Recurrent <scp><i>FLNA</i> p.Gly1554Arg</scp> Variant Associated With Familial Ebstein Anomaly and Joint Stiffness

Ebstein anomaly is a rare congenital heart disease characterized by a downward tricuspid valve position, valvular dysplasia, and hypoplastic right ventricle. The genetic basis is unclear in most patients, although several genes are associated with Ebstein anomaly, such as NKX2-5, MYH7, TPM1, LAMA3, and FLNA (Benson et al. 1999; Mercer et al. 2017; Postma et al. 2011; Zhou et al. 2023). FLNA is an X-linked gene and encodes filamin A, actin-binding protein crosslinking actin filaments and organizing cytoskeleton, which is involved in cell migration, signaling and adhesion (Wade et al. 2020). FLNA deficiency causes congenital heart disease, cardiovascular disease, and periventricular nodular heterotopia. FLNA deficiency is generally found in heterozygous females with loss-of-function variants, and hemizygous males are thought to be prenatally lethal. Some specific FLNA variants are associated with distinct phenotypes such as isolated cardiac valvular dysplasia in males and females. The FLNA (p.Gly1554Arg) variant has been reported in one family with Ebstein anomaly (Mercer et al. 2017). Here, we report the second family with Ebstein anomaly associated with the FLNA (p.Gly1554Arg) variant in affected female and male family members. The patient was a 2-year-old male and the first child born to healthy nonconsanguineous parents. His maternal grandmother had Ebstein anomaly and finger joint stiffness, while his maternal uncle died at the age of 3 months due to cardiac valvular disease (Figure 1A). His mother did not have any symptoms and echocardiography was normal. He was delivered at a gestational age of 38 weeks 1 day via elective caesarean section due to breech position after an uneventful pregnancy, with a birth weight of 2840 g (−0.6 standard deviation [SD]), birth length of 46.0 cm (−1.6 SD), and an occipital-frontal head circumference (OFC) of 34.5 cm (+0.9 SD). Hearing screening was normal. Ebstein anomaly was noted including a lightly hypoplastic right ventricle, tricuspid valve dysplasia, mitral valve stenosis, and patent foramen ovale (Figure 1B). He also had strabismus, hyperopia, right cryptorchidism, bilateral talipes equinovarus, and arthrogryposis of the hip, knee, and fingers of the hand and foot. He was referred to our department at 1 year 2 months of age for the assessment of multiple congenital anomalies. At 1 year 7 months of age, his height, weight, and OFC were 81.7 cm (+0.1 SD), 9735 g (−0.8 SD), and 49.3 cm (+1.1 SD), respectively. His dysmorphic features included a wide forehead, deep set eyes, short palpebral fissures, infraorbital ridge, prominent nose and philtrum, wide face, retrognathia, camptodactyly, and malalignment of the right fourth and left fifth toes of the foot. His mother had similar facial gestalt to his. Motor development was delayed due to joint stiffness. He controlled his head at 3 months, rolled over at 8–9 months, sat up and crawled at 10 months, walked with support at 12 months, and walked alone at 15 months of age. He started speaking words at 13 months of age. Chromosomal microarray analysis was normal. The patient and family members were enrolled in an Institutional Review Board-approved study with written informed consent, which included permission to publish images, at Kanagawa Children's Medical Center. DNA was obtained from blood of the proband and his parents, and saliva from his grandmother. DNA libraries were enriched for sequencing using SureSelect XT (Agilent Technologies Inc., Santa Clara, CA). Samples were sequenced by NovaSeq (Illumina Inc., San Diego, CA) using 151-bp pair-end reads. The data were analyzed by the Burrows-Wheeler alignment tool and the Genome Analysis Toolkit pipeline (Broad Institute, Cambridge, MA). Of all the named variants within exons or at ±10 bp from exon–intron boundaries, those registered as having an allele frequency of > 1% in the Genome Aggregation Database (gnomAD), the Japanese Multi Omics Reference Panel (jMorp) genome variant database, and our in-house Japanese exome database (752 individuals) were removed. For de novo variants, variants absent in these control and in-house databases were called. An X-inactivation assay was performed according to previous reports (Allen et al. 1992; Machado et al. 2014). Briefly, 200 ng genomic DNA was digested by methylation-sensitive restrictive enzymes (HpaII and HhaI) at 37°C for 4 h. The target region, including the flanking HpaII and HhaI sites and a repeat element (GAAA repeat in the X-linked retinitis pigmentosa 2 promotor and CAG repeat in the human androgen-receptor gene), respectively, was amplified by PCR. Each PCR product was sized and quantified by fragment analysis. A ratio greater than 20:80 was