeloid-origin DCs (Population E: CD11b⁺ CD11c⁺) detected in the liver at 72 h post injection. For (A) and (B), one representative image out of 4 mice within each group is shown. For the phenotyping plots in (C), (E), and (F–M), data points from all mice within the same group (in most cases, n = 4) are concatenated. For (F), (G), and (K), the parent gate is CD45⁺ Ly6G^–/low. For (M), the parent gate is CD45⁺Ly6G^−/lowF4/80⁻. Data are presented as mean ± SD.

Fig. 5.

mRNA expression profiles upon intravenous administration of PBAE/mRNA particles. (A) Scattered plots quantifying tdTomato⁺ cells (i.e., % transfected) among Ly6C^high inflammatory monocytes at 24 h, induced F4/80⁺ macrophages at 72 h and induced myeloid-origin DCs at 72 h in Ai-9 mice following injection of mCre-loaded particles. (B–E) Transfection efficiency (% transfected) within each immune cell type (B and C), and proportion of transfected cells within a particular cell type relative to all counts of transfected cells for 400-nm particles (D and E), for cells isolated from the liver (B and D) and spleen (C and E). (F and G) Transgene expression levels in major organs after i.v. injection of mLuc-carrying particles imaged by IVIS at 24 h (F) and quantified in homogenized organs (G). In (A), data from all animals within the same group were concatenated; in (D and E), average results from each group are plotted. In (F), one representative mouse from four mice within each group is shown.

Fig. 6.

Proof-of-concept monocyte-mediated anti-tumor activity in the B16-OVA model, and proposed schematic overview of the delivery process of PBAE/mRNA particles. (A) Tumor cell inoculation and particle dosing schedule. (B) Survival plots of mice in different treatment groups. The control mRNA (mCtrl) refers to mLuc. (C) Percentages of OVA-specific T cells detected in the tumor (Left) and the tumor-draining lymph node (dLN, Right). (D) The gating strategy to identify Mo-DCs in dLN (Population 1: CD11b⁺CD11c⁺MHC^-II^int) showing their characteristic MHC-II and Ly6C expression levels. (E) Comparison of the abundance of the induced Mo-DCs among different treatment groups. (F) CD209a⁺ and TNF-α levels of Mo-DCs. (G) Comparison of the abundance of populations by their MHC-II or Ly6C expression levels, as defined in (D). (H) The responses to different treatment groups characterized by comprehensive phenotyping panels specified in SI Appendix, Table S1. White bar: value higher than the highest presented by the scale. In (B) n = 8 (4 male, 4 female mice); In (D–H), n = 4 (2 male, 2 female mice), and data from all mice in the same group are concatenated to generate the density plots; Data are presented as mean ± SD. (I and J) Schematic overview of the mRNA delivery process mediated by PBAE/mRNA particles (generated using BioRender.com).