Introduction

Mucopolysaccharidoses (MPSs) are a group of lysosomal storage diseases resulting from a deficiency in one of several acid hydrolases (lysosomal enzymes) necessary for the normal catabolism of glycosaminoglycans (GAGs) (reviewed in¹). The MPSs are characterized by progressive accumulation of partially degraded or undegraded GAGs within lysosomes. In general, MPSs lead to bone and joint abnormalities, enlargement of the visceral organs, cardiovascular disease, and neurologic impairment. Current therapies, including hematopoietic cell transplantation² and enzyme replacement³, have been effective only in selected lysosomal storage diseases and do not consistently lead to improvements in all patients, especially those with central nervous system (CNS) manifestations. MPS type VII or Sly disease is caused by a deficiency of the lysosomal enzyme -glucuronidase (GUSB; EC 3.2.1.31)⁴. This enzyme is involved in the degradation of heparan, dermatan, and chondroitin sulfates. The availability of a murine model of this disease⁵ has led to many important insights into the treatment of this and other mucopolysaccharidoses⁶. The MPS VII mouse has a single base pair deletion in the murine -glucuronidase gene (Gus)⁷. This mutation leads to a shift in the reading frame beginning at codon 490 and a termination signal at codon 497 of the normally 648-amino-acid protein. These mice have nearly absent -glucuronidase activity levels⁵. Lysosomal accumulation of GAGs in the mutant mice is progressive and occurs within the circulating and fixed-tissue macrophage system and the parenchyma of several organs, including the liver, spleen, heart, kidneys, skin, cartilage, and brain^5,8. MPS VII mice have a shortened life span, with 50% mortality by 6 months of age^5,9.

-Glucuronidase and the majority of lysosomal enzymes are transported to their appropriate intracellular locations via mannose 6-phosphate (M6P)-specific receptors^10,11,12. This transport is not complete and small proportions of the enzymes continue through the secretory pathway to the exterior of the cell. At the cell surface, the secreted enzyme can be internalized and retargeted to the lysosomal compartment through a mannose 6-phosphate receptor (MPR)-mediated process. The application of gene transfer to the treatment of MPS VII relies, in part, upon secretion and M6P/MPR-dependent uptake of lysosomal enzymes; cells targeted for gene transfer will secrete the deficient lysosomal enzyme, which will be taken up by non-genetically modified cells. With the goal of global disease correction, peripheral in vivo administration of recombinant retrovirus (rRV)^13,14,15, adenovirus (rAd)^{16,17,18,19,20}, and adeno-associated virus (rAAV)^{9,21,22,23,24} vectors has been studied in MPS VII mice. In neonatal animals, significant reductions of lysosomal storage within most organs, including the brain, is observed following systemic or liver-directed gene transfer by rAAV^9,22,24 and rRV^13,15. In addition, gene transfer during the neonatal period leads to improvements in the clinical disease course, including an increase in longevity^9,24. In adult MPS VII mice, the peripheral administration of rRV^13,14, rAd^{16,17,18,19,20}, and rAAV²¹ vectors also leads to reductions in lysosomal storage. However, the histologic changes reported in these studies were limited to peripheral organs and improvements in the course of the disease were not observed.

We hypothesized that failure to achieve a more pronounced therapeutic effect in the adult mice might be due to low levels²¹ and short durations of transgene expression¹⁷. In this work, we investigated the effect of peripheral administration of an rAAV type 2 vector in adult MPS VII mice. Our goal was to achieve long-term, supraphysiologic levels of -glucuronidase expression within the liver. We found rAAV2 vectors to be capable of transducing and maintaining long-term expression of murine -glucuronidase within the diseased liver of adult MPS VII mice. The level of gene transfer was sufficient to ameliorate disease manifestations within several organs, including the brain. This is the first demonstration of a gene therapy strategy employing peripheral in vivo administration of a vector in adult MPS VII mice resulting in correction of CNS disease.

Results

rAAV Vector Construction and Production

We constructed an rAAV type 2 vector carrying the murine homolog of the human GUSB cDNA (designated as Gus) under the transcriptional direction of the promoter region of the human elongation factor-1 (EF-1) gene. In vitro studies demonstrated that this vector, rAAV-Gus.2.1, was capable of mediating -glucuronidase expression within hepatocytes (data not shown). For in vivo studies, we used two vector preparations purified by affinity chromatography²⁵. Titers of these preparations were 1.8 10¹² DNase-resistant particles (DRP; or 7.0 10⁹ infectious units, IU) and 1.5 10¹³ DRP (2.8 10¹¹ IU). We used single-vector preparations to complete each of two complementary studies. These studies differed in the method used to determine the timing of analysis of the mice. For the first study, groups of mice were sacrificed and analyzed at scheduled intervals (study 1: scheduled analysis). For the second, in which the higher titer vector preparation was used, mice were sacrificed upon the development of end-point criteria indicating significant illness (study 2: survival analysis).

Scheduled Analysis (Study 1)

rAAV-mediated transduction of the liver in adult MPS VII mice

Adult MPS VII mice (n = 9, age 7–8 weeks, body weight 20.7 3.5 g) received 3.6 10¹¹ DRP in a volume of 200 l by direct intrahepatic injection. To control for vector effects, we injected mice (n = 3) with an rAAV carrying an Escherichia coli -galactosidase expression cassette (rAAV--gal). At the time of vector administration, mice were assigned to be analyzed at 4 (rAAV--gal, n = 3; rAAV-Gus.2.1, n = 3), 12 (rAAV-Gus.2.1, n = 3), or 24 (rAAV-Gus.2.1, n = 3) weeks postinjection. One mouse scheduled to be sacrificed at 12 weeks postinjection died unexpectedly (unrelated to the surgical procedure) and was not available for analysis.

Transgene-derived -glucuronidase activity was present within the livers of the rAAV-Gus.2.1-injected mice at each time point (Fig. 1A; mice designated as L1–L8). Overall, enzyme activity within the injected lobe was 232 109% (mean standard deviation) of wild-type levels and was greater than that present in the livers of rAAV--gal-injected and uninjected MPS VII control mice (P = 0.003; Student's t test, rAAV-Gus.2.1-injected group versus all control mice; enzyme activity levels found in the tissues of wild-type, normal mice are presented under Materials and Methods). We observed the highest enzyme levels at the longest time postinjection evaluated (24 weeks). To evaluate the distribution of enzyme within the liver, we measured -glucuronidase activity in three additional (uninjected) liver lobes (medial, right lateral, and caudate lobes) from the mice sacrificed 24 weeks postinjection (mice L6–L8). Considering all four lobes, hepatic -glucuronidase activity within mice L6, L7, and L8 was 314 26, 304 24, and 282 57% of wild-type levels, respectively. There was no difference in the level of enzyme activity among the injected and uninjected lobes (P = 0.12; Kruskal–Wallis test). There was no more than a 1.8-fold difference between the lobe with the highest activity and that with the lowest in any of the mice. We evaluated transgene expression, also, by histochemical staining for -glucuronidase activity (Figs. 2A–2C). Consistent with the results of the fluorometric assay, -glucuronidase activity was present within the livers of the rAAV-Gus.2.1-injected mice and absent in the control mice. The majority of enzyme-positive cells had the characteristic morphology of hepatocytes (i.e., large polygonal cells with granular cytoplasm and a central nucleus). Considering all eight mice, -glucuronidase was present within 35 11% of hepatocytes within the injected lobe (Fig. 1B). The percentage of enzyme-positive cells correlated directly with the quantitative level of enzyme activity (r_s = 0.90; Spearman's rank correlation).

Figure 1.