defined as a skewed (nonrandom) X-inactivation pattern. The mean depth of coverage in the patient's exome sequencing was > 82.98 per base, and 98.5% of the coding region was covered by at least 10 reads. A hemizygous FLNA variant (NM_001110556.2:c.4660G>A (p.Gly1554Arg) NC_000023.11:g.154358294C>T) was identified by trio exome sequencing. The hemizygous variant was inherited from his heterozygous carrier mother and symptomatic grandmother, which was confirmed by Sanger sequencing after trio exome sequencing (Figure 1C,D). No other candidate variants were detected. The variant (rs1603360542) was not present in gnomAD and jMorp and was registered as a likely pathogenic variant in the ClinVar database. In silico analysis indicated that the Combined Annotation Dependent Depletion score was 26.2. The X-inactivation pattern was random in his grandmother (nearly 50:50) and tended to be skewed (20:80) in his mother. Here, we identified an FLNA p.G1554R variant with incomplete penetrance in familial Ebstein anomaly patients of three generations, that is, proband, maternal uncle, and maternal grandmother. The proband male presented with Ebstein anomaly and talipes equinovarus. The same variant has been reported in one family of Ebstein anomaly with joint stiffness (Mercer et al. 2017). FLNA loss-of-function variants are associated with various phenotypes of FLNA deficiency including periventricular nodular heterotopia, epilepsy, congenital heart disease (mainly septal defects), vascular disease, and Hirschsprung disease. Generally, loss-of-function variants of FLNA cause lethality in males. Also, gain-of-function variants lead to X-linked otopalatodigital spectrum disorders (OPD) including otopalatodigital syndrome, frontometaphyseal dysplasia, and Melnick-Needles syndrome. These X-linked skeletal dysplasia is characterized by cleft palate, digital abnormalities, limited joint mobility, and genitourinary abnormalities. The FLNA p.G1554R variant was found in males and females of the present and previous families, so it is likely to have a hypomorphic loss-of-function or gain-of-function effect. Valvular dysplasia was involved in FLNA deficiency. Digital abnormalities, joint stiffness, and cryptorchidism were involved in X-linked OPD. The phenotypic spectrum in this family is likely dependent on the X-inactivation pattern. His grandmother showed a random pattern, while his mother showed a mildly skewed pattern. Specific FLNA variants are associated with isolated polyvalvular dysplasia predominantly including the mitral valve. Myxomatous valvular dysplasia has been reported with FLNA variants (p.Pro637Gln, p.His743Pro, and p.Gly288Arg) in familial cases (Figure 1D) (Bernstein et al. 2011; Kyndt et al. 2007; Le Tourneau et al. 2018). In these cases of isolated polyvalvular dysplasia, the clinical phenotypes were different from those of classic FLNA deficiency since they did not have any other phenotypes. The two phenotypes do not overlap, and they are thought to be different disorders with different underlying mechanisms. Further, the p.G1554R variant is associated with other phenotypes of Ebstein anomaly predominantly including tricuspid valve dysplasia and joint stiffness. This report suggests that there is another manifestation of FLNA deficiency that presents with a distinct phenotype including polyvalvular dysplasia predominantly of the tricuspid valve resembling Ebstein anomaly and joint stiffness. Kenji Kurosawa designed the study. Takuya Naruto performed the bioinformatics experiments and contributed to the genetics experiments. Yukiko Kuroda, Takuya Wakamiya, Koki Nagai, and Yasuhiro Kawai recruited and evaluated the study subjects. Kenji Kurosawa supervised Yukiko Kuroda. Yukiko Kuroda wrote the manuscript. This study was supported by The Initiative on Rare and Undiagnosed Diseases (Grant Number JP24ek0109760) from the Japan Agency for Medical Research and Development, Ministry of Health, Labour and Welfare (MHLW) Grant (number 23FC1052), and JSPS KAKENHI Grant (number 23K14966 to YK). We thank the patient and his family for their cooperation. The patient and family members were enrolled in an Institutional Review Board-approved study with written informed consent at Kanagawa Children's Medical Center. The authors declare no conflicts of interest. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Figure S1. (A) Integrated Genome Viewer image of the filamin A (FLNA) variant detected by trio exome sequencing. Variant allele frequency (VAF) were indicated. (B) X-inactivation pattern in the human androgen-receptor (HUMURA) gene and retinitis pigmentosa 2 (RP2) promotor by a methylation-sensitive PCR-based assay with HpaII and HhaI digestion of genomic DNA from the proband’s mother and maternal grandmother. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.