GM₁-gangliosidosis is a lysosomal storage disease that is inherited as an autosomal recessive disorder, predominantly caused by structural defects in the ß-galactosidase gene (GLB1). The molecular cause of GM₁-gangliosidosis in Alaskan huskies was investigated and a novel 19-bp duplication in exon 15 of the GLB1 gene was identified. The duplication comprised positions +1688–+1706 of the GLB1 cDNA. It partially disrupted a potential exon splicing enhancer (ESE), leading to exon skipping in a fraction of the transcripts. Thus, the mutation caused the expression of two different mRNAs from the mutant allele. One transcript contained the complete exon 15 with the 19-bp duplication, while the other transcript lacked exon 15. In the transcript containing exon 15 with the 19-bp duplication a premature termination codon (PTC) appeared, but due to its localization in the last exon of canine GLB1, nonsense-mediated RNA decay (NMD) did not occur. As a consequence of these molecular events two different truncated GLB1 proteins are predicted to be expressed from the mutant GLB1 allele. In heterozygous carrier animals the wild-type allele produces sufficient amounts of the active enzyme to prevent clinical signs of disease. In affected homozygous dogs no functional GLB1 is synthesized and G_M1-gangliosidosis occurs.

---

THE canine lysosomal acidic ß-galactosidase (GLB1, EC 3.2.1.23) is an exoglycosidase that removes ß-ketosidically linked galactose residues from glycoproteins, sphingolipids, and keratan sulfate (VAN DER SPOEL et al. 2000). GM₁-gangliosidosis is a lysosomal storage disease, inherited as an autosomal recessive disorder, predominantly caused by structural defects in the ß-galactosidase gene (GLB1) (THOMAS and BEAUDET 1995; CALLAHAN 1999). Mutations in the GLB1 gene were identified in Portuguese waterdogs with GM₁-gangliosidosis (WANG et al. 2000) and in Shiba inus with GM₁-gangliosidosis (YAMATO et al. 2002). The GLB1 gene is located on chromosome 3p21 in humans (NCBI MapViewer, human genome build 35.1) and on chromosome 23 in dog (PRIAT et al. 1998; BREEN et al. 2001). Both orthologous GLB1 genes contain 16 exons and share 86.5% identity at the nucleotide level and 81% identity at the amino acid level (WANG et al. 2000). In both species GLB1 is synthesized as an 85-kDa precursor protein, which is subsequently processed into a 64- to 66-kDa mature form and a 22- to 24-kDa cleavage fragment (VAN DER SPOEL et al. 2000). Comparative studies carried out on human, mouse, and bovine GLB1 revealed that the released 22- to 24-kDa proteolytic fragment remains associated to the 64- to 66-kDa chain to form the catalytically active ß-galactosidase (D'AZZO et al. 1982; BOUSTANY et al. 1993). The significance of the 22- to 24-kDa C-terminal GLB1 domain, encoded partially by exons 15 and 16, is supported by the identification of several amino acid substitutions in different forms of GM₁-gangliosidosis in canine (WANG et al. 2000; YAMATO et al. 2002) and human patients (BOUSTANY et al. 1993; MORRONE et al. 2000; VAN DER SPOEL et al. 2000).

Many disease-associated mutations affect pre-mRNA splicing, usually causing incorrect exon assembly (CARTEGNI et al. 2002). Up to 15% of point mutations responsible for genetic diseases in humans cause aberrant splicing (KRAWCZAK et al. 1992). The most common consequence of these mutations is exon skipping. In constitutive splicing all exons are included in the mature mRNA, whereas in the skipped pattern one or more exons are missing. The regulation of this process is still not very well understood; so far, cis-regulatory elements such as exonic splicing enhancers (ESEs) were mostly identified in individual cases (SCHAAL and MANIATIS 1999; BLACK 2003; FAUSTINO and COOPER 2003). Analysis of candidate sequences demonstrated that purine-rich motifs (GGAGA/GGGA/AGAGA) and CA-rich consensus motifs (C)CACC(C) are frequently used as splicing enhancer elements (DU et al. 1997; LIU et al. 1998; ROMANO et al. 2002; MIRIAMI et al. 2003; FAIRBROTHER et al. 2004).

In a previous study we reported clinical and pathological findings in Alaskan huskies with GM₁-gangliosidosis (MüLLER et al. 1998, 2001). The objective of the present study was to identify the causative genetic defect. Therefore, the GLB1 gene and GLB1 mRNA processing were investigated in a family of Alaskan huskies with GM₁-gangliosidosis.

Animals:

A large pedigree of Alaskan huskies segregating for GM₁-gangliosidosis was available (MüLLER et al. 2001). Affected and related heterozygous carrier animals previously classified by clinical, biochemical, and pathological investigations were used for nucleic acid isolation (Figure 1). RNA from DH82 cells (ATCC CRL-10389) and canine primary skin fibroblasts from two healthy control dogs (Airedale terrier C₁ and dachshund C₂) were used as controls in RT-PCR experiments.

View larger version (29K): In this window In a new window Download PPT slide   FIGURE 1.— Animals and results of the mutation analysis. (A) Alaskan husky pedigree used in this study (modified from MüLLER et al. 2001). (B) RT-PCR amplification of the complete coding region of GLB1 mRNAs with primers Ex1_F and Ex16_R. Samples are numbered according to the pedigree in A. The sizes of the RT-PCR products are indicated beneath the lanes. Note that in the heterozygous dog (1) three different mRNA species are expressed. The agarose gel does not resolve the 2052-bp and the 2071-bp bands; however, several clones from each band were sequenced to confirm the data. M, 250-bp ladder. (C) Mutation analysis of exon 15 on mRNA. Primers Ex15_F and Ex15_R were used for RT-PCR to detect the 19-bp duplication on the mRNA/cDNA level. The presence of the 19-bp duplication in exon 15 of the GLB1 gene leads to a 202-bp product while the wild-type product is only 183 bp. M, 100-bp ladder; neg., negative control without cDNA; DH, DH82 cells from a normal dog; C1 and C2, normal control dogs. (D) Mutation analysis on genomic DNA. PCR was again performed with primers Ex15_F and Ex15_R on genomic DNA. C1, normal control animal; M, 100-bp ladder. (E) Schematic of the 19-bp duplication in exon 15 of the canine GLB1 gene that leads to GM1-gangliosidosis in Alaskan huskies. The position of the diagnostic PCR primers Ex15_F and Ex15_R for this mutation is indicated.

 

DNA analysis:

Genomic DNA was isolated from canine skin fibroblast cell cultures using the Puregene kit (Biozym, Germany). All enzymatic reactions were carried out in a DNA programmable thermal controller (PTC-200; MJ Research, Cambridge, MA). PCR reactions were performed using primers Ex15_F and Ex15_R (Table 1) and the KOD Hot Start DNA polymerase (Novagen, Darmstadt, Germany). The PCR amplifies a 182-bp product within exon 15 of the GLB1 gene (Table 1). The resulting PCR products were cloned and sequenced to confirm their identity. The sequencing reactions were performed by MWG Biotech (Ebersberg, Germany).

View this table: In this window In a new window   TABLE 1 Primers used for amplification of canine GLB1 fragments

 

RNA analysis:

Mutation analysis of the whole coding region of the canine GLB1 cDNA was performed on total RNA isolated from canine skin fibroblasts using TRIzol Reagent (Invitrogen, Carlsbad, CA). The SMART-RACE method (Clontech, Palo Alto, CA) was used for cDNA synthesis. For first-strand cDNA synthesis 100 ng total RNA and SuperScript II reverse transcriptase (Invitrogen) were used. Second-strand synthesis was performed using 10 µl first-strand cDNA and Advantage GC PCR polymerase mix (Clontech). The double-stranded cDNA obtained was purified with the QIAquick PCR purification kit (QIAGEN, Hilden, Germany). In subsequent 25-µl PCR reactions 100 ng double-stranded cDNA, primers Ex1_F and Ex16_R (Table 1), and the KOD Hot Start polymerase were used. The resulting RT-PCR products were cloned and sequenced (two independent clones each) or directly sequenced.

Sequence analysis and bioinformatics:

Sequences were assembled and analyzed for mutations with Sequencher 4.2 (Genecodes, Ann Arbor, MI). To predict the possible localization of SR-protein-specific putative ESEs, a web-based tool, the ESEfinder Release 2.0, was used (http://rulai.cshl.edu/tools/ESE/). The default threshold values were set as the median of the highest score for each sequence in a set of 30 randomly chosen 20-nucleotide sequences previously used for functional SELEX (CARTEGNI et al. 2003).

