Selection of an Efficient AAV Vector for Robust CNS Transgene Expression

Figure 1. iTransduce Library for Selection of Novel AAV Capsids Capable of Efficient Transgene Expression in Target Tissue (A) Two-component system of the library construct. (1) Cre recombinase is driven by a minimal chicken β-actin (CBA) promoter. (2) p41 promoter-driven AAV9 capsid with random heptamer peptide is inserted between amino acids 588 and 589, cloned downstream of the Cre cassette. (B) Selection strategy. (Bi) The iTransduce library comprised of different peptide inserts expressed on the capsid (represented by different colors) is injected intravenously (i.v.) into an Ai9 transgenic mouse with a loxP-flanked STOP cassette upsteam of the tdTomato reporter gene, inserted into the Gt(ROSA)26Sor locus. AAV capsids able to enter the cell of interest but that do not functionally transduce the cell (no Cre expression) do not turn on tdTomato expression. Capsids that can mediate functional transduction (express Cre) will turn on tdTomato expression. (Bii) Cells are isolated from the organ of interest (e.g., brain), and transduced cells are sorted for tdTomato expression and optionally cell markers. (Biii) Capsid DNA is PCR amplified from the sorted cells, cloned back to the library…

Figure 2. Identification of AAV-S and AAV-F after Two Rounds of In Vivo Selection for Brain Transduction Donut charts indicate the frequency of particular peptide inserts determined by NGS. (A) Chart of round 2 vector sequenced after production but before intravenous injection. (B) Chart of peptide frequency appearing in iTransduce isolation after round 2 injection. “Others” indicates sequence variants appearing as less than 1% of the total pool (in A, variants isolated after the round 2 screen are also highlighted, appearing at less than 1%). The asterisk (*) indicates a stop codon.

Figure 3. AAV-F Efficiently Transduces the Brain of Mice after Systemic Injection (A) Single-stranded AAV-GFP expression cassette used to compare the transduction potential of capsids. ITR, inverted terminal repeat; CBA, hybrid CMV enhancer/chicken β-actin promoter; WPRE, woodchuck hepatitis virus posttranscriptional regulatory element; pA, poly(A) signals (both SV40 and bovine growth hormone derived). (B) Representative low-magnification images of whole-brain sagittal sections from C57BL/6 mice (males) transduced with 1 × 1011 vg (low dose) of AAV9, AAV9-PHP.B, AAV-S, or AAV-F. (C) Representative images of sagittal section of brains after injection of 8 × 1011 vg (high dose) of each vector in C57BL/6 males. (D) Example sections of spinal cords transduced by each of the four vectors administered intravenously at the higher dose (8 × 1011 vg/mouse). (E) Quantitation of native GFP expression from each vector by the percentage of sections covered by fluorescence at low (left panel) and high (right panel) doses. (F) Multiregional comparison of transduction in the brain at the higher dose. ***p < 0.001; ****p < 0.0001 after one-way ANOVA with Tukey’s multiple comparison test (n = 3 mice/group).

Figure 4. AAV Vector Comparison of Neuron and Astrocyte Transduction and Biodistribution High-magnification images of AAV9-, AAV-PHP.B-, AAV-S-, and AAV-F-transduced cells (GFP positive) after co-immunostaining with markers for (A) neurons (NeuN) and (B) astrocytes (glutamine synthetase [GS]). Merged cells also include nuclear staining by DAPI. (C) Stereological evaluation of the percentage of transduced cortical astrocytes and neurons after intravenous (i.v.) delivery of 1 × 1011 vg of each vector. p < 0.0001 by one-way ANOVA. One (*) and two (**) asterisks represent the significant differences between each vector group after Tukey’s multiple comparisons test (n = 3 mice/group). (D) Biodistribution of vectors in the brain and liver as measured by qPCR of vector genomes, normalized by GAPDH genomic DNA levels (input DNA). (E) Transduction of AAV9, AAV9-PHP.B, AAV-S, and AAV-F in peripheral organs following intravenous administration at 8 × 1011 vg in C57BL/6 males. Retinal images: RPE, retinal pigment epithelium; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer. ****p < 0.0001 (n = 3 mice/group, one-way…

Figure 5. AAV-F Mediates High Transduction in Male and Female C57BL/6 Mice and Also in BALB/c Mice (A) Representative images of GFP signal across sagittal brain sections in male and female mice (n = 3) transduced by AAV-F at 1 × 1011 vg/mouse. (B) Sagittal brain sections of male BALB/c mice injected with AAV-F (left) or AAV9-PHP.B (right) at 1 × 1011 vg/mouse. DAPI (blue) is provided as a counterstain alongside GFP (green) to visualize PHP.B-treated brain sections. (C) Quantitation of native GFP expression from each vector by the percentage of sections covered by fluorescence. **p < 0.01. Unpaired t test (n = 3 mice/group).

Figure 6. AAV-F Mediates Higher Transduction Efficiency than AAV9 in Human Cortical Neurons (A) GFP expression in fetal-derived primary human neurons, transduced by AAV-F. Neurons were co-labeled with an antibody to β-tubulin to quantify transduction. (B) Quantitation of transduction efficiency of human neurons by AAV9, AAV-S, and AAV-F. *p < 0.05.