HUMAN LYSOSOMAL SIALIDASE AND THERAPEUTIC USES THEREOF
BACKGROUND OF THE INVENTION
( a) Field of the Invention The invention relates to the identification of a complete cDNA coding for human lysosomal sialidase, its cloning, sequencing and expression, and to the identification of mutations found in sialidosis patients and chromosomal mapping of the sialidase gene. (b) Description of Prior Art
Sialidosis is an autosomal recessive disease caused by genetic deficiency of a lysosomal sialidase (EC 3.2.1.18, neuraminidase ) which catalyzes the hydrolysis of the terminal sialic acid residues of glycoconjugates . Sialidase ( acetylneuraminyl hydrolase or neuraminidase, EC 3.2.1.18) catalyzes the hydrolysis of terminal sialic acid residues of oligosaccharides, glycoproteins and glycolipids. Sialidase has been well studied in viruses and bacteria where it destroys the sialic acid receptors on host cells, and mobilize bacterial nutrients. In mammals, three types of sialidases, localized to the lysosome, plasma membrane and cytosol, have been described. They have different substrate specificities and biochemical properties and are probably encoded by different genes. The lysosomal acid-sialidase is associated with cathepsin A (also named "protective protein"), and β-galactosidase in a 1.27 MDa multienzyme complex. This complex is essential for the expression of sialidase activity and for the stabilisation of β-galactosidase in the lysosome as demonstrated by the existence of an autosomal recessive disease, galactosialidosis, characterized by combined deficiency of β-galactosidase and sialidase activities secondary to cathepsin A deficiency.
The genetic deficiency of lysosomal sialidase in humans causes an autosomal recessive lysosomal storage disease, sialidosis, associated with tissue accumulation and urinary excretion of sialylated oligosaccharides and glycolipids. Sialidosis includes two main clinical variants with different ages of onset and severity. Sialidosis type I is the late-onset form characterized by bilateral macular cherry-red spots and myoclonus. Sialidosis type II is the infantile-onset form which is also associated with skeletal dysplasia, Hurler-like phenotype, mental retardation and hepatosplenomegaly. A severe form of the disease also manifests prenataly and is associated with ascites and hydrops fetalis. The molecular defects in sialidosis have not been characterized since the lysosomal sialidase has never been purified.
It would be highly desirable to be provided with the identification of a complete cDNA coding for human lysosomal sialidase, its cloning, sequencing and expression.
It would also be highly desirable to be provided with the identification of mutations found in sialidosis patients and chromosomal mapping of the sialidase gene.
SUMMARY OF THE INVENTION
One aim of the present invention is to provide the identification of a complete cDNA coding for human lysosomal sialidase, its cloning, sequencing and expression.
Another aim of the present invention is to provide the identification of mutations found in sialidosis patients and chromosomal mapping of the sialidase gene.
In accordance with the present invention there is provided a human lysosomal sialidase encoded by the DNA sequence set forth in Fig. 1 and having the amino acid sequence depicted in Fig. 1. In accordance with the present invention there is provided a method of mutations analysis in patients affected with sialidosis or similar diseases, which comprises the steps of: a) isolating DNA from a biological sample of said patients; b) comparing the DNA of step a) with the DNA of the non-mutated protein to determine the presence of any mutation, whereby the presence of a mutation is indicative of sialidosis or similar diseases. In accordance with the present invention there is provided a method of treatment of lysosomal storage disorders in patients, which comprises administering to said patients recombinant human lysosomal sialidase. The disorders treated include, without limitation, Sialidosis, Tay-Sachs and Sandhof diseases.
In accordance with the present invention there is provided the use of the human lysosomal sialidase of the present invention for screening pharmaceutical agents against viral sialidase without side effects to the human, wherein said pharmaceutical agents are useful in the treatment of viral infections.
In accordance with the present invention there is provided the use of the human lysosomal sialidase of the present invention as poly-His and GST fusion proteins for therapeutic protein expression.
In accordance with the present invention there is provided the use of the human lysosomal sialidase of the present invention for the digestion of sialylated oligosaccharides and glycolipids in milk.
In accordance with the present invention there is provided the use of an human lysosomal sialidase of the present invention inactivated as an anti-viral agent, wherein the inactivated sialidase can bind with high affinity to surface sialic acid residues thereby preventing the viral protein from binding with the cells .
