Virginie Vignard

We report an autosomal recessive neurodevelopmental disorder exclusively affecting females carrying biallelic loss-of-function variants in CIZ1 , a gene encoding a nuclear matrix protein essential for the maintenance of X-chromosome inactivation. Eight unrelated affected females were identified, whereas male siblings with biallelic variants were asymptomatic. We showed that loss of CIZ1 compromises the maintenance of X-chromosome inactivation, leading to an abnormal overexpression of a subset of X-linked genes.

Poly(rC)-binding protein 1 (PCBP1), a splicing factor and key member of the hnRNP E family, was initially characterized for its tumor suppressive properties. More recently, its role in regulating gene expression in the brain and nervous system has attracted growing interest. Through an international multicenter collaboration, we identified 13 de novo pathogenic variants in PCBP1 across 13 unrelated families. All affected individuals exhibited intellectual disability (ID), with autism spectrum disorder (ASD) as a prominent feature. Functional analysis in primary hippocampal mouse neuron cultures demonstrated that PCBP1 variants impair dendritic arborization, underscoring their deleterious effects. Transcriptomic profiling by RNA sequencing of subjects-derived T cells showed a distinctive signature characterized by significantly increased exon skipping. These results highlight the essential role of PCBP1 in neurogenesis and neuritogenesis, revealing the impact of loss-of-function variants in cells harboring PCBP1 pathogenic variants, thereby confirming the link between splicing defects and neurodevelopmental disorders. Collectively, our findings demonstrate the critical role of PCBP1 in neurodevelopment, reaffirming the importance of splicing regulation in mammalian neurodevelopment.

Objective: De novo variants in cullin‐3 ubiquitin ligase (CUL3) have been strongly associated with neurodevelopmental disorders (NDDs), but no large case series have been reported so far. Here, we aimed to collect sporadic cases carrying rare variants in CUL3, describe the genotype–phenotype correlation, and investigate the underlying pathogenic mechanism.Methods: Genetic data and detailed clinical records were collected via multicenter collaboration. Dysmorphic facial features were analyzed using GestaltMatcher. Variant effects on CUL3 protein stability were assessed using patient‐derived T‐cