A single intravenous injection with 4 × 107 PFU of recombinant adenovirus encoding mouse β-galactosidase cDNA to newborn mice provided widespread increases of β-galactosidase activity, and attenuated the development of the disease including the brain at least for 60 days. The β-galactosidase activity showed 2–4 times as high a normal activity in the liver and lung, and 50 times in the heart. In the brain, while the activity was only 10–20% of normal, the efficacy of the treatment was distinct. At the 30th day after the injection, significant attenuation of ganglioside GM1 accumulation in the cerebrum was shown in three out of seven mice. At the 60th day after the injection, the amount of ganglioside GM1 was above the normal range in all treated mice, which was speculated to be the result of reaccumulation. However, the values were still definitely lower in most of the treated mice than those in untreated mice. In the histopathological study, X-gal-positive cells, which showed the expression of exogenous β-galactosidase gene, were observed in the brain. It is noteworthy that neonatal administration via blood vessels provided access to the central nervous system because of the incompletely formed blood–brain barrier.
A number of therapeutic experiments using the murine models have been performed for various lysosomal enzyme deficiencies, such as mucopolysaccharidosis VII (MPS VII) mouse.1,2 For the treatment of lysosomal storage disease (LSD), the enzyme does not necessarily need to be produced within the affected cells, but can be taken up from the extracellular milieu via binding to mannose-6-phosphate receptors on the cell surface in many of the lysosomal enzymes.3 Thus, the deficient enzyme can be supplied with the administered enzyme protein or the secreted enzyme from the transplanted normal cells or genetically reconstructed cells secreting the enzyme protein. This ability of cells to internalize lysosomal enzymes and direct them to the lysosomal compartment forms the biochemical basis of the potential for the therapeutic strategies for LSD.
Enzyme replacement therapy (ERT),4,5 bone marrow transplantation (BMT),1,6,7 and gene transfer2,8 have been studied in animals and in humans with LSDs. ERT is clinically available for Gaucher disease9 and Fabry disease10 as medicine in many countries, and they are very effective. ERTs for MPS I, II, and glycogen storage disease type II are ongoing to clinical uses. However, the effects of ERT are transient, requiring repeated administrations of the enzyme protein throughout life to maintain activity and to prevent the disease. Moreover, no effect is shown on the brain because of the blood–brain barrier. BMT is effective on the somatic involvement of LSDs.6,11 However, the uses of BMT are limited for the lack of appropriate donors, or morbidity associated with allogeneic transplantation. Moreover, bone and the brain are the exceptions to the effects of BMT. Gene transfer using virus vectors via blood vessels is also effective on various organs.2 The transduced cells would make the enzyme protein from the transferred gene to be reconstituted, and the enzyme protein could be secreted from the cells and be delivered to uninfected cells locally or distantly. However, only direct injections of the gene into the brain or into the ventricle have been shown to be effective on the central nervous system.12,13,14 The blood–brain barrier blocks the enzyme protein-mediated correction or the gene transfer into the brain in any of these approaches. According to the previous literatures, neonatal treatments of ERT or BMT provide a more complete correction in many organs, even in the brain.15,16,17 It has been reported that neonatal gene transfer in MPS VII mouse provided sufficient effects on most tissues including the brain.18,19
We focused our studies on the treatment of the brain for GM1 gangliosidosis. GM1 gangliosidosis is a member of LSDs, which shows a progressive neurological disease in humans caused by the genetic defect of lysosomal acid β-galactosidase that hydrolyses the terminal β-galactosidic residue of ganglioside GM1 and other glycoconjugates.20 The defects of the β-galactosidase activity result in an accumulation of ganglioside GM1 in various organs especially in the brain, which causes the progressive neurodegeneration. Since the gene transfer by intracranial injection is an invasive therapy for the patients, we carried out our study of treating the brain by intravenous gene transfer for the murine model of GM1 gangliosidosis. As it is speculated that protective protein/cathepsin A is needed for β-galactosidase stabilization, β-galactosidase may not be amenable to treatment by the overexpression from the exogenous gene. However, we could show that intravenous gene transfer into the brain was reliable in neonatal period and a small amount of β-galactosidase activity would be sufficient for the brain to prevent the disease progression. It is obvious that the success of the brain therapy in GM1 gangliosidosis mouse provides a promising tool for the brain treatment of LSDs in general.
