Genotype-Phenotype Correlations in Human Diseases Caused by Mutations of LINC Complex-Associated Genes: A Systematic Review and Meta-Summary

Abstract

Mutations in genes encoding proteins associated with the linker of nucleoskeleton and cytoskeleton (LINC) complex within the nuclear envelope cause different diseases with varying phenotypes including skeletal muscle, cardiac, metabolic, or nervous system pathologies. There is some understanding of the structure of LINC complex-associated proteins and how they interact, but it is unclear how mutations in genes encoding them can cause the same disease, and different diseases with different phenotypes. Here, published mutations in LINC complex-associated proteins were systematically reviewed and analyzed to ascertain whether patterns exist between the genetic sequence variants and clinical phenotypes. This revealed LMNA is the only LINC complex-associated gene in which mutations commonly cause distinct conditions, and there are no clear genotype-phenotype correlations. Clusters of LMNA variants causing striated muscle disease are located in exons 1 and 6, and metabolic disease-associated LMNA variants are frequently found in the tail of lamin A/C. Additionally, exon 6 of the emerin gene, EMD, may be a mutation "hot-spot", and diseases related to SYNE1, encoding nesprin-1, are most often caused by nonsense type mutations. These results provide insight into the diverse roles of LINC-complex proteins in human disease and provide direction for future gene-targeted therapy development.

Keywords: EMD; LINC complex; LMNA; SYNE1; emerin; lamin A/C; laminopathies; nesprin; nuclear envelope.

Figure 1

The Linker of Nucleoskeleton and Cytoskeleton (LINC) complex and associated human diseases. The Linker of Nucleoskeleton and Cytoskeleton (LINC) complex primarily consists of SUN1/SUN2 and nesprin proteins, that are located at the inner nuclear membrane (INM) and outer nuclear membrane (ONM), respectively. The respective proteins interact via their Sad1-UNC-84 and Klarsicht/Anc-1/Syne homology (KASH) domains in the lumen of the NE, creating a physical connection between the nuclear lamina (NL) and the cytoskeleton. The NL comprises of type-V intermediate filament proteins, the nuclear lamins. Lamins A/C are A-type nuclear lamins, whilst lamin B1/B2 are B-type lamins. Various other NE proteins interact with the LINC complex including integral INM protein emerin, TMEM43/LUMA and isoforms of four-and-a-half LIM domain protein 1 (FHL1). Mutations in the genes encoding these proteins have been associated with a range of diseases exhibiting different phenotypes. Abbreviations: Emery-Dreifuss muscular dystrophy (EDMD), cerebellar ataxia (SCAR8), dilated cardiomyopathy (DCM), limb-girdle muscular dystrophy (LGMD), familial partial lipodystrophy (FPLD), congenital muscular dystrophy (L-CMD), Charcot-Marie-Tooth disease (CMT), Hutchinson-Gilford progeria syndrome (HGPS), mandibuloacral dysplasia (MADA), atypical Werner syndrome (WS). Created with BioRender.com.

Figure 2

PRISMA flow diagram outlining the search and selection process of identifying records for the systematic review. The databases that were searched were Medline (n = 2048), CINAHPlus (n = 165) and AMED (n = 5), yielding a total of 2218 records. Duplicate records (n = 875) were removed prior to screening, then 1254 records were subjected to blind screening by two independent reviewers. 878 reports were excluded, and 464 records were retrieved. This left a total of 464 reports that were assessed for eligibility. There were 402 studies that were included in the review, whilst 62 were excluded. Reports were excluded for the following reasons: foreign language (n = 12), multiple mutations (n = 6), mutations were not specified, unclear or unrelated (n = 42) or contained a disease type that was listed within the exclusion criteria (n = 2). Created with BioRender.com.

Figure 3

LMNA mutations. (A) Phenotypes associated with LMNA mutations. There were 37 different conditions that were found to be associated with LMNA mutations, some which were caused by identical LMNA variants. These can be broadly classified into diseases affecting cardiac muscle, skeletal muscle, the metabolic system and the nervous system. A few variants caused diseases that fell into other categories. (B) Coding vs. intronic mutations. A total of 519 LMNA mutations were identified. 477 (91.91%) were coding mutations, whilst 42 (8.09%) were splice-site variants. (C) Coding region mutations. Missense mutations were the most frequently occurring type of LMNA mutations (342, 71.70%). Frameshift (59, 12.37%), nonsense (30, 6.29%), deletion (28, 5.87%), deletion/insertion (7, 1.47%), insertion (6, 1.26%) and duplication (5, 1.05%) mutations were also identified. (D) Mutations associated with different disease types and their frequency in LMNA exons. The most commonly occurring disease types arising from LMNA mutations were diseases affecting skeletal (262) and cardiac (260) muscle. These disease types clustered at exons 1 and 6. A total of 55 mutations were associated with metabolic disease, and these mostly appeared to increase in frequency from exon 7 onwards. Only 16 mutations caused nervous system disease and four caused other conditions which fell into different disease categories. Created with GraphPad Prism.

Figure 4

The structure of the LMNA gene and protein product lamin A. LMNA encodes lamin A/C. Lamin C is produced via a splice site located at intron 10 and therefore it’s C-terminus differs to lamin A. LMNA consists of 12 exons. Each coding region LMNA mutation was assigned to a disease category based on the disease type it caused. If it caused more than one type of disease, it was considered in multiple categories. These mutations were then marked below the LMNA gene at the appropriate codon, in a colour that corresponds to its related disease type(s). Above the LMNA gene is the structure of lamin A, consisting of a head, central rod and tail domain. Created with BioRender.com.

Figure 5

EMD mutations. (A) Coding vs. intronic mutations. A total of 83 EMD mutations were identified. 73 (86.75%) were coding region mutations and eleven (13.25%) were intronic variants. (B) Coding region mutations. The largest percentage of coding region mutations were frameshift (32, 44.44%), deletion (9, 12.50%) and nonsense (15, 20.83%) variants. Missense mutations (14, 19.44%) also accounted for a considerable proportion of the mutations identified. A couple of insertion (2, 2.78%) mutations were also identified. (C) Mutation frequency in EMD exons. Mutations most frequently occurred within exon 6, accounting for 45.21% (33) of total coding region mutations. Exon 2 contained the second highest proportion of mutations (17, 23.29%), followed by exons 3 (7, 9.59%) and 4 (6, 8.22%). Exon 1 and 5 contained the fewest EMD variants (5, 6.85% and 4, 5.48%, respectively). Created with GraphPad Prism.

Figure 6

SYNE1 mutations. (A) Coding vs. intronic mutations. There were 127 (90.07%) mutations identified in the coding region of SYNE1 and 14 (9.93%) were intronic. (B) Coding region mutations. Almost half (48.82%, 62) of coding region mutations were nonsense. Many mutations were also missense (32, 25.20%) and frameshift (30, 23.62%) variants. Three (2.36%) deletion mutations were also identified. (C) Diseases associated with SYNE1 variants. The most common disease resulting from SYNE1 variants was autosomal recessive spinocerebellar ataxia (SCAR8). 122 (85.82%) mutations were related to this disease. Amyotrophic lateral sclerosis was also found to be frequently caused by SYNE1 mutations (19, 13.48%). Seven cases of Emery-Dreifuss muscular dystrophy (EDMD) were related to SYNE1 variants, and a handful of mutations were associated with myopathic arthrogryposis (2), dilated cardiomyopathy (DCM) (5) and general muscle weakness (1). Created with GraphPad Prism.