In accordance with the present invention there is provided the use of recombinant human lysosomal sialidase for the treatment of lysosomal storage disorders in patients.
The disorders include, without limitation, Sialidosis, Tay-Sachs and Sandhof diseases.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 illustrates the cDNA (SEQ ID N0:1) and deduced amino acid (SEQ ID NO: 2) sequences of human lysosomal sialidase;
Fig. 2 illustrates the amino acid sequence alignment of human lysosomal (hum) (SEQ ID NO: 18) and Salmonella typhimurium (salty) (SEQ ID NO: 19) siali- dases by the Lipman-Pearson algorithm;
Fig. 3 illustrates an insertion frameshift and two missense mutations identified in sialidosis patients;
Fig. 4 illustrates the detection of identified mutations in genomic DNA;
Fig. 5 illustrates the allele-specific amplification of DNA from sialidosis patient GM01718A; Fig. 6 illustrates the Northern blot analysis of
RNA from cultured skin fibroblasts of sialidosis patients and control; and
Fig. 7 illustrates FISH mapping of human lysosomal sialidase on chromosome 6 (6p21.3).
DETAILED DESCRIPTION OF THE INVENTION
The identification of lysosomal sialidase is necessary for the understanding of the molecular bases of two severe human genetic diseases, sialidosis and galactosialidosis. However, the identification and sequencing of sialidase has been hampered by low tissue content, instability and complete inactivation of the enzyme after dissociation of the 1.27 MDa complex. We identified a cDNA coding for the lysosomal sialidase using a different approach - an EST database search for proteins containing the Asp-box and the Phe-Arg-Ile-Pro amino acid sequence motifs conserved in both bacterial and mammalian sialidases. The identity of the clone was confirmed by expression of sialidase activity, structural analysis of the deduced amino acid sequence, the characterization of mutations in sialidosis patients and chromosomal mapping.
In accordance with the present invention, the human lysosomal sialidase cDNA was identified, cloned and sequenced. The deduced amino acid sequence of human sialidase is homologous to those of bacterial sialidases, suggesting that the human enzyme shares the same "β-propeller" fold. Expression of the cloned cDNA in sialidosis fibroblasts produced a 15-fold increase of intracellular sialidase activity. Three mutations, one frameshift insertion and two missense, were identified in six patients affected with the neurodegenerative disease. The gene coding for lysosomal sialidase was mapped to human chromosome 6 (6p21.3), which is consistent with the previous chromosomal assignment of the sialidase gene in proximity to the HLA locus.
RNA isolation, analyses and probes Human skin fibroblasts from sialidosis patients were obtained from NIGMS Human Genetic Mutant Cell
Repository (GM1718A, GM11604, GM02685, GM02837, GM02921) and the Montreal Children's Hospital Cell Repository (WG544). Cells were cultured to confluency in Eagle's Minimal Essential Medium (Mediatech, Washington DC), supplemented with 10% (v/v) fetal calf serum (MultiCell) and antibiotics. Total RNA or DNA were isolated as described (Maniatis,T. , Fritsch,E.F. and Sambrook,J. (eds) (1982) Molecular cloning: a laboratory manual . Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.).
Northern blotting was performed as described (Maniatis,T. , Fritsch,E.F. and Sambrook,J. (eds) (1982) Molecular cloning: a laboratory manual . Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), using a sialidase cDNA fragment obtained from clone 31259 by a NotI and Hindlll double digestion, and a human tubulin Pstl/SacI cDNA fragment, both radiolabelled with 3 P dCTP using the Oligolabelling Kit (Pharmacia).
Generation and cloning of sialidase cDNA
The human sialidase cDNA was obtained by reverse transcription of total human fibroblast RNA using the Marathon cDNA amplification Kit (Clonetech) according to the manufacturer's protocol, followed by cDNA amplification using 5 ' -CCCAAGCTTAGATCTTGGAGTCTAGCTGCC AGGGT-3' (SEQ ID NO: 3) and 5 ' -AGTTCCCTGAGTTCACATTG-3 ' ( SEQ ID NO : 4 ) primers complementary to the cDNA sequence of the I.M.A.G.E. Consortium clones 31259 and HIBBL61, respectively. The amplification conditions were: denaturation at 94 °C for 6 min, 30 cycles each consisting of 1 min at 94°C, 45 sec at 58°C and 2 min at 72°C followed by 8 min at 72°C.
