Seven Novel Acid Sphingomyelinase Gene Mutations in Niemann-Pick Type A and B Patients
*Correspondence: Dr. Jakub Sikora, Institute of Inherited Metabolic Disorders, Div. B, Ke Karlovu 2, 128 08, Praha 2, Czech Republic. Fax: + 420/2/2491 9392. E-mail: firstname.lastname@example.org
We have analyzed acid sphingomyelinase (SMPD1; E.C. 184.108.40.206) gene mutations in four Niemann-Pick disease (NPD) type A and B patients of Turkish ancestry and in three patients of Dutch origin.
Among four NPD type A patients we found two homozygotes for the g.1421C > T (H319Y) and g.3714T > C (Y537H) mutations and two compound heterozygotes, one for the g.3337T > C (F463S) and g.3373C > T (P475L) mutations and the other for the g.84delC (G29fsX74) and g.1208A > C (S248R) mutations.
One of the type B patients was homozygous for the g.2629C>T (P371S) mutation. The last two type B patients were homozygotes for the common g.3927_3929delCGC (R608del) mutation.
The G29fsX74, S248R, H319Y, P371S, F463S, P475L and Y537H SMPD1 mutations are all novel and were verified by PCR/RFLP and/or ARMS. All of the identified mutations are likely to be rare or private, with the exception of R608del which is prevalent among NPD type B patients from the North-African Maghreb region. Geographical and/or social isolation of the affected families are likely contributing factors for the high number of homozygotes in our group.
Niemann-Pick disease is an autosomal recessive sphingolipidosis caused by the deficiency of lysosomal acid sphingomyelinase (ASM, E.C. 220.127.116.11) resulting in lysosomal accumulation of sphingomyelin. At the extremes of the phenotype spectrum lie two common presentations: infantile neurovisceral fatal type A, and type B, characterized by purely visceral involvement and by survival till adulthood (Elleder, 1989; Kolodny, 2000; Schuchman & Desnick, 2001; Vanier & Suzuki, 1996). Patients with intermediate phenotypes have been reported (Elleder & Cihula, 1983; Elleder et al. 1986; Sperl et al. 1994; Takada et al. 1987). Both type A and B variants have higher prevalence in the Ashkenazi Jewish population, reaching 1:40 000 for NPD type A (Goodman, 1979); NPD type B incidence is estimated to be significantly less than that of NPD type A (Schuchman & Desnick, 2001) in this population. Little is known about the frequency of NPD types A and B in non-Jewish populations. Individual reports on the number of NPD type A and B patients and estimated birth prevalence data are available from several European countries, and Australia (Czartoryska et al. 1994; Krasnopolskaya et al. 1993; Meikle et al. 1999; Ozand et al. 1990; Poorthuis et al. 1999).
The SMPD1 gene is 5kb long, consists of six exons (Schuchman et al. 1992; Schuchman et al. 1991), and is located on chromosome 11p15.1–11p15.4 (Da Veiga Pereira et al. 1991). More than 20 mutations associated with both types of NPD in Ashkenazi and non-Ashkenazi patients have been published. Most of the mutations are single base substitutions and small deletions with or without a frameshift. Three mutations, g.3592G > T (R496L) (Levran et al. 1991a), g.1372T > C (L302P) (Levran et al. 1992) and a single nucleotide deletion in a C rich region of exon 2 (fsP330) (Levran et al. 1993), are prevalent among Ashkenazi NPD type A patients, comprising 92% of all mutations in this population. A three base in-frame deletion in exon 6, g.3927_3929delCGC (R608del) (Levran et al. 1991b) is the most prevalent mutation in the population of NPD type B patients from the North-African region of Maghreb (87% of mutated alleles in this area) (Vanier et al. 1993). Another mutation with ethnic prevalence is the c.677delT among Israeli Arab NPD type A patients from the lower Galilee and Samaria region (Gluck et al. 1998). The remainder of the mutations are rare or private. The total number of NPD type A and B patients diagnosed in the Netherlands in the years 1970–1996 was 27 (14 NPD type A and 13 in whom the type was not determined). The calculated birth prevalence for both NPD variants in the Netherlands is 0.53/100000 newborns (Poorthuis et al. 1999). This is the first report on mutation profiles in a subset of NPD type A and B patients diagnosed in the Netherlands.
