Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Retrovirally expressed human arylsulfatase A corrects the metabolic defect of arylsulfatase A-deficient mouse cells


A deficiency of arylsulfatase A (ASA) causes the lysosomal storage disease metachromatic leukodystrophy (MLD) which is characterized primarily by demyelination of the central nervous system. ASA-deficient mice develop a disease which resembles MLD in many respects and thus serve as an appropriate animal model for this disease. To establish gene therapy protocols for ASA-deficient mice, we constructed two retroviral vectors based on the murine stem cell virus. Both vectors harbor the human ASA cDNA controlled by the retroviral promoter/enhancer element, but differ by the presence or absence of a neomycin resistance gene driven by an internal promoter. A comparative analysis of the one- versus the two-gene vector and an amphotropic versus an ecotropic producer cell line revealed that the amphotropic producer cell line for the one-gene vector transfers ASA overexpression to the target cells most efficiently. The human ASA encoded by this vector is correctly expressed in heterologous mouse cells and corrects the metabolic defect of transduced ASA-deficient murine cells. The constructed one-gene vector might thus be a potentially useful tool for the development of a gene-based therapy for ASA-deficient mice.


A deficiency of the lysosomal enzyme arylsulfatase A (ASA; E.C. causes the lysosomal storage disease metachromatic leukodystrophy (MLD) in which the desulfation of cerebroside 3-sulfate (sulfatide) and other 3-O- sulfogalactosyl-containing glycolipids is defective.1 The metabolic defect results in the intralysosomal accumulation of these lipids mainly in glial cells and leads to a progressive and widespread demyelination of the nervous system. Affected individuals die within a few years after onset of the neurological symptoms.

ASA-deficient mice which were generated by homologous recombination in embryonic stem cells develop a number of histological, biochemical and neurological alterations closely related to MLD and thus represent an appropriate model for the human disease.2

Due to the lack of therapeutic options for MLD, we are interested in the potential of gene therapy approaches for ASA knockout mice. This interest was stimulated by recent gene transfer experiments in β-glucuronidase-deficient mice, which develop a severe central nervous system pathology similar to humans suffering from the corresponding lysosomal storage disease mucopolysaccharidosis type VII (Sly disease). In this mouse model, intracerebral grafting of fibroblasts overexpressing β-glucuronidase from a retroviral vector resulted in the clearance of storage granules from neuronal and glial cells.3 Improvement in storage was also observed following the administration of a recombinant adenoviral vector or a recombinant adeno-associated viral vector.456

Here we report the generation and in vitro analysis of producer cell lines for two retroviral vectors encoding the human ASA polypeptide. These vectors are derived from the murine stem cell virus (MSCV) which, originally constructed to overcome transcriptional silencing of more conventional, Moloney murine leukemia virus-based vectors in murine embryonic stem cells, is highly active also in a variety of other cell types such as differentiated cells of the hematopoietic system.7 In comparative analyses, we evaluated the vector and the vector tropism which yields highest levels of ASA overexpression in transduced target cells. We demonstrate that human ASA expressed from this vector corrects the metabolic defect of ASA-deficient mouse cells. The retroviral vector might thus be suitable for gene therapy approaches in correcting the disease of ASA knockout mice.


Retroviral vectors and producer cells

We constructed two MSCV-based retroviral vectors bearing the full-length human ASA (hASA) cDNA under transcriptional control of the retroviral promoter/ enhancer element.8 One vector, designated MSCV-asa-neo, harbors in addition to the hASA cDNA an internal phosphoglycerokinase promoter which drives the transcription of a bacterial neomycin phosphotransferase gene located downstream from the hASA cDNA (Figure 1). The other vector, MSCV-asa, is a one-gene vector lacking the phosphoglycerokinase promoter-neomycin resistance cassette. Producer cells were generated by stable transfection (MSCV-asa-neo) or crossinfection (MSCV-asa) of amphotropic (GP+AM12) and ecotropic packaging cells (GP+E86). Three producer clones which exported recombinant retroviral vector at a titer of 1 × 104 (GP+E86: MSCV-asa-neo), 1 × 106 (GP+E86: MSCV-asa) and 5 × 105 infectious particles/ml (GP+AM12: MSCV-asa) were subsequently analyzed.

Figure 1

Structures of the retroviral vector plasmids pMSCV-asa and pMSCV-asa-neo. The plasmids were derived from pMSCVneoEB by insertion of the hASA cDNA (HT14/CP8; asa) with or without subsequent deletion of the phosphoglycerokinase promoter (pgk)-neomycin phosphotransferase (neo) cassette.811 In MSCV-asa and MSCV-asa-neo the hASA cDNA is expressed from the long terminal repeat (LTR) of MSCV (arrows). The position of a SmaI fragment used as a probe for quantitative Southern blotting and restriction sites used for the plasmid construction are indicated. For further details see Materials and methods.