The amplified sialidase cDNA was directly cloned into pCRII vector using the TA cloning kit (InVitrogen) according to the manufacturer's protocol.
Nucleotide sequencing
Both strands of sialidase cDNA and of PCR- amplified fragments from patients ' total RNA were sequenced by PRISM Ready Reaction Dye Deoxy Terminator cycle sequencing kit on an Applied Biosystems 373A automated sequencer, and analyzed using the SeqED software.
Identification and cloning of sialidase cDNA Bacterial sialidases share a conserved motif of amino acid sequence: the so-called "Asp-box" (Ser/Thr- X-Asp-X-Gly-X-X-Trp/Phe) , which is repeated three to five times at topologically equivalent positions, and a Phe-Arg-Ile-Pro sequence motif which is part of the active site. Since Asp-boxes and the Phe-Arg-Ile-Pro motif have been conserved in the mammalian cytosolic sialidases (Crennell, S.J. et al., (1993) Proc. Natl . Acad. Sci . USA, 90:9852-9856), we hypothesized that such motifs could also be present in the lysosomal sialidase. Therefore, we searched the expressed sequence tags database (dbEST, National Center for Biotechnology Information) for proteins containing these sequences and found several overlapping clones of human fetal brain, spleen and placenta which included three or four Asp-boxes, and the Phe-Arg-Ile-Pro motif. The longest clone, 31259, was obtained from the I.M.A.G.E. Consortium (Lawrence Livermore National Laboratory, Livermore CA) and sequenced.
A complete cDNA was obtained by RT-PCR amplification of total RNA from human fibroblast and placenta using primers complementary to the sequences of two overlapping clones, 31259 and HIBBL61, cloned into a pCRII vector and sequenced. The resulting cDNA includes 1245 bp of open reading frame and a polyadenylation site within 3' untranslated region (Fig. 1). The predicted signal peptide is underlined,
lysosomal C-terminal targeting motif is double- underlined. Potential glycosylation sites are marked with asterisks. Asp-boxes are boxed. The deduced amino acid sequence contains four Asp-boxes (Fig. 1) and a similar sequence motif where the conserved Asp residue is replaced by a Ser residue. The presence of three potential N-glycosylation sites (indicated by asterisks in Fig. 1) suggests that the enzyme is glycosylated, which is in accordance with the observation that lysosomal sialidase binds to concanavalin A-SEPHAROSE™ affinity matrix. The sequence also revealed a C-terminal "Tyr-X-X- hydrophobic residue" motif (double-underlined in Fig. 1) similar to that found in other lysosomal integral membrane proteins. We predict that the first 42 amino acid residues of sialidase may represent the signal peptide (underlined in Fig. 1) since its length and composition are similar to those of other lysosomal glycosidases . Neither bacterial nor mammalian cytosolic sialidases contain a signal peptide or a lysosomal targeting motif.
The analysis of the deduced amino acid sequence of lysosomal sialidase revealed the presence of a possible signal peptide, three glycosylation sites and a Gly-Tyr-X-X-hydrophobic-residue lysosomal-targeting signal, common to several lysosomal integral membrane proteins such as glucocerebrosidase, LAMP-1, LAMP-2, and LGP-85. All these proteins are transported to the lysosome by association of their C-terminus with cytoplasmic proteins called adapters. The same pathway is also used by the lysosomal acid phosphatase, which is synthesized as a transmembrane protein but is cleaved inside the lysosome into a soluble form. The dual localization of sialidase in the lysosomal membrane and in matrix and the demonstration that the
intralysosomal processing of sialidase is necessary for its activation suggest that it may be also transported by a similar mechanism.
Homology with bacterial sialidases
A database search for proteins homologous to the deduced sialidase amino acid sequence using BLASTP program revealed several bacterial sialidases, sialidase from Trypanosoma cruzi , as well as rat and hamster cytosolic sialidases. Sialidase from Salmonella typhimurium, which X-ray-structure has been determined at 1.6 A resolution ( PDB file 2sil) had the highest similarity with human sialidase (P value of 1.6 X 10-' in BLASTP comparison). The alignment of their amino acid sequences (Fig. 2) suggested that the basic fold of bacterial and viral sialidases which consists of six four-stranded antiparallel β-sheets arranged as the blades of a propeller on an axis of symetry passing through the active site is conserved in the human enzyme. This finding supports the hypothesis that mammalian, viral and bacterial sialidases have a common phylogenetic origin. The deduced amino acid sequence of lysosomal sialidase was analysed using the protein-fold prediction program THREADER, which determines how the sequence fits into a library of unique protein folds derived from the database of protein structures ( Jones, D.T. et al. (1992) Nature, 358:86-89). Using the default parameters, THREADER gave the highest probability (pairwise energy Z-score = -3.5) for fitting the human lysosomal sialidase sequence into the Salmonella typhimurium sialidase fold. All important active site residues which bind the carboxylate group of the sialic acid substrate (Arg 37, 246 and 309) and one of three N-acetyl-/N-glycolyl- binding active site residues (Asp 97) of bacterial
sialidase are conserved in the human enzyme (Arg 78, 280, 347 and Asp 135, respectively).