Quantitative analyses of -glucuronidase expression and vector genomes after rAAV-Gus.2.1 administration: scheduled analysis (study 1). Groups of mice (each mouse receiving 3.6 10¹¹ DRP of rAAV-Gus.2.1) were sacrificed at 4, 12, and 24 weeks post-vector injection. Data points represent results from individual mice (L1–L8). (A) Hepatic -glucuronidase activity was determined by a fluorometric assay using equivalent amounts of protein from the injected lobe of each liver. -Glucuronidase activity was calculated as the percentage of activity found in the liver of wild-type mice. Enzyme activity in control mice (uninjected, n = 4; rAAV--gal-injected, n = 3) was less than 0.1% of wild-type levels. (B) The percentage of -glucuronidase-positive hepatocytes was determined in cryosections from each liver following histochemical staining for -glucuronidase activity. Enzyme-positive cells were not observed in control tissues. (C) Hepatic vector genome levels were determined by a real-time PCR assay. Equivalent amounts of genomic DNA (10 ng; approximately 1667 cells based upon 6 pg of total genomic DNA per murine cell) were assayed from the liver of each mouse. DNA from control animals consistently demonstrated no detectable amplification product. -Glucuronidase activity in the (D) spleen, (E) kidney, and (F) brain of MPS VII mice injected with rAAV-Gus.2.1 is shown. For each organ, equivalent amounts of protein were assayed. Enzyme activity is represented as the percentage of wild-type levels.

Full figure and legend (188K)

Figure 2.

Histologic evaluation of rAAV-Gus.2.1-mediated -glucuronidase activity. Tissue sections were stained for -glucuronidase activity and counterstained with hematoxylin. Enzyme activity is indicated by the red pigmentation. Representative images of (A) wild-type mouse liver, (B) uninjected MPS VII mouse liver, (C) rAAV-Gus.2.1-injected MPS VII mouse liver evaluated 24 weeks post-vector administration, (D) spleen from an rAAV-injected MPS VII mouse evaluated 12 weeks post-vector administration, (E) kidney from an rAAV-injected MPS VII mouse evaluated 24 weeks post-vector administration, and (F) brain from an rAAV-injected MPS VII mouse evaluated 24 weeks post-vector administration are shown. Enzyme-positive cells were not present in tissues from uninjected MPS VII mice. Scale bars, 100 (A–C) and 25 m (D–F).

Full figure and legend (569K)

Vector genomes within the liver were measured using a quantitative, real-time, fluorescence-based polymerase chain reaction (PCR) assay. The conditions of this assay were designed to detect specifically vector sequences in the presence of the endogenous murine -glucuronidase gene²⁶. Total genomic DNA from uninjected MPS VII mice did not result in an amplification product (i.e., no vector genome copies). Vector genomes were present in the rAAV-Gus.2.1-injected mice livers at each time point (Fig. 1C). Within the injected lobe, the levels were between 0.5 and 4.1 vector genomes per cell genome equivalent analyzed (based upon 6 pg of total genomic DNA per murine cell or 1667 cells per 10 ng of DNA). After the 4-week time point, the number of vector genomes remained stable at 1.0 0.3 genomes per cell.

Correction of lysosomal storage within the liver

To determine whether the level of transgene-mediated enzyme activity was sufficient to reverse lysosomal storage within the liver at the time of injection and maintain correction over the period of observation, we compared specimens from the rAAV-injected mice to those from age-matched control mice (Figs. 3A–3D). Lysosomal storage was qualitatively evaluated by bright-field and electron microscopy. At 12 weeks of age (equivalent to 4 weeks postinjection), control mice had widespread lysosomal storage within the liver as manifested by cytoplasmic vacuolation on bright-field microscopy. An extensive degree of storage was found within cells that line the sinusoidal spaces (e.g., Kupffer cells). Lower levels were present within the hepatocytes. The degree and extent of lysosomal storage in all cell types increased over time and by 32 weeks of age (24 weeks postinjection) cytoplasmic vacuolation was readily observed within hepatocytes (Figs. 3A and 3C). In contrast, the livers from the rAAV-Gus.2.1-injected mice (n = 8) had low levels of lysosomal storage at each time point evaluated. At 24 weeks postinjection, storage was nearly absent within the hepatocytes and significantly reduced within the cells lining the sinusoidal space (Figs. 3B and 3D).

Figure 3.

Evaluation of lysosomal storage following rAAV-Gus.2.1 administration. Tissue sections were imaged by bright-field and electron microscopy. Representative micrographs of liver, spleen, kidney, and brain from aged-matched control (A, C, E, G, I, K, M, O, Q, S) and rAAV-Gus.2.1-injected (B, D, F, H, J, L, N, P, R, T) MPS VII mice are shown. The liver, spleen, and kidney images are from study 1 mice (24 weeks post-vector administration). The brain images are from study 2 mice (rAAV-Gus.2.1-injected >250 days post-vector administration; control age-matched to 200 days postinjection). The degree of lysosomal storage, manifested by cytoplasmic vacuolation, is reduced in tissues from rAAV-injected mice. (A–D) Histopathology of the liver. The dashed lines (A, B) encircle characteristic hepatocytes and arrowheads (A–D) indicate sinusoidal spaces. (E–H) Histopathology of the spleen. (I–L) Histopathology of the kidney. Bright-field images (I, J) are of renal tubules. Arrow (J) indicates residual storage within a cross section of a renal tubule. The electron micrographs (K, L) are images of glomeruli. Asterisks (K, L) indicate podocytes. (M–P) Histopathology of the brain cortex. Neuronal cell bodies are present in each image. Asterisk (N) indicates a blood vessel. (Q–T) Histopathology of the brain striatum. Asterisks (Q, R) indicate blood vessels. The electron micrographs (S, T) are images of glial cells. Scale bars, 25 m.

Full figure and legend (856K)

Effect of intrahepatic rAAV delivery on extrahepatic organs

As widespread disease correction is the therapeutic goal of peripheral gene transfer for the MPSs, we evaluated the spleen, kidney, and brain for transgene-expressed -glucuronidase activity. We examined these organs in this initial study because of the large degree of GAG storage within the spleen and kidney and the significance of the disease process within the brain. Tissue sections stained for -glucuronidase activity from each of these organs contained rare, enzyme-positive cells (Figs. 2D–2F). -Glucuronidase-positive cells were not observed in tissues from control MPS VII mice even with prolonged exposure (approximately 16 h) to the enzyme substrate. Quantitative measurements of -glucuronidase activity within these tissues confirmed the presence of increased levels of enzyme within the spleen and kidney (Figs. 1D–1F) of several mice. Overall, -glucuronidase activity was 2.6 1.1, 1.1 1.1, and 0.0% of wild-type activity in the spleen, kidney, and brain, respectively. -Glucuronidase activity within these tissues might have derived from direct transduction of the organs or uptake of enzyme secreted from transduced hepatocytes. We examined these possibilities by evaluating the presence of vector genomes within extrahepatic organs and -glucuronidase activity within the serum in subsequent experiments (described later in this article). We performed histologic evaluation (bright-field and electron microscopy) for lysosomal storage abnormalities on the spleens and kidneys from the three mice sacrificed 24 weeks postinjection. Reductions in the degree of cytoplasmic vacuolation compared to tissues from age-matched control mice were present (Figs. 3E–3L). In each spleen, we observed low levels of storage, primarily requiring electron microscopy. In each kidney, glomerular and cortical tubular cells had a reduced degree of storage, while the medullary tubular cells continued to have significant histopathologic abnormalities.