Transfer of ASA overexpression to target cells

To compare the capacities of MSCV-asa and MSCV-asa-neo to transfer expression of ASA activity to target cells, NIH/3T3 cells were transduced with GP+E86-packaged retroviral vector and the increase of intracellular ASA activity was determined (Figure 2a). Transduction with MSCV-asa yielded approximately a 30-fold overexpression of ASA.

Figure 2

Intracellular ASA activity of cells after transduction with MSCV-asa-neo and MSCV-asa, respectively. (a) Activity of non-transduced NIH/3T3 cells (black bar) and NIH/3T3 cells transduced with MSCV-asa-neo (grey bar) or MSCV-asa (open bar). Cells transduced with MSCV-asa-neo were selected with 0.4 mg/ml geneticin. Means ± s.d. (n = 3) (b) Activity of non-transduced control cells (black bars) and cells transduced with GP+E86- (grey bars) or GP+AM12-packaged (open bars) MSCV-asa. nd, not determined. Cell lines: means ± s.d. (n = 3); primary cells: n = 1. The artificial ASA substrate p-nitrocatechol sulfate used for this assay is also hydrolysed by non-ASA sulfatases explaining the activity background in the ASA-deficient cell line 17−/−A1. (c) Specific activity of retrovirally encoded hASA in different cell types following transduction with amphotropic MSCV-asa. The specific hASA activity was calculated by the ratio: [(mU ASA per mg protein following transduction) – (mU ASA per mg protein of non-transduced control cells)]/μg hASA per mg protein. The ASA activity was determined by the p-nitrocatechol sulfate assay and the intracellular mass of the hASA by ELISA. The origin of the cells is indicated.

The two-gene vector MSCV-asa-neo was used in a 100-fold lower multiplicity of infection than MSCV-asa which might result in a less efficient transduction of target cells. To exclude a possible limiting effect of a low transduction efficiency on the hASA expression, the NIH/3T3 cells transduced with MSCV-asa-neo were selected with 0.4 mg/ml G418. Selected cells showed a six-fold overexpression of ASA.

In addition to NIH/3T3 cells, ecotropic as well as amphotropic MSCV-asa was also applied to primary murine cells (BMCs, glial cells), two human glioma cell lines (U118, U373), two murine astrocytoma cell lines (17−/−A1, 11+/+C1) and a hamster cell line (BHK21) (Figure 2b). Transduction did not result in the transfer of ASA activity to BHK21 cells which served as a negative control due to the lack of ecotropic and amphotropic retrovirus receptors. Human cells lack the ecotropic retrovirus receptor and were therefore not transducable with GP+E86-packaged retroviral vector. However, the amphotropic retroviral vector mediated a more than 150-fold overexpression of ASA in U118 and U373 cells. In murine NIH/3T3, 11+/+C1 and 17−/−A1 cells the level of hASA expression was independent of the tropism of the retroviral vector. Finally, using amphotropic producer cells, co-culture transduction of pre-stimulated murine BMCs and supernatant transduction of mixed cultures of murine glial cells yielded a 30- and 3.5-fold ASA overexpression, respectively. The absolute expression levels of hASA ranged between approximately 100 (11+/+C1 cells and primary BMCs) and 500 mU ASA per milligram cellular protein (U373 cells). The intracellular levels of the lysosomal enzymes β-hexosaminidase, α-galactosidase, β-galactosidase and β-glucuronidase were in the normal range in hASA-overexpressing cells (data not shown).

We then determined the specific activity of the retrovirally expressed hASA in different target cells. The ratio between the intracellular activity and mass of hASA was 44 and 40 mU/μg on average in the two human and the five murine cell lines, respectively (Figure 2c).

Selection of an efficient producer cell line

Since the one-gene vector was superior to the two-gene vector in mediating high ASA expression in vitro, MSCV-asa-neo was excluded from subsequent experiments. When compared in initial stem cell gene therapy experiments performed in parallel, the amphotropic, but not the ecotropic producer clone for MSCV-asa mediated efficient transgene expression in vivo.9 Therefore, all further experiments were confined to the amphotropic producer cell line GP+AM12: MSCV-asa.

Targeting of retrovirally encoded hASA

To examine the intracellular localization of the retrovirally encoded hASA, NIH/3T3 and 17−/−A1 cells were transduced with amphotropic MSCV-asa and subsequently analyzed by indirect immunofluorescence. In both cell types specific staining was detectable in a perinuclear granular distribution which is typical for a lysosomal localization (shown for NIH/3T3 cells in Figure 3). Targeting of the retrovirally encoded transgene product was also analyzed by quantifying the amount of secreted hASA by ELISA. When confluent 17−/−A1 cells were cultured for 96 h without a change of medium, 30% and 70% of hASA was detectable extra- and intracellularly, respectively (data not shown).

Figure 3

Immunofluorescent staining of hASA. (a) Non-transduced NIH/3T3 cells, (b) NIH/3T3 cells transduced with amphotropic MSCV-asa. Cells were permeabilized, fixed and incubated with an hASA-specific antibody. Transduced cells show a granular perinuclear staining which indicates lysosomal localization of the retrovirally encoded hASA.

