RESUMOS DO 51^O CONGRESSO ANUAL DA SOCIEDADE AMERICANA DE GENÉTICA HUMANA

Dp260 isoform of dystrophin is able to bind actin and produces a Becker-like phenotype in transgenic mdx mice. L.E. Warner¹, C. Dello Russo¹, R.W. Crawford¹, J.M. Ervasti², J.S. Chamberlain¹. 1) Dept Neurology, Univ Washington, Seattle, WA; 2) Dept Physiology, Univ Wisconsin, Madison, WI.
   Duchenne muscular dystrophy (DMD) is caused by defects in the dystrophin gene. DMD is clinically characterized by severe, progressive muscle degeneration and weakness. In muscle, dystrophin is thought to play a mechanical role linking the actin cytoskeleton to the extracellular matrix through its assembly in a larger complex called the dystrophin glycoprotein complex (DGC). This link helps maintain muscle membrane integrity in part by dissipating the forces of muscle contraction into the extracellular matrix. Loss of the DGC results in contraction-induced injury and muscle degeneration in mdx mice and DMD patients. Dystrophin has an N-terminal actin-binding domain (ABD) as well as an internal ABD within the central rod region (repeats 11-17). Deletion of the internal ABD does not significantly affect the ability of dystrophin to bind actin or prevent dystrophy. However, it is not known whether the internal ABD alone could bind actin in vivo and reverse the dystrophic phenotype in mdx mice. To test this concept, we constructed transgenic mdx mice expressing Dp260 in skeletal muscle. Dp260 is the retinal-specific isoform of dystrophin and lacks the N-terminal ABD and the first 9 repeats of the rod region, but retains the internal ABD. Our results indicated that the internal ABD is able to bind actin in the absence of the N-terminal ABD and with similar affinity. While Dp260 did not fully prevent dystrophy in mdx mice, it slowed progression of the disease and prevented development of inflammation and fibrosis. The difference in functionality between constructs deleted for the two ABD may be related to differences in specificity of actin binding and the particular repeats deleted. Importantly, Dp260 was highly effective in protecting muscles from contraction-induced injury, demonstrating a mechanically functional link with the cytoskeleton. In contrast, force development in transgenic muscle was not different from mdx muscles. These results suggest that the dystrophin rod domain may facilitate the formation of a muscle architecture optimized for maximal force development.

Polyglutamine-expanded ataxin-7 induces a cone-rod dystrophy in transgenic mice by antagonizing the function of the nuclear transcription factor CRX. A.R. La Spada¹, Y.H. Fu², B.L. Sopher¹, R.T. Libby¹, X. Wang³, L.Y. Li³, D.D. Einum⁴, J. Huang⁵, D.E. Possin⁵, J.B. Hurley^6,7, L.J. Ptacek^4,8, S. Chen³. 1) Dept Lab Medicine, Univ Washington, Seattle, WA; 2) Dept Neurobiol & Anat, Univ Utah, Salt Lake City, UT; 3) Dept Ophthal & Vis Sciences, Washington Univ, St Louis, MO; 4) Dept Human Genet, Univ Utah, Salt Lake City, UT; 5) Dept Ophthal, Univ Washington, Seattle, WA; 6) Dept Biochem, Univ Washington, Seattle, WA; 7) HHMI, Univ Washington, Seattle, WA; 8) HHMI, Univ Utah, Salt Lake City, UT.
   Spinocerebellar ataxia type 7 (SCA7) is caused by the expansion of a CAG repeat. Although all polyglutamine diseases share a common mutational motif, the basis of cell-type specificity in each disorder remains elusive. To determine the mechanism of ataxin-7 neurotoxicity, we produced transgenic mice that express ataxin-7 with 24 or 92 glutamines. Histological analysis of the 92Q mice revealed periodic thinning of the photoreceptor cell layer of the retina. Whole mounts and immunostaining with pigment-specific antibodies indicated that this periodic thinning is due to preferential loss of cones. ERGs performed on the 92Q mice revealed that they were blind. These results indicate that the process of retinal degeneration in our SCA7 transgenic mice is a cone-rod dystrophy phenotype remarkably akin to what occurs in SCA7 patients. When a yeast two-hybrid assay indicated that the cone-rod homeodomain protein (CRX) interacts with ataxin-7, we performed further studies to assess the significance of this interaction. We found that ataxin-7 and CRX co-localize in nuclear aggregates and can be co-immunoprecipitated. We observed that polyglutamine-expanded ataxin-7 can suppress CRX transactivation. Electrophoretic mobility shift assays performed on retinal nuclear extracts revealed a marked reduction in the ability of CRX to bind its consensus sequence. We carried out real-time RT-PCR analysis on presymptomatic mice and found significant decreases in the expression of genes regulated by CRX. Our results suggest that CRX transcription interference accounts for SCA7 retinal degeneration, and thus may provide an explanation for how cell-type specificity is achieved in this disease.

Myotonia is associated with loss of transmembrane chloride conductance and aberrant splicing of Clcn1, the skeletal muscle chloride channel, in a transgenic model of myotonic dystrophy (DM1). A.K. Mankodi¹, M. Takahashi², C. Beck³, S. Cannon², C.A. Thornton¹. 1) Univ Rochester, Rochester, NY; 2) MGH Hospital, Boston, MA; 3) Jefferson Univ, Philadelphia, PA.
   DM1 is caused by expansion of an untranslated CTG repeat in the DMPK gene. Pathogenic effects of this mutation are likely mediated, at least in part, by the expanded CUG repeat in mutant mRNA. DM1 is characterized by myotonia, a state of hyperexcitability in muscle fibers. In other disorders, myotonia results from defects in the muscle chloride or sodium channels. The pathophysiology of myotonia in DM1, however, is unclear. Lines of transgenic mice (HSA-LR mice) that express expanded CUG repeats develop myopathy and myotonia (Mankodi et al, 2000). We investigated the pathogenesis of myotonia in HSA-LR mice. Intracellular recordings show that hyperexcitability is intrinsic to the muscle fibers, independent of their innervation. Transmembrane chloride conductance is reduced 7.5-fold in HSA-LR mice. Immunofluorescence studies show segmental loss of full-length Clcn1 protein from muscle fibers. DM1 is associated with aberrant splicing of cardiac troponin T (Phillips et al, 1998). We examined the splicing of Clcn1 mRNA in HSA-LR skeletal muscle. RT-PCR experiments show inclusion of a novel Clcn1 exon in HSA-LR mice, similar to the pattern of splicing in muscle from neonatal wild-type mice. Inclusion of this exon results in frame-shift and premature termination at codon 340 of 994. Mis-splicing of this exon correlates with the presence of myotonia in different HSA-LR founder lines. Aberrant splicing of Clcn1, however, is not confined to this single exon. 29 Clcn1 cDNA clones generated from HSA-LR muscle show 11 different splice variants, none of which are observed in wild-type muscle. By contrast, 19 of 20 alpha-sarcoglycan cDNA clones from HSA-LR mice are spliced correctly, an indication that the fidelity of splicing is not universally impaired. We conclude that the nuclear accumulation of expanded CUG repeats triggers aberrant splicing of Clcn1 mRNA. The resulting loss of Clcn1 protein and transmembrane chloride conductance is sufficient to account for the myotonia in HSA-LR mice.

Myotonic Dystrophy Type 2 is Caused by a CCTG Expansion in Intron 1 of ZNF9. L.P.W. Ranum¹, C.L. Liquori¹, M.L. Moseley¹, J.F. Jacobsen¹, A.V. Phillips², R. Savkur², W. Kress³, S.L. Naylor⁴, T.A. Cooper², K. Ricker³, J.W. Day¹. 1) Institute of Human Genetics, Univ Minnesota, Minneapolis, MN; 2) Department of Pathology, Baylor College of Medicine, Houston, TX; 3) Department of Neurology, University of Würzburg, Germany; 4) Department of Cellular and Structural Biology, University of Texas Health Science Center, San Antonio, TX.
   Myotonic dystrophy (DM) can be caused by a mutation on chromosome 19 (DM1) or 3 (DM2/PROMM). DM1 is caused by a CTG expansion in the 3' UTR of the dystrophia myotonica-protein kinase gene (DMPK). Several mechanisms have been suggested to explain how this mutation causes the multisystemic effects of DM including: DMPK haploinsufficiency; reduced expression of regional genes (eg SIX5); and pathogenic effects of the CUG expansion in RNA. Mouse models have suggested that each of these mechanisms contributes to DM1 pathogenesis and that DM1 is a regional gene disorder. To clarify the pathogenic mechanism of DM, we have identified a second human mutation that causes the same multisystemic effects. Linkage disequilibrium analysis (see Liquori, et al.) refined the DM2 region. One of our markers, CL3N58, showed an aberrant segregation pattern by PCR. Southern analysis demonstrated that all affected individuals in six DM2 families (LOD=31.6 at Q=0.00) had an expanded allele that was not found in controls (n=1360). Sequence analysis showed that the CCTG portion of the compound repeat (TG)n(TCTG)n(CCTG)n expands in affected alleles. The range of expanded alleles is broad (~75-11,000 CCTGs, mean~5,000). The expansion is located in intron 1 of the zinc finger protein 9 (ZNF9) gene, whose normal function as an RNA binding protein appears unrelated to any of the genes found in the DM1 region. Similar to the CUG RNA foci in DM1, intense CCUG-containing nuclear foci were found in DM2 but not control muscle. RT-PCR of insulin receptor (IR) mRNA in skeletal muscle showed a marked reduction in the IR-B isoform in DM2 (18%, n=3) and DM1 (21%, n=9) compared to controls (73%, n=5). Parallels between DM1 and DM2 indicate that the CUG and CCUG expansions in RNA can themselves be pathogenic and cause the multisystemic features common to both diseases.

Muscular dystrophy and neuronal migration disorder caused by mutations in a novel glycosyltransferase. K. Kobayashi¹, A. Yoshida², H. Manya³, M. Mamoru⁴, T. Inazu⁴, H. Mitsuhashi², H. Topaloglu⁵, M. Takeuchi², T. Endo³, T. Toda¹. 1) Division of Functional Genomics, Department of Post-Genomics and Diseases, Course of Advanced Medicine, Osaka University Graduate School of Medicine, Osaka, Japan; 2) Central Laboratories for Key Technology, Kirin Brewery Co., Ltd., Yokohama, Japan; 3) Department of Glycobiology, Tokyo Metropolitan Institute of Gerontology, Tokyo, Japan; 4) Research Department, The Noguchi Institute, Tokyo, Japan; 5) Department of Paediatric Neurology, Hacettepe Children's Hospital, Ankara, Turkey.
   Muscle-eye-brain disease (MEB) is an autosomal recessive disorder characterized by congenital muscular dystrophy, ocular abnormalities, and lissencephaly. Mammalian O-Mannosyl glycosylation is a novel protein modification observed in a limited number of glycoproteins of brain, nerve, and skeletal muscle. Here we isolated a human cDNA for O-linked mannose b1,2-N-acetylglucosaminyltransferase (OMGnT), which is a novel enzyme for mammalian type O-mannosyl glycan synthesis. The cDNA encoded a putative type II membrane protein of 660 amino acids having 23.2% sequence identity to human a-3-D-mannoside b-1,2-N-acetylglucosaminyltransferase I (GnT-I). By transfecting this cDNA into HEK293T cells, OMGnT activity increased approximately 100-fold. However, OMGnT did not show any detectable GnT-I activity. Northern blot analysis revealed that OMGnT is constitutively expressed in all human tissues tested. These results clearly demonstrate that OMGnT is a new b1,2-N-acetylglucosaminyltransferase functionally different from GnT-I. Moreover, we identified two independent mutations of this gene in two patients with MEB, indicating that the OMGnT gene is responsible for MEB. These findings would suggest a novel pathomechanism, glycosylation, on muscular dystrophy as well as neuronal migration disorder.

