Description of the data and file structure

CRISPR/Cas9-based gene-editing technologies offer promise for treating inherited retinal diseases (IRDs); however, safe and efficient ocular delivery of precision editors remains challenging. To address this challenge, we report a new class of Coomassie brilliant blue (CBB)-derived lipidoids that bind and deliver proteins. Subretinal injection of Cre complexed with these lipidoids into mT/mG mice leads to robust recombination in the retinal pigment epithelium and photoreceptors. We employ the CBB-lipidoid platform to deliver adenine base editor (ABE) ribonucleoproteins (RNP). Incorporating CBB lipidoids into liposomes improves delivery efficiency. CBB11 stands out for facilitating precise in vivo ABE-mediated gene editing. Delivery of liposome-CBB11-RNP complexes results in a 120-fold increase in base editing compared to RNP alone and restores the scotopic ERG b-wave response in the rd12 mouse model. These results demonstrate the potential of CBB-augmented, liposome-RNP systems for therapeutic gene editing in the eye, paving the way for single-dose precision medicines to treat IRDs.

The data is organized in compressed subdirectories whose name correspond to figures in the main directory (Figure_2b, Figure_3, Figure_S1, etc.).

Abbreviations:

• ABE - adenine base editor

• CBB - Coomassie Brilliant Blue

• CBBZ - Coomassie Brilliant Blue with alkyne moiety

• DAPI - 4',6-diamidino-2-phenylindole

• DMSO - dimethylsulfoxide

• EndoV - endonuclease V

• ERG - electroretinography

• GFP - green fluorescent protein

• HEK - human embryonic kidney cells

• IRDs - inherited retinal diseases

• kDa - kilodaltons

• m/z - mass to charge ratio

• MW - molecular weight

• OCT - optical coherence tomography

• PAGE - polyacrylamide gel electrophoresis

• PdI - polydispersity index

• RFP - red fluorescent protein

• RPE - retinal pigment epithelium

• RNP - ribonucleoprotein

• TAS - tris-acetate-sucrose buffer

• TEM - transmission electron microscopy

Source_data.xlsx: contains data used to draw graphs and the statistical analyses as reported by GraphPad. The contents are described below.

Figure_2b.zip. CBB-mediated delivery of Cre recombinase in vitro into color-switching Cre reporter cells (CS cells). Fluorescence microscopic images of HEK color-switch (CS) cells treated with Cre complexed with CBB lipidoids. C, Cre without CBB. (c, d) Quantification of GFP-to-RFP conversion in CS cells treated with the complexes of Cre recombinase with CBB compounds (c) by flow cytometry or with CBBZ compounds (d) by fluorescence microscopy. [Cre]/[CBB(Z)] = 1:8; Cre delivery mediated by Lipofectamine 3000 (LF) as a positive control is also shown. C, Cre recombinase only. NE, no enzyme.

Figure_3b.zip Cell-specific delivery of Cre recombinase mediated by CBB lipidoids in the mT/mG mouse retina. Representative two-photon microscopy images indicating the switch in expression from tdTomato to eGFP, one week after subretinal injection of Cre recombinase complexed with CBB lipidoids. The RPE layer is orientated toward the top. C, Cre recombinase only. The scale is provided in micrometers. Images are representative of at least 3 eyes.

Figure_4.zip. Delivery of ABE-RNPs to rd12-reporter cells mediated by CBB lipidoids. Fluorescence microscopy images of rd12 reporter cells treated with ABE8e NG RNP-CBB complexes. A = ABE RNP only; LF = ABE RNP with Lipofectamine 3000, CBB = ABE with specified CBB compound). (c,d) Quantitative assessment of ABE-RNP delivery into rd12-reporter cells, mediated by lipidoid complexes of ABE8e with CBB compounds (c) by flow cytometry; or CBBZ compounds (d) by fluorescence-microscopic imaging only. [ABE8e]/[CBB(Z)] = 1:8. A, ABE8e-RNP only. NE, no enzyme. (f - Figure_4f) Representative urea-PAGE-gel image of products of ABE deamination cleaved by EndoV. The gel was imaged with a fluorescein filter. N, no enzyme; P, positive control: 60 bp DNA with inosine in the middle; A, ABE-RNP in TAS; D, ABE-RNP in TAS with 10% DMSO. Numbers indicate molar excess of the CBB compound versus ABE. Left to right: N; P; A; D; CBB22: 4, 8, 16; CBB11: 4, 8, 16; CBB14: 4, 8, 16. (g) Fluorescence quantification of substrate and products of deamination and EndoV cleavage. Two independent samples per data point were assayed in parallel. Deaminated (%) is the ratio of intensity of the 30 nt band to total intensity of 30 and 60 nt bands.