Size and intracellular stability of retrovirally encoded hASA

To assay the stability of retrovirally encoded hASA, the ability of cells to internalize ASA from the medium via mannose 6-phosphate receptor-mediated endocytosis was exploited.10 35S-labelled hASA included in secretions of metabolically labelled GP+AM12: MSCV-asa producer cells was endocytosed by human fibroblasts within a 24 h pulse and was detectable by immunoprecipitation up to 10 days later (Figure 4a). Logarithmic extrapolation of signal intensities suggested that half of the internalized and immunoprecipitable hASA was lost within approximately 2 weeks and the same result was obtained following pulse-feeding of murine NIH/3T3 cells (data not shown). In both cell types, the immunoprecipitated hASA had a size of approximately 62 kDa. For the murine NIH/3T3 cells the time-dependent decline of intracellular hASA levels could be further determined by ELISA. Using this assay, only 39% of the initial hASA level was detectable at day 10 after feeding indicating a structural half-life of approximately 1 week (data not shown).

Figure 4

Intracellular stability and size of hASA in primary human fibroblasts. (a) MSCV-asa-encoded hASA, (b) hASA expressed from BHK21 cells stably transfected with the eukaryotic expression plasmid pBEH-HT14/CP8. Cells were pulse-fed with 35S-labelled hASA and chased for different times as indicated.

hASA encoded by the eukaryotic expression plasmid pBEH-HT14/CP8 was used as a control and metabolically labelled secretions of stably transfected BHK21 cells were applied to human primary fibroblasts and murine NIH/3T3 cells.11 Immunoprecipitation revealed that the size of the hASA polypeptide as well as the kinetics of hASA uptake and degradation resembled that of retrovirally encoded hASA (shown for human fibroblasts in Figure 4b). Further, as with retrovirally encoded hASA, ELISA of the NIH/3T3 cells incubated with plasmid-encoded hASA indicated that only 44% of immunoreactive hASA was still present after a chase of 10 days (data not shown). Thus, retrovirally encoded hASA has a normal half-life in murine cells.

Correction of the metabolic defect of ASA-deficient mouse cells

Normal cultured cells internalize exogenously applied sulfatide and degrade this lipid to galactocerebroside and ceramide by successive hydrolyses catalyzed by ASA and galactocerebrosidase. In a sulfatide loading assay, cells are incubated with sulfatide which is fluorescently labelled in the fatty acid moiety so that accumulating degradation products can be visualized by UV illumination of lipid extracts separated by thin layer chromatography. If ASA-expressing cells, like cells of the murine astrocytoma cell line 11+/+C1, are submitted to such a sulfatide loading assay, ceramide accumulates at a rate directly dependent on the ASA activity, while for ASA-deficient cells such as the astrocytoma cells 17−/−A1, no sulfatide breakdown product is detectable (Figure 5a,b). However, following transduction of 17−/−A1 cells with MSCV-asa the ceramide production was restored to a level identical to that of 11+/+C1 control cells, indicating a complete correction of the defective sulfatide catabolism (Figure 5a).

Figure 5

Hydrolysis of fluorescently labelled sulfatide by wild-type control cells (11+/+C1) and ASA-deficient cells (17−/− A1) (sulfatide loading assays). Galactocerebroside is hardly visible since it is immediately hydrolysed yielding ceramide. The intensity of the ceramide signal is a measure of ASA activity. As a control, sulfatide used for cell loading was also chromatographed. (a) ASA activity of ASA-deficient 17−/−A1 cells following transduction with MSCV-asa compared with control cells. (b) ASA activity of 17−/−A1 cells after pre-incubation with recombinant hASA polypeptide as indicated. A faint ceramide signal for 17−/−A1 cells fed with 5 mU/ml hASA for 48 h (lane 4) was lost during photographic reproduction.

Correction of the metabolic defect was also achieved when non-transduced 17−/−A1 cells were pre-incubated with recombinant hASA polypeptide that had been purified from medium conditioned by BHK21 cells overexpressing the hASA from the expression plasmid pBEH-HT14/CP8 (Figure 5b).11 Using a 48 h feeding period a partial correction was detectable after addition of 5 (not visible in Figure 5b) or 40 mU hASA per milliliter of medium, while a complete correction was obtained following pre-incubation with 300 mU/ml. More precise analyses indicated that sulfatide degradation cannot be enhanced by an increase of enzyme levels above 100 mU/ml (data not shown) and, above this saturating level, by extension of feeding times from 48 up to 144 h (Figure 5b).


In this study we have established a functional retroviral vector and a corresponding producer cell line intended for somatic gene therapy of ASA-deficient mice. Producer cells for two different MSCV-based retroviral vectors were generated and tested. The two vectors, designated MSCV-asa-neo and MSCV-asa, both harbor the full length coding region of the hASA cDNA driven by the retroviral long terminal repeat, but differ by the presence or absence of the neomycin phosphotransferase gene controlled by an internal phosphoglycerokinase promoter (Figure 1).

