Portal de Periódicos da CAPES

Alternative Splicing in the α-Galactosidase A Gene: Increased Exon Inclusion Results in the Fabry Cardiac Phenotype

2002; Elsevier BV; Volume: 70; Issue: 4 Linguagem: Inglês

10.1086/339431

1537-6605

Seiichi Ishii, Shoichiro Nakao, Reiko Minamikawa‐Tachino, Robert J. Desnick, Jianqiang Fan,

Trypanosoma species research and implications

Fabry disease is an inborn error of glycosphingolipid catabolism, resulting from deficient activity of lysosomal α-galactosidase A (α-Gal A). A rare alternative splicing that introduces a 57-nucleotide (nt) intronic sequence to the α-Gal A transcript from intron 4 of the gene has been identified. In addition, a novel midintronic base substitution that results in substantially increased alternative splicing has been identified in a patient with Fabry disease who has the cardiac variant phenotype. The sequence of the patient's intron 4 contains a single G→A transversion at genomic nt 9331 (IVS4+919G→A), located at the −4 position of the 3′ end of the intronic insertion (nts 9278–9334 in the genomic sequence). Minigene constructs containing the entire intron 4 sequence with G, A, C, or T at nt 9331 within an α-Gal A complementary DNA expression vector were prepared and expressed in COS-1 cells. Whereas transfection of the G or T minigenes transcribed predominantly normal-sized transcripts, the transfection of the A or C minigenes produced a large amount of the alternatively spliced transcript. These results suggest that the G→A mutation, within an A/C-rich domain, results in increased recognition of the alternative splicing by an A/C-rich enhancer–type exonic splicing enhancer. The intronic mutation was not observed in 100 unrelated unaffected men but was present in 6 unrelated patients with cardiac Fabry disease. Reverse-transcriptase polymerase chain reaction of total RNA of various normal human tissues revealed that the alternatively spliced transcript was present in all of the samples, and especially at a higher ratio in the lung and muscle. The normal transcript was present in the patients' lymphoblasts and resulted in ∼10% residual enzyme activity, leading to a cardiac phenotype of Fabry disease. Fabry disease is an inborn error of glycosphingolipid catabolism, resulting from deficient activity of lysosomal α-galactosidase A (α-Gal A). A rare alternative splicing that introduces a 57-nucleotide (nt) intronic sequence to the α-Gal A transcript from intron 4 of the gene has been identified. In addition, a novel midintronic base substitution that results in substantially increased alternative splicing has been identified in a patient with Fabry disease who has the cardiac variant phenotype. The sequence of the patient's intron 4 contains a single G→A transversion at genomic nt 9331 (IVS4+919G→A), located at the −4 position of the 3′ end of the intronic insertion (nts 9278–9334 in the genomic sequence). Minigene constructs containing the entire intron 4 sequence with G, A, C, or T at nt 9331 within an α-Gal A complementary DNA expression vector were prepared and expressed in COS-1 cells. Whereas transfection of the G or T minigenes transcribed predominantly normal-sized transcripts, the transfection of the A or C minigenes produced a large amount of the alternatively spliced transcript. These results suggest that the G→A mutation, within an A/C-rich domain, results in increased recognition of the alternative splicing by an A/C-rich enhancer–type exonic splicing enhancer. The intronic mutation was not observed in 100 unrelated unaffected men but was present in 6 unrelated patients with cardiac Fabry disease. Reverse-transcriptase polymerase chain reaction of total RNA of various normal human tissues revealed that the alternatively spliced transcript was present in all of the samples, and especially at a higher ratio in the lung and muscle. The normal transcript was present in the patients' lymphoblasts and resulted in ∼10% residual enzyme activity, leading to a cardiac phenotype of Fabry disease. Fabry disease (MIM 301500) is an X-linked recessive inborn error of glycosphingolipid catabolism, caused by lysosomal α-galactosidase A (α-Gal A, or GLA; Enzyme Commission [EC] number 3.2.1.22) deficiency. The enzymatic defect results in the accumulation of neutral glycosphingolipids with terminal α-galactosyl residues—predominantly globotriaosylceramide—in body fluids and in tissue lysosomes (Desnick et al. Desnick et al., 2001Desnick RJ Ioannou YA Eng CM Fabry disease: α-galactosidase A deficiency.in: Scriver C Beaudet A Sly W Valle D The metabolic and molecular bases of inherited disease. 