A novel glycosyltransferase is mutated in a form of congenital muscular dystrophy with secondary laminin a2 deficiency and abnormal glycosylation of a-dystroglycan. M. Brockington¹, D.J. Blake², P. Prandini¹, S.C. Brown¹, S. Torelli^1,3, M.A. Benson², C.P. Ponting², B. Estournet⁴, N. Romero⁵, T. Voit⁶, C.A. Sewry⁷, P. Guicheney⁵, F. Muntoni¹. 1) The Dubowitz Neuromuscular Centre, ICSM,Hammersmith Hospital, London, UK; 2) Department of Human Anatomy and Genetics, University of Oxford, UK; 3) Department of Cytomorphology, University of Cagliari, Italy; 4) Hpital Raymond Poincar, Garches, France; 5) Inserm U 523, Pitie-Salpetriere, Paris, France; 6) Department of Paediatrics, University of Essen, Germany; 7) 7Department of Histopathology, Robert Jones & Agnes Hunt Orthopaedic Hospital, Oswestry, UK.
   The congenital muscular dystrophies (CMD) are a heterogeneous group of autosomal recessive disorders, presenting in infancy with muscle weakness, contractures and dystrophic changes on skeletal muscle biopsy. Brain involvement is also present in several CMD syndromes. Approximately 40% of patients have a primary deficiency of laminin a2 due to mutations in the LAMA2 gene. A secondary deficiency of laminin a2 is apparent in some CMD syndromes.
   Here we identify a gene encoding a novel putative glycosyltransferase. We report the genomic organization of this locus and its pattern of tissue expression. Mutations in this gene have been identified in 7 families with a severe form of CMD that does not involve the brain. Patients have markedly elevated serum creatine kinase and a secondary deficiency of laminin a2. We observed a marked decrease of muscle a-dystroglycan immunostaining and a significant drop in its molecular weight on Western blot. These observations are interpreted to be the result of altered glycosylation of a-dystroglycan. This is likely to be integral to the pathology seen in these patients.

Gene expression analysis of muscle from Duchenne muscular dystrophy patients. J.N. Haslett, D. Sanoudou, S.A. Greenberg, H.G.W. Lidov, I.S. Kohane, A.H. Beggs, L.M. Kunkel. Genetics Division, Children's Hospital, Harvard Medical School, Boston, MA.
   Muscular dystrophy (MD) refers to a clinically and genetically heterogenous group of myopathyies defined by progressive degeneration of skeletal muscle fibers, leading to loss of muscle function. Mutations in the dystrophin gene were identified as causative of the most common muscular dystrophy, Duchenne MD. Dystrophin is linked to a group of integral membrane proteins, forming the dystrophin-associated protein complex (DAPC). To examine the DMD pathogenic pathways and identify new or modifying factors involved in muscular dystrophy we used Affymetrix oligonucleotide arrays to analyze the expression patterns of 12,600 genes in 12 DMD patients. The DMD data were compared to data from 10 normal skeletal muscle samples and to data obtained from other myopathic samples. Conventional statistical and fold-ratio comparisons and automated classification and clustering techniques were used to analyze the data, allowing identification of both disease-specific changes and changes due to the non-specific response of diseased muscle. This will enable additional analysis of DMD pathophysioloical mechanisms and those of skeletal muscle more generally. When compared to other datasets, the DMD samples show a distinctive clustering pattern, presumed to reflect differences in the pathogenic pathways. Fold-ratio analysis suggests approximately 80 of the genes examined differ significantly in expression level between normal and DMD muscle. More genes were found to be overexpressed than underexpressed in DMD, potentially reflecting an increase in protein turnover, probably due to the degenerative and regenerative nature of the disease. As expected dystrophin was found to be substantially underexpressed in DMD muscle. In contrast, a number of other muscle genes (myosin, troponinT, alpha- and beta-tubulin, alpha-actinin) were found to be overexpressed in DMD muscle. The observed changes are being subjected to further statistical analysis and classification, in addition to experimental verification, to provide insight into the molecular mechanisms behind the DMD pathogenic pathways.

The alternative splicing of DMPK is affected in cis by the CTG expansion of Myotonic Dystrophy Type-1.. A.J. Gibb, M.G. Hamshere. School of Life & Env. Sciences, University of Nottingham, Nottingham, Nottinghamshire, UK.
   
Myotonic Dystrophy Type-1 (DM1) is the most common muscular dystrophy affecting 1/8000 live births in Western European and North American populations. The condition is characteristically multisystemic with symptoms including myotonia, progressive muscle weakness, cardiac conduction defects, cataracts, frontal balding in males, testicular atrophy, diabetes, hypersomnia and mental retardation. The condition segregates in an autosomal dominant manner and is caused by a CTG triple-repeat expansion found in the 3UTR of the gene for a serine-threonine protein kinase, DMPK.

DMPK has several alternative splice forms. The frequency of their use and the nature of any effect mediated by the repeat, either in cis or in trans, has not yet been fully investigated. Using affected and control fibroblasts heterozygous for a SNP in Exon 10 of DMPK, we have been able to develop a fluorescence based quantitative RT-PCR method in order to discriminate between transcripts derived from the affected and unaffected chromosomes. By a combination of fluorescence and size discrimination, we have established that the profile of alternative splicing is affected in cis by the myotonic dystrophy triplet repeat expansion.

Epidemiological impact of genetic counselling and predictive testing for myotonic dystrophy. J. Mathieu, C. Tremblay, C. Prevost. Dept Neurology, Complexe Hosp de la Sagamie, Chicoutimi, PQ., Quebec, Canada.
   The prevalence of myotonic dystrophy type 1 (DM) is very high in the Saguenay-Lac-Saint-Jean (SLSJ) region, a geographically isolated region in the northeastern part of the province of Quebec (Canada). In SLSJ, a program of genetic counselling for DM patients and their families was introduced in 1981 and a program of predictive testing is offered to the population since 1988. In order to determine the effect of genetic counselling and predictive testing on the prevalence of the DM phenotypes, we compared the demographic characteristics and the phenotypes of the DM patients over the last 15-year interval (1985-2000). During this period, the median age of the DM patients increased from 34.5 years to 44.0 years (p<0.00001). The median age of the congenital, childhood and classic DM patients increased respectively from 9 to 24 years (p<0.001), 24 to 33 years (p=0.01) and 35 to 44 years (p<0.001). In this time interval, the proportion of patients with a congenital or a childhood phenotype remained stable but the proportion of patients with a classic phenotype decreased from 75% to 61% and, the proportion of patients with a mild phenotype increased from 6% to 20% (p<0.001). In the last 5 years, among the 63 newly diagnosed patients, 43% had a mild DM phenotype detected only by DNA analysis. The reasons for the significant ageing of the DM population in SLSJ region are complex and include a low natality rate throughout the Quebec population, a reduction in births at risk due to genetic counselling and an increase in the number of mildly affected patients often recognized at an older age only by predictive testing. The shift in the distribution of the DM phenotypes toward mildy affected individuals is expected to accelerate in the future.

Founder effect in North-Eastern Quebec and the extent of the genetic signature associated with the myotonic dystrophy mutation. V. Yotova¹, E. Zietkiewicz¹, E. Lemieux-Blanchard¹, M. Labuda¹, S. Bourgeois¹, D. Labuda¹, J. Fortin², P. Lepage², T.J. Hudson², A. Lescault³, C. Laberge³. 1) Research Center, Hopital Sainte-Justine, Montreal, Quebec, Canada; 2) MGH Research Center, McGill University, Montreal Genome Center, Montreal, Quebec, Canada; 3) Centre hospitalier, Universit Laval, Ste-Foy, Quebec, Canada.
   Myotonic dystrophy is frequent in French-Canadian population of North-Eastern Quebec (1:650). This dominant disorder is due to a CTG-triplet expansion in the DMPK gene on chromosome 19q13 (OMIM160900). Rather than directly analyzing the expansion, we wanted to define a unique haplotype composed of single-nucleotide polymorphisms, a SNP signature, eventually amenable to automatic screening. By the same token, we investigated the extent of the ancestral carrier-haplotype conservation in a young population issued from a founder effect and addressed the hypothesis of a single introduction of the DM-mutation in North-Eastern Quebec. SNPs were found in silico as well as experimentally in a population panel by DHPLC and subsequent sequencing. Twenty of these SNPs were typed in 51 DM-families by allele-specific oligonucleotide hybridization. The resulting haplotype extends over 2.3 Mb according to the recent UCSC map. There are two groups of DM haplotypes. The first, presumably young, is represented by a haplotype seen in 34 copies and a minor recombinant (2 copies). The second is represented by four variants (5, 5, 3 and 2 copies) that can be mutually related through single recombinations. These two groups could therefore represent two separate entries of the DM-chromosomes. On the other hand, they share a shorter core haplotype, suggesting a relatively recent common origin for both groups; genealogical data will help to trace it back in time. In conclusion, the DM-mutation appears to be associated with a limited number of unique SNP-signatures that are not shared with non-affected chromosomes; shorter segments of these haplotypes are relatively common in the population; our data demonstrate an important increase in the extent of linkage disequilibrium that can be related to the young age of the population of Quebec. (Supported by RMGA FRSQ).

Psychosocial impact of predictive testing for myotonic dystrophy. C. Prevost¹, J. Villeneuve¹, M. Tremblay², S. Veillette³, M. Perron³, J. Mathieu¹. 1) Complexe Hosp de la Sagamie, Chicoutimi, Qc, Canada; 2) Quebec University at Chicoutimi, Chicoutimi, Qc, Canada; 3) Cegep of Jonquiere, Qc, Canada.
   In the Saguenay-Lac-Saint-Jean region (Quebec, Canada), a predictive DNA-testing program for myotonic dystrophy type 1 (DM) has been available as a clinical service since 1988. From 1 to 12 years (median, 5 years) after receiving predictive testing, a total of 308 participants (44 carriers and 264 non carriers) answered a questionnaire to determine reasons for testing and recall of test result, to assess their perception of the psychosocial impact of predictive testing and to measure their actual general well-being, self-esteem and psychological distress. The reasons for wanting to be tested were to learn if children are at risk for DM or for reproductive decision making (75%) and to relieve the uncertainty for themselves (17%). The majority of participants (96.1%) remembered correctly their test result. Among carriers, 19% consider themselves in a worse psychological state, 49% are more worried about their future health, 25% feel less satisfaction about life, 25% present a lower self-esteem and 53% are more concerned about their childrens risk. Among non carriers, 36% find themselves in a better psychological situation, 49% are less worried about their future health, 48% feel more satisfaction about life, 20% present a better self-esteem and 85% are reassured about their childrens risk. The actual general well-being, the self-esteem (Rosenberg Self-Esteem Scale) and the psychological status (Psychiatric Symptom Index) are similar in carriers, in non carriers and in the reference (Quebec) population; these results are not influenced by the number of years elapsed since predictive testing was done. All respondents believe that predictive testing should be available for the at-risk population and the vast majority of carrier (95%) and of noncarriers (93%) would recommend the use of predictive testing to their family members.