Figure_5.zip. Liposome delivery of Cre- and ABE-RNPs to reporter cells. (b) The size distribution of the CBB11-alone and CBB11-doped liposomes (Lipo1), in TAS buffer. Average of 3 replicates. (c) Fluorescence microscopic quantification of the recombination activity triggered by Cre-recombinase delivery, mediated by Lipo1 or Lipo2. CBB14 and Lipofectamine 3000 served as positive controls. C, Cre recombinase only; NE, no enzyme. Data are presented as a mean of 2 replicates. (d) The zeta potentials of Lipo1 and Lipo2 in the absence or the presence of ABE-RNP. The ratio of ABE-RNP to CBB11 is 1:8. Means of two biological replicates. (e - Figure_5e) Representative images of cryo-TEM of Lipo1, ABE-RNP only and Lipo1 with ABE-RNP (f) Titration of in vitro delivery of ABE8e-RNPs mediated by CBB11, CBB14, and Lipo1 ~5 h after the RNP was complexed with the CBB compounds and liposomes. The ratio [CBB]/[RNP] = 8:1. The total lipid amounts of Lipo 2 and Lipofectamine 3000 were comparable to those for Lipo1. (g, h) Titration of in vitro delivery of ABE8e-RNPs mediated by Lipo3/Lipo4 (g); and Lipo5/Lipo6 (h), respectively, after allowing 15 min for complex formation in each case. The ionizable lipid concentration was always 80-fold greater than the concentration of ABE-RNP. ABE-RNP in TAS buffer served as negative control. (i - Figure_5i) Fluorescence-microscopic images of rd12-reporter cells treated with liposome-RNP complex. The RNP concentration was 100 nM. Representative of 2 biological replicates. Fluorescence microscopy images of rd12 reporter cells treated with ABE8e NG RNP-CBB liposome complexes. e = enhanced. CH1 = green channel, GFP. CH2 = red channel, TxRed. Overlay = overlay of CH1 and CH2.
CBB11 = 10x_XY08
CBB14 = 10x_XY14
Lipo1 = 10x_XY20
Lipo2 = 10x_XY26
LF = 10x_XY02
NE = 10x_XY38.

Figure 6. ABE-RNP delivery mediated by CBB-containing liposomes in rd12 mice. Four-week-old rd12 mice were treated via subretinal injection with 20 µM ABE8e-RNPs with indicated reagents, with 8-fold molar excess of CBB and 80-fold molar excess of SM-102, where applicable. (a) B-wave amplitudes of ERG responses of rd12 mice recorded two weeks after treatments. (b) Quantification of the visual chromophore 11-cis retinal in the eyes from ABE-treated rd12 mice obtained after a 24-h dark adaptation. n = 6 eyes. (c-f) Quantification of (c) total DNA editing, (d) total cDNA editing, (e) precise DNA editing and (f) precise cDNA editing in the retinal pigment epithelium isolated from the treated rd12 mice. (g) Off-target adenine genome-editing in rd12 mouse RPE after delivery of ABE-RNPs by Lipo1. N = 4 isolates picked randomly from samples analyzed in c-d.

Supplementary Figure 2. Identification of oxidation sites on oxCBB11 with mass spectrometry. (a,d) Two isomers of oxCBB11 that eluted from the silica gel column were injected into the LC-MS and analyzed by selected-ion monitoring at m/z 543.5. (b, e) The MS1 spectrum of the peak that eluted at 4.65 min (b) and the MS1 spectrum of the peak at 4.93 min (e) were similar. (c, f) The MS2 spectrum (c) of the ion at m/z 543.02 in panel b suggested hydroxylation events on one of the polyunsaturated fatty acid moieties. The MS2 spectrum (f) of the ion at m/z 543.02 in panel e is consistent with hydroxylation events on the tertiary amine of CBB11.