When applied to NIH/3T3 target cells, the one-gene vector MSCV-asa was clearly superior to the two-gene vector MSCV-asa-neo and mediated a five-fold higher ASA expression (Figure 2a). The discrepancy between the two vectors might be explained by competitive interference of the retroviral promoter/enhancer element and the phosphoglycerokinase promoter in MSCV-asa-neo.12 Promoter interference has been documented for a number of internal promoter vectors and can result in an almost mutually exclusive expression pattern of the two retrovirally encoded genes in transduced target cell clones.13

In human as well as in an extended set of murine target cells, MSCV-asa-mediated expression of ASA activity varied in a cell type-specific manner between approximately 100 and 500 mU per milligram cellular protein (Figure 2b). Previously described retroviral vectors harboring the hASA cDNA transferred 10- to 52-fold overexpression of ASA to human primary fibroblasts.1415 Human fibroblasts were not transduced in this experiment and a direct comparison of the retroviral vectors is therefore not possible. However, a high activity of MSCV-asa in human target cells is clearly indicated by the mediation of 150- and 180-fold overexpression of hASA in two glioma cell lines of human origin (Figure 2b).

The comparison of an amphotropic (GP+AM12-based) and an ecotropic producer cell clone (GP+E86-based) for MSCV-asa revealed equal performance in cultured mouse cells (Figure 2b), however, in a parallel study the GP+AM12-derived producer cell clone transferred significantly higher transgene expression levels in vivo.9 As a result of these transduction experiments the retroviral vector MSCV-asa in the combination with the corresponding amphotropic GP+AM12 producer cell line was chosen as the most promising candidate for gene transfer experiments and further analyzed.

In a number of experiments the specific activity (Figure 2c), sorting (Figure 3), stability (Figure 4a) and size (Figure 4a) of the hASA polypeptide encoded by GP+AM12: MSCV-asa was determined. The results of these analyses are identical for human and murine target cells and in agreement with data from recombinant control hASA (Figure 4b and results) and data published for gene-encoded hASA.116 This excludes the possibility that the retroviral hASA cDNA had acquired mutations in the process of retroviral vector and producer cell construction which might interfere with a normal expression and processing of the polypeptide. Such mutations frequently occur within the retroviral genome during infection and, since the producer cell line was obtained by transduction of a single GP+AM12 packaging cell, would affect all retroviral vector particles released from this clone.17

The correct maturation, targeting and function of the hASA polypeptide is dependent on a variety of enzymatic interactions between the sulfatase and accessory polypeptides, such as mannose 6-phosphate receptors and the phosphotransferase complex.10 The correct expression of the human polypeptide in heterologous mouse cells therefore also indicates that the structural requirements for these interactions are conserved between mouse and man. This notion is clearly supported by the capacity of hASA to correct the metabolic defect of ASA-deficient mouse cells following transduction or feeding (Figure 5a, b). The observation that the hASA can functionally substitute for the murine ASA in cultured cells suggests that MSCV-asa might also be suitable for the development of gene therapy approaches in mice using transplantation of retrovirally transduced donor cells.

A deficiency of ASA causes the lysosomal storage disease MLD which predominantly affects the central nervous system. The fact that the blood–brain barrier is impermeable to lysosomal enzymes and allows only limited access of cells to the brain, presents a major obstacle to the success of gene therapy in correcting MLD. In the mouse, only one in 30 brain cells is of donor type 6 to 8 months following bone marrow transplantation and similar frequencies might be achieved by intracerebral grafting.1819 Transmission of lysosomal enzymes from enzyme-expressing platform cells to deficient cells is favored by a well established secretion/recapture pathway.20 However, in most animal models for lysosomal storage diseases, only low intracerebral levels of the therapeutic enzyme were achieved by allogeneic bone marrow transplantation.21 This level differs among the lysosomal storage diseases and might be governed by the varying amounts of individual lysosomal enzymes in the secretions of intracerebral platform cells.22 On the other hand, ASA levels as low as 5% of the normal ASA activity are sufficient to prevent the manifestation of MLD and subtle differences of residual ASA activities below this threshold decide on the severity of the disease.2324 Thus, the therapeutic success or failure of gene transfer might also depend on minor differences in the transgene expression levels. The transcriptional activity of the retroviral vector in genetically modified brain cells might therefore be of crucial importance in gene therapy experiments.

The retroviral vector MSCV-asa might be superior to previously described retroviral vectors constructed for ASA gene transfer for two reasons.14152526 First, most of the described vectors are two-gene vectors bearing a dominant selectable marker gene controlled by an internal promoter. The superiority of one-gene vectors such as MSCV-asa is indicated by this (Figure 2a) and other studies.2728 Second, all previous vectors are derivatives of the Moloney murine leukemia virus (MoMLV) and a direct comparison of MoMLV- and MSCV-driven transgene expression revealed that MSCV overcomes transcriptional silencing which is often observed for MoMLV-based vectors especially in differentiated cells.7 MSCV-asa is thus more likely to be therapeutically effective than the conventional MoMLV-based vectors.