Materials and Methods
The diagnosis of Niemann-Pick disease was based on the clinical data and biochemically measured ASM deficiency in peripheral white blood cells and/or in cultured skin fibroblasts (Table 1). Samples of parental DNA were not available for mutation analysis. The control group consisted of 25 healthy males and 25 healthy females of Czech origin. Participation in the project was based on informed consent.
Table 1. Clinical, biochemical and genetic characteristics
|1||A||male||Dutch||died at 20||Hepatosplenomegaly,||0,0||2,0||848||[g.3337T>C]+||F463S/P475L||exon 5,|
| ||months|| growth and developmental|| ||[g.3373C>T]||463 Phe>Ser||exon 5|
| || retardation at time of|| ||475 Pro>Leu|| |
| || death, cherry red macula|| |
|2||A||male||Turkish||died at||Psychomotor retardation||0,0||0,6||515||[g.3714T>C]+||Y537H/Y537H||exon 6|
| ||3 years|| at one year of age,|| ||[g.3714T>C]||537 Tyr>His|| |
| || hepatosplenomegaly|| |
|3||A||male||Turkish||died at||Hepatosplenomegaly at 2||0,6||1,0||1868||[g.1421C>T]+||H319Y/H319Y||exon 2|
| ||1 year|| months of age,|| ||[g.1421C>T]||319 His>Tyr|| |
| || consanguineous parents|| |
|4||A||male||Dutch||died at||Hepatosplenomegaly,||1,0||3,0||data not||[g.1208A>C]+||S248R/||exon 2,|
| ||(mild)|| || ||5 years|| psychomotor retardation,|| || ||available||[g.84delC]||G29fsX74||exon 1|
| || foam cells in bone|| ||248 Ser>Arg|| |
| || marrow|| ||Stop 74 Ile|| |
|5||B||female||Turkish||adult, >||Hepatosplenomegaly,||0,4||Fibroblasts not||492||[g.2629C>T]+||P371S/P371S||exon 3|
| ||20 years|| no mental retardation at|| ||available|| ||[g.2629C>T]||371 Pro>Ser|| |
| || at 20 years of age|| |
|6||B||female||Turkish||adult, >||Hepatosplenomegaly, sea||0,1||18,0||not||[g.3927_3929delCGC]+||R608del/||exon 6|
| ||30 years|| blue histiocytes in bone|| || ||expressed||[g.3927_3929delCGC]||R608del|| |
| || marrow, no neurological|| ||608 del Arg|| |
| || impairment|| |
|7||B||male||Dutch||adult, >||Splenomegaly and||1,1||35,0||194||[g.3927_3929delCGC]+||R608del/||exon 6|
| ||20 years|| hyperbilirubinemia, foam|| ||[g.3927_3929delCGC]||R608del|| |
| || cells in bone marrow, no|| ||608 del Arg|| |
| || neurological impairment|| |
| ||reference values||2,2–6,8||260–700||10–103|| |
| || (nmol mg−1 h−1)|| |
ASM Activity Measurements
ASM activity was measured in homogenates of leukocytes and cultured skin fibroblasts (Table 1) using N-methyl-14C sphingomyelin (Amersham) as a substrate (Vanier et al. 1980).
Chitotriosidase Activity Measurements
Chitotriosidase activity was measured in blood plasma with the artificial substrate 4-methylumbelliferyl β-D-N, N′, N′′-triacetylchitotrioside hydrate (Sigma-Aldrich) according to Guo et al. (1995).