Molecular analysis of Duchenne and Becker muscular dystrophy patients in Saudi Arabia. R. Majumdar¹, M. Al Jumah², S. Al Rajeh³, E. Chaves-Carballo⁴, M.M. Salih⁵, A. Awada², S. Shahwan⁶, S. Al Uthaim¹. 1) Neurogenetics Laboratory, Department of Medicine, King Fahad Natl Guard Hosp, Riyadh, Saudi Arabia; 2) Division of Neurology, Department of of Medicine, King Fahad Natl Guard Hosp, Riyadh, Saudi Arabia; 3) Division of Neurology, King Saud University, Riyadh, Saudi Arabia; 4) Department of Neurosciences, King Faisal Specialist Hospital & Research Center, Riyadh, Saudi Arabia; 5) Department of Pediatrics, King Saud University, Riyadh, Saudi Arabia; 6) Department of Pediatrics, Military Hospital, Riyadh, Saudi Arabia.
   Objective:The deletion in the dystrophin gene has been reported for many ethnic groups, but until now the mutations in this gene have not been thoroughly investigated in Saudi Duchenne and Becker muscular dystrophy (DMD/BMD) patients. Methods: We examined the deletion pattern in the dystrophin gene of the Saudi patients applying multiplex-polymerase chain reaction (PCR). Genomic DNA was isolated from twenty two patients with DMD/BMD confirmed by dystrophin staining on muscle biopsy, eleven patients with clinical suspicion of DMD without muscle biopsy, three patients with limb girdle muscular dystrophy, twelve relatives of the patients, and five healthy Saudi volunteers. Specific exons around the deletion prone regions (hot spots) of the dystrophin gene were amplified. Results: The deletion of one or more exons was found in sixteen of twenty two DMD/BMD patients. The deletion in the gene was detected in seven of eleven patients with suspected DMD diagnosis, but not confirmed by dystrophin staining of muscle biopsy. No deletion in the dystrophin gene was detected in control Saudi volunteers, the limb girdle dystrophy patients, and the relatives of patients, as expected. Conclusion: The present study suggests that intragenic dystrophin gene deletions (70%) occur with the same frequency in Saudi patients compared with other ethnic groups.
   (This study was supported by King Fahad National Guard Hospital and a grant from King Abdulaziz City for Science and Technology, Project AT-18-03).

Choosing a novel strategy for point mutation detection in the dystrophin gene : DHPLC or BESS ? A preliminary comparison. S. Chambert, C. Saquet, M. Claustres, S. Tuffery-Giraud. Lab Genetique Moleculaire, CHU montpellier, France.
   In Duchenne and Becker Muscular Dystrophy (D/BMD), one third of the mutations are not detectable as intragenic deletions or duplications of the dystrophin gene. Because full gene sequencing (79 exons) would be laborious and time-consuming in order to identify point mutations occurring in this gene, we used an approach based on the analysis of dystrophin transcripts (RT-PCR) coupled with the protein truncation test (PTT) during these last years. As a result of these studies, the families of these patients are currently benefiting from accurate carrier-status assessment. Also, some patients carrying a point mutation have been selected for inclusion in clinical trials in view of therapy. However, this technique was limited to a detection rate of 86% in our series. In the aim at identifying point mutation in more patients, we set out to assess the sensitivity of Denaturing High-Performance Liquid Chromatography (DHPLC) and Base Excision Sequence Scanning (BESS) as an approach to mutation screening in DMD. We first assessed a cohort of 15 patients with previously identified mutations (frameshift or stop mutations) spread all over the gene. The comparison of the two methods was based on the analysis of the same amplified fragment of cDNA (~ 500bp) known to contain the mutation. A sequenced cDNA was used as control, and for heteroduplex formation in DHPLC. The sensitivities of DHPLC and BESS were very similar when respectively more than one column temperature was used for the detection of mismatches, and when BESS T and BESS G cleavage reactions were performed. Both allowed the rapid detection of single base substitutions as well as small deletions/insertions. From these findings, we conclude that these two methods provide a reliable alternative to the PTT. Up to now, we have not yet established which of these two technology (BESS or DHPLC) would be the most appropriate for the molecular diagnosis of DMD. We plan to analyze DMD patients (clinical diagnosis confirmed by immunohistochemical analysis of dystrophin) in whom extensive RT-PCR/PTT analysis failed to detect the mutation. Supported in part by a grant of Association Franaise contre les Myopathies (AFM).

A comprehensive molecular analysis of myotonic dystrophy in Iran. M.S.M.E. Mousavi, K. Kahrizi, S.S Hosseini Amini, H. Najmabadi. Genetics Research center, Universitiy of Social Welfare and Rehabilitation Sciences , Tehran, Iran.
   Myotonic dystrophy (DM) is the most common form of adult muscle dystrophies with varrying frequecies in different populations. So far there has been no comprehensive population screening on this disease in Iran. DM has multisystemic manifestatioin including muscle weakness and myotonia. It has pattern of Trinucleotide (CTG) repeats, in 3' untranslated region of the serine Thereonine protein kinase gene located on chromosome 19q13.3. Due to similarity in clinical picture with the other dystrophic patients and the lack of reliable diagnostics we decided to set up a molecular analysis technique for DM for the first time in Iran. PCR and southern bloting were perfomed using nonradioactive material. 100 patients were investigated. DM gene mutations were detected in 35% of the clincally diagnosed DM patients.

Six novel point mutations in the Dystrophin gene identified in Brazilian Duchenne patients. D.N.F. Vagenas^1,2, A. Cerqueira¹, A. Torres¹, R. Pavanello¹, M.R. Passos-Bueno¹, M. Zatz¹. 1) Centro de Estudo Genoma Humano, Universidade de So Paulo, So Paulo, S.P, Brazil; 2) Centro em Interunidades em Biotecnologia, Universidade de So Paulo, S.P. Brazil.
   Duchenne (DMD) and Becker (BMD) type muscular dystrophies are allelic X-linked conditions caused by mutations in the gene encoding dystrophin. This gene contains 79 exons, which are distributed in approximately 2,400 Kb. The phenotype is caused in about 60% of the cases by deletions, 5 - 6% by duplications, while the remaining cases are due to point mutations or small deletions or rearrangements. We have ascertained so far 1010 families with DMD patients in our center. Among them 632 had deletions in the dystrophin gene. We are currently trying to identify the mutations in the non-deleted cases through SSCP, PTT (Protein Truncation Test) and sequencing of abnormal fragments. For PTT analysis, the dystrophin gene was divided in 10 fragments, each one with 8 exons, encompassing the 79 exons. In 99 patients, about 4 fragments were randomly analyzed leading to the identification of 10 mutations. Among these, six are novel mutations: one deletion of exon 38, three frameshifting involving the exons: 5 (494-497 delAGTA); 16 (2067 - 2068 del TG and 2080 - 2083 del ACAA); and two nonsense in the exons 17 (2345 C®T; Q 713 X) and 45 (6785 T®C; Q 2193 X). We are also assessing the intellectual capacity in our patients in order to verify what mutations/deletions are more commonly associated with mental impairment. CEPID-FAPESP, PRONEX, CNPq, IAEA.

Resolution of equivocal molecular genetic test results in the diagnosis of Facioscapulohumeral Muscular Dystrophy (FSHD). W.K. Seltzer¹, K. Felice², J. Durocher¹, A.R. Yesley¹, M.A. Boss¹, S.E. Hallam¹. 1) Athena Diagnostics, Inc, Worcester, MA; 2) University of Connecticut Health Center, Farmington, CT.
   Ninety to 95% of Facioscapulohumeral muscular dystrophy (FSHD) cases are associated with a deletion within a repeat array. Athena Diagnostics, Inc. performs FSHD analysis by EcoRI and EcoRI/Bln1 restriction endonuclease digestions followed by Pulsed-Field Gel Electrophoresis to separate and resolve the 8 to > 48 kb sized DNA fragments. Southern analysis is then performed using probe, p13E-11, which hybridizes to repeat arrays at both 4q35 and 10q26. Bln1 specifically digests the chromosome 10 repeat array to small fragments allowing discrimination between chromosome 10 and chromosome 4 arrays. Twenty percent of the general population carry a benign translocation resulting in either three chromosome 4q35 repeat arrays and a single 10q26 array, or alternatively, three 10q26 arrays and a single 4q35 array. This complicates result interpretation since it is the location of the repeat array deletion, and not its chromosomal origin, which determines the pathogenicity of the deletion. The presence of these translocations can potentially lead to false positive and false negative test results. We reviewed 275 cases of which 49 (18%) yielded findings that could not be unequivocally interpreted. In order to achieve a better understanding of the remaining equivocal results, we carried out XapI analyses which specifically digest the chromosome 4 repeat array. This provided significant utility in interpretation of these previously equivocal results.

New tools to study proteins involved in disease: FSHD as paradigm. S. van Koningsbruggen¹, H. de Haard², R.W. Dirks¹, J.T. den Dunnen¹, G.W. Padberg³, G.J. van Ommen¹, C.T. Verrips², R.R. Frants¹, S.M. van der Maarel². 1) Center for Human & Clinical Genetics and Dept. of Molecular Cell Biology, LUMC Leiden, Leiden, Netherlands; 2) Unilever Research, Vlaardigen, Netherlands; 3) Dept. of Neurology, UMCN Nijmegen, Nijmegen, The Netherlands.
   Facioscapulohumeral muscular dystrophy (FSHD) is caused by a complex genetic rearrangement. Likely, partial deletion of the subtelomeric D4Z4 repeat on 4qter causes the transcriptional deregulation of nearby genes by a position effect variegation-like mechanism. Consequently, there are no obvious structural mutations within the FSHD candidate genes, complicating their positive identification.
   Next to expression profiling of candidate genes, we have embarked on protein profiling employing llama-derived phage-display single (heavy) chain antibody fragments. Camelidae carry apart from a conventional antibody repertoire, a unique repertoire of heavy-chain antibodies that only consist of two heavy chains and are devoid of light chains. Using this antibody repertoire as source for phage-display libraries circumvents the in vitro combination of heavy and light chains, one of the major drawbacks of conventional phage-display libraries. Moreover, these antibodies have unique properties regarding affinity and stability. Current strategies are directed in optimizing high-throughput selection and screening protocols and design of phage-display vectors allowing uniform arraying, visualization, and intracellular expression. Due to their small size (14kDa) and high affinities, autofluorescent heavy-chain antibody fragments will be eminently suited for a variety of applications.
    As proof of principle, we successfully generated immune-libraries raised against cocktails of proteins and whole human skeletal muscle homogenate. We selected high-affinity antibody fragments against FRG1P, the gene product of one of the candidate genes for FSHD. These antibody fragments perform well in a series of immunological techniques: for example, co-localization was observed for EGFP-tagged FRG1P and selected heavy-chain antibody fragments.