In summary, we have succeeded in efficient transduction of murine cells with the murine stem cell virus-based retroviral vector MSCV-asa encoding the human ASA polypeptide. The vector is highly active in bone marrow cells as well as glial cells both of which cell types might be utilized for gene therapy approaches in ASA-deficient mice, using bone marrow transplantation and intracerebral grafting, respectively. The structural and functional integrity of the human ASA in heterologous mouse cells was verified and its capacity to correct the metabolic defect of ASA-deficient mouse cells was demonstrated. These data and the suggested superiority of MSCV- over MoMLV-based vectors provide a rationale for testing MSCV-asa in gene transfer experiments in ASA-deficient mice.

Materials and methods


The murine fibroblastic cell line NIH/3T3, the human glioma cell lines U118 and U373, the murine astrocytoma cell lines 17−/−A1 and 11+/+C1 (Habetha, unpublished) and the packaging cell lines GP+AM12 and GP+E862930 were maintained in Dulbecco's modified Eagle medium (DMEM) with 4.5 g/l glucose, 3.7 g/l NaHCO3, 10% (vol/vol) heat-inactivated fetal calf serum, 100 U/ml penicillin, 100 mg/ml streptomycin and 2 mM L-glutamine (all media and supplements from Life Technologies, Eggenstein, Germany) in a humidified atmosphere containing 5% CO2/95% air at 37°C. Packaging cells were cultured on gelatine-coated surfaces prepared by pre-incubation of cell culture dishes with 0.1% gelatine (type A; Sigma, Deisenhofen, Germany) in H2O for 10 min. Before transfection or crossinfection packaging cells were selected with 15 μg/ml hypoxanthin, 250 μg/ml xanthin and 25 μg/ml mycophenolic acid (Sigma) as described.2930

Pre-stimulated primary BMC and mixed cultures of primary glial cells (astrocytes and oligodendrocytes) were generated as described.931

Retroviral vectors

The entire coding region of the hASA cDNA was obtained by digestion of the plasmid pBEH-HT14/CP8 with EcoRI.11 Cloning of the ASA cDNA fragment (HT14/CP8) into the EcoRI site of the retroviral vector plasmid pMSCVneoEB yielded pMSCV-asa-neo which harbors the bacterial neomycin phosphotransferase gene driven by an internal phosphoglycerokinase promoter as a dominant selectable marker (Figure 1).8 Excision of the phosphoglycerokinase promoter-neomycin phosphotransferase cassette by digestion with BamHI and XhoI and subsequent ligation of the blunted plasmid ends yielded the retroviral vector plasmid pMSCV-asa. pMSCV-asa-neo and pMSCV-asa were partially sequenced to verify the integrity of the plasmid/insert boundaries. Both plasmids were purified by the CsCl method.32

Producer cell lines

Ecotropic GP+E86 and amphotropic GP+AM12 packaging cells were transfected with pMSCV-asa-neo alone or co-transfected with pMSCV-asa and pSV2Neo using the calcium phosphate co-precipitation method and stably transfected cells were selected with 0.4 mg/ml G418.32 For MSCV-asa-neo, individual producer cell clones were isolated and pre-screened for high retroviral titers by functional titering (see below). For MSCV-asa, conditioned medium from pooled G418-resistant producer clones was used for crossinfection of the packaging cell line with the opposite tropism. For crossinfection, 1 × 103 target cells were seeded on a 100-mm dish and transduction was repeated three times in 6 h intervals using centrifuged medium conditioned for 24 h and supplemented with 8 μg/ml polybrene (hexadimethrine bromide; Aldrich, Steinheim, Germany) according to Cepko.33 One to 2 weeks later, individual producer cell clones were isolated and pre-screened by functional titering (see below).

Producer cell lines were assayed on replication-competent retroviruses by quantification of reverse transcriptase activity in supernatants of transduced NIH/3T3 cells. One week following supernatant transduction (see below) 5 ml medium was harvested from transduced NIH/3T3 cells and analyzed by a non-radioactive reverse transcriptase assay (Boehringer Mannheim, Germany, Cat. No. 1468120). Reverse transcriptase activity was undetectable in all samples analyzed (data not shown).

Pre-screening of producer cell clones by functional titering

Producer cell clones were pre-selected by their capacity to mediate efficient transgene expression by supernatant transduction of NIH/3T3 cells.