Genomic DNA Isolation and PCR Amplification of SMPD1 Gene
Genomic DNA was isolated from cultured skin fibroblasts or leukocytes using the standard phenol/chlorophorm method (Strauss, 2000). PCR amplifications of the SMPD1 coding region were performed in four fragments using five pairs of primers (Table 2). The final volume of PCR reactions was 50 μl. The reaction mixture contained 100 ng of genomic DNA, 0.2 mM dNTPs, 1×PCR buffer Sigma (protocol 1 and 3), 1× LA PCR buffer (protocols 2a, 2b, 4) (Barnes, 1994), DMSO 4–5% (protocols 1, 2a, 2b, 4), 10 μg/ml gelatine (protocol 1), 1.0–3.0 mM MgCl2, 0.12–0.2 μM of each primer. Thermal conditions for all PCR amplifications were as follows: initial denaturation for 2 min (94°C), then 35 cycles, starting with denaturation for 55 s (94°C), followed by 1 min annealing at primer specific temperatures (Table 2), extension for 2 min 20 s (72°C). Protocol 3 included annealing and extension in one step at 72°C for 3 min; final extension lasted 10 min at 72°C for all protocols. Taq DNA polymerase 2.5U (Promega - protocol 1, Sigma - protocol 3), Klen Taq 1.5-3U (Ab Peptides, protocols 2a, 2b, 4) together with Deep Vent 0.1U (NEB, protocols 2a, 2b, 4) were used as well as Perfect Match 1U (Stratagene, protocol 3) PCR enhancer. PCR amplifications were carried out in GeneAmp 2400 (Perkin-Elmer) or PHC-3 (Techne) thermocyclers. PCR products were purified using High PureTM PCR Product Purification Kit (Boehringer Manheim).
Table 2. PCR and sequencing primers
|NPD 307||5′-TGA CAG CCG CCC GCC ACC GAG AGA-3′||5′→ 3′||1||645||−211||60||1|
|OV-NP 2||*5′-CAG GAA ACA GCT ATG ACC TCC ATC AGG GAT GCA TT-3′||3′→ 5′||1|| ||417||60||1|
|NPD 1249||5′-TC CTC TGC TCT GCC TCT GAT TTC TCA CCA T-3′||5′→ 3′||2||926||731||68||2a|
|OV-NPD 2157||*5′-CAG GAA ACA GCT ATG ACA ATC AGA GAC AAT GCC CCA||3′→ 5′||2|| ||1639||68||2a|
| || GGT TCC CTT CT-3′|| |
|OV-NPD 1249||*5′-CAG GAA ACA GCT ATG ACT CCT CTG CTC TGC CTC TGA||5′→ 3′||2||926||731||68||2b|
| || TTT CTC ACC AT-3′|| |
|NPD 2157||5′-AAT CAG AGA CAA TGC CCC AGG TTC CCT TCT-3′||3′→ 5′||2|| ||1639||68||2b|
|NPD 3061||5′-CCC AGC ACA GGA GGA CCA GGA TTG GAA-3′||5′→ 3′||3,4||645||2543||72||3|
|OV-NPD 3688||*5′-CAG GAA ACA GCT ATG ACG GGA CAA CAG GGA TGG TGA||3′→ 5′||3,4|| ||3170||72||3|
| || GAT GCT CA-3′|| |
|NPD 3658||5′-GCA TCT CAC CAT CCC TGT TGT CCC ATG-3′||5′→ 3′||5,6||920||3147||68||4|
|OV-NPD 4567||*5′-CAG GAA ACA GCT ATG ACG CTT TTT CAC CCT TTC CTA||3′→ 5′||5,6|| ||4049||68||4|
| || CAT CAA GAA CT-3′|| |
|Sequencing primers|| |
|SNP1Cy5||5′Cy5-GGA CGG GAC AGA CGA ACC-3′||5′→ 3′|| || ||−41||55|| |
|SNP2CCy5||5′Cy5-CAC TGG GAC ATT TTC TC-3′||5′→ 3′|| || ||965||55|| |
|SNPR2Cy5||5′Cy5-GGC TTC GGC ACA GTA GG-3′||3′→ 5′|| || ||1017||55|| |
|SNP7Cy5||5′Cy5-CTC ACC ATC CCT GTT GTC C-3′||5′→ 3′|| || ||3151||55|| |
|SNP6Cy5||5′Cy5-CCA GTC AGC CCC ACA TC-3′||5′→ 3′|| || ||3562||55|| |
|CyNPD 3061||5′Cy5-CCC AGC ACA GGA GGA CCA GGA TTG GAA-3′||5′→ 3′|| || ||2543||55|| |
|SNP8Cy5||5′Cy5-GAC CCA GGC AAA CAT AC-3′||5′→ 3′|| || ||3668||55|| |
|UP||5′Cy5-CAG GAA ACA GCT ATG AC-3′|| || || ||Universal||55|| |
| || primer|| || |
Sequencing of SMPD1 Gene
