Severe Phenotype of Cutis Laxa Type 1B with Antenatal Signs due to a Novel Homozygous Nonsense Mutation in EFEMP2

Abstract

EFEMP2 mutations are known to be responsible for autosomal recessive cutis laxa type 1B (ARCL1B), a rare multisystem disease affecting skin, skeleton, and vascular structures. We report 2 additional related cases of ARCL1B of particular severity leading to termination of pregnancy. Cardinal signs of this connective tissue disease were already seen during the second trimester of pregnancy, then confirmed and clarified at autopsy. Anomalies included cutis laxa, arachnodactyly, clubfoot, wormian bones, moderate bowing of long bones with slender bone trabeculae, rib fractures, undermuscularized diaphragm, hiatal hernia, and arterial tortuosity with thick vascular walls and disorganized elastic fibers. Sequencing of the EFEMP2 gene revealed a novel homozygous nonsense mutation: c.639C>A (p.Cys213*). We performed a thorough histological analysis and discuss differential diagnoses, genotype-phenotype correlations, and the challenge of prenatal diagnosis of this disease.

Severe congenital cutis laxa: Identification of novel homozygous LOX gene variants in two families

“…First, both variants are located in the catalytic domain, absent in control databases and predicted deleterious by all in silico prediction algorithms including Polyphen2, SIFT, Provean and Mutation Taster. Second, Lysyl oxidase directly interacts with Fibulin‐4 ( FBLN4 ), the protein deficient in ARCL1B, with similar clinical effects on the vasculature, skin (cutis laxa), skeleton and respiratory system (including diaphragmatic abnormalities) 9,10,13,14 . Third, several LOX knockout mouse models ( Lox −/− ) have been generated that presented with a very similar clinical phenotype to that seen in our patients 5,6,15 .…”

“…Second, Lysyl oxidase directly interacts with Fibulin-4 (FBLN4), the protein deficient in ARCL1B, with similar clinical effects on the vasculature, skin (cutis laxa), skeleton and respiratory system (including diaphragmatic abnormalities). 9,10,13,14 Third, several LOX knockout mouse models (Lox À/À ) have been generated that presented with a very similar clinical phenotype to that seen in our patients. 5,6,15 Fourth, the ultrastructural findings can be mice was <20% in skin fibroblasts and aortic smooth muscle cells compared to Lox wt/wt mice.…”

“…Table 1 compares the phenotypic features of the cases presented here and the three known subtypes of ARCL type 1. The associated vascular dilatation and tortuosity, along with bone fragility, resembles most closely the FBLN4 ‐associated subtype (ARCL1B, OMIM #614437), 1,3,9‐11 while the pulmonary hypoplasia resulting in early lethality indicates overlap with ARCL1A and ARCL1C due to FBLN5 (OMIM #604850) and LTBP4 (OMIM #604710) variants respectively. There was no obvious emphysema present in Patients 1 and 2, which is similar to ARCL1B (Unpublished data, B. Callewaert, MD, PhD & A. Beyens, MD, 2020), although X‐ray features of emphysema were present in Patient 3.…”

“…Table 1 compares the phenotypic features of the cases presented here and the three known subtypes of ARCL type 1. The associated vascular dilatation and tortuosity, along with bone fragility, resembles most closely the FBLN4-associated subtype (ARCL1B, OMIM #614437), 1,3,[9][10][11] while the pulmonary hypoplasia resulting in early lethality indicates overlap with ARCL1A and ARCL1C due to no relationship to the rate of fracture. 12 Several lines of evidence support the pathogenicity of the LOX variants in the patients' phenotype.…”

Clinical and molecular characterization of an 18‐month‐old infant with autosomal recessive cutis laxa type 1C due to a novel LTBP4 pathogenic variant, and literature review

“…Given the phenotypic overlap among the different forms, proper differential diagnosis might be challenging as well as predicting the causative gene according to the clinical picture of patients, mainly due to the lack of detailed comparative phenotype data (Berk et al, 2012;Gardeitchik & Morava, 2013;. ARCL1 type 1 is subdivided in ARCL1A (MIM #219100), which is caused by pathogenic variants in the fibulin-5 gene (FBLN5, MIM *604,580) (Elahi et al, 2006;Hu et al, 2006;Loeys et al, 2002;Tekedereli et al, 2019), ARCL1B (MIM #614437) caused by mutations in the EGFcontaining fibulin-like extracellular matrix protein 2 gene or fibulin-4 gene (EFEMP2 or FBLN4; MIM *604,633) (Dasouki et al, 2007;Hucthagowder et al, 2006;Letard et al, 2018), and ARCL1C (MIM #613177), a.k.a. Urban-Rifkin-Davis syndrome, which is due to biallelic variants in the latent transforming growth factor-beta binding protein 4 gene (LTBP4, MIM *604,710) (Callewaert et al, 2013;Su et al, 2015;Urban et al, 2009).…”