Portal de Periódicos da CAPES

Extracellular Matrix Protein 1 Gene (ECM1) Mutations in Lipoid Proteinosis and Genotype-Phenotype Correlation

2003; Elsevier BV; Volume: 120; Issue: 3 Linguagem: Inglês

10.1046/j.1523-1747.2003.12073.x

1523-1747

Takahiro Hamada, Vesarat Wessagowit, Andrew P. South, G H S Ashton, Ien Chan, Noritaka Oyama, Apatorn Siriwattana, Prachiya Jewhasuchin, Somyot Charuwichitratana, Devinder Mohan Thappa, Patsy Lenane, Bernice R. Krafchik, Kanokvalai Kulthanan, Hiroshi Shimizu, Tamer Kaya, Mehmet Emin Erdal, Mauro Paradisi, Amy S. Paller, Mariko Seishima, Takashi Hashimoto, John A. McGrath,

Urologic and reproductive health conditions

The autosomal recessive disorder lipoid proteinosis results from mutations in extracellular matrix protein 1 (ECM1), a glycoprotein expressed in several tissues (including skin) and composed of two alternatively spliced isoforms, ECM1a and ECM1b, the latter lacking exon 7 of this 10-exon gene (ECM1). To date, mutations that either affect ECM1a alone or perturb both ECM1 transcripts have been demonstrated in six cases. However, lipoid proteinosis is clinically heterogeneous with affected individuals displaying differing degrees of skin scarring and infiltration, variable signs of hoarseness and respiratory distress, and in some cases neurological abnormalities such as temporal lobe epilepsy. In this study, we sequenced ECM1 in 10 further unrelated patients with lipoid proteinosis to extend genotype-phenotype correlation and to add to the mutation database. We identified seven new homozygous nonsense or frameshift mutations: R53X (exon 3); 243delG (exon 4); 507delT (exon 6); 735delTG (exon 7); 785delA (exon 7); 892delC (exon 7) and 1190insC (exon 8), as well as two new compound heterozygous mutations: W160X/F167I (exon 6) and 542insAA/R243X (exons 6/7), none of which were found in controls. The mutation 507delT occurred in two unrelated subjects on different ECM1 haplotypes and may therefore represent a recurrent mutation in lipoid proteinosis. Taken with the previously documented mutations in ECM1, this study supports the view that exons 6 and 7 are the most common sites for ECM1 mutations in lipoid proteinosis. Clinically, it appears that mutations outside exon 7 are usually associated with a slightly more severe mucocutaneous lipoid proteinosis phenotype, but neurological features do not show any specific genotype-phenotype correlation. The autosomal recessive disorder lipoid proteinosis results from mutations in extracellular matrix protein 1 (ECM1), a glycoprotein expressed in several tissues (including skin) and composed of two alternatively spliced isoforms, ECM1a and ECM1b, the latter lacking exon 7 of this 10-exon gene (ECM1). To date, mutations that either affect ECM1a alone or perturb both ECM1 transcripts have been demonstrated in six cases. However, lipoid proteinosis is clinically heterogeneous with affected individuals displaying differing degrees of skin scarring and infiltration, variable signs of hoarseness and respiratory distress, and in some cases neurological abnormalities such as temporal lobe epilepsy. In this study, we sequenced ECM1 in 10 further unrelated patients with lipoid proteinosis to extend genotype-phenotype correlation and to add to the mutation database. We identified seven new homozygous nonsense or frameshift mutations: R53X (exon 3); 243delG (exon 4); 507delT (exon 6); 735delTG (exon 7); 785delA (exon 7); 892delC (exon 7) and 1190insC (exon 8), as well as two new compound heterozygous mutations: W160X/F167I (exon 6) and 542insAA/R243X (exons 6/7), none of which were found in controls. The mutation 507delT occurred in two unrelated subjects on different ECM1 haplotypes and may therefore represent a recurrent mutation in lipoid proteinosis. Taken with the previously documented mutations in ECM1, this study supports the view that exons 6 and 7 are the most common sites for ECM1 mutations in lipoid proteinosis. Clinically, it appears that mutations outside exon 7 are usually associated with a slightly more severe mucocutaneous lipoid proteinosis phenotype, but neurological features do not show any specific genotype-phenotype correlation. lipoid proteinosis extracellular matrix protein 1 Polymerase chain reaction Lipoid proteinosis (LP), also known as hyalinosis cutis et mucosae or Urbach-Wiethe disease (OMIM 247100), is a rare autosomal recessive genodermatosis characterized by a variable degree of scarring and infiltration of skin and mucosae (Hofer, 1973Hofer P. Urbach-Wiethe disease (lipoglycoproteinosis; lipoid proteinosis; hyalinosis cutis et mucosae).Acta Derm. 1973; 53 (A review (Stockh)): 1-52Google Scholar;Hamada, 2002Hamada T. Lipoid proteinosis.Clin Exp Dermatol. 