At the beginning of the study a large pedigree of Alaskan huskies segregating for GM₁-gangliosidosis was available (Figure 1A). We had previously measured the biochemical activity of GLB1 in these dogs, which confirmed that the affected dogs had very little GLB1 activity (MüLLER et al. 1998, 2001). Therefore, we started to investigate whether mutations in the GLB1 gene are causative for this defect. We amplified full-length cDNAs from dogs with the three different postulated genotypes (homozygous mutant, heterozygous, and homozygous wild type). Surprisingly, when the RT-PCR products were separated on agarose gels, a homozygous mutant and a heterozygous animal seemingly showed the same pattern of two different bands, while only the homozygous wild-type animal showed a single band of the expected size (Figure 1B). Sequencing of these RT-PCR products revealed that the smaller band seen in the affected and heterozygous dogs corresponded to a transcript, in which exon 15 was missing. Sequencing of the larger RT-PCR product in the affected dog showed that this band corresponded to a full-length GLB1 transcript; however, 19 bp within the region encoded by exon 15 were duplicated in transcripts from the affected dog. This duplication was additionally confirmed by direct RT-PCR and PCR analyses of exon 15 (Figure 1, C and D). The duplication corresponded to positions +1688–+1706 in the canine GLB1 cDNA and had the sequence 5'-TCCCAGACTTGCCCCAGGA-3' (Figure 1E). To ensure that the observed mutation is indeed the causative pathological mutation, a sample of 30 normal dogs of different breeds was tested for the presence of the 19-bp duplication. The duplication was not present on any of the 60 chromosomes studied.

Our results indicate that the 19-bp duplication in exon 15 leads to exon skipping in a portion of the GLB1 transcripts and two different mRNA species are expressed from the mutant allele (Figure 2). The results of the affected dog were confirmed in the heterozygous dog, where the presence of three different mRNA species was established after DNA sequencing. Thus the DNA sequencing of the transcripts rectified the unexpected agarose gel picture from the heterozygous dog, where only two bands had been discernible initially (Figure 1B). The heterozygous dog expressed three different GLB1 mRNAs. In addition to the normal GLB1 mRNA from the wild-type allele, heterozygous dogs also express mRNAs lacking exon 15 and full-length mRNAs with the 19-bp duplication from the mutant allele.

View larger version (17K): In this window In a new window Download PPT slide   FIGURE 2.— Schematic of the splicing events in dogs with the GLB1 mutation. In the top part of the figure the genomic organization of the 16-exon GLB1 gene is depicted. Note that exon and intron sizes are not drawn to scale. The newly identified 19-bp duplication is located in the 255-bp exon 15. In the middle part the 3'-end of the primary transcript is represented. The primary transcript is spliced into two different mRNAs (bottom). One mRNA corresponds to the expected message with exon 15 containing 19 extra nucleotides from the genomic duplication. The other mRNA species lacks the entire exon 15. In the mRNA containing exon 15 the 19-bp duplication leads to a frameshift that causes the apparition of a PTC in exon 16. As the PTC is located in the last exon the mRNA is not subject to NMD and is expressed in amounts similar to that of the wild-type RNA. The frameshift mutation leads to the truncation of 78 amino acids from the C terminus of the GLB protein (indicated by shading of the truncated part of the ORF). The transcript variant without exon 15 leads to the translation of a protein lacking 85 amino acids encoded by this exon.

 
The 19-bp duplication within exon 15 is predicted to have dramatic consequences for the GLB1 protein. The exon skipping of the 255-bp exon 15 leads to the deletion of 85 amino acids in the functionally important C-terminal region of GLB1. This causes the loss of post-translational processing sites including the proteolytic cleavage site, necessary for transformation of precursor into mature acidic GLB1 protein.

In the other aberrant transcript variant, which contains exon 15 with the duplication, the insertion of 19 additional nucleotides leads to a frameshift and the apparition of a premature termination codon (PTC). As the PTC resides in the last exon, the transcript is not subject to nonsense-mediated decay (NMD). The band intensities of the gels with the RT-PCR products suggest that it is present in roughly comparable amounts to the transcript lacking exon 15. Because of the PTC, the transcript containing exon 15 with the 19-bp duplication gives rise to a protein that is 78 amino acids shorter than the wild-type GLB1 and additionally has an entirely different C-terminal sequence. Both transcripts from the mutated allele are thus likely to encode completely nonfunctional GLB1 proteins.

In this study a novel 19-bp duplication in exon 15 of the canine GLB1 gene was identified as a causative mutation for GM₁-gangliosidosis in Alaskan huskies. The mutation corresponded to nucleotides +1688–+1706 of the GLB1 cDNA. The GLB1 mutation causing GM₁-gangliosidosis in Alaskan huskies is thus different from earlier described GLB1 mutations in Portuguese waterdogs and Shiba inus (WANG et al. 2000; YAMATO et al. 2002) The Portuguese waterdog mutation is a point mutation (R60H) whereas the Shiba inu mutation represents a 1-bp deletion leading to a frameshift and PTC.

