(250d) Inhalation Delivery of RNA-Loaded Lipid Nanoparticles Against Sars-Cov-2

Problem: Despite ongoing vaccination efforts, SARS-CoV-2 continues to present a significant global health burden. Furthermore, long-term immunity following vaccination is not guaranteed and the emergence of virus variants could further hinder the efficacy of this approach. These challenges suggest a continuing need for antiviral therapies that can reduce disease severity and lethality; however, available antiviral medications have shown limited impact on symptomatic disease. Efforts to identify drugs that are effective against SARS-CoV-2 have been complicated by its use of endogenous cellular components for viral replication. RNA interference (RNAi) is a promising alternative approach for antiviral therapy, where oligonucleotides with sequence homology to the viral RNA can selectively result in degradation of viral RNA transcripts and translational inhibition. RNAi sequences can be computationally determined based on well understood design rules; therefore, this approach could be rapidly adapted to different viruses or virus variants. Lipid nanoparticles (LNPs), a leading technology for RNAi delivery, protect RNA cargo from nucleolytic degradation and promote cellular uptake and endosome escape of RNA, leading to efficient gene knockdown. LNP formulations are usually administered through injection, whereas SARS-CoV-2 primarily infects epithelial cells in the respiratory tract, which are better accessed by topical administration via nebulization. A major obstacle in this delivery modality is the shear forces produced during nebulization can cause nanoparticle disruption and agglomeration, which may hinder the efficacy of nebulized LNPs. Here, we develop an RNAi strategy targeting SARS-CoV-2 and demonstrate conditions that stabilize LNPs during nebulization.

Methods: Lipid nanoparticles (LNPs) encapsulating RNA were prepared by microfluidic mixing, which enables production of particles that are about 100 nm in diameter with high RNA encapsulation efficiency (>90%). For nebulization studies, model RNA cargo was loaded in LNPs, which was either firefly luciferase-targeting siRNA (siLuc) or cyanine 5-labelled universal negative control siRNA (Cy5-NC). LNPs were diluted in buffer solutions of varying pH with or without selected excipients prior to nebulization. Nebulization of LNPs was achieved using an Omron MicroAir vibrating mesh nebulizer. Nebulizer output was collected in a conical tube and LNPs in this solution were evaluated in terms of hydrodynamic diameter, encapsulation efficiency and RNA recovery. Additionally, nebulized LNPs were assessed for cellular uptake, measured by flow cytometry (Cy5-NC-LNPs), and luciferase knockdown, determined by luminescence (siLuc-LNPs).

Computational tools were used to design a set of siRNAs that target SARS-CoV-2 transcripts. siRNA targets were inspected in the folded RNA transcript structure using RNAstructure software. To evaluate the potency of these siRNAs, plasmids were prepared that express enhanced green fluorescent protein (EGFP) flanked by the 3’ and 5’ untranslated regions (UTR) of SARS-CoV-2 under the control of a constitutive promoter. Plasmids representing the positive sense viral transcripts (5’UTR-EGFP-3’UTR) were chemically transfected into Vero cells, which are commonly used to grow SARS-CoV-2. Following drug selection, stably expressing uniform clones were identified by fluorescence microscopy and flow cytometry. Knockdown of viral RNA constructs could be quantified as a reduction in EGFP mean fluorescence intensity (MFI) measured by flow cytometry. siRNA against EGFP (siEGFP) dosed using lipofectamine or LNPs was used as a positive control in knockdown experiments to determine concentration (tested range 20 nM - 200 nM), duration and efficiency. Candidate siRNAs targeting the 5’UTR of SARS-CoV-2 were synthesized by Integrated DNA Technologies. The full set of designed siRNAs were evaluated by transfecting 50 nM with RNAiMAX Lipofectamine into transgenic Vero cells individually and in combination. Varying dosage concentrations (tested range 1 pM - 50 nM) were also evaluated. Promising siRNAs were encapsulated in LNPs prior to dosing Vero cells containing viral RNA-EGFP constructs. siRNAs were also evaluated in Vero cells infected with SARS-CoV-2 virus. Vero cells were treated with siRNA for 6 hours prior to viral infection. Sub-genomic and total viral RNA were quantified by quantitative polymerase chain reaction (qPCR) at 24 hours post-infection.

