Open Access

Thuy Duong Nguyen

• Laminopathies are a group of rare genetic disorders caused by mutations in genes encoding proteins of the nuclear lamina. One can distinguish between primary and secondary laminopathies. Primary laminopathies representing at least fourteen disease phenotypes arise through pleiotropic mutations in LMNA - the gene that codes for the A-type lamins A and C, mutations in LMNB1 encoding lamin B1 and mutations in LMNB2 encoding lamin B2. Secondary laminopathies including disease phenotypes also observed in primary laminopathies are caused by genes encoding proteins related to the nuclear lamina like ZMPSTE24 (FACE1), LAP2, LBR and thus reflecting genetic heterogeneity in laminopathies. The goal of the present investigation was to study pleiotropy and genetic heterogeneity in primary and secondary laminopathies by analysis of genotype/phenotype correlations. Emery-Dreifuss muscular dystrophy (EDMD), dilated cardiomyopathy with conduction disturbances (CMD1A), familial partial lipodystrophy (FPLD), mandibuloacral dysplasia (MAD), progeroid syndrome, atypical Werner syndrome (aWRN), restrictive dermopathy (RD) and Hallermann-Streiff syndrome (HSS) were included as disease phenotypes to look for their association with LMNA (primary laminopathies) and ZMPSTE24 (secondary laminopathies). Additionally, EDMD patients without STA or LMNA mutations were tested for ZMPSTE24 mutations. A functional candidate gene approach was applied using NARF and SREBF1 in patients suffering secondary laminopathies including FPLD, MAD, HGPS and RD, who were excluded from having LMNA and ZMPSTE24 mutations. Finally, practical consequences of the present study have been considered in genetic counseling and prevention of primary and secondary laminopathies. Screening for mutations in LMNA, ZMPSTE24 (FACE1), NARF and SREBF1 was carried out by PCR using intronic primers flanking each of the exons of the genes tested. The PCR products were tested for changes by heteroduplex analysis and directly sequenced by a cycle-sequencing procedure. Each DNA variation found was checked for its frequency in 386 chromosomes of an ethnically matched control population. For primary laminolathies, 249 unrelated individuals suffering EDMD, CMD1A (DCM), FPLD, MAD, HGPS, aWRN, RD, Hallermann-Streiff syndrome or only partially showing clinical features of the afore mentioned disease phenotypes were tested for LMNA mutations. Eighteen independent LMNA mutations were found in 249 unrelated patients resulting in a general detection rate of 7.2% Summary Dissertation 83 Among the 79 unrelated Caucasian patients and seven families suffering EDMD or EDMD-like disease phenotypes, 14 were found with LMNA mutations, including p.E33G, p.R249Q, p.L263P, p.R377H, p.M348I, p.R249W, p.R453W, p.R527P, p.L530R and p.R644C have been found, resulting in a detection rate of 17.7%. Of the ten different mutations, the three mutations p.L263P, p.M348I and p.L530R are novel. The other seven mutations have been reported before to be pathogenic. There is strong evidence that indicates the pathogenicity of the three novel mutations, p.L263P, p.M348I and p.L530R. Firstly, the mutations exchanged evolutionary highly conserved amino acids as shown by orthologous gene comparisons. Secondly, they were not found in 386 alleles of a reference population. Moreover, the mutations are located in the α- helical rod or globular domains of lamins A and C that might lead to the disruption of their nuclear function causing in skeletal and cardiac muscular malfunction. The LMNA p.M348I mutation was found in a Belgian male patient (G-13730) who also carried a STA c.1A>G, p.0 mutation. The STA mutation leading to a loss of emerin has previously shown to be causative for X-linked recessive EDMD and would explain the lack of emerin and a pathogenic effect found in the affected male by itself. But co-segregation of LMNA p.M348I with cardiac conduction disturbances in female family members showed an additional cardiac effect of this mutation to the pathology. This observation is one of the very rare pieces of evidence for digenic (oligo-allelic) pathogenesis in a neuromuscular disease phenotype of laminopathies. It points to related pathogenic mechanisms in EDMD and CMD1A that are not associated with STA and LMNA but with other so far unknown genes functionally related to the nuclear envelope. The known mutation p.R453W of the LMNA gene represents a mutational hot spot. So, it was not unexpectedly found in four unrelated EDMD patients of this study. Other recurrent mutations p.R249Q and p.R377H were found in two patients each. Variable phenotypic expression of the LMNA p.R644C mutation, ranging from no clinical signs to fully expressed EDMD was observed in an Austrian family in the present study. This mutation has reportedly been associated with strikingly diverse phenotypes in unrelated patients including left ventricular hypertrophy, limb girdle muscle weakness, CMD1A, FPLD or atypical lipodystrophy, neuropathy and atypical progeria. But the mechanism of pathogenesis is unknown. The apparent non-penetrance in relatives raises questions about the clinical significance of this missense mutation. However, the observations Summary Dissertation 84 in the present family and in those previously published provide evidence that the risk to express a laminopathy in close relatives is likely to be low but reasonable. Of the 49 unrelated Caucasian patients suffering CMD1A four mutations, p.E161K, c.