Multimodal imaging of capsid and cargo reveals differential brain targeting and liver detargeting of systemically-administered AAVs

Abstract

The development of gene delivery vehicles with high organ specificity when administered systemically is a critical goal for gene therapy. We combine optical and positron emission tomography (PET) imaging of 1) reporter genes and 2) capsid tags to assess the temporal and spatial distribution and transduction of adeno-associated viruses (AAVs). AAV9 and two engineered AAV vectors (PHP.eB and CAP-B10) that are noteworthy for maximizing blood-brain barrier transport were compared. CAP-B10 shares a modification in the 588 loop with PHP.eB, but also has a modification in the 455 loop, added with the goal of reducing off-target transduction. PET and optical imaging revealed that the additional modifications retained brain receptor affinity. In the liver, the accumulation of AAV9 and the engineered AAV capsids was similar (∼15% of the injected dose per cc and not significantly different between capsids at 21 h). However, the engineered capsids were primarily internalized by Kupffer cells rather than hepatocytes, and liver transduction was greatly reduced. PET reporter gene imaging after engineered AAV systemic injection provided a non-invasive method to monitor AAV-mediated protein expression over time. Through comparison with capsid tagging, differences between brain localization and transduction were revealed. In summary, AAV capsids bearing imaging tags and reporter gene payloads create a unique and powerful platform to assay the pharmacokinetics, cellular specificity and protein expression kinetics of AAV vectors in vivo, a key enabler for the field of gene therapy.

Fig. 1.

⁶⁴Cu-AAV pharmacokinetics from PET image analysis of the radiolabeled capsids in the brain and blood (AAV9, PHP.eB, and CAP-B10). A. Capsid protein sequence of AAVs with enhanced brain targeting and reduced liver transduction. B. Experimental scheme of PET imaging and biodistribution of AAV9, PHP.eB, and CAP-B10 in C57BL/6 mice. C. Representative maximum intensity projected (MIP) PET/CT images of mice systemically administered ⁶⁴Cu-AAV9 (n = 3), -PHP.eB (n = 4) and -CAP-B10 (n = 4). Mice were scanned for 30 min post-injection (p.i.) and rescanned at 4 and 21 h. D. Representative sliced whole body (left) and brain (right) PET/CT images of ⁶⁴Cu-PHP.eB and -CAP-B10 accumulation. E. Time-activity curve (TAC) (%ID/cc) of AAVs (AAV9: black, PHP.eB: magenta, CAP-B10: green) from blood (left) and blood subtracted radioactivity in the brain (right). F. Blood subtracted radioactivity in the brain (%ID/cc) at 21 h p.i. G. The early binding kinetics of AAVs in the brain over 30 min p.i. H–I. Blood subtracted radioactivity in the brain (%ID/cc) of individual mice from 0.5 to 4 h (H) and 4–21 h (I). Significant changes in radioactivity value represent the gradual accumulation of AAVs. J. Logan plots of brain uptake rate after AAVs administration. Abbreviations: %ID/cc: Percent injected dose per cubic centimeter, B: brain, L: liver, H: heart, V: vena cava, BL: bladder, S: spleen, IT: intestinal track, CB: cerebellum, MB: midbrain, CTX: cerebellar cortex, OB, olfactory bulb. Data and error bars are presented as mean ± SEM. One-way ANOVA formed in Fig. 1G with Tukey’s multiple hypothesis correction. Unpaired two-tailed t-test with Welch’s correction was performed in Fig. 1F, H and I n.s.: not significant, *P < 0.05, **P < 0.01, ****P < 0.0001.

Fig. 2.