Survival Analysis (Study 2)

rAAV-mediated transduction of the liver in adult MPS VII mice

Having shown that peripheral administration of rAAV-Gus.2.1 via intrahepatic injection resulted in gene transfer levels sufficient to correct lysosomal storage, we sought to examine the effects of a higher dose of vector and a more prolonged period of observation. As a secondary goal, we designed this study to investigate whether administration of this vector would lead to an increase in survival of the MPS VII mice. Progress made in vector production methods enabled us to deliver approximately eightfold more particles than in the previous experiment. MPS VII mice (n = 13) at 8 weeks of age (age 58.5 3.6 days, body weight 21.8 2.7 g) received 3.0 10¹² DRP of rAAV-Gus.2.1 via intrahepatic injections. All mice were daily and weighed three times per week. Mice were sacrificed if they developed signs of debilitating illness (e.g., loss of greater than 20% of body weight, difficulty ambulating, or abnormal respiratory pattern). Tissues from two mice that died unexpectedly were not available for analysis. Within the injected lobe of the liver, enzyme activity was 95 to 1032% (473 254%) of wild-type levels (Fig. 4; mice designated as L9–L19). A stable level of -glucuronidase activity was maintained for the duration of the study. Vector genome levels were 4 to 336 (60 96) vector copies per cell genome equivalent (Fig. 4). The highest genome levels were present during the first 2 months following vector administration. After that time, lower levels (15 11 copies per cell genome equivalent) were present for greater than 1 year. In comparison to study 1, the livers of these mice had greater levels of -glucuronidase activity (P = 0.03; Student's t test). Though the mean number of vector genomes was greater in study 2, the difference did not reach significance (P = 0.10; Student's t test). To expand upon our initial observation that transgene-derived -glucuronidase activity was equally distributed throughout the lobes of the liver and, importantly, to address whether rAAV2 vectors diffuse away from the site of injection, we measured the levels of -glucuronidase and vector genomes in four hepatic lobes from mice surviving 30, 122, 147, 283, and 375 days postinjection (L10, L13, L14, L17, and L19). There were no differences in either enzyme activity or vector genome levels among the lobes (P = 0.83 for enzyme and P = 0.96 for vector genomes; Kruskal–Wallis test).

Figure 4.

Quantitative analyses of -glucuronidase expression and vector genomes in the liver following rAAV-Gus.2.1 injection: survival analysis (study 2). Mice (each receiving 3.0 10¹² DRP of rAAV-Gus.2.1) were sacrificed according to preestablished end-point criteria based upon the health of the animal. Hepatic -glucuronidase activity was determined by a fluorometric assay using equivalent amounts of protein from each liver lobe. -Glucuronidase activity was calculated as the percentage of activity found in the liver of wild-type mice. Vector genome levels were determined by a real-time PCR assay. Equivalent amounts of genomic DNA (10 ng) were assayed for each liver lobe. -Glucuronidase activity and vector genomes were measured in the injected lobe of each mouse. Symbols represent individual mice (L9–L19).

Full figure and legend (82K)

-Glucuronidase and vector genomes within extrahepatic organs

Given the facts that the second group of mice received a greater amount of vector and had higher levels of hepatic -glucuronidase activity, we hypothesized that this would result in greater levels of transgene-mediated enzyme activity in extrahepatic organs (Fig. 5A). The mean levels of activity within the spleen (4.2 3.1% of wild-type levels) and kidney (3.0 2.5%) were not significantly different from those observed in the initial study. However, greater enzyme levels were found within the brains of mice receiving the higher dose of virus (2.0 1.0% versus 0%; P = 0.0002; Student's t test). In the heart and lung, -glucuronidase levels were 15 17 and 1.0 1.0%, respectively. Histochemical staining revealed -glucuronidase-positive cells scattered throughout the individual organs (data not shown). To evaluate the possible sources contributing to -glucuronidase activity within the extrahepatic organs, we determined enzyme levels within the serum and vector genome levels within the tissues. We measured serum -glucuronidase activity in four mice (L9, L11, L13, and L19) at the time of sacrifice. Activity levels were 8, 172, 789, and 123% of wild-type in these mice surviving 16, 51, 122, and 375 days post-vector administration, respectively. Serum activity correlated directly with liver -glucuronidase levels (r_s = 1.00; Spearman's rank correlation). Vector genomes were detected within each organ over a wide range (Fig. 5B), with levels trending downward over time. Control tissues, assayed in parallel, consistently demonstrated the absence of an amplification product. Based upon these results, we cannot differentiate between the uptake of circulating enzyme or direct transduction of cells within the individual organs as the source of -glucuronidase within the extrahepatic organs.

Figure 5.

Extrahepatic -glucuronidase activity and vector genomes: survival analysis (study 2). The spleen, kidney, heart, lung, and brain of each mouse (n = 11) receiving rAAV-Gus.2.1 were evaluated for enzyme activity and vector genomes. Symbols represent results obtained from individual mice (L9–L19). (A) -Glucuronidase activity within extrahepatic organs was determined by a fluorometric assay using equivalent amounts of protein for each organ. Enzyme activity is represented as the percentage of wild-type levels. Note that the scale of the y axis is different for each organ. (B) Vector genome levels were determined by a real-time PCR assay. Equivalent amounts of genomic DNA (10 ng; approximately 1667 cells based upon 6 pg of total genomic DNA per murine cell) were assayed from each organ. The symbol below the x axis for the brain indicates that the number of vector genomes was below the limit of detection of the assay. DNA from control animals consistently demonstrated no detectable amplification product.

Full figure and legend (159K)

Correction of disease pathology

In regard to lysosomal storage, this study enabled us to (i) expand upon the observation that correction of hepatic and extrahepatic storage in adult MPS VII mice can be mediated by rAAV vectors, (ii) evaluate whether histopathologic disease correction persists in long-term surviving mice, and (iii) correlate improvement in pathology with the level of enzyme activity. We evaluated the livers (injected lobe) and extrahepatic organs of the 11 mice for lysosomal storage and compared them to organs from uninjected control mice. Similar to what we observed in the first study (scheduled analysis), lysosomal storage within hepatocytes was eliminated by 30 days post-vector administration and remained so in the longest surviving mice. Within Kupffer cells, the degree of cytoplasmic vacuolation followed a biphasic pattern with respect to time postinjection. At the earliest time analyzed, there was extensive vacuolation. The level was reduced to less than control levels by 30 days postinjection. Afterward, cytoplasmic vacuolation within these cells progressively increased. However, this abnormality remained less than that in uninjected MPS VII mice.

We evaluated the spleen, kidney, heart, lung, and brain, also, for lysosomal storage. Improvements in the histopathologic abnormalities within the spleen and kidney were similar to what we observed in study 1. For the spleen, we observed complete resolution of storage on bright-field microscopy in the mice surviving 50 days or more postinjection. For the kidney, storage was improved in all mice surviving longer than 50 days postinjection, though low levels of storage persisted within the medullary tubular cells. In addition, we found reduced levels of storage within the endocardium of the hearts and bronchial epithelium of the lungs in all mice analyzed after 30 days postinjection (data not shown). The brains of the rAAV-Gus.2.1-injected mice, also, had evidence of reduced levels of lysosomal storage within the regions examined (Figs. 3M–3T). Improvements in storage were present in neurons, glia, and perivascular cells in the cortex, white matter (corpus callosum), and subcortical gray matter (striatum) in mice surviving greater than 50 days after vector administration. For each mouse, the degree of cytoplasmic vacuolation was greatest in perivascular cells and least in neurons (perivascular > glia > neurons). Quantitative assessment of storage within the cortex and striatum demonstrated inverse relationships between the degree of storage and time post-vector administration (cortex, r_s = -0.91; striatum, r_s = -0.90; Spearman's rank correlation; Figs. 6A and 6B). In the three longest surviving mice (L17–L19), cytoplasmic vacuolation was absent by bright-field microscopy in the majority of cortical and striatal neurons and glia. The histologic appearance of these cell types was indistinguishable from those in wild-type, normal mice. Perivascular cells had readily identifiable, but reduced levels of storage.

Figure 6.