2002; 27: 624-629Crossref PubMed Scopus (120) Google Scholar). The disorder usually presents in early infancy with hoarseness and thickening of vocal cords. During childhood, the skin may be easily damaged by minor trauma or friction, resulting in blisters and varicelliform or acneiform scar formation. Subsequently, infiltration of the mucosae may lead to respiratory difficulty, especially in association with upper respiratory tract infection, sometimes requiring tracheostomy (Ramsey et al., 1985Ramsey M.L. Tschen J.A. Wolf Jr, J.E. Lipoid proteinosis.Int J Dermatol. 1985; 24: 230-232Crossref PubMed Scopus (20) Google Scholar). In addition, recurrent episodes of inflamed parotid or submandibular glands, poor dental hygiene, and shortening of the tongue with a thickened frenulum may also occur (Disdier et al., 1994Disdier P. Harle J.R. Andrac L. Swiader L. Weiller P.J. Specific xerostomia during Urbach-Wiethe disease.Dermatology. 1994; 188: 50-51Crossref PubMed Scopus (15) Google Scholar;Aroni et al., 1998Aroni K. Lazaris A.C. Papadimitriou K. Paraskevakou H. Davaris P.S. Lipoid proteinosis of the oral mucosa: Case report and review of the literature.Pathol Res Pract. 1998; 194: 855-859Crossref PubMed Scopus (28) Google Scholar). Infiltration of the skin may present as grouped warty plaques on the dorsa of the hands and elbows. Infiltration of the eyelid margins results in beaded papules referred to as moniliform blepharosis. Other extracutaneous features may include epilepsy and neuropsychiatric abnormalities, sometimes in association with calcification in the temporal lobes or hippocampi (Friedman et al., 1984Friedman L. Mathews R.D. Swanepoel P.D. Radiographic and computed tomographic findings in lipid proteinosis: A case report.S Afr Med J. 1984; 65: 734-735PubMed Google Scholar;Kleinert et al., 1987Kleinert R. Cervos-Navarro J. Kleinert G. Walter G.F. Steiner H. Predominantly cerebral manifestation in Urbach–Wiethe's syndrome (lipoid proteinosis cutis et mucosae): A clinical and pathomorphological study.Clin Neuropathol. 1987; 6: 43-45PubMed Google Scholar). Thus the clinical manifestations of lipoid proteinosis are protean and may vary considerably between affected individuals (Hofer, 1973Hofer P. Urbach-Wiethe disease (lipoglycoproteinosis; lipoid proteinosis; hyalinosis cutis et mucosae).Acta Derm. 1973; 53 (A review (Stockh)): 1-52Google Scholar). The molecular basis of lipoid proteinosis was recently shown to result from mutations in the gene encoding extracellular matrix protein 1 (ECM1) (Hamada et al., 2002Hamada T. McLean W.H.I. Ramsay M. et al.Lipoid proteinosis maps to 1q21 and is caused by mutations in the extracellular matrix protein 1 gene (ECM1).Hum Mol Genet. 2002; 11: 833-840Crossref PubMed Scopus (207) Google Scholar). There are two alternatively spliced forms of ECM1, ECM1a and ECM1b, the latter lacking exon 7 of this 10-exon gene. Both transcripts are expressed in skin and upper respiratory tract, although ECM1a is also distributed more widely, including the placenta, heart, liver, small intestine, lung, ovary, prostate, testis, skeletal muscle, pancreas and kidney (Smits et al., 1997Smits P. Ni J. Feng P. et al.The human extracellular matrix gene 1 (ECM1): Genomic structure, cDNA cloning, expression pattern and chromosomal localization.Genomics. 1997; 45: 487-495Crossref PubMed Scopus (68) Google Scholar). Previously, six different homozygous mutations in ECM1 have been described, four that affect ECM1a (but not ECM1b), and two that are predicted to disrupt both ECM1a and ECM1b. In this study, we investigated ECM1 gene pathology in 10 further unrelated individuals with lipoid proteinosis to extend the mutation spectrum in this genodermatosis and to help define genotype-phenotype correlation. Following informed consent, DNA was extracted from peripheral blood samples taken from affected individuals, their parents and clinically normal siblings (where possible) using a standard cold water lysis method. Polymerase chain reaction (PCR) amplification of the ECM1 gene was performed using eight pairs of primers spanning all 10 exons and situated in flanking introns (Table I). For PCR amplification, 250 ng of genomic DNA was used as the template in an amplification buffer containing 6.25 pmol of the primers, 