Evidence of further genetic heterogeneity for both autosomal dominant and autosomal recessive limb-girdle muscular dystrophy. A.L. Starling, M. Vainzof, R. Pavanello, M. Canovas, A. Cerqueira, M.R. Passos-Bueno, M. Zatz. Center for the study of the human genome, IBUSP, University of Sao Paulo, Sao Paulo, Brazil.
   Fifteen genes responsible for limb-girdle muscular dystrophy have been identified to date including six autosomal dominant (LGMD1A-F) and nine autosomal recessive (LGMD2A-I). We have ascertained three Brazilian families that were excluded by linkage analysis (and protein studies when possible) for these 15 LGMD causing genes. The first family with nine affected members (7 males and 2 females) in 3 generations displays AD inheritance. Two of the affected patients were submitted to muscle biopsy and electromyography which showed a myopathic pattern in both exams. In all affected members the onset occurred after age 30, with proximal weakness in the upper and lower limbs and a variable presence of calf hypertrophy. Serum creatine kinase (CK) was increased 2 - 9 fold. The second family also with AD inheritance has four affected members (3 males and 1 female) in 2 generations. All of them have a history of cardiac problems, weakness of proximal limbs and a CK increase 3 - 6 fold. The age at onset ranged from 25 to 30 years old. The last family, with four affected males has an atypical inheritance. The affected patients have weakness of proximal limbs (predominantly in the lower limbs in three patients but in the upper limbs in the last one), calf hypertrophy and the age at onset ranged from 10 to 15 years old. The CK was grossly elevated (50-80 fold). Electromyography showed a myopathic pattern and the biopsy showed a normal pattern for dystrophin, calpain, dysferlin, the four sarcoglicans and telethonin. We are currently performing a genome-wide scan to identify the disease loci in these three families. Supported by FAPESP-CEPID, PRONEX and CNPq.

Mutations in the dystrophin gene are associated with sporadic dilated cardiomyopathy. J. Yan¹, J. Feng¹, C.H. Buzin¹, J.A. Towbin*², S.S. Sommer*¹. 1) Department of Molecular Genetics, City of Hope Medical Center, Duarte, CA; 2) Department of Pediatrics (Cardiology), Baylor College of Medicine, Houston, TX. *These authors contributed equally to the work.
   Dilated cardiomyopathy is the major indication for heart transplantation. Approximately 30% of all DCM is thought to be inherited, while 70% is sporadic. Mutations in the dystrophin gene have been associated with the uncommon X-linked form of dilated cardiomyopathy. We hypothesized that missense mutations and other less severe mutations might predispose to sporadic dilated cardiomyopathy. To test this hypothesis, we examined 22 patients with sporadic dilated cardiomyopathy by DOVAM-S (Detection of Virtually All Mutations-SSCP)( Liu et al, 1999. BioTechniques 26:932-942; Buzin et al, 2000. BioTechniques 28:746-753), a form of SSCP in which there is sufficient redundancy to detect virtually all mutations. Twenty-two kb of genomic dystrophin DNA was scanned in the 22 patients with sporadic DCM, including all 79 coding sequences and splice junctions, as well as six alternative exon 1 dystrophin isoforms. Three putative new mutations (IVS5+1 G>T, N2299T and F3228L) and six polymorphisms were identified. The splice site mutation IVS5+1 is predicted to cause skipping of exon 5, a region containing an actin binding site. In the missense mutation F3228L in exon 67, the amino acid phenylalanine is highly conserved in dog, mouse, chicken, Torpedo, dogfish, starfish, scallop, and amphioxus dystrophins. Screening of 141 control individuals failed to identify these 3 mutations, hinting that milder mutations in the dystrophin gene are associated with sporadic DCM, but clinical manifestation may require some environmental effector or a modifying gene. These mutations may have reduced penetrance and generally appear as sporadic DCM. Future studies are needed to confirm that mutations in the dystrophin gene are a frequent cause of DCM and to analyze families with these dystrophin mutations to search for environmental or genetic modifiers that may help to generate the disease phenotype

Muscleblind localizes to nuclear foci of aberrant RNA in myotonic dystrophy (DM) types 1 and 2. C.A. Thornton¹, A. Mankodi¹, C. Urbinati², R. Moxley¹, V. Sansone¹, M. Swanson². 1) Univ Rochester, Rochester, NY; 2) Univ Florida, Gainesville, FL.
   There is locus heterogeneity in DM. DM1 is caused by expansion of a CTG repeat in the DMPK gene on chr. 19. DM2 is linked to chr. 3. Pathogenic effects in DM1 are likely mediated, at least in part, by the expanded CUG repeat in mutant mRNA. The mutant transcripts are retained in the nucleus in multiple foci. Various proteins interact with CUG repeats in vitro, but none have been shown to interact in vivo. We investigated the possibility that DM2 is caused by expression of a CUG repeat or related sequence. RNase protection with a CAG-29 probe showed an expanded CUG repeat in DM1 but not in DM2 or normal controls (n=4 in each group, myoblast RNA). However, FISH using CAG probes on sections of muscle tissue showed nuclear foci in DM2 similar to those in DM1. Nuclear foci were present in muscle tissue from all patients with symptomatic DM1 (n=9) or DM2 (n=9) but not in disease or healthy controls (n=23). Foci were not seen with CUG- or GUC-repeat probes. Foci in DM2 were distinguished from DM1 by lower stability of the probe-target duplex, suggesting that a sequence related to CUG repeats accumulates in the DM2 nucleus. Furthermore, muscleblind, homologue of a protein required for muscle development in Drosophila, localized to the nuclear foci in both DM1 and DM2. In contrast to previous studies showing hundreds of RNA foci per nucleus in DM1 myoblasts, we found 1 to 3 foci per nucleus in muscle tissue from DM1 and DM2 patients. The foci did not associate with a specific nuclear structure, as determined by markers for nucleoli, coiled bodies, PML bodies, or perinucleolar complex. To identify proteins that interact with expanded CUG repeats in vivo, we examined muscle from transgenic mice that express expanded CUG repeats. Among 9 proteins examined, including 7 dsRNA binding proteins and CUGBP1, only muscleblind colocalized with foci of expanded CUG repeats. These results support the idea that nuclear accumulation of mutant mRNA is pathogenic in DM1, suggest that a similar disease process occurs in DM2, and point to a role for muscleblind in the pathogenesis of both disorders.

Variable phenotype in monozygotic twins with Duchenne Muscular Dystrophy. K. Radha Mani, G.R. Chandak. Centre for Cellular and Molecular Biology, Hyderabad, Andhra Pradesh, India.
   Duchenne and Becker Muscular Dystrophy are X-linked recessive allelic neuromuscular disorders where affected individuals typically have pseudohypertrophy of calf muscles with development of secondary atrophy and contraction and cardiac or respiratory muscle involvement in the later stages. The disease is caused by mutation in the DMD gene and approximately 65% of patients have intragenic deletions, the rest have point mutations or duplications. We report an interesting case of monozygotic twins with variable phenotype on a similar genetic background. The patients presented to us at the age of 12 years, with one of them almost immobilized with an age of onset at 4 years (Twin I) whereas the other fairly mobile with an age of onset only at the age of 8 years (Twin II). Although both had similar pseudohypertrophy of the calf muscles but the twin I had comparatively flabby muscles consistent with degenerating muscles. The CPK levels were 28,000 and 13,000 respectively for twin I and twin II. Cytogenetic study for both was consistent with a male karyotype with a normal banding pattern. DNA fingerprinting using a 10-loci microsatellite showed similar profile and matched with the parents thus confirming paternity. Deletion analysis of the DMD gene showed exon 49-50 deletion out of 25 exons analysed. Analysis of entire DMD gene using cDNA probes failed to show any other missing or abnormal restriction fragment. Both twins showed deletion of 1.6 Kb and 3.7 Kb fragments on a HindIII blot hybridized with cDNA 8 of the DMD gene. Neither the mother nor the relatives showed this deletion suggesting the deletion to be a new mutation in this pair of monozygotic twin. Two possible explanations can be advanced to account for this variability on a seemingly similar genetic background. This may be an evidence for somatic mosaicism or it may suggest the presence of other factors modulating the severity of the DMD phenotype.

FSHD Myoblasts Possess Reduced Resistance to Oxidative Stress. K.A. Barrett^1,2, R. Tawil², R.C. Griggs², D.A. Figlewicz^1,2. 1) Dept Neurobiology & Anatomy, Univ Rochester, Rochester, NY; 2) Dept Neurology, Univ Rochester, Rochester, NY.
   Facioscapulohumeral muscular dystrophy (FSHD), the third most common muscular dystrophy, is inherited in an autosomal dominant manner. A variable deletion in a repeat region (D4Z4) of chromosome 4q35 has been associated with the disorder, however, the pathogenesis of FSHD has yet to be established. Myoblasts from FSHD patients possess a necrotic appearing morphology, with a swollen cytoplasm and perinuclear vacuoles. Highly confluent cells appear to lack the organization of normal myoblasts, and FSHD cells fuse to form disorganized, swollen myotubes. Although necrotic appearing cells can be found in cultures of normal myoblasts, a larger percentage of FSHD cells possesses the phenotype (2.6 ±1.5% for normals, and 15.7± 6.1% for FSHD, p=0.046, n=4.), controlling for cell density and replicative age. Preliminary observations indicate that this phenotype can be elicited in normal cells exposed to the superoxide anion generator, paraquat. Additionally, previous studies have demonstrated an increased susceptibility of FSHD myoblasts to paraquat relative to normal and disease control (other myopathies) cells. The cyclin dependent kinase (cdk) inhibitor, p21, appears to be upregulated in FSHD cells under normal growth conditions (22.2% of FSHD myoblast nuclei stain strongly positive for p21 compared to 14.2% of normal myoblast nuclei, p=0.004, n=4.); a similar phenomenon in fibroblasts was associated with oxidative stress. Rescue from oxidative stress was attempted by exposing normal and FSHD myoblasts to the membrane permeable anti-oxidant, glutathione ethyl ester (GSH-OEt). Low concentrations of GSH-OEt (0.89mM) achieved full rescue of normal cells exposed to 20mM paraquat. In contrast, GSH-OEt was unable to rescue FSHD cells exposed to 10mM paraquat; the viability of FSHD myoblasts in test wells (containing paraquat plus GSH-OEt) was equivalent to that in wells containing paraquat alone. The current study demonstrates an enhanced irreversible vulnerability of FSHD myoblasts to oxidative stress, suggesting a biochemical marker for FSHD early in myocyte development

Development of a cDNA microarray to study musclar distrophy. T. Tsukahara¹, H. Nagasawa², K. Arahata^{1, 3, 4}. 1) Dept Neuromuscular Research, National Institute of Neuroscience, NCNP, Tokyo, Japan; 2) Ebara Co., Kanagawa, Japan; 3) CREST, JST, Tokyo, Japan; 4) Dr. Kiichi Arahata passed away on December 20, 2000.
   Defects of plasma membrane or extracellular matrix associated proteins, dystrophin or laminin alpha 2 and so on, due to fragile sarcolemma and muscular dystrophy have been identified. However, there are some muscular dystrophy genes whose products are not associated with the plasma membrane. In particular, deficiencies in two nuclear membrane associated proteins, emerin and lamin A/C, result from Emery-Dreifuss muscular dystrophy, EDMD. Moreover, genes for some cytosolic enzymes, calpain 3 and myotonin protein kinase are also responsible genes for muscular dystrophy. However, the relationship between abnormalities in these genes and muscular dystrophy is unclear. Recently, the characterization of pathological features by a comprehensive examination of gene expression in the patients tissue has become possible. To clarify the gene expression profile and to help therapeutic studies, we developed a human muscle cDNA microarray.
   To develop a low-background microarray, we constructed a highly nonredundant human singleton database for virtual cDNAs expressed in skeletal or cardiac muscle. The database was constructed from public sequence data, and then homologous sequences with rRNA and with mtDNA, and with repetitive sequences were excluded to increase the reliability of each probe. PCR primers were designed as amplification lengths of 450550b in the vicinity of 3 ends of each singleton. Each fragment of genes in the database was amplified with specific primers and muscle cDNA pools, cloned and confirmed by sequencing. 1536 clones were then amplified, purified and spotted on a CMT-GAPS coated slideglass to make microarrays.
   To confirmed sensitivity and reproducibility of our microarray, RNAs were labeled and then analyzed by using tyramide signal amplification system and the ScanArray 5000. One microgram of total RNA was enough to analyze, and our microarray showed low background and good resolution. This microarray was considered to be a suitable device for gene expression analysis of the muscular disease with biopsied samples.

