Abstract
Fabry disease is a systemic disease caused by genetic deficiency of a lysosomal enzyme, α-galactosidase A (α-gal A), and is thought to be an important target for enzyme replacement therapy. We studied the feasibility of gene-mediated enzyme replacement for Fabry disease. The adeno-associated virus (AAV) vector containing the α-gal A gene was injected into the right quadriceps muscles of Fabry knockout mice. A time course study showed that α-gal A activity in plasma was increased to ≈25% of normal mice and that this elevated activity persisted for up to at least 30 weeks without development of anti-α-gal A antibodies. The α-gal A activity in various organs of treated Fabry mice remained 5–20% of those observed in normal mice. Accumulated globotriaosylceramide in these organs was completely cleared by 25 weeks after vector injection. Reduction of globotriaosylceramide levels was also confirmed by immunohistochemical and electronmicroscopic analyses. Echocardiographic examination of treated mice demonstrated structural improvement of cardiac hypertrophy 25 weeks after the treatment. AAV vector-mediated muscle-directed gene transfer provides an efficient and practical therapeutic approach for Fabry disease.
Fabry disease is an X-linked lysosomal disease caused by deficiency of α-galactosidase A (α-gal A) (1). Undigested glycosphingolipids with terminal α-galactosyl moieties, mainly globotriaosylceramide (Gb3), progressively accumulates in various types of cells in systemic organs including vascular endothelial cells, renal epithelial cells, and myocardial cells. The classical form of Fabry patient with no detectable enzyme activity develops angiokeratoma, hypohidrosis, and episodic pain crises in the extremities during childhood or adolescence. With increasing age, vital organs are also affected, and patients eventually succumb to renal or cardiac complications (2). Recently, a variant form of Fabry disease with mild clinical manifestations primarily limited to cardiac symptoms has been described. In contrast to patients with the classical form, atypical variants have residual activity of α-gal A (3).
One possible therapeutic approach for Fabry disease is enzyme replacement (4). Like most lysosomal enzymes, α-gal A targets the lysosome through an interaction between mannose 6 phosphate (M6P) moieties of enzyme molecules and M6P receptor molecules in the Golgi apparatus and endosomes. Enzymes escaping from this pathway are secreted and often recaptured by M6P receptors on the cell membrane of surrounding cells that direct the enzyme to the lysosome (5). It has been confirmed that exogenously added α-gal A is taken up and catabolizes Gb3 in fibroblasts from Fabry patients (6). This phenomenon is referred to as metabolic cooperativity or cross-correction and is the rationale behind enzyme replacement therapy of Fabry disease. Preclinical studies using α-gal A-deficient mice also demonstrated that i.v. infusion of α-gal A clears the accumulated lipid in various organs (7). Based on these studies, a phase I/II clinical trial of enzyme replacement therapy was recently initiated (8, 9). Preliminary evaluation has suggested the safety and potential therapeutic value of enzyme replacement for Fabry disease. However, a serious limitation is the short half-life of enzyme molecules. Repeated administration of large amounts of the enzyme is required for long-term metabolic correction.
An alternative approach is gene-mediated enzyme replacement therapy in which the expression vector for α-gal A is inserted into the patient's cells, instead of direct infusion of α-gal A protein. Bone marrow cells were transduced with a retroviral vector containing the α-gal A gene and were transplanted into sublethally irradiated Fabry mice (10). Increased α-gal A activity and decreased Gb3 storage were observed in multiple organs of recipient animals up to 26 weeks. In this protocol, bone marrow cells were used as a reservoir for systemic and continuous secretion of α-gal A. Unfortunately, the efficiency of retrovirus-mediated gene transfer into human bone marrow cells is still too low to apply this approach to clinical trials. Because retroviral vectors can transduce only dividing cells, it is difficult to find other target cells susceptible to retrovirus-mediated gene transfer. Another possibility is the use of adeno-associated virus (AAV) vectors, which are able to transduce nondividing cells and achieve long-term expression of a therapeutic gene from either integrated or episomal vector genome (11). The AAV vector carrying the α-gal A gene was delivered to the liver via the hepatic portal vein, resulting in long-term correction of both the enzyme and the glycolipid storage defects in the Fabry mice (12). However, the safety of i.v. injection of viral vectors must be carefully evaluated before clinical application (13).
We examined a simple and clinically applicable strategy for gene-based enzyme replacement of Fabry disease. The AAV vector was directly injected into muscle of the Fabry mice, resulting in long-term expression of α-gal A and complete normalization of lipid metabolism. Structural correction of the heart was also confirmed by an echocardiographic examination.
Materials and Methods
Vector Production.
AAV vector plasmids, pAAV.CAαGTN and pAAV.CAαGBE, are derivatives of psub201 (14). Each contains the human α-gal A cDNA driven by the CAG promoter composed of the chicken β-actin promoter and the CMV enhancer (15). pAAV.CAαGTN has the neoR expression unit from pMC1neo (16), whereas pAAV.CAαGBE has the enhanced green fluorescence protein (EGFP) gene driven by the B19 promoter (17). Recombinant AAV vectors (AAV.CAαGTN and AAV.CAαGBE) were generated by the adenovirus-dependent classical transfection–infection method and concentrated by the combination of sulfonated-cellulose column chromatography (Seikagaku Kougyo, Tokyo) and ultrafiltration using Microcon-YM30 (Millipore) (18). All vector stocks were assayed for adenovirus contamination by using a plaque assay on permissive HeLa cells and were completely free of adenovirus. The titer of AAV vectors was determined by slot-blot hybridization assay. The titer of final preparation of each AAV vector was ≈1 × 1012 vector genomes per ml.