Quantitative analyses of disease pathology within the brain: survival analysis (study 2). (A) Evaluation of storage within the cortex of rAAV-Gus.2.1-injected mice. Cortical tissue specimens from rAAV-Gus.2.1-injected mice were evaluated and assigned a relative score for histopathologic abnormalities characteristic of lysosomal storage. The rAAV-injected MPS VII mice, except for the three long-term survivors (L17, L18, and L19), were compared to age-matched (28 days of age) uninjected MPS VII mice. L17, L18, and L19 were compared to uninjected MPS VII mice matched to 200 days postinjection. The pathologic scores (defined under Materials and Methods) ranged from 0 (no different from normal mice; least amount of storage) to 4 (no different from untreated, age-matched MPS VII mice; greatest amount of storage). (B) Evaluation of storage within the striatum of rAAV-Gus.2.1-injected mice. Tissue specimens from the striatum of rAAV-Gus.2.1-injected mice were evaluated as described for the cortex. (C) -Galactosidase activity within the brain. Enzyme activity within the brain was determined by a fluorometric assay using equivalent amounts of protein from each brain specimen. Control (untreated) MPS VII mice were age-matched to the time post-injection of the rAAV-injected mice. Enzyme activity is represented as the percentage of wild-type levels (P = 0.0011 for rAAV-injected compared to control; Wilcoxon rank sum test). (D) -Hexosaminidase activity within the brain. Enzyme activity was determined as described for -galactosidase (P = 0.0003 for rAAV-injected compared to control, Wilcoxon rank sum test). Symbols represent results obtained from individual rAAV-Gus.2.1-injected (A–D) and control (C, D) mice.

Full figure and legend (187K)

We further evaluated manifestations of disease pathology within the brain by measuring -galactosidase and -hexosaminidase activity within the brain of each mouse (Figs. 6C and 6D). These lysosomal enzymes are secondarily elevated in MPS VII mice and normalize following effective therapeutic intervention^15,27,28. In the rAAV-Gus.2.1-injected mice, brain -galactosidase levels were 134 50% (median 109%) of wild-type levels and -hexosaminidase levels 217 128% (median 153%) of wild type. The -galactosidase and -hexosaminidase levels in uninjected MPS VII mice (n = 7) were 232 38 (median 240%) and 517 80% (median 516%) of wild-type levels, respectively. The differences between the injected and the uninjected mice for both enzymes were significant (P = 0.0011 for -galactosidase, P = 0.0003 for -hexosaminidase, Wilcoxon rank sum tests). In the mice sacrificed greater than 100 days post-rAAV administration, the enzyme levels approached those found in normal, wild-type mice.

To correlate tissue -glucuronidase and lysosomal storage levels, we rank-ordered the extrahepatic organs according to enzyme activity from mice surviving more than 100 days post-AAV administration. There were no qualitative histologic differences between the organs with the lowest enzyme levels and those with the highest. Moreover, organs with -glucuronidase activity near or at uninjected MPS VII levels had reduced levels of storage. For example, in the spleens of mice L16 and L18 with enzyme levels of 0.8 and 0.4% of wild-type, respectively, nearly complete histopathologic correction of lysosomal storage was observed. Also, there were reduced levels of storage in brains with absent and low levels of enzyme activity (0.0 and 0.7% of wild-type levels in mice L17 and L18, respectively) at the time of sacrifice.

Survival post-vector administration

To determine whether peripheral administration of the rAAV-Gus.2.1 vector leads to an increase in the survival of MPS VII mice, we compared the injected (n = 13; 11 male) to the uninjected (n = 36; 19 male) mice (Fig. 7). All mice were born in our breeding colony during the same period of time and were cared for in a similar manner. Mice surviving to 60 days of age, the approximate age at which the experimental mice received the rAAV-Gus.2.1 vector (58.5 3.6 days), were included in this analysis. Thus, this study includes all mice born in our facility that reached the predetermined age for vector administration. The rAAV-Gus.2.1-injected mice survived to 213 110 days of age and the control mice to 162 73 days. The median age of survival was 196 (range 71 to 427 days) and 168 days (range 66 to 379 days) for rAAV-Gus.2.1-injected and control mice, respectively. The three long-term rAAV-injected survivors were healthy at the time of sacrifice (345, 386, and 427 days of age). All control mice died or were sacrificed according to the criteria established prior to study initiation (see Materials and Methods). The length of their survival was similar to past reports⁵. The fraction of surviving mice was greater in the injected group at each time point greater than 30 days postinjection; however, a statistically significant difference between the two groups was not observed (P = 0.063; Kaplan–Meier survival estimates).

Figure 7.

Survival of mice following rAAV-Gus.2.1 administration: survival analysis (study 2). The survival of rAAV-Gus.2.1-injected MPS VII and control MPS VII mice was compared by Kaplan–Meier survival estimates; P = 0.0629.

Full figure and legend (66K)

Discussion

As part of the therapeutic strategy used in these studies, it was critical to achieve high levels of gene transfer upon peripheral vector administration in adult MPS VII mice. We chose to target the liver for gene transfer based upon the tropism of rAAV2 vectors for this organ and the potential of hepatocytes to secrete transgene-expressed -glucuronidase. Previous investigations have shown that the efficiency of rAAV2 transduction of the liver is dependent on the injection site, with portal vein being more efficient than tail vein infusion^29,30,31,32. However, adult MPS VII mice have decreased tolerance for surgical procedures requiring prolonged periods of anesthesia^33,34. This is exemplified by the report by Lau and colleagues in which the mortality rate of adult MPS VII mice undergoing intra-abdominal surgery approached 70%³⁵. Thus, we chose to deliver the virus via direct hepatic injection as a means to expose the liver to the greatest amount of vector possible²⁹, while avoiding a more complex surgical procedure required for portal vein injection. This approach proved to be a well tolerated and an effective means of vector delivery. No mouse died as a result of the surgery and high levels of gene transfer were achieved. Based upon the sinusoidal structure of the liver, we predicted that intraparenchymal injection would result in diffusion of the vector away from the site of administration. Indeed, rAAV2-mediated transduction was widespread throughout the liver, with vector genomes and transgene expression equally distributed in all lobes.

We chose to use the promoter region of the human EF-1 gene in our vector based upon its ability to direct transgene expression within the murine liver³⁶. The genetic elements from the EF-1 gene carried by the rAAV-Gus.2.1 vector included exon 1, intron 1, and a portion of exon 2 (11 bp). Our studies demonstrated that the EF-1 promoter–-glucuronidase transgene carried by the rAAV-Gus.2.1 vector was capable of mediating prolonged, high-level transgene expression within the liver of MPS VII mice. In the first study presented in this article, we examined the pattern of transgene-derived -glucuronidase activity by a histochemical staining assay. The majority of cells staining positive for -glucuronidase had the morphologic appearance of hepatocytes. This is consistent with past investigations demonstrating that rAAV2 vectors preferentially transduce hepatocytes following intravascular (via systemic and portal circulation) administration^30,31,37,38. In our study, 20 to 50% of hepatocytes were positive for -glucuronidase activity. This is greater than what would be expected from rAAV2 transduction alone^{31,37,38,39,40,41}. A possible explanation is that a portion of the -glucuronidase-positive hepatocytes acquired secreted enzyme through the mannose 6-phosphate receptor-mediated endocytotic pathway.

-Glucuronidase activity within the livers of mice receiving the two different vector doses exceeded levels normally found in heterozygous normal MPS VII mice and, in the second study, hepatic levels approached or exceeded wild-type levels in all mice. To achieve the higher level of -glucuronidase expression within the liver, the delivery of eightfold more vector particles was required. However, in long-term surviving mice, the increased vector dose accounted for no more than a twofold greater level of enzyme within the liver. This is consistent with the reported nonlinear dose response above 3 10¹¹ rAAV2 vector genomes per mouse⁴¹. The levels of -glucuronidase activity achieved in these two current experiments were sufficient to correct hepatic lysosomal storage. Significantly lower enzyme levels were present within the extrahepatic organs; nonetheless the levels were sufficient to result in widespread improvements in storage. Xu and colleagues¹⁵ observed that specific cell types in MPS VII mice are relatively resistant to correction following neonatal gene transfer. Similar to their findings, the medullary tubular cells were more resistant to correction than other cell types within the kidneys. In contrast to their observations, correction of the Kupffer cell compartment within the liver was incomplete in mice treated during adulthood.

