Limb girdle muscular dystrophies (LGMDs) are a clinically and genetically heterogeneous group of myopathies characterized by weakness and wasting of the proximal musculature. There are currently seventeen loci associated with different LGMDs, seven with an autosomal dominant mode of inheritance (LGMD1A–1G) and 10 with an autosomal recessive mode of inheritance (LGMD2A– 2J). The cumulative worldwide prevalence of LGMD is thought to be ~1/15,000. In the Hutterite population of North America there is an over-representation of autosomal recessive LGMD with a prevalence estimated to be >1/400. The objective of this work was to delineate the genetic basis of LGMD in this large genetically isolated population. A genome-wide scan was performed on Hutterite LGMD patients and their families in order to locate the mutant gene. This allowed us to identify a novel locus at chromosome region 9q31-33 that was named LGMD2H. Extensive haplotyping and mutation screening led to the discovery of c.1459G>A in TRIM32 as the causative mutation of LGMD2H. We then found that this same mutation was the cause of another previously described myopathy in the Hutterites, sarcotubular myopathy (STM)[reference awaiting publishers decision]. Analysis of the TRIM32 gene product revealed that it is a potential E3-ubiquitin ligase, is expressed in many human tissues including muscle and brain, and has a punctate cytoplasmic distribution. During the analysis of the LGMD2H region, it became apparent that there were Hutterite LGMD patients not linked to the LGMD2H locus. In order to identify the causative gene(s) in the remaining families, we performed a genome-wide scan. A locus at chromosome 19q13 was found to correspond to disease inheritance, the site of a previously described LGMD locus, LGMD2I. No causative gene had yet been identified at this locus so haplotyping and mutation screening was performed. We were able to identify c.826C>A in FKRP as the causative mutation in our remaining cohort of LGMD patients. The same mutation has since been found in many other populations, and is apparently a relatively common cause of LGMD. We obtained DNA from 19 non-Hutterite LGMD2I patients of diverse origins with c.826C>A and determined that it is an old founder mutation. There is no further evidence of any other loci causing autosomal recessive myopathy in the Hutterites. With the identification of c.1459G>A in TRIM32 and c.826C>A in FKRP we appear to have delineated the genetic cause of all myopathies of increased prevalence in the Hutterite population. To date, we have been able to provide accurate, non-invasive, diagnosis to over 70 patients and have provided carrier testing to approximately 120 at-risk family members. This kind DNA-based approach is not feasible in the general population due the enormous amount of locus, allelic, and clinical heterogeneity among myopathy patients. / May 2005

Limb girdle muscular dystrophies (LGMDs) are a clinically and genetically heterogeneous group of myopathies characterized by weakness and wasting of the proximal musculature. There are currently seventeen loci associated with different LGMDs, seven with an autosomal dominant mode of inheritance (LGMD1A–1G) and 10 with an autosomal recessive mode of inheritance (LGMD2A– 2J). The cumulative worldwide prevalence of LGMD is thought to be ~1/15,000. In the Hutterite population of North America there is an over-representation of autosomal recessive LGMD with a prevalence estimated to be >1/400. The objective of this work was to delineate the genetic basis of LGMD in this large genetically isolated population. A genome-wide scan was performed on Hutterite LGMD patients and their families in order to locate the mutant gene. This allowed us to identify a novel locus at chromosome region 9q31-33 that was named LGMD2H. Extensive haplotyping and mutation screening led to the discovery of c.1459G>A in TRIM32 as the causative mutation of LGMD2H. We then found that this same mutation was the cause of another previously described myopathy in the Hutterites, sarcotubular myopathy (STM)[reference awaiting publishers decision]. Analysis of the TRIM32 gene product revealed that it is a potential E3-ubiquitin ligase, is expressed in many human tissues including muscle and brain, and has a punctate cytoplasmic distribution. During the analysis of the LGMD2H region, it became apparent that there were Hutterite LGMD patients not linked to the LGMD2H locus. In order to identify the causative gene(s) in the remaining families, we performed a genome-wide scan. A locus at chromosome 19q13 was found to correspond to disease inheritance, the site of a previously described LGMD locus, LGMD2I. No causative gene had yet been identified at this locus so haplotyping and mutation screening was performed. We were able to identify c.826C>A in FKRP as the causative mutation in our remaining cohort of LGMD patients. The same mutation has since been found in many other populations, and is apparently a relatively common cause of LGMD. We obtained DNA from 19 non-Hutterite LGMD2I patients of diverse origins with c.826C>A and determined that it is an old founder mutation. There is no further evidence of any other loci causing autosomal recessive myopathy in the Hutterites. With the identification of c.1459G>A in TRIM32 and c.826C>A in FKRP we appear to have delineated the genetic cause of all myopathies of increased prevalence in the Hutterite population. To date, we have been able to provide accurate, non-invasive, diagnosis to over 70 patients and have provided carrier testing to approximately 120 at-risk family members. This kind DNA-based approach is not feasible in the general population due the enormous amount of locus, allelic, and clinical heterogeneity among myopathy patients.

