Targeted disruption of the lysosomal α-mannosidase gene results in ...

© 1999 Oxford University Press

Human Molecular Genetics, 1999, Vol. 8, No. 8 1365–1372

ARTICLE

Targeted disruption of the lysosomal α-mannosidase gene results in mice resembling a mild form of human α-mannosidosis Sofia Stinchi1, Renate Lüllmann-Rauch2, Dieter Hartmann2, Ruth Coenen2, Tommaso Beccari1, Aldo Orlacchio1, Kurt von Figura3 and Paul Saftig3,+ 1Dipartimento

di Biologia Cellulare e Molecolare, Sezione di Biochimica e Biologia Molecolare, Università degli Studi di Perugia, Via del Giochetto, 06126 Perugia, Italy, 2Anatomisches Institut der Universität Kiel, Otto-Hahn-Platz 8, 24043 Kiel, Germany and 3Zentrum Biochemie und Molekulare Zellbiologie, Abteilung Biochemie II, Universität Göttingen, Heinrich Düker Weg 12, 37073 Göttingen, Germany Received March 22, 1999; Revised and Accepted May 6, 1999

α-Mannosidosis is a lysosomal storage disease with autosomal recessive inheritance caused by a deficiency of the lysosomal α-mannosidase, which is involved in the degradation of asparagine-linked carbohydrate cores of glycoproteins. An α-mannosidosis mouse model was generated by targeted disruption of the gene for lysosomal α-mannosidase. Homozygous mutant animals exhibit α-mannosidase enzyme deficiency and elevated urinary secretion of mannose-containing oligosaccharides. Thin-layer chromatography revealed an accumulation of oligosaccharides in liver, kidney, spleen, testis and brain. The cellular alterations were characterized by multiple membrane-limited cytoplasmic vacuoles as seen for instance in liver, exocrine pancreas, kidney, thyroid gland, smooth muscle cells, osteocytes and in various neurons of the central and peripherial nervous systems. The morphological lesions and their topographical distribution, as well as the biochemical alterations, closely resemble those reported for human α-mannosidosis. This mouse model will be a valuable tool for studying the pathogenesis of inherited α-mannosidosis and may help to evaluate therapeutic approaches for lysosomal storage diseases.

INTRODUCTION α-Mannosidosis is a rare lysosomal storage disease with autosomal recessive inheritance. It has been shown to cause a collection of clinical symptoms characteristic of lysosomal storage diseases: progressive mental retardation, impaired hearing, dysostosis multiplex, immune defects, elevation of serum and urinary oligosaccharide levels and an enlargement of lysosomes in most cell types resulting from the accumulation of undegraded oligosaccharides (1–3). Other findings are lens opacities, muscular hypotonia, macroglossia, prognathism, vacuolated lymphocytes and pancytopenia. α-Mannosidosis is caused by a deficiency of lysosomal α-mannosidase (LAMAN; EC 3.2.1.24). α-Mannosidase is an exoglycosidase which cleaves α-linked mannose residues from the non-reducing end during the ordered degradation of N-linked glycoproteins (4). It catalyses the hydrolysis of α1,2-, α1,3- and α1,6-mannoside linkages present in complex, hybrid and highmannose asparagine-linked glycans (5). In human and mouse tissues, α-D-mannosidase activity can be separated by DEAE– cellulose chromatography into two major isoenzymes, A and B, +To

with similar pH optima (pH 4.5), thermal stabilities, molecular masses (6) and broad specificities towards natural and synthetic substrates (7). Both in human and mouse, lysosomal α-mannosidase is encoded by a single gene (8,9). The cDNA of human (8– 10) and mouse (6,7) and the gene of mouse α-mannosidase (11) have been isolated and characterized. A deficiency of the enzyme causes intralysosomal accumulation of mainly unbranched oligosaccharides of which the major storage product is the trisaccharide Manα1–3Manβ1–4GlcNAc (1,2,12–13). The clinical severity of α-mannosidosis ranges in a continuum from mildly affected to severely affected patients, and heterogeneity has been observed between affected siblings (14). Multiple mutations in the α-mannosidase genes of αmannosidosis patients and naturally occurring animal models have been detected documenting the heterogeneity of the disease also at the molecular level (10,15–18). There is no apparent correlation between the types of mutation and the clinical manifestations. Environmental and epigenic factors may contribute to the clinical variation observed in α-mannosidosis (16).

whom correspondence should be addressed. Tel: +49 551 395932; Fax: +49 551 395979; Email: [email protected]

1366 Human Molecular Genetics, 1999, Vol. 8, No. 8

Figure 1. Targeted disruption of the α-mannosidase gene. (A) Strategy for inactivation of the α-mannosidase gene by homologous recombination in ES cells. (I) Structure of the genomic α-mannosidase gene region. Exons are indicated by black boxes, flanking introns are indicated by solid lines. The bar designates the 5' DNA probe used for Southern blot analysis. (II) Targeting vector pαM6Bneo with 4 kbp homology to the α-mannosidase gene locus. The neo cassette (open box) was inserted into an NheI restriction site in exon 2. The arrow marks the direction of transcription of the neo gene. Broken line, plasmid vector pBluescript SK–. (III) Predicted α-mannosidase gene locus after homologous recombination. Restriction sites: E, EcoRI; B, BamHI; H, HincII. (B) Southern blot analysis of ES cell clones. Probe A was hybridized to EcoRI-digested genomic DNA from ES cell clones EαM-27 and EαM-29 and targeted ES cell clone EαM-28. An additional 5.6 kb DNA fragment indicates a targeted allele. (C) Southern blot analysis of tail DNA. DNA was digested with EcoRI and analysed with the 5' probe (wild-type allele, 17 kb; mutant allele, 5.6 kb). (D) PCR analysis of tail genomic DNA with an exon-specific PCR amplifying a 0.4 kb fragment in wild-type (+/+) mice, 0.4 and 1.6 kb fragments in heterozygous (+/–) mutants and a 1.6 kb fragment in homozygous (–/–) α-mannosidase-deficient animals.

