Lung SORT LNPs enable precise homology-directed repair mediated CRISPR/Cas genome correction in cystic fibrosis models

Abstract

Approximately 10% of Cystic Fibrosis (CF) patients, particularly those with CF transmembrane conductance regulator (CFTR) gene nonsense mutations, lack effective treatments. The potential of gene correction therapy through delivery of the CRISPR/Cas system to CF-relevant organs/cells is hindered by the lack of efficient genome editor delivery carriers. Herein, we report improved Lung Selective Organ Targeting Lipid Nanoparticles (SORT LNPs) for efficient delivery of Cas9 mRNA, sgRNA, and donor ssDNA templates, enabling precise homology-directed repair-mediated gene correction in CF models. Optimized Lung SORT LNPs deliver mRNA to lung basal cells in Ai9 reporter mice. SORT LNP treatment successfully corrected the CFTR mutations in homozygous G542X mice and in patient-derived human bronchial epithelial cells with homozygous F508del mutations, leading to the restoration of CFTR protein expression and chloride transport function. This proof-of-concept study will contribute to accelerating the clinical development of mRNA LNPs for CF treatment through CRISPR/Cas gene correction.

Fig. 1. Schematic illustration of SORT LNP-mediated gene correction therapy in CF mouse models.

SORT LNPs encapsulating three nucleic acid cargos (Cas9 mRNA, sgRNA, and ssDNA HDR template) effectively corrected CFTR gene mutations in a G542X CF mouse model (in vivo) and in a patient-derived F508del HBE cell model (ex vivo).

Fig. 2. Optimized DOTAP LNP formulation achieved enhanced mRNA delivery in mouse lungs with low toxicity.

a mDLNP-2 was optimized by adjusting the internal molar ratio among 5A2-SC8/DOPE/Chol/PEG-DMG to 36/20/40/4. b Higher luciferase expression in mouse liver was observed using mDLNP-2 compared to mDLNP-1, measured using IVIS (0.1 mg kg⁻¹ Luc mRNA, i.v., 6 h). c Quantification data further confirmed enhanced mRNA delivery efficiency in mouse liver using mDLNP-2. Data are shown as mean ± s.e.m. (n = 3 biologically independent animals). d Permanently cationic lipid structures were screened for lung-targeting LNPs using IVIS. A series of 5A2-SC8 LNPs with 20% to 50% different permanently cationic lipids (fraction of total lipids) were prepared and administered into mice for testing (0.1 mg kg⁻¹ Luc mRNA, i.v., 6 h). e Quantification data suggested that DOTAP LNPs and DDAB LNPs showed higher luciferase expression in mouse lungs. f The lung-targeting specificity was evaluated by calculating the relative luciferase expression of the Lung/Liver and Lung/Spleen. DOTAP40 and DDAB30 LNPs were selected by comparing lung delivery efficacy and specificity. Data are shown as mean ± s.e.m. (n = 3 biologically independent animals). g In vivo toxicity assessments revealed that DOTAP40 LNPs are less toxic compared to DDAB30 LNPs (1.5 mg kg⁻¹ mCherry mRNA, i.v., 40/1(total lipid/total RNA)), as indicated by liver function (ALT and AST) after 24 h of treatment (g), body weight changes after the treatment (h) and organ-to-body weight ratio (i). PBS treatment (i.v.) was used as a negative control, and lipopolysaccharide (LPS, i.p., 5 mg kg⁻¹) treatment was used as a positive control. LPS group mice were sacrificed at day 2, other groups were sacrificed at day 5. Data of g, h, i are shown as mean ± s.e.m. (n = 5 biologically independent animals). Source data are provided as a Source Data file. A two-tailed unpaired t-test was used to determine the significance of the comparisons of data (*P <  0.05; **P <  0.01; ***P <  0.001).

Fig. 3. DOTAP10 LNPs, encapsulating Cas9 mRNA, sgRNA, and ssDNA HDR template, successfully induced HDR in HEK293 cells with Y66H GFP mutation.

a The HEK293 cells have a Y66H mutation in the GFP sequence, which alters the fluorescence from GFP to BFP. Once corrected, the fluorescence will turn back to GFP. The gene editing efficiency was analyzed using TIDER analysis with DNA sequencing data obtained after testing with a series of DOTAP10 LNPs containing Cas9 mRNA, sgRNA, and donor ssDNA template with Cas9:sgRNA weight ratio fixed at 1:1 (b), then with sgRNA:HDR weight ratio fixed at 1:6 (total NA at 0.8 ng μL⁻¹) (c). The optimal weight ratio among Cas9 mRNA, sgRNA and HDR was 0.5/1/6 (named as DOTAP10-HDR). Data of b, c are shown as mean±s.e.m. (n = 3 biologically independent samples). d, DNA sequencing result after treatment of DOTAP10-HDR demonstrated that the editing area was successfully corrected from “CCAT” to “GTAC”. e DOTAP10-HDR efficiently corrected Y66H mutation in HEK293 cells and turned on bright GFP fluorescence. NC group (PBS) and DOTAP10 LNPs encapsulating Cas9 mRNA and sgRNA (named as DOTAP10-NHEJ) group were used as controls. The total NA was fixed at 0.4 ng μL⁻¹. Scale bar: 100 μm. The data was repeated three times independently with similar results. f Ai14 mice were used to evaluate Lung SORT LNP-mediated CRISPR/Cas gene editing in vivo. tdTOM mice were treated once a week with (g) DOTAP40-NHEJ (Cas9 mRNA:sgTOM1 = 2:1) or (h) DOTAP40-HDR (Cas9 mRNA: sgTOM1: HDR template = 2:1:3) formulations with a total NA at 1 mg kg⁻¹ IV. One week following the third injection, mice were sacrificed, and the lungs were collected for flow cytometry analysis to determine the proportion of tdTOM⁺ cells in different cell types. There is a significant difference in tdTom⁺ between bulk lung and lung epithelium cell populations with a P value equal to 0.0218 (DOTAP40-HDR). Data are shown as mean ± s.e.m. (n = 3 biologically independent animals). A two-tailed unpaired t-test was used to determine the significance of the comparisons of data (*P <  0.05; **P <  0.01; ***P <  0.001; ****P <  0.0001). Source data of b, c, g, h are provided in Source Data file.