The newly identified GLB1 mutation causes very peculiar aberrations in RNA processing. In animals carrying the 19-bp duplication normal and abnormal splicing of GLB1 pre-mRNA coexist. The 19-bp duplication interferes with the recognition of exon 15 leading to skipping of this exon in a fraction of all transcripts, while the remaining transcripts are normally spliced. Exon recognition is becoming more and more accepted as an important mechanism in the splicing of mammalian transcripts (BERGET 1995). The mechanism by which exons are recognized or defined is not completely understood. So far, it seems clear that there is an interaction between splicing factors across exons and that these interactions are often mediated by auxiliary proteins such as SR-proteins that bind to the exon sequence itself (BERGET 1995; BIANCA and HERTEL 2002). The in silico analysis of potential binding sites for SR-proteins in exon 15 of the canine GLB1 gene revealed that the 19-bp duplication leads to the presence of potential additional binding sites for SF2/ASF, SC35, and SRp40, respectively. Thus, it seems conceivable that the precise binding of one or more of these proteins is essential for the correct exon recognition of exon 15. This implies a hypothetical explanation for the peculiar observation that two different mRNAs are produced from the mutant allele: In pre-mRNAs with the 19-bp duplication a splicing factor might bind to one of two alternative binding sites. However, only one of the two alternative binding sites is in the correct distance to the 3'-end of exon 15 and only binding to this site leads to inclusion of exon 15 in the mature transcript. If the binding site with the wrong distance to the 3'-end of exon 15 is occupied this would lead to exon skipping. As the sequences in the two duplicated 19-bp segments are identical the chance that a potential binding protein binds to the correct binding site is 50%, which corresponds to the experimental observation that the amounts of the two transcripts are roughly equal in fibroblasts.

In conclusion, we describe a novel 19-bp mutation in the canine GLB1 gene that causes interesting aberrations of the normal splicing process. The molecular elucidation of GM₁-gangliosidosis in Alaskan huskies provides another model for human GM₁-gangliosidosis. Knowledge of the causative mutation for GM₁-gangliosidosis in Alaskan huskies will also allow genetic testing of these dogs to avoid accidental matings of carrier animals. Thus breeders will be enabled to eliminate this disease from their breeding lines.

The authors thank Werner Hecht for very useful advice. This work was supported by a German Research Council [Deutsche Forschungsgemeinschaft (DFG)] grant from the Graduate College (GK 455 "Molecular Veterinary Medicine") and by DFG grant Ba815/7-1.

BERGET, S. M., 1995 Exon recognition in vertebrate splicing. J. Biol. Chem. 270: 2411–2414.[Free Full Text]

BIANCA, J. L., and K. J. HERTEL, 2002 A general role for splicing enhancers in exon definition. RNA 8: 1230–1241.

BLACK, D. L., 2003 Mechanisms of alternative pre-messenger RNA splicing. Annu. Rev. Biochem. 72: 291–336.[CrossRef][Medline]

BOUSTANY, R. M., W. H. QIAN and K. SUZUKI, 1993 Mutations in acid ß-galactosidase cause GM₁-gangliosidosis in American patients. Am. J. Hum. Genet. 53: 881–888.[Medline]

BREEN, M., S. JOUQUAND, C. RENIER, C. S. MELLERSH, C. HITTE et al., 2001 Chromosome-specific single-locus FISH probes allow anchorage of an 1800-marker integrated radiation-hybrid/linkage map of the domestic dog genome to all chromosomes. Genome Res. 11: 1784–1795.[Abstract/Free Full Text]

CALLAHAN, J. W., 1999 Molecular basis of GM₁-gangliosidosis and Morquio disease type B. Structure-function studies of lysosomal ß-galactosidase like protein. Biochim. Biophys. Acta 1455: 83–105.

CARTEGNI, L., S. L CHEW and A. R. KRAINER, 2002 Listening to silence and understanding nonsense: exonic mutations that affect splicing. Nat. Genet. 3: 285–298.[CrossRef][Medline]

CARTEGNI, L., J. WANG, Z. ZHU, M. Q. ZHANG and A. R. KRAINER, 2003 ESEfinder: a web resource to identify exonic splicing enhancers. Nucleic Acids Res. 31: 3568–3571.[Abstract/Free Full Text]

D'AZZO, A., A. T. HOOGEVEEN, A. J. REUSER, D. ROBINSON and H. GALJAARD, 1982 Molecular defect in combined GLB1 and neuraminidase deficiency in man. Proc. Natl. Acad. Sci. USA 79: 4535–4539.[Abstract/Free Full Text]