Lamin C mutation associated with variable expression of dilated cardiomyopathy and conduction-system disease (D-CM) and limb girdle muscular dystrophy type 1B (LGMD1B) in four generations of a large kindred. S. Dyack^1,2, K. Collins¹, P. Welch², C. Riddell^3,4. 1) Division of Medical Genetics, IWK Health Centre, Halifax, NS, Canada; 2) Department of Pediatrics, IWK Health Centre, Halifax, NS, Canada; 3) Department of Laboratory Medicine, IWK Health Centre, Halifax, NS, Canada; 4) Department of Pathology, Dalhousie University, Halifax, NS, Canada.
   LMNA mutations have been described in four distinct medical conditions including dilated cardiomyopathy with conduction-system defects (D-CM), autosomal dominant Emery-Dreifuss muscular dystrophy (EDMD-AD), limb girdle muscular dystrophy type 1B (LGMD1B), and familial partial lipodystrophy (FPLD). Lamin A and C are nuclear proteins that are encoded by the LMNA gene and differ from each other only through alternative splicing of exon 10. We report here a unique family with a novel mutation, R541S, in exon 10 of the lamin C coding region of LMNA. There is variable expression of D-CM and LGMD1B in affected family members. Three affected individuals had severe, progressive D-CM resulting in sudden cardiac death. A 13 year old male with severe D-CM is now the youngest reported child with a LMNA mutation to develop D-CM and to require a cardiac transplant. Two members presented with a myopathy, subsequently discovered to be LGMD1B, and had less severe D-CM. One affected individual with mild LGMD1B had delayed ambulation as a child. Several asymptomatic individuals have cardiac conduction defects. Many affecteds have muscle pain and most have had an elevated CPK. Also noted is a distinct physical phenotype in females with a decrease in adipose tissue of the extremities, but not associated with the biochemical features of insulin resistance as seen in FPLD. This family illustrates that a single LMNA mutation may lead to the expression of several distinct phenotypes and, in light of these findings, it is suggested that individuals with LMNA mutations should be investigated for features of all of the above named conditions. In particular, it should be recognized that alterations in the lamin C sequence can lead to the development of severe D-CM and children with these mutations may require a heart transplant at a young age.

Functional analysis of lamin A/C mutations leading to dilated cardiomyopathy, Emery-Dreifuss muscular dystrophy and partial lipodystrophy. G.L. Brodsky^1,2, A.C. Moss², L. Mestroni^1,2, M.R.G. Taylor^1,2. 1) Dept Cardiology/Medicine, Univ Colorado Health Sci Ctr, Denver, CO; 2) University of Colorado Cardiovascular Institute, Aurora, CO.
   Lamin A/C gene mutations have been identified in four different human diseases including Emery Dreifuss muscular dystrophy (EDMD), limb-girdle muscular dystrophy (LGMD), dilated cardiomyopathy (DCM) and partial lipodystrophy (PLD). The lamin A/C gene encodes two proteins which are the primary components of the nuclear lamina; a proteinascious meshwork lining the nucleoplasmic face of the inner nuclear membrane. Lamins have also been shown to form intranuclear filaments. The nuclear lamina has been implicated in nuclear structural integrity, chromatin organization, cell cycle regulation and transcriptional unit organization.
    As a first step in elucidating the molecular pathway by which lamin A/C mutations lead to human disease, we have examined the expression and localization of disease causing mutant lamin A proteins in mammalian cell culture. Single nucleotide substitutions responsible for EDMD, DCM and PLD were introduced into a wild type lamin A cDNA and transfected into mouse F9 cells. The expression and sub-cellular localization of the mutant lamin A proteins were determined by indirect immunofluorescence deconvoluting confocal microscopy.
    All of the mutant lamin A proteins examined were found to be expressed and localized to the cell nucleus. The mutant proteins responsible for EDMD and DCM produced a wild type pattern of distribution, localizing to the nuclear lamina as well as intranuclear filaments. In contrast, the mutant lamin A proteins responsible for PLD were found to be concentrated in disk-like or punctate structures at the nuclear periphery. No diffuse lamina staining or intranuclear filaments were observed.
    This is the first demonstration of altered lamin A function resulting from disease causing mutations. Furthermore, these results support a model where expression of dominant negative lamin proteins is responsible for the observed disease phenotypes.

Histone Hypoacetylation of Subtelomeric DNA Repeats (D4Z4) Whose Deletion Causes Fascioscapulohumeral Muscular Dystrophy. G. Jiang, F. Tsien, M. Ehrlich. Human Genetics, Tulane Cancer Center, Tulane Medical School, New Orleans, LA.
   Fascioscapulohumeral muscular dystrophy (FSHD) is an unusual autosomal dominant syndrome caused by the loss of some copies of a complex repeat (D4Z4) in the subtelomeric region of one chromosome 4 homologue. The number of copies of this 3.3-kb repeat at 4q35 arm is polymorphic. Unaffected individuals have 11 to about 95 copies on each 4q35 while > 90% of FSHD patients have <10 copies at one 4q35. Many investigators have proposed that normally this region is heterochromatic but that when the number of tandem copies of D4Z4 is <10, the region loses its condensed chromatin structure. This is hypothesized to induce inappropriate gene expression in the affected muscle cells. However, there have been no reports of the chromatin structure in this region. Because constitutive heterochromatin has hypoacetylated core histones, which lead to chromatin compaction, we are testing the acetylation of D4Z4 repeat chromatin using chromatin immunoprecipitation (ChIP) with an antibody for tetra-acetylated histone H4 followed by PCR with various amounts of immunoprecipitated sample for DNA amplification. We compared H4 acetylation of chromatin containing the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene, D4Z4 repeats, or centromeric satellite a repeats by ChIP assays on a normal lymphoblastoid cell line. GAPDH served as a euchromatic standard and satellite a repeats as the heterochromatic standard. H4 acetylation was quantitated from the immunoprecipitated chromatin normalizing for the PCR efficiency. The relative amounts of H4 acetylation were 1.0, 0.01, and <0.01, for GAPDH, D4Z4, and satellite a, respectively, which is consistent with heterochromatinization of D4Z4. We will analyze chromatin containing D4Z4, satellite a, or various euchromatic genes for mono- and tetra-acetylation of histone H4 to determine if D4Z4 chromatin is normally hypoacetylated compared to bulk euchromatin, as is satellite a heterochromatin. We will also compare cells from FSHD patients and analogous cells from unaffected individuals to determine if the D4Z4 repeats from the deletion-containing 4q35 are abnormally hyperacetylated in FSHD cells. Supported by FSH Society Grant FSHS-MB-06.

Myotonic dystrophy in the North of Portugal: a significant but still overlooked problem. T. Coelho¹, A. Tuna², M. Freijo², M.C. Palmares³, R. Santos⁴, M. Santos⁵, A. Sousa⁶. 1) Neurophysiology, HGSA; 2) Neurology, HGSA; 3) Centro Gentica Clinica; 4) Inst. Genetica Medica Jacinto Magalhaes; 5) Neuropediatria, Hospital Maria Pia; 6) Inst. Ciencias Biomedicas Abel Salazar, Porto, Portugal.
   Myotonic dystrophy, one of the most frequent hereditary muscular disorders has never been systematically studied in Portugal. Complete molecular diagnosis became available only recently. Since 1991, 74 cases of myotonic dystrophy from --35 different families were diagnosed at one paediatric (Hospital Maria Pia) and one adult (Hospital Santo Antnio) integrated neuromuscular outpatient clinics in Porto. Recently 32 families (69 patients, 37 female and 32 male) had molecular confirmation of the CTG expansion in chromosome 19. The size of this expansion was quantified in 36 patients. Clinical type, age-of-onset, family history and molecular studies were reappraised in these 69 patients. Age-at-onset was established in 57: congenital form presented in 10 patients (8 probands) and varied from 1 to 63 years in the non-congenital forms. Concerning parental transmission, 28 patients inherited the disease from their fathers and 29 from their mothers; 12 cases had unidentified transmitter. Mean age-of-onset was higher (23.3 years) in offspring of affected fathers than in offspring of affected mothers (13.8 years). Myotonic dystrophy is a relevant problem in the North of Portugal. Systematic molecular diagnosis of all patients will help in ascertainement of families and genetic counselling.

Mutation screening of caveolin-3 and myotilin in autosomal dominant Limb Girdle Muscular Dystrophy and other dominant myopathies. R.D. Dancel¹, J.M. Stajich¹, J.M. Vance¹, M.A. Pricak-Vance¹, P.C. Gaskell¹, J.M. Gilchrist², R.W. Tim³, M.C. Speer¹, M.A. Hauser¹. 1) Section of Medical Genetics, Department of Medicine, Duke University Medical Center, Durham, NC; 2) Rhode Island Hospital, Providence, RI; 3) Raleigh Neurology Associates, Raleigh, NC.
   Limb-girdle muscular dystrophy (LGMD) is a genetically heterogeneous condition with both autosomal dominant and autosomal recessive forms and extensive heterogeneity demonstrated within each of the two forms. In the autosomal dominant form of LGMD, loci have been identified on chromosomes 5 (myotilin), 3 (caveolin-3), 1, and 6. To date, only one family has been identified as having a mutation in myotilin. Thus, we undertook a screen of myotilin and caveolin-3 in 32 autosomal dominant LGMD families. We also investigated patients affected with other autosomal dominant myopathies including individuals affected with non-chromosome 4 linked facioscapulohumeral muscular dystrophy, scapuloperoneal muscular dystrophy, and other non-specific dominant myopathies. Each of the 9 coding exons in the myotilin gene and 2 coding exons in the caveolin-3 gene was PCR amplified from genomic DNA in pools of 5 unrelated individuals. The PCR products were analyzed by denaturing high performance liquid chromatography using the Transgenomic WAVE DHPLC system. Several different temperatures were tested for each PCR amplicon to optimize mutation detection sensitivity. Any pools displaying altered retention time were separated and the constituent samples were reanalyzed individually. All observed changes were confirmed by sequencing both strands using the Beckman CEQ2000 capillary electrophoresis sequencer. A single polymorphism, present in both affected and control individuals, was detected in the caveolin-3 gene. To date, no caveolin-3 or myotilin mutations have been detected, excluding mutations in these genes as a common underlying cause of disease inthese families.

LAMA2 mutation thought to be molecular basis of congenital muscular dystrophy in the merosin-deficient dy2j mouse is found to be splicing variant present in normal mice. D.M. Pillers, J. Pang. Dept Pediatrics and Molecular & Medical Genetics, Oregon Health Sci Univ, Portland, OR.
   Laminin is a trimeric extracellular glycoprotein that is a component of the extracellular matrix. It plays an important role in cell adhesion and migration, in addition to signaling through transmembrane molecules such as integrin and dystroglycan. Laminin-2, also known as merosin, is critical in skeletal muscle development and function. Mutations in laminin-a2 are associated with human congenital muscular dystrophy (CMD) and with muscular dystrophy in the mouse models dy and dy2j. A G to A splice donor mutation at exon 2 in dy2j was defined by Xu et al. (1994) as the site of the mutation resulting in the muscular dystrophy phenotype based upon the creation of a group of aberrant laminin-a2 transcripts. In one, exon 2 is skipped but the reading frame is retained. In others, a stop codon is predicted to result in a truncated protein. We performed RT-PCR analysis of the alternatively spliced products for laminin-a2 and found that the spliced-in fragment described by Xu et al. is present in both normal C57Bl/6J mice and the muscular dystrophy mouse model dy2j suggesting that the mutation may actually be a polymorphism. Furthermore, we found that the truncated products predicted by Xu et al. could be translated into near full-length proteins, lacking only domain VI of the laminin-a2 subunit which contains the binding sites for heparin and cell surface receptors. We also found that the skipping pattern of exon 2 resulted in skipping the entire exon 2 while retaining the reading frame, as opposed to skipping part of exon 2 and the entire exon 3 as suggested by Xu et al. Thus, we propose that the splicing-in variant is expressed in normal mouse and likely has a specific function. In addition, we propose that the alternative splice pattern occurs independent of the G to A polymorphism and is the result of a novel regulatory mechanism.

Evaluation of real-time quantitative PCR for carrier detection of dystrophin gene rearrangements. S. Tuffery-Giraud, C. Saquet, S. Chambert, M. Claustres. Lab Genetique Moleculaire, CHU Montpellier, France.
   Large deletions and duplications in the dystrophin gene are detected in about two third of patients with Duchenne (DMD) or Becker (BMD) muscular dystrophy. This enables accurate carrier detection and prenatal diagnosis in those families. Diagnosis of heterozygous deletions is mainly based on haplotype analysis, involving detection of loss of heterozygosity for microsatellite alleles or other polymorphic markers, whereas Southern blot is the method currently used to assess gene dosage for duplications. To meet the need for an easy-to-perform, rapid direct diagnosis of BMD/DMD carrier status, we report the development of a method based on measurement of gene copy number. Gene copy number analysis by real-time quantitative PCR can be targeted directly to a selected sub region of the gene of interest, and does not depend on the availability of informative flanking markers or parental DNA. Also, it involves no radioisotopes and requires no post-PCR handling. Heterozygosity for duplication increases the gene dose from two to three, whereas heterozygosity for a deletion reduces the gene dose from two to one. The amplification was performed in the Light-CyclerTM PCR system using the LightCycler FastStart DNA master SYBR Green I kit. In order to demonstrate the accuracy and reliability of the method for genetic testing, we analyzed 15 patients and relatives females with previously identified deletions or duplications including either exon 11 or 44. Fifteen normal samples were also tested as controls. All samples were run in triplicate. In order to correct for differences in the amount of input genomic DNA, the method involves amplification of a reference locus with known copy number and the use of a common standard DNA for amplification of the test and reference loci. The presence or absence of the genomic deletion or duplication was unambiguously diagnosed in all individuals. Hence, our initial data demonstrated the effectiveness of this approach, which could be extended to the analysis of multiple dystrophin exons in a cost- effective manner. Support: the Association Franaise contre les Myopathies (AFM).

DNA rearrangements at the D4Z4 locus in South African facioscapulohumeral muscular dystrophy families. A. van der Merwe¹, C-M. Schutte¹, S.M. van der Maarel², M. Alessandrini¹, E. Honey¹, R.R. Frants², A. Olckers¹. 1) University of Pretoria, South Africa; 2) Leiden University Medical Centre, The Netherlands.
   Facioscapulohumeral muscular dystrophy (FSHD) is an autosomal dominant, late-onset degenerative neuromuscular disorder. Disease onset occurs generally within the first or second decade of life. The phenotypical expression varies in the age of onset and clinical severity. FSHD is characterised at onset by the progressive weakness and atrophy of the facial and shoulder girdle muscles. Linkage was established to the sub-telomeric region of chromosome 4q35. This sub-telomeric region contains a polymorphic repeat array locus (D4Z4) consisting of 3.3 kb repeats. An integral number of repeats are deleted in affected individuals. Integrated haplotypes were constructed for 177 individuals from five extended South African FSHD families. Southern Blot analysis was performed to determine DNA rearrangements in 162 individuals from this selected group of families. Deletion fragments were observed in all families investigated. A BlnI resistant deletion fragment of 24 kb segregated with the postulated FSHD-haplotype in families F10, F30 and F40. The FSHD-haplotype in all three families was identical, with the entire nine-allele haplotype segregating in one sub-family of F10 (F15) and in family F30. Six alleles of this haplotype co-segregated with the FSHD phenotype in F13 (sub-family of F10) and three alleles co-segregated in F40. It was previously reported that an identical six allele FSHD-haplotype segregated in families F20 and F60. The deletion fragments segregating in these families were also of identical size. To date, only two FSHD haplotypes have been observed in the South African population - each co-segregating with a deletion fragment of a specific size.

Temporal expression profiling in a progressive disease, Duchenne muscular dystrophy. Y.-W. Chen¹, M. Bakay¹, R. Borup¹, S. Toppo², G. Valle², G. Lanfranchi², E.P. Hoffman¹. 1) Center for Genetic Medicine, Children's Natioanl Medical Center, Washington, DC; 2) CRIBI Biotechnology Center, University of Padua, Padua, Italy.
   Duchenne muscular dystrophy (DMD) is a progressive muscular disorder caused by mutations in the dystrophin gene. Although the primary biochemical defect is known, the downstream biochemical changes leading to later patient weakness and death are not understood. To identify differentially expressed genes involved in the progression of the disease, we report expression profiling of pre-symptomatic, early-stage and mid- to end-stage DMD muscles by using custom made Affymetrix MuscleChip containing approximately 1000 full-length genes and 2000 muscle ESTs. Among the total 4,654 probe sets on the MuscleChip, expression of 3688 (80%) genes was detected by the MucleChip at least once out of 46 chips in this study. We found a large number of significant gene expression changes in fetal muscles (193 up- and 8 down-regulated genes) years before the onset of symptoms. By applying both correlation and K-means analysis, 48 genes were identified up-regulated during the progression of the DMD while down-regulated in the age-matched control individuals. One hundred and fifty seven genes were down-regulated during disease progression while up-regulated in control individuals. Among the down-regulated genes, 55% of the known genes were mitochondrial and metabolic genes. Genes involved in regulation of oxidative stress, protein metabolism were also identified. The major up-regulated genes are genes involved in fibrosis and extracellular matrix remodeling, muscle fiber structure, protein synthesis and degradation. Interestingly, expression levels of many genes that we previously reported misregulated by comparing pooled 5-9-year-old DMD vs. control (ex. a-cardiac actin and chondroitin sulfate proteoglycan versican) showed correlation to the progression of the disease. By comparing the pre-symptomatic, early- and mid- to late-stage DMD profiles, we set the stage for progression-specific therapeutic targets for this common and lethal inherited disease.

Novel Dysferlin Mutations in Brazilian LGMD2B Patients. F. Paula¹, M. Vainzof¹, E.S. Moreira¹, M.R. Passos-Bueno¹, K. Bushby², R. Bashir², M. Zatz¹. 1) Ctr Study Human Genome, Univ de São Paulo, São Paulo, SP, Brazil; 2) Dept of Biochemistry and Genetics, Newcastle Upon Tyne, Newcastle Univ, UK.
   The limb girdle muscular dystrophy 2B (LGMD2B) and Miyoshi myopathy (MM) are caused by allelic mutations in the dysferlin gene. Only few pathogenic changes have been reported to date. In order to establish a genotype-phenotype correlation we are analyzing the distribution of dysferlin mutations in patients from 18 LGMD2B and MM Brazilian families , classified by linkage analysis (10) or dysferlin deficiency (8).
   Through SSCP, dHPLC and sequencing of abnormal fragments we identified to date 5 different mutations in 7 unrelated families: 2 missense changes in heterozigosity (the L189V in 2 families and D396H in one, exons 6 and 13 respectively), one stop codon in exon 28 (W999X) in 2 families (one homozigous and another in heterozigous state) and 2 frameshift mutations in the last 2 families: 3522-3523delTC (in one allele) and 3446-3453insCAGTGCTT (a duplication of 8pb in homozigosity state), both in exon 29. Dysferlin protein analyzed in muscle from all of these families showed absence of this protein. No hot spot for mutations was identified.
   All these patients related weakness first in the distal muscles of the lower limbs and involvement in upper limbs on average 10 years later , but with a variable course even in patients belonging to the same family. All showed slow progression and on average, the phenotype in patients with nonsense mutations was not more severe than among those with missense mutations. In one of the families with the W999X mutation , the parents (deceased in their sixties) who were first degree cousins, were both affected. They had 10 children, also with a variable course. The intrafamilial variability , despite the high degree of consanguinity in this unusual family gives further support to the existence of modifying factors modulating the severity of the phenotype. Supported by FAPESP/CEPID, CNPq, PRONEX and IAEA.
   .

Expression profiling in facioscapulohumeral muscular dystrophy (FSHD) indicates a defect in myogenic differentiation. S. Winokur¹, J. Martin¹, Y-W. Chen², J. Ehmsen¹, K. Flanigan³. 1) Dept Biological Chemistry, Univ California, Irvine, CA; 2) Children's National Medical Center, Washington, D.C; 3) Eccles Institute of Genetics, University of Utah, Salt Lake City, UT.
   Facioscapulohumeral muscular dystrophy (FSHD) is an autosomal dominant neuromuscular disorder resulting from integral deletions of a 3.3 kb tandem repeat (D4Z4) in the subtelomeric region of chromosome 4q. Although the specific genes affected by this deletion have not been identified, a regional disruption of chromatin structure affecting local gene expression is thought underlie the pathophysiology. The global gene expression profiling of mature muscle tissue in FSHD presented here elucidates aberrant cellular processes previously concealed by this unusual molecular mechanism. FSHD expression profiles were compared to those generated from normal muscle and other types of muscular dystrophies (DMD, aSGD, JDM) in order to determine FSHD specific changes. In addition, matched biopsies (affected and unaffected muscle) from patients with FSHD served to monitor expression changes during the progression of the disease and to reduce non-specific changes resulting from individual variability. Among 15 genes upregulated in an FSHD specific and highly significant manner were several involved in cellular differentiation and proliferation. These include muscle LIM protein (MLP), delta homolog (DLK1) and the mitotic inhibitor huWEE1. As MLP and DLK were not increased further in the affected to unaffected FSHD muscle, these may represent a primary change in FSHD rather than an effect of progressing dystrophy. Increased MLP and DLK expression was confirmed by both real time RT-PCR and a second GeneChip format. The majority of FSHD-specific genes with reduced expression reflect a diminished capacity to buffer oxidative stress. These FSHD specific downregulated genes include SNAPC1 (HHCPA78 homolog) and nuclear factor of kappa light polypeptide gene alpha (NFKB1A). Enhanced vulnerability to oxidative stress is a characteristic previously demonstrated in cultured FSHD undifferentiated myoblasts. We propose a model linking aberrant myogenic differentiation and oxidative stress to subtelomeric shortening in FSHD.

