testing-vitro-vivo-efficiency-mrna-lipid-nanoparticles-formulated

Testing the <em>In Vitro</em> and <em>In Vivo</em> Efficiency of mRNA-Lipid Nanoparticles Formulated by Microfluidic Mixing

Lipid nanoparticles (LNPs) have attracted widespread attention recently with the successful development of the COVID-19 mRNA vaccines by Moderna and ...

University of Pennsylvania

Messenger RNA (mRNA) has recently emerged as a promising class of nucleic acid therapy, with the potential to induce protein production to treat and prevent a range of diseases. However, the widespread use of mRNA as a therapeutic requires safe and effective in vivo delivery technologies. Libraries of ionizable lipid nanoparticles (LNPs) have been designed to encapsulate mRNA, prevent its degradation, and mediate intracellular delivery. However, these LNPs are typically characterized and screened in an in vitro setting, which may not fully replicate the biological barriers that they encounter in vivo. Here, we designed and evaluated a library of engineered LNPs containing barcoded mRNA (b-mRNA) to accelerate the screening of mRNA delivery platforms in vivo. These b-mRNA are similar in structure and function to regular mRNA, and contain barcodes that enable their delivery to be quantified via deep sequencing. Using a mini-library of b-mRNA LNPs formulated via microfluidic mixing, we show that these different formulations can be pooled together, administered intravenously into mice as a single pool, and their delivery to multiple organs (liver, spleen, brain, lung, heart, kidney, pancreas, and muscle) can be quantified simultaneously using deep sequencing. In the context of liver and spleen delivery, LNPs that exhibited high b-mRNA delivery also yielded high luciferase expression, indicating that this platform can identify lead LNP candidates as well as optimal formulation parameters for in vivo mRNA delivery. Interestingly, LNPs with identical formulation parameters that encapsulated different types of nucleic acid barcodes (b-mRNA versus a DNA barcode) altered in vivo delivery, suggesting that the structure of the barcoded nucleic acid affects LNP in vivo delivery. This platform, which enables direct barcoding and subsequent quantification of a functional mRNA, can accelerate the in vivo screening and design of LNPs for mRNA therapeutic applications such as CRISPR-Cas9 gene editing, mRNA vaccination, and other mRNA-based regenerative medicine and protein replacement therapies.

Ionizable lipid nanoparticles encapsulating barcoded mRNA for accelerated in vivo delivery screening.

Nucleic acids, such as messenger RNAs, antisense oligonucleotides, and short interfering RNAs, hold great promise for treating previously 'undruggable' diseases. However, there are numerous biological barriers that hinder nucleic acid delivery to target cells and tissues. While lipid nanoparticles (LNPs) have been developed to protect nucleic acids from degradation and mediate their intracellular delivery, it is challenging to predict how alterations in LNP formulation parameters influence delivery to different organs. In this study, we utilized high-throughput in vivo screening to probe for structure-function relationships of intravenously administered LNPs along with quartz crystal microbalance with dissipation monitoring (QCM-D) to measure the binding affinity of LNPs to apolipoprotein E (ApoE), a protein implicated in the clearance and uptake of lipoproteins by the liver. High-throughput in vivo screening of a library consisting of 96 LNPs identified several formulations containing the helper lipid 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine (DOPE) that preferentially accumulated in the liver, while identical LNPs that substituted DOPE with the helper lipid 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC) preferentially accumulated in the spleen. Using QCM-D, it was found that one DOPE-containing LNP formulation (LNP 42) had stronger interactions with ApoE than an identical LNP formulation that substituted DOPE with DSPC (LNP 90). In order to further validate our findings, we formulated LNP 42 and LNP 90 to encapsulate Cy3-siRNA or mRNA encoding for firefly luciferase. The DSPC-containing LNP (LNP 90) was found to increase delivery to the spleen for both siRNA (two-fold) and mRNA (five-fold). In terms of liver delivery, the DOPE-containing LNP (LNP 42) enhanced mRNA delivery to the liver by two-fold and improved liver transfection by three-fold. Understanding the role of the helper lipid in LNP biodistribution and ApoE adsorption may aid in the future design of LNPs for nucleic acid therapeutics.

Helper lipid structure influences protein adsorption and delivery of lipid nanoparticles to spleen and liver.