Fig. 4. DOTAP LNPs successfully corrected CFTR gene mutation in G542X-CF mouse model and restored CFTR function in a mouse intestinal organoid cell model with the G542X mutation.

a Optimized DOTAP40 LNPs showed higher luciferase expression in mouse lungs, compared to previously reported Lung SORT LNPs, and mRNA delivery efficiency remained comparable when the weight ratio of total lipid/total NA was reduced from 40/1 to 20/1 (0.1 mg kg⁻¹ Luc mRNA, n = 3 biologically independent animals). b DOTAP40 LNPs efficiently delivered Cre mRNA (2 mg kg⁻¹ Cre mRNA, 20/1) to basal cells of tdTOM mouse lung, activating tdTOM fluorescence. Two days after the second injection, the tdTOM positive cell in the mouse lung basal cell (NGFR + ) population was analyzed using flow cytometry (n = 4 biologically independent animals in DOTAP40-Cre group). NP only group was used as negative control (NC). c DOTAP40-HDR LNPs encapsulating Cas9 mRNA, sgRNA, and HDR template were I.V. injected into CF mouse containing G542X mutation over three weeks (2 mg kg⁻¹ total NA, 20:1). d DOTAP40-HDR treatment successfully corrected G542X mutation in mouse lungs analyzed by NGS deep sequencing and CRISPResso2 analysis. NP only group was used as negative control (NC). Data are shown as mean±s.e.m. (n = 6 biologically independent animals). e Representative sequencing result showed the successful HDR correction event at G542X mutation site was observed in mouse lungs after DOTAP40-HDR, by subsequent cloning and DNA sequencing. f Schematic illustration shows the mechanism of ex vivo intestinal organoid based forksolin-induced swelling (FIS) assay. g Intestinal organoid swelling was observed after treatment of DOTAP10-HDR, suggesting successful CFTR function restoration. No organoid swelling was detected after NC (NP only) treatment. Scale bar: 500 μm. The data was repeated three times independently with similar results. More than 20% of organoids swelled (h) and successful HDR correction by Sanger sequencing (i) were detected after treated with DOTAP10-HDR. Data are shown as mean ± s.e.m. (n = 8 biologically independent samples). Two-tailed unpaired t-tests were used to determine the significance of the comparisons of data (*P <  0.05; **P <  0.01; ***P < 0.001). Source data are provided as a Source Data File.

Fig. 5. DOTAP LNPs successfully corrected CFTR mutation and restored CFTR function in patient-derived human bronchial epithelial (HBE) cells with F508del mutation.

a Undifferentiated HBE cells with F508del mutation (passage 1, P1) were treated with DOTAP10-HDR for 4 days before cell expansion (passage 2, P2). One week after cell expansion, cells were established on transwell inserts for cell differentiation (passage 3, P3). The untreated group was used as negative control (NC) and HBE cells from a healthy donor expressing wild-type (WT) CFTR was used as positive control. b DOTAP10 LNPs encapsulating mCherry mRNA (DOTAP10-mCherry) were efficiently internalized into P1 HBE cells, showing bright mCherry fluorescence. Scale bar: 500 μm. c, The HDR correction efficiency maintained at as high as 16% after a single DOTAP10-HDR treatment, as evaluated in P1, P2, and P3 HBE cells. DOTAP10-HDR treatment successfully restored CFTR activity, compared to WT CFTR activity (d), as confirmed by area under the curve (AUC) analysis (e). Data are shown as mean ± s.e.m. (n = 4–6 biologically independent samples). f Expression of CFTR protein was detected after treatment of DOTAP10-HDR (1, 2 means repeats). Co-treatment of DOTAP10-HDR and VX-809 enhanced Cl^- transport and improved CFTR function (g), with corresponding AUC values (h). There was a significant difference between VX-809 and DOTAP10-HDR + VX-809 groups with a P value < 0.0001. Data are shown as mean ± s.e.m. (n = 4 biologically independent samples). i Increased expression of CFTR protein was detected after co-treatment of DOTAP10-HDR + VX-809. j, DOTAP10-HDR were directly added onto the surface of differentiated HBE cells. k DOTAP10-mCherry was efficiently internalized into fully differentiated HBE cells, resulting in bright mCherry fluorescence. Scale bar: 500 μm. l DOTAP10-HDR successfully corrected F508del mutation in differentiated HBE cells. Data are shown as mean ± s.e.m. (n = 5 biologically independent samples). m Co-treatment of DOTAP10-HDR and VX-809 significantly enhanced Cl^- transport and improved CFTR function in differentiated HBE cells, as confirmed with corresponding AUC values (P = 0.0114) (n). Data are shown as mean ± s.e.m. (n = 3 biologically independent samples). Two-tailed unpaired t-tests were used to determine the significance (*P <  0.05; **P <  0.01; ***P <  0.001; ****P <  0.0001). Data of (b, k) were repeated three times independently with similar results.