There is no effective therapy for α-mannosidosis. Bone marrow transplantations (BMTs) have been performed in α-mannosidosis with varying outcomes. Since HLA-matched donors are not available for the majority of α-mannosidosis patients, BMT is still associated with high mortality rates (19). Remarkable success has been obtained with feline α-mannosidosis, a naturally occurring animal model (20), where BMT has led to replacement of α-mannosidase activity in cells of the central nervous system (21). Enzyme replacement (22) and somatic gene therapy (23) have promising prospects for therapy of lysosomal storage diseases, but suitable animal models are required for their evaluation. Although cat (20) and cattle (24) strains with α-mannosidase deficiencies have been described, they are not as appropriate for testing the pathophysiology of the disease and a variety of therapeutic strategies as a murine model. Here we report the generation of an α-mannosidosis

mouse model by disrupting the α-mannosidase gene using homologous recombination. RESULTS Generation of α-mannosidase-deficient mice A gene targeting vector was constructed to disrupt the α-mannosidase mouse gene (Fig. 1A, I–III). A homologous recombined embryonic stem (ES) cell clone (EαM-28; Fig. 1B) was used to generate chimeras which transmitted the introduced mutation to their offspring (Fig. 1C and D). All mice used for experiments were derived from ES cell clone EαM-28. Outbred mice (129SV×C57B6J) were used for phenotypic characterization. Northern blotting (Fig. 2A) and RT–PCR (data not shown) demonstrated a complete absence of α-mannosidase

Human Molecular Genetics, 1999, Vol. 8, No. 8 1367

Table 1. Increased amount of neutral sugars in α-mannosidase-deficient tissues Neutral carbohydrate (mg neutral carbohydrate/g wet weight) (± SD) Tissue α-mannosidase+/+

α-mannosidase–/–

Kidney 0.34 (±0.1)

0.95 (±0.13)

Spleen 0.36 (±0.08)

0.93 (±0.06)

Brain

0.32 (±0.13)

0.71 (±0.08)

Liver

12.22 (±2.91)

13.61 (±2.31)

Testis

0.42 (±0.16)

0.67 (±0.33)

An increased amount of neutral sugars was found in homogenates from kidney, spleen, brain and testis of α-mannosidase-deficient (–/–) mice. In liver tissue, only a moderate increase in the content of neutral sugars was observed. Values represent means ± SD taken from three independent experiments.

crosses (Fig. 1D) revealed a frequency of 23.2% for homozygous and 48.5% for heterozygous mutant mice, which are in accordance with the expected Mendelian frequencies. Urinary excretion of oligosaccharides in α-mannosidasedeficient mice

Figure 2. Inactivation of the α-mannosidase gene and loss of lysosomal α-mannosidase activity in α-mannosidase-deficient mice. (A) Northern blot analysis of α-mannosidase expression. Total RNA (15 µg) was hybridized using murine αmannosidase and glyceraldehyde-3-phosphate dehydrogenase cDNA probes. (B) Determination of lysosomal α-mannosidase activity at pH 4.5 and of endoplasmic reticulum and Golgi α-mannosidase at pH 6.0. The results are shown for brain, liver and kidney of wild-type and α-mannosidase–/– mice. No α-mannosidase was observed at pH 4.5 in α-mannosidase–/– mice. As a control, β-mannosidase activities were included. (C) Separation of α-mannosidase isoenzymes from kidney homogenates from control (closed circles) and α-mannosidase–/– mice (open circles) by DEAE ion-exchange chromatography. Unretained protein was eluted with the column buffer and a linear gradient of NaCl (0–0.5 M) was applied. Fractions were assayed for activity at pH 4.5. Note that no isoenzyme activities could be observed in α-mannosidase–/– kidney.

RNA in homozygous α-mannosidase-deficient mice. Lysosomal α-mannosidase enzyme activity was not detectable, when measured at pH 4.5, in brain, liver and kidney from α-mannosidase–/– mice, whereas it was readily detectable in the respective homogenates from wild-type mice. No significant differences were observed between the genotypes when α-mannosidase activity was measured at pH 6.0 (Fig. 2B). This pH corresponds to the pH optimum of non-lysosomal mannosidases residing in the endoplasmic reticulum, Golgi and cytosol (1). Our observation confirms that the mannosidase activities at pH 4.5 and 6.0 arise from different gene products. Northern blot analysis and determination of lysosomal α-mannosidase activity showed that the α-mannosidase gene has been inactivated and that homozygous mutant mice are devoid of lysosomal α-mannosidase enzyme activity. Ion-exchange chromatography of kidney homogenates of α-mannosidasedeficient and control mice demonstrated that both isoforms of the enzyme had been inactivated, verifying that they are encoded by a single gene (Fig. 2C). Heterozygotes exhibit a normal phenotype and fertility (data not shown). Genotyping of 95 offspring from heterozygote

