New Insights on the Genetic Basis Underlying SHILCA Syndrome: Characterization of the NMNAT1 Pathological Alterations Due to Compound Heterozygous Mutations and Identification of a Novel Alternative Isoform

Abstract

This study aims to genetically characterize a two-year-old patient suffering from multiple systemic abnormalities, including skeletal, nervous and developmental involvements and Leber congenital amaurosis (LCA). Genetic screening by next-generation sequencing identified two heterozygous pathogenic variants in nicotinamide mononucleotide adenylyltransferase 1 (NMNAT1) as the molecular cause of the disease: c.439+5G>T and c.299+526_*968dup.This splice variant has never been reported to date, whereas pathogenic duplication has recently been associated with cases displaying an autosomal recessive disorder that includes a severe form of spondylo-epiphyseal dysplasia, sensorineural hearing loss, intellectual disability and LCA (SHILCA), as well as some brain anomalies. Our patient presented clinical manifestations which correlated strongly with this reported syndrome. To further study the possible transcriptional alterations resulting from these mutations, mRNA expression assays were performed in the patient and her father. The obtained results detected aberrant alternative transcripts and unbalanced levels of expression, consistent with severe systemic involvement. Moreover, these analyses also detected a novel NMNAT1 isoform, which is variably expressed in healthy human tissues. Altogether, these findings represent new evidence of the correlation of NMNAT1 and SHILCA syndrome, and provide additional insights into the healthy and pathogenic expression of this gene.

Keywords: Leber congenital amaurosis (LCA); developmental delay; hypomyelination; inherited retinal diseases; macular coloboma; nicotinamide mononucleotide adenylyltransferase 1 (NMNAT1); sensorineural hearing loss; spondylo-epiphyseal dysplasia; spondylo-epiphyseal dysplasia, sensorineural hearing loss, intellectual disability and Leber congenital amaurosis (SHILCA); transcriptional alteration.

Figure 1

Clinical images of the patient. Frontal (A) and lateral (B) photographs of the patient’s head show facial appearances with mild coarsening and a deep nasal bridge. Retinographies of the left (C) and right (D) eyes present macular colobomatous involvement due to a retinal degeneration. Upper images were obtained at the age of six months whereas lower images were taken at age two years. Cerebral magnetic resonance imaging (MRI) (E–G) show hypomyelinating leukoencephalopathy, and progressive brain and cerebellum atrophy. Left images were obtained at the age of six months, whereas right images were obtained at age two years. Lateral spine MRI image and radiograph (H,I, respectively) present small epiphyseal changes due to a spondylo-epiphyseal dysplasia at six months of age. Skeletal radiographs (J,K) show severe ossification delay when the patient was 20 months old.

Figure 2

Genetic screenings allowed the discovery of the molecular cause of disease in the patient. Pedigree from family Fi21/01 (A) show the identified nicotinamide mononucleotide adenylyltransferase 1 (NMNAT1) pathogenic mutations M1: c.439+5G>T and M2: c.299+526_968*dup. The mutant allelic combination, analyzed by whole exome sequencing, is shown in the patient (individual III.4) and in the father (individual II.4, of Bulgarian origin). The confirmation of M1 was obtained by Sanger sequencing (B). The detection of the genetic rearrangement M2, corresponding to the duplication of exons 4 and 5 of NMNAT1, was detected by analyzing the BAM file from the whole genome sequencing of the patient (C). The upper scheme shows the genetic localization within the gene (exons 2 to 5, numbered), whereas the reading coverage is depicted in the lower graph. The red arrows show the region of the duplication, from the beginning of intron 3 to the middle of the 3′ UTR.

Figure 3

Study of the nicotinamide mononucleotide adenylyltransferase 1 (NMNAT1) expression levels by real-time reverse transcriptase (RT)-PCR was performed in the patient and her father, and compared with those of three healthy control subjects. The first graph (A) shows slightly higher levels of total NMNAT1 mRNA in the father and lower levels in the patient. The second graph (B) represents the level of two concrete NMNAT1 transcript variants: the alternative isoform (transcript variant NM_001297779.1, left bars) and a newly identified isoform with the intron 4 retention (right bars). Only the patient showed decreased expression of the alternative isoform, whereas both tested family members presented dramatically higher levels of expression of the new isoform. In all cases, control error bars correspond to the standard deviation between the three biological replicates. The concrete amplicon localization is shown under the graph in each condition. Arrows represent forward and reverse primers, whereas the black boxes are the concrete target exons.

Figure 4

Transcriptional study of the different nicotinamide mononucleotide adenylyltransferase 1 (NMNAT1) isoforms in human tissues. PCR amplification with specific primers (Table S3) using a panel of human tissues (A) detected variable ubiquitous expressions of three different isoforms: canonical or 1 (transcript variants NM_022787.3 and NM_001297778.1, upper panel), alternative or 2 (transcript variant NM_001297779.1, lower panel, bottom band) and a newly described isoform 3 (lower panel, top band). ret: retina. I1: liver. I2: skeletal muscle. I3: kidney. I4: pancreas. I5: heart. I6: brain. I7: placenta. I8: lung. II1: ovary. II2: small intestine. II3: colon. II4: leukocyte. II5: spleen. II6: thymus. II7: prostate. II8: testis. In order to quantify these expressions, real-time reverse transcriptase (RT)-PCR assays were performed for some of these human tissues (B). Levels of expression of isoform 1 (white bars) were higher in all tested tissues, followed by isoform 2 (gray bars). The novel isoform 3 presented lower levels of expression, although it was always detected. (C)The sequence of coding exons for each isoform were created using GeneDrawer software (Insilicase). Primer pairs used for PCR amplification (seen in A) are represented with black arrows in each isoform. (D)The alignment of the resulting protein sequences (in order: isoforms 1, 2 and 3, as indicated within the figure), using Jalview software [21], showed an identical sequence along the first 146 amino acids and some conservation of the last residues (lower yellow line).