β-Galactosidase activity in HeLa cells and in the culture medium infected at multiplicity of infection (MOI) 40 by the recombinant adenovirus vector carrying mouse β-galactosidase cDNA were 3–5 times and 40–50 times higher than in the cells and in the medium with mock infection, respectively (data not shown).
β-Galactosidase activity in each organ obtained at the 30th and 60th days after the injection was shown in Table 1. In treated mice, the β-galactosidase activity of the liver and the heart increased definitely in every sample, while the activity of other organs did not increase in a few samples. The maximum increase of the activity was remarkable in the liver, lung, and the heart. The activities at the 60th day were lower than those at the 30th day in most of the samples. Very large deviation values in the increased activities were seen in every organ. Figure 1 shows the activities in each organ of each mouse at the 30th day. The levels of the activity in each organ showed a similar feature among the mice, but the ratio of the increased activity in each organ was very different and did not show a parallel view among the mice. A slight increase of β-galactosidase activity in the brain was detected, and the increases in some mice were significant as shown in Table 2.
Ganglioside GM1 analysis by thin-layer chromatography (TLC)
Figure 2 shows the thin-layer chromatogram of the brain extracts. The accumulation of ganglioside GM1 in the brain was almost corrected in some of the treated mice at the 30th day after the injection. The accumulation of ganglioside GM1 in the liver was corrected in every treated mouse (data not shown).
Table 2 shows the β-galactosidase activity and the amount of ganglioside GM1 of brain in each treated mouse, and the mean±1SD values in untreated knockout mice and normal control mice. At the 30th day, four (mouse nos. 1, 5, 6, and 7) out of seven mice showed higher activity of β-galactosidase and a significant decrease of ganglioside GM1 accumulation in the cerebrum. Three mice of the remaining (nos. 2, 3, and 4) did not show any increase of the β-galactosidase activity or attenuation of the disease in the brain. At the 60th day, five (mouse nos. 3, 7, 8, 9, and 10) out of 10 mice showed higher activity of β-galactosidase corresponding to the decrease of ganglioside GM1 in cerebrum. An excellent efficacy was shown in mouse no. 3, which still had a significant higher activity of β-galactosidase and a lower amount of ganglioside GM1 accumulation. The amounts of ganglioside GM1 were significantly lower in most of the treated mice than those in the untreated mice, but were definitely higher than those in normal mice.
In the liver, β-galactosidase activity showed 73–857% of the mean value of normal activity at the 30th day, which resulted in the prevention of ganglioside GM1 accumulation almost completely (data not shown). At the 60th day, β-galactosidase activity in the liver still showed 25–225% of the normal activity, and the continuous prevention of ganglioside GM1 accumulation was observed (data not shown).
β-Galactosidase-positive cells by X-gal staining were detected in various organs (data not shown) including the brain. X-gal stainings of the brain and the liver are shown in Figures 3 and 5. Figures 4a and b show anti-GM1 ganglioside staining of the brain in treated and untreated mouse, respectively. The amount of immuno-reactive materials was definitely less in the treated mouse than in the untreated mouse, indicating attenuation of ganglioside GM1 accumulation by the neonatal gene transfer. The percentages of the infected cells counted in the specimen with X-gal staining were 0–1% in the brain and 5–10% in the liver.
A single intravenous injection of recombinant adenovirus encoding mouse β-galactosidase cDNA for newborn mice provided widespread increases of β-galactosidase activity and attenuated the development of the disease including the brain at least for 60 days.