Detection of Mutations in the Dystrophin Gene Via Automated DHPLC Screening and Direct Sequencing. R.R. Bennett¹, J. denDunnen², K. O"Brien³, B. Darras¹, L. Kunkel^1,3,4. 1) Division of Genetics, Children's Hospital, Boston, MA; 2) Department of Human and Clinical Genetics, Leiden University Medical Center, Leiden, Netherlands; 3) Department of Genetics, Harvard Medical School, Boston, MA; 4) Howard Hughes Medical Institute, Childrens Hospital, Harvard Medical School, Boston, Massachusetts,USA.
   Duchenne Muscular Dystrophy (DMD) is a genetic disorder which manifests itself in one of every 1600 to 2000 live births. One in 3200 to 4000 will be severly disabled boys and one in 3200 to 4000 will be carrier females with reproductive complications and potential late-onset cardiac complications. Approximately 65% of mutations causing this disease are large deletions or duplications, most of which are found by the existing set of multiplex PCR primers. The other 35% have remained undetected in most patients. The purpose of this project was to develop an efficient and inexpensive process for detecting mutations large and small. We have obtained sequence upstream and downstream of each exon as well as the 5' and 3' UTRs from the NIH human genome database and have designed primers to include 30 to 100 bases on either side of each exon. These primers were designed to produce a single clean band on agarose gel, to create DNA fragments with melting characteristics appropriate for DHPLC analysis using the WAVE from Transgenomic Inc. for DNA variation screening, and finally to sequence well on automated DNA sequencers. We have tested eight patients --selected from medical records as clearly having DMD but no large mutation-- using these new primers in conjuction with the WAVE and PE-ABI 373 sequencer. We have detected six disease-causing mutations. The remaining two patients have been completely sequenced. Several polymorphisms have been detected in each, most of which are known to be non disease-causing. Further investigation is required to determine the cause of DMD in these two patients.

MAPH detection of deletions/duplications in Duchenne/Becker Muscular Dystrophy: an alternative to quantitative Southern blotting. M.H. Breuning, S. White, M. Kalf, M. Villerius, E. Bakker, G.-J.B. van Ommen, J.T. den Dunnen. Center for Human and Clinical Genetics, LUMC, Leiden, Netherlands.
   Currently most genetic diagnostic protocols are PCR-based and do not readily yield quantitative data. As a consequence, potential deletions and duplications of the regions analyzed go undetected unless specific methods are applied. Southern blotting is most commonly used, but is time consuming and laborious. Recently an alternative method was published, called Multiplex Amplifiable Probe Hybridization (MAPH). In this technique a series of short DNA fragments are cloned in such a way that all can be PCR-amplified using one pair of primers. The probes are hybridized to genomic DNA immobilized on nylon filters, and after stringent washing the probes are recovered off the filters and PCR- amplified in a quantitative manner. We have applied this technique to the diagnosis of Duchenne/Becker Muscular Dystrophy (DMD/BMD), diseases caused by mutations in the dystrophin gene. This gene is the largest known, covering 2.4 Mb, containing 79 exons. In approximately 65 % of cases the mutation is a deletion or duplication of one or more exons. Frame-shift mutations cause the lethal DMD, whereas maintenance of the reading frame leads to the less severe BMD. For this reason it is important to assess the boundaries of rearrangements, a potentially arduous task. A technique allowing simultaneous analysis of all exons would greatly simplify this procedure. We cloned all 79 exons into the same vector and divided the PCR products into 2 pools. Following hybridization the secondary PCR was performed using a fluorescently labeled primer, allowing the products to be analyzed on a 96 capillary sequencer. This allows parallel analysis of 96 samples in ~48 hours. The number of copies of each exon could be determined by comparing the appropriate peaks between controls and patients. Using this technique we were able to detect exon deletions and duplications missed using current methods. Further probe sets are being developed to cover other areas of interest such as breast cancer and deletion syndromes. The possibility of analyzing many more regions simultaneously using microarrays is being investigated.

Point mutations in the dystrophin gene: a super hot spot of mutation at a CpG dinucleotide and a difference in the pattern of microdeletions. C.H. Buzin¹, J. Yan¹, J. Feng¹, J.R. Mendell², S.S. Sommer¹. 1) Molecular Genetics, City of Hope Medical Center , Duarte, CA; 2) Ohio State University Medical Center, Columbus, OH.
   About 35% of patients with Duchenne muscular dystrophy have point mutations in the dystrophin gene, one of the largest known genes that spans nearly 2.4 Mb of genomic DNA, contains 79 exons, and encodes a 14 kb transcript. Comprehensive mutation scanning was performed with DOVAM-S, a partially automated method that detects virtually all mutations. In this study, genomic DNA from 141 patients with DMD, ascertained both clinically and by muscle biopsy, was scanned in all coding exons, associated intronic splice regions, and promoter sequences. Although prior screening had excluded most patients with large deletion and duplication mutations, eight of the patients were found to have a large deletion. Of the remaining 133 patients, 98 (74%) had truncating point mutations, including nonsense, small deletions or insertions, and splice mutations. Of the 98 causative point mutations, 94 were independent as determined by haplotype analysis. No causative missense mutations were found, although many missense polymorphisms were discovered, all of which were seen also as second site changes in patients in which a truncating mutation was found. For frameshift and nonsense mutations (86 independent mutations), a mutation target could be determined within the coding region. One super hot spot mutation, a transition at CpG in exon 59 (arg2905ter), was detected in six patients, all with different haplotypes. When the point mutation types are normalized for their target, the relative rates of microdeletions/insertions, transitions and transversions at and not at CpG were similar to the factor IX gene. However, the size distribution of microdeletions was significantly different in DMD and factor IX.

Laminin is required to produce membrane instability in muscular dystrophy mediated by sarcoglycan loss. E.M. McNally¹, K. Lapidos², M. Hadhazy¹, A.A. Hack². 1) Department of Medicine, The University of Chicago, Chicago, IL; 2) Department of Molecular Genetics Cell Biology, The University of Chicago, Chicago, IL.
   The sarcoglycans, a group of transmembrane proteins in muscle and heart, are part of the dystrophin glycoprotein complex (DGC). Mutations in sarcoglycan genes cause cardiomyopathy and muscular dystrophy in humans and in mice by producing instability of the remaining sarcoglycans while leaving dystrophin intact. In contrast, mutations in dystrophin cause a reduction in not only the sarcoglycans, but also in additional DGC proteins such as the syntrophins and dystrobrevins. We used a genetic analysis to evaluate whether sarcoglycan is sufficient to mediate membrane instability and by generating mice lacking both dystrophin (mdx) and g-sarcoglycan. Mice lacking both g- sarcoglycan and dystrophin (gdx) had an identical phenotype to mice lacking only g-sarcoglycan with a similar degree of central nucleation and serum creatine kinase elevation. This demonstrates that sarcoglycan loss is sufficient to produce alterations in membrane permeability and the dystrophic process. In parallel experiments, we evaluated the role of the extracellular matrix protein laminin in the dystrophic process by producing mice mutant for both g-sarcoglycan and laminin-a2. Laminin-a2 is the major site of attachment to the extracellular matrix for the DGC. Mice mutant for both g-sarcoglycan and laminin-a2, (gdy), surprisingly displayed a severe dystrophic process yet had little evidence for disruption of membrane integrity. Little to no Evans Blue Dye uptake was seen and serum creatine kinase was only minimally elevated in (gdy) mice. These genetic analyses demonstrate that sarcoglycan loss is sufficient to produce the dystrophic process, and that laminin is required for disruption of membrane integrity and increased membrane permeability that is seen when sarcoglycan is absent.

PFGE analysis of 4qter-10qter interchromosomal exchanges in Italian families with Facioscapulohumeral muscular dystrophy (FSHD). L. Felicetti^1,2, G. Galluzzi^1,2, L. Colantoni^1,2, M. Rossi^1,2, B. Merico², F. Mangiola¹, P. Tonali², E. Ricci². 1) Ctr Neuromuscular Diseases, UILDM, Rome, Italy; 2) Institute of Neurology, Catholic University, Rome , Italy.
   In 95% of FSHD patients, molecular diagnosis is based on the detection by probe p13E-11 of EcoRI, BlnI-resistant fragments of 10 to 35 kb, shorter than those found in normal individuals (35-300 kb). The size reduction is due to a deletion of a variable number of KpnI repeats at 4q35 region. 4q-10q interchromosomal exchanges occur both in normal and FSHD subjects, resulting in the reshuffling of 4q-type BlnI-resistant and 10q-type BlnI-sensitive repeats from one chromosome to the other. We analyzed the segregation of BlnI-resistant and BlnI-sensitive alleles in 55 FSHD Italian families for a total of 230 individuals (116 affected, 70 unaffected and 44 spouses). DNA extraction and subsequent restriction steps with EcoRI, BlnI and Tru9I were performed directly in agarose blocks. After separation by Pulsed Field Gel Electrophoresis (PFGE), the alleles were identified with p13E-11 and KpnI cloned sequences as probes. We observed different types of exchanges: total 4q to 10q transfers resulting in trisomy and tetrasomy; partial 10q to 4q transfers leading to monosomy and nullisomy; complex rearrangements such as multiple translocations, even in subjects with p13E-11 standard allele configuration. In addition, we detected two patients carrying 10q-4q translocations involving the short fragments associated with FSHD. Among 8 sporadic cases carrying a de novo p13E-11 small fragment (BlnI-resistant), we found four patients, with no evidence of somatic mosaicism, with one or both parents showing various types of rearrangements (trisomy, monosomy, etc.). The remaining four were somatic mosaics, all carrying 4q-10q translocations. Our results show the high frequency of 4qter-10qter interchromosomal exchanges in FSHD families and confirm that the instability of subtelomeric regions can play a role in the molecular mechanism of the disease. Telethon Italy grant n.1296.

Detection of expanded CAG repeats at the myotonic dystrophy (DM) locus in single cells by triplet primed PCR (TP-PCR) for preimplantation genetic diagnosis (PGD). P.F. Ray¹, N. Frydman², R. Frydman², M. Vekemans¹, A. Munnich¹. 1) Departement de genetique medicale, Hopital Necker, PARIS; 2) Service de genicologie-obstetrique, Hopital Antoine Beclere, Clamart, France.
   The amplification of large number of triplet repeats is challenging from microgram amounts of DNA but is impossible from single cells. Large pathologic alleles responsible for myotonic dystrophy (DM) can expand up to several thousands triplets but a maximum of 50-100 repeats can reliably be amplified from single cells. Thus, genetic analysis in the course PGD for couples at risk of transmitting myotonic dystrophy has so far mostly been based on the detection of the normal allele of the transmitting parent. This semi-indirect strategy, however, is only applicable when that allele is different in size from the proposituss partner two alleles, from our experience, only about one fourth of the couples. Warner et al. (1996)* described a fluorescent assay in which a primer internal to the triplet repeat itself was used and allowed the detection of expanded alleles. We utilized this strategy and developed a single cell hemi nested TP-PCR assay. The outer reaction mix contained a DM specific forward primer and the TP primer made of 7 CAG repeats tailed with an overhanging 20 bp specific primer. Aliquots of this reaction were reamplified with the same 5 fluorescently labeled primer and the 20 nucleotides specific primer. Fluorescent TP-PCR products were analyzed in an ABI 310 DNA analyzer. Expanded alleles were detected in 100% of the control lymphocytes analyzed and one PGD cycle was carried out. Following this procedure two embryos which had not shown amplification of the expanded allele were transferred. Positive hCG were detected two weeks later, indicating a successful uterine implantation of an embryo but the pregnacy was not sustained. Reliable detection of expanded CAG repeats was achieved from single lymphocytes and blastomeres by fluorescent TP-PCR. This single cell analysis technique is applicable to all couples at risk of transmitting myotonic dystrophy and similar protocols could be developed for the diagnosis of other trinucleotide expansion diseases. * J Med Genet, 1996, (33) 1022-6.