Reductions in the level of lysosomal storage were observed in several extrahepatic organs in which -glucuronidase activity was below 1% of wild-type levels. Moreover, there was no discernable difference between organs with low levels of enzyme activity and those with higher levels. At least two possibilities might account for this observation. First, the lack of correlation between tissue enzyme levels and degree of correction might be due to events that occurred prior to sacrifice; in other words, higher transient enzyme levels were present within these tissues prior to analysis. Second, prolonged exposure to low levels of enzyme might be sufficient to reverse the pathologic abnormalities. The delineation of which possibility is correct will have significant impact on the application of gene therapy to patients with MPSs. If the latter were true, there would be a greater possibility of achieving therapeutic gene transfer levels in humans. In either case, enzyme activity within the brain and other extrahepatic organs might have resulted from direct transduction or uptake of circulating, secreted enzyme. Our data do not differentiate between these two possibilities; the extrahepatic organs contained vector genomes and -glucuronidase was present within the serum. We did not investigate whether transgene mRNA was present within the extrahepatic tissues. In general, given the low enzyme levels within these organs, low levels of transcription can be expected. Thus, failure to detect transgene-derived mRNA would not definitively exclude direct transduction as contributing to the -glucuronidase within the organ. This question should be addressed by the use of a liver-specific promoter and the administration of lower doses of rAAV to reduce extrahepatic vector distribution³⁰.

In the brains of MPS VII mice receiving the higher dose of rAAV-Gus.2.1, we found improvements in histologic and biochemical manifestations of disease. Cytoplasmic vacuolation was reduced in all neural cell types (e.g., neurons, glia, and perivascular cells) within the cortex, corpus callosum, and striatum. The lowest levels of storage were observed in the longest surviving mice. Further evidence of disease correction within the brain was obtained by measuring -galactosidase and -hexosaminidase activity levels. The decreases in these enzymes following rAAV administration paralleled the histopathologic changes, consistent with the hypothesis that elevations of these enzymes are secondary to storage-induced lysosomal abnormalities. However, these biochemical improvements do not necessarily indicate global correction of lysosomal storage; reductions in -galactosidase and -hexosaminidase activity might occur despite the continued presence of significant GAG storage. In addition, our histologic investigation involved a limited area of the brain. Thus, further studies are necessary to evaluate possible regional differences in the degree of correction and whether reduction in storage occurs in neural cell types known to be more resistant to therapy¹⁵. The observation that -glucuronidase activity was present only within the brains of mice receiving the higher dose of vector suggests a dose dependency in the ability of a peripherally administered rAAV to correct the brain. Those mice receiving the higher vector dose had evidence of direct vector transduction of the brain, albeit at a relatively low level. Overall, the brain had the lowest level of vector genomes (approximately 0 to 0.06 vector genome per cell genome equivalent or 0 to 105 copies per 10 ng of genomic DNA analyzed) of the organs studied. The vector copy level was 10³- to 10⁴-fold lower than that found within the liver and near the limit of sensitivity of the assay (5 vector copies per 1667 cell genome equivalents or 10 ng of DNA). A similar level relative to that within the liver was observed by Grimm and colleagues³² following portal vein injection of an rAAV2 vector. Thus, it is possible that if administered in a sufficient amount, rAAV2 vectors cross the MPS VII blood–brain barrier and contribute to disease correction.

Though not addressed directly, our data provide insight into the immune response of the MPS VII mice to transgene-expressed -glucuronidase. The stability of transgene expression and the lack of an overt inflammatory cell infiltrate within the liver following rAAV-Gus.2.1 administration suggest that the MPS VII mice did not mount a significant cytotoxic T lymphocyte response directed toward -glucuronidase. Moreover, the persistence of circulating -glucuronidase in long-term surviving mice indicates that antibodies with the ability to remove -glucuronidase from the serum were not produced at a significant level. In contrast, Ross et al.²⁸ reported the development of "inactivating" antibodies to -glucuronidase in adult MPS VII mice after implantation of microencapsulated cells engineered to secrete -glucuronidase. These antibodies led to the rapid elimination of circulating -glucuronidase. Also, adult MPS VII mice have been shown to develop antibodies to human -glucuronidase following administration of purified enzyme⁴² or retrovirus-¹³ and adenovirus-¹⁷ mediated gene transfer. If an immune response occurred in our studies, it was not of sufficient vigor to prevent the widespread therapeutic effect of intrahepatic administration of the vector. Lack of an immune response might be accounted for by at least three mechanisms. First, rAAV-mediated expression of a foreign protein within the liver can result in the induction of tolerance^43,44,45. Second, adult MPS VII mice have a known defect in immunity⁴⁶ and might not generate a vigorous response to a foreign antigen within the liver. Third, though the MPS VII mouse mutation leads to a 200-fold decrease in mRNA levels⁵, a sufficient amount of partial-length, inactive -glucuronidase might be produced in the MPS VII mice to lead to the recognition of the transgene-expressed -glucuronidase as a "self" protein²¹ under the conditions of these studies (i.e., rAAV-mediated expression within the liver). The possible contribution of any of these mechanisms will require further study.

Study 2 was designed to evaluate whether survival could be improved in adult MPS VII mice following administration of the rAAV-Gus.2.1 vector. Previous studies have demonstrated that gene transfer (as well as other therapeutic interventions) during the neonatal period leads to improved survival^9,24. Although three treated mice survived to approximately 1 year of age and were healthy at the time of sacrifice, we were unable to detect an increase in overall survival in the mice receiving the vector. Inclusion of the rAAV-Gus.2.1-injected MPS VII mice from the first study in the survival analysis resulted in a significant improvement in overall survival (P = 0.003; Kaplan–Meier survival estimates). Thus, the statistical power of study 2 might not be sufficient to detect an effect of gene transfer. Nevertheless, failure to demonstrate an increase in the survival of mice in which widespread histopathologic improvements were observed indicates that an irreversible pathologic process occurred prior to or soon after gene transfer. This will require further studies evaluating the cause of death in MPS VII mice after gene transfer outside of the immediate neonatal period.

In these studies we found that the peripheral administration of an rAAV2 vector capable of expressing -glucuronidase resulted in widespread correction of the lysosomal storage abnormality in adult MPS VII mice. The histopathologic improvements were maintained in each organ evaluated for periods of at least 1 year post-vector injection. Importantly, histologic and biochemical measures of disease correction were observed within the brains of the rAAV-injected mice. Based upon these data, we conclude that the blood–brain barrier of adult MPS VII mice does not prevent at least one type of therapy targeted to the periphery from reversing preexisting lysosomal storage within the CNS. As the majority of patients with MPSs are diagnosed only after the development of a significant degree of lysosomal storage, this finding might have significant impact on the development of novel therapies for this and other lysosomal storage diseases.

Materials and methods

Cell culture and chemical reagents

HeLa cells (ATCC CCL-2) were maintained in Dulbecco's modified Eagle medium containing 10% heat inactivated fetal calf serum and antibiotics (penicillin and streptomycin). HeLa-derived cell lines were maintained in HeLa growth medium supplemented with G418 (500 g/ml, active component). Unless otherwise indicated, molecular biology and cell culture reagents were obtained from Gibco Life Technologies (Gaithersburg, MD, USA) and chemicals were obtained from Sigma Chemical Co. (St. Louis, MO, USA).

rAAV construction and production

A recombinant adeno-associated virus serotype 2 vector was engineered to express murine -glucuronidase. The -glucuronidase cDNA was derived by reverse transcription-polymerase chain reaction amplification of RNA isolated from BALB/c kidney (Gus^a; GenBank Accession No. M19279)^47,48. The final vector genome consists of the AAV type 2 terminal repeat elements, the promoter region of the human EF-1 gene (-202 to +988; the first nucleotide of exon 1 designated as +1⁴⁹), the Gus^a cDNA, and an SV40 polyadenylation signal [subcloned from pCMV (Clontech Laboratories, Inc., Palo Alto, CA, USA)]. The vector is referred to as rAAV-Gus.2.1. To produce this vector, HeLa-derived cell lines were constructed containing the above sequences, the AAV rep and cap genes, and a neomycin resistance selectable marker using previously described methods⁵⁰.

