An acidic oligopeptide displayed on AAV2 improves axial muscle tropism after systemic delivery

Abstract

Background: The appropriate tropism of adeno-associated virus (AAV) vectors that are systemically injected is crucial for successful gene therapy when local injection is not practical. Acidic oligopeptides have been shown to enhance drug delivery to bones.

Methods: In this study six-L aspartic acids (D6) were inserted into the AAV2 capsid protein sequence between amino acid residues 587 and 588. 129SVE mice were injected with double-stranded wild-type- (WT-) or D6-AAV2 mCherry expression vectors (3.24 x 1010 vg per animal) via the superficial temporal vein within 24 hours of birth.

Results: Fluorescence microscopy and quantitative polymerase chain reaction confirmed higher levels of mCherry expression in the paraspinal and gluteus muscles in the D6-AAV2 injected mice. The results revealed that although D6-AAV2 was less efficient in the transduction of immortalized cells stronger mCherry signals were detected over the spine and pelvis by live imaging in the D6-AAV2-injected mice than were detected in the WT-AAV2-injected mice. In addition, D6-AAV2 lost the liver tropism observed for WT-AAV2.

Conclusions: An acidic oligopeptide displayed on AAV2 improves axial muscle tropism and decreases liver tropism after systemic delivery. This modification should be useful in creating AAV vectors that are suitable for gene therapy for diseases involving the proximal muscles.

Figure 1

Production of D6-AAV2 vectors andin vitroinfection assay. (A) Electrophoresis and silver staining for the WT- and D6-AAV2 capsid proteins. The picture shows capsid proteins VP1, VP2, and VP3. The three capsid proteins of D6-AAV2 were present at similar relative quantities but had higher molecular weights than those of WT-AAV2. (B) Heparin agarose chromatography. When known amounts of WT- and D6-AAV2 were loaded onto a heparin agarose column, the D6-AAV2 viruses did not bind to the column and were present in the flowthrough fraction. (C) Infectivity in cells shown by the mCherry red fluorescence. HT1080 cells (a and e), HEK293 cells (b and f), C2C12 cells (c and g), and human chondrocytes (d and h) were infected with either WT- or D6-AAV2. Compared with the WT-AAV2-infected cells, the D6-AAV2-infected cells showed either weak fluorescence (HEK293) or no fluorescence (all other types of cells).

Figure 2

IVIS live imaging for three mice at four months after injection. The D6-AAV2-injected mouse is on the left, the lactated Ringer-injected mouse is in the middle, and the WT-AAV2-injected mice is on the right. (A) An image with the fur removed. (B) An image with the skin removed. (C) Dissected gluteus muscles. (D) Dissected paraspinal muscles. A scale bar was placed on each image. Stronger mCherry fluorescence was observed in the D6-AAV2-injected mouse over the spine and pelvis (A and B) and over the dissected gluteus (C) and paraspinal (D) muscles.

Figure 3

Direct fluorescence microscopy for sections of the paraspinal and gluteus muscles from a D6-AAV2-injected mouse (A and D), a control mouse (B and E), and a WT-AAV2-injected mouse (C and F). Red mCherry fluorescence, visualized using a FITC/Texas Red dual-band filter, was stronger in muscles from the D6-AAV2-injected mouse.

Figure 4

Biodistribution of the WT- (n = 4) and D6-AAV2 (n = 4) vectors in mice four weeks after injection. The vg copy numbers are shown on a log scale on the left. The error bars indicate one standard deviation. *P < 0.05.