It is noteworthy that neonatal administration via blood vessels provided access to the brain, and it is a noninvasive method. Although the activity in the treated brain was not remarkable, the accumulation of ganglioside GM1 was attenuated definitely in three out of seven mice at the 30th day (Table 2, mouse nos. 5, 6 and 7). The efficacy was clearly confirmed by biochemical analysis of TLC (Figure 2) and histochemical analysis of immunostaining (Figure 4a and b). It is consistent with the clinical fact that a very slight residual catalytic activity in the patients attenuates the symptoms, and results in a mild phenotype.20 At the 60th day after the treatment, the amount of ganglioside GM1 was above the normal range in all treated mice, which was speculated to be the result of reaccumulation. However, the values were still definitely lower in most of the treated mice than those in untreated mice (Table 2). The continuous activity of β-galactosidase about 15–20 nmol/mg/h (20% of normal activity) in the brain would be sufficient to prevent the disease.
A small number of cells stained with X-gal, which means the expression of β-galactosidase activity in the cells, were found in the cerebrum of the mice with good efficacy at the 30th day after treatment (Figure 3). These cells were away from the blood vessels and appeared to be neuronal cells from their shape, which leads to the speculation that adenovirus particles got into the brain through the blood–brain barrier via blood vessels and infected the cells.
The efficacy on the attenuation of ganglioside GM1 accumulation was consistent with the increase in β-galactosidase activity. Good efficacy was observed in the liver of every treated mouse, while the efficacy was varied and limited in the brain. The difference in the efficacy on the brain among the treated mice might be caused by the different efficiency of adenovirus infection, which resulted from the different permeability of the blood–brain barrier. Although the adenovirus injections were carried out at 24–48 h after birth, the maturation of blood–brain barrier would be definitely different between at 24 h and at 48 h because the brain of the newborn mouse develops very rapidly.
The activities of β-galactosidase were very much increased in the liver, lung, and the heart (Table 1), where the high vascularity and abundant blood flow may give predominance of adenovirus infection. Moreover, large deviations of activity among the treated mice would suggest that the activity resulted from the infectious efficiency. The primary source of β-galactosidase activity is speculated to be direct infection by adenovirus, not cross-correction by secreted enzyme. This hypothesis is supported by the fact that the activities in each organ in treated mice were not parallel among the organs (Figure 1). In other words, the mouse having the highest activity in the liver did not necessarily have the highest activity in the brain, and vice versa.
Although the direct infection of tissues by adenovirus vector seems to be the primary source of β-galactosidase activity, interorgan and intraorgan cross-correction by circulating enzyme must also be a factor. The interorgan cross-correction might play a major role in the early stage of infection when high levels of serum activity are present. Since the blood–brain barrier is not fully intact until 10–14 days of life in rodents,21 it would be possible for the proteins to enter from blood flow into the brain, as well as for the virus particles. The evidence of intraorgan cross-corrections was shown both in the brain and in the liver of this study. The remarkable attenuation of ganglioside GM1 accumulation in the brain and the liver was shown in some treated mice by biochemical analysis, while only a small number of cells (less than 1 and 10% of the cells in the brain and in the liver, respectively) were positive with X-gal staining (Figures 3 and 5). Since the infected cells are the only source of the β-galactosidase activity in the brain after the closure of the blood–brain barrier, the vector expressing the enzyme activity constitutively is needed for the successful prevention of the disease. As the maturation of blood–brain barrier in the human, which is completed by birth, is definitely different from that in the mouse, intrauterine treatment may be needed for human patients. However, the treatment before the massive accumulation occurs is important definitely, and it would be possible that the low level but persisting activity of the enzyme could prevent the further accumulation and the disease development throughout life.
Materials and methods
GM1-gangliosidosis mouse model (β-galactosidase-deficient mouse)
Mouse model of GM1 gangliosidosis was generated by targeting of β-galactosidase gene in ES cells as previously described.22,23 Newborn mice were obtained by mating the heterozygous female mice with the homozygous male mice. Identification of newborn mutants was accomplished by quantitative analysis of β-galactosidase activity in the tail tip homogenates on the day of birth. Age-matched wild-type mice of C57BL/6 strain were used as the control.
All surgical and care procedures were carried out according to the Guidelines for Use and Care of Experimental Animals approved by the Animal Committee of Osaka City University School of Medicine.