Cell Lines.
HeLa cells and fibroblasts from Fabry patients, and normal subjects were grown in DMEM supplemented with 10% FBS.
Animals.
Fabry model mice were bred from C57BL/6 hemizygous male mice (−/0) and homozygous female mice (−/−) obtained from Kulkarni (19). All animal experiments were performed according to protocols approved by the Nippon Medical School Animal Ethics Committee. Female mice (−/−) were anesthetized with diethyl ether, and 1.5 × 1011 vector genomes in 100 μl of AAV vectors were injected into the right quadriceps muscle by using a 23 gauge needle. Blood samples were collected from the retro-orbital puncture of ether-anesthetized mice every 2 weeks postinjection. The animals were perfused with PBS before removing the organs.
Assay of α-Gal A.
Fluorometric assay of α-gal A was performed as described (20) with minor modification. HeLa cells or confluent fibroblasts were harvested by trypsin treatment, and pelleted by low-speed centrifugation. The cell pellets were washed three times with PBS, frozen, and thawed three times in homogenization buffer [28 mM citric acid/44 mM disodium phosphate/2% (wt/vol) Triton X-100, pH 4.4] and then centrifuged at 14,000 rpm (15,000 × g) for 30 min. The supernatant was assayed for α-Gal A activity by incubation with 5 mM 4-methylumbelliferyl-α-d-galactopyranoside in the presence of 100 mM N-acetylgalactosamine, a specific inhibitor of N-acetylgalactosaminidase (α-galactosidase B) activity (21). Protein concentrations were determined by the method of Bio-Rad Protein Assay. One unit of α-gal A activity is equivalent to the hydrolysis of 1 nmol of substrate in 1 h at 37°C.
Quantification of Gb3 Levels.
The crude lipids were extracted from tissue homogenates by the method of Folch et al. (22). Briefly, the lipids were successively extracted with mixtures of chloroform/methanol (2/1, vol/vol), chloroform/methanol (1/1, vol/vol), and chloroform/methanol/water (60/30/8, vol/vol/vol). Glycosphingolipids recovered in Folch's lower phase were dried under a stream of nitrogen and then treated with mild alkaline (0.1 M NaOH in methanol) at 40°C for 1 h. After neutralizing the solution with 1 M acetic acid, glycosphingolipids were again recovered in Folch's lower phase and then applied to precoated high-performance thin-layer chromatography (HPTLC)-silica gel 60F254 plates (100 × 200 mm, Art. 13728; E. Merk, Darmstadt, Germany). The developing solvent used was chloroform/methanol/water (65/35/8, vol/vol/vol). Glycosphingolipids were visualized with orcinol reagent, and their amounts were quantitatively determined by densitometric scanning with AE6920M-CX (Atto, Tokyo). Standard Gb3 was obtained from Matreya (State College, PA).
Anti-α-Gal A Immune Response.
The antibody response against α-Gal A was measured as described (12), with some modifications. Briefly, 96-well optical bottom flat plate for ELISA (Nunc) were coated with 1 μg/ml of recombinant α-Gal A produced by a Pichia pastoris expression system using an expression vector pPIC9 (HIS4; Invitrogen). After washing in 0.05% (vol/vol) Tween 20-PBS, plates were blocked in 0.1% BSA in sodium phosphate buffer (0.1 M, pH 7.5) for 1 h at 4°C. Subsequently, sera at 5, 15, and 25 weeks were serially diluted in antibody buffer (0.5% BSA-PBS, pH 8.0) and incubated for 1 h at 37°C. After washing with Tween 20-PBS, anti-mouse IgG (H+L) alkaline phosphatase-conjugated antibody developed in goat (Promega) was added in a volume of 100 μl per well and incubated for 1 h at 37°C. After washing with Tween 20-PBS, substrate solution (10 mM p-nitrophenylphosphate/5 mM magnesium chloride/0.1 M 2-amino-2-methyl-1,3-propanediol-HCl buffer, pH 10.0) was added (150 μl per well) and incubated for 30 min. The enzyme reaction was terminated with addition of 100 μl 1 M sodium hydroxide. The optical density was determined at 420 nm. Titers were estimated from a standard curve derived with serial dilutions of an anti-recombinant human α-gal A mouse IgG1 (#6G9.6.1.6; kindly supplied by Seng H. Cheng, Genzyme). The lower limit of detection was equivalent to 1 μg/ml of the mouse monoclonal antibody.
DNA Analysis.
DNA was extracted from each organs by using QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) and suspended in 10 mM Tris-Cl/0.5 mM EDTA-Na (pH 9.0). PCR amplification was done in 50 μl of PCR buffer (50 mM KCl/10 mM Tris-Cl, pH 9.0/0.1% Triton X; Promega), containing 2.5 μg of template DNA, 10 pmol each of the primers, 0.2 mM dNTP, and 2.5 units of Taq polymerase, for 30 cycles of 94°C for 1 min, 60°C for 2 min, and 72°C for 5 min. The primers for α-gal A were 5′-GACACATCAGCCCTCAAGCCAAAGC-3′ (sense) and 5′-CAATCTCCTGCCGGTTTATCATAGC-3′ (antisense). These primers amplify 177 bp of human α-gal A cDNA derived from the transgene and 1.3 kb of mouse genomic α-gal A gene. Integrity of DNA was determined by amplifying a 604-bp region of the murine β-actin gene, using appropriate primers (5′-TGACGGGGTCACCCACACTGTGCCCATCTA-3′ and 5′-CTAGAAGCATTTGCGGTGGACGATGGAGGG-3′) (23).
Immunohistochemical Analysis.