In recent years, immune cell-based cancer therapeutics have been utilized broadly in the clinic. Through advances in cellular engineering, chimeric antigen receptor (CAR) T-cell therapies have demonstrated substantial success in treating hematological tumors and have become the most prominent cell-based therapy with three commercialized products in the market. However, T-cell-based immunotherapies have certain limitations, including a restriction to autologous cell sources to avoid severe side-effects caused by human leukocyte antigen (HLA) mismatch. This necessity for personalized treatment inevitably results in tremendous manufacturing and time costs, reducing accessibility for many patients. As an alternative strategy, natural killer (NK) cells have emerged as potential candidates for improved cell-based immunotherapies. NK cells are capable of killing cancer cells directly without requiring HLA matching. Furthermore, NK cell-based therapies can use various allogeneic cell sources, allowing for the possibility of "off-the-shelf" immunotherapies with reduced side-effects and shortened manufacturing times. Here we provide an overview of the use of NK cells in cancer immunotherapy, their current status in clinical trials, as well as the design and implementation of delivery technologies-including viral, non-viral, and nanoparticle-based approaches-for engineering NK cell-based immunotherapies.

Delivery technologies to engineer natural killer cells for cancer immunotherapy.

A major challenge to advance lipid nanoparticles (LNPs) for RNA therapeutics is the development of formulations that can be produced reliably across the various scales of drug development. Microfluidics can generate LNPs with precisely defined properties, but have been limited by challenges in scaling throughput. To address this challenge, we present a scalable, parallelized microfluidic device (PMD) that incorporates an array of 128 mixing channels that operate simultaneously. The PMD achieves a >100× production rate compared to single microfluidic channels, without sacrificing desirable LNP physical properties and potency typical of microfluidic-generated LNPs. In mice, we show superior delivery of LNPs encapsulating either Factor VII siRNA or luciferase-encoding mRNA generated using a PMD compared to conventional mixing, with a 4-fold increase in hepatic gene silencing and 5-fold increase in luciferase expression, respectively. These results suggest that this PMD can generate scalable and reproducible LNP formulations needed for emerging clinical applications, including RNA therapeutics and vaccines.

Scalable mRNA and siRNA Lipid Nanoparticle Production Using a Parallelized Microfluidic Device.

Myocardial infarction (MI) is a leading cause of death worldwide, largely because efficient interventions to restore cardiac function after MI are currently lacking. Here, we characterize vascular aberrancies induced by MI, and propose to target acquired endothelial cell (EC) changes to normalize vessels and promote cardiac repair after MI. Single-cell transcriptome analyses of MI-associated ECs indicates that ECs acquire mesenchymal gene signature that result in phenotypic and functional changes and lead to vessel abnormalities. We identify a PDGF/NF-κB/HIF-1α axis that induces Snail expression and mesenchymal phenotypes in ECs under hypoxia, altogether causing aberrant vascularization. EC-specific knockout of PDGFR-β, pharmacological PDGFR inhibition or nanoparticle-based targeted PDGFR-β siRNA delivery in mice attenuates vascular abnormalities in the infarcted tissue and improves cardiac repair after MI. These findings illustrate a mechanism controlling aberrant neovascularization after ischemia, and suggest that targeting PDGF/Snail-mediated endothelial plasticity may offer opportunities for normalizing vasculature and treating ischemic heart diseases.

Endothelial plasticity drives aberrant vascularization and impedes cardiac repair after myocardial infarction.

Lipid nanoparticle-mediated RNA delivery holds great potential to treat various liver diseases. However, targeted delivery of RNA therapeutics to activated liver-resident fibroblasts for liver fibrosis treatment remains challenging. Here, we develop a combinatorial library of anisamide ligand-tethered lipidoids (AA-lipidoids) using a one-pot, two-step modular synthetic method and adopt a two-round screening strategy to identify AA-lipidoids with both high potency and selectivity to deliver RNA payloads to activated fibroblasts. The lead AA-lipidoid AA-T3A-C12 mediates greater RNA delivery and transfection of activated fibroblasts than its analog without anisamide and the FDA-approved MC3 ionizable lipid. In a preclinical model of liver fibrosis, AA-T3A-C12 enables ~65% silencing of heat shock protein 47, a therapeutic target primarily expressed by activated fibroblasts, which is 2-fold more potent than MC3, leading to significantly reduced collagen deposition and liver fibrosis. These results demonstrate the potential of AA-lipidoids for targeted RNA delivery to activated fibroblasts. Furthermore, these synthetic methods and screening strategies open a new avenue to develop and discover potent lipidoids with targeting properties, which can potentially enable RNA delivery to a range of cell and tissue types that are challenging to access using traditional lipid nanoparticle formulations.

Rakan El-Mayta

Department of Bioengineering

University of Pennsylvania

Testing the In Vitro and In Vivo Efficiency of mRNA-Lipid Nanoparticles Formulated by Microfluidic Mixing