Virus-like particle delivery of Cas9 targeting Vegfa reduces abnormal blood vessel growth in wet age-related macular degeneration

NIH/3T3 (CRL-1658) and HEK293T (CRL-3216) cells were obtained from ATCC. After thawing, the cells were verified to be mycoplasma negative before further culture. Cells were cultivated in Dulbecco's modified Eagle's medium (LM007-01; Welgene). The media contained 10% fetal bovine serum (16000044; Thermo Fisher Scientific) and 100 units/mL penicillin/streptomycin (15140–122; Thermo Fisher Scientific). The cells were maintained in a 5% CO₂ incubator at 37 °C.

To construct Vegfa sgRNA plasmid, an sgRNA-expressing vector (104174; Addgene) was digested with BsaI. sgRNA oligomers were annealed, phosphorylated with T4 PNK, and Ligated with the Linearized vector. The oligomer sequences are shown in Additional file 2: Table S1. VSV-G plasmid (8454), MMLVgag-pol plasmid (35614), and MMLVgag-3xNES-Cas9 plasmid (181752) were obtained from Addgene.

To produce Cas9-eVLPs, 1 × 10⁷ HEK293T cells were seeded on 150-mm cell culture dishes containing DMEM. After incubation for 16 h, DMEM was exchanged with fresh medium and a mixture of plasmids expressing VSV-G (400 ng), MMLVgag–pol (3375 ng), MMLVgag–3xNES–Cas9 (1125 ng), and an Vegfa sgRNA (4400 ng) were co-transfected per dish using polyethyleneimine. At 48 h after transduction, cell supernatant was harvested and centrifuged for 5 min at 500 g to remove cell debris. The clarified supernatant was filtered through a Millex-HV 0.45-μm low protein-binding membrane (Millipore). For Cas9-eVLPs that were used in cell culture, the filtered supernatant was concentrated 100-fold using PEG-it Virus Precipitation Solution (LV825A-1; System Biosciences) according to the manufacturer's protocols. For Cas9-eVLPs that were injected into mice, the filtered supernatant was concentrated 2000-fold via ultracentrifugation using a cushion of 20% (w/v) sucrose in PBS. Ultracentrifugation was performed at 26,000 rpm for 2 h (4 °C) using an SW28 rotor in an Optima XPN Ultracentrifuge (Beckman Coulter). Following ultracentrifugation, eVLP pellets were resuspended in cold PBS (pH 7.4) and centrifuged at 1000 g for 10 min to remove debris. eVLPs were frozen at a rate of − 1 °C/min and stored at − 80 °C.

The concentration of eVLPs (number of VLP particles per microliter) was measured by quantifying MuLV p30 protein with the MuLV Core Antigen ELISA kit (Cell Biolabs; VPK-156) following the manufacturer's protocol. Recombinant MuLV p30 standard included in the kit was used to generate the standard curve and the concentration of VLP-associated p30 protein was calculated with the assumption that 20% of the observed p30 in solution was associated with VLPs, as was previously reported [65].

Genomic DNA was extracted either from cultured cells using the Wizard Genomic DNA Purification Kit (A2920; Promega) or from mouse RPE using the DNeasy Blood & Tissue Kit (69504; QIAGEN). Target sequences were PCR-amplified using Phusion® High-Fidelity DNA Polymerase (M0530S; NEB). The PCR primers used for targeted deep sequencing are listed in Additional File 2: Table S1. To evaluate indel frequencies in HEK293T cells, an initial PCR was performed using 200 ng of genomic DNA and primers containing Illumina adaptor sequences. Then, 2 µL of a 1:10 dilution of the PCR amplicon was subjected to a second PCR using primers with barcode sequences. After gel purification with the MEGAquick-spin Total Fragment DNA Purification Kit (17290; iNtRON Biotechnology), the amplicons were analyzed using the MiniSeq platform (Illumina). For evaluation of indel frequencies in mouse RPE, PCR was performed using 50 ng of genomic DNA, following the same protocol described above. After gel purification, the resulting amplicons were sequenced using both the MiniSeq and MiSeq platforms (Illumina).

The top-four predicted off-target sites for Vegfa sgRNA were identified using the CCTop-CRISPR/Cas9 target online predictor [41]. Predicted off-target sites were amplified using 2 × Taq PCR smart mix (STD01-M50h; Solgent). An initial PCR was performed to amplify the target and potential off-target sites in a 30-µL reaction volume containing 30 ng of genomic DNA from the RPE. To attach Illumina index sequences, a second round of PCR was performed using 1 µL of the initial PCR product in a 30-µL reaction volume. The purified PCR products were pooled at equal molar ratios for sequencing on the MiniSeq platform (Illumina). Cas-analyzer was used to analyze indel frequencies at the off-target sites [66]. The sequences of the off-target sites and the primers used for deep sequencing are listed in Additional file 2: Table S2.

We conducted Digenome-seq as previously described [67, 68]. Genomic DNA was extracted using the DNeasy Tissue Kit (QIAGEN) following the manufacturer's protocol. In brief, recombinant Cas9 nuclease (100 nM) was pre-incubated with Vegfa sgRNA at room temperature for 10 min to generate Cas9-sgRNA complex. Next, the complex was mixed with 8 µg of mouse genomic DNA in a reaction buffer (100 mM NaCl, 50 mM Tris–HCl, 10 mM MgCl₂, 100 µg mL⁻¹ bovine serum albumin, at pH 7.9) and incubated for 8 h at 37 °C. Following digestion, the DNA was treated with RNase A and proteinase K, and then re-purified using the DNeasy Tissue Kit (QIAGEN). The digested DNA was subsequently fragmented using the Covaris system. The resulting DNA fragments were blunt-ended using End Repair Mix (illumina) and then ligated with sequencing adaptors for library preparation. Whole-genome sequencing (WGS) was performed on the Illumina HiSeq X Ten platform at a sequencing depth of 30 × . DNA cleavage scores were calculated with previously used source codes (https://github.com/chizksh/digenome-toolkit2↗) [67]. All sites captured by Digenome-seq are shown in Additional file 2: Table S3.

NIH/3T3 cells were seeded into 8-well culture slide (30508; SPL Life Sciences) at 3 × 10⁴ cells per well. One day after Cas9-eVLPs transduction, cells were fixed in 4% paraformaldehyde (PFA, P2031; Biosesang) for 15 min at room temperature and stained with the Cas9 mouse monoclonal antibody (1:200, 14697S; Cell Signaling Technology) overnight at 4 °C. On the following day, the cells were rinsed three times with PBS, then incubated in Alexa Fluor 488 Goat anti-Mouse IgG (H + L) Highly Cross-Adsorbed (1:250, A32723; Invitrogen) at room temperature for 2 h. After rinsed with PBS three times, cells were counterstained with Phalloidin Labeling Probes (1:400, A12381; Invitrogen) and DAPI (1:1000, D9542; Sigma-Aldrich) at room temperature, respectively. After rinsing three times, the culture slide was mounted with a sufficient amount of mounting solution. The imaging was performed using confocal microscopy (Leica Microsystems).

In vitro, NIH/3T3 cells were seeded into a 60-mm culture dish (20060; SPL Life Sciences) at 3 × 10⁵ cells per well. After Cas9-eVLPs transduction, cells were fractionated using Nuclear and Cytoplasmic Extraction Reagents (78833; ThermoFisher) according to the manufacturer's protocols. In vivo, the eye was harvested at 1, 3, and 7 days after Cas9-eVLPs treatment, and protein was extracted from the RPE tissue using a RIPA buffer (RC2002-050; Biosesang) containing protease inhibitor (p3100-001; GenDEPOT). The protein concentration of each sample was determined using the BCA Protein Quantitation Kit (23228; ThermoFisher). Protein samples were then separated via SDS-PAGE and transferred to polyvinylidene difluoride filter membranes. The membranes were blocked with 5% skim milk in TBST (Tris buffered saline containing 0.1% Tween) at room temperature for 1 h, and then incubated with primary antibodies overnight at 4 °C. The following primary antibodies were used: Lamin A/C (1:1000, 4777S; Cell Signaling Technology), α-Tubulin (1:1000, sc-23950; Santa Cruz Biotechnology), β-actin (1:10000, A1978; Sigma-Aldrich), and Cas9 (1:1000, 14697S; Cell Signaling Technology). On the following day, membranes were rinsed with TBST 3 times and incubated with horse anti-mouse IgG antibody (1:5000, 7076; Cell Signaling Technology) for 1 h at room temperature. The blots were treated with enhanced chemiluminescence reagent (34095; ThermoFisher) and detect bands using Chemi image system2 (GE Life Sciences). Gel band quantification using ImageJ software, following the steps, converting the gel image to 8-bit grayscale, inverting the image so that bands appear as peaks, and using the gel analysis tool to select regions of interest, according to the protein size (Cas9, 162 kDa; β-actin, 42 kDa). After plotting the lanes, the area under each peak (representing band intensity) is measured.