Immunofluorescent staining of Gb3 was performed according to the method described (24) with some modifications. Cells grown on cover slips or frozen-sectioned tissues were fixed with 4% paraformaldehyde in PBS for 30 min at 4°C. After removal of fixing solution, the specimens were rinsed in PBS and blocked with 5% BSA in PBS for 1 h at room temperature. The specimens were then incubated with monoclonal mouse anti-Gb3 IgG (25) at 4°C overnight, followed by washing with ice-cold PBS. Specific binding was visualized with fluorescein isothiocyanate-conjugated F(ab′)2 goat anti-mouse IgG under a confocal laser microscope (LSM 510-Axiovert 100M, Zeiss).
Electron Microscopy.
Thin slices of tissues were fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4, at 4°C, cut into small cubes, rinsed overnight in buffer, postfixed with 1% osmium tetroxide in phosphate buffer, dehydrated, and embedded in epoxy resin (Epok 812, Nagase, Tokyo). One-micrometer-thick sections were stained with alkaline toluidine blue and used to select appropriate areas for ultrathin sectioning. The ultrathin sections were stained with tannic acid, uranyl acetate, and lead citrate, and were observed with an Hitachi H7100 electron microscope (Hitachi, Tokyo).
Echocardiographic Examination.
To assess structural and functional effects of the treatment, 2D and Doppler transthoracic echocardiography was performed using a Powervision 8000 SSA-390A (Toshiba, Tokyo) with a 12.0-MHz imaging transducer (PLM-12023 or PLM-1204T, Toshiba). Mice were lightly sedated by i.p. injection of 20 mg/kg pentobarbital and secured to an imaging platform. Intraventricular septal thickness (IVST) and posterior wall thickness (PWT) were measured by the blinded-manner with no information about the treatments. The heart weight/body weight ratio (HW/BW) was measured at the time the mice were killed.
Results
α-Gal A Activity in Cultured Cells.
The biological activities of recombinant AAV vectors were examined by measuring α-gal A activity in cultured cells after transduction. Intracellular α-gal A activity of nontransduced HeLa cells was 373 ± 11 units/mg protein, and no extracellular (secreted) enzyme activity was detected. After incubation with AAV.CAαGNT at a multiplicity of infection (moi) of 1.0 × 104, the intracellular activity of nonselected HeLa cells was increased to 2,750 ± 215 units/mg protein. The highest enzyme level in neoR-selected clones was 3,150 ± 111 units/mg protein in clone H#13. Secreted α-gal A activity in the conditioned medium derived from 2 × 106 H#13 cells was 130 ± 11 units/ml.
Cultured skin fibroblasts derived from a moderately affected patient with Fabry disease exhibited an α-gal A enzyme activity of 9.4 ± 3.1 units/mg protein, corresponding to 6.7% of the normal fibroblast level (139 ± 19 units/mg protein). After transduction with AAV.CAαGBE at a moi of 1 × 105, intracellular α-gal A activity was increased to 428 ± 22 units/mg protein (Fig. 1A).
Figure 1.
AAV vector-mediated expression of α-gal A in patient's fibroblasts. (A) Fibroblasts from a Fabry patient was transduced with AAV.CAαGBE. α-Gal A activity was determined 3 days after transduction. 1, Nontreated normal fibroblasts; 2, nontreated Fabry fibroblasts; 3, Fabry fibroblasts transduced with 5 × 104 AAV.CAαGBE genomes; 4, Fabry fibroblasts transduced with 1 × 105 AAV.CAαGBE genomes. (B) Fabry fibroblasts were incubated with the conditioned medium (CD) of HeLa cell clone (H#13) secreting α-gal A for overnight. 1, Nontreated fibroblasts; 2, incubation with CD; 3, incubation with CD in the presence of 1 mM mannose-6 phosphate; 4, incubation with CD in the presence of 1 mM glucose-6-phosphate. Values are presented as means ± SD.
We also tested the uptake of α-gal A synthesized in AAV-transduced HeLa cells. The Fabry fibroblasts were cultured with the conditioned medium of H#13 cells containing α-gal A (130 ± 11 units/ml) and washed with PBS, and then intracellular α-gal A activity was measured (Fig. 1B). High-level α-gal A activity was detected in the fibroblasts. The addition of mannose 6 phosphate, but not glucose 6 phosphate, inhibited 88% of the enzyme uptake. These results indicate that AAV vectors are useful for overexpression of functional α-gal A in human cells.
α-Gal A Activity in AAV-Treated Fabry Mice.
AAV.CAαGBE (1.5 × 1011 vector genomes) was injected at a single site in the right quadriceps muscle of 12-week-old Fabry knockout mice. Blood samples were collected by retro-orbital puncture, and α-gal A activity in plasma was monitored every 2 weeks. The enzyme levels increased and reached to a plateau at 6–8 weeks after injection. Circulating α-gal A activity at the plateau in the treated Fabry mice was ≈25% of that in normal mice (8.3 ± 0.3 units/ml) and persisted for up to at least 30 weeks (Fig. 2). An experiment using a lower dose of AAV.CAαGBE (0.5 × 1011 genomes) suggested that the expression levels are dose dependent.
Figure 2.
α-Gal A activity in plasma of AAV-treated mice. AAV.CAαGBE (1.5 × 1011 or 0.5 × 1011 vector genomes) was injected into the right quadriceps muscle of 12-week-old Fabry mice. α-Gal A activity in plasma was determined every 2 weeks. Values are expressed as the mean ± SD (n = 3).