Construction of adenovirus vector with β-galactosidase cDNA
AdEasy system,24 the system for constructing adenovirus vector, was kindly provided by Dr. Vogestein at Johns Hopkins Oncology Center. Mouse β-galactosidase cDNA was cloned and constructed as previously reported.25 Construction of adenovirus vector carrying mouse β-galactosidase cDNA was performed by the protocol described at http://www.coloncancer.org/adeasy/protocol.htm. Mouse β-galactosidase cDNA was cloned into the shuttle vector pAdTrackCMV having GFP gene as a reporter, linearized by digesting with PmeI, and cotransformed by electroporation with an adenoviral backbone plasmid pAdEasy-1 into E. coli BJ5183 cells. Recombinants were selected for kanamycin resistance, and linearized by digesting with PacI, then transfected by Lipofectamine (GIBCO BRL, Rockville, MD, USA) into the 293 adenovirus packaging cells. Recombinant adenoviruses were generated within 7–12 days. The recombinant virus solution was obtained by removing the host cells by sonication and centrifugation. The virus was amplified by repeating the infection with the viral supernatant. The concentration of viral solution was achieved finally at 3.93 × 109 PFU/ml.
To confirm the virus activity, infection to HeLa cells was performed at an MOI of 40. The infection and the protein formation by the recombinant virus were determined by GFP fluorescence on microscope and the analysis of β-galactosidase activity in the cell homogenate and in the culture medium.
Administration of adenovirus vector and the preparation of tissues
Each mouse received a single intravenous injection of 100 μl of viral suspension with 4 × 107 PFU via the superficial temporal vein from 24 to 48 h after birth. Mice were examined on the 30th and the 60th days after birth. One hemisphere of the brain was used for biochemical analyses, and the other was used for histological examinations in each mouse.
For the biochemical analysis, mice were anesthetized with diethylether and the blood was washed out with normal saline by perfusing through the heart, and each organ was wrapped with Parafilm and kept at −80°C until use. For the histological studies, the organs were fixed by perfusing through the heart with 4% paraformaldehyde in 0.1 M phosphate buffer pH 7.4 (PB) for 20 min, after washing out the blood with normal saline. The organs were placed in the same fixing mixture overnight for paraffin sections. For the use of frozen sections, the tissues were placed in 0.1 M phosphate buffer pH 7.4 containing 30% sucrose, then frozen in liquid nitrogen.
β-Galactosidase activity was analyzed in the tissue homogenate with the artificial substrate of 2 mM 4-methylumbelliferyl β-galactoside at pH 4.0 in 0.1 M sodium citrate-phosphate buffer according to the method described by Suzuki.26 The protein was analyzed by using Bio-Rad protein assay system (Bio-Rad Laboratories, Hercules, CA, USA) of Bradford.27
Lipid extraction, ganglioside isolation, and lipid quantification were performed as described by Fujita et al28 and Hahn et al29. Total lipids were extracted from tissue homogenate in chloroform–methanol (1:2, v/v). Neutral and acidic fractions were separated by reverse-phase column of Varian Bond Elute C-18 (GL Sciences Inc, Tokyo, Japan). The column was preconditioned with chloroform–methanol (1:2, v/v) and 99.5% methanol. The lipids extracted from 100 to 200 mg in wet weight of each tissue were applied onto the column, and the column was washed with methanol–0.9% saline (1:1, v/v). Ganglioside/acidic lipids were eluted with methanol–water (12:1, v/v) and collected for the study. The content of sialic acid in the ganglioside/acidic lipid fraction was analyzed by the colorimetric assay by the resorcinol method30 with N-acetylneuraminic acid as the standard. Lipids were separated by TLC with high-performance thin-layer plates (Merck High-Performance TLC 60; Merck KGaA, Darmstadt, Germany) and developed in the solvent of chloroform–methanol–0.2% CaCl2 (55:45:10; v/v/v). Gangliosides were visualized by resorcinol spray and heating. Each sample was applied in two lanes and repeated twice to show the reliability of the TLC technique. Densitometric quantification of gangliosides was performed using Kodak Digital Science™ EDAS 120 system with 1D Image Analysis software (Eastman Kodak Company, NY, USA). The analysis was carried out within the linear range with respect to the quantity of the lipid using commercially purchased ganglioside GM1 (Sigma G7641, Sigma, MO, USA) as the standard.