The Neu-1 gene, which controls lysosomal sialidase activity in liver, was identified and mapped within the major histocompatibility gene complex in H-2 locus of mouse and RT-1 locus of rat. In humans, the gene causing sialidosis cosegregates with a specific HLA haplotype, also indicating the close linkage of lysosomal sialidase gene locus to HLA at 6p21.3. Using the FISH method (Lichter,P. and Cremer,T. (1992) In : Rooney,D. E. , and Czepulkowski ,B. H. (eds) Human cytogenetics . A practical approach, Oxford University Press, N.Y. vol. 1:157-192) we demonstrated that the gene coding for human sialidase is located on chromosome 6 (6p21.3) which confirms the location of sialidase gene near the HLA locus.
SSCP/PCR mutation analysis
2 μg of total RNA isolated from patients ' fibroblasts were reverse transcribed after denaturation for 3 min at 65 °C, using 140 ng of oligonucleotide primer dt2o* 20 μl of IX H-RT buffer (Gibco BRL), containing 1 mM dNTP, 39 units of rRNasin (Promega, Madison), 1 mM DTT and 200 units of Superscript RT. The mixture was incubated 30 min at 42 °C followed by 5 min at 95°C. Patients' sialidase cDNA, was amplified into 3 overlapping fragments using the following primer sets: 5 ' -CCCAAGCTTAGATCTTGGAGTCTAGCTGCCAGGGT-3 ' (SEQ ID NO: 3) and 5 ' -CCAGGGGCAAACACTTCAGT-3 ' (SEQ ID NO: 5) (fragment 1); 5 ' -GAGCGATGTTGAGACAGGAG-3 ' (SEQ ID NO: 6) and 5'-TCGCAGGGTCAGGTTCACTC-3' (SEQ ID NO: 7) (fragment 2); and 5 ' -CGCTACGGAAGTGGGGTCAG-3 ' (SEQ ID NO: 8) and 5 ' -AGTCCTGAAGGCAGAATACC-3 ' ( SEQ ID NO: 9 ) ( fragment 3 ) . 15 μl of the radioactive fragments were directly digested for 3 h with 10 U of Ncol (in case of fragments 1 and 2), and PstI (in case of fragment 3),
to generate smaller fragments . Reaction mixtures were added to an equal volume of 95% formamide, 20 mM EDTA, 0.05% bromphenol blue and 0.05 % xylene-cyanol, denatured 3 min at 90°C and separated on a 6% (w/v) acrylamide gel containing 10% (v/v) glycerol.
Patient (identified as GM02921) cDNA was amplified as above and entirely sequenced, instead of using SSCP analysis.
Confirmation of mutations
The insertion frameshift mutation of patient GM11604 was confirmed by heteroduplex analysis. 50 ng of genomic DNA from patient (identified as GM11604) containing the ACTGδ-11 insertion and from control were PCR amplified using the following primers: 5 ' -CCCAAGCTTAGATCTTGGAGTCTAGCTGCCAGGGT-3 ' (SEQ ID NO: 3) and 5 ' -GCGGCCCCCAGCGTCTGTCC-3 ' (SEQ ID NO: 10). PCR conditions were as for the amplification of cDNA except the annealing temperature was 62°C. 5 μl of PCR reaction mix was subjected to PAGE (12% (w/v) of polyacrylamide) . 10 μl of PCR products from control and patient were mixed, heated at 94 °C for 6 min and slowly cooled to room temperature prior to PAGE.