- 3_+12del, p.Y259C and p.R377H, were found resulting in a detection rate of 8.2%, which did not significantly differ from the 2.5% found in 197 dilated cardiomyopathy patients of an earlier study. This overall low detection rate reflects the wide genetic and environmental heterogeneity of the pathogenesis in dilated cardiomyopathy. Otherwise, LMNA mutations may cause dilated cardiomyopathy in about 5% of the cases. The wide overlapping phenotypic and genetic similarities between Hallermann-Streiff syndrome (HSS) and HGPS, made HSS a good candidate disease for a primary laminopathy caused by LMNA mutations. But there was no co-segregating disease causing mutation identified. Thus, this study excluded HSS for the first time to be associated to LMNA and adds to the molecular genetic differentiation by excluding HSS from primary laminopathies. Among 32 individuals of 12 families suffering restrictive dermopathy, 22 individuals have been found to carry the ZMPSTE24 mutations c.50delA, c.209_210delAT, c.1085 - 1086insT or c.1385T>G. The mutation c.1085 -1086insT is a recurrent mutation that occurred in the present sample with a frequency of 68% in all RD patients with a ZMPSTE24 mutation. Three mutations, c.50delA, c.209_210delAT and c.1385T>G, are novel mutations. Like the c.1085 -1086insT mutation, c.209_210delAT and c.50delA lead to a frame shift, which putatively results in a non-functional truncated peptide. As an additional indication for a pathogenic effect, the novel mutations c.50delA and c.209_210delAT were not found in 386 alleles of a normal reference population. The first ZMPSTE24 missense mutation c.1385T>G (p.L462R) changing a highly conserved amino acid was found in patient from Guinea suffering from a clinically unequivocal case of restrictive dermopathy. The mutation was heterozygous in the patient but also in the healthy mother. A second pathogenic mutation should be expected. This hypothesis could not be proven, as there was no sufficient test material available from the patient and other family members. Moreover, there was no appropriate African (Guinea) reference population available, which could have been used to estimate the frequency of p.L462R. Thus, it cannot be excluded that p.L462R might be a polymorphism or rare non-pathogenic variant in the ethnic group the patient belongs to. Genetic instability in ZMPSTE24 has interfered with the molecular genetic diagnosis of restrictive dermopathy leading to the inability to distinguish between homozygotes and heterozygotes for the ZMPSTE24 mutation c.1085-1086insT. The reason is a repeated Summary Dissertation 85 thymine (T)9 c.1076-1085 in ZMPSTE24 that can cause a slippage of DNA polymerases. By sequencing cDNAs obtained from homozygous wild-type [(T)9], heterozygous [(T)9/(T)10] and homozygous mutant [(T)10] individuals by using regular Taq polymerase (Fermentas) or high fidelity polymerase (Pfu) for the sequencing reaction the genetic instability was quantified. High error rates up to 23% were found if regular Taq polymerase (Fermentas) was used for sequencing while using high fidelity polymerase (Pfu) resulted in error rates of 6.2 % or lower. As a practical consequence, high fidelity polymerase should always be used to distinguish homozygous mutant [(T)10] individuals from heterozygous [(T)9//(T)10] by sequencing. A high percentage of EDMD patients was tested negative for mutations in STA or LMNA (Bonne et al., 2003). Therefore, other genes are supposed to be involved in the molecular pathology of EDMD. ZMPSTE24 was considered as a promising functional candidate gene in this study, as the gene product - the ZMPSTE24 peptide - takes part in the post-translational modification of lamin A. The negative result of the present study points to a rather unlikely association of EDMD with ZMPSTE24. Additionally, NARF can very likely be excluded by this study from being associated with FPLD, MAD, HGPS and RD, while SREBF1 has obviously no association with FPLD. By the present study, diagnostic tools have been established for molecular genetic diagnosis of several very rare primary and secondary laminopathies, which has a direct practical impact on disease management of laminopathies. Now, the molecular definition of the diseases by association with a specific mutation can be used for genotype/phenotype correlation, predictive diagnosis and prenatal diagnosis.