Direct cycle sequencing reactions contained 100–200 fmol of purified PCR products, 1.0 μM of 5′Cy5 labeled primer (intragenic or universal, Table 2), 0.1 mM dNTP's, 0.1 mM deaza dGTP, 0.83 μM ddNTP's and 6.5 mM MgCl2, 1 × LA PCR buffer (Barnes 1994) and AmpliTaqFS (Perkin-Elmer). Thermal conditions for all cycle sequencing reactions were as follows: initial denaturation for 2 min (94°C), then 35 cycles, starting with denaturation for 15 s (95°C), followed by 30 s (55°C) annealing, and extension 30 s (68°C). Sequence analysis was performed on the automated fluorescent sequencer Alfexpress (Pharmacia).
Confirmation of Novel Mutations
Five of the novel mutations altered the recognition site for one of the following restriction endonucleases: NlaIII, ApoI, BslI, SfcI and EagI (all NEB). The patient and control PCR products were digested by the corresponding endonuclease (Table 3). The Amplification-Refractory Mutation System (ARMS) (Little, 1998) method was used for confirmation of the G29fsX74 and S248R mutations. ARMS reaction conditions were identical to that of PCR protocol 4, and ARMS primer sequences are listed in the Table 4.
Table 3. Restriction endonucleases and PCR products cleaved for mutation confirmation.
Table 4. ARMS primers
|G29fsX74||5′-GCC AGG CCC ATC CAA AGG AGT CCG GGA C-3′||3′→ 5′||mutated||106||NPD 307|
| ||5′-GCC AGG CCC ATC CAA AGG AGT CCG GGA G-3′||3′→ 5′||normal||106||NPD 307|
|S248R||5′-AGG GTC CTC AGG GGC AGG TCA CAC TTG TG-3′||3′→ 5′||mutated||1236||NPD1249|
| ||5′-AGG GTC CTC AGG GGC AGG TCA CAC TTG TT-3′||3′→ 5′||normal||1236||NPD1249|
Each of the novel mutations was assessed in 100 wild type SMPD1 alleles to exclude the possibility of a common genetic polymorphism.
All mutations and genes are described according to recent mutation nomenclature (Den Dunnen & Antonarakis 2000, 2001; Wain et al. 2002). Nucleotide change numbers and primer positions are derived from the genomic sequence of the SMPD1 gene (GenBank accession no. X63600), nucleotide A from the methionine initiator codon (ATG) being nucleotide +1, both for mutation description and primer location.
Fourteen mutant alleles were characterized (Table 1) as well as the presence of the three common polymorphisms in the SMPD1 gene (Table 5). We found seven novel mutations (G29fsX74, S248R, H319Y, P371S, F463S, P475L, Y537H) and identified one previously known (R608del) mutation. Mutations were distributed throughout the whole gene (Table 1). Six of the novel mutations (S248R, H319Y, P371S, F463S, P475L, Y537H) are single base substitutions; the single nucleotide deletion (G29fsX74) is a frameshift mutation. Information on genotype/phenotype associations is summarized in Table 1. Mutation S248R was found in one patient from the cohort of Czech and Slovak NPD type A and B patients analysed in parallel by Hana Pavlu-Pereira at our institution (manuscript in preparation).
Table 5. Common polymorphisms in SMPD1 gene
Significant elevations of plasma chitotriosidase activity in Niemann-Pick disease and some other lysosomal storage diseases have already been reported by Guo et al. (1995). This enzyme derives from activated macrophages and if elevated may be an indicator of lysosomal storage. Table 1 shows elevated activity of blood plasma chitotriosidase in all cases where it was measured.