Limb girdle muscular dystrophies (LGMDs) are a clinically and genetically heterogeneous group of myopathies characterized by weakness and wasting of the proximal musculature. There are currently seventeen loci associated with different LGMDs, seven with an autosomal dominant mode of inheritance (LGMD1A–1G) and 10 with an autosomal recessive mode of inheritance (LGMD2A– 2J). The cumulative worldwide prevalence of LGMD is thought to be ~1/15,000. In the Hutterite population of North America there is an over-representation of autosomal recessive LGMD with a prevalence estimated to be >1/400. The objective of this work was to delineate the genetic basis of LGMD in this large genetically isolated population. A genome-wide scan was performed on Hutterite LGMD patients and their families in order to locate the mutant gene. This allowed us to identify a novel locus at chromosome region 9q31-33 that was named LGMD2H. Extensive haplotyping and mutation screening led to the discovery of c.1459G>A in TRIM32 as the causative mutation of LGMD2H. We then found that this same mutation was the cause of another previously described myopathy in the Hutterites, sarcotubular myopathy (STM)[reference awaiting publishers decision]. Analysis of the TRIM32 gene product revealed that it is a potential E3-ubiquitin ligase, is expressed in many human tissues including muscle and brain, and has a punctate cytoplasmic distribution. During the analysis of the LGMD2H region, it became apparent that there were Hutterite LGMD patients not linked to the LGMD2H locus. In order to identify the causative gene(s) in the remaining families, we performed a genome-wide scan. A locus at chromosome 19q13 was found to correspond to disease inheritance, the site of a previously described LGMD locus, LGMD2I. No causative gene had yet been identified at this locus so haplotyping and mutation screening was performed. We were able to identify c.826C>A in FKRP as the causative mutation in our remaining cohort of LGMD patients. The same mutation has since been found in many other populations, and is apparently a relatively common cause of LGMD. We obtained DNA from 19 non-Hutterite LGMD2I patients of diverse origins with c.826C>A and determined that it is an old founder mutation. There is no further evidence of any other loci causing autosomal recessive myopathy in the Hutterites. With the identification of c.1459G>A in TRIM32 and c.826C>A in FKRP we appear to have delineated the genetic cause of all myopathies of increased prevalence in the Hutterite population. To date, we have been able to provide accurate, non-invasive, diagnosis to over 70 patients and have provided carrier testing to approximately 120 at-risk family members. This kind DNA-based approach is not feasible in the general population due the enormous amount of locus, allelic, and clinical heterogeneity among myopathy patients.

Les dystrophies musculaires des ceintures (ou limb-girdle muscular dystrophy, LGMD) sont un groupe hétérogène de dystrophies musculaires chez l’adulte et sont définies par une atrophie et une faiblesse progressive qui surviennent dans les muscles proximaux. Chez une cohorte canadienne-française, nous avons précédemment décrit une nouvelle forme récessive, désignée LGMD2L et marquée par une atrophie asymétrique du quadriceps, que nous avions cartographiée au chromosome 11p12-p13 grâce à des analyses de liaison. L’objectif de ce projet de thèse était de raffiner l’intervalle candidat, puis d’identifier et de caractériser le gène muté responsable de la LGMD2L. Grâce à une cartographie par homozygotie de polymorphismes de nucléotide simple (SNPs) réalisée sur une grande famille consanguine, nous avons redéfini l’intervalle candidat à une région du chromosome 11p14.3-p15.1. Par séquençage de l’ADN génomique et complémentaire au gène Anoctamine 5 (ANO5) inclus dans cet intervalle, nous avons identifié trois mutations, chez autant de familles: une substitution créant un site d’épissage aberrant, une insertion d’un nucléotide et une mutation faux-sens. Les deux premières mutations étaient associées à une hausse de la dégradation de l’ARN messager médiée par une troncation prématurée. Nous avons également identifié des mutations ANO5 chez une seconde dystrophie musculaire de type distal cartographiant au même locus que la LGMD2L, nommée MMD3, et dont