• Laminopathies are a group of rare genetic disorders caused by mutations in genes encoding proteins of the nuclear lamina. One can distinguish between primary and secondary laminopathies. Primary laminopathies representing at least fourteen disease phenotypes arise through pleiotropic mutations in LMNA - the gene that codes for the A-type lamins A and C, mutations in LMNB1 encoding lamin B1 and mutations in LMNB2 encoding lamin B2. Secondary laminopathies including disease phenotypes also observed in primary laminopathies are caused by genes encoding proteins related to the nuclear lamina like ZMPSTE24 (FACE1), LAP2, LBR and thus reflecting genetic heterogeneity in laminopathies. The goal of the present investigation was to study pleiotropy and genetic heterogeneity in primary and secondary laminopathies by analysis of genotype/phenotype correlations. Emery-Dreifuss muscular dystrophy (EDMD), dilated cardiomyopathy with conduction disturbances (CMD1A), familial partial lipodystrophy (FPLD), mandibuloacral dysplasia (MAD), progeroid syndrome, atypical Werner syndrome (aWRN), restrictive dermopathy (RD) and Hallermann-Streiff syndrome (HSS) were included as disease phenotypes to look for their association with LMNA (primary laminopathies) and ZMPSTE24 (secondary laminopathies). Additionally, EDMD patients without STA or LMNA mutations were tested for ZMPSTE24 mutations. A functional candidate gene approach was applied using NARF and SREBF1 in patients suffering secondary laminopathies including FPLD, MAD, HGPS and RD, who were excluded from having LMNA and ZMPSTE24 mutations. Finally, practical consequences of the present study have been considered in genetic counseling and prevention of primary and secondary laminopathies. Screening for mutations in LMNA, ZMPSTE24 (FACE1), NARF and SREBF1 was carried out by PCR using intronic primers flanking each of the exons of the genes tested. The PCR products were tested for changes by heteroduplex analysis and directly sequenced by a cycle-sequencing procedure. Each DNA variation found was checked for its frequency in 386 chromosomes of an ethnically matched control population. For primary laminolathies, 249 unrelated individuals suffering EDMD, CMD1A (DCM), FPLD, MAD, HGPS, aWRN, RD, Hallermann-Streiff syndrome or only partially showing clinical features of the afore mentioned disease phenotypes were tested for LMNA mutations. Eighteen independent LMNA mutations were found in 249 unrelated patients resulting in a general detection rate of 7.2% Summary Dissertation 83 Among the 79 unrelated Caucasian patients and seven families suffering EDMD or EDMD-like disease phenotypes, 14 were found with LMNA mutations, including p.E33G, p.R249Q, p.L263P, p.R377H, p.M348I, p.R249W, p.R453W, p.R527P, p.L530R and p.R644C have been found, resulting in a detection rate of 17.7%. Of the ten different mutations, the three mutations p.L263P, p.M348I and p.L530R are novel. The other seven mutations have been reported before to be pathogenic. There is strong evidence that indicates the pathogenicity of the three novel mutations, p.L263P, p.M348I and p.L530R. Firstly, the mutations exchanged evolutionary highly conserved amino acids as shown by orthologous gene comparisons. Secondly, they were not found in 386 alleles of a reference population. Moreover, the mutations are located in the α- helical rod or globular domains of lamins A and C that might lead to the disruption of their nuclear function causing in skeletal and cardiac muscular malfunction. The LMNA p.M348I mutation was found in a Belgian male patient (G-13730) who also carried a STA c.1A>G, p.0 mutation. The STA mutation leading to a loss of emerin has previously shown to be causative for X-linked recessive EDMD and would explain the lack of emerin and a pathogenic effect found in the affected male by itself. But co-segregation of LMNA p.M348I with cardiac conduction disturbances in female family members showed an additional cardiac effect of this mutation to the pathology. This observation is one of the very rare pieces of evidence for digenic (oligo-allelic) pathogenesis in a neuromuscular disease phenotype of laminopathies. It points to related pathogenic mechanisms in EDMD and CMD1A that are not associated with STA and LMNA but with other so far unknown genes functionally related to the nuclear envelope. The known mutation p.R453W of the LMNA gene represents a mutational hot spot. So, it was not unexpectedly found in four unrelated EDMD patients of this study. Other recurrent mutations p.R249Q and p.R377H were found in two patients each. Variable phenotypic expression of the LMNA p.R644C mutation, ranging from no clinical signs to fully expressed EDMD was observed in an Austrian family in the present study. This mutation has reportedly been associated with strikingly diverse phenotypes in unrelated patients including left ventricular hypertrophy, limb girdle muscle weakness, CMD1A, FPLD or atypical lipodystrophy, neuropathy and atypical progeria. But the mechanism of pathogenesis is unknown. The apparent non-penetrance in relatives raises questions about the clinical significance of this missense mutation. However, the observations Summary Dissertation 84 in the present family and in those previously published provide evidence that the risk to express a laminopathy in close relatives is likely to be low but reasonable. Of the 49 unrelated Caucasian patients suffering CMD1A four mutations, p.E161K, c.