To confirm the presence of TiQ88~'tc,--C. substitution, a fragment surrounding the mutation was PCR-amplified from 100 ng of genomic DNA from GM01718A and controls as described above using the 5 ' -CCCAGCACATCCAGAGTTCC-3 ' (SEQ ID NO: 11) and 5 ' -AGTCCTGAAGGCAGAATACC-3 ' (SEQ ID NO: 9) primers. 10 μl of PCR products from control and patient were digested with 10 U of Ncol for 1 h at 37 °C and analyzed by electrophoresis in 1% (w/v) agarose gel.
To confirm the presence of the W-j-j^-to-A substitution, a fragment containing the mutation was PCR-amplified from 100 ng of genomic DNA from GM01718A and controls as described above using the forward
primer, 5 ' -CGCTACGGAAGTGGGGTCAG-3 ' (SEQ ID NO:8), and reverse primer, 5 ' -TCGGCAGTGGCAGTGGTAGT-3 ' (SEQ ID NO: 12). 1 μl of the resulting PCR product was reamplified using the same forward primer and the reverse primer, 5 ' -GGGCTGGCATTCATCAGGATTT-3 ' ( SEQ ID NO: 13) with a single nucleotide mismatch (T for G) which creates a Dral restriction site in normal sequence. 8 μl of PCR products from control and patient were digested with 20 U of Dral for 2 hrs at 37 °C and analyzed by electrophoresis in 12% polyacrylamide gel.
In order to demonstrate that T-j-j^-to-A and T1088-to-C mutations in GM01718A are present on different alleles, the sequence containing these mutations was amplified from 100 ng of genomic DNA of GM01718A using four combinations of the following primers. Two forward primers, 5'-
TCAGCCCAAGCAGGAAAATGATTT-3 ' (SEQ ID NO: 14) and 5'- TCAGCCCAAGCAGGAAAATGATTA-3 ( SEQ ID NO: 15), corresponded to normal sialidase sequence and to Ty^g-to-A mutation. Two reverse primers, 5 ' -GCCACTGGGGCCTGGCCATA-3 ' (SEQ ID NO: 16) and 5 ' -GCCACTGGGGCCTGGCCATG-3 ' (SEQ ID NO: 17) corresponded to normal sequence and to ιø88-to-C mutation. The PCR conditions were as described above except that annealing temperature was 66 °C. 7 μl of the resulting PCR products were analyzed by electrophoresis in 1% (w/v) agarose gel.
Mutations in sialidosis patients Sialidase cDNA was PCR-amplified as three overlaping fragments from reverse-transcribed total fibroblast RNA of six sialidosis patients. SSCP analysis of the first fragment (Fig. 3A, top panel) revealed a band with altered migration for cell line GM11604, as compared to other cell lines. The β- strands localized in the Salmonella typhimurium
sialidase X-ray structure (Crennell, S.J. et al., (1993) Proc . Natl . Acad. Sci . USA, 90:9852-9856) are underlined and indicated. Active site residues which interact with the inhibitor, 2, 3-dehydro-N- acetylneuraminic acid, are indicated with asterisks: N indicate N-acetyl-binding residues; C, carboxylate- binding residues; and Y, anomeric cycle-binding residue. Sequencing of the variant band demonstrated that the patient was homozygous for a frameshift mutation, an ACTG duplication after nucleotide 7 (+ACTG8-11, Fig. 3A, middle panel). To confirm the mutation, the surrounding sequence was amplified from the genomic DNA of GM11604 and of a control- The PCR products were mixed, heated to 94°C, and analyzed by PAGE to test for heteroduplex formation. As expected, the normal and mutant homoduplexes migrated as an unresolved doublet of the expected size along with two heteroduplex bands of slower mobility due to the presence of the insertion on one strand and the normal sequence on the other (Fig. 3A, lower panel).
The presence of additional bands in the SSCP analysis of PCR products of the second and the third cDNA fragments of GM01718A (Figs. 3B and 3C, top) suggested heterozygous mutations. Indeed, direct sequencing of the PCR products revealed two nucleotide substitutions, ^g-to-A and τιo88-to-c (F gs. 3B and 3C, middle). The presence of these mutations in genomic DNA of GM01718A was confirmed by restriction analysis. The T10gg-to-C substitution creates an Ncol restriction site which was present in fragments amplified from genomic DNA of GM01718A but not of a control (Fig. 4B, lower panel). The mutation T-j-jy-to-A does not create or remove any restriction site. Therefore we amplified the fragment containing this mutation using a reverse primer with a single mismatch
which itself creates Dral site only in amplification products from normal genomic DNA. The Tyyg-to-A mutation removes the Dral site in one allele of GM01718A (Fig. 4C, lower panel). T-j-j q-to-A and Tχo88- to-A mutations were not found in genomic DNA of 28 normal unrelated individuals when tested by the above methods .