7th ed. McGraw-Hill, New York2001: 3733-3774Google Scholar). Classically affected men with no α-Gal A activity have angiokeratoma, acroparesthesias, hypohidrosis, and characteristic corneal and lenticular opacities. In contrast, those afflicted with a milder form have residual α-Gal A activity and do not have the classical manifestations but typically present with left-ventricular hypertrophy (Nakao et al. Nakao et al., 1995Nakao S Takenaka T Maeda M Kodama C Tanaka A Tahara M Yoshida A Kuriyama M Hayashibe H Sakuraba H Tanaka H An atypical variant of Fabry's disease in men with left ventricular hypertrophy.N Engl J Med. 1995; 333: 288-293Crossref PubMed Scopus (624) Google Scholar) and have late-onset cardiomyopathy (Elleder et al. Elleder et al., 1990Elleder M Bradova V Smid F Budesinsky M Harzer K Kustermann-Kuhn B Ledvinova J Cardiocyte storage and hypertrophy as a sole manifestation of Fabry's disease: report on a case simulating hypertrophic non-obstructive cardiomyopathy.Virchows Arch A Pathol Anat Histopathol. 1990; 417: 449-455Crossref PubMed Scopus (188) Google Scholar; Ogawa et al. Ogawa et al., 1990Ogawa K Sugamata K Funamoto N Abe T Sato T Nagashima K Ohkawa S-I Restricted accumulation of globotriaocylceramide in the hearts of atypical cases of Fabry's disease.Hum Pathol. 1990; 21: 1067-1073Abstract Full Text PDF PubMed Scopus (35) Google Scholar; von Scheidt et al. von Scheidt et al., 1991von Scheidt W Eng CM Fitzmaurice TF Erdmann E Hubner G Olsen EGJ Christomanou H Kandolf R Bishop DF Desnick RJ An atypical variant of Fabry's disease with manifestations confined to the myocardium.N Engl J Med. 1991; 324: 395-399Crossref PubMed Scopus (326) Google Scholar). To date, >190 α-Gal A mutations that cause Fabry disease have been described (Human Gene Mutation Database Web site), and all of these mutations are located within the coding or the flanking intronic sequences. Mutations that cause the classic phenotype include missense and nonsense mutations, large and small insertions or deletions, and splicing defects (Desnick et al. Desnick et al., 2001Desnick RJ Ioannou YA Eng CM Fabry disease: α-galactosidase A deficiency.in: Scriver C Beaudet A Sly W Valle D The metabolic and molecular bases of inherited disease. 7th ed. McGraw-Hill, New York2001: 3733-3774Google Scholar). In contrast, mutations that cause cardiac Fabry disease include missense mutations, a small in-frame deletion, and decreased α-Gal A mRNA (Desnick et al. Desnick et al., 2001Desnick RJ Ioannou YA Eng CM Fabry disease: α-galactosidase A deficiency.in: Scriver C Beaudet A Sly W Valle D The metabolic and molecular bases of inherited disease. 7th ed. McGraw-Hill, New York2001: 3733-3774Google Scholar). Among the mutations reported, all of the splicing mutations have been found in the flanking regions of exons, resulting in exon skipping or in misreading of the splice site (Yokoi et al. Yokoi et al., 1991Yokoi T Shinoda K Ohno I Kato K Miyawaki T Taniguchi N A 3′ splice site consensus sequence mutation in the intron 3 of the α-galactosidase A gene in a patient with Fabry disease.Jinrui Idengaku Zasshi. 1991; 36: 245-250Crossref PubMed Scopus (18) Google Scholar; Sakuraba et al. Sakuraba et al., 1992Sakuraba H Eng CM Desnick RJ Bishop DF Invariant exon skipping in the human α-galactosidase A pre-mRNA: Ag+1 to t substitution in a 5′-splice site causing Fabry disease.Genomics. 1992; 12: 643-650Crossref PubMed Scopus (46) Google Scholar; Eng et al. Eng et al., 1993Eng CM Resnick-Silverman LA Niehaus DJ Astrin KH Desnick RJ Nature and frequency of mutations in the α-galactosidase A gene that cause Fabry disease.Am J Hum Genet. 