Twenty-four hour medium collected from producer cell lines grown to approximately 80% confluency was centrifuged to sediment non-adherent cells. Twenty, 10 and 2 μl, respectively, of the 24 h supernatant was mixed with fresh medium supplemented with 8 μg/ml polybrene to a final volume of 250 μl. Each dilution was applied to 1 × 104 NIH/3T3 cells plated per well of a 24-well plate 12 h before. After 6 h, 1 ml fresh medium was added. Following transduction, the NIH/3T3 cells were grown until confluency, trypsinized, washed in 1 × TBS pH 7.4 and lysed in 100 μl 1 × TBS pH 7.4, 0.5% Triton X-100 by sonication. Triton-insoluble material was sedimented and 20 μl of the supernatant was analyzed for ASA activity (see below). The intracellular ASA activity was normalized on the intracellular activity of the lysosomal enzyme β-hexosaminidase, which was determined by a rapid chromogenic assay using 5 μl homogenate (see below).

Titering by quantitative Southern blot or transfer of geneticin resistance

For MSCV-asa, the titers of cross-infected producer cell lines which transferred ASA overexpression to target cells most efficiently were determined by quantitative Southern blotting. 1 × 105 NIH/3T3 cells were plated per 60-mm cell culture dish. Twelve hours later, 50, 250 and 500 μl of medium aspirated from subconfluent producer cells 24 h after the last change of medium was filled up to 5 ml with fresh medium and supplemented with 8 μg/ml polybrene. Each dilution was applied to one dish with target cells. The number of NIH/3T3 cells per dish at the time point of transduction was counted in dishes not used for transduction and was the same as 12 h before. After 6 h, 5 ml of fresh medium was added per dish. The transduced NIH/3T3 cells were passaged to 100-mm cell culture dishes, grown until confluency and genomic DNA was isolated by the proteinase K method.32

For standardization 15 μg genomic DNA from non-transduced NIH/3T3 cells was mixed with 0, 1.7, 3.5, 8.6 and 17.1 pg pMSCV-asa plasmid DNA, respectively. These amounts yield copy numbers of MSCV-asa per cell of 0, 0.1, 0.2, 0.5 and 1.0. The standard DNAs as well as genomic DNAs obtained from transduced NIH/3T3 cells were digested with KpnI which has one restriction site in each long terminal repeat of MSCV-asa (Figure 1). The digested DNAs were electrophoresed, blotted on to Hybond N filter and the filter was hybridized to a 32P-labelled SmaI fragment of HT14/CP8 (Figure 1).

For MSCV-asa one amphotropic and one ecotropic clone which exported recombinant retroviral vector at a titer of 5 × 105 and 1 × 106 infectious particles per milliliter, respectively, were used for transduction experiments.

For MSCV-asa-neo, G418-resistant producer cell clones obtained by transfection of GP+E86 packaging cells were pre-screened by functional titering and the titers of a subset of producer clones was then evaluated by G418-resistance of supernatant transduced NIH/3T3 cells.33 A clone with a titer of 1 × 104 infectious particles per milliliter was chosen.

Transduction of target cells

Cell lines and primary glial cells were transduced by supernatant transduction. Medium was aspirated from subconfluent producer cells 24 h after the medium was changed. Following centrifugation the supernatant was supplemented with 8 μg/ml polybrene and 10 ml were applied to 5 × 104 target cells seeded on a 100-mm dish. Transduction was repeated three times in 12 h intervals. Under these conditions the multiplicity of infection was 600 (GP+E86: MSCV-asa), 300 (GP+AM12: MSCV-asa) and 6 (GP+E86: MSCV-asa-neo).

BMCs were transduced by co-cultivation with producer cells as described.9

Determination of the activity, mass, intracellular localization and half-life of hASA

ASA and β-hexosaminidase activities of cell homogenates were determined with the artificial substrate p-nitrocatechol sulfate and p-nitrophenyl-2-acetamido-2-desoxy-β-D-glucopyranosid (Sigma), respectively.3435 Human ASA was further quantified by an indirect sandwich ELISA with absolute specificity for the human polypeptide (Matzner, in preparation).

Indirect immunofluorescence was performed with the murine monoclonal antibody α-hASA IgG 19c2.36

The intracellular half-life of endocytosed recombinant hASA was determined according to Waheed et al.16 Briefly, target cells were incubated with metabolically labelled secretions of cells which had been transfected or transduced with vectors encoding hASA. The 35S-methionine-containing labelling medium was supplemented with ammonium chloride to induce hypersecretion of lysosomal enzymes. Secreted proteins were precipitated with ammonium sulfate, dissolved in DMEM and dialysed against DMEM. Target cells seeded in 35-mm dishes were incubated with 1 × 106 c.p.m. 35S-labelled proteins for 24 h and subsequently cultured in the absence of radioactivity. Cells harvested at different chase times were lysed and hASA was immunoprecipitated with a polyclonal rabbit α-hASA antiserum and/or quantified by indirect ELISA. Immunoprecipitates were submitted to SDS-PAGE and 35S-labelled hASA was detected by fluorography.