la manifestation initiale était une faiblesse des mollets, mais qui pouvait progresser vers une atrophie des quadriceps. Une réparation membranaire défective avait été observée chez les fibroblastes de deux patients MMD3, suggérant un rôle pour ANO5 dans ce mécanisme. La localisation et la fonction d’ANO5 dans le muscle sont inconnues, mais cette protéine fait partie d’une famille conservée de protéines à huit domaines transmembranaires, les Anoctamines, dont certains membres sont des transporteurs chloriques activés par le calcium. Les résultats de nos études d’immunofluorescence suggèrent qu’ANO5 se localise peu au sarcolemme, mais plutôt à une structure intracellulaire qui suit la ligne Z des myofibrilles. De façon étonnante, cette localisation était préservée chez un patient LGMD2L porteur homozygote de la mutation d’épissage, en dépit du fait que cette dernière était considérée comme une mutation nulle. Néanmoins, nous avons identifié un épissage alternatif de l’exon 15 qui se produisait sur une proportion des transcrits porteurs de la mutation d’épissage, ce qui rétablirait le cadre de lecture, soulignant la complexité de la régulation de l’épissage d’ANO5 et laissant croire que la LGMD2L pourrait être causée par une perte de fonction partielle, et non complète, d’ANO5. Des études subséquentes par des groupes européens ont montré que les anoctaminopathies 5 sont une cause fréquente de dystrophies musculaires des ceintures chez l’adulte. Notre découverte de mutations au gène Anoctamine 5 a mis en évidence une nouvelle classe de protéines importantes pour la biologie du muscle et a ouvert la voie à de nouvelles pistes pour étudier les mécanismes par lesquels un défaut de réparation membranaire progresse en une dystrophie musculaire. / Limb-girdle muscular dystrophies (LGMD) encompass a broad spectrum of muscular dystrophies in which the initial weakness arises in proximal muscles. We previously described in French-Canadian (FC) families a new form of LGMD characterized by asymmetrical quadriceps femoris atrophy, named LGMD2L, which we mapped to chromosome 11p12-p13 using linkage analyses. The objectives of this thesis project were to refine the candidate interval, identify and characterize the LGMD2L gene. Using single nucleotide polymorphisms (SNPs) homozygosity mapping in a large consanguineous family, we narrowed down the LGMD2L candidate interval to a region on chromosome 11p14.3-p15.1, and identified three mutations in the Anoctamin 5 (ANO5) gene located in the interval. These mutations consisted of a missense, a one-bp duplication and a splice site mutation. We demonstrated that the latter two triggered the nonsense-mediated RNA decay pathway. In addition, we identified ANO5 mutations in cases affected by a non-dysferlin Miyoshi muscular dystrophy mapped also to chromosome 11, termed MMD3. In two MMD3 families of European descent, patients presented with calf weakness as the initial symptoms, sometimes evolving to quadriceps atrophy. Fibroblasts from one MMD3 family were shown to be defective for membrane repair. ANO5 localization and function in muscle are unknown, but it is a member of the conserved Anoctamin family of proteins with eight transmembrane domains, of which some function as calcium-activated chloride channel. Our immunofluorescence studies on longitudinal muscle sections suggest that ANO5 is not importantly localized to the sarcolemma, but rather to a structure following the Z-line. To our surprise, this localization was preserved for a LGMD2L patient homozygous for the splice site mutation, previously considered as a null mutation. By studying the splicing isoforms in this patient, we observed that skipping of exon 15 occurs on a proportion of transcripts, in addition to the aberrant splicing caused by the mutation. This alternative splicing event would recover the reading frame, thus underlining the complexity of ANO5 splicing and suggesting that LGMD2L could be the consequence of a partial, rather than complete, loss-of-function. Subsequent studies by other groups have shown that anoctaminopathies 5 are a common cause of adult-onset LGMD. Our discovery of ANO5 mutations has shed light on a new class of proteins important for the muscle biology and opened new research avenues to study how defective membrane repair progresses into muscular dystrophies.