1993; 53: 1186-1197PubMed Google Scholar, Eng et al., 1997Eng CM Ashley GA Burgert TS Enriquez AL D'Souza M Desnick RJ Fabry disease: thirty-five mutations in the α-galactosidase A gene in patients with classic and variant phenotypes.Mol Med. 1997; 3: 174-182PubMed Google Scholar; Germain and Poenaru Germain et al., 1996Germain D Biasotto M Tosi M Meo T Kahn A Poenaru L Fluorescence-assisted mismatch analysis (FAMA) for exhaustive screening of the α-galactosidase A gene and detection of carriers in Fabry disease.Hum Genet. 1996; 98: 719-726Crossref PubMed Scopus (40) Google Scholar, Germain and Poenaru, 1999Germain DP Poenaru L Fabry disease: identification of novel α-galactosidase A mutations and molecular carrier detection by use of fluorescent chemical cleavage of mismatches.Biochem Biophys Res Commun. 1999; 257: 708-713Crossref PubMed Scopus (42) Google Scholar; Okumiya et al. Okumiya et al., 1996Okumiya T Takenaka T Ishii S Kase R Kamei S Sakuraba H Two novel mutations in the α-galactosidase gene in Japanese classical hemizygotes with Fabry disease.Jpn J Hum Genet. 1996; 41: 313-321Crossref PubMed Scopus (8) Google Scholar; Matsumura et al. Matsumura et al., 1998Matsumura T Osaka H Sugiyama N Kawanishi C Maruyama Y Suzuki K Onishi H Yamada Y Morita M Aoki M Kosaka K Novel acceptor splice site mutation in the invariant AG of intron 6 of α-galactosidase A gene, causing Fabry disease: mutation in brief no. 146. Online.Hum Mutat. 1998; 11: 483Crossref PubMed Google Scholar; Topaloglu et al. Topaloglu et al., 1999Topaloglu AK Ashley GA Tong B Shabbeer J Astrin KH Eng CM Desnick RJ Twenty novel mutations in the alpha-galactosidase A gene causing Fabry disease.Mol Med. 1999; 5: 806-811PubMed Google Scholar). During the course of mutant analysis of a patient with cardiac Fabry disease (proband 1) who had residual enzyme activity of 8.7 U/mg protein (9.1% of normal) in lymphoblasts, we were unable to identify any mutation in the exonic or flanking intronic regions of the α-Gal A gene. Therefore, we directed our efforts toward the determination of the responsible mutation, by means of RT-PCR of the RNA from the cardiac variant. Direct sequencing of the RT-PCR product revealed an insertion between exons 4 and 5 (data not shown). To further characterize the abnormal splicing, RT-PCR of the region that includes exons 4 and 5 was performed. Two bands were visualized on gel electrophoresis: the expected wild-type 215-bp fragment and a larger fragment, ∼270 base pairs in length (fig. 1A). This abnormal DNA fragment was also found in the normal cells, at a very low level. From the intensity of the DNA bands, it was estimated that the abnormal α-Gal A mRNA in the patient's lymphoblasts was ⩾70% of the total α-Gal A mRNA, whereas it was <5% of the total in normal lymphoblasts. These results suggested that the abnormal DNA fragment was generated by a rare splicing event within the α-Gal A gene. To determine the sequence of the abnormal fragment, the RT-PCR product generated from the patient's lymphoblast RNA through use of primer set 5′-GTCCTTGGCCCTGAATAG-3′ and 5′-GTCCAGCAACATCAACAATT-3′, which amplified a fragment that included exons 4 and 5, was subcloned into the TA cloning pCR vector (Invitrogen). Sequencing of the RT-PCR product revealed a 57-nucleotide (nt) insertion at the junction between exons 4 and 5 of the α-Gal A cDNA (fig. 1B), consistent with the observed size of the RT-PCR product from the patient's lymphoblasts (fig. 1A). This insertion corresponded to an intronic sequence in the gene at nts 9278–9334, which is located in the middle of intron 4. This in-frame insertion caused a premature termination at the 27th nucleotide downstream from exon 4. Therefore, the predicted product of the mutant α-Gal A mRNA was a truncated protein of 222 amino acid residues. To examine whether the truncated protein exhibited enzyme activity, an expression construct (pBaIns57) containing the mutant insertion between exons 4 and 5 was prepared and expressed in COS-1 cells. A truncated protein of 30 kD was detected by western blot analysis (fig. 2A). The enzyme activities in the mock-transfected COS-1 cells and COS-1 cells transfected with