PET reporter gene imaging and quantification, comparing AAV9 and PHP.eB transduction of the PKM2 gene in the brain. A. Schematic illustration of the longitudinal PET reporter gene (PKM2) expression study. AAVs, PHP.eB (n = 4) and AAV9 (n = 4) with EF1A-PKM2, and saline (n = 4) are systemically injected in C57BL/6 mice, and PKM2 transduction was assessed by [¹⁸F]DASA-23 at 1, 3, 8, 16, and 34 weeks p.i. B. Representative sliced PET/CT images of brains from mice (at 3 weeks p.i. of AAVs) systemically injected with [¹⁸F]DASA-23. C. Comparison of PET/CT and fluorescent microscope images of the mouse brain. 1) PET/CT intensity image represents accumulated ⁶⁴Cu-PHP.eB (6 × 10¹¹ vg) capsid location at 21 h p.i. 2) PET/CT intensity image represents retained [¹⁸F]DASA-23 indicating where the PKM2 protein is overexpressed at 3 weeks after tail-vein administration of PHP.eB:EF1A-PKM2 (2 × 10¹¹ vg). 3) Fluorescence microscopy intensity (FMI) of mouse brain overexpressing mNG at 3 weeks after tail-vein administration of PHP.eB:CAG-mNG (7 × 10¹¹ vg). Green fluorescence represents mNG. Blue represents nuclear (DAPI) staining. CB, cerebellum; CTX, cerebral cortex; MB, midbrain; P, Pons; TH, Thalamus. D. Time-activity curve for the initial 30-min dynamic brain acquisition following [¹⁸F]DASA-23 administration. Dynamic data are from mice systemically injected with PHP.eB, AAV9, or saline at 3 weeks p.i. (n = 3–4 per group). [¹⁸F]DASA-23 retention increased in PHP.eB-treated mice compared to both AAV9 and saline treated mice. E. [¹⁸F]DASA-23 uptake (25–30 min) in the brain from PET/CT images acquired at 1 week (left) and 3 weeks (middle) p.i. of AAVs, and whole body (WB) retention (right) at 3 weeks p.i. F. Normalized fold uptake of [¹⁸F]DASA-23 (%ID/cc) over saline control mice (%ID/cc) at 1, 3, 8, 16 and 34 weeks p.i. of PHP.eB (left) and AAV9 (right). G. Gene expression of PKM2 from the mouse brain at 1 and 3 weeks p.i. of AAV9, PHP.eB, and saline (each week, n_AAV9 = 4, n_PHP.eB = 4, and n_saline = 4). H. Protein bands of PKM2 and β-actin control from the brain of mice at 1 and 3 weeks p.i. of AAV9, PHP.eB, and saline (each week, n_AAV9 = 2, n_PHP.eB = 2, n_saline = 2). I. Gene expression of TSPO (translocator protein) from the mouse brain at 1 and 3 weeks p.i. of AAV9, PHP.eB, and saline (each week, n_AAV9 = 4, n_PHP.eB = 4, and n_saline = 4). Data and error bars are presented as mean ± SEM. Ordinary one-way ANOVA with Tukey’s multiple comparison test (E and G) and multiple unpaired t-test corrected for multiple comparisons using the Holm-Sidak method (F) were performed for the statistical analysis. *P < 0.05, **P < 0.01, ****P < 0.0001.

Fig. 3.

The pharmacokinetics of ⁶⁴Cu-AAVs in the liver and entire body. A. Blood subtracted accumulation of ⁶⁴Cu-AAV9 (n = 3), -PHP.eB (n = 4), and –CAP-B10 (n = 4) in the mouse liver (C57BL/6). B-D. Bar graphs displaying statistical significance of liver radioactivity from Fig. 3A at 0.5 h (B) 4 h (C) and 21 h (D) time points. E. The early binding kinetics of AAVs in the liver over 30 min. F. Logan plots of liver uptake rate after AAVs administration. CB(t) and CT(t) are the radioactivity concentration in the blood and target at a given time, and IntCB(t) and IntCT(t) are the accumulated radioactivity in the blood and target, respectively, from the time of injection to 30 min. G. The clearance of ⁶⁴Cu-AAV9 (n = 3), -PHP.eB (n = 4), and –CAP-B10 (n = 4) from the body. RA: radioactivity. Data and error bars are presented as mean ± SEM. Ordinary one-way ANOVA (B, C, D) and two-way ANOVA (E) with Tukey’s multiple comparison test were performed. *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 4.

