Hypomorphic variants of SEL1L-HRD1 ER-associated degradation are associated with neurodevelopmental disorders

**Background:** The SEL1L-HRD1 complex is a key component of endoplasmic reticulum-associated degradation (ERAD), a quality-control mechanism that clears misfolded proteins from the ER. While mouse models have shown its vital importance in various physiological processes, no disease-associated variants in humans had been identified. This study aimed to identify and characterize human variants in SEL1L and HRD1 in patients with neurodevelopmental disorders.

**Methods:** Whole-exome sequencing (WES) was performed on six patients from three unrelated families (Saudi Arabia, Morocco, Italy) with similar neurodevelopmental phenotypes. Variants were filtered stringently and confirmed by Sanger sequencing. CRISPR/Cas9 knockin (KI) HEK293T cells carrying the biallelic variants were generated to study ERAD function. Protein stability, substrate accumulation, and interactions were assessed using Western blot, cycloheximide chase, immunoprecipitation, and ubiquitination assays. Structural modeling was performed using AlphaFold2.

**Key Results:** Three autosomal recessive variants were identified: SEL1L p.Gly585Asp (patient 1), SEL1L p.Met528Arg (patients 2–5), and HRD1 p.Pro398Leu (patient 6). All six patients presented with infantile-onset neurodevelopmental disorder, developmental delay, intellectual disability, and facial dysmorphisms. Four out of six had microcephaly. Two patients showed hypotonia, while four from the Moroccan family exhibited severe progressive ataxia. In KI HEK293T cells, all three variants caused accumulation of endogenous ERAD substrates (IRE1α, OS9, CD147) and increased high molecular weight aggregates of a model substrate proAVP(G57S), but to a lesser extent than complete knockout, indicating hypomorphic function. The SEL1L M528R variant reduced SEL1L and HRD1 protein levels by approximately 80% and 60%, respectively, due to decreased stability. The SEL1L G585D variant impaired substrate recruitment, reducing interactions with lectins ERLEC1 and OS9 by approximately 70–80% and with HRD1 by approximately 50%. The HRD1 P398L variant attenuated HRD1 ubiquitination without affecting interactions with other ERAD components. No overt unfolded protein response was observed in KI cells, though ER chaperones BiP and PDI accumulated.

**Clinical Implications:** This study establishes the first human disease associated with core ERAD components, termed ERAD-associated neurodevelopmental disorder with onset in infancy (ENDI). The hypomorphic nature of the variants suggests a threshold requirement for SEL1L-HRD1 ERAD function in normal neuronal development. The distinct mechanisms of dysfunction (substrate recruitment, complex stability, HRD1 activity) may explain phenotypic variations among patients. The findings highlight the diagnostic value of evaluating SEL1L-HRD1 ERAD in patients with intellectual disability, developmental delay, and ataxia. Currently, treatment options are limited, but this study provides a framework for future therapeutic targeting of this complex.