Design and screening of novel endosomal escape compounds that enhance functional delivery of oligonucleotides in vitro

Abstract

Antisense oligonucleotides (ASOs), including splice-switching oligonucleotides (SSOs), are promising therapeutic approaches for targeting genetic defects. ASOs act in the nucleus and the cytosol to cleave mRNAs via the RNaseH1 mechanism (e.g., gapmers), while SSOs alter transcript splicing to restore or inhibit protein function. RNA interference (RNAi) is another approach to down-regulate gene expression via the RISC complex. However, a major challenge is the effective delivery of these nucleic acid-based therapeutics. Recent developments focus on enhancing cellular uptake and endosomal release, including the use of small-molecule endosomal escape enhancers (EEEs) such as chloroquine. Here, we disclose a next generation of EEEs, which efficiently enhance SSOs and gapmers in vitro activity. We identify proton sponge-mediated endosomal leakage as a mechanism of action and observe, by Gene Ontology analysis on bulk RNA sequencing, that EEE treatment increased gene expression of markers associated with vesicle organization. Additionally, using primary human hepatocytes, we demonstrate that EEEs enhance small interfering RNA (siRNA) activity. Unconjugated siRNA reached similar levels of mRNA knockdown to the observed GalNAc-conjugated siRNA. Substantial GalNAc conjugated siRNA enhancement was also observed when used together with EEE. Our results indicate that these EEEs constitute a promising strategy to enhance the activity of multimodal oligonucleotide therapeutics.

Keywords: MT: Oligonucleotides: Therapies and Applications; antisense; endosomes; enhancers; leakage; lysosomes; mitophagy; oligonucleotides; small molecules; therapeutic.

Graphical abstract

Figure 1

Structure activity relationship studies led to the design of new compounds that can enhance endosomal escape of SSOs (A) Main strategy and chemical substitutions used for the structure activity relationship analysis. (B) Representative EEE structures generated. The boxes highlight leading EEEs. (B and C) Screening of functional SSO delivery in HeLa_Luc705 cells. Cells were pretreated with SSOs for 24 h, followed by 24-h EEE treatment at the indicated concentration, rinsed, 0.1% Triton X-100 lysed, and luciferase intensity measured. When indicated, cell proliferation reagent WST-1 (1:10 ratio) was used, the 4-h pre-incubation step was included, and absorbance measured against background control. Bars indicate mean with SD. (D) Representative images from eight cell lines treated with EEE displaying GAL9 translocation, from cytosolic to punctate. GAL9 events are observed as puncta (yellow), and nuclei were Hoechst stained (blue); scale bars, 5 μm. GAL9 maximum quantitation is displayed as nuclei normalized signal. Significance was calculated for single variable using the Mann-Whitney U test and for multiple variables comparison t test using the Sidak-Bonferroni correction method (α = 0.05) (∗p < 0.05; ∗∗p < 0.005; ∗∗∗p < 0.0005; ∗∗∗∗p < 0.00005).

Figure 2

Functional studies of EEE4 as an endosomal escape enhancer (A) Efficient EEE4 concentration showed no to mild-acute cell toxicity upon 1- to 24-h treatment. For luciferase expression analysis, SSO-pretreated cells were EEE treated and then rinsed, 0.1% Triton X-100 lysed, and luciferase intensity measured. Conditioned media was collected 1 or 24 h after treatment initiated for lactate dehydrogenase cell toxicity assay and performed according to the manufacturer’s instructions. (B) (Bar plots) Selected Gene Ontology term analysis on bulk RNA sequencing data, fold enrichments, and p values for genes upregulated by 2.5-μM (light gray) or 5-μM (dark gray) treatments. (Plot) Force directed t-stochastic neighbor embedding (t-SNE) nearest neighbor plot showing connections between each experiment and its four nearest neighbors. (C) EEE4 potency is dependent on endosomal acidification. H⁺ATPase inhibition was performed by EEE4 + bafilomycin (0.5 μM) co-incubation. (D) The 24-h SSO pre-incubation enhanced endosomal escape EEE4 activity. Cells were either SSO + EEE4 cotreated or 24 h SSO pre-incubated and then EEE4 treated, followed by rinse, lysis, and luciferase readout at five different time points after EEE4 treatment. (A, C, and D) Bars indicate mean and SD. Significance was calculated for single variable using the Mann-Whitney U test (∗p < 0.05; ∗∗p < 0.005; ∗∗∗p < 0.0005).

Figure 3

EEE4 piperidine chemical substitutions are well tolerated and have positive impact on cell viability (A) HeLa_Luc705 cells were SSO pretreated, followed by either 1- or 24-h EEE treatment, rinse, lysis and luciferase measurements. (B) Conditioned media was collected 1 or 24 h after treatment initiated for lactate dehydrogenase cell toxicity assay and performed according to the manufacturer’s instructions. For tetrazolium salt WST-1 assay, the cytotoxic effect was measured 1 or 24 h after treatment initiation, considering WST-1 prior incubation time (4 h) according to the manufacturer’s instructions. Bars indicate mean and SD.

Figure 4

EEEs as endosomal escape enhancer for SSO and gapmer (A) Similar to EEE4, 24-h SSO pre-incubation enhanced EEE32-induced endosomal escape activity. Cells were either cotreated or 24-h SSO pre-incubated and then EEE4 treated, followed by rinse, lysis, and luciferase readout at indicated time points. (B) No changes in luciferase expression were observed upon SSO delivery with EEE32 enantiomers. HeLa_Luc705 cells were treated and luciferase expression measured as previously described. (C) Cells were seeded, 24 h after ASO + EEE treatment, followed by rinse and sample collection at 24 h. Dose response in moles for MALAT1 ASO (1e−5, 3e−5, …, 3e−9, 1e−9). Bars indicate mean with SD. Significance for multiple variables comparison was calculated using multiple t test Sidak-Bonferroni correction method (α = 0.05) (n.s, not significant; ∗p < 0.05).

Figure 5

EEEs as endosomal escape enhancer for siRNA therapeutics (A–D) Cells were seeded 24 h after siRNA + EEE treatment, followed by rinse and sample collection at 48 h. (A, B, and D) Dose response for siRNA (1e−6, 1e−7, …, 1e−12). (C) Dose response for EEE 1, 2, 2.5, 3, 3.5, 4, and 5 μM. The heatmap shows RNA yield average of three replicates, calculated as (measured sample yield/measured control yield). Bars indicate mean with SD. Significance for multiple variables comparison was calculated using multiple t test Sidak-Bonferroni correction method (α = 0.05) (n.s, not significant; ∗p < 0.05).