Functional correction of adult mdxmouse muscle using gutted adenoviral vectors expressing full-length dystrophin. C. DelloRusso^1,3, J. Scott³, D. Hartigan-O'Connor³, C. Barjot², G. Salvatori², A. Robinson², S. Brooks¹, J. Chamberlain³. 1) Departments of Physiology and; 2) Human Genetics, University of Michigan, Ann Arbor; 3) Department of Neurology, University of Washington, Seattle.
   Duchenne muscular dystrophy is a degenerative lethal muscle disorder caused by mutations in the dystrophin gene. Adenoviral vectors are promising tools that may be used to express dystrophin in affected muscle. We have constructed 3 gutted adenoviral vectors devoid of all viral genes and containing a full-length human (HDys) or mouse (MDys and GEbDys) dystrophin cDNA driven by a muscle specific promoter. Virus was injected into 1 year old mdxmouse TA muscles and a novel lengthening contraction (LC) protocol was used to test for restored muscle function. This protocol reveals the high susceptibility of mdxTA muscles to contraction-induced injury; after 1 LC, wild type muscle force is reduced by 10% while mdxmuscles show a 72% force loss. After 5 days, HDys injected muscles were significantly protected from injury and expressed high levels of dystrophin in contrast to sham injected control muscles. However, 25 days after injection, significant decreases in force generating capacity were detected. This loss of force was similar to that observed after injection of a first generation virus containing a LacZ transgene (CNb). MDys and GEbDys injected muscles demonstrated high levels of dystrophin protein expression and no functional defects after 25 days. In addition, MDys injected muscles were able to produce 62% of wild type force levels after 1 LC that, in contrast, reduced mdxand sham injected control muscles to 38 and 29% of wild type force, respectively. FACS sorting revealed the least amount of CD4+ and CD8+ cells in MDys injected muscles, 1/3 more immune cells in HDys injected muscles, and the highest number in muscles injected with CNb. Collectively, these data demonstrate 1) gutted adenoviral vectors are successful in transducing dystrophin in TA muscles of 1 year old mdxmice, 2) evidence for an immune response against human dystrophin that causes atrophy and loss of function, and 3) a partial functional correction of adult mdxmouse muscle after delivery of full-length mouse dystrophin

Delivery of Functional Four-Repeat Micro-Dystrophin to Mdx Muscle Via Different AAV Serotypes. S. Harper¹, C. DelloRusso¹, R. Crawford¹, H. Harper¹, J. Engelhardt², D. Duan², J. Chamberlain¹. 1) Dept of Neurology, University of Washington, Seattle, WA; 2) Dept of Anatomy and Cell Biology, University of Iowa, Iowa City, IA.
   The goal of this study was to evaluate the correction of Duchenne muscular dystrophy (DMD) in the mdx mouse model using different serotypes of AAV carrying extremely small dystrophin genes (~3.5 kb vs. the full-length 14 kb). The rod domain of dystrophin is composed of 24 spectrin-like repeats, and natural, in-frame deletions of this region lead to a milder form of dystrophy called Becker MD. Based on our prior analysis in transgenic mice, we created several rod domain and C terminal deletions that resulted in dystrophin cDNAs small enough to fit into AAV. AAV is a highly efficient vector that does not elicit an immune response in dystrophic muscle if it is carrying a gene driven by a muscle-specific promoter. AAV-2 is the most prevalent and well-characterized system, but recent studies show that serotypes 1, 5, and 6 boost gene expression 100- to 1000-fold in mouse muscle. We generated transgenic mice expressing several different micro-dystrophin clones containing only four repeats, and analyzed their ability to correct muscular dystrophy by both morphological and physiological assays. Two of three transgenes significantly reduced the amount of dystrophy observed in the mdx limb and diaphragm muscles. The best transgene (DR4-R23) produced muscle with wild-type levels of central nuclei, membrane integrity, resistance to contraction-induced injury, and the ability to run on a treadmill. No areas of fibrosis nor monocyte infiltration were observed. We subsequently demonstrated that highly functional micro-dystrophins can be successfully delivered via AAV-2 to muscles of young adult mdx mice. We are currently testing gene expression of our best micro-dystrophin with alternate AAV serotypes, a more efficient Kozak sequence, and a highly active, mutant form of the muscle creatine kinase promoter (CK6). Functional correction of muscular dystrophy using these vectors will be compared to results obtained with our gutted adenovirus vector system that expresses full-length dystrophin.

Amelioration of dystrophic phenotypes in transgenic mdx mice expressing truncated dystrophin cDNA. M. Sakamoto, K. Yuasa, T. Yokota, S. Masuda, Y. Miyagoe-Suzuki, S. Takeda. Molecular Therapy, Natl Inst Neurosci NCNP, Kodaira, Tokyo, Japan.
   Duchenne Muscular Dystrophy (DMD) is an X-linked, lethal disorder caused by a defect in the dystrophin gene. Progressive muscle weakness, cardiomyopathy and early death characterize the disease. Dystrophin is localized at the inner surface of the plasma membrane and forms the complex with dystrophin-associated proteins (DAPs) to stabilize the sarcolemma by linking the cytoskeleton to the extracellular matrix. Dystrophin has a central rod domain, which consists of 24 triple-helical repeats and 4 hinge segments, and accounts for 76% of the molecule. It has been shown that a large in-frame deletion in this domain results in a mild allelic form of the disease, Becker muscular dystrophy (BMD).
   Adeno-associated virus (AAV) vector-mediated micro-dystrophin cDNA transfer is one of attractive approaches for the treatment of DMD. AAV vector allows long-term expression of the transferred gene without significant immune responses, while it has a limited insertion size up to 4.7-4.9 kb. Therefore, we cannot accommodate a full-length dystrophin cDNA (14 kb) into AAV vector. To find a functional, but small-sized dystrophin, we generated a series of rod-truncated micro-dystrophin cDNAs with one rod repeat and two hinges (named M3 construct), three rod repeats and two hinges (AX11 construct), and four rod repeats and three hinges (CS1 construct). We have generated transgenic mdx mice expressing micro-dystrophins and evaluated their function to improve dystrophic phenotype. Histological examination showed that CS1 construct recovered DAPs at the sarcolemma and significantly reduced muscle degeneration in the hind limb muscles and diaphragm. On the other hand, M3 construct could not ameliorate mdx phenotype in spite of recovery of DAPs at the sarcolemma. These data suggest that the rod structure, especially its length is critical for function of micro-dystrophin. We are now investigating the phenotypes of AX11-transgenic mdx mice. Our goal is to identify a functional micro-dystrophin and transfer it into dystrophin-deficient dystrophic muscles using AAV vector

Modulation of muscular dystrophy by inhibitors of apoptosis. S. Abmayr, R.W. Crawford, J.S. Chamberlain. Dept Neurology, Univ Washington, Seattle, WA.
   Duchenne muscular dystrophy (DMD) is a human X-linked recessive disorder caused by mutations in the dystrophin gene. DMD is characterized by a progressive loss of muscle function, leading to death in the early twenties. The relationship between the protein deficiency and the late clinical onset of muscle pathology remains unclear. Several lines of evidence support the hypothesis that cell death of dystrophin deficient muscle might be initiated by apoptosis, followed by necrotic processes. We have been studying the role of proteins that repress apoptosis or/and enhance muscle regeneration for their ability to modulate the dystrophic phenotype in conjunction with gene replacement therapy. We have generated transgenic mice that overexpress ARC, an apoptosis inhibitor, almost exclusively expressed in skeletal muscle and heart. ARC was shown previously to interact selectively with caspases and to prevent hypoxia-induced release of cytochrome c from mitochondria. Morphology studies were performed to determine the level of de-and regeneration of muscle fibers of ARC transgenic/mdx mice in comparison with mdx controls. Tg/mdx mice displayed a clear pattern of mdx pathology. We performed immunohistochemistry analysis to compare the level of activated caspase-3 in mdx and transgenic/mdx muscle and detected no significant difference in the levels of activation. Caspase-3 positive fibers displayed membrane damage as assessed by uptake of the vital dye Evans blue, suggesting apoptotic pathways may be activated as a result of membrane leakage. Together the results suggest that overexpression of ARC does not overtly protect mdx muscle from dystrophic pathology. Currently we are investigating if ARC plays an inhibitory role in hypoxia-induced apoptosis in muscle, which might contribute to improved muscle function in less sedentary animals. In a complementary approach, we are examining the role of IGF-I in muscle and its ability to modulate the mdx muscle pathology. IGF-I has been shown to enhance muscle regeneration and to prevent age-related declines in muscle mass and function. We have generated adenoviral vectors expressing IGF-I and dystrophin and are studying their effects in preventing and reversing dystrophic pathology in mdx mice.

Adenoviral Vector Systems for Delivery of Dystrophin cDNA Expression Constructs to Mouse Skeletal Muscle. J.M. Scott, D.J. Hartigan-O'Connor, C. Barjot, S.Q. Harper, A.S. Robinson, C. DelloRusso, R.W. Crawford, J.S. Chamberlain. Neurology, University of Washington, Seattle, WA.
   Gene therapy using viral vectors holds promise for several human genetic diseases but much remains to be learned about which vector system will meet all of the requirements of such a treatment. Our laboratory has focused on modification of adenoviral (Ad) vectors in order to improve the efficiency of delivering expression constructs to skeletal muscle with respect to transgene expression levels, persistence, immune response, and tissue-specific vs. constitutive promoters. Improvements in first generation Ad vectors that carry mini-dystrophin cassettes was achieved by deleting the viral polymerase gene which rendered the vector replication defective and eliminated viral late gene expression in transduced cells. We have also made gutted Ad vectors lacking all viral ORFs, and adeno-associated virus (AAV) containing a micro-dystrophin expression cassette. These systems are compared using the mdx mouse model of Duchenne muscular dystrophy whereby skeletal muscle is injected and assays are performed to evaluate dystrophin protein expression levels, persistence of the transgene, immune cell infiltration, and integrity of the muscle fiber membranes. Our results indicate that gutted Ad vectors containing a dystrophin expression cassette driven by a strong muscle-specific promoter (modified MCK) lead to significantly increased transgene expression and decreased immune response compared to the first generation viruses. We find that the MCK promoter is highly effective in reducing the antigen-specific immune response against Ad vectors despite the muscle cell death and antigen-presenting cell infiltration in dystrophic muscle. A gutted Ad vector expressing full-length utrophin is also being compared with the dystrophin vector for function, persistence and immunogenicity. Additional improvements have been noted using an AAV vector which exhibits robust expression of dystrophin and even less immune response than that elicited by Ad vectors. These results show that modified Ad and AAV vectors can lead to significant and prolonged gene expression in dystrophic muscle.