Homozygous mutant, heterozygous and wild-type mice resulting from heterozygote crosses did not exhibit differences in growth and weight development (data not shown). α-Mannosidase-deficient animals were fertile, did not show an elevated mortality nor obvious altered behaviour, and lacked phenotypic changes on inspection or autopsy up to the age of 10 months. Regular radiological examinations and determination of clinical blood and serum parameters did not reveal abnormalities (data not shown). Urine collected from control, heterozygote and α-mannosidasedeficient mice was subjected to thin-layer chromatography. Homozygous α-mannosidase-deficient mice demonstrated a chromatographic oligosaccharide pattern, which compared well with a sample from a human α-mannosidosis patient. Consistent with the findings from healthy human controls, mannose-containing oligosaccharides were not detectable in the urine of heterozygote or wild-type mice (Fig. 3A). Storage of neutral oligosaccharides in tissues In homogenates from kidney, spleen, brain and testis of α-mannosidase-deficient mice, an increased amount of neutral sugars was found. The increase was most pronounced in kidney, spleen, brain and testis, whereas in liver tissue only a moderate increase in the content of neutral sugars was observed (Table 1). Thin-layer chromatography revealed the presence of excessive amounts of neutral oligosaccharides in α-mannosidase-deficient spleen, kidney, liver (Fig. 3B), testis and brain (data not shown). These oligosaccharides were susceptible to digestion with jack bean α-mannosidase (Fig. 3B, shown for brain). Morphological findings In order to evaluate lysosomal storage, control and α-mannosidase-deficient animals were analysed morphologically at the ages of 2 and 6 months. Macroscopic examination of visceral organs of homozygous α-mannosidase-deficient mice revealed

1368 Human Molecular Genetics, 1999, Vol. 8, No. 8

Figure 3. Urinary excretion and storage in tissues of oligosaccharides in α-mannosidase-deficient animals. (A) Thin-layer chromatography of equal amounts of urine from α-mannosidase +/+, +/– and –/– mice. As additional controls, urine samples from a mannosidosis patient and healthy control were included. st, standard of raffinose:lactose:fructose 1:1:1 (from bottom to top). (B) Thin-layer chromatography of equal amounts of extracts from α-mannosidase–/– spleen, kidney and liver. The silica gels were stained with 0.2% orcinol in sulfuric acid to detect oligosaccharides. (C) Thin-layer chromatography of equal amounts of extracts from α-mannosidase–/– brain with (+) and without (–) overnight digestion with jack bean α-mannosidase. Note that, after digestion, the mannose-containing oligosaccharides (open arrows) are removed and converted to mannose monomers (closed arrow).

no difference in the sizes of the liver, kidney, spleen or brain as compared with age-matched heterozygous and wild-type mice. Although no immediate macroscopic abnormalities were noticed in the brain or visceral organs at that early stage, various tissues of α-mannosidase-deficient mice displayed clearly detectable abnormalities at the microscopic level. Signs of lysosomal storage were observed in liver, pancreas, spleen, kidney, eye, thyroid gland, smooth muscle cells, bone and the peripheral and central nervous systems, as indicated by the accumulation of translucent vacuoles. Using Pisum sativum (25) and Lens culinaris (26) lectins specific for terminal α-D-mannosyl residues, a clearly increased staining was demonstrated in liver sections from α-mannosidase-deficient mice (Fig. 4B, shown for L.culinaris), whereas the staining intensity in control sections was low (Fig. 4A). Ultrastructural analysis of α-mannosidase-deficient liver revealed the accumulation of clear vacuoles in hepatocytes (Fig. 4C), Kupffer cells and sinus endothelial cells (Fig. 4D). Despite the lack of immediate radiological abnormalities and alterations of overall bone histological architecture, lysosomal storage could also be identified easily in osteocytes and endosteal cells of several bones (Fig. 4E). Clear vacuoles were also observed in epithelial cells of the proximal and distal tubules of the kidney of α-mannosidase-deficient mice (Fig. 4F). Additionally, a prominent vacuolization was observed in the acinar cells of the exocrine pancreas (Fig. 4H), whereas such alterations were not observed in cells of controls (Fig. 4G). Lysosomal storage was seen in spleen, most conspicuously in sinus endothelia, reticulum cells, macrophages and trabecular fibroblasts of the red pulpa. In lymphocytes, only single small vacuoles were found (data not shown). In the eye, storage was observed in retinal photoreceptors, ciliary epithelium and smooth muscle, corneal stroma cells and endothelium (data not shown). The cytoplasmic vacuoles in the null mutant tissues were membrane-limited and appeared either totally empty or contained some fine granular material, thus closely resembling the histopathological finding described in tissues of human αmannosidosis (3).

It is of note that when tissues of α-mannosidase-deficient mice were stained for the lysosomal markers cathepsin D and LAMP-1 (data not shown), a cell type-specific increase in immunoreactivity was found for either LAMP-1 or cathepsin D, but not for both antigens. Neuropathology of α-mannosidase-deficient animals Vacuolized cells were seen in all brain regions and were most prominent in neurons, yet with regional differences. Brain sections were immunostained for the lysosomal markers LAMP-1 (data not shown) and cathepsin D. In the dorsal isocortex of control mice (Fig. 5A), only neurons were stained to varying degrees. In α-mannosidase-deficient mice, the number of intensely stained neurons was clearly increased (Fig. 5B). Glial cells also showed immunoreactivity for cathepsin D (Fig. 5B, insert). Neurons of the isocortex and, most prominently, those of layers II/III and V/VI displayed clear vacuoles (data not shown). The L.culinaris lectin revealed storage of mannosecontaining material in neurons of the CA2 and CA3 region of the hippocampal cornu ammonis (Fig. 5D). This was paralleled by numerous abnormal vacuoles in these neurons as seen in semi-thin sections (data not shown). Neuronal somata did not stain with the lectin in control sections (Fig. 5C). In the cerebellum, Purkinje and granule neurons displayed lysosomal storage (data not shown). Furthermore, clear vacuoles were observed in some nuclei of the brain stem, e.g. the inferior olive (data not shown). Additionally, abnormal axonal spheroids representing axonal swellings with diameters up to 45 µm, either myelinated or unmyelinated, occurred in the central white matter (see Fig. 5F for the nucleus gracilis). The axoplasm contained an accumulation of polymorphous material (Fig. 5G). Similar axonal dystrophy was not seen in any of the wild-type controls (Fig. 5E). The storage phenotype was also observed in the peripheral nervous system. In peripheral neurons, such as trigeminal, spiral and vestibular ganglia, numerous storage lysosomes were prominent findings (data not shown). Axonal spheroids were