Large-scale, affinity column-purified rAAV vector preparations were produced in Columbus Children's Research Institute Viral Vector Core Laboratory using previously described methods²⁵. The purified virus was dialyzed against a 150 mM NaCl, 1 mM MgCl₂, 20 mM Tris, pH 8.0, solution. Recombinant virus titers were determined using two methods. A physical (particle) titer was determined using a quantitative PCR assay to measure the number of DNase-resistant (i.e., encapsidated) vector genomes²⁵. Final titers are expressed as DNase-resistant particles/ml. An infectious titer was obtained using a HeLa-derived cell line containing copies of the AAV rep and cap genes in a replication-based infectious center assay^26,51. Final titers, expressed as infectious units/ml, were based upon the number of positive-staining foci. The rAAV vector preparations were assayed for the presence of contaminating replication-competent viruses^25,50,52. The rAAV vector preparations used in this study were free of adenovirus and wild-type AAV at the limit of detection of each assay (<1 plaque-forming unit of adenovirus or infectious unit of AAV in 1/100 of the vector preparation).

Experimental animals

Mice were obtained from an MPS VII colony maintained at our facility. The colony was established using breeding pairs of B6.C-H2^bm1/ByBir-Gus^mps/+ (Gus^mps/Gus⁺) mice obtained from The Jackson Laboratory (Bar Harbor, ME, USA). Normal (Gus⁺/Gus⁺ and Gus^mps/Gus⁺) and MPS VII (Gus^mps/Gus^mps) offspring were identified by a PCR-based assay on tail biopsies³³. All experiments involving animals were approved by Columbus Children's Research Institute's (Columbus, OH) Institutional Animal Care and Use Committee.

Liver injections

Mice were anesthetized with Avertin (tribromoethyl alcohol/tertiary amyl alcohol). Under sterile conditions, a midline abdominal incision was made exposing the lateral left lobe of the liver. The rAAV was delivered via a direct intrahepatic injection into the liver parenchyma using a 29-gauge needle attached to an insulin syringe. The vector was slowly administered, under gentle, intermittent pressure, to allow the injectate to flow (within the hepatic sinusoidal spaces or blood vessels) from the site of injection. During this procedure the surface of the liver was visualized through a surgical stereoscope. If subcapsular hemorrhage or extravasation of the injectate was observed, the needle was removed from the liver and an alternate site chosen for injection. Approximately three to five closely spaced injection sites were used to administer the vector in each mouse. Following vector injection, bleeding from the surface of the liver was controlled by gentle pressure with a sterile cotton wick.

Preparation and analysis of tissues

At the time of sacrifice the liver, extrahepatic organs, and serum were harvested. Portions of each organ were frozen in cryoprotective medium (Tissue-Tek OCT Compound, Sakura Finetek U.S.A., Inc., Torrance, CA, USA), frozen in liquid nitrogen, fixed in 2.5% glutaraldehyde, or fixed in 10% formalin. For the liver, each lobe (left lateral, medial (right and left), right lateral, and caudate lobes) was processed in this manner. For each brain, the cerebral hemispheres were separated; a 1-mm coronal section through each hemisphere was obtained 4 mm caudal to the olfactory bulbs. The regions rostral to the coronal section were frozen in liquid nitrogen and the caudal regions embedded in cryoprotective medium. The coronal sections were further dissected to separate the cortex from the striatum and fixed in 2.5% glutaraldehyde in cacodylate buffer (0.2 M, pH 7.4).

-Glucuronidase activity within specimens was determined by one of two methods. To detect activity by histochemical staining, 10-m-thick sections were obtained from the OCT-embedded tissues and incubated in a solution consisting of 0.5 mg/ml naphthol AS-BI -D-glucuronide, 1.6 mg/ml pararosaniline hydrochloride, 0.1% sodium nitrite in a veronal acetate buffer, pH 4.8, for 3 h at 37°C^33,53. The sections were counterstained with hematoxylin. Within the liver, the percentage of hepatocytes positive for enzyme activity was determined by counting the number of stained cells within four adjacent high-magnification fields along an imaginary line bisecting the length of the tissue section and dividing this by the total number of hepatocytes within the same field. Hepatocytes were identified as polygonal cells with large central nuclei. The observer was blinded to the experimental conditions. Quantitative -glucuronidase activity measurements were performed using a fluorometric assay on protein lysates of tissue samples frozen in liquid nitrogen at the time of harvesting^33,53. Equal amounts of protein from each tissue and equal volumes of serum were assayed using the substrate 4-methylumbelliferyl -D-glucuronide. Low levels of nonspecific activity in MPS VII mouse tissues (mean level from at least three MPS VII mice) were subtracted from the results obtained from the vector-injected mice. Enzyme activity is reported within the text as the percentage of wild-type levels to allow for comparison across studies. The wild-type levels were from tissues obtained from homozygous normal mice (homozygous Gus⁺/Gus⁺; at least three for each organ) in our breeding colony and were analyzed in parallel with the MPS VII mice. Wild-type -glucuronidase levels within the organs were as follows: liver 209 19 (standard error of the mean), spleen 858 24, kidney 213 5.5, heart 23 0.8, lung 313 13, and brain 49 4.8 U/mg. For the serum wild-type -glucuronidase levels were 16 1.9 U/ml. One unit is defined as the amount of enzyme activity required to release 1 nmol of 4-methylumbelliferone per hour at 37°C. In a similar manner, -galactosidase and -hexosaminidase activity levels were measured using the substrates 4-methylumbelliferyl--D-galactopyranoside and 4-methylumbelliferyl-N-acetyl--D-glucosaminide, respectively. Wild-type -galactosidase and -hexosaminidase activity levels within the brain were 53 0.8 and 2594 100 U/mg, respectively.

To evaluate for lysosomal storage, the tissue specimens fixed in 2.5% glutaraldehyde were postfixed in 1% osmium tetroxide in cacodylate buffer (0.2 M, pH 7.4). Tissues were washed in buffer, dehydrated in a graded series of ethanol solutions, cleared in propylene oxide, and embedded in Epon–Araldite resin. Semithin (0.5 m) sections were stained with toluidine blue and examined under a light microscope. For each mouse, at least four sections from two random, 1-mm³ specimens from the liver, spleen, kidney, heart, and lung and four random, 1-mm³ specimens from the brain were examined. An investigator blinded to the experimental conditions performed quantitative assessment of the degree of lysosomal storage in comparison to normal, wild-type mice and age-matched, uninjected MPS VII mice. The degree of storage was graded on a scale of 0 to 4: 0, no different from normal, wild-type mice; 1, the majority of the areas within the specimens examined contained cells that were no different from normal mice; 2, the majority of areas had cells with evidence of storage at low levels; 3, histopathologic abnormalities consistent with lysosomal storage were present in all areas of the specimens examined, but were less extensive than those in age-matched uninjected-MPS VII mice; 4, no different from the degree of histopathologic abnormalities observed in age-matched uninjected MPS VII mice. Electron microscopic evaluation for lysosomal storage was performed on selected tissue specimens. Ultrathin sections (60 nm) from these specimens were stained with lead citrate and uranyl acetate and examined with a Hitachi H600 electron microscope.

To evaluate for the presence of rAAV genomes within tissue specimens, total genomic DNA was isolated using a column purification procedure (DNeasy Tissue Kit; Qiagen, Inc., Valencia, CA, USA). For each tissue specimen, equal amounts (10 ng) of genomic DNA were evaluated. This amount of DNA represents approximately 1667 murine cells (based upon 6 pg of genomic DNA per mouse cell). Vector genomes were detected with a fluorescence-based, real-time, quantitative polymerase chain reaction assay (ABI Prism 7700 Sequence Detection System; Perkin–Elmer, Foster City, CA, USA). Primers were designed to anneal to DNA sequences flanking the junction of the first and second exons of the Gus^a cDNA. Sequences of the primers were 5'-CGGCAGCCGCTACGG-3 (forward/sense) and 5'-GATGTCATTGAAGCTAGAAGGGACA-3' (reverse/antisense). The fluorescent probe sequence was 5'-FAM (6-carboxy-fluorescein)-AGTCGGGCCCAGTCTTGGACATGC-TAMRA (6-carboxy-tetramethyl-rhodamine)-3' (sense). The quantity of the amplified products was determined by comparison to serial dilutions of plasmid DNA containing the Gus^a cDNA.