Sulfatide loading assays were done according to Monti et al.37 Fluorescently labelled sulfatide (N-lissamin rhodaminyl-(12-aminododecanyl) cerebroside 3-sulfate) was kindly provided by S Marchesini, Department of Biomedical Sciences and Biotechnology, University of Brescia, Italy.

Recombinant hASA was purified by immunoaffinity chromatography with the murine monoclonal antibody A2 as described.38


  1. 1

    Kolodny EH, Fluharty AL . Metachromatic leukodystrophy and multiple sulfatase deficiency: sulfatide lipidosis. In: Scriver CR, Beaudet AL, Sly WS, Valle D (eds) The Metabolic and Molecular Bases of Inherited Disease McGraw-Hill: New York 1995 2693–2740

    Google Scholar 

  2. 2

    Hess B et al. Phenotype of arylsulfatase A-deficient mice: relationship to human metachromatic leukodystrophy Proc Natl Acad Sci USA 1996 93: 14821–14826

    CAS  Article  Google Scholar 

  3. 3

    Taylor RM, Wolfe JH . Decreased lysosomal storage in the adult MPS VII mouse brain in the vicinity of grafts of retroviral vector-corrected fibroblasts secreting high levels of beta-glucuronidase Nature Med 1997 3: 771–774

    CAS  Article  Google Scholar 

  4. 4

    Daly 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: 2296–2300

    CAS  Article  Google Scholar 

  5. 5

    Stein CS, Ghodsi A, Derksen T, Davidson BL . Systemic and central nervous system correction of lysosomal storage in mucopolysaccharidosis type VII mice J Virol 1999 73: 3424–3429

    CAS  PubMed  PubMed Central  Google Scholar 

  6. 6

    Ghodsi A et al. Extensive beta-glucuronidase activity in murine central nervous system after adenovirus-mediated gene transfer to brain Hum Gene Ther 1998 9: 2331–2340

    CAS  Article  Google Scholar 

  7. 7

    Cheng L et al. Sustained gene expression in retrovirally transduced, engrafting human hematopoietic stem cells and their lympho-myeloid progeny Blood 1998 92: 83–92

    CAS  Google Scholar 

  8. 8

    Hawley RG, Lieu FHL, Fong AZC, Hawley TS . Versatile retroviral vectors for potential use in gene therapy Gene Therapy 1994 1: 136–138

    CAS  PubMed  PubMed Central  Google Scholar 

  9. 9

    Matzner U et al. Long term expression and transfer of arylsulfatase A into brain in arylsulfatase A-deficient mice transplanted with bone marrow expressing the arylsulfatase A cDNA from a retroviral vector (submitted for publication)

  10. 10

    Kornfeld S . Structure and function of the mannose 6-phosphate/insulinlike growth factor II receptors Annu Rev Biochem 1992 61: 307–330

    CAS  Article  Google Scholar 

  11. 11

    Stein C et al. Cloning and expression of human arylsulfatase A J Biol Chem 1989 264: 1252–1259

    CAS  PubMed  Google Scholar 

  12. 12

    Emmerman M, Temin HM . Genes with promoters in retrovirus vectors can be independently suppressed by an epigenetic mechanism Cell 1986 95: 459–467

    Google Scholar 

  13. 13

    Sanes JR et al. Lineage, migration, and phenotype in avian optic tectum: analysis with recombinant retroviral vectors. In Gage FH, Christen Y (eds) Gene Transfer and Therapy in the Nervous System Springer: Berlin 1992 pp 59–75

    Google Scholar 

  14. 14

    Rommerskirch W et al. Restoration of arylsulphatase A activity in human-metachromatic-leucodystrophy fibroblasts via retroviral-vector-mediated gene transfer Biochem J 1991 280: 459–461

    CAS  Article  Google Scholar 

  15. 15

    Ohashi T, Matalon R, Barranger JA, Eto Y . Overexpression of arylsulfatase A gene in fibroblasts from metachromatic leukodystrophy patients does not induce a new phenotype Gene Therapy 1995 2: 363–368

    CAS  PubMed  Google Scholar 

  16. 16

    Waheed A, Hasilik A, von Figura K . Enhanced breakdown of arylsulfatase A in multiple sulfatase deficiency J Biochem 1982 123: 317–321

    CAS  Google Scholar 

  17. 17

    Varmus H . Retroviruses Science 1988 240: 1427–1435

    CAS  Article  Google Scholar 

  18. 18

    Krall WJ et al. Cells expressing human glucocerebrosidase from a retroviral vector repopulate macrophages and central nervous system microglia after murine bone marrow transplantation Blood 1994 83: 2737–2748

    CAS  PubMed  PubMed Central  Google Scholar 

  19. 19

    Kennedy DW, Abkowitz JL . Kinetics of central nervous system microglial and macrophage engraftment: analysis using a transgenic bone marrow transplantation model Blood 1997 90: 986–993