Supplementary Figure 4. Concentration dependence of the in vitro delivery of gene editors mediated by representative CBB compounds. (a) Quantification by fluorescence microscopic imaging of GFP-to-RFP conversion in CS cells. Cells were treated with Cre recombinase (0.5 μM) complexed with CBB compounds at increasing concentrations (1-8 μM). The cells emit green fluorescence at baseline and red fluorescence after editing. (b) Quantification by fluorescence microscopic imaging of the area ratio of eGFP to mCherry in rd12-reporter cells treated with ABE-RNPs (100 nM) complexed with CBB compounds at increasing concentrations (0.2-1.6 μM).

Supplementary Figure 5. Relative viability of reporter cells treated with Cre or ABE-RNPs, along with CBB compounds. Color-switching cells treated with Cre recombinase and CBB compounds 1-22 (a), or rd12-reporter cells treated with ABE-RNPs and CBB compounds 1-22 (b), were assessed for viability by DAPI exclusion, measured with flow cytometry. LF, lipofectamine 3000; C, Cre recombinase only; A, ABE-RNP only; NE, no enzyme.

Supplementary Figure 6: Cre is complexed by CBB compounds into large macromolecular complexes. The particle size distributions of Cre complexed with CBB11 and CBB14. CBB11-Cre formed homogenous particles (d = 1443 nm, PdI = 0.182), whereas CBB14-Cre formed heterogenous particles (average d = 1514 nm, PdI = 0.33) with two species whose d were 167 and 2129 nm. d = diameter; PdI = polydispersity index.

Supplementary Figure 7 - Figure_S7.pptx: In vivo two‐photon excitation microscopy demonstrating Cre recombinase delivery via CBB lipidoids or lipid-nanoparticle formulations in the intact mT/mG mouse eye. One week after subretinal injection of the Cre‐containing vector, tdTomato‐to‐eGFP expression switching was assessed in the mT/mG model. The left panels show the eGFP channel (green), and the right panels display the overlay of the tdTomato (red) and eGFP (green) channels. The RPE layer, oriented toward the top, is marked by open yellow triangles, and the photoreceptor layer is denoted by open blue triangles. Scale numbers correspond to micrometers. Images are representative of n = at least 3 eyes.

Supplementary Figure 8 - Figure_S8.zip: Size-exclusion chromatography of CBB liposomes. (a) Coomassie fluorescence scan of a 96-well plate containing samples collected from PD10 desalting columns. Ve = elution volume. (b,c) A photograph (b) and a Coomassie fluorescence scan (c) of the PD10 desalting columns after the size-exclusion chromatography.

Column Ve (ml)

1 2.00

2 2.25

3 2.50

4 2.75

5 3.00

6 3.25

7 3.50

8 3.75

9 4.00

10 4.25

11 4.50

12 4.75

Figure_S9.zip: Addition of CBB11 boosts the liposomal delivery of ABE8e-RNPs in rd12-reporter cells. Lipo3-6 correspond to formulations as in Fig 5. (a) The particle size distribution of Lipo3 (dark blue) was average diameter = 155 nm, with PDI = 0.19; and Lipo4 (orange) was average diameter = 155 nm, with PDI = 0.20. (b) The particle size distribution of Lipo5 (pale blue) was average diameter = 154 nm, with PDI = 0.17; and Lipo6 (yellow) was average diameter = 161 nm, with PDI = 0.15. (c) The zeta potential of Lipo3 (dark blue) was -2.64 mV, slightly higher than that of Lipo4 (orange), -4.11 mV. (d) The zeta potential of Lipo5 (pale blue) was -3.25 mV, slightly lower than that of Lipo6 (yellow), -2.25 mV. The measurements for a-d were done 3 replicates. (e) Representative fluorescence-microscopic images of rd12-reporter cells treated with liposome-RNP complex. The RNP concentration was 50 nM. Lipofectamine 3000 (LF) served as positive control, and ABE8e-RNPs delivered without vehicles (A) served as negative control. The images are representative of 4 biological replicates. CH1 = green channel, GFP. CH2 = red channel, TxRed.
XY16 = Lipo3
XY22 = Lipo4
XY28 = Lipo5
XY34 = Lipo6
XY40 = Lipofectamine 3000
XY45 = ABE RNP only.

Figure_S10.zip: Representative cryo-TEM images of Lipo1. The round vesicles in the images represent lipid bilayers of Lipo1 surrounding aqueous cores. Scale bar = 100 nm.

