Submission Details

NRAP was evaluated for autosomal recessive dilated cardiomyopathy (DCM). In addition to DCM, there were publications of one case of HCM with homozygous missense NRAP variants (Maurer et al. 2023 PMID: 36672924) and one case of myofibrillar myopathy with three NRAP missense variants identified in the presence of additional clinically relevant variants (D’Avlia et al. 2016 PMID: 27443559). This was considered in accordance with ClinGen lumping and splitting criteria. Due to distinct differences in mechanism and phenotypic variability, the curation was split and curated for autosomal recessive, idiopathic DCM. Human genetic evidence supporting this gene-disease relationship includes case level and case-control data. At least ten rare variants (nonsense, frameshift) have been reported in humans with DCM. A total of 17 scorable cases were reviewed from 8 primary publications (Names et al., 2023 DOI:10.24911/JBCGenetics/183-1668575222; Zhang et al., 2023 PMID 36815016; Maurer et al., 2023 PMID: 36672924; Lesurf et al., 2022 PMID: 35288587; Al-Hassnan et al., 2023 PMID: 32870709; Ahmed et al., 2019 DOI: 10.24911/JBCGenetics/183-1542267981; Vasilescu et al., 2018 PMID: 30384889; Truszowska et al., 2019 PMID: 28611399). The cases reviewed were from a series of consanguineous and non-consanguineous families; however, all NRAP cases reviewed were identified to be homozygous for a rare truncating variants with 7 of the primary publications utilizing whole exome sequencing and Lesurf et al. utilizing whole genome sequencing (Lesurf et al., 2022 PMID: 35288587). While ancestry information was limited in many of the publications reviewed, 6 of the cases reviewed were homozygous for the NRAP variant p.C134Sfs12 (Maurer et al., 2023 PMID: 36672924; Al-Hassnan et al., 2023 PMID: 32870709; Ahmed et al., 2019) and 3 cases reviewed were homozygous for the NRAP variant p.Glu1190 (Names et al., 2023) suggestive of possible founder variants. In addition to case level data, one case-control study was reviewed (Koskenvuo et al., 2021 PMID: 33534821). Here, aggregate variant analysis was completed through investigation of a total of 31,639 patients recruited in Finland who completed genetic testing via WES or a 4,600 gene high-quality next generation sequencing assay through a clinical laboratory. Of this cohort, 577 patients had suspected or confirmed diagnoses of DCM, 5150 had non-DCM cardiac indications, and 25,912 were referred for non-cardiac indications and served as controls. Case-control analysis showed enrichment of the presence of two NRAP variants in 1.9% of DCM cases compared to controls and non-DCM cases (OR = 1052 (62-17876), p < 0.0001). While Finland is a known founder population, data available in FinnGen, a publicly available, large database of 500,000 Finnish biobank participants, supports rarity of NRAP variants and enrichment in cardiomyopathy cases (https://r11.finngen.fi/gene/NRAP). Via FinnGen browser, 33 NRAP variants are reported. Loss-of-function variants in NRAP are significantly associated with cardiomyopathy phenotypes in FinnGen (p=1.3e-7). Additional evidence is available in the literature, but the maximum score for genetic evidence has been reached.

In addition, this gene-disease assertion is supported by expression and protein-interaction data. Following the identification of human NRAP from Luo et al. 1997, northern blot analysis in human and mouse tissues identified NRAP expression in heart tissue (Mohiddin et al., 2003 PMID: 12789664). Protein-interaction data in additional publications supports NRAP interaction with previously DCM-associated proteins vinculin (VCL) and filamin-c (FLNC). In Luo et al., 1999, cDNA fragments from mouse skeletal muscle coding for predefined regions of NRAP were flagged using an N-terminal histidine tag and expressed via E coli host cells to screen for interaction with various candidate proteins. Using enzyme-linked immunosorbent (ELISA) assays, NRAP super repeat regions showed significant binding affinity to the C-terminal of glutathione-S-transferase (GST)-tagged VCL (Luo et al., 1999 PMID: 10320340). In Lu et al., 2003, a yeast two-hybrid model was utilized to screen mouse skeletal muscle DNA for proteins capable of binding to NRAP in eukaryotic cells. After a series of extensive steps to develop double transformants harboring either NRAP-bait plasmid or negative controls, end-sequencing of positive clones exhibiting activation of both reporter genes identified FLNC as an interacting protein with NRAP (Lu et al., 2003 PMID: 12692149). Additional experiments via gel-overlay assays and NI-NTA bead assays further identified FLNC binds to NRAP super repeat regions (Lu et al., 2003 PMID: 12692149).

Limitations in available experimental evidence due to lack of animal models recapitulating a DCM phenotype and lack of functional data was considered during gene-disease relationship review. While NRAP literature reviewed shows replication of publications over time, these limitations were considered for final evidence summary.

In summary, there is strong evidence to support the relationship between NRAP and autosomal recessive DCM. This classification was approved by the ClinGen Dilated Cardiomyopathy Working Group on 7/26/2024 (SOP Version 10).