To examine whether the circulating α-gal A is taken up by systemic organs, the treated Fabry mice and age-matched nontreated and normal mice were killed at different time points for measurement of α-gal A activity (Table 1). At 5 weeks after injection, significant increased enzyme activity was detected in all organs except the brain of the treated animals. The α-gal A activity levels in the liver, spleen, heart, and lung were ≈10% of normal, whereas the enzyme activity in the kidney reached 20% of normal. These levels were stably maintained up to 25 weeks after treatment.
Table 1.
α-Gal A activities in organs from Fabry mice treated with AAV.CAαGBE (nmol/h/mg protein)
| Organ | Animals | 5 weeks | 15 weeks | 25 weeks |
|---|---|---|---|---|
| Liver | Untreated | 0.2 ± 0.1 | 0.2 ± 0.1 | 0.1 ± 0.0 |
| Wild | 9.2 ± 1.3 | 8.7 ± 2.0 | 9.0 ± 0.9 | |
| Treated | 1.0 ± 0.1 | 0.8 ± 0.1 | 1.0 ± 0.1 | |
| (Treated/wild) | (10.6%) | (9.7%) | (10.9%) | |
| Spleen | Untreated | 0.4 ± 0.1 | 0.4 ± 0.2 | 0.4 ± 0.1 |
| Wild | 15.4 ± 2.4 | 15.7 ± 2.0 | 16.3 ± 2.2 | |
| Treated | 1.5 ± 0.2 | 1.2 ± 0.4 | 1.4 ± 0.1 | |
| (Treated/wild) | (9.9%) | (7.8%) | (8.5%) | |
| Heart | Untreated | 0.6 ± 0.1 | 0.4 ± 0.1 | 0.6 ± 0.1 |
| Wild | 45.9 ± 7.1 | 36.4 ± 6.4 | 52.3 ± 5.0 | |
| Treated | 3.9 ± 0.4 | 3.4 ± 1.1 | 3.9 ± 0.3 | |
| (Treated/wild) | (8.4%) | (9.4%) | (7.5%) | |
| Lung | Untreated | 0.9 ± 0.1 | 0.8 ± 0.1 | 0.9 ± 0.1 |
| Wild | 67.6 ± 6.4 | 67.0 ± 7.3 | 69.0 ± 5.4 | |
| Treated | 7.3 ± 1.0 | 6.2 ± 0.1 | 7.3 ± 1.0 | |
| (Treated/wild) | (10.8%) | (9.2%) | (10.5%) | |
| Kidney | Untreated | 0.5 ± 0.1 | 0.5 ± 0.1 | 0.5 ± 0.1 |
| Wild | 16.8 ± 3.1 | 16.1 ± 1.2 | 17.5 ± 3.1 | |
| Treated | 3.3 ± 0.2 | 3.2 ± 1.0 | 3.2 ± 0.3 | |
| (Treated/wild) | (19.6%) | (20.0%) | (18.5%) | |
| Brain | Untreated | 0.4 ± 0.1 | 0.5 ± 0.1 | 0.4 ± 0.1 |
| Wild | 29.3 ± 7.5 | 35.2 ± 6.0 | 32.8 ± 2.4 | |
| Treated | 1.6 ± 0.1 | 1.3 ± 0.1 | 1.5 ± 0.2 | |
| (Treated/wild) | (5.4%) | (3.6%) | (4.5%) |
Organs were obtained from treated mice killed 5, 15, and 25 weeks after vector injection. Values are expressed as mean ± SD (n = 3).
Genomic DNA was extracted from various organs at 25 weeks after an intramuscular injection and analyzed for detection of vector genome by PCR (Fig. 3). The human α-gal A specific 177-bp fragment was detectable only in the right thigh muscle where the AAV vector was injected. There was no evidence of spill over to other organs. These data indicate that the injected muscle remained the predominant site for expression of α-gal A.
Figure 3.
PCR analysis of the transduced α-gal A gene in Fabry mice. DNA was extracted from organs of Fabry mice 25 weeks after vector injection and analyzed by PCR. The 177-bp fragment corresponds to the human α-gal A cDNA derived from AAV.CAαGBE, whereas the 1.3-kb fragment corresponds to the mouse genomic α-gal A gene. Integrity of DNA was determined by amplifying a 604-bp region of the murine β-actin gene, using appropriate primers. 1, Heart; 2, kidney; 3, liver; 4, right muscle; 5, left muscle; 6, brain; 7, spleen; 8, lung; 9, positive control.
The levels of anti-α-gal A antibodies in sera of treated animals were determined by an ELISA using a mouse anti-human α-gal A monoclonal antibody as a standard. No specific antibodies against α-gal A were detected (<1 μg/ml) at 5, 15, and 25 weeks after vector injection (data not shown). Lack of neutralizing antibodies is likely to be a major reason for sustained expression of α-gal A in treated mice.
Gb3 Levels in AAV-Treated Fabry Mice.
Next, changes in the Gb3 concentration of various organs were determined (Table 2). At 5 weeks after treatment, moderate reduction of Gb3 levels was observed in heart, lungs, and kidneys. Gb3 levels decreased to less than half of the levels seen in nontreated Fabry mice at 15 weeks in liver, spleen, heart, lung, and kidney. Total clearance of the accumulated Gb3 in all organs was achieved at 25 weeks after AAV injection (Fig. 4).
Table 2.