Human Molecular Genetics, 1999, Vol. 8, No. 8 1369

seen in pre-terminal portions of sensory trigeminal and vestibular nerve fibres (data not shown). DISCUSSION A deficiency of the lysosomal α-mannosidase is the basis of the lysosomal storage disease α-mannosidosis. Mice homozygous for the mutant gene, which express neither specific α-mannosidase mRNA nor α-mannosidase activity, develop a biochemical and morphological phenotype closely resembling that of naturally occurring α-mannosidosis in human, as well as in cat and cattle (20,21,24). Drastically increased urinary excretion of mannose-containing oligosaccharides, which has served as a diagnostic parameter in human α-mannosidosis (2,3,12), is a characteristic finding in α-mannosidase-deficient mice. The increase of neutral sugars in most organs as well as the accumulation of oligosaccharides in spleen, kidney, liver, testis and brain are further abnormalities of α-mannosidase-deficient mice. Staining with lectins specific for mannose residues confirmed the presence of mannose-containing oligosaccharides in many different cell types. These cells also displayed a cell type-specific increased immunoreactivity against LAMP-1 and/or cathepsin D, which is a common finding in lysosomal storage disorders (27,28). The significance of the increase in immunoreactivity for either LAMP-1 or cathepsin D, but not for both antigens, remains obscure. Similar observations have been made in other animal models with lysosomal storage (27,29). Electron microscopic examination of α-mannosidase-deficient tissues confirmed a prominent lysosomal storage in different cells of liver, kidney, spleen, pancreas, testis, eye, thyroid gland, smooth muscle, bone and the central and peripheral nervous systems. The apparently high specificity regarding the vulnerability of cells to the development of lysosomal storage is exemplified in the case of hippocampal pyramidal neurons in which storage is restricted to neurons of certain sectors of the cornu ammonis whereas immediately neighbouring similar cells are not affected at all. Likewise, endothelia display lysosomal storage in, for example, spleen and liver, but not in pancreas and other parenchymal organs. In the light of the ubiquitous occurrence of glycoproteins with mannose-containing N-linked oligosaccharides and the ubiquitous expression of lysosomal α-mannosidase, the highly selective restriction of lysosomal storage to certain cell types within tissues is somewhat unexpected. It points to hitherto unrecognized cell type-dependent differences in glycoprotein catabolism. While the histological and biochemical results underline the similarities between α-mannosidosis in human and mice, the clinical presentation exhibits some differences. In human, α-mannosidosis is presenting as a disease exhibiting a continuum of symptoms ranging from severe to mild forms. In its most frequent form, the disease becomes overt within the first year of life, mostly associated with psychomotor retardation, lens opacities and signs of dysostosis multiplex leading to, for example, facial coarsening. These clinical symptoms were not observed in αmannosidase-deficient mice up to the age of 12 months. It therefore appears that the phenotype of murine α-mannosidosis is somewhat attenuated compared with the most frequent form of αmannosidosis in human in spite of the fact that these mice are homozygous for a null mutation in the α-mannosidase gene. Such attenuation has also been described for other animal models of

Figure 4. Histopathology of α-mannosidase–/– mice. (A and B) Liver of a 6month-old control (A) and α-mannosidase-deficient (B) mouse stained with the mannose-binding lectin from L.culinaris. In controls, a subset of slender cells in the sinusoideal region stains positively. In deficient mice, not only is the density of the stained cells increased, but they also are considerably enlarged (inserts). CV, central vein. Bar, 150 µm; insert: bar, 20 µm. (C) Electron micrograph of liver from an α-mannosidase–/– mouse at 6 months. The peribiliary region of a hepatocyte is shown. Arrows point to abnormal clear vacuoles. BC, bile canaliculus. Bar, 2 µm. (D) Liver sinusoid of the same mouse as in (C). The Kupffer cell (Ku) and the sinus-lining endothelial cell (SE) contain abnormal clear vacuoles (arrows). Bar, 6 µm. (E) Osteocyte in temporal bone of an α-mannosidase–/– mouse at 2 months. The arrow points to abnormal vacuoles. Bar, 2 µm. (F) Kidney of an α-mannosidase–/– mouse at 6 months. The straight part of a distal tubule (DT) in a medullary ray. The epithelium displays abnormal vacuoles (arrow). Bar, 6 µm. (G and H) Exocrine pancreas of (G) an α-mannosidase+/+ and (H) an α-mannosidase–/– mouse at 2 months. The acinar cells of the deficient animal show abnormal clear vacuoles. The electron-dense particles are zymogen granules. Bar, 40 µm (light micrographs). Inserts show representative electron micrographs; bar, 20 µm.

