Identification of a novel neutralization epitope in rhesus AAVs

Figure 1. Discovery of novel AAV capsids in SIV-infected rhesus macaques (A) Phylogenetic tree of novel rhesus AAVs discovered in this study (red) and contemporary AAV serotypes (black). Scale bar represents mean number of amino acid changes per site. (B) Capsid percent similarity between contemporary serotypes and novel AAV capsids. (C) Negative stain EM of RhAAV4282 capsid packaging a luciferase transgene. (D) In vitro transduction of 293T cells with RhAAV4282 expressing EGFP.

Figure 2. In vitro receptor and capsid testing (A) Cells were treated with increasing concentrations of heparinase III. At 1 h later, cells were transduced with RhAAV4282.eGFP. At 72 h later, GFP+ cells were detected by flow cytometry. (B and C) CMAS (B) and (C) KIAA0319L KO cells were normalized to WT cells for RhAAV4282.EGFP transduction after 72 h. (D) Capsid stability of RhAAV4282, AAV7, and AAV9 was compared by incubating each capsid with SYPRO Orange dye and incrementally increasing temperature. The derivative signal/temperature was used to determine the Tm of the capsid. Separate biological replicates are denoted as black dots; black lines represent group medians. SEM for each temperature is denoted by colored horizontal lines in (D).

Figure 3. Human seroprevalence Human serum samples from around the Boston area were measured for RhAAV4282, AAV7, and AAV2 NAb titers. (A) Responses were categorized for each virus. (B) All responders’ (categorized as having NAbs against RhAAV4282, AAV7, or AAV2) NAb titers plotted for each virus. Black bars represent medians. p values represent result of one-way ANOVA test. ∗p < 0.05.

Figure 4. RhAAV4282 tissue tropism Mice were injected with 5E11 GC of AAVs expressing luciferase. Four weeks later, tissues were removed and used to measure (A) viral genome localization by qPCR, (B) viral transcript localization by RT-PCR, and (C) protein levels by measuring luminescence. (D) Representative image of tissues extracted from mice and immersed in PBS with luciferin. (E) Representative image of whole-mouse luminescence after luciferin injection. Individual mice in (A)–(C) denoted as dots, with group median values denoted by black lines. p values represent result of one-way ANOVA test for each tissue for (A) DNA, (B) RNA, and (C) luciferase. ∗∗∗∗p < 0.0001. Significance comparisons with AAV9 not shown.

Figure 5. Cryo-EM model of RhAAV4282 capsid (A) Atomic model from a cryo-EM reconstruction of the empty RhAAV4282 capsid to 2.57 Å resolution. (B) Structural comparison of RhAAV4282 VP3 monomer (red) aligned with an AAV7 VP3 monomer (blue) (7L5Q). Enlargement shows differences in the hypervariable region IV. RhAAV4282 electron density (gray mesh) and residues for one monomer of RhAAV4282 capsid (green) shown for amino acids (C) 280–288 and (D) 449–457 using RhAAV4282 VP1 amino acid numbering.

Figure 6. Sequence and structural comparison to AAV7 (A–C) Structural alignment of RhAAV4282 capsid amino acids (A) 292–295, (B) 715–720, and (C) 451–461 (using AAV7 amino acid numbering). (D) Sequence alignment of RhAAV4282 and AAV7 VP1 amino acid sequence. Hypervariable regions are numbered (I–IX) and denoted by bold letters. Sequences displayed in (A)–(C) are shown by red lettering. HRNE is designated by black brackets. “∗” signifies an exact match of amino acids; “:” signifies an amino acid change of amino acids with strongly similar properties; “.” denotes an amino acid change of amino acids with weakly similar properties; “” denotes a swap of amino acids that share no similar properties or a deletion.

Figure 7. Heterologous mouse boost Mice were injected with 5E11 GC of AAVs expressing EGFP. Four weeks later, mice were injected with 5E11 GC of AAVs expressing luciferase in either a homologous or heterologous dose. (A) Four weeks post-boost, whole-mouse imaging was conducted to determine the success of AAVs expressing luciferase. (B) Whole-mouse signal was quantified with regions of interest covering each mouse body, excluding the head. Individual mice denoted as separate dots. Black bars represent group medians. p values represent result of one-way Mann-Whitney test. ∗p < 0.05.

Figure 8. NAb titers HRNE region can significantly change NAb titers between AAV7 and RhAAV4282. Mice were injected with 5E11 GC of AAVs. Four weeks post-injection, serum was collected from mice and tested for NAb titers. Mouse groups are shown on the x axis (A–C). Serum was tested against (A) AAV7, (B) RhAAV4282, and (C) AAV9 in an AAV neutralization assay. Individual mice in (A)–(C) denoted as dots, with group median values denoted by black lines. p values represent results of one-way ANOVA tests for NAbs against each virus. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. Limit of quantification was defined as 2 SDs from the mean for sham in each plot and is denoted by black dotted line.