DU, K., Y. PENG, L. E. GREENBAUM, B. A. HABER and R. TAUB, 1997 HRS/SRp-40 mediated inclusion of fibronectin EIIIB exon, a possible cause of increased EIIIB expression in proliferating liver. Mol. Cell. Biol. 17: 4096–4104.[Abstract/Free Full Text]

FAIRBROTHER, W. G., D. HOLSTE, C. B. BURGE and P. A. SHARP, 2004 Single nucleotide polymorphism-based validation of exonic splicing enhancers. PLoS Biol. 2: E268.[CrossRef][Medline]

FAUSTINO, N. A., and T. A. COOPER, 2003 Pre-mRNA splicing and human disease. Genes Dev. 17: 419–437.[Free Full Text]

KRAWCZAK, M., J. REISS and D. N. COOPER, 1992 The mutational spectrum of single base-pair substitutions in mRNA splice junctions of human genes: causes and consequences. Hum. Genet. 90: 41–54.[Medline]

LIU, H. X., M. ZHANG and A. R. KRAINER, 1998 Identification of functional exonic splicing enhancer motifs recognized by individual SR proteins. Genes Dev. 12: 1998–2012.[Abstract/Free Full Text]

MIRIAMI, E., H. MARGALIT and R. SPERLING, 2003 Conserved sequence elements associated with exon skipping. Nucleic Acids Res. 31: 1974–1983.[Abstract/Free Full Text]

MORRONE, A., T. BARDELLI, M. A. DONATI, M. GIORGI, M. DI ROCCO et al., 2000 GLB1 gene mutations affecting the lysosomal enzyme and the elastin-binding protein in GM₁-gangliosidosis patients with cardiac involvement. Hum. Mutat. 15: 354–366.[CrossRef][Medline]

MüLLER, G., W. BAUMGäRTNER, A. MORITZ, A. SEWELL and B. KUSTERMANN-KUHN, 1998 Biochemical findings in a breeding colony of Alaskan huskies suffering from GM₁-gangliosidosis. J. Inherit. Metab. Dis. 21: 430–431.[CrossRef][Medline]

MüLLER, G., S. ALLDINGER, A. MORITZ, A. ZURBRIGGEN, N. KIRCHHOF et al., 2001 GM₁-gangliosidosis in Alaskan huskies: clinical and pathologic findings. Vet. Pathol. 38: 281–290.[Medline]

PRIAT, C., C. HITTE, F. VIGNAUX, C RENIER, Z. JIANG et al., 1998 A whole-genome radiation hybrid map of the dog genome. Genomics 54: 361–378.[CrossRef][Medline]

ROMANO, M., R. MARCUCCI, E. BURATTI, Y. M. AYALA, G. SEBASTIO et al., 2002 Regulation of 3' splice site selection in the 844ins68 polymorphism of the cystathionine ß-synthase gene. J. Biol. Chem. 277: 43821–43829.[Abstract/Free Full Text]

SCHAAL, T. D., and T. MANIATIS, 1999 Selection and characterization of pre-mRNA splicing enhancers: identification of novel SR protein-specific enhancer sequences. Mol. Cell. Biol. 19: 1705–1719.[Abstract/Free Full Text]

THOMAS, G. H., and A. L. BEAUDET, 1995 Lysosomal disorders, pp. 2529–2561 in The Metabolic and Molecular Bases of Inherited Diseases, Ed. 7, edited by C. R. SCRIVER, A. L. BEAUDET, W. S. SLY and D. VALLE. McGraw Hill, New York.

VAN DER SPOEL, A., E. BONTEN and A. D'AZZO, 2000 Processing of lysosomal ß-galactosidase. J. Biol. Chem. 275: 10035–10040.[Abstract/Free Full Text]

WANG, Z. H., B. ZENG, H. SHIBUYA, G. S. JOHNSON, J. ALROY et al., 2000 Isolation and characterization of the normal canine GLB1 gene and its mutation in a dog model of GM₁-gangliosidosis. J. Inherit. Metab. Dis. 23: 593–606.[Medline]

YAMATO, O., D. ENDOH, A. KOBAYASHI, Y. MASUOKA, M. YONEMURA et al., 2002 A novel mutation in the gene for canine acid beta-galactosidase that causes GM₁-gangliosidosis in Shiba dogs. J. Inherit. Metab. Dis. 25: 525–526.[Medline]

Communicating editor: G. GIBSON

International Access Link

Online ISSN: 1943-2361  Print ISSN: 0016-6731
Copyright 2009 by the Genetics Society of America
phone: 412-268-1812   fax: 412-268-1813   email: genetics-gsa{at}andrew.cmu.edu