Results: Nebulization of siLuc-LNPs in phosphate buffered saline (PBS, pH 7.4) resulted in significant loss of siRNA encapsulation and more than a doubling in particle size. This confirmed that a stabilization strategy is needed for LNP nebulization. LNPs contain an ionizable lipid that is protonated at low pH. It was hypothesized that reducing nebulization buffer pH could impart particle stability by increasing surface charge and hence electrostatic repulsion between particles. Additionally, ionization of LNP lipids at low pH could increase retention of encapsulated RNA by maintaining electrostatic complexes between the ionizable lipid and the RNA phosphate backbone. Encapsulation efficiency was recovered to near pre-nebulization levels for LNPs nebulized at reduced pH. However, reduced nebulization buffer pH did not prevent LNP particle growth. Further, low pH buffer resulted in a reduction in total RNA recovery from the nebulizer, presumably due to adsorption of charged LNPs to the nebulizer plastic surface. To overcome these remaining issues, we evaluated several excipients to be included in the low pH nebulization buffer, including surfactants, sugars, and hydrophilic polymers. Of the screened excipients, we identified an excipient that largely prevented particle growth and significantly improved RNA recovery. We further validated that LNPs nebulized under these optimized conditions retained cellular uptake and efficient gene knockdown. Fluorescently labelled Cy5-NC-LNPs were taken up in Vero cells to a similar degree before and after nebulization. Additionally, siLuc-LNPs exhibited no significant change in luciferase knockdown potency when nebulized in the optimized conditions.

We computationally designed a set of ten siRNA oligonucleotides that target the UTR of SARS-CoV-2 RNA transcripts. Transgenic Vero cells expressing viral RNA-EGFP constructs were generated to evaluate siRNA potency. Generation of cell lines stably expressing plasmids was confirmed by observation of EGFP fluorescence by confocal microscopy and quantification of EGFP+ cells by flow cytometry. The cell line exhibiting the greatest proportion of EGFP+ cells (97%) was selected for evaluation of siRNAs. First, this system was validated by demonstrating efficient knockdown of the viral RNA-EGFP construct with siEGFP as a positive control. EGFP fluorescence was reduced by 68.2% or 56.2% when siEGFP was dosed with lipofectamine or LNPs, respectively. Next, we evaluated our set of ten candidate siRNAs in this viral construct model with transfection via lipofectamine. Of these siRNAs, five resulted in significant (>55%) EGFP knockdown. To test whether the siRNAs behave synergistically to knock down expression, we also transfected with combinations of the top candidates; this approach yielded no beneficial effect. Further studies evaluating different siRNA doses allowed the identification of three promising siRNA sequences for further study. These three siRNAs also exhibited efficient knockdown of viral RNA-EGFP constructs when encapsulated in LNPs. LNP delivery of as little as 10 nM siRNA yielded >60% knockdown. Importantly, in Vero cells infected with SARS-CoV-2 virus, pre-infection treatment with one siRNA candidate via lipofectamine resulted in a 2-log reduction in sub-genomic viral RNA compared to untreated cells.

Implications: Overall, inhalation RNAi therapy is a promising strategy to manage viral infections. We demonstrate an approach to stabilize LNPs during nebulization which will enable RNAi therapies against SARS-CoV-2 and other respiratory viruses. Insights gained from these studies may be applicable for stabilization of nanoparticle systems for pulmonary delivery. siRNAs designed in the present work showed promise through knockdown of model viral RNA constructs in vitro. Further studies will evaluate this delivery strategy for efficacy in an in vivo model of viral infection. The ability to effectively deliver potent RNAi via inhalation could be broadly clinically useful for a range of respiratory diseases.