pBaIns57 were 153±3 U/mg and 147±3 U/mg, respectively, compared with 2,670±200 U/mg in cells transfected with the wild-type α-Gal A cDNA (pBaN; fig. 2B). These results indicated that the C-terminal truncated protein had no enzyme activity. To determine the ratio of mutant and total (both mutant and normal) α-Gal A mRNA in the patient's lymphoblasts, a quantitative competitive RT-PCR was performed. DNA competitors were prepared by PCR amplification of a λ-DNA provided by the Competitive DNA Construction Kit (Takara Shuzo), through use of sense and antisense primers 5′-GCTCCACACTATTTGGAAGTACGGTCATCATCTGACAC-3′ and 5′-AAAGGAGCAGCCATGATAGAACTATCGACATCATTACGC-3′, for specific determination of the mutant α-Gal A mRNA, or sense and antisense primers 5′-GGTTGGAATGACCCAGATAGTACGGTCATCATCTGACAC-3′ and 5′-TGCGATGGTATAAGAGCGCAGTTAATCGAACAAGAC-3′, for specific determination of total α-Gal A mRNA. A 298-bp PCR product or a 324-bp PCR product was amplified in the presence of the respective competitors to detect either mutant α-Gal A mRNA or both normal and mutant α-Gal A mRNAs, respectively (fig. 3A). The total α-Gal A mRNA amount was essentially the same in both the normal cells and the patient's cells (fig. 3B). On the other hand, the 298-bp mutant fragment was the predominant product in the patient's cells, whereas this fragment was present at a very low level in the normal cells (fig. 3C). On the basis of the intensity of the DNA bands, the mutant α-Gal A mRNA was estimated to be present at ∼115 copies, compared with the DNA competitor; this amount was ∼10-fold higher than that in the normal cells (10 copies) (fig. 3D). To determine whether the alternative splicing occurred in human tissues, RT-PCR between exons 4 and 5 was performed using total tissue RNA (fig. 4). The alternatively spliced transcript (the 272-bp fragment) was observed in a trace amount (<1% of total α-Gal A mRNA) in heart and small intestine, and a small amount ( 87 nts in length were included in the splice product, whereas complete skipping of an internal exon was observed when the exons were <33 nt in length. A 51-nt exon was processed in vitro to result in both skipped products and included products, and the importance of these elements has been shown experimentally in studies employing pre-mRNAs that contain point mutations or deletions within an existing sequence (Aebi et al. Aebi et al., 1986Aebi M Hornig H Padgett RA Reiser J Weissmann C Sequence requirements for splicing of higher eukaryotic nuclear pre-mRNA.Cell. 1986; 47: 555-565Abstract Full Text PDF PubMed Scopus (244) Google Scholar). In the present case, a slightly shorter sequence, generated by deletion of 3 bp from the 57-nt sequence, was exclusively skipped from the transcript (data not shown), suggesting that the length of 57-nt probably limits the inclusion of this sequence. This may be the reason that only low-level splicing of this intronic sequence (∼5% of α-Gal A mRNA) occurred in normal lymphoblasts and tissues. To evaluate whether the G→A transition at nt 9331 was responsible for the mutant α-Gal A mRNA in the patient's cells, minigene constructs containing the entire normal intron 4 sequence (pCXN2Gal/int4N) or the patient's intron 4 sequence (pCXN2Gal/int4P) were prepared and then expressed in COS-1 cells. To confirm that the pCXN2Gal/int4P construct contained the G→A transition, the minigene construct was digested with BfaI. Presence of a 427-bp fragment (nts 9116–9543) after the enzyme digestion indicated the presence of the mutation in pCXN2Gal/int4P (data not shown), since this fragment would be cleaved into two smaller fragments by the enzyme if the wild-type sequence was present. α-Gal A activity in COS-1 cells transfected with both normal and mutant constructs was detected, although the enzyme activity was substantially lower in the cells transfected with pCXN2Gal/int4P, suggesting that the minigene constructs were transcribed, spliced, and translated in the COS-1 cells. α-Gal A mRNA was analyzed by RT-PCR of total RNA isolated from the transfected COS-1 cells. A large amount of the 272-bp RT-PCR fragment was obtained from the COS-1 cells transfected with pCXN2Gal/int4P, indicating that alternative splicing of the α-Gal A transcript occurred in COS-1 cells (fig. 5B). This indicated that the IVS4+919G→A mutation is responsible for the drastic increase of the alternative splicing in intron 4. Intronic mutations have been found to cause many human diseases, including β-thalassemia (Treisman et al. Treisman et al., 1983Treisman R Orkin SH Maniatis T Specific transcription and RNA splicing defects in five cloned β-thalassaemia genes.Nature. 