lysosomal storage diseases (30–33). The reason for this phenotypic attenuation could be related to different rates of storage of oligosaccharides between species and occasionally has been shown to be related to the existence of metabolic bypasses (34). It may be that the phenotype of α-mannosidase-deficient mice will worsen as they become older (presently the oldest animals are 12

1370 Human Molecular Genetics, 1999, Vol. 8, No. 8

months of age) and continue to accumulate storage material. In fact we have observed a progression of the histological findings with age in the α-mannosidosis mice. In spite of the milder clinical phenotype, the lysosomal storage and excretion of mannosecontaining oligosaccharides suggest that murine α-mannosidosis represents a valid mouse model for the human disease. Although there are also cattle and cat models of α-mannosidosis (20,24), which are useful for some purposes, the mouse model has considerable advantages over mannosidosis models in cat and cattle in terms of lifespan, ease of breeding and control of the genetic background of the affected animals. The murine α-mannosidosis model offers the possibility to investigate systematically the pathogenesis of the disease and of therapeutic approaches for lysosomal storage disorders. Mouse models (29,35,36) have proven to be extremely valuable for studying the effects of BMT and enzyme replacement. BMT in feline α-mannosidosis has been used successfully to replace α-mannosidase activity also in cells of the central nervous system (21). In light of this observation, the murine α-mannosidosis model will be used to evaluate the therapeutic effect of αmannosidase replacement and transplantation of genetically engineered bone marrow stem cells. MATERIALS AND METHODS Generation of mice with targeted disruption of αmannosidase The isolation of the murine α-mannosidase gene (Man2b) has been described recently (13). For construction of a targeting vector, a 4 kbp BamHI DNA fragment was subcloned into pBluescript SK–. This fragment encompasses 1.3 kb of the 5'-flanking region and 2.7 kb of the Man2b gene containing exons 1–4. The neomycin phosphotransferase gene (neo) was inserted in the NheI site, generated by in vitro mutagenesis in exon 2 under the control of the PGK promoter (37). Insertion of the neo cassette introduces a premature translational stop codon into the open reading frame of the Man2b gene. The target construct, pαM-BamHI(neo) [Fig. 1A (II)], was used to disrupt the Man2b gene in ES cells, which were cultured as previously described (38). G418-resistant clones were screened by Southern analysis of genomic DNA, which was digested with EcoRI and probed with the 5' probe [Fig. 1A (I and III)]. Mutated ES cells (EαM-28) were microinjected into blastocysts of C57BL/6J females. The resulting chimeras were used to generate heterozygous and subsequently homozygous mutant offspring of inbred (129SV) and outbred (129SV×C57B6J) genetic background. All subsequent experiments used outbred mice derived from ES cell clone EαM-28. Mice were genotyped for the introduced Man2b gene mutation by analysis of genomic tail DNA using exon 2-specific PCR (primers α-Man2: 5'-gcc agg caa ggg ttc tac cgc ag-3' and αMan3: 5'-gaa cag acg cgt gtt gaa cat ca-3') or by Southern hybridization of EcoRI-digested tail DNA and hybridization with the 5' probe. Northern blot analysis Total RNA from spleen and testis of 4-week-old mice was prepared and RNA (15 µg) separated on a formaldehyde–agarose gel and processed as described previously (39). Filters were hybrid-

Figure 5. Lysosomal storage in the central nervous system. (A and B) Dorsal isocortex of a 6-month-old control (A) and α-mannosidase-deficient (B) mouse stained with antibodies to cathepsin D. Normally only neurons are stained to varying degrees; specifically within the molecular zone (insert), even these cells are barely recognizable. In deficient mice (B), the number of intensely staining neurons is considerably increased; moreover, glial cells also become immunoreactive for cathepsin D (insert). Bar, 200 µm; insert: bar, 20 µm. (C and D) Section of the hippocampal cornu ammonis stained with L.culinaris lectin. Whereas in 6-month-old control animals (C) neuronal somata are unstained, in deficient mice of the same age (D) the lectin reveals the storage of mannose-containing oligosaccharides in specific segments of Ammon’s horn. The central vertical vessel in (D) demarcates the approximate border between CA2 and CA3. Bar, 100 µm. (E and F) Nucleus gracilis of (E) α-mannosidase+/+ and (F) α-mannosidase–/– mice at 6 months, light micrographs. In the deficient mouse, numerous axonal spheroids (arrows) are seen, which are absent in the age-matched control. Bar, 60 µm. (G) Nucleus gracilis of the same mouse as in (F), electron micrograph. Two axonal spheroids are shown, one of which (top) is surrounded by a myelin sheath. The axoplasm contains an accumulation of polymorphous material. Bar, 20 µm.

ized subsequently with the murine Man2b cDNA probe and then with glyceraldehyde-3-phosphate dehydrogenase (40). α-Mannosidase activity assay and ion-exchange chromatography Tissue homogenates from 16-week-old mice were generated and processed as described (6). Enzyme activity was deter-

Human Molecular Genetics, 1999, Vol. 8, No. 8 1371

mined using the fluorogenic substrate 4-methyl-umbelliferylα-D-mannopyranoside as described (6). As a control, β-mannosidase activities were determined using the fluorogenic substrate 4-methyl-umbelliferyl-β-D-mannopyranoside. pH optima were determined in 0.1 M/0.2 M citric acid/sodium phosphate buffers in the pH range 3.0–7.6. Ion-exchange chromatography of homogenates of α-mannosidase-deficient and control kidneys was performed as described (6). The resulting fractions were monitored for α-mannosidase activity. Isolation of neutral oligosaccharides from tissues Tissues from 8-week-old knockout and control mice were homogenized (20% w/v) and sonicated in distilled water to release any water-soluble material. Proteins and lipids were precipitated by the addition of 4 vol of methanol and chloroform:water (1:3), respectively (41). The supernatant was passed to the mixed bed resin AG 501-X8 (Bio-Rad, Hercules, CA) to remove charged molecules. The isolated carbohydrates were dried and resuspended in distilled water. Determination of neutral sugars The neutral sugar content was determined by a colorimetric assay, using a solution of sulfuric acid and resorcinol, according to the assay described by Monsigny et al. (42).