Gb3 levels in organs from Fabry mice treated with AAV.CAαGBE (nmol/mg protein)
| Organ | Animals | 5 weeks | 15 weeks | 25 weeks |
|---|---|---|---|---|
| Liver | Untreated | 1.3 ± 0.2 | 2.1 ± 0.5 | 2.8 ± 0.9 |
| Wild | 0.2 ± 0.1 | 0.2 ± 0.1 | 0.2 ± 0.0 | |
| Treated | 1.2 ± 0.4 | 0.7 ± 0.1 | 0.2 ± 0.1 | |
| Spleen | Untreated | 4.0 ± 0.8 | 2.0 ± 0.4 | 4.7 ± 0.2 |
| Wild | 0.3 ± 0.1 | 0.2 ± 0.0 | 0.2 ± 0.1 | |
| Treated | 3.6 ± 0.6 | 0.9 ± 0.1 | 0.2 ± 0.0 | |
| Heart | Untreated | 2.5 ± 1.1 | 1.5 ± 0.0 | 4.6 ± 0.7 |
| Wild | 0.4 ± 0.0 | 0.3 ± 0.1 | 0.2 ± 0.1 | |
| Treated | 1.3 ± 0.2 | 0.7 ± 0.2 | 0.3 ± 0.2 | |
| Lung | Untreated | 4.9 ± 2.4 | 4.3 ± 3.0 | 6.8 ± 2.2 |
| Wild | 0.5 ± 0.1 | 0.4 ± 0.1 | 0.3 ± 0.1 | |
| Treated | 2.0 ± 1.6 | 1.9 ± 0.7 | 0.5 ± 0.3 | |
| Kidney | Untreated | 3.9 ± 0.7 | 2.7 ± 0.7 | 3.0 ± 0.7 |
| Wild | 0.2 ± 0.1 | 0.3 ± 0.1 | 0.1 ± 0.0 | |
| Treated | 2.9 ± 0.4 | 1.2 ± 0.7 | 0.1 ± 0.0 | |
| Brain | Untreated | 0.6 ± 0.1 | 0.6 ± 0.1 | 0.9 ± 0.2 |
| Wild | 0.6 ± 0.1 | 0.2 ± 0.0 | 0.1 ± 0.0 | |
| Treated | 1.1 ± 0.4 | 0.5 ± 0.3 | 0.1 ± 0.0 |
Organs were obtained from treated mice killed 5, 15, and 25 weeks after vector injection. Values are expressed as mean ± SD (n = 3).
Figure 4.
Gb3 levels in organs of treated Fabry mice. Gb3 levels of various organs of treated and untreated Fabry mice, and age-matched wild-type mice at 15 (A) and 25 (B) weeks postinjection. Values are expressed as the mean ± SD (n = 3).
Reduction of Gb3 in the kidney was confirmed by histological examination. Immunohistochemical staining with the anti-Gb3 antibody showed that Gb3 was detectable only in the cell membrane fraction of the renal tubular cells from wild-type mice (Fig. 5A). However, many Gb3-positive granules had accumulated in the cytosol fraction of Fabry kidney (Fig. 5B). These granules were efficiently cleared by a single injection of the AAV vector (Fig. 5C). Electron microscopic study of the kidney tissues showed that lipid inclusions with electron-dense concentric lamellar structures were accumulated in the swollen glomerular epithelial cells. The lipid inclusions were also detected in some endothelial and mesangial cells of glomeruli and tubular epithelial cells in Fabry mice (Fig. 5D). Foot processes of glomerular epithelial cells were enfaced, and thick basement membrane was observed. In contrast, no lipid inclusions were detected in glomeruli and tubules in the Fabry mice at 32 weeks after treatment with the AAV vector (Fig. 5E). Less enfacement of foot processes of glomerular epithelial cells and normalized thinner glomerular basement membrane were observed in these treated Fabry mice compared with the untreated ones.
Figure 5.
Histological examination of kidneys of treated Fabry mice. (A–C) Immunohistochemical staining of kidney tissues from a wild-type mouse (A), an untreated Fabry mouse (B), and a treated Fabry mouse 25 weeks after vector injection (C) were immunostained with a monoclonal anti-Gb3 antibody, followed by FITC-labeled anti-mouse IgG. (D and E) Electron microscopic analysis of the kidney. Lipid inclusions with electron-dense concentric lamellar structures in glomerular epithelial cells were observed in the kidneys of untreated Fabry mice (D). The lipid inclusions were cleared 32 weeks after treatment with the AAV vector (E).
Echocardiographic Examination.
The morphological and functional phenotype of the heart of Fabry mice were assessed by high-resolution 2D and Doppler transthoracic echocardiography (Table 3). At 25 weeks post treatment (37 weeks of age), the HW/BW value was 4.80 ± 0.60 for normal mice and 5.48 ± 0.52 for nontreated Fabry mice, indicating that cardiac hypertrophy had been progressed in α-gal A-deficient mice. AAV-mediated expression of α-gal A significantly decreased the HW/BW value to the normal level (4.76 ± 0.31). Apparent increases in intraventricular septal thickness (IVST) and posterior wall thickness (PWT) were also observed in the untreated Fabry mice compared with age-matched normal mice (1.83 ± 0.10 vs. 1.48 ± 0.60 and 1.68 ± 0.31 vs. 1.35 ± 0.17, respectively). IVST was significantly suppressed by AAV-mediated treatment in the Fabry mice (IVST, 1.65 ± 0.10; PWT, 1.58 ± 0.10).
Table 3.