- 3_+12del, p.Y259C and p.R377H, were found resulting in a detection rate of 8.2%, which did not significantly differ from the 2.5% found in 197 dilated cardiomyopathy patients of an earlier study. This overall low detection rate reflects the wide genetic and environmental heterogeneity of the pathogenesis in dilated cardiomyopathy. Otherwise, LMNA mutations may cause dilated cardiomyopathy in about 5% of the cases. The wide overlapping phenotypic and genetic similarities between Hallermann-Streiff syndrome (HSS) and HGPS, made HSS a good candidate disease for a primary laminopathy caused by LMNA mutations. But there was no co-segregating disease causing mutation identified. Thus, this study excluded HSS for the first time to be associated to LMNA and adds to the molecular genetic differentiation by excluding HSS from primary laminopathies. Among 32 individuals of 12 families suffering restrictive dermopathy, 22 individuals have been found to carry the ZMPSTE24 mutations c.50delA, c.209_210delAT, c.1085 - 1086insT or c.1385T>G. The mutation c.1085 -1086insT is a recurrent mutation that occurred in the present sample with a frequency of 68% in all RD patients with a ZMPSTE24 mutation. Three mutations, c.50delA, c.209_210delAT and c.1385T>G, are novel mutations. Like the c.1085 -1086insT mutation, c.209_210delAT and c.50delA lead to a frame shift, which putatively results in a non-functional truncated peptide. As an additional indication for a pathogenic effect, the novel mutations c.50delA and c.209_210delAT were not found in 386 alleles of a normal reference population. The first ZMPSTE24 missense mutation c.1385T>G (p.L462R) changing a highly conserved amino acid was found in patient from Guinea suffering from a clinically unequivocal case of restrictive dermopathy. The mutation was heterozygous in the patient but also in the healthy mother. A second pathogenic mutation should be expected. This hypothesis could not be proven, as there was no sufficient test material available from the patient and other family members. Moreover, there was no appropriate African (Guinea) reference population available, which could have been used to estimate the frequency of p.L462R. Thus, it cannot be excluded that p.L462R might be a polymorphism or rare non-pathogenic variant in the ethnic group the patient belongs to. Genetic instability in ZMPSTE24 has interfered with the molecular genetic diagnosis of restrictive dermopathy leading to the inability to distinguish between homozygotes and heterozygotes for the ZMPSTE24 mutation c.1085-1086insT. The reason is a repeated Summary Dissertation 85 thymine (T)9 c.1076-1085 in ZMPSTE24 that can cause a slippage of DNA polymerases. By sequencing cDNAs obtained from homozygous wild-type [(T)9], heterozygous [(T)9/(T)10] and homozygous mutant [(T)10] individuals by using regular Taq polymerase (Fermentas) or high fidelity polymerase (Pfu) for the sequencing reaction the genetic instability was quantified. High error rates up to 23% were found if regular Taq polymerase (Fermentas) was used for sequencing while using high fidelity polymerase (Pfu) resulted in error rates of 6.2 % or lower. As a practical consequence, high fidelity polymerase should always be used to distinguish homozygous mutant [(T)10] individuals from heterozygous [(T)9//(T)10] by sequencing. A high percentage of EDMD patients was tested negative for mutations in STA or LMNA (Bonne et al., 2003). Therefore, other genes are supposed to be involved in the molecular pathology of EDMD. ZMPSTE24 was considered as a promising functional candidate gene in this study, as the gene product - the ZMPSTE24 peptide - takes part in the post-translational modification of lamin A. The negative result of the present study points to a rather unlikely association of EDMD with ZMPSTE24. Additionally, NARF can very likely be excluded by this study from being associated with FPLD, MAD, HGPS and RD, while SREBF1 has obviously no association with FPLD. By the present study, diagnostic tools have been established for molecular genetic diagnosis of several very rare primary and secondary laminopathies, which has a direct practical impact on disease management of laminopathies. Now, the molecular definition of the diseases by association with a specific mutation can be used for genotype/phenotype correlation, predictive diagnosis and prenatal diagnosis.