Biocell, London, UK). Central sections of each series were stained with haematoxylin and eosin for standard light microscopy. Adjacent sections were used for immunohistological screening for tissue alterations using a panel of antibodies to glial fibrillary acidic protein (GFAP) (astroglia; Dako, Hamburg, Germany), F4/80 (microglia/macrophages; clone obtained from ATTC, Rockford, IL), vimentin (fibroblastic cells; Sigma) and lectins (all from Vector, Burlingame, CA) of Solanum tuberosum (endothelia), RCA-1 (endothelia and activated macrophages), of P.sativum and of L.culinaris (specific for terminal α-D-mannosyl residues). Correlative sections of all tissues were stained for lysosomal marker antigens cathepsin D (44) and LAMP-1 (DSHB, University of Iowa, IA). Detection of primary antibody binding was performed by the use of biotin-labelled secondary antibodies and either avidin– biotin complexes or tyramide signal amplification. ACKNOWLEDGEMENTS We thank N. Hartelt, M. Grell and D. Niemeier for excellent technical assistance, and K. Rajewski (Köln, Germany) for providing the E-14-1 cell line. S.S. was supported by a VIGONI-programme fellowship. This work was supported by the Deutsche Forschungsgemeinschaft grant LU172/8-1, the Fonds der Chemischen Industrie, CNR Target Project on Biotechnology and Cofin 97, MURST.

α-Mannosidase digestion of oligosacharides The isolated neutral carbohydrates were resuspended in either distilled water or 100 mM Na acetate, 2 mM ZnCl2 at pH 5.0 containing 30 U/ml of jack bean α-mannosidase (Sigma, Deisenhofen, Germany) and incubated for 16 h. The samples were desalted, dried, resuspended in distilled water and applied on a thin-layer chromatography plate. Thin-layer chromatography of neutral sugars and urine Thin-layer chromatography of equal amounts of urine and neutral oligosaccharides from tissues was carried out on silica gel 60 plates (Merck, Darmstadt, Germany) according to Sewell (43). Oligosaccharides were separated with two different solvent systems: n-butanol:acetic acid:water (100:50:50, w/v) and n-propanol:nitromethane:water (100:80:60, by volume). The silica gels were stained with 0.2% orcinol in sulfuric acid (20% in water) to detect oligosaccharides. Histological examinations For electron microscopy, the animals were perfused via the left ventricle with 6% glutardialdehyde in 0.1 M phosphate buffer using the right atrium and the jugular veins as outlet. Tissue samples were post-fixed with 2% osmium tetroxide, dehydrated and embedded in Araldite. Semi-thin sections were stained with toluidine blue. Ultrathin sections were processed according to standard techniques. For histochemical investigations, animals were perfused with Bouin’s solution diluted 1:4 in phosphate-buffered saline (PBS). After dissection of individual organs and dehydration, embedding was performed in low melting point paraffin (Wolff, Wetzlar, Germany). Serial sections (7 µm) were cut and mounted on glass slides covered with Biobond (British

REFERENCES 1. Daniel, P.F., Winchester, B. and Warren, C.D. (1994) Mammalian alphamannosidases—multiple forms but a common purpose? Glycobiology, 4, 551–566. 2. Bennet, J.K., Dembure, P.P. and Elsas, L.J. (1995) Clinical and biochemical analysis of two families with type I and type II mannosidosis. Am. J. Med. Genet., 55, 21–26. 3. Neufeld, E.F. (1991) Lysosomal storage diseases. Annu. Rev. Biochem., 60, 257–280. 4. Aronson, N.N.J. and Kuranda, M.J. (1989) Lysosomal degradation of Asn-linked glycoproteins. FASEB J., 3, 2615–2622. 5. Al Daher, S., de Gasperi, R., Daniel, P., Hall, N., Warren, C.D. and Winchester, B. (1991) The substrate-specificity of human lysosomal α-Dmannosidase in relation to genetic α-mannosidosis. Biochem. J., 277, 743–751. 6. Beccari, T., Appolloni, M.G., Costanzi, E., Stinchi, S., Stirling, J.L., Fazia, M.A., Servillo, G., Viola, M.P. and Orlacchio, A. (1997) Lysosomal α-mannosidases of mouse tissues: characteristics of the isoenzymes, and cloning and expression of a full-length cDNA. Biochem. J., 327, 45–49. 7. Merkle, R.K., Zhang, Y., Ruest, P.J., Lal, A., Liao, Y.F.and Moremen, K.W. (1997) Cloning, expression, purification, and characterization of the murine lysosomal acid α-mannosidase. Biochim. Biophys. Acta, 1336, 132–146. 8. Nebes, V.L. and Schmidt, M.C. (1994) Human lysosomal α-mannosidase: isolation and nucleotide sequence of the full-length cDNA. Biochem. Biophys. Res. Commun., 200, 239–245. 9. Liao, Y.F., Lal, A. and Moremen, K.W. (1996) Cloning, expression, purification, and characterization of the human broad specificity lysosomal acid α-mannosidase. J. Biol. Chem., 8, 28348–28358. 10. Nilssen, O., Berg, I., Riise, H.M., Ramachandran, U., Evjen, G., Hansen, G.M., Malm, D., Tranebjaerg, L. and Tollersrud, O.K. (1997) α-Mannosidosis: functional cloning of the lysosomal α-mannosidase cDNA and identification of a mutation in two affected siblings. Hum. Mol. Genet., 6, 717–726. 11. Stinchi, S., Orlacchio, A., Costanzi, E., Stirling, A.R., Menghini, A.R., Orlacchio, A. and Beccari, T. (1998) Promoter characterization and structure of the gene encoding mouse lysosomal α-D-mannosidase. Mamm. Genome, 9, 869–873.