On optical imaging at 4 h after injection, liver distribution of Cy5-labeled AAV9 variants displays a distinct cellular and zonal pattern. A. Labeling scheme for Cy5-AAVs (AAV9, PHP.eB and CAP-B10). B. Schematic of study for identifying viral particle distribution and transduction in liver. C. Representative fluorescence microscope images (FMI) of mice (C57BL/6) liver sections stained with Clec4F (green) and DAPI (blue) and the quantitation of the accumulated Cy5-AAVs in whole liver tissue (right) at 4 h p.i. Red and green channel represent the presence of Cy5-labeled viral particles and Kupffer cells, respectively. Bar graph represents a normalized mean fluorescent inteisity (MFI) of Cy5-AAVs in whole liver tissues. Scale bar: 100 μm. D. Maximum intensity projection (MIP) of liver tissue stained with Clec4F (green) and DAPI (blue). White arrow indicates Cy5-AAVs (red) in Kupffer cells. Scale bar: 20 μm. E. Manders’ overlap coefficient of AAVs. Colocalization of Cy5-AAV (red channel) and Kupffer cells stained with Clec4F (binarized green channel). F. A Z-section of Cy5-PHP.eB in Fig. 4D (left) and magnified square region of interest from left image (right). Cy5-PHP.eB (red) is found on the surface and cytoplasm of Kupffer cell (green). Scale bar: 20 and 5 μm. G. FMIs of mouse (C57BL/6) liver sections collected at 4 h p.i. of Cy5-AAV9, -PHP.eB, and -CAP-B10. Sliced tissues were stained with DAPI (blue). Maximum threshold was adjusted to 10% of raw image to enhance the Cy5 fluorescent channel (red). Scale bar: 2 mm (whole liver image) and 100 μm (magnified image). H. Relative fluorescent intensity of cells in zone 1 portal vein (PV) over zone 3 central vein (CV) in liver tissues. I. Representative FMI of mouse liver sections stained with DAPI (blue) and Clec4F (green) with Cy5 fluorescence (red) around the CV and PV. Arrow indicates blood flow from PV to CV. Scale bar: 100 μm. Data and error bars are presented as mean ± SEM. Ordinary one-way ANOVA with Tukey’s multiple comparison test was performed for the statistical analysis (C, E, and H). **P < 0.01, ****P < 0.0001.

Fig. 5.

Liver transduction of Cy5-labeled AAV9 variants displays the same zonal patterns as AAV localization. A. Fluorescence microscope images (FMIs) of liver and brain sections from mice collected at 3 weeks systemic p.i. of 7 × 10¹¹ vg Cy5-AAV9, -PHP.eB, and -CAP-B10 packaging CAG-mNeonGreen (mNG). Sliced tissues were stained with DAPI (blue) for morphology. Scale Bar: 2 mm. B. Median fluorecence intensity (MFI) of mNG from whole liver section (n = 4 from 2 mice per group). C. Magnified FMI of mouse liver obtained at 3 weeks p.i. of AAV9:CAG-mNG. In left image, hepatocytes around central veins (CV) express more mNG compared to those around the portal vein (PV). Yellow area is magnified in right image. Sliced tissues were stained with DAPI (blue), allowing the differentiation of portal vein and central vein. Arrow indicates blood flow from PV to CV. D. Comparison of MFI (mNG) of zone 1 (PV) and zone 3 (CV) in livers of mice administered Cy5-AAV9:CAG-mNG. mNG expression in cells around CV was significantly higher than near PV (p < 0.0001). ROI data (n = 20) was pooled from livers of two mice administered Cy5-AAV9. Data and error bars are presented as mean ± SEM. Ordinary one-way ANOVA with Tukey’s multiple comparison test (B) and unpaired two-tailed t-test with Welch’s correction (D) were performed for the statistical analysis. ****P < 0.0001.