Statistical analysis

Means standard deviations and, where appropriate, medians are reported. The statistical test used for analysis of each data set is indicated within the text. Groups were compared using the Student t test or the nonparametric Kruskal–Wallis and Wilcoxon rank sum tests. Spearman's correlation, a distribution-free analog to the Pearson correlation coefficient, was used for all correlations due to the small sample sizes (the assumption of normality did not hold) and the ordinal scale of the storage scores. Kaplan–Meier survival estimates using the equal weighting (log-rank) test were used to evaluate the survival of mice following vector administration. An = 0.05 level of significance was used for all tests.

References

1. Neufeld, E. F. and Muenzer, J. (2001). The mucopolysaccharidoses. In: Scriver C. R., Beaudet A. L., Sly W. S., and Valle D. (Eds), The Metabolic & Molecular Bases of Inherited Disease. McGraw–Hill: New York. 3421–3452.
2. Peters, C. and Steward, C. G. (2003). Hematopoietic cell transplantation for inherited metabolic diseases: an overview of outcomes and practice guidelines. Bone Marrow Transplant. 31: 229–239. | Article | PubMed | ISI | ChemPort |
3. Kakkis, E. D. (2002). Enzyme replacement therapy for the mucopolysaccharide storage disorders. Expert Opin. Invest. Drugs. 11: 675–685. | Article | ChemPort |
4. Sly, W. S., Quinton, B. A., McAlister, W. J. and Rimoin, D. L. (1973). -Glucuronidase deficiency: report of clinical, radiologic and biochemical features of a new mucopolysaccharidosis. J. Pediatr. 82: 249–257. | PubMed | ISI | ChemPort |
5. Birkenmeier, E. H., et al. (1989). Murine mucopolysaccharidosis type VII: characterization of a mouse with -glucuronidase deficiency. J. Clin. Invest. 83: 1258–1266. | PubMed | ISI | ChemPort |
6. Vogler, C., Barker, J., Sands, M. S., Levy, B., Galvin, N. and Sly, W. S. (2001). Murine mucopolysaccharidosis VII: impact of therapies on the phenotype, clinical course, and pathology in a model of lysosomal storage disease. Pediatr. Dev. Pathol. 4: 421–433. | Article | PubMed | ChemPort |
7. Sands, M. S. and Birkenmeier, E. H. (1993). A single-base-pair deletion in the -glucuronidase gene accounts for the phenotype of murine mucopolysaccharidosis type VII. Proc. Natl. Acad. Sci. USA. 90: 6567–6571. | Article | PubMed | ChemPort |
8. Vogler, C., et al. (1990). A murine model of mucopolysaccharidosis VII: gross and microscopic findings in beta-glucuronidase-deficient mice. Am. J. Pathol. 136: 207–217. | PubMed | ISI | ChemPort |
9. Daly, T. M., Ohlemiller, K. K., Roberts, M. S., Vogler, C. A. and Sands, M. S. (2001). Prevention of systemic clinical disease in MPS VII mice following AAV-mediated neonatal gene transfer. Gene Ther. 8: 1291–1298. | Article | PubMed | ChemPort |
10. Kornfeld, S. (1992). Structure and function of the mannose 6-phosphate/insulin like growth factor II receptors. Annu. Rev. Biochem. 61: 307–330. | Article | PubMed | ISI | ChemPort |
11. Hille-Rehfeld, A. (1995). Mannose 6-phosphate receptors in sorting and transport of lysosomal enzymes. Biochim. Biophys. Acta. 1241: 177–194. | Article | PubMed | ChemPort |
12. Dahms, N. M. and Hancock, M. K. (2002). P-type lectins. Biochim. Biophys. Acta. 1572: 317–340. | Article | PubMed | ISI | ChemPort |
13. Gao, C., Sands, M. S., Haskins, M. E. and Ponder, K. P. (2000). Delivery of a retroviral vector expressing human -glucuronidase to the liver and spleen decreases lysosomal storage in mucopolysaccharidosis VII mice. Mol. Ther. 2: 233–244. | Article | PubMed | ChemPort |
14. Stein, C. S., et al. (2001). In vivo treatment of hemophilia A and mucopolysaccharidosis type VII using nonprimate lentivirus vectors. Mol. Ther. 3: 850–856. | Article | PubMed | ISI | ChemPort |
15. Xu, L., Mango, R. L., Sands, M. S., Haskins, M. E., Ellinwood, N. M. and Ponder, K. P. (2002). Evaluation of pathological manifestations of disease in mucopolysaccharidosis VII mice after neonatal hepatic gene therapy. Mol. Ther. 6: 745–758. | Article | PubMed | ISI | ChemPort |
16. Ohashi, T., Watabe, K., Uehara, K., Sly, W. S., Vogler, C. and Eto, Y. (1997). Adenovirus-mediated gene transfer and expression of human -glucuronidase gene in the liver, spleen, and central nervous system in mucopolysaccharidosis type VII mice. Proc. Natl. Acad. Sci. USA. 94: 1287–1292. | Article | PubMed | ChemPort |
17. Stein, C. S., Ghodsi, A., Derksen, T. and Davidson, B. L. (1999). Systemic and central nervous system correction of lysosomal storage in mucopolysaccharidosis type VII mice. J. Virol. 73: 3424–3429. | PubMed | ISI | ChemPort |
18. Kosuga, M. (2000). Adenovirus-mediated gene therapy for mucopolysaccharidosis VII: involvement of cross-correction in wide-spread distribution of the gene products and long-term effects of CTLA-4Ig coexpression. Mol. Ther. 1: 406–413. | Article | PubMed | ChemPort |
19. Xia, H., Mao, Q. and Davidson, B. L. (2001). The HIV Tat protein transduction domain improves the biodistribution of beta-glucuronidase expressed from recombinant viral vectors. Nat. Biotechnol. 19: 640–644. | Article | PubMed | ISI | ChemPort |
20. Li, X.-K., et al. (2002). Prolongation of transgene expression by coexpression of cytokine response modifier A in rodent liver after adenoviral gene transfer. Mol. Ther. 5: 262–268. | Article | PubMed | ChemPort |
21. Watson, G. L., et al. (1998). Treatment of lysosomal storage disease in MPS VII mice using a recombinant adeno-associated virus. Gene Ther. 5: 1642–1649. | Article | PubMed | ChemPort |
22. Daly, T. M., Vogler, C., Levy, B., Haskins, M. E. and Sands, M. S. (1999). Neonatal gene transfer leads to widespread correction of pathology in a murine model of lysosomal storage disease. Proc. Natl. Acad. Sci. USA. 96: 2296–2300. | Article | PubMed | ChemPort |
23. Daly, T. M., Okuyama, T., Vogler, C., Haskins, M. E., Muzyczka, N. and Sands, M. S. (1999). Neonatal intramuscular injection with recombinant adeno-associated virus results in prolonged beta-glucuronidase expression in situ and correction of liver pathology in mucopolysaccharidosis type VII mice. Hum. Gene Ther. 10: 85–94. | Article | PubMed | ISI | ChemPort |
24. Elliger, S. S., Elliger, C. A., Lang, C. and Watson, G. L. (2002). Enhanced secretion and uptake of beta-glucuronidase improves adeno-associated viral-mediated gene therapy of mucopolysaccharidosis type VII mice. Gene Ther. 5: 617–626. | ChemPort |
25. Clark, K. R., Liu, X., McGrath, J. P. and Johnson, P. R. (1999). Highly purified recombinant adeno-associated virus vectors are biologically active and free of detectable helper and wild-type viruses. Hum. Gene Ther. 10: 1031–1039. | Article | PubMed | ISI | ChemPort |
