AAV production

HEK293T cells were transfected with 3 different plasmids using polyethylenimine (PEI; Polysciences, Warringtion, Pennsylvania), and AAVs were harvested after 3 days. Purification and titration were performed according to standard procedures [15]. For the production of barcoded AAV genomes, the pscAAV-CMV-eYFP-BC-bGHpolyA was used as genome plasmid and was kindly provided by the Grimm lab (Heidelberg, Germany [16]). The desired capsid motif is determined by the p5E18vd2/9-SfiI1759 helper carrying rep/cap genes and lacking inverted terminal repeats (ITRs) [11]. In this plasmid, the cap9 gene contains Sfi restriction sites for cloning the heptamer insertion at the A589 position. Production of barcoded WT AAVs of the serotypes 4, 5, 6, and 9 as well as benchmark controls was performed with the respective rep/cap helper plasmid [17]. The pDGΔVP is lacking the cap gene and was used for its adenovirus helper functions in all productions [18]. Ten 15 cm dishes with 5.4 × 106 cells/dish were used for separate production of each capsid variant. To generate a random AAV library, the library plasmid backbone pKV-AAV9Lib/BB was used as genome plasmid [11]. One library was produced in 40 × 15 cm dishes of HEK293T cells.

Cloning of random plasmid library

The pKV-AAV9Lib/BB carries the cap9 gene within ITRs and is packed accordingly in the genome for expression and capsid building. Insertion of random (NNK)7 sequences was introduced in the SfiI restriction site. Ligation of the double stranded (NNK)7 oligonucleotide and SfiI digested backbone was performed at 1:30 molar ratio and transformed into electro-competent 10-beta E. coli (NEB, Frankfurt am Main, Germany). After overnight incubation at 37 °C and 225 rpm in MaxQ 8000 incubator shaker (Thermo Fisher Scientific, Waltham, Massachusetts), the plasmid library was purified with the Plasmid Plus Maxi Kit (QIAGEN, Hilden, Germany) and used for AAV library production.

Subcloning of enriched variants

Isolated gDNA and cDNA of mVSMCs after each selection round contained internalized vector genomes that were used for production of the selected AAV library for the next infection cycle. For this, the insertion region of the cap9 was PCR-amplified and inserted into the p5E18vd2/9-SfiI1759 plasmid backbone generating a plasmid library which was used for production of a the selected AAV library.

Cell culture

Murine vascular smooth muscle cells (mVSMCs) were isolated from adult murine aorta of both sexes (C57BL/6). Permission for organ removal was granted by the animal ethics board of the Ministry of Energy, Agriculture, the Environment, Nature and Digitalization Schleswig-Holstein (MELUND, V312-7224.121-4). Randomization and blinding were not performed nor required. Isolated aortae were flushed with PBS and cut into 2 mm2 pieces. Next, mVSMCs were isolated via 6 h incubation with 0.1% collagenase type II (Worthington Biochemical Corporation, Lakewood, New Jersey) at 37 °C. mVSMCs were routinely stained with smooth muscle actin (SMA) marker in order to prove the purity of the primary cells. Murine cardiac fibroblasts were isolated from murine adult heart. Isolated heart was cut into 2 mm2 pieces and incubated with collagenase type II and DNase I (Sigma-Aldrich, St. Louis, Missouri) for 45 min at 37 °C under continuous shaking. NRVCMs were isolated from neonatal rat hearts as previously published [19]. All cell types were maintained in Dulbecco’s modified Eagle’s medium (Sigma-Aldrich, Hamburg, Germany) supplemented with 10% heat-inactivated fetal bovine serum (FBS; PAA, Coelbe, Germany), 2 mM L-glutamine (Gibco, Karlsruhe, Germany), and 1 x Pen/Step (Life Technologies, Carlsbad, California).

HAoSMC and HUVECs were provided cryo-preserved by Promocell and maintained in the recommended medium supplied by the company (Promocell, Heidelberg, Germany).

For passaging and seeding of all cell types 0.05% Trypsin-EDTA (Gibco, Karlsruhe, Germany) was used. Viral transduction was performed overnight in serum-free medium. Cells were further cultured for 3 days in 10% FBS-supplemented medium.

Animal study

Animal experiments were carried out under the guidelines from directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes with the approval of the regional council (MELUND, permission no. V242-56596/2022). Animals were housed under standard conditions with a 12h-light- / 12h-night-cycle. Water and food were offered ad libitum. Barcoded AAV vector libraries were injected systemically via tail vein injection into eight-weeks-old male C57BL/6 N mice (Charles River). Each mouse was injected with 5 × 1011 vg (viral genomes) in a total volume of 100 µl. Mice were sacrificed two weeks after injection by cervical dislocation and harvested organs were snap-frozen in liquid nitrogen.

Transduction of human aortic tissue

Permission for using aortic tissues from patients undergoing surgery at the University Hospital Schleswig-Holstein, Kiel, Germany was granted from the Ethical Board of the Medical Faculty of the University of Kiel (permission no. P2N 2024-011). Informed consent was obtained from all patients. Tissue was collected, cut into 3–5 mm2 pieces, and maintained in Ham’s F12 Medium (Corning, Glendale, Arizona) supplemented with 10% FBS and 1× Pen/Strep as previously described [20]. AAV transduction with 1010 vg for RNA quantification was performed overnight in serum-free medium followed by 3 days of incubation in FBS-supplemented medium. For protein determination tissues were incubated with 5 × 1010 vg for 4 days. Injection of 1010 vg into the tissue were performed with insulin injections in serum-free medium and medium was changed after overnight expression to FBS-supplemented medium for further 3 days.

Total DNA and RNA isolation and qRT-PCR

DNA and RNA were simultaneously extracted from cells or tissues with AllPrep DNA/RNA Mini Kit (QIAGEN, Hilden, Germany) according to manufacturer’s instructions. Before isolations, approx. 5 mg human aortic tissue per sample was snap-frozen and pulverized in liquid nitrogen using a pestle and mortar. Subsequently 1 µg RNA was transcribed to cDNA with LunaScript® Reverse Transcriptase SuperMix Kit (NEB, Frankfurt am Main, Germany) according to manufacturer’s instructions. For eYFP quantification in human aortic tissues, 12.5 µg cDNA was used for qRT-PCR with SensiFAST SYBR No-ROX Kit (Bioline, Luckenwalde, Germany) in a CFX96 real-time PCR detection system (Bio-Rad, Hercules, California). The following primers were used for eYFP quantification: forward primer: 5ʹ-GCATCAAGGTGAACTTCAAGATCC-3ʹ and reverse primer: 5ʹ-ATGTGATCGCGCTTCTCGTTG-3ʹ. EYPF values were normalized to human ribosomal protein, large, P0 (hRPLR0). The eYFP expression from a vector was indicated as x-fold over untransduced cells or tissues.

NGS of amplified insertion region and barcodes

Region of interest of cDNA and gDNA samples was amplified by PCR reactions comprising Q5 High-Fidelity DNA Polymerase (NEB, Frankfurt am Main, Germany) to generate a 215 bp product of barcode regions and a 245 bp product for the insertion region in the cap9 gene. The PCR reaction mix was purified with NucleoSpin Mini Kit (Macherey-Nagel, Düren, Germany) and amplicons were subsequently sent for deep sequencing via amplicon sequencing service provided by Genewiz (Leipzig, Germany).

NGS data processing

For data processing, the python 2.7 script published by the Grimm lab [16] was applied to fastq files to count the 15 nt barcode sequences. The output files were used to normalize read counts and calculate the proportion of all variants in one tissue and the proportion considering the variation in the input library displaying the efficiency of individual variants on cells. For insert region amplicons, the configuration file of the script was adapted according to the flanking regions. The output provided the read counts of non-determent nucleotide sequences that were normalized on flow cell variation and ranked based on their frequency.

Flow cytometry

For analysis of transduction efficiency, cells were harvested and washed with PBS. Cells were monitored in fluorescence-activated cell sorting in a Flow Cytometry Analyser (Institute for Immunology, UKSH Kiel) and the percentage of eYFP positive cells (transduced cells) out of 10,000 events in technical replicates was evaluated with BD FACSDiva software (BD Biosciences).

Western blot analysis

Western blot analysis was performed according to standard protocols using 50 µg protein and antibody against GFP (1:1,000, 2956S, Cell Signaling, Danvers, Massachusetts) and corresponding secondary antibody anti-rabbit IgG-HRP (1:10,000, Santa Cruz Biotechnology, Dallas, Texas). Tubulin was used as housekeeping protein and detected with primary antibody (1:8000, MAB-9344-100, R&D Systems, Minneapolis, Minnesota) and secondary mouse IgG-HPR (1:10,000, Santa Cruz Biotechnology, Dallas, Texas). Chemiluminescence was detected with the Chemidoc™ MP Imaging System (Bio-Rad, Hercules, California). Densitometric analysis was performed using ImageJ software (version 1.54).

Immunohistochemistry

Human aortic tissue pieces were fixed in 4% paraformaldehyde for 4 h at 4 °C prior to embedding and freezing in TissueTek®. 7 µm sections were blocked with 2.5% BSA solution containing 0.01% Triton-X-100 for 1.5 h. 647-conjugated anti GFP antibody (Invitrogen, Eugene, Oregon, catalog no. A31852) was diluted 1:500 in blocking buffer and used to incubate sections overnight at 4°C in a humidified atmosphere. Images were taken by a fluorescence microscope (Keyence BZ-X810, Neu-Isenburg, Germany).

Statistics

Sample sizes were chosen based on previous experience [16]. Data are presented as means ± SD and analysed for statistical significance with ordinary one-way ANOVA with Tukey’s multiple comparisons for three or more groups. Calculation was performed in GraphPad Prism version 9.5.0 and p values are defined as <0.05 = *, <0.01 = **, <0.001 = ***, and <0.0001 = ****.