1372 Human Molecular Genetics, 1999, Vol. 8, No. 8

12. Daniel, P.F., Defeudis, D.F. and Lott, I.F. (1981) Mannosidosis: isolation and comparison of mannose-containing oligosaccharides from gingiva and urine. Eur. J. Biochem., 114, 235–237. 13. Michalski, J.C., Haeuw, J.F., Wieruszeski, J.M., Montreuil, J. and Strecker, G. (1990) In vitro hydrolysis of oligomannosyl oligosaccharides by the lysosomal α-D-mannosidases. Eur. J. Biochem., 189, 369–379. 14. Chester, M.A., Lundblad, A., Öckerman, P.A. and Autio, S. (1982) Mannosidosis. In Durand, P. and O’Brian, J. (eds), Genetic Errors of Glycoprotein Metabolism. Edi-Ermes, Milano, Springer-Verlag, Berlin, pp. 89– 121. 15. Berg, T., Tollersrud, O.K., Walkley, S.U., Siegel, D. and Nilssen, O. (1997) Purification of feline lysosomal α-mannosidase, determination of its cDNA sequence and identification of a mutation causing α-mannosidosis in Persian cats. Biochem. J., 328, 863–870. 16. Berg, T., Riise, H.M., Hansen, G.M., Malm, D., Tranebjarg, L., Tollersrud, O.K. and Nilssen, O. (1999) Spectrum of mutations in α-mannosidosis. Am. J. Hum. Genet., 64, 77–88. 17. Gotoda, Y., Wakamatsu, N., Kawai, H., Nishida, Y. and Matsumoto, T. (1998) Missense and nonsense mutations in the lysosomal α-mannosidase gene (MANB) in severe and mild forms of α-mannosidosis. Am. J. Hum. Genet., 63, 1015–1024. 18. Tollersrud, O.K., Berg, T., Healy, P., Evjen, G., Ramachandran, U. and Nilssen, O. (1997) Purification of bovine lysosomal α-mannosidase, characterization of its gene and determination of two mutations that cause αmannosidosis. Eur. J. Biochem., 246, 410–419. 19. Neufeld, E.F. and Muenzer, J. (1995) Disorders of glycoprotein degradation and structure. In Scriver, C.R., Beaudet, A.L., Sly, W.S. and Valle, D. (eds), The Metabolic and Molecular Bases of Inherited Disease. McGrawHill, New York, Vol. II, pp. 2529–2561. 20. Burditt, L.J., Chtai, K., Hirani, S., Nugent, P.G., Winchester, B. and Blakemore, W.F. (1980) Biochemical studies on a case of feline mannosidosis. Biochem. J., 189, 467–473. 21. Walkley, S.U., Thuall, M.A., Dobrenis, K., Huang, M., March, P.A., Siegel, D.A. and Wurzelmann, S. (1994) Bone marrow transplantation corrects the enzyme defect in neurons of the central nervous system in a lysosomal storage disease. Proc. Natl Acad. Sci. USA, 91, 2970–2974. 22. Barton, N.W., Brady, R.O., Dambrosia, J.M., Di Bisceeglie, A.M., Doppelt, S.H., Hill, S.C., Mankin, H.J., Murray, G.J., Parker, R.I., Argoff, C.E., Grewal, R.P., Yu, K.-T. et al. (1991) Replacement therapy for inherited enzyme deficiency–macrophage-targeted glucocerebrosidase for Gaucher’s disease. N. Engl. J. Med., 324, 1464–1470. 23. Kay, M.A. and Woo, L.C. (1994) Gene therapy for metabolic disorders. Trends Genet., 10, 253–257. 24. Hocking, J.D., Jolly, R.D. and Batt, R.D. (1972) Deficiency of α-mannosidase in Angus cattle. An inherited lysosomal storage disease. Biochem. J., 128, 69–78. 25. Trowbridge, I.S. (1974) Isolation and chemical characterisation of a mitogenic lectin from Pisum sativum. J. Biol. Chem., 249, 6004–6012. 26. Howard, I.K., Sage, H.J., Stein, M.D., Young, N.M., Leon, M.A., Dyckes, D.F. (1971) Studies on a phytohemagglutinin from the lentil. II. Multiple forms of Lens culinaris hemagglutinin. J. Biol. Chem., 246, 1590–1595. 27. Saftig, P., Hartmann, D., Lüllmann-Rauch, R., Wolff, J., Evers, M., Köster, A., Hetman, M., von Figura, K. and Peters, C. (1997) Mice deficient in lysosomal acid phosphatase develop lysosomal storage in the kidney and central nervous system. J. Biol. Chem., 272, 18628–18635. 28. Meikle, P.J., Brooks, D.A., Ravenscroft, E.M., Yan, M., Williams, R.E., Jaunzems, A.E., Chataway, T.K., Karageorgos, L.E., Davey, R.C., Boulter, C.D., Carlsson, S.R. and Hopwood, J.J. (1997) Diagnosis of lysosomal storage disorders: evaluation of the lysosome-associated membrane protein LAMP-1 as a diagnostic marker. Clin. Chem., 43, 1325– 1335.