Paraffin sections (10 μm thick) were processed and used for hematoxylin/eosin staining. Frozen sections (15 μm thick) were reacted with X-gal by β-Gal staining Kit (Invitrogen Corp., Carlsbad, CA, USA) to visualize β-galactosidase activity. Ganglioside GM1 accumulated in the brain was visualized in frozen sections by immuno-staining of avidin:biotinylated enzyme complex method with anti-GM1 ganglioside monoclonal antibody conjugated with biotin (Seikagaku Corp., Tokyo, Japan) and VECTASTAINTM ABC Kit (Vector Laboratories, CA, USA). The staining procedures were carried out according to the instructions of the company.
The percentage of the infected cells was analyzed in the brain and the liver specimen stained with X-gal by measuring the blue-stained areas in the pictures.
Birkenmeier EH et al. Increased life span and correction of metabolic defects in murine MPSVII after syngeneic bone marrow transplantation. Blood 1991; 78: 3081–3092.
Ohashi T et al. Adenovirus-mediated gene transfer and expression of human β-glucuronidase gene in the liver, spleen, and central nervous system in mucopolysaccharidosis type VII mice. Proc Natl Acad Sci USA 1997; 94: 1287–1292.
Konfeld S . Trafficking of lysosomal enzyme. FASEB J 1987; 1: 462–468.
Sands MS et al. Murine mucopolysaccharidosis type VII: long term therapeutic effects of enzyme replacement and enzyme replacement followed by bone marrow transplantation. J Clin Invest 1997; 99: 1596–1605.
Kakkis ED, Muenzer J, Tiller GE, Waber L . Enzyme replacement therapy in mucopolysacharidosis I. N Engl J Med 2001; 344: 182–188.
Hoogerbrugge PM et al. Allogenic bone marrow transplantation for lysosomal storage diseases. Lancet 1995; 345: 1398–1402.
Suzuki K et al. The Twitcher mouse: central nervous system pathology after bone marrow transplantation. Lab Invest 1998; 58: 302–309.
Matzner U et al. Long-term expression and transfer of arylsulfatase A into brain of arylsulfatase A-deficient mice transplanted with bone marrow expressing the arylsulfatase A cDNA from a retroviral vector. Gene Therapy 2000; 7: 1250–1257.
Pastores GM, Sibille AR, Grabowski GA . Enzyme therapy in Gaucher disease type I: dosage efficacy and adverse and effects in 33 patients treated for 6 to 24 months. Blood 1993; 82: 408–416.
Desnick RJ, Banikazemi M, Wasserstein M . Enzyme replacement therapy for Fabry disease, an inherited nephropathy. Clin Nephrol 2002; 57: 1–8.
Imaizumi M et al. Long term effects of bone marrow transplantation for inborn errors of metabolism: a study of four patients with lysosomal storage disease. Acta Pediat Jpn 1994; 36: 30–36.
Elliger SS et al. Elimination of lysosomal storage in brains of MPS VII mice treated by intrathecal administration of an adeno-associated virus vector. Gene Therapy 1999; 6: 1175–1178.
Frisella WA et al. Intracranial injection of recombinant adeno-associated virus improves cognitive function in a murine model of mucopolysaccharidosis type VII. Mol Ther 2001; 3: 351–358.
Shen JS, Watabe K, Ohashi T, Eto Y . Intraventricular administration of recombinant adenovirus to neonatal Twitcher mouse leads to clinicopatholgical improvements. Gene Therapy 2001; 8: 1081–1087.
Sands MS et al. Syngenic bone marrow transplantation reduces hearing loss associated with murine mucopolysaccharidosis type VII. Blood 1995; 86: 2033–2040.
O'Connor LH et al. Enzyme replacement therapy for murine mucopolysaccharidosis type VII leads to improvements in behavior and auditory function. J Clin Invest 1998; 101: 1394–1400.