1983; 302: 591-596Crossref PubMed Scopus (383) Google Scholar), ornithine δ-aminotransferase deficiency (Mitchell et al. Mitchell et al., 1991Mitchell GA Labuda D Fontaine G Saudubray JM Bonnefont JP Lyonnet S Brody LC Steel G Obie C Valle D Splice-mediated insertion of an Alu sequence inactivates ornithine delta-aminotransferase: a role for Alu elements in human mutation.Proc Natl Acad Sci USA. 1991; 88: 815-819Crossref PubMed Scopus (115) Google Scholar), and ataxia-telangiectasia (Teraoka et al. Teraoka et al., 1999Teraoka SN Telatar M Becker-Catania S Liang T Onengut S Tolun A Chessa L Sanal O Bernatowska E Gatti RA Concannon P Splicing defects in the ataxia-telangiectasia gene, ATM: underlying mutations and consequences.Am J Hum Genet. 1999; 64: 1617-1631Abstract Full Text Full Text PDF PubMed Scopus (230) Google Scholar). These mutations often exist in consensus AG/GT dinucleotides, the branch point, or the polypyrimidine tract of a cryptic exon, resulting in inclusion of the cryptic exon or in changing of the normal splicing site upstream of the cryptic exon. In the present case, the IVS4+919G→A mutation that drastically changes the splicing of the 57-nt sequence is located within the alternative splicing exon. To assess the mechanism of the alternative splicing, additional minigenes with mutations at the −4 position and its surrounding region were generated by site-directed mutagenesis, and COS-1 cells were transfected with minigenes of α-Gal A cDNA containing intron 4 with IVS4+919G→C (pCXN2Gal/int4C) or IVS4+919G→T (pCXN2Gal/int4T) transversions. Only the IVS4+919G→C transversion, and not the G→T transversion, was shown to increase the inclusion of the 57-nt sequence (fig. 5B), indicating that an A or C at the position activates the alternative splicing. Exonic splicing enhancers (ESEs) that direct the specific recognition of splicing sites during constitutive and alternative splicing are found within both coding and noncoding exons (Blencowe Blencowe, 2000Blencowe BJ Exonic splicing enhancers: mechanism of action, diversity and role in human genetic diseases.Trends Biochem Sci. 2000; 25: 106-110Abstract Full Text Full Text PDF PubMed Scopus (516) Google Scholar). The most common feature of ESEs is that most are purine rich (Eldridge et al. Eldridge et al., 1999Eldridge AG Li Y Sharp PA Blencowe BJ The SRm160/300 splicing coactivator is required for exon-enhancer function.Proc Natl Acad Sci USA. 1999; 96: 6125-6130Crossref PubMed Scopus (64) Google Scholar). Sequences other than purine-rich repeats have also been characterized as exon enhancers. For example, novel A/C-rich enhancers (ACEs) have been identified both in natural genes and in experiments designed to select exon enhancers. The ACEs were reported from human calcitonin exon 4 (van Oers et al. van Oers et al., 1994van Oers CC Adema GJ Zandberg H Moen TC Baas PD Two different sequence elements within exon 4 are necessary for calcitonin-specific splicing of the human calcitonin/calcitonin gene-related peptide I pre-mRNA.Mol Cell Biol. 1994; 14: 951-960Crossref PubMed Scopus (49) Google Scholar), chicken cTNT exon 16 (Wang et al. Wang et al., 1995Wang Z Hoffmann HM Grabowski PJ Intrinsic U2AF binding is modulated by exon enhancer signals in parallel with changes in splicing activity.RNA. 1995; 1: 21-35PubMed Google Scholar), and Drosophila doublesex exon 4 (Ryner and Baker Ryner and Baker, 1991Ryner LC Baker BS Regulation of doublesex pre-mRNA processing occurs by 3′-splice site activation.Genes Dev. 1991; 5: 2071-2085Crossref PubMed Scopus (120) Google Scholar; Lynch and Maniatis Lynch and Maniatis, 1995Lynch KW Maniatis T Synergistic interactions between two distinct elements of a regulated splicing enhancer.Genes Dev. 1995; 9: 284-293Crossref PubMed Scopus (147) Google Scholar). Disruption of an ESE in the coding sequence can cause inappropriate exon skipping in vitro (Liu et al. Liu et al., 2001Liu HX Cartegni L Zhang MQ Krainer AR A mechanism for exon skipping caused by nonsense or missense mutations in BRCA1 and other genes.Nat Genet. 