37.5 nmol MgCl2, 5 mmol of each nucleotide triphosphate and 2.5 U Taq polymerase (Applied Biosystems, Warrington, U.K.) in a total volume of 50 μL in an OmniGene thermal cycler (Hybaid, Basingstoke, U.K.). The amplification conditions were 95°C for 5 min, followed by 38 cycles of 95°C for 45 s, annealing temperature (see Table I) for 45 s, 72°C for 45 s. Aliquots (5 μL) of the PCR products were analysed by 2.5% agarose gel electrophoresis. PCR products were then purified using QIAquick PCR Purification Kit (Qiagen, Crawley, U.K.) and sequenced directly using Big Dye labeling in an ABI 310 genetic analyser (Applied Biosystems). Potential mutations were confirmed by restriction endonuclease digestion or bi-directional sequencing and assessed in up to 200 control chromosomes, selected to match the patients' ethnic backgrounds. The control samples were selected from 3000 archival DNA samples kept in the Department of Cell and Molecular Pathology at St John's Institute of Dermatology.Table IIntronic genomic primers used for PCR amplification of ECM1Exon no.Forward primer (5′→3′)Reverse primer (5′→3′)Annealing temp. (°C)Product size (bp)1agctgggactgagtcatggctaaaggctccactggcctag624162/3tcctacactcttgatctccaggtgtcaacaggatccatag606224/5cagtgaccctccaggtttctcagagcccaccgtcttgtct604846agccttgagaagcaggaggaagtgaacgggacctgaggtt606717ttatctgcctgcccagtgtcacatggatggatggactggc605488cacatcaacagttgcctcctggcatcttctggcatcagat604999agttgcctagtccttccccaaggccaggtcagagtgaaga6040810aatccagctgtgcaaggcaggtaatgagtgttcagatggg62469 Open table in a new tab To search for common sequence variants in ECM1, genomic DNA from 40 control individuals was amplified using the PCR primers and methods detailed above and then screened using heteroduplex analysis, as described elsewhere (Ganguly et al., 1993Ganguly A. Rock M.J. Prockop D.J. Conformation-sensitive gel electrophoresis for rapid detection of single-base differences in double-stranded PCR products and DNA fragments: Evidence for solvent-induced bends in DNA heteroduplexes.Proc Natl Acad Sci USA. 1993; 90: 10325-10329Crossref PubMed Scopus (604) Google Scholar). PCR products displaying bandshifts were then investigated further by direct nucleotide sequencing and subsequent calculation of allelic frequencies using restriction endonuclease digestions, according to the manufacturer's recommendations (New England BioLabs, Hitchin, U.K.). Following informed consent, biopsies from non-lesional, upper arm skin were taken under local anesthetic from two affected individuals (patients 1 and 2) and from the same site in control subjects of similar age. The epidermis was separated from the dermis using Dispase (Sigma-Aldrich, Poole, U.K.) in Dulbecco's phosphate-buffered saline (Gibco-BRL, Paisley, U.K.) for 2 h at 37°C, as described previously (Bleck et al., 1999Bleck O. Abeck D. Ring J. et al.Two ceramide subfractions detectable in Cer (AS) position by HPTLC in skin surface lipids of non-lesional skin of atopic eczema.J Invest Dermatol. 1999; 113: 894-900Crossref PubMed Scopus (116) Google Scholar). Keratinocytes from the epidermal extract were cultured, in the presence of a 3T3 feeder layer (Navsaria et al., 1994Navsaria H.A. Sexton C. Bouvard V. Leigh I.M. Growth of keratinocytes with a 3T3 feeder layer: Basic techniques.in: Leigh I.M. Watt Keratinocyte Methods. Cambridge University Press, Cambridge1994: 5-12Google Scholar), in Dulbecco's modified Eagle's medium (Gibco-BRL) and Ham's F12 medium (Gibco-BRL) in a ratio of 3:1 (v/v) supplemented with 10% fetal bovine serum (ICN Biomedicals, Hampshire, U.K.), 0.4 μg/ml hydrocortisone (Sigma-Aldrich), 10−10 M cholera toxin (Sigma-Aldrich), 10 ng/ml epidermal growth factor (Gibco-BRL), 5 mg/ml insulin (Gibco-BRL), 1% penicillin/streptomycin (Sigma-Aldrich) and 1% pimaricin (Autogenbioclear, Wiltshire, U.K.). For RNA extraction from cultured keratinocytes, a confluent T75 flask was used. RNA extraction with on-column DNase digestion was performed using a commercial kit (RNeasy Midi Kits and Rnase-Free DNase Set, Qiagen). RNA was quantified in a Model 6200 fluorimeter (Jenway, Dunmow, U.K.) using RiboGreen dye (Molecular Probes Europe, Leiden, The Netherlands) and a standard curve fitting created using ExcelTM (Microsoft, Reading, U.K.) for a range of known control concentrations. For reverse transcription, two identical reactions for each RNA sample were established comprising 5 μg DNase I-treated RNA, 7 μL 5X M-MLV RT-buffer (Promega, Southampton, U.K.), 4 μL 10 mM dNTPs, 1 μL RNasin Riobonuclease