Figure_S11_RIMG3165.JPG: : Digestion of ABE8e-RNPs (without, or with Lipo1) with proteinase K. ABE8e-RNPs were incubated with proteinase K (protein:enzyme weight ratio = 100:1) at room temperature for 5, 10, 15, or 60 min. ABE8e-RNPs with Lipo1 in the absence of protease K served as a negative control.  Left to right: Protein molecular weight marker Bio-Rad Precision Plus Prestained; Lipo1 + Proteinase K - 5 min, 10 min, 60 min; Lipo1 - Proteinase K + 5 min, 15 min, 60 min; Lipo1 + Proteinase K + 5 min, 10 min, 60 min; Protein molecular weight marker. From top to bottom, protein molecular weight marker bands represent molecular weight (MW) in kDa: 250, 150, 100, 75, 50, 37, 25, 20, 15. Molecular weight of ABE: 187.1 kDa.

Figure_S12.zip. (a) Fluorescence microscopic quantification of delivery of ABE8e RNP into rd12 reporter cells mediated by liposomes with CBB11 (Lipo1, Lipo7) and SM102 (Lipo1, Lipo2). Lipofectamine 3000 (LF) was used as a positive control, buffer (TAS) as a negative control. (b) Representative fluorescence microscopic images of rd12 reporter cells treated with ABE8e RNP complexed with liposomes. The images are representative of 4 replicates.

Figure_S13.zip.. Treatment of TIGER-HEK cells with ABE-RNPs. (b, c) The conversion of non-fluorescent TIGER HEK cells to fluorescent after the treatment with ABE8e-NG-RNPs (b) or ABE8e-N108Q-NG-RNPs (c), complexed with CBB11, CBB14, Lipo1, or Lipo2. [CBB]/[RNP] ratio = 8:1. Results for Lipofectamine 3000 (as a positive control) were comparable to Lipo1. Cells treated with ABE-RNP without any vehicles (TAS) served as a negative control.

Figure_S14.zip: Evaluation of the safety and efficacy of Lipo1. (a) Particle size distributions of fresh Lipo1, diameter 148 nm and PDI = 0.23; and stored Lipo1, size = 140 nm, PDI = 0.16. (b) Fluorescence microscopic quantification of the ratio of eGFP to mCherry for rd12 reporter cells, treated with the fresh Lipo1-RNPs (100 nM). RNPs with Lipofectamine 3000 (LF) served as positive control, and RNPs alone (A) as negative control; [CBB]/[RNP] ratio = 8:1. (c) Representative optical coherence tomography (OCT) images of female WT mice treated with 20 μM ABE-RNPs alone (R20), 160 μM CBB11 as Lipo1 (L160), with 10 μM RNPs and 80 μM CBB11 as Lipo1 (L80R10), or with 20 μM RNPs with 160 μM CBB11 as Lipo1 (L160R20); control WT mice (WT N) were non-treated. The injection volume was 1 μl. N = 4 eyes per treatment. (d) ERG b-wave amplitudes of WT mice whose OCT scans are in panel c. (e) 11-cis-retinal levels in WT mice whose ERG responses are presented in d. (f) ERG b-wave amplitudes of rd12 mice treated with 0.5 μl of L80R10; 1 μl of L80R10; or 1 μl of L160R20. Untreated WT mice (WT N) and rd12 mice (rd12 N) served as positive and negative controls, respectively. (g) Quantification of 11-cis-retinal levels in rd12 mice whose ERG responses are presented in panel f.

Figure_S15.zip. Flow cytometry gating strategies. Gating strategies used in flow cytometry quantification of (a) CS cells and (b) rd12 reporter cells. SSC-A: side scatter area; FSC-A: forward scatter area; SSC-H: side scatter, height; FSC-H: forward scatter height; FITC-A: green fluorescence intensity; PE-A: red fluorescence intensity; Pacific Blue-A: blue fluorescence intensity; P1, P2, P4: gates.

Synthesis of CBB: Mass spectrometry traces of the CBB compounds. m/z = mass to charge ratio; Intensity = signal intensity.

Sharing/Access information.

Amplicon sequencing data are available in NCBI Sequencing Read Archive: PRJNA1240374 and PRJNA1358750.

Code/Software.

ImageJ was used for image-based estimation of conversion of fluorescent reporter cells.