Volger C et al. Enzyme replacement therapy in murine mucopolysaccharidosis type VII: neuronal and glial response to β-glucuronidase requires early inhibition of enzyme replacement therapy. Ped Res 1999; 45: 838–844.
Day TM et al. Neonatal gene transfer leads to widespread correction of pathology in a murine model of lysosomal storage disease. Proc Natl Acad Sci USA 1999; 96: 11288–11294.
Day TM et al. Prevention of systemic clinical disease in MPS VII mice following AAV-mediated neonatal gene transfer. Gene Therapy 2001; 8: 1291–1298.
Suzuki Y, Oshima A, Nanba E . β-Galactosidase deficiency (β-galactosidosis): GM1 gangliosidosis and Morquio B disease. In: Scriver CR, Beaudet AL, Sly WS, Valle D (eds) The Metabolic and Molecular Bases of Inherited Disease, 8th edn. McGraw-Hill: New York, 2001, pp 3775–3803.
Stewart PA, Hayakawa EM . Interendothelial junctional change underlie the developmental ‘tightening’ of the blood–brain barrier. Dev Brain Res 1987; 32: 271–281.
Matsuda J et al. Neurological manifestations of knockout mice with β-galactosidase deficiency. Brain Dev 1997; 19: 19–20.
Matsuda J et al. β-Galactosidase-deficient mouse as an animal model for GM1-gangliosidosis. Glycoconjugate J 1997; 14: 729–736.
He TC et al. A simplified system for generating recombinant adenoviruses. Proc Natl Acad Sci USA 1998; 95: 2509–2514.
Nanba E, Suzuki K . Molecular cloning of mouse acid β-galactosidase cDNA: sequence, expression of catalytic activity and comparison with human enzyme. Biochem Biophys Res Comm 1990; 173: 141–148.
Suzuki K . Enzymatic diagnosis of sphingolipidosis. Methods Enzymol 1987; 138: 727–762.
Bradford MM . A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 1976; 72: 255–260.
Fujita N et al. Targeted disruption of the mouse sphingolipid activator protein gene: a complex phenotype, including severe leukodystrophy and wide-spread storage of multiple sphingolipids. Hum Mol Genet 1996; 5: 711–725.
Hahn CN et al. Generalized CNS disease and massive GM1-ganglioside accumulation in mice defective in lysosomal acid β-galactosidase. Hum Mol Genet 1997; 6: 205–211.
Svennerholm L . Estimation of sialic acids. II. A colorimetric resorcinol-hydrochloric acid method. Biochem Biophys Acta 1957; 24: 604–611.
We thank Dr Junichiro Matsuda at the National Institute of Infectious disease, Tokyo, for providing us β-galactosidase knockout mouse. We also thank Dr Kunihiko Suzuki in the University of North Carolina at Chapel Hill, and Dr Marie T Vanier in University of Lyon, France, for their critical readings of the manuscript and many helpful suggestions.
This work was supported by grants AT-11694306 and AT-11557060 from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
About this article
Cite this article
Takaura, N., Yagi, T., Maeda, M. et al. Attenuation of ganglioside GM1 accumulation in the brain of GM1 gangliosidosis mice by neonatal intravenous gene transfer. Gene Ther 10, 1487–1493 (2003) doi:10.1038/sj.gt.3302033
- GM1 gangliosidosis
- neonatal gene transfer
- intravenous administration
- brain gene therapy
Genetics and Molecular Biology (2019)
Journal of Pharmacy and Pharmacology (2015)
Systemic AAV9 gene transfer in adult GM1 gangliosidosis mice reduces lysosomal storage in CNS and extends lifespan
Human Molecular Genetics (2015)
Evaluation of N-nonyl-deoxygalactonojirimycin as a pharmacological chaperone for human GM1 gangliosidosis leads to identification of a feline model suitable for testing enzyme enhancement therapy
Molecular Genetics and Metabolism (2012)
Chemical chaperone therapy: chaperone effect on mutant enzyme and cellular pathophysiology in β-galactosidase deficiency
Human Mutation (2011)