According to the manufacturer's protocols, the Quantikine murine VEGF ELISA kit (MMV00; R&D Systems) was used to detect the levels of VEGFA released from NIH/3T3 cells and RPE tissues after Cas9-eVLPs transduction. In brief, collected sample, control, and standard sample were added to each well and mixed by gently tapping the plate frame for 1 min. Thereafter, wells were covered with an adhesive strip and incubate for 2 h at room temperature. Each well was aspirated and washed with Wash Buffer using an autowasher. Thereafter, 100 μL of Mouse VEGF Conjugate was added to each well, which was then covered with a new adhesive strip to incubate for 2 h at room temperature. The aspiration and wash process was then repeated, 100 μL of Substrate Solution were added to each well and incubate for 30 min without light stimulation. Finally, we added 100 μL of Stop Solution and determined the optical density using microplate reader (ELR08IFL; AID Reader Systems).

The C57BL/6 (000664) mice were purchased from the Jackson Laboratory. Eight-week-old male mice were employed in all experiments. All animal experiments in this study were approved by the Seoul National University Animal Care and Use Committee (Permit Number: SNU-230731–4-4) and conducted in strict accordance with the guidelines of the Association for Research in Vision and Ophthalmology Statement. Mice were kept under cyclic light (12-on/12-off) with ad libitum access to food and water in approved cages.

Mice were anesthetized with an intraperitoneal injection of tiletamine (25 mg/mL)/zolazepam (25 mg/mL) mixture. After anesthesia, mouse eyes were placed in the proper position, and pupils were dilated with an eye drop containing phenylephrine hydrochloride (5 mg/mL) and tropicamide (5 mg/mL). After opening the eyelid and protruding the eye to expose the equator for convenient injection, a small hole was punctured at the slight posterior of the limbus using a sterile 30-gauge needle. The 33-gauge blunt needle of a microliter syringe was placed through the pre-punctured hole. The needle was then inserted into the subretinal space until the point when mild resistance was felt. The solution was injected slowly with low pressure, and the retinal bleb was checked under the indirect ophthalmoscopy. Mice received a titer of 4.3 × 10¹⁰ or 2.15 × 10¹¹ Cas9-eVLPs into the subretinal space. Moreover, age-matched mice that were not injected served as the untreated group, and mice injected with PBS served as the vehicle group, respectively.

Eight-week-old male mice were utilized to conduct LI-CNV following treatment with Cas9-eVLPs for 4 weeks. Upon anesthesia, their eyes were dilated using an eye drop solution containing phenylephrine hydrochloride (5 mg/mL) and tropicamide (5 mg/mL). Subsequently, a laser photocoagulator equipped with an indirect headset delivery system (Ilooda) was employed to visualize the retina. The laser parameters were as follows: wavelength: 810 nm; spot size: 200 μm; power: 1 W; exposure time: 100 ms. Adequate laser energy was applied to four locations per eye to induce rupture of Bruch's membrane. Only burns that resulted in a bubble formation without vitreous hemorrhage were considered for inclusion in the study. On the 7 days post-laser photocoagulation, the eyes were enucleated, fixed in 4% PFA, and prepared for whole-mounted RCS complex formation. Following the immunofluorescence staining procedure with Alexa Fluor 568-conjugated anti-IB4 antibody (1:250, I21413; Invitrogen), the CNV area of all the laser burn sites (four laser burn sites for each eye) was quantitatively analyzed using a built-in measuring tool, the LAS X systems (TCS SP8; Leica Microsystems). CNV volume quantitied using Imaris software (Oxford Instruments).

Prior to checking, phenylephrine hydrochloride/tropicamide mixture were applied topically, and general anesthesia was induced in mice. The Vantage Plus head-worn binocular indirect ophthalmoscope (1205-P-1020; Keeler) with non-contact 78D slit lamp lens (Volk Optical) to retinal general diagnosis while maintaining corneal moisture. After subretinal injection or laser photocoagulation, when complications such as subvitreal hemorrhage were observed by ophthalmoscopy, the mice were discharged from the experiment.

Pups from each group euthanized via CO₂ inhalation. The ocular globe was enucleated and fixed in 4% PFA for 30 min at room temperature. The cornea and lens were removed, then the retina was dissociated from the RCS complex. The RCS complex was incubated in blocking solution (BP150; Biosolution) at room temperature for 2 h and stained with Cas9 mouse monoclonal antibody (1:200, 14697S; Cell Signaling Technology) overnight at 4 °C. On the following day, we rinsed the stained RCS complex three times and incubated with Alexa Fluor 488 Goat anti-Mouse IgG (H + L) Highly Cross-Adsorbed (1:250, A32723; Invitrogen) at room temperature for 2 h. We then rinsed three times and incubated with an Alexa Fluor 594-conjugated anti-ZO-1 antibody (1:250, 339194; Invitrogen) at room temperature for 2 h. The samples were counterstained with DAPI (1:1000, D9542; Sigma-Aldrich) at room temperature for 15 min. Afterward, the stained RCS complex was placed on a glass slide with the retinal pigment epithelial layer against the glass slide. An adequate amount of mounting solution was added, and a cover slide was placed. For detect activated retinal microglia, the paraffin ribbons followed deparaffinization and heat-induced epitope retrieval procedure. Then, anti-Iba1 antibody (1:50, EPR16588; Abcam) was incubated overnight at 4 °C. After rinsed with PBS 3 times and incubated with Alexa Fluor 488 Goat anti-Rabbit IgG (H + L) Highly Cross-Adsorbed (1:400, A32731; Invitrogen) at room temperature for 2 h. The nuclei were counterstained with DAPI and carry out sealing. Immunostained tissues were observed using a confocal microscope.

