AAV vectors engineered to target insulin receptor greatly enhance intramuscular gene delivery

Figures

• 

  Figure 1. Insulin mimetic peptide S519 and its insertion into the AAV9 capsid (A) A segment of the S519 peptide was aligned to the human insulin B chain using the MUSCLE algorithm and visualized in JalView;38,39 amino acid similarity is plotted below the alignment. (B) A diagram of the AAV genome shows the assembly activating protein (AAP) and three VP proteins translated from a single Cap gene; VP1, VP2, and VP3. All display the same C-terminal region and thus all contain variable regions IV and VIII (VR-IV and VR-VIII). (C) The AAV9 cryo-EM structure (PDB: 3UX1) was visualized with ChimeraX software.40 Virion surface is colored by distance from the center of the virion. In the expanded view of the threefold axis of symmetry, VR-IV and VR-VIII are colored in green and blue, respectively.

• 

  Figure 2. Insertion of the insulin-mimetic peptide S519 into the capsid markedly enhances AAV9 transduction efficiency in vitro (A) Mosaic vectors were produced with the indicated ratios of WT to S519(IV) or S519(VIII) Rep/Cap plasmid. Vectors were quantified by qPCR. Each dot represents the vector titer from one experiment, and trend lines represent the mean of three independent experiments. (B) HEK293T cells transduced with CRISPR-Cas9 and untargeted guide RNA (Control 293T) or guide RNA directed against IR (IR-KO 293T) were analyzed for IR expression by flow cytometry. Representative data from three independent experiments are shown. (C) GFP-encoding mosaic vectors produced with the indicated ratios were used to transduce cells at an MOI of 105 vg/cell. GFP expression was measured by flow cytometry after 24 h. Trend lines are the mean of three independent experiments. (D) Cells were transduced with the indicated vector at an MOI of 105 vg/cell or a comparable quantity of LASV pseudovirus. S519(IV)-AAV9 and S519(VIII)-AAV9 were produced as mosaic vectors at a 10:1 ratio of WT to mutant capsid. GFP expression was analyzed 24 h post transduction (hpt). Data are shown as mean ± SEM of three…

• 

  Figure 3. eAAV9 is markedly more efficient in transducing primary human skeletal muscle cells compared to WT AAV9 Primary human skeletal muscle cells were differentiated to form myotubes and transduced with the indicated vectors expressing GFP that were preincubated or not with 30 nM IR-Fc. After 3 days of culture in growth media, cells were fixed and stained with Hoechst 33342 to label nuclei. Green, GFP reporter; red, cellular autofluorescence for morphology visualization; blue (included in Merge), Hoechst 33342. Scale bar (bottom panels) represents 100 μm. Representative images are shown from one of 3 independent experiments with cells sourced from 2 different donors.

• 

  Figure 4. eAAV9 transduces mouse skeletal muscle in vivo with much greater efficiency than WT AAV9 (A and B) Mice with ad libitum access to food (A, Fed) or fasted for 4 h before and 4 h after transduction (B, Fasted) were injected with 109 vg of luciferase-expressing WT AAV9 or eAAV9 vector in the gastrocnemius muscle of the right hindleg and imaged at the indicated time points. Top, representative images of mice at 14 dpt. Bottom, quantification of luminescence; dots represent individual mice and horizontal bars indicate mean values. Data are derived from experiments performed with two independent vector preparations (A, n = 6–9 mice per condition; B, n = 14–15 mice per condition). Statistical significance was analyzed by Mann-Whitney U-test (∗∗p < 0.01; ∗∗∗p < 0.001). Transduction efficiency of eAAV9 relative to WT AAV9: (A) 6.3-, 5.9-, and 6.5-fold enhancement for 14, 21, and 28 dpi, respectively; (B) 17.6-, 17.9-, and 18.4-fold enhancement for 14, 21, and 28 dpi, respectively. (C) Mice were fasted for 4 h, and blood glucose was measured before (−0.25 h) and at the indicated times after the injection of AAV9, eAAV9, or 0.5 U/kg human insulin. Representative data from three independent…

• 

  Figure 5. Insertion of the S519 peptide into the capsid enhances the transduction efficiency of a wide range of AAV serotypes (A) Control 293T or IR-KO 293T cells were transduced with WT or S519-modified vector of the indicated serotypes, encoding luciferase. Luciferase activity was measured 24 hpt. Data are shown as mean ± SEM of three independent experiments, and statistical significance was analyzed by Student’s t test (∗p < 0.05; ∗∗p < 0.01). (B) Fasted mice were injected in the gastrocnemius muscle with 109 vg of WT or S519-modified vector of the indicated serotypes, expressing luciferase, and imaged at 14 and 21 dpt. Top, representative images of mice at 14 dpt. Bottom, quantification of luminescence; dots represent individual mice and horizontal bars indicate mean values. Data are derived from experiments performed with two independent vector preparations (AAV1, n = 9–10 mice per condition; all others, n = 6–8 mice per condition). Statistical significance was analyzed by Mann-Whitney U-test (∗∗p < 0.01; ∗∗∗p < 0.001). Transduction efficiency of eAAV relative to parental AAV for 14 and 21 dpt, respectively: 2.4- and 2.4-fold (AAV1), 6.3- and 7.6-fold (AAV2), 14.5- and 14.2-fold…