2001; 27: 55-58PubMed Google Scholar). In the 57-nt sequence, an A/C-rich sequence (nts 9323–9332: CCCCACUAGA) was observed in the alternative splicing exon within intron 4 of the α-Gal A gene. The IVS4+919G→A/C substitutions could further enrich the A/C predominance for the sequence and, thus, could increase the recognition of the alternative splice site. Therefore, the ACE recognition mechanism may be involved in the alternative splicing of the α-Gal A gene. ACE-type ESEs are not as common as purine-rich–type ESEs. Recently, the RNA-binding protein YB-1 was reported to bind ACEs and to stimulate the alternative splicing of the CD44 exon v4 (Stickeler et al. Stickeler et al., 2001Stickeler E Fraser SD Honig A Chen AL Berget SM Cooper TA The RNA binding protein YB-1 binds A/C-rich exon enhancers and stimulates splicing of the CD44 alternative exon v4.EMBO J. 2001; 20: 3821-3830Crossref PubMed Scopus (153) Google Scholar), although the detailed structural mechanism was not revealed. Our finding of the increased alternative splicing in α-Gal A gene could be valuable for the elucidation of the mechanism of the ACE enhancement. Mutations in another patient with Fabry disease (proband 2) with the cardiac variant and in four patients with Fabry disease of the cardiac phenotype (or atypical Fabry disease) previously characterized as expressing a decreased level of α-Gal A mRNA (Nakao et al. Nakao et al., 1995Nakao S Takenaka T Maeda M Kodama C Tanaka A Tahara M Yoshida A Kuriyama M Hayashibe H Sakuraba H Tanaka H An atypical variant of Fabry's disease in men with left ventricular hypertrophy.N Engl J Med. 1995; 333: 288-293Crossref PubMed Scopus (624) Google Scholar) were analyzed in the present study. Proband 2 had residual lymphoblast α-Gal A activity that was 10.6% of normal. The residual enzyme activities in other patients were 8%–15% of normal. BfaI digestion of the genomic fragment containing the substitution site did not cleave the fragment from these patients (fig. 6A), clearly indicating the presence of the G→A substitution. These results were confirmed by sequencing of the appropriate PCR products (data not shown). RT-PCR amplification of total RNA from these patients' lymphoblasts revealed a high level of mutant α-Gal A mRNA, detected as a 272-bp fragment (fig. 6B). The percentage of mutant α-Gal A mRNA was 64%–88%. In contrast, only 5%–8% of the alternatively spliced mRNA was detected in the lymphoblasts from unaffected subjects or from patients with Fabry disease who had missense mutations. These results indicated that the increased alternative splicing in these patients with cardiac Fabry disease was caused by the G→A substitution at nt 9331 and that the deficient enzyme activity was due to a decreased level of normal α-Gal A transcript. Accordingly, these patients should appropriately be classified as having the alternative splicing mutations. The cardiac variant of Fabry disease is usually associated with residual α-Gal A activity at 5%–10% of normal level. The residual enzyme activities in the patients with cardiac Fabry disease in the present study were determined to be ∼10% of normal. On the basis of RT-PCR amplification, the amount of mutant α-Gal A mRNA in the patients' lymphoblasts was 64%–88% of the total α-Gal A mRNA, and the product of the mutant α-Gal A mRNA was a truncated polypeptide of 222 amino acid residues that had no detectable enzymatic activity (fig. 2B). The remainder of the normal α-Gal A mRNA was transcribed to result in the 10% residual α-Gal A activity and cause the cardiac variant phenotype. The truncated α-Gal A polypeptide may be also translated from the small amount of the abnormal α-Gal A mRNA in normal cells. However, it may be rapidly