Echocardiographic examination of heart of Fabry mice treated with AAV.CAαGBE
| Untreated | Wild type | Treated | |
|---|---|---|---|
| Plasma α-Gal A, nmol/h/ml | 0.6 ± 0.2 | 8.1 ± 0.4 | 2.2 ± 0.3* |
| Heart Gb3, nmol/mg | 4.6 ± 0.7 | 0.2 ± 0.1 | 0.3 ± 0.2* |
| HW/BW | 5.48 ± 0.52 | 4.80 ± 0.60 | 4.76 ± 0.31** |
| IVST, mm | 1.83 ± 0.10 | 1.48 ± 0.60 | 1.65 ± 0.10** |
| PWT, mm | 1.68 ± 0.31 | 1.35 ± 0.17 | 1.58 ± 0.10 |
Echocardiographic examination of heart was done before being killed at 25 weeks post vector injection. Values are expressed as the mean ± SD (n = 4). HW, heart weight; BW, body weight; IVST, intraventricular septal thickness; PWT, posterior wall thickness.
, P < 0.001;
, P < 0.05 (vs. untreated; unpaired t test).
Discussion
Gene-mediated enzyme replacement is a reasonable and highly promising approach for the treatment of Fabry disease. Because Fabry is a systemic disease, many organs in the body must be treated. However, we are unable to correct all involved organs genetically with current gene-transfer technology. An important feature is that overexpressed α-gal A is efficiently secreted and taken up by the surrounding cells. Based on this metabolic cooperativity, clinical trials of direct enzyme replacement therapy have been initiated and preliminary data appear to be promising (8, 9), although repeated i.v. administration of costly recombinant α-gal A is necessary. For continuous and long-term enzyme supply, gene-mediated enzyme replacement is much superior to direct injection of enzyme molecules.
A single intramuscular injection of the AAV vector resulted in an increase in the plasma α-gal A activity to ≈20% of that of the normal mice. This plasma level of α-gal A persisted up to at least 30 weeks and was sufficient to maintain the tissue α-gal A level of 5–20% of normal. It was reported that heterozygotes with as little as 10% of normal enzyme activity are phenotypically normal (2). In our animal experiments, 10% enzyme activity was sufficient to totally clear deposit of Gb3 in all organs examined.
The utility of AAV vector-mediated muscle-directed gene transfer has been studied in various animal models (26–28). Long-term expression of bacterial β-galactosidase (29) and human proteins including coagulation factor IX (hFIX) (27), erythropoietin (26), and α1-antitrypsin (30) has been successfully achieved by direct injection of AAV vectors in muscle of immunocompetent animals. Unexpectedly, AAV vectors often do not elicit both humoral and cellular immune response to neoantigenic transgene products when injected into muscle (27, 29, 31, 32), whereas adenovirus expressing the identical transgene efficiently activate immune response (33). Three hemophilia patients have now been treated by intramuscular injection with the AAV vector without evidence of formation of neutralizing antibodies against FIX (34). Our results also demonstrated that AAV-mediated expression of α-gal A failed to develop anti-α-gal A antibodies. The mechanisms by which AAV evade immunological responses are not clearly understood, although it was reported that AAV vectors do not efficiently transduce dendritic cells, which are essential for CTL and CD4+ T-cell responses (35).
Recently, Nathwani reported that in three strains of immunocompetent mice, therapeutic levels of hFIX were detected after i.v. administration of the AAV vector, whereas the expression of hFIX was markedly inhibited by intramuscular injection (23). In the latter case, neutralizing anti-hFIX antibodies were detected in all mice. They speculated that ectopically expressed proteins, such as hFIX in muscle cells, may be more immunogenic because the transgene protein is improperly processed to generate neoantigens (23). Expression of α-gal A occurs in all tissues, and therefore it may be less immunogenic. The extent of immune reaction seems to depend on the mouse strain (30) and the transgene protein. Further studies are required for elucidation of the immunogenicity of AAV vector-mediated gene expression. A final evaluation, however, should be done in careful clinical trials in humans, not in mice.
The liver is another possible depot organ for gene-mediated enzyme replacement therapy. Recently, Jung et al. (12) reported on liver-directed enzyme replacement therapy, wherein an AAV vector containing the α-gal A cDNA was injected into the hepatic portal vein of Fabry mice. The activity of α-gal A in the liver increased to 30% of that in normal mice at 2 weeks postinjection, and progressively declined to 5% at 15 weeks. Although the levels of α-gal A activity in each organ were not significantly different from those observed in our present experiments, the kinetics of Gb3 levels are quite different. Following portal vein injection, maximum reduction of Gb3 levels (66–97%) was observed in most organs, except the kidney at 2–5 weeks, and re-accumulation of Gb3 was detected at 15 weeks. Gb3 in the kidneys was highly resistant to liver-directed therapy. In contrast, the muscle-directed therapy as presented here showed that accumulated Gb3 in various organs including the kidneys was completely cleared and remained at the normal level up to 25 weeks.
Although the exact reason for this discrepancy has not yet been clarified, we believe that a continuous systemic supply of the enzyme is important for efficient clearance of accumulated glycolipids. As shown in Fig. 2, after muscle-directed gene transfer, the level of α-gal A in plasma is highly stable beyond 6 weeks. PCR analysis suggested that the AAV vector was predominantly inserted in the injected muscle. Therefore, the α-gal A activity in organs detected in our study (Table 1) seems to be derived from the homogeneously distributed enzyme taken up from the circulation. On the other hand, it may be possible that AAV vectors directly transduced various tissues after i.v. injection. The α-gal A activity in each organ reflects the enzyme expressed in a small number of transduced cells rather than the uptaken enzyme. If this is the case, the α-gal A activity in the majority of nontransduced cells may be too low to clear the accumulated Gb3. The α-gal A concentrations in plasma and the distribution of the AAV vector were not reported after i.v. injection (12).
