Cloning of the gene encoding a novel integral membrane protein, mucolipidin-and identification of the two major founder mutations causing mucolipidosis type IV

Abstract

Mucolipidosis type IV (MLIV) is an autosomal recessive lysosomal storage disorder characterized by severe psychomotor retardation and ophthalmologic abnormalities, including corneal opacity, retinal degeneration, and strabismus. Unlike the situation in other lysosomal disorders, the accumulation of heterogeneous storage material observed in MLIV does not result from a block in the catabolic pathways but is due to an ill-defined transport defect in the late steps of endocytosis. With the aim of cloning the MLIV gene, we searched in the 19p13.2-13.3 region, where the locus previously had been assigned by linkage mapping. In this region, we have identified a novel gene that is mutated in all patients with MLIV who were enrolled in our study. One patient was homozygous for the splice-acceptor mutation, and another was homozygous for a deletion removing the first six exons of the gene. In addition, four compound heterozygotes for these two mutations were identified. Haplotype analysis indicates that we have identified the two major founder mutations, which account for >95% of MLIV chromosomes in Ashkenazi Jewish patients. The gene, ML4, encodes a protein named "mucolipidin, " which localizes on the plasma membrane and, in the carboxy-terminal region, shows homologies to polycystin-2, the product of the polycystic kidney disease 2 gene (PKD2) and to the family of transient receptor potential Ca(2+) channels. Mucolipidin is likely to play an important role in endocytosis.

Figure 1

a, ORF map of ML4 cDNA. Short vertical bars denote possible translation start sites; long bars denote stop codons. The ML4 coding region is gray shaded. The 2,272-bp ML4 transcript contains an ORF of 1,635 bp encoding a protein of 545 amino acids. A 2,037-bp alternative spliced form, missing a 235-bp region that is denoted by the thick bar, also is shown. Since te size of the spliced region is not an exact multiple of 3, this alternative ML4 transcript presents a longer ORF (1,740 bp, encoding a 580-amino-acid protein), with the first Met located at nucleotide position 81. The proteins encoded by the two forms differ in their NH2 portions, whereas they are identical for the last 345 residues, because of the usage of the same reading frame in the two transcripts (the region denoted by the double-arrowheaded line) downstream of the splicing event. b, Northern blot analysis of the ML4 gene. A cDNA probe spanning the entire coding sequence of ML4 was hybridized to commercially available northern blot containing 2 μg of poly(A)⁺ RNA from various human adult tissues. A band of ∼2.2 kb, consistent with the size of the cDNA contig, is detected in all tissues tested, with the exception of colon and thymus. Hybridization to a human β-actin cDNA revealed almost equal amounts of poly(A)⁺ RNA in each lane.

Figure 2

Mutation Δ6432 in patients with MLIV. a, ML4 gene, which is organized into 13 exons. The unblackened boxes denote the ML4 exons removed by the Δ6432 deletion; the blackened boxes denote the exons that are retained. b, Δ6432 Interstitial deletion removing 6,432 bp of genomic DNA, denoted by the dotted line. The primer pair (i.e., ZFP2F oligonucleotide, designed on the basis of the sequence 2.5 kb upstream of the first base of the ML4 cDNA, and 1296R, designed on the basis of intron 7; see the Patients, Material, and Methods section) used to amplify the breakpoint is also shown. The electropherogram at the bottom left corresponds to the nucleotide sequence across the breakpoint as determined in patient GM02525. c, Segregation of the 1.87-kb junction fragment in patients WG909, GM02525, GM02526, GM02529, and GM02533A (see text) and in the carrier father, WG987. The wild-type allele could not be amplified by PCR, because of its large size (8.3 kb), as shown in control genomic DNA (lane C).

Figure 3

Mutation 486-2A→G in patients with MLIV. a, Lack of 165 bp, corresponding to exon 4 skipping, observed, at the cDNA level, by RT-PCR. The electropherogram shows the cDNA sequence across the deletion. The mutated allele encodes a truncated protein that retains only the first 21 amino acids of the wild-type allele. The region of homology to the PKD2 gene family is deleted in this mutant. b, 486-2A→G mutation, which segregates with the disease in the family of patient WG909. The electropherograms indicate the genomic sequences surrounding the 486-2A→G mutation. The mother is a carrier of the mutation, showing heterozygosity at the 486-2 position, whereas the father, who is a carrier of the Δ6432 allele, is hemizygous for a portion of the ML4 gene (see fig. 2b and text). The WG909 proband is a compound heterozygote for the 486-2A→G and Δ6432 mutations. c, RT-PCR of RNA from fibroblast cell lines, performed with primers located on exons 3 (MC21E) and 5 (MC21B) (see the Patients, Material, and Methods section). An altered size fragment, of 142 bp (lower arrow), is amplified in patients WG909, GM02526, GM02527, and GM02529 and in the carrier mother, WG988. The wild-type fragment, of 307 bp (upper arrow), is present only in the unaffected parents—WG987 and WG988—and in the fibroblast control cDNA sample. A plus sign (+) and a minus sign (−) denote, respectively, the presence and absence of RT in the RT-PCR. H₂O = the negative control. No amplification product is observed in patient GM02525, who is homozygous for the Δ6432 deletion, which removes this portion of the gene. The cDNA of patient GM02533A was not available.

Figure 4

MSA using the Clustal W algorithm and hydropathy plot of the mucolipidin protein. a, MSA between mucolipidin and related amino acid sequences found in protein databases. Mucolipidin = human mucolipidin polypeptide (545-amino-acid form); FLJ11006 = hypothetical protein FLJ11006 (GenBank accession number BAA91951) predicted from a cDNA clone identified by a large-scale cDNA sequencing approach; CG8743 (GenBank accession number AE003516) and CE25082 (GenBank accession number not available) = predicted proteins from genomic sequences of Drosophila melanogaster and Caenorhabditis elegans, respectively. Identical residues (shaded) and conservatively substituted residues (unshaded) are boxed. The gaps inserted by the Clustal W program are denoted by broken lines. b, Top, Hydropathy plot of the mucolipidin protein (545-amino-acid form). The amino acid position is plotted as a function of the average hydrophobicity of an 11-amino-acid peptide (Kyte and Doolittle 1982). The region corresponding to the TRPC-like profile is shown above the plot, as a thick diagonally striped line. Bottom, MSA between a 136-amino-acid sequence of mucolipidin (amino acids 354–490, also indicated as a shaded box in the hydropathy plot) and a region of human proteins belonging to the polycystic kidney disease 2 family. PKDREJ = polycystic kidney disease and receptor for egg jelly (sea urchin homologue)–like protein (GenBank accession number NP_006062); PKD2L2 = polycystic kidney disease 2–like 2 (GenBank accession number AAF65622); PKD2 = polycystic kidney disease 2 protein (GenBank accession number NP_032887); PKD2L1 = polycystic kidney disease 2–like 1 protein (GenBank accession number NP_057196). The first and last amino acids of each sequence, numbered according to their database entries, are at the end of each sequence.

Figure 5

Immunofluorescence of ML4 encoded protein. COS7 cells were transfected with either 129-pMHA (a) or 129-pMT21 (b) constructs and were grown 48 h prior to fixation and immunofluorescence staining. Cells were fixed with paraformaldehyde and were permeabilized with Triton X-100 and then were treated with either anti-HA (A) or anti-Myc (B) monoclonal antibodies. Staining was performed with fluorescein 5–isothiocyanated isotype-specific antibodies.