The different clinical facets of SYN1-related neurodevelopmental disorders

Abstract

Synapsin-I (SYN1) is a presynaptic phosphoprotein crucial for synaptogenesis and synaptic plasticity. Pathogenic SYN1 variants are associated with variable X-linked neurodevelopmental disorders mainly affecting males. In this study, we expand on the clinical and molecular spectrum of the SYN1-related neurodevelopmental disorders by describing 31 novel individuals harboring 22 different SYN1 variants. We analyzed newly identified as well as previously reported individuals in order to define the frequency of key features associated with these disorders. Specifically, behavioral disturbances such as autism spectrum disorder or attention deficit hyperactivity disorder are observed in 91% of the individuals, epilepsy in 82%, intellectual disability in 77%, and developmental delay in 70%. Seizure types mainly include tonic-clonic or focal seizures with impaired awareness. The presence of reflex seizures is one of the most representative clinical manifestations related to SYN1. In more than half of the cases, seizures are triggered by contact with water, but other triggers are also frequently reported, including rubbing with a towel, fever, toothbrushing, fingernail clipping, falling asleep, and watching others showering or bathing. We additionally describe hyperpnea, emotion, lighting, using a stroboscope, digestive troubles, and defecation as possible triggers in individuals with SYN1 variants. The molecular spectrum of SYN1 variants is broad and encompasses truncating variants (frameshift, nonsense, splicing and start-loss variants) as well as non-truncating variants (missense substitutions and in-frame duplications). Genotype-phenotype correlation revealed that epileptic phenotypes are enriched in individuals with truncating variants. Furthermore, we could show for the first time that individuals with early seizures onset tend to present with severe-to-profound intellectual disability, hence highlighting the existence of an association between early seizure onset and more severe impairment of cognitive functions. Altogether, we present a detailed clinical description of the largest series of individuals with SYN1 variants reported so far and provide the first genotype-phenotype correlations for this gene. A timely molecular diagnosis and genetic counseling are cardinal for appropriate patient management and treatment.

Keywords: SYN1; autism spectrum disorders; genotype-phenotype correlation; neurodevelopmental disorders; reflex epilepsy; synapsins.

FIGURE 1

Schematic representation of the SYN1 gene and SYN1 protein with positions of the identified variants relative to exon and domains distribution. Variants identified in the present cohort are depicted in red, those reported in the literature in black, and those reported both in the literature and in our cohort in blue. The conserved A (N-terminal highly phosphorylated domain), B (linker), and C (functional domain) domains are depicted in yellow, orange, and green, respectively; the variable D, E, and F domains in blue, pink, and red. Classification of the variants was performed based on ACMG criteria. Missense substitutions are subdivided in likely pathogenic, VUS, and benign based on the presence of functional data and their frequency in the healthy population.

FIGURE 2

Comparison of the distribution of the SYN1 variants identified in our cohort and/or in the literature with the variants reported in gnomAD. The number of individuals with gnomAD variants was plotted for each combined codon. gnomAD variants were stratified by both sex and damage potential (based on variant type and CADD score). Specifically, gnomAD variants were subdivided in the following two groups: one comprising synonymous variants and non-synonymous variants with CADD scores lower than 22 and one comprising non-synonymous variants with CADD scores equal or higher than 22 (putatively damaging gnomAD variants). Numbers above bars represent the positions of residues affected by putatively damaging non-synonymous variants in at least two males or two females. Males are depleted in putatively damaging SYN1 variants compared to females (p = 7.3 × 10^–4, OR = 0.67, Fisher’s test). Regions A and B are almost devoid of putatively damaging variants in gnomAD.

FIGURE 3

Pedigrees of the SYN1 familial cases of our cohort. Black arrows indicate the index cases initially referred to the genetic center. Individuals within the same family tree are variably affected, pointing to the existence of incomplete penetrance and clinical heterogeneity.

FIGURE 4

Frequency of the main clinical features associated with SYN1 variants. Numbers inside the bars are ratios between the number of patients presenting with the specific clinical sign and the number of patients for which information on that specific feature was available. (A) Percentage of SYN1 patients in our cohort and/or in the literature manifesting the clinical features of interest. (B) Percentage of male and female SYN1 patients manifesting specific clinical features. Data from patients in our cohort and the literature were combined. (C) Percentage of patients with either truncating variants (TV) or non-truncating variants (NTV) manifesting specific clinical features. Data from patients in our cohort and the literature were combined. TV: nonsense, frameshift and splicing variants; NTV: missense substitutions and in-frame duplications. (D) Per clinical feature enrichment/depletion of TV versus NTV. Only significant p-values are shown inside bars (Fisher’s test followed by Bonferroni correction for multiple testing; Supplementary Table S4). DD, developmental delay; ID, intellectual disability; ASD, autism spectrum disorder; ADHD, attention deficit hyperactivity disorder.

FIGURE 5

Combined clinical features observed in patients with specific SYN1 variants. T: truncating variants (nonsense, frameshift and splicing variants); N: non-truncating variants (missense substitutions and in-frame duplications). Black squares indicate that a given clinical feature was described in association with the corresponding variant. The protein region has been subdivided in the four protein domains affected by the variants, namely domains A-D. No association can be detected between the position of the variant and the presence of a given clinical feature.

FIGURE 6

Epilepsy types and associations (A) Age of epilepsy onset according to the severity of ID. Patients were grouped into three categories of ID severity: borderline/mild (n = 14), moderate (n = 5) and severe/profound (n = 5). Each point represents one patient for which information was obtained. Box plot elements are defined as follows: center line: median; box limits: upper and lower quartiles; whiskers: 1.5× interquartile range; smaller point outside whiskers: outlier (B) Categorization of epilepsy types. Numbers inside the bars are shown as percentages and ratios for epilepsy patients. RE, reflex epilepsy. (C) Triggers reported in association with SYN1 variants sorted from the most to the less frequent.