Characterization of Fabry mice treated with recombinant adeno-associated virus 2/8-mediated gene transfer

Abstract

Background: Enzyme replacement therapy (ERT) with alpha-galactosidase A (alpha-Gal A) is currently the most effective therapeutic strategy for patients with Fabry disease, a lysosomal storage disease. However, ERT has limitations of a short half-life, requirement for frequent administration, and limited efficacy for patients with renal failure. Therefore, we investigated the efficacy of recombinant adeno-associated virus (rAAV) vector-mediated gene therapy for a Fabry disease mouse model and compared it with that of ERT.

Methods: A pseudotyped rAAV2/8 vector encoding alpha-Gal A cDNA (rAAV2/8-hAGA) was prepared and injected into 18-week-old male Fabry mice through the tail vein. The alpha-Gal A expression level and globotriaosylceramide (Gb3) levels in the Fabry mice were examined and compared with Fabry mice with ERT. Immunohistochemical and ultrastructural studies were conducted.

Results: Treatment of Fabry mice with rAAV2/8-hAGA resulted in the clearance of accumulated Gb3 in tissues such as liver, spleen, kidney, heart, and brain with concomitant elevation of alpha-Gal A enzyme activity. Enzyme activity was elevated for up to 60 weeks. In addition, expression of the alpha-Gal A protein was identified in the presence of rAAV2/8-hAGA at 6, 12, and 24 weeks after treatment. alpha-Gal A activity was significantly higher in the mice treated with rAAV2/8-hAGA than in Fabry mice that received ERT. Along with higher alpha-Gal A activity in the kidney of the Fabry mice treated with gene therapy, immunohistochemical studies showed more alpha-Gal A expression in the proximal tubules and glomerulus, and less Gb3 deposition in Fabry mice treated with this gene therapy than in mice given ERT. The alpha-gal A gene transfer significantly reduced the accumulation of Gb3 in the tubules and podocytes of the kidney. Electron microscopic analysis of the kidneys of Fabry mice also showed that gene therapy was more effective than ERT.

Conclusions: The rAAV2/8-hAGA mediated alpha-Gal A gene therapy provided improved efficiency over ERT in the Fabry disease mouse model. Furthermore, rAAV2/8-hAGA-mediated expression showed a greater effect in the kidney than ERT.

Figure 1

PCR analysis of transduced α-Gal A gene in Fabry mice. DNA was extracted from the organs of Fabry mice 6, 12, and 24 weeks after vector injection and analyzed by PCR. rAAV-hAGA, whereas the 1.4-kb fragment corresponds to the mouse genomic α-Gal A gene. The distribution was identified in liver (A) and kidney (B) at 6, 12, and 24 weeks after treatment.

Figure 2

Western blot analysis of α-galactosidase A expression in liver and kidney at 6, 12, and 24 weeks after treatment in Fabry mice. Liver and kidney tissue lysates were immunoblotted using anti-α-galactosidase A antibodies. α-Gal A protein expression in liver and kidney at 6, 12, and 24 weeks after injection is demonstrated in (A), and levels of α-Gal A in liver (B) and kidney (C) were quantified using a bioimaging analyzer. Experiments were repeated approximately three to five times using each sample. Values are expressed as means ± SD (n = 3 or 4). *p < 0.05, **p < 0.01 and ***p < 0.001 vs. GAPDH using Student's t test. Black bar: 6 weeks after injection (n = 5), gray bar: 12 weeks after injection (n = 3), white bar: 24 weeks after injection (n = 3).

Figure 3

α-Gal A immunostaining in the kidney. Kidney sections were stained with peroxidase-conjugated rabbit anti-human α-galactosidase A shown as browning to plasmic staining. (A) Wild-type mice, (B) untreated Fabry mice, (C) Fabry mice treated with ERT, (D) Fabry mice treated with gene therapy (2 × 10¹² particles of rAAV2/8-hAGA) (×200).

Figure 4

Immunohistochemistry of CD77/Gb3 in the kidney of Fabry mice. (A) Wild-type mice unstained, (B) staining appeared in glomeruli and tubules of untreated Fabry mice, (C) stained tubules and glomeruli in Fabry mice treated with ERT, (D) No detection after 6 weeks in Fabry mice injected with 2 × 10¹² particles of rAAV2/8-hAGA (×200).

Figure 5

Gb3 clearance from proximal renal tubules of kidney of rAAV 2/8-treated Fabry mice. (A) Wild-type mice, (B) Fabry mice (at 24 weeks), (C) 6 weeks after ERT, and (D) 6 weeks after gene therapy (2 × 10¹² rAAV2/8-hAGA). The mice were killed 6 weeks after injection and kidney tissue was examined by electron microscopy. Gb3 containing myeloid bodies were recognized in proximal tubules (×8000).

Figure 6

Ultrastructure of podocytes in the kidney of Fabry mouse. Compared with wild-type mice (A, × 15000), foot process effacement and thickening of the basement membrane were noted in Fabry mice kidneys (B, × 12000). After 6 weeks ERT (C, × 12000), foot process fusion appeared in a few glomerular podocytes. The glomerular podocytes of kidneys from mice with gene therapy (2 × 10¹² particles of rAAV2/8-hAGA) appeared normal (D, ×12000).