Allele-specific PCR on genomic DNA (Fig. 5A) was used to determine whether the mutations are present on the same or different alleles in GM01718A. Fig. 5A, the allele-specific forward and reverse primers terminate in a nucleotide mismatches relative to Υηη q- to-A and Tχo88~ to-C mutations.
The combinations of normal and mutant oligonucleotide primers, where 3' nucleotide matches the nucleotide of the normal and of mutant sequence, respectively, were used. When the PCR was conducted with two normal (Fig. 5B, lane 1) or two mutant oligonucleotides (Fig. 5B, lane 3) the PCR product was weaker in intensity than in the case of PCR ' s conducted with the normal/mutant ( lane 2 ) and the mutant/normal (lane 4) oligonucleotides. Fig. 5B, 7 μl of the products of the PCR reactions using the combinations of normal sense/normal antisense (1), normal sense/mutated antisense (2) mutated sense/mutated antisense (3) and mutated sense/normal antisense ( 4 ) primers are analyzed by electrophoresis in 1% (w/v) agarose gel. These results indicate that the two mutations are on different alleles . For the other cell lines alterations were not detected by SSCP, although the PCR products were additionally digested with the restriction enzymes to generate smaller fragments. For WG544, the level of sialidase mRNA was dramatically reduced (less that 10% of normal), as analyzed by Northern blotting by
comparison with tubulin mRNA as a control (Fig. 6), which suggests a transcriptional regulation abnormality. (Fig. 6) 10 μg of total RNA from sialidosis fibroblasts and 5 μg from control were applied on gel. The membranes were hybridized with 32P-labelled sialidase (A) and tubulin (B) cDNA probes as described above. The size of sialidase mRNA (1.9 kb) is consistent with that of its cDNA, indicating that the cloned cDNA is probably full-length. Fig. 6 (Top) shows SSCP patterns of PCR- amplified cDNA fragments from four sialidosis patients and from control (Figs. 3A, fragment 1; 3B, fragment 2; 3C, fragment 3). Patient GM11604 shows a lower mobility band corresponding to fragment 1 indicating the homozygous insertion; patient GM01718A shows duplication of bands corresponding to fragments 2 and 3 indicating two heterozygous nucleotide changes.
Fig. 6 (Bottom) shows the nucleotide sequences of cDNA fragments amplified from patients (upper panels) and controls (lower panels). Patient GM11604 is homozygous for ACTGg-n insertion; patient GM01718A is heterozygous for two mutations: a T'A transversion at nucleotide 779 in fragment 2 and a T*C transition at nucleotide 1088 in fragment 3. The sialidosis patients with identified exonic mutations belong to the severe early-onset clinical form, sialidosis type II. The almost complete deficiency of sialidase activity (less than 1% of normal) in cultured skin fibroblasts from these patients is consistent with the predicted termination of protein synthesis due to the frameshift mutation in one patient and the dramatic decrease in mRNA level in the other.
Table 1
Expression of sialidase activity in sialidosis cultured skin fibroblasts
Cells Enzyme activity in cellular homogenates 24 h after transfection* β-hexosaminidase Sialidase+ (nmol/min mg of protein) (pmol/min mg of protein)
Untransfected 60.6 ± 5.0 4 ± 0.3
Transfected with pCMV 47.9 ±4.4 3 ± 0.1
Transfected with pCMV-SIAL 48.6 ± 4.6 45 ± 13
* Data represent the average of at least 3 independent measurements ± SD. + A highly reproducible background of 32 ± 2 pmol/min due to spontaneous hydrolysis of the substrate at acidic pH was substracted from the data.
Further in vitro expression studies will be necessary to confirm that the identified T- -jg- -o-A substitution (Phe260 to Tyr) and Tjøgg-to-C substitution (Leu3g3 to Pro) also cause deficiency of sialidase activity. Nevertheless the latter mutation, Leu3β3 to Pro, is likely to induce the structural change which affects the sialidase activity or stability. The ability to genotype patients with sialidosis will permit DNA-based diagnosis and genotype-phenotype correlations .