The disease-causing character of the novel mutations was evaluated with respect to the accepted criteria for sequence variation discovery and reporting (Cotton & Horaitis, 2000):
there was no other nucleotide change in the SMPD1 coding region, except for the previously described common polymorphisms.
all mutations, with the exception of P475L, result in non-conservative amino acid exchanges, the frameshift mutation G29fsX74 results in an early stop codon.
the affected amino acid residues are conserved in the human and murine (GenBank accession no. Z14132) SMPD1 gene, with the exception of glycine 29 which is located in the signal part of the protein.
at least two (usually more) independent PCR products were used for DNA sequence analysis.
negative results in restriction analysis and ARMS of one hundred control alleles diminished the possibility that the sequence variations found are common genetic polymorphisms. However, in this regard we must note that the controls were of Czech origin because DNA samples of healthy subjects from Dutch and Turkish populations were not available.
The seven novel mutations are most probably private mutations. For these reasons sequencing is the most suitable method for mutation analysis in NPD families.
In patient 7 we found in addition to R608del a synonymous heterozygous mutation (g.54G > A) in exon 1 in one of the alleles. This single nucleotide substitution changes glutamic acid codon GAG, with a frequency of 40,2 per 1000 of codon usages, and to GAA with a frequency of 29,1 per 1000 of codon usages (data based on GenBank Release 129.0, 15th April 2002). A BLAST search set at default values (http://www.ncbi.nlm.nih.gov/BLAST/) using the SMPD1 gene exon 1 as a query sequence did not reveal a single human EST with this pattern out of 174 BLAST hits. We thus regard this polymorphic change as relatively unimportant, for the above reasons.
Of considerable interest is the high number of homozygotes in our cohort of patients. All four patients of Turkish origin, and one patient of Dutch origin, were found to be homozygous for mutations H319Y, P371S, Y537H, and R608del, respectively. Geographical and/or social isolation (probably more important in our group) of the affected families together with a high rate of intra-ethnic marriages may explain these findings.
The results suggest that the mutations H319Y and Y537H are associated with a type A phenotype, as these two mutations were found homozygously in patients 3 and 2 respectively. Both patients presented with a progressive infantile course of the disease and with residual ASM activities in the range of commonly accepted criteria for type A NPD (Schuchman & Desnick, 2001).
We base association of mutation P371S with type B NPD on the phenotype of patient 5, whose clinical presentation includes only hepatosplenomegaly and no mental retardation in early adulthood.
The influence of mutations G29fsX74, S248R, F463S and P475L on phenotype cannot be derived from our data due to compound heterozygosity of patients 1 and 4. We assume that G29fsX74 is a null mutation as it results in a premature stop codon.
The R608del, found in patients 6 and 7, is the prevalent mutation (87% of mutated alleles) among North-African NPD type B patients from the region of Maghreb (Morocco, Algeria, Tunisia) (Vanier et al. 1993). Mutation R608del was always associated with milder type B NPD, even when identified heteroallelically with a severe type A mutation (Levran et al. 1991b). The in-frame character of this deletion, and the localisation at the end of the coding sequence of the SMPD1 gene, may be the reason why the mutation associates exclusively with NPD type B. Both patients 6 and 7 suffer from a typical type B phenotype with corresponding higher residual ASM activities in cultured skin fibroblasts (Table 1). At this point we should note that in-situ cell-loading assay provides better estimation of lysosomal sphingomyelin degradative capacity than plain in-vitro activity measurement (Graber et al. 1994). Despite this, genotype/phenotype correlation, as shown for R608del, may play an important role in future decisions about enzyme replacement therapy of NPD patients.
This work was funded by the Research Project of the Czech Ministry of Education No.11110003. We would like to thank Dr. Martin Hrebicek, Dr. Lenka Dvorakova (both from Institute of Inherited Metabolic Disorders, Prague, Czech Republic) and Prof. Jan P. Kraus (Department of Pediatrics, University of Colorado School of Medicine, USA) for critical reading of the manuscript.
Received: 11 June 2002
Accepted: 11 November 2002