Efficient clearance of Gb3 in the kidney of Fabry mice seems to be an important advantage of our gene therapy approach. The kidney is an important target organ for treatment of Fabry disease, but the efficiency of α-gal A uptake is moderate, and Gb3 was only partially digested in the kidney after adenoviral-mediated (36) and AAV-mediated (12) gene transfer into liver. In contrast, α-gal A expressed in the muscle was efficiently uptaken in the kidney, resulting in complete clearance of Gb3. It is unknown why α-gal A synthesized in the liver is not efficiently transferred into the kidney. One possibility is that the difference in the sugar chains may affect the uptake of the enzyme into the kidney. It is well known that posttranscriptional modification of lysosomal enzymes is organ specific (37, 38). Another possibility is that anti-α-gal A antibodies may interfere with the uptake of the enzyme into the kidney. In contrast to muscle-directed gene transfer, expression of α-gal A in liver elicits low titers of anti-α-gal A antibody (12).
Echocardiographic examination clearly demonstrated that Fabry mice have cardiomegaly involving the left ventricular wall and intraventricular septum. The cardiac hypertrophy was significantly decreased by the present gene therapy approach. Gb3 levels of treated mice were completely normalized 25 weeks after vector injection, although the wall thickness was still higher than that of age-matched normal mice. It may be possible that irreversible structural changes had already occurred before the enzyme replacement became effective. We are now conducting a series of experiments to determine when the gene therapy should be started to obtain maximum efficacy.
In conclusion, we demonstrated that AAV-mediated muscle-directed gene transfer is very effective for long-term systemic delivery of α-gal A. The safety of intramuscular injection of AAV vectors has been confirmed in gene therapy protocols for hemophilia B (34). This strategy is a highly promising and practical therapeutic approach for Fabry disease.
Acknowledgments
We thank Ashok B. Kulkarni and Toshio Ohshima for providing Fabry knockout mice, Roscoe O. Brady and Seng H. Cheng for valuable discussion, and Kumi Adachi-Takahashi for technical assistance. This work was supported by a grant from the Ministry of Health and Welfare of Japan and the Ministry of Education, Science and Culture of Japan.
Abbreviations
- α-gal A
α-galactosidase A
- Gb3
globotriaosylceramide
- AAV
adeno-associated virus
- hFIX
human coagulation factor IX
Footnotes
This paper was submitted directly (Track II) to the PNAS office.
References
- 1.Brady R O, Gal A E, Bradley R M, Martensson E, Warshaw A L, Laster L. N Engl J Med. 1967;276:1163–1167. doi: 10.1056/NEJM196705252762101. [DOI] [PubMed] [Google Scholar]
- 2.Desnick R J, Ioannou Y A, Eng C M. In: The Metabolic and Molecular Bases of Inherited Disease. Scriver C R, Beaudet A L, Sly W S, Valle D, editors. New York: McGraw–Hill; 2001. pp. 3733–3774. [Google Scholar]
- 3.Nakao S, Takenaka T, Maeda M, Kodama C, Tanaka A, Tahara M, Yoshida A, Kuriyama M, Hayashibe H, Sakuraba H, et al. N Engl J Med. 1995;333:288–293. doi: 10.1056/NEJM199508033330504. [DOI] [PubMed] [Google Scholar]
- 4.Brady R O, Tallman J F, Johnson W G, Gal A E, Leahy W R, Quirk J M, Dekaban A S. N Engl J Med. 1973;289:9–14. doi: 10.1056/NEJM197307052890103. [DOI] [PubMed] [Google Scholar]
- 5.Von Figura K, Hasilik A. Annu Rev Biochem. 1986;55:167–193. doi: 10.1146/annurev.bi.55.070186.001123. [DOI] [PubMed] [Google Scholar]
- 6.Medin J A, Tudor M, Simovitch R, Quirk J M, Jacobson S, Murray G J, Brady R O. Proc Natl Acad Sci USA. 1996;93:7917–7922. doi: 10.1073/pnas.93.15.7917. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ioannou Y A, Zeidner K M, Gordon R E, Desnick R J. Am J Hum Genet. 2001;68:14–25. doi: 10.1086/316953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Eng C M, Guffon N, Wilcox W R, Germain D P, Lee P, Waldek S, Caplan L, Linthorst G E, Desnick R J. N Engl J Med. 2001;345:9–16. doi: 10.1056/NEJM200107053450102. [DOI] [PubMed] [Google Scholar]
- 9.Schiffmann R, Kopp J B, Austin H A, III, Sabnis S, Moore D F, Weibel T, Balow J E, Brady R O. J Am Med Assoc. 2001;285:2743–2749. doi: 10.1001/jama.285.21.2743. [DOI] [PubMed] [Google Scholar]