Table 2 Mutations identified in sialidosis patients
Cell line(s) Clinical type Mutation Result
GM 11604 Sialidosis type II +ACTG8.11 Val28→Stop
GM01718A Sialidosis type II T779→A Phe260→Tyr
GM01718A Sialidosis type II T1088→c Leu363-»Pro
WG544 Sialidosis type II low mRNA no activity*
* based on enzymatic assay
Expression of sialidase in human fibroblasts
Full-length human sialidase cDNA was obtained as a Xmalll fragment (1.9 kb) from pCRII vector and cloned into the NotI site of a pCMV expression vector (MacGregor,G.R. and Caskey,C.T. (1989) Nucleic Acids Res, 17:2365).
Skin fibroblasts of sialidosis type II, WG544 cell line, were cultured as described, trypsinized, suspended at a density of 6 X 10° cells per ml in OPTI- MEM medium (BRL) supplemented with 5% (v/v) of fetal calf serum and electroporated twice using a Gene Zapper device (IBI). 24 h after the transfection, sialidase and hexosaminidase activities in cellular homogenates were assayed using the corresponding fluorogenic 4-methylumbelliferyl-glycoside substrates according to published methods (Potier,M. et al. (1979) Anal . Biochem. 94:287-296). One unit of enzyme activity (U) is defined as the conversion of 1 μmol of substrate per min. The WG544 fibroblast cell line which showed the reduced sialidase mRNA level and negligible sialidase activity was used for the expression of the cloned cDNA. The full-length cDNA was inserted into a pCMV vector containing the CMV immediate-early promoter (MacGregor,G.R. and Caskey,C.T. (1989) Nucleic Acids Res, 17:2365). The resulting construct was electroporated into fibroblasts and, 24 h after
transfection, sialidase and β-hexosaminidase (control) activities were assayed in fibroblast homogenates . The cells transfected with pCMV-SIAL showed a 15-fold increase of sialidase activity (45±15 pmol/min μg of protein) as compared with untransfected cells (4±0.3 pmol/min μg of protein) or with pCMV-transfected cells (3±0.1 pmol/min μg of protein).
Chromosomal mapping of sialidase gene The gene encoding lysosomal sialidase was mapped to chromosome 6 at 6p21.3 using the FISH method (Fig. 7).
Heteroduplexes (shown by arrows, Fig. 4A) were identified when the PCR products amplified from GM11604 and a control genomic DNA were mixed, heated at 94 °C for 6 min, slowly cooled to room temperature and analyzed by electrophoresis in 12% polyacrylamide gel. Lane 1-control, lane 2 - GM11604, lane 3 - control plus GM11604. The Ncol restriction site created in one of two alleles by T'C transition was found in PCR product amplified from genomic DNA of GM01718A but not of the control (Fig. 4B) . Lanes 1 and 2 - non-digested control and GMD1718A, lanes 3 and 4 - control and GM01718A, digested with 10 U of Ncol.
The Dral restriction site created in DNA fragments amplified from control genomic DNA using the reverse primer with a single nucleotide T for G mismatch (Fig. 4C) was not found in one allele of DNA amplified from GM01718A. Lanes 1 and 2 - non-digested control and GM01718A, lanes 3 and 4 - control and GM01718A, digested with 20 U of Dral.
This result supports the previous colocalisation of the sialidase gene with the HLA locus based on the clinical association of adrenal hyperplasia and sialidosis in a patient.
FISH
Sialidase cDNA subcloned in pCRII vector was nick-translated with biotin-16-dUTP (Bionick, BRL), and hybridized with lymphocyte chromosomes overnight at 37 °C according to Lichter and Cremer (Lichter, P. and Cremer, T. (1992) In : Rooney,D. E. , and Czepulkowski ,B. H. (eds) Human cytogenetics . A practical approach, Oxford University Press, N.Y. vol. 1:157-192). The biotinylated DNA was detected using rabbit anti-biotin antibodies (Enzo Diagnostic), biotinylated goat anti-rabbit antibodies (Sigma) and streptavidin- FITC (BRL) as recommended by manufacturer. Preparations were counter-stained with propidium iodide and R-banding was obtained with an anti-fade solution of p-phenylenediamine as described by Lemieux et al . (Lemieux,N. et al. (1992) Cytogenet . Cell . Genet . 59:311-312). Fluorescence microscopy was performed using a Axioskop microscope (Zeiss). Chromosomal localization by FISH method is shown on Fig. 7.