degraded in situ, since we were unable to detect the truncated protein in the patient's lymphoblasts, although it was barely detected in COS-1 cells transfected with pBaIns57 (fig. 3A). Combining six unrelated patients who had cardiac Fabry disease retained the same mutation; this mutation may be considered to be a major mutation in Fabry disease, particularly in the patients who have no mutation in the coding and flanking regions. Until now, all splicing mutations reported in Fabry disease have been nucleotide substitutions or a single base deletion at or near the invariant dinucleotides of an intron, resulting in exon skipping or misreading of the splicing site (Yokoi et al. Yokoi et al., 1991Yokoi T Shinoda K Ohno I Kato K Miyawaki T Taniguchi N A 3′ splice site consensus sequence mutation in the intron 3 of the α-galactosidase A gene in a patient with Fabry disease.Jinrui Idengaku Zasshi. 1991; 36: 245-250Crossref PubMed Scopus (18) Google Scholar; Sakuraba et al. Sakuraba et al., 1992Sakuraba H Eng CM Desnick RJ Bishop DF Invariant exon skipping in the human α-galactosidase A pre-mRNA: Ag+1 to t substitution in a 5′-splice site causing Fabry disease.Genomics. 1992; 12: 643-650Crossref PubMed Scopus (46) Google Scholar; Eng et al. Eng et al., 1993Eng CM Resnick-Silverman LA Niehaus DJ Astrin KH Desnick RJ Nature and frequency of mutations in the α-galactosidase A gene that cause Fabry disease.Am J Hum Genet. 1993; 53: 1186-1197PubMed Google Scholar, Eng et al., 1997Eng CM Ashley GA Burgert TS Enriquez AL D'Souza M Desnick RJ Fabry disease: thirty-five mutations in the α-galactosidase A gene in patients with classic and variant phenotypes.Mol Med. 1997; 3: 174-182PubMed Google Scholar; Germain and Poenaru Germain et al., 1996Germain D Biasotto M Tosi M Meo T Kahn A Poenaru L Fluorescence-assisted mismatch analysis (FAMA) for exhaustive screening of the α-galactosidase A gene and detection of carriers in Fabry disease.Hum Genet. 1996; 98: 719-726Crossref PubMed Scopus (40) Google Scholar, Germain and Poenaru, 1999Germain DP Poenaru L Fabry disease: identification of novel α-galactosidase A mutations and molecular carrier detection by use of fluorescent chemical cleavage of mismatches.Biochem Biophys Res Commun. 1999; 257: 708-713Crossref PubMed Scopus (42) Google Scholar; Okumiya et al. Okumiya et al., 1996Okumiya T Takenaka T Ishii S Kase R Kamei S Sakuraba H Two novel mutations in the α-galactosidase gene in Japanese classical hemizygotes with Fabry disease.Jpn J Hum Genet. 1996; 41: 313-321Crossref PubMed Scopus (8) Google Scholar; Matsumura et al. Matsumura et al., 1998Matsumura T Osaka H Sugiyama N Kawanishi C Maruyama Y Suzuki K Onishi H Yamada Y Morita M Aoki M Kosaka K Novel acceptor splice site mutation in the invariant AG of intron 6 of α-galactosidase A gene, causing Fabry disease: mutation in brief no. 146. Online.Hum Mutat. 1998; 11: 483Crossref PubMed Google Scholar; Topaloglu et al. Topaloglu et al., 1999Topaloglu AK Ashley GA Tong B Shabbeer J Astrin KH Eng CM Desnick RJ Twenty novel mutations in the alpha-galactosidase A gene causing Fabry disease.Mol Med. 1999; 5: 806-811PubMed Google Scholar). The IVS4+919G→A mutation was unique because it occurred in the middle of an intron, and it increased the recognition of a normally weak splice site, resulting in the insertion of an additional sequence into the α-Gal A transcript, which leads to the cardiac phenotype of Fabry disease. In summary, we report for the first time that there is a normally weakly regulated alternative splice site in the human α-Gal A gene. In addition, we determined that a novel intronic mutation causes a remarkable increase in the alternatively spliced α-Gal A transcript and, consequently, results in the cardiac phenotype of Fabry disease. We thank Dr. E. Nanba (Gene Research Center, Tottori University, Japan) for supplying the DNA samples from unaffected Japanese men, and Mr. K. Inoue for excellent technical assistance. This work was supported, in part, by Ministry of Health and Welfare of Japan, Mizutani Foundation for Glycoscience grant 010092 and by American Heart Association grant AHA 0130522T.