inhibitor (Promega) and 1 μL random primers (ICN). Reactions were incubated at 65°C for 10 min and then placed on ice. After cooling, 2 μL M-MLV reverse transcriptase (400 units) was added to one of the duplicate reactions, whereas 2 μL water was added to the other (negative control). Samples were then incubated at 42°C for 90 min and then stored at – 20°C. PCR for house-keeping genes L32 and GAPDH were used to compare the quantity of the cDNA template and to exclude genomic contamination. A pair of primers for RT-PCR across the sites of the 243delT and 507delT mutations in patients 1 and 2 (see Results) comprised: forward primer 5′-TACAGGACAGAGGCAGCTGA-3′ (nucleotides 81–100; GenBank XM_002030, with initiation codon ATG as nucleotide 1-2-3), reverse primer 5′-GAACTCGGCCTCACAGAATC-3′ (nucleotides 747–728). The reverse primer for RT-PCR was placed within exon 7. Therefore, ECM1a but not ECM1b was specifically amplified by this RT-PCR reaction using these primers. The size of the expected RT-PCR product was 667-bp. The ampli-fication conditions were 95°C for 5 min, followed by 38 cycles of 95°C for 45 s, 60°C for 45 s, 72°C for 45 s. For the PCR reactions, 1 μL of the solution with the first strand cDNA was used as a template in a 50μL reaction containing 6.25 pmol of each primers, 37.5 nmol of MgCl2, 5 mmol of each nucleotide and 2.5 U of Taq polymerase (Applied Biosystems) in an OmniGene thermal cycler (Hybaid). The RT-PCR products were subcloned into a PCR-compatible cloning vector (TOPO TA Cloning Kit, Invitrogen, Paisley, U.K.) and sequenced directly using Big-Dye labeling in an ABI 310 genetic analyser (Applied Biosystems). Patient 1 was a 38-year-old Thai male (Fig 1a–c). He had severe hoarseness since birth, as well as extensive skin scarring and infiltration with prominent hyperkeratotic erosions on both legs. He had a history of occasional convulsions and aura in keeping with temporal lobe epilepsy. He had a similarly affected 19-year-old brother (Fig 1e). One other affected sibling died aged 34-years from acute-on-chronic upper airway obstruction. His parents were not known to be consanguineous but came from the same rural district. Patient 2 was a 30-year-old Thai female, unrelated to patient 1. She had mild hoarseness since early childhood but minimal skin signs apart from a few beaded papules on both upper eyelids (Fig 1d), starting aged 5-years, and slight thickness of the frenulum with restricted tongue movement. She also suffered from occasional episodes of collapse with transient loss of consciousness suggestive of petit mal epilepsy. There was no family history and no reported consanguinity, although the parents originated from the same province in Southern Thailand. Patient 3 was a 23-year-old Indian male who had hoarseness and progressive infiltrated papules and scarring (starting on the face) since infancy. He also had pock-like scarring on the face and upper trunk, warty infiltrated plaques in both groins and thickening of his oral mucosa. He had no history of epilepsy but a computed tomography scan showed bilateral temporal lobe calcification. He had a younger sister with LP and the parents were known to be first cousins. Patient 4 was a 16-year-old Japanese male, for whom full clinico-pathological details have been described recently (Nagasaka et al., 2000Nagasaka T. Tanaka M. Ito D. Tanaka K. Shimizu H. Protean manifestations of lipoid proteinosis in a 16-year-old boy.Clin Exp Dermatol. 2000; 25: 30-32Crossref PubMed Scopus (10) Google Scholar). Briefly, he had hoarseness, refractory temporal lobe epilepsy, blisters, scars, eyelid papules and extensive papular skin infiltration. He also had radiographic signs of temporal lobe calcification. His parents were non-consanguineous. Patient 5 was a 3-year-old Canadian male from an Iranian family, in which the parents were first cousins. He had a hoarse voice since birth, and scattered skin crusts and erosions particularly on his upper trunk and face starting in the second year of life. These lesions were followed by scarring that resembled hydroa vacciniforme. There were no neurological abnormalities. No other family members were affected. Patient 6 was a 14-year-old Turkish female from a consanguineous family. A 4-year-old sister was also affected. Clinico-pathological details of this family have been published recently (Kaya et al., 2002Kaya T.I. Gunduz O. Kokturk A. Tursen U. Ikizoglu G. A life-threatening exacerbation of lipoid proteinosis.J Eur Acad Dermatol Venereol. 