26. Sferra, T. J., et al. (2000). Recombinant adeno-associated virus-mediated correction of lysosomal storage within the central nervous system of the adult mucopolysaccharidosis type VII mouse. Hum. Gene Ther. 11: 507–519. | Article | PubMed | ISI | ChemPort |
27. Birkenmeier, E. H., et al. (1991). Increased life span and correction of metabolic defects in murine MPS VII following syngeneic bone marrow transplantation. Blood. 78: 3081–3092. | PubMed | ISI | ChemPort |
28. Ross, C. J. D., Bastedo, L., Maier, S. A., Sands, M. S. and Chang, P. L. (2000). Treatment of a lysosomal storage disease, mucopolysaccharidosis VII, with microencapsulated recombinant cells. Hum. Gene Ther. 11: 2117–2127. | Article | PubMed | ChemPort |
29. Nakai, H., Iwaki, Y., Kay, M. A. and Couto, L. B. (1999). Isolation of recombinant adeno-associated virus vector cellular DNA junctions from mouse liver. J. Virol. 73: 5438–5447. | PubMed | ISI | ChemPort |
30. Nathwani, A. C., Davidoff, A., Hanawa, H., Zhou, J.-F., Vanin, E. F. and Nienhuis, A. W. (2001). Factors influencing in vivo transduction by recombinant adeno-associated viral vectors expressing the human factor IX cDNA. Blood. 97: 1258–1265. | Article | PubMed | ChemPort |
31. Mingozzi, F., et al. (2002). Improved hepatic gene transfer by using an adeno-associated virus serotype 5 vector. J. Virol. 76: 10497–10502. | Article | PubMed | ISI | ChemPort |
32. Grimm, D., et al. (2003). Preclinical in vivo evaluation of pseudotyped adeno-associated virus vectors for liver gene therapy. Blood. 102: 2412–2419. | Article | PubMed | ISI | ChemPort |
33. Wolfe, J. H. and Sands, M. S. (1996). Murine mucopolysaccharidosis type VII: a model system for somatic gene therapy of the central nervous system. In: Lowenstein P. R., and Enquist L. W. (Eds), Protocols for Gene Transfer in Neuroscience: Towards Gene Therapy of Neurologic Disorders. Wiley: Essex. 263–274.
34. Taylor, R. M. and Wolfe, J. H. (1997). Decreased lysosomal storage in the adult MPS VII mouse brain in the vicinity of grafts of retroviral vector-corrected fibroblasts secreting high levels of -glucuronidase. Nat. Med. 3: 771–774. | Article | PubMed | ISI | ChemPort |
35. Lau, C., et al. (1995). Retroviral gene transfer into the intestinal epithelium. Hum. Gene Ther. 6: 1145–1151. | PubMed | ChemPort |
36. Guo, Z. S., Wang, L. H., Eisensmith, R. C. and Woo, S. L. (1996). Evaluation of promoter strength for hepatic gene expression in vivo following adenovirus-mediated gene transfer. Gene Ther. 3: 802–810. | PubMed | ISI | ChemPort |
37. Snyder, R. O., et al. (1997). Persistent and therapeutic concentrations of human factor IX in mice after hepatic gene transfer of recombinant AAV vectors. Nat. Genet. 16: 270–276. | Article | PubMed | ISI | ChemPort |
38. Xiao, W., Berta, S. C., Lu, M. M., Moscioni, A. D., Tazelaar, J. and Wilson, J. M. (1998). Adeno-associated virus as a vector for liver-directed gene therapy. J. Virol. 72: 10222–10226. | PubMed | ISI | ChemPort |
39. Miao, C. H., et al. (2000). Nonrandom transduction of recombinant adeno-associated virus vectors in mouse hepatocytes in vivo: cell cycling does not influence hepatocyte transduction. J. Virol. 74: 3793–3803. | Article | PubMed | ISI | ChemPort |
40. Chen, S.-J., Tazelaar, J. and Wilson, J. M. (2001). Selective repopulation of normal mouse liver by hepatocytes transduced in vivo with recombinant adeno-associated virus. Hum. Gene Ther. 12: 45–50. | Article | PubMed | ChemPort |
41. Nakai, H., et al. (2002). A limited number of transducible hepatocytes restricts a wide-range linear vector dose response in recombinant adeno-associated virus-mediated liver transduction. J. Virol. 76: 11343–11349. | Article | PubMed | ISI | ChemPort |
42. Sly, W. S., et al. (2001). Active site mutant transgene confers tolerance to human beta-glucuronidase without affecting the phenotype of MPS VII mice. Proc. Natl. Acad. Sci. USA. 98: 2205–2210. | Article | PubMed | ChemPort |
43. Chao, H. and Walsh, C. E. (2001). Induction of tolerance to human factor VII in mice. Blood. 97: 3311–3312. | Article | PubMed | ChemPort |
44. Mount, J. D., et al. (2002). Sustained phenotypic correction of hemophilia B dogs with factor IX null mutation by liver-directed gene therapy. Blood. 99: 2670–2676. | Article | PubMed | ISI | ChemPort |
45. Mingozzi, F., et al. (2003). Induction of immune tolerance to coagulation factor IX antigen by in vivo hepatic gene transfer. J. Clin. Invest. 111: 1347–1356. | Article | PubMed | ISI | ChemPort |
46. Daly, T. M., Lorenz, R. G. and Sands, M. S. (2000). Abnormal immune function in vivo in a murine model of lysosomal storage disease. Pediatr. Res. 47: 757–762. | PubMed | ChemPort |
47. Wawrzyniak, C. J., et al. (1989). DNA determinants of structural and regulatory variation within the murine beta-glucuronidase gene complex. Mol. Cell. Biol. 9: 4074–4078. | PubMed | ChemPort |
48. Funkenstein, B., Leary, S. L., Stein, J. C. and Catterall, J. F. (1988). Genomic organization and sequence of the gus-s^a allele of the murine -glucuronidase gene. Mol. Cell. Biol. 8: 1160–1168. | PubMed | ChemPort |
49. Uetsuki, T., Naito, A., Nagata, S. and Kaziro, Y. (1989). Isolation and characterization of the human chromosomal gene for polypeptide chain elongation factor-1 alpha. J. Biol. Chem. 264: 5791–5798. | PubMed | ISI | ChemPort |
50. Clark, K. R., Voulgaropoulou, F., Fraley, D. M. and Johnson, P. R. (1995). Cell lines for the production of recombinant adeno-associated virus. Hum. Gene Ther. 6: 1329–1341. | PubMed | ISI | ChemPort |
51. Clark, K. R., Voulgaropoulou, F. and Johnson, P. R. (1996). A stable cell line carrying adenovirus-inducible rep and cap genes allows for infectivity titration of adeno-associated virus vectors. Gene Ther. 3: 1124–1132. | PubMed | ISI | ChemPort |
52. Sferra, T. J., McNeely, D. and Johnson, P. R. (1997). Gene transfer to the intestinal tract: a new approach using selective injection of the superior mesenteric artery. Hum. Gene Ther. 8: 681–687. | PubMed | ISI | ChemPort |
53. Kyle, J. W., Grubb, J. H., Galvin, N. and Vogler, C. (1992). -Glucuronidase (GUS) assay in animal tissue. In: Gallagher S. R. (Ed), GUS Protocols: Using the GUS Gene as a Reporter of Gene Expression. Academic Press: San Diego. 189–203.

Acknowledgements

The authors thank Cynthia McAllister for technical assistance with the preparation of the electron microscopic images, Amy Dutcher for secretarial support, and Soledad Fernandez, Ph.D., for her support in performing the statistical analyses. We acknowledge the Columbus Children's Research Institute Viral Vector Core Laboratory for production of the viral vector stocks. The National Institutes of Health, National Institute of Neurological Disorders and Stroke (R01NS39071), supported this work.