29. Gieselmann, V., Matzner, U., Hess, B., Lüllmann-Rauch, R., Coenen, R., Hartmann, D., D’Hooge, R., DeDeyn, P. and Nagels, G. (1998) Metachromatic leukodystrophy: molecular genetics and an animal model. J. Inherited Metab. Dis., 21, 564–574. 30. Evers, M., Saftig, P., Schmidt, P., Hafner, A., McLoghlin, D., Schmahl, W., von Figura, K. and Peters, C. (1996) Targeted disruption of the arylsulfatase B gene results in mice resembling the phenotype of mucopolysaccharidosis VI. Proc. Natl Acad. Sci. USA, 93, 8214–8219. 31. Hess, B., Saftig, P., Hartmann, D., Coenen, R., Lüllmann-Rauch, R., Goebel, H., Evers, M., von Figura, K., D’Hooge, R., Nagels, G., De Deyn, P., Peters, C. and Gieselmann, V. (1996) Phenotype of arylsulfatase Adeficient mice: relationship to human metachromatic leukodystrophy. Proc. Natl Acad. Sci. USA, 93, 14821–14826. 32. Haskins, M.E., Jezyk, P.F., Desnick, R.J., McDonough, S.K. and Patterson, D.F. (1979) Alpha-L-iduronidase deficiency in a cat: a model of mucopolysaccharidosis I. Pediatr. Res., 13, 1294–1297. 33. Yamanaka, S., Johnson, M.D., Grinberg, A., Westphal, H., Crawley, J.N., Taniike, M., Suzuki, K. and Proia, R.L. (1994) Targeted disruption of the Hexa gene results in mice with biochemical and pathologic features of Tay–Sachs disease. Proc. Natl Acad. Sci. USA, 91, 9975–9979. 34. Phaneuf, D., Wakamatsu, N., Huang, J.Q., Borowski, A., Peterson, A.C., Fortunato, S.R., Ritter, G., Igdoura, S.A., Morales, C.R., Benoit, G., Akerman, B.R., Leclerc, D., Hanai, N., Marth, J.D., Trasler, J.M. and Gravel, R.A. (1996) Dramatically different phenotypes in mouse models of human Tay–Sachs and Sandhoff diseases. Hum. Mol. Genet., 5, 1–14. 35. Hahn, C.N., del Pilar Martin, M., Zhou, X.Y., Mann, L.W. and d’Azzo, A. (1998) Correction of murine galactosialidosis by bone marrow-derived macrophages overexpressing human protective/cathepsin A under control of the colony-stimulating factor-1 receptor promoter. Proc. Natl Acad. Sci. USA, 95, 14880–14885. 36. Vogler, C., Sands, M.S., Galvin, N., Levy, B., Thorpe, C., Barler, J. and Sly, W.S. (1998) Murine mucopolysaccharidosis type VII: the impact of therapies on the clinical course and pathology in a murine model of lysosomal storage disease. J. Inherited Metab. Dis., 21, 575–586. 37. Thomas, K.R. and Capecchi, M.R. (1987) Site-directed mutagenesis by gene targeting in mouse embryo-derived stem cells. Cell, 51, 503–512. 38. Köster, A., Saftig, P., Matzner, U., von Figura, K., Peters, C. and Pohlmann, R. (1993) Targeted disruption of the M(r) 46,000 mannose 6-phosphate receptor gene in mice results in misrouting of lysosomal proteins. EMBO J., 12, 5219–5223. 39. Isbrandt, D., Arlt, G., Brooks, D.A., Hopwood, J.J., von Figura, K. and Peters, C. (1994) Mucopolysaccharidosis VI (Maroteaux–Lamy syndrome): six unique arylsulfatase B gene alleles causing variable disease phenotypes. Am. J. Hum. Genet., 54, 454–463. 40. Lyons, K., Graycar, J.L., Lee, A., Hashmi, S., Lindquist, P.B., Chen, E.Y., Hogan, B.L.M. and Derynck, R. (1989) Vgr-1, a mammalian gene related to Xenopus Vg-1, is a member of the transforming growth factor beta gene superfamily. Proc. Natl Acad. Sci. USA, 86, 4554–4558. 41. Wessel, D. and Flügge, U.I. (1984) A method for the quantitative recovery of protein in dilute solution in the presence of detergents and lipids. Anal. Biochem., 138, 141–143. 42. Monsigny, M., Petit, C. and Roche, A.C. (1988) Colorimetric determination of neutral sugars by a resorcinol sulfuric acid micromethod. Anal. Biochem., 175, 525–530. 43. Sewell, A.C. (1979) An improved thin-layer chromatographic method for urinary oligosaccharide screening. Clin. Chim. Acta, 92, 411–414. 44. Kasper, D., Dittmer, F., von Figura, K. and Pohlmann, R. (1996) Neither type of mannose 6-phosphate receptor is sufficient for targeting of lysosomal enzymes along intracellular routes. J. Cell Biol., 134, 615–623.