The mice were euthanized, and the globes were fixed in Hartman's fixative solution (H0290; Sigma-Aldrich) and 4% PFA for 20 h at room temperature, respectively. After fixation, "windows" were made on the eyeball at the location of the anterior and posterior chambers, as described previously [69]. Briefly, the eye was held in place with a forceps, and a slight incision with a 26-gauge needle was made into the anterior chamber. This step was repeated in the posterior part of the eyeball, which was in line with the window in the anterior chamber. The globes were embedded in paraffin. After 4-μm-thick paraffin sections were prepared, the sections were deparaffinized and hydrated via sequential immersion in graded ethyl alcohol solutions and xylene substitute (6764506; Epredia). H&E staining was performed for histological examination. The retinal thickness from the inner limiting membrane to the retinal pigment epithelial layer was measured using NIS-Elements Imaging Software (Nikon Instruments).

After deparaffinization and hydration of retinal paraffin sections via sequential immersion in graded ethyl alcohol solutions and xylene substitute, the deparaffinized slides were immersed in heated citrate buffer for 10 min. The slides were removed from the hot citrate buffer and cooled in ddH₂O for 10 min. TUNEL staining was performed using an in situ cell death detection kit (11684809910; Sigma-Aldrich) according to the manufacturer's protocols, and the nuclei were counterstained with DAPI. Retinal TUNEL-positive cells were counted in randomly selected fields in each group under a confocal microscope.

OCT imaging was performed using a custom-built OCT machine developed by the Korea Research Institute of Standards and Science. Mice were anesthetized with an intraperitoneal injection of tiletamine (25 mg/mL)/zolazepam (25 mg/mL) mixture. After anesthesia, mouse eyes were placed in the proper position, and pupils were dilated with an eye drop containing phenylephrine hydrochloride (5 mg/mL) and tropicamide (5 mg/mL). The fully dilated pupil was directed toward the subjective lens. Eyes were kept moisturized with PBS solution during the entire procedure to ensure optimized images. The following OCT system specifications were used, axial resolution: 2.60 µm in tissue (n = 1.37), image depth: 1.45 mm in tissue, scan range: 3.0 mm (mouse eye), SLD: 850 nm ± 100 nm and optical power: 0.78 mW.

After anesthesia and mydriasis, a recording electrode was placed on the corneal surface, and the reference needle electrode was subcutaneously inserted in the head, with an electrode in the tail serving as the ground. The ffERG was performed using the electrophysiologic system 3000 (UTAS E-3000; LKC Technologies). Mice were adapted in the dark for over 16 h. In the dark-adapted condition, the scotopic responses were recorded using a single dim flash of 0 dB using a notch filter at 60 Hz and a digital bandpass filter ranging from 0.3 to 500 Hz. After recording scotopic responses, the mice were exposed to light for a minimum of 15 min. In the light-adapted condition, photopic responses were recorded in response to a single flash of 0 dB, utilizing the same notch filter and digital bandpass filter settings. The amplitude of the a-wave was measured from the baseline to the lowest negative going voltage, whereas peak b-wave amplitudes were measured from the trough of the a-wave to the highest peak of the positive b-wave. The ERG waveforms were visualized using GraphPad PRISM 9 (GraphPad Software).

A virtual optomotor system (OptoMotry apparatus; CerebralMechanics) was used to assess visual function. Briefly, the mice were placed on an elevated platform positioned in the middle of an arena created by four inward-facing display monitors. Spatial frequency thresholds were assessed using a video camera to monitor the elicitation of the optokinetic reflex through virtual stimuli projected with sine-wave gratings (100% contrast) on the computer monitors. Experimenters were blinded to the treatment and each animal's previously recorded thresholds.

Statistical significance was calculated using GraphPad Prism 9. All data were first assessed for normal distribution, an unpaired t test was used to identify significant differences. Considering the normal distribution and homogeneity of variance data, one-way ANOVA with Tukey's post hoc tests was used for multiple comparisons. The data sets are presented as mean ± standard deviations (SD). Differences were considered significant at the level of p < 0.05.