    CAS  PubMed  Google Scholar 

  20. 20

    Neufeld EF . Lysosomal storage diseases Annu Rev Biochem 1991 60: 257–280

    CAS  Article  Google Scholar 

  21. 21

    Hoogerbrugge PM, Valerio D . Bone marrow transplantation and gene therapy for lysosomal storage diseases Bone Marrow Transplant 1998 21: (Suppl.2) 34–36

    Google Scholar 

  22. 22

    Walkley SU, Dobrenis K . Bone marrow transplantation for lysosomal diseases Lancet 1995 345: 1382–1383

    CAS  Article  Google Scholar 

  23. 23

    Penzien JM et al. Compound heterozygosity for metachromatic leukodystrophy and arylsulfatase A pseudodeficiency alleles is not associated with progressive neurological disease Am J Genet 1993 52: 557–564

    CAS  Google Scholar 

  24. 24

    Leinekugel P, Michel S, Conzelmann E, Sandhoff K . Quantitative correlation between the residual activity of beta-hexosaminidase A and arylsulfatase A and the severity of the resulting lysosomal storage disease Hum Genet 1992 88: 513–523

    CAS  Article  Google Scholar 

  25. 25

    Learish R et al. Retroviral gene transfer and sustained expression of human arylsulfatase A Gene Therapy 1996 3: 343–349

    CAS  PubMed  Google Scholar 

  26. 26

    Sangalli A et al. Transduced fibroblasts and metachromatic leukodystrophy lymphocytes transfer arylsulfatase A to myelinating glia and deficient cells in vitro Hum Gene Ther 1998 9: 2111–2119

    CAS  Article  Google Scholar 

  27. 27

    Hock RA, Miller AD, Osborne WR . Expression of human adenosine deaminase from various strong promoters after gene transfer into human hematopoietic cell lines Blood 1989 74: 876–881

    CAS  PubMed  Google Scholar 

  28. 28

    Plavec I et al. High transdominant RevM10 protein levels are required to inhibit HIV-1 replication in cell lines and primary T cells: implication for gene therapy of AIDS Gene Therapy 1997 4: 128–139

    CAS  Article  Google Scholar 

  29. 29

    Markowitz D, Goff S, Bank A . A safe packaging line for gene transfer: separating viral genes on two different plasmids J Virol 1988 62: 1120–1124

    CAS  PubMed  PubMed Central  Google Scholar 

  30. 30

    Markowitz D, Goff S, Bank A . Construction and use of a safe and efficient amphotropic packaging cell line Virology 1988 167: 400–406

    CAS  Article  Google Scholar 

  31. 31

    Aloisi F, Agresti C, Levi GJ . Establishment, characterization, and evolution of cultures enriched in type-2 astrocytes J Neurosci Res 1988 21: 188–198

    CAS  Article  Google Scholar 

  32. 32

    Maniatis T, Fritsch EF, Sambrook J . Molecular Cloning: A Laboratory Manual Cold Spring Harbor Laboratory Press: Plainview, New York 1989

    Google Scholar 

  33. 33

    Preparation of a specific retrovirus producer cell line. In: Ausubel FM et al (eds) Current Protocols in Molecular Biology John Wiley. Chichester 1995 unit 9.10

  34. 34

    Baum H, Dodgsen KS, Spencer B . The assay of aryl sulfatase A and B in human urine Clin Chim Acta 1959 4: 453–455

    CAS  Article  Google Scholar 

  35. 35

    Levvy GA, Conchie J . Mammalian glycosidases and their inhibition by aldonolactones Meth Enzymol 1966 8: 571–584

    CAS  Article  Google Scholar 

  36. 36

    Sommerlade HJ, Hille-Rehfeld A, von Figura K, Gieselmann V . Four monoclonal antibodies inhibit the recognition of arylsulphatase A by the lysosomal enzyme phosphotransferase Biochem J 1994 297: 123–130

    CAS  Article  Google Scholar 

  37. 37

    Monti E et al. Uptake and metabolism of a fluorescent sulfatide analogue in cultured skin fibroblasts Biochim Biophys Acta 1992 1124: 80–87

    CAS  Article  Google Scholar 

  38. 38

    Schierau A et al. Interaction of arylsulfatase A with UDP-N-acetylglucosamine:lysosomal enzyme-N-acetylglucosamine-1-phosphotransferase J Biol Chem 1999 274: 3651–3658

    CAS  Article  Google Scholar 

Download references


We wish to thank C Fischer for technical assistance and L Shaw for his comments on the manuscript. This work was supported by the Bundesministerium für Forschung und Technologie.

Author information



Corresponding author

Correspondence to U Matzner.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Matzner, U., Habetha, M. & Gieselmann, V. Retrovirally expressed human arylsulfatase A corrects the metabolic defect of arylsulfatase A-deficient mouse cells. Gene Ther 7, 805–812 (2000).

Download citation


  • murine stem cell virus
  • gene therapy
  • arylsulfatase A
  • metachromatic leukodystrophy
  • lysosomal storage disease

Further reading


Quick links