Les dystrophies musculaires des ceintures (ou limb-girdle muscular dystrophy, LGMD) sont un groupe hétérogène de dystrophies musculaires chez l’adulte et sont définies par une atrophie et une faiblesse progressive qui surviennent dans les muscles proximaux. Chez une cohorte canadienne-française, nous avons précédemment décrit une nouvelle forme récessive, désignée LGMD2L et marquée par une atrophie asymétrique du quadriceps, que nous avions cartographiée au chromosome 11p12-p13 grâce à des analyses de liaison. L’objectif de ce projet de thèse était de raffiner l’intervalle candidat, puis d’identifier et de caractériser le gène muté responsable de la LGMD2L. Grâce à une cartographie par homozygotie de polymorphismes de nucléotide simple (SNPs) réalisée sur une grande famille consanguine, nous avons redéfini l’intervalle candidat à une région du chromosome 11p14.3-p15.1. Par séquençage de l’ADN génomique et complémentaire au gène Anoctamine 5 (ANO5) inclus dans cet intervalle, nous avons identifié trois mutations, chez autant de familles: une substitution créant un site d’épissage aberrant, une insertion d’un nucléotide et une mutation faux-sens. Les deux premières mutations étaient associées à une hausse de la dégradation de l’ARN messager médiée par une troncation prématurée. Nous avons également identifié des mutations ANO5 chez une seconde dystrophie musculaire de type distal cartographiant au même locus que la LGMD2L, nommée MMD3, et dont la manifestation initiale était une faiblesse des mollets, mais qui pouvait progresser vers une atrophie des quadriceps. Une réparation membranaire défective avait été observée chez les fibroblastes de deux patients MMD3, suggérant un rôle pour ANO5 dans ce mécanisme. La localisation et la fonction d’ANO5 dans le muscle sont inconnues, mais cette protéine fait partie d’une famille conservée de protéines à huit domaines transmembranaires, les Anoctamines, dont certains membres sont des transporteurs chloriques activés par le calcium. Les résultats de nos études d’immunofluorescence suggèrent qu’ANO5 se localise peu au sarcolemme, mais plutôt à une structure intracellulaire qui suit la ligne Z des myofibrilles. De façon étonnante, cette localisation était préservée chez un patient LGMD2L porteur homozygote de la mutation d’épissage, en dépit du fait que cette dernière était considérée comme une mutation nulle. Néanmoins, nous avons identifié un épissage alternatif de l’exon 15 qui se produisait sur une proportion des transcrits porteurs de la mutation d’épissage, ce qui rétablirait le cadre de lecture, soulignant la complexité de la régulation de l’épissage d’ANO5 et laissant croire que la LGMD2L pourrait être causée par une perte de fonction partielle, et non complète, d’ANO5. Des études subséquentes par des groupes européens ont montré que les anoctaminopathies 5 sont une cause fréquente de dystrophies musculaires des ceintures chez l’adulte. Notre découverte de mutations au gène Anoctamine 5 a mis en évidence une nouvelle classe de protéines importantes pour la biologie du muscle et a ouvert la voie à de nouvelles pistes pour étudier les mécanismes par lesquels un défaut de réparation membranaire progresse en une dystrophie musculaire. / Limb-girdle muscular dystrophies (LGMD) encompass a broad spectrum of muscular dystrophies in which the initial weakness arises in proximal muscles. We previously described in French-Canadian (FC) families a new form of LGMD characterized by asymmetrical quadriceps femoris atrophy, named LGMD2L, which we mapped to chromosome 11p12-p13 using linkage analyses. The objectives of this thesis project were to refine the candidate interval, identify and characterize the LGMD2L gene. Using single nucleotide polymorphisms (SNPs) homozygosity mapping in a large consanguineous family, we narrowed down the LGMD2L candidate interval to a region on chromosome 11p14.3-p15.1, and identified three mutations in the Anoctamin 5 (ANO5) gene located in the interval. These mutations consisted of a missense, a one-bp duplication and a splice site mutation. We demonstrated that the latter two triggered the nonsense-mediated RNA decay pathway. In addition, we identified ANO5 mutations in cases affected by a non-dysferlin Miyoshi muscular dystrophy mapped also to chromosome 11, termed MMD3. In two MMD3 families of European descent, patients presented with calf weakness as the initial symptoms, sometimes evolving to quadriceps atrophy. Fibroblasts from one MMD3 family were shown to be defective for membrane repair. ANO5 localization and function in muscle are unknown, but it is a member of the conserved Anoctamin family of proteins with eight transmembrane domains, of which some function as calcium-activated chloride channel. Our immunofluorescence studies on longitudinal muscle sections suggest that ANO5 is not importantly localized to the sarcolemma, but rather to a structure following the Z-line. To our surprise, this localization was preserved for a LGMD2L patient homozygous for the splice site mutation, previously considered as a null mutation. By studying the splicing isoforms in this patient, we observed that skipping of exon 15 occurs on a proportion of transcripts, in addition to the aberrant splicing caused by the mutation. This alternative splicing event would recover the reading frame, thus underlining the complexity of ANO5 splicing and suggesting that LGMD2L could be the consequence of a partial, rather than complete, loss-of-function. Subsequent studies by other groups have shown that anoctaminopathies 5 are a common cause of adult-onset LGMD. Our discovery of ANO5 mutations has shed light on a new class of proteins important for the muscle biology and opened new research avenues to study how defective membrane repair progresses into muscular dystrophies.