- 10.Takenaka T, Murray G J, Qin G, Quirk J M, Ohshima T, Qasba P, Clark K, Kulkarni A B, Brady R O, Medin J A. Proc Natl Acad Sci USA. 2000;97:7515–7520. doi: 10.1073/pnas.120177997. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Russell D W, Miller A D, Alexander I E. Proc Natl Acad Sci USA. 1994;91:8915–8919. doi: 10.1073/pnas.91.19.8915. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Jung S C, Han I P, Limaye A, Xu R, Gelderman M P, Zerfas P, Tirumalai K, Murray G J, During M J, Brady R O, et al. Proc Natl Acad Sci USA. 2001;98:2676–2681. doi: 10.1073/pnas.051634498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Donsante A, Vogler C, Muzyczka N, Crawford J M, Barker J, Flotte T, Campbell-Thompson M, Daly T, Sands M S. Gene Ther. 2001;8:1343–1346. doi: 10.1038/sj.gt.3301541. [DOI] [PubMed] [Google Scholar]
- 14.Samulski R J, Chang L S, Shenk T. J Virol. 1987;61:3096–3101. doi: 10.1128/jvi.61.10.3096-3101.1987. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Miyazaki J, Takaki S, Araki K, Tashiro F, Tominaga A, Takatsu K, Yamamura K. Gene. 1989;79:269–277. doi: 10.1016/0378-1119(89)90209-6. [DOI] [PubMed] [Google Scholar]
- 16.Thomas K R, Capecchi M R. Cell. 1987;51:503–512. doi: 10.1016/0092-8674(87)90646-5. [DOI] [PubMed] [Google Scholar]
- 17.Liu J M, Fujii H, Green S W, Komatsu N, Young N S, Shimada T. Virology. 1991;182:361–364. doi: 10.1016/0042-6822(91)90682-2. [DOI] [PubMed] [Google Scholar]
- 18.Tamayose K, Hirai Y, Shimada T. Hum Gene Ther. 1996;7:507–513. doi: 10.1089/hum.1996.7.4-507. [DOI] [PubMed] [Google Scholar]
- 19.Ohshima T, Murray G J, Swaim W D, Longenecker G, Quirk J M, Cardarelli C O, Sugimoto Y, Pastan I, Gottesman M M, Brady R O, et al. Proc Natl Acad Sci USA. 1997;94:2540–2544. doi: 10.1073/pnas.94.6.2540. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kusiak J W, Quirk J M, Brady R O. J Biol Chem. 1978;253:184–190. [PubMed] [Google Scholar]
- 21.Mayes J S, Scheerer J B, Sifers R N, Donaldson M L. Clin Chim Acta. 1981;112:247–251. doi: 10.1016/0009-8981(81)90384-3. [DOI] [PubMed] [Google Scholar]
- 22.Folch J, Lees M, Sloane-Stanley G H. J Biol Chem. 1957;226:497–509. [PubMed] [Google Scholar]
- 23.Nathwani A C, Davidoff A, Hanawa H, Zhou J F, Vanin E F, Nienhuis A W. Blood. 2001;97:1258–1265. doi: 10.1182/blood.v97.5.1258. [DOI] [PubMed] [Google Scholar]
- 24.Itoh K, Kotani M, Tai T, Suzuki H, Utsunomiya T, Inoue H, Yamada H, Sakuraba H, Suzuki Y. Clin Genet. 1993;44:302–306. doi: 10.1111/j.1399-0004.1993.tb03905.x. [DOI] [PubMed] [Google Scholar]
- 25.Kotani M, Kawashima I, Ozawa H, Ogura K, Ariga T, Tai T. Arch Biochem Biophys. 1994;310:89–96. doi: 10.1006/abbi.1994.1144. [DOI] [PubMed] [Google Scholar]
- 26.Bohl D, Bosch A, Cardona A, Salvetti A, Heard J M. Blood. 2000;95:2793–2798. [PubMed] [Google Scholar]
- 27.Chao H, Samulski R, Bellinger D, Monahan P, Nichols T, Walsh C. Gene Ther. 1999;6:1695–1704. doi: 10.1038/sj.gt.3301024. [DOI] [PubMed] [Google Scholar]
- 28.Rudich S M, Zhou S, Srivastava R, Escobedo J A, Perez R V, Manning W C. J Surg Res. 2000;90:102–108. doi: 10.1006/jsre.2000.5859. [DOI] [PubMed] [Google Scholar]
- 29.Fisher K J, Jooss K, Alston J, Yang Y, Haecker S E, High K, Pathak R, Raper S E, Wilson J M. Nat Med. 1997;3:306–312. doi: 10.1038/nm0397-306. [DOI] [PubMed] [Google Scholar]
- 30.Song S, Morgan M, Ellis T, Poirier A, Chesnut K, Wang J, Brantly M, Muzyczka N, Byrne B J, Atkinson M, et al. Proc Natl Acad Sci USA. 1998;95:14384–14388. doi: 10.1073/pnas.95.24.14384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Kessler P D, Podsakoff G M, Chen X, McQuiston S A, Colosi P C, Matelis L A, Kurtzman G J, Byrne B J. Proc Natl Acad Sci USA. 1996;93:14082–14087. doi: 10.1073/pnas.93.24.14082. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Xiao X, Li J, Samulski R J. J Virol. 1996;70:8098–8108. doi: 10.1128/jvi.70.11.8098-8108.1996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Yang Y, Li Q, Ertl H C, Wilson J M. J Virol. 1995;69:2004–2015. doi: 10.1128/jvi.69.4.2004-2015.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kay M A, Manno C S, Ragni M V, Larson P J, Couto L B, McClelland A, Glader B, Chew A J, Tai S J, Herzog R W, et al. Nat Genet. 2000;24:257–261. doi: 10.1038/73464. [DOI] [PubMed] [Google Scholar]
- 35.Jooss K, Yang Y, Fisher K J, Wilson J M. J Virol. 1998;72:4212–4223. doi: 10.1128/jvi.72.5.4212-4223.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ziegler R J, Yew N S, Li C, Cherry M, Berthelette P, Romanczuk H, Ioannou Y A, Zeidner K M, Desnick R J, Cheng S H. Hum Gene Ther. 1999;10:1667–1682. doi: 10.1089/10430349950017671. [DOI] [PubMed] [Google Scholar]
- 37.Yamashita K, Hitoi A, Tateishi N, Higashi T, Sakamoto Y, Kobata A. Arch Biochem Biophys. 1983;225:993–996. doi: 10.1016/0003-9861(83)90116-9. [DOI] [PubMed] [Google Scholar]
- 38.Kobata A. Glycoconj J. 2000;17:443–464. doi: 10.1023/a:1011006122704. [DOI] [PubMed] [Google Scholar]