2002; 16: 286-288Crossref PubMed Scopus (15) Google Scholar). Briefly, she had hoarseness since birth and progressive varicelliform scars and warty skin papules. She also showed radiographic temporal lobe calcification. Patient 7 was a 17-year-old American female from an Iranian family. Her parents were third cousins. She developed erosive skin lesions and hoarseness from infancy. The skin erosions resolved as linear and cribriform scarring. Periorbital infiltrated papules were first noted at the age of 9 years, and have progressively increased in number. Patient 8 was a 5-year-old Italian female born to non-consanguineous parents. She developed hoarseness during the first few months of life. At the age of 3 years, she started to get post-traumatic crusting and scarring on her face as well as signs of beaded eyelid papules. To date, there have been no other significant skin or mucosal changes. Patient 9 was a 2-year-old Canadian male of central European origin. His parents were non-consanguineous. He had hoarseness since birth, as well as progressive skin scarring and infiltration. Although his mother was clinically normal, his maternal grandmother was reported to have LP (further details not available). No other family members were affected. Patient 10 was a 53-year-old Japanese male. He had severe hoarseness since early childhood and progressive skin scarring and infiltration. The degree of skin thickening in some sites was particularly prominent and warty skin plaques were present on the elbows and knees. Several papules were also noted on the eyelid margins. No neurological features had been noted at any stage. He had an 80-year-old brother with similar skin features. However, the brother had very severe laryngeal infiltration and had a tracheostomy inserted at the age of 67 years. Their parents were not known to be consanguineous but came from the same region. To facilitate detection of pathogenetic mutations in LP, we designed genomic DNA primers to amplify exons 1–10 of ECM1, with primers sited >50 bp away from the intron-exon borders. Primer sequence details, the optimized amplification conditions and the expected sizes of PCR products are indicated in Table I. Heteroduplex analysis of amplified DNA from normal controls identified similar bandshifts in several of the PCR products spanning exon 6 and exon 8. Sequencing revealed two point mutations, a C>T transition at nucleotide 389 in exon 6 that converts a threonine residue (ACG) to methionine (ATG), and a G>A substitution at nucleotide 1243 in exon 8 that changes glycine (GGT) to serine (AGT). These sequence variants altered cut sites for the restriction endonucleases Bsm I and BsiE I, respectively. For the 389C/T polymorphism, the major allele was C (0.61) and the minor allele was T (0.39); 72 chromosomes (control from mixed ethnic backgrounds) were examined. For the 1243G/A polymorphism, the major allele was G (0.53) and the minor allele was A (0.47); 64 chromosomes were examined. Sequencing of genomic DNA in the 10 unrelated patients with LP disclosed pathogenic mutations in each case, as detailed in Table II. Cases 1–7 were homozygous for small internal deletion/insertion mutations that lead to frame-shift and downstream premature termination codons: three of these mutations occured within the alternatively spliced exon 7. The mutation 507delT (Fig 2) was identified in two unrelated patients but was shown to have arisen on different haplo-types: patient 1 was homozygous for methionine at amino acid position 130, whereas patient 6 was homozygous for threonine (see details of intragenic exon 6 polymorphism above). Case 8 was a compound heterozygote for a frameshift/nonsense combination of mutations. Case 9 was a compound heterozygote for a nonsense/missense combination of mutations. Case 10 was homozygous for a nonsense mutation. All nonsense and frameshift mutations were verified by restriction endo-nuclease digestion and/or bi-directional nucleotide sequencing, confirmed in heterozygous family members, and excluded in 80–120 control chromosomes. At least 80 of the control chromosomes used to exclude the frameshift or nonsense mutations were accurately matched with the ethnic background of the case of LP harboring that particular mutation. To exclude the possibility of the missense mutation F167I being a non-pathogenic polymorphism, this sequence change was excluded in 200 ethnically matched control chromosomes. This control group comprised Caucasian Southern European DNA. These new mutations, and the previously documented mutations in other patients with LP, are illustrated in Fig 3.Table IINew pathogenic ECM1 mutations detected in this studyPatientMutation (s)Exon (s)Verification1507delT/507delT6Hpa II2243delG/243delG4Hpy178 III3785delA/785delA7Fau I4892delC/892delC7Seq5507delT/507delT7Hpa II6735delTG/735delTG6Hpy188 I71190insC/1190insC8Bgl I8542insAA/R243X6/7Hpa II/ Seq9W160X/F167I6/6Mnl I /SfaN I10R53X/R53X3SeqSeq=sequencing Open table in a new tab Figure 3Illustration of the positions of 11 new mutations and the previous ECM1 gene mutations. New mutations are depicted below and previously reported mutations above the gene structure. Most mutations comprise frameshift or nonsense mutations located in exon 6 or the alternatively spliced exon 7. The homozygous mutation 507delT has been detected in two unrelated individuals on different ECM1 haplotypes. Most of the mutations in LP patients are homozygous (double arrow head).View Large Image Figure ViewerDownload (PPT) Seq=sequencing In general, patients with mutations in exon 7 tended to have milder upper respiratory tract abnormalities and less skin scarring/infiltration compared to patients with mutations in other parts of the gene. However, the differences were often subtle and evidence for intrafamilial variability was seen in families with more than one affected member (patients 1, 3 and 6) as well as interfamilial variability for unrelated patients with the same mutation (patients 1 and 5). Direct comparisons between some affected individuals were difficult because of the different ages of the patients and the typical progressive nature of the skin scarring and infiltration with increasing age. The mildest clinical signs were noted in patient 2, who was homozygous for the mutation 243delG in exon 4. In cultured keratinocyte cDNA from patients 1 and 2, RT-PCR identified multiple different PCR bands compared to a single 667-bp band in the control sample (Fig 4a). Subcloning and direct sequencing disclosed multiple aberrant mutant transcripts, most of which were out-of-frame. However, evidence for restoration of the open reading frame was noted for one mutant transcript in patient 2 with the clinically mild phenotype. Specifically, sequencing of a 484-bp band revealed evidence for aberrant splicing with in-frame skipping of exons 3 and 4 leading to deletion of 61 amino acids from the cysteine-free domain (Fig 4b). No other in-frame transcripts corresponding to the remaining multiple different PCR bands from both samples were identified. In this study, we have identified 11 new mutations in the ECM1 gene in patients with LP. Details of these and the previous ECM1 mutations are illustrated in Fig 3. The combined data thus far suggest that the majority of mutations occur in exon 7. This exon is alternatively spliced and we have previously shown that frameshift mutations in exon 7 lead to ablation of the ECM1a transcript, but not the shorter ECM1b transcript that normally lacks this exon (Hamada et al., 2002Hamada T. McLean W.H.I. Ramsay M. et al.Lipoid proteinosis maps to 1q21 and is caused by mutations in the extracellular matrix protein 1 gene (ECM1).Hum Mol Genet. 2002; 11: 833-840Crossref PubMed Scopus (207) Google Scholar). Outside exon 7, most other mutations are found in exon 6. Collectively, these two exons contain 24/32 of the mutated ECM1 alleles delineated thus far. These findings have implications for mutation detection strategies in LP and we recommend initial sequencing of exon 7, then exon 6, and then the remainder of the ECM1 gene in the investigation of any new patients with this genodermatosis. Homozygous nonsense or frameshift mutations in exon 6 are predicted to affect both full-length ECM1a and ECM1b transcripts, whereas ECM1b should be unaffected for similar types of mutation in exon 7 (Hamada et al., 2002Hamada T. McLean W.H.I. Ramsay M. et al.Lipoid proteinosis maps to 1q21 and is caused by mutations in the extracellular matrix protein 1 gene (ECM1).Hum Mol Genet. 2002; 11: 833-840Crossref PubMed Scopus (207) Google Scholar). This provides an opportunity to review the phenotype of LP patients harboring ECM1a/ECM1b genotype pathologies. Clinical assessment of our series of patients suggests that individuals with mutations in exon 7 have a slightly milder phenotype than individuals with mutations outside exon 7. These observations are true for respiratory and skin manifestations of LP, but do not appear to relate to neuropathological abnormalities such as temporal lobe epilepsy and/or brain calcification. Indeed, a calcium-binding domain has been identified in exon 7 of ECM1 and this might explain the lack of specific genotype-phenotype correlation for these particular features (Bhalerao et al., 1995Bhalerao J. Tylzanowski P. Filie J.D. Kozak C.A. Merregaert J. Molecular cloning, characterization and genetic mapping of the cDNA coding for a novel secretory protein of mouse. Demonstration of alternative splicing in skin and cartilage.J Biol Chem. 1995; 270: 16385-16394Crossref PubMed Scopus (40) Google Scholar;Smits et al., 1997Smits P. Ni J. Feng P. et al.The human extracellular matrix gene 1 (ECM1): Genomic structure, cDNA cloning, expression pattern and chromosomal localization.Genomics. 1997; 45: 487-495Crossref PubMed Scopus (68) Google Scholar). A further difficulty in comparing clinical signs in different patients relates to the patients' age. Typically in LP, there is an initial phase of skin blisters and crusting followed by later changes of progressive scarring and skin infiltration. Thus, comparing signs in an infant with those in an adult is likely to reflect more than just the inherent ECM1 genotype. Likewise, it is evident that trauma does contribute to the severity or extent of skin changes since many of the scars and infiltrated skin plaques are induced or exacerbated by physical trauma. Nevertheless, the observations that patients with mutations in exon 7 have a similar (if very slightly milder) phenotype to other cases with mutations elsewhere does provide support for the ECM1a transcript having a more significant overall biological role in human skin and mucosae when compared to the ECM1b transcript. Interestingly, our patient with the mildest clinical features had a homozygous frameshift mutation in exon 4 that would be predicted to disrupt both ECM1a and ECM1b. However, we found evidence for partial rescue of the mutant transcript through in-frame skipping of the exon containing this mutation and the adjacent upstream exon, giving a shorter transcript lacking 61 amino acids. This part of the protein is upstream from most of the known functional domains of ECM1 (e.g., ligand and calcium binding and predicted protein loops) (Bhalerao et al., 1995Bhalerao J. Tylzanowski P. Filie J.D. Kozak C.A. Merregaert J. Molecular cloning, characterization and genetic mapping of the cDNA coding for a novel secretory protein of mouse. Demonstration of alternative splicing in skin and cartilage.J Biol Chem. 1995; 270: 16385-16394Crossref PubMed Scopus (40) Google Scholar;Johnson et al., 1997Johnson M.R. Wilkin D.J. Vos H.L. de Ortiz Luna R.I. Dehejia A.M. Polymeropoulos M.H. Francomano C.A. Characterization of the human extracellular matrix protein 1 gene on chromosome 1q21.Matrix Biol. 1997; 16: 289-292Crossref PubMed Scopus (28) Google Scholar;Smits et al., 1997Smits P. Ni J. Feng P. et al.The human extracellular matrix gene 1 (ECM1): Genomic structure, cDNA cloning, expression pattern and chromosomal localization.Genomics. 1997; 45: 487-495Crossref PubMed Scopus (68) Google Scholar). Similar skipping of exons containing frameshift mutations has been described in other genodermatoses to account for unexpectedly mild phenotypes (McGrath et al., 1999McGrath J.A. Ashton G.H. Mellerio J.E. Salas-Alanis J.C. Swensson O. McMillan J.R. Eady R.A. Moderation of phenotypic severity in dystrophic and junctional forms of epidermolysis bullosa through in-frame skipping of exons containing non-sense or frameshift mutations.J Invest Dermatol. 1999; 113: 314-321Crossref PubMed Scopus (62) Google Scholar). Thus, the effects of RNA processing are important in evaluating genotype-phenotype correlation in LP. The mutation 507delT in exon 6 was identified in two unrelated patients and was shown to have arisen on different ECM1 haplotypes. Specifically, the patients had different amino acid polymorphisms at codon 130 in the same exon harboring the pathogenic mutations. Review of the nucleotide sequence surrounding the deleted T reveals palindromic runs of multiple C or G repeats, suggesting that this part of the ECM1 gene might be prone to slipped mispairing during replication, as reported for several other recurrent mutations throughout the genome (Denoyelle et al., 1997Denoyelle F. Weil D. Maw M.A. et al.Prelingual deafness: High prevalence of a 30delG mutation in the connexin 26 gene.Hum Mol Genet. 1997; 6: 2173-2177Crossref PubMed Scopus (535) Google Scholar;Schimmenti et al., 1997Schimmenti L.A. Cunliffe H.E. McNoe L.A. et al.Further delineation of renal–coloboma syndrome in patients with extreme variability of phenotype and identical PAX2 mutations.Am J Hum Genet. 1997; 60: 869-878PubMed Google Scholar). Nearly all the ECM1 mutations in LP reported thus far are homozygous, reflecting that most cases of this rare genodermatosis occur in consanguineous families. However, in this study we identified two compound heterozygote mutations. One of these patients had a nonsense/frameshift combination of mutations but the other patient harbored nonsense/missense mutations. The missense mutation F167I was excluded as a common or rare polymorphism by screening 200 control chromosomes. Although this amino acid change represents substitution of one hydrophobic neutral amino acid by another, F167 is a conserved residue (present in mouse ECM1 sequence) and the amino acid change does markedly alter the size of the amino acid side chain. The substitution occurs near the start of the first tandemly-duplicated domain of ECM1 (regions of the protein that have homology to albumin-binding loops) and forms part of a highly conserved FPPG amino acid sequence characteristic of these tandem duplications (Bhalerao et al., 1995Bhalerao J. Tylzanowski P. Filie J.D. Kozak C.A. Merregaert J. Molecular cloning, characterization and genetic mapping of the cDNA coding for a novel secretory protein of mouse. Demonstration of alternative splicing in skin and cartilage.J Biol Chem. 1995; 270: 16385-16394Crossref PubMed Scopus (40) Google Scholar;Smits et al., 1997Smits P. Ni J. Feng P. et al.The human extracellular matrix gene 1 (ECM1): Genomic structure, cDNA cloning, expression pattern and chromosomal localization.Genomics. 1997; 45: 487-495Crossref PubMed Scopus (68) Google Scholar). Nevertheless, the precise consequences of this missense mutation on ECM1 protein function are not yet known. Indeed, the biological function of ECM1 is only starting to become clear, largely through elucidation of ECM1 as the pathogenic gene in LP. Initially, ECM1 had been shown to inhibit bone mineralization, to contribute to epidermal differentiation, and to stimulate angiogenesis (Smits et al., 2000Smits P. Poumay Y. Karperien M. et al.Differentiation-dependent alternative splicing and expression of the extracellular matrix protein 1 gene in human keratinocytes.J Invest Dermatol. 2000; 114: 718-724Crossref PubMed Scopus (51) Google Scholar;Deckers et al., 2001Deckers M.M. Smits P. Karperien M. et al.Recombinant human extracellular matrix protein 1 inhibits alkaline phosphatase activity and mineralization of mouse embryonic metatarsals in vitro.Bone. 2001; 28: 14-20Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar;Han et al., 2001Han Z. Ni J. Smits P. et al.Extracellular matrix protein 1 (ECM1) has angiogenic properties and is expressed by breast tumor cells.FASEB J. 2001; 15: 988-994Crossref PubMed Scopus (103) Google Scholar). However, demonstration of mutations in LP has emphasized the significance of ECM1, particularly the ECM1a isoform, in the physiology and biology of the dermis. It is likely that ECM1 contributes to protein binding of interstitial collagens, basement membrane collagens and glycosaminoglycans since lack of the protein (as occurs in LP) results in massive reduplication of basement membrane (Olsen et al., 1988Olsen D.R. Chu M.L. Uitto J. Expression of basement membrane zone genes coding for type IV procollagen and laminin by human skin fibroblasts in vitro: Elevated alpha 1 (IV) collagen mRNA levels in lipoid proteinosis.J Invest Dermatol. 1988; 90: 734-738Abstract Full Text PDF PubMed Google Scholar) as well as the clinical features of scarring and skin infiltration. However, thus far there are no reports detailing the specific protein–protein interactions of ECM1 in skin or other tissues. In this study, we have elucidated the molecular basis of LP in 10 unrelated patients, identified two common intragenic polymorphisms, and developed a mutation detection strategy for screening DNA from patients with this genodermatosis. The results demonstrate the mutation spectrum in LP and identify some of the practical difficulties in correlating genotype with phenotype. This work was supported by the British Skin Foundation and also by the Dystrophic Epidermolysis Bullosa Research Association (DEBRA U.K.) and Action Research, as well as a Butterfield Award from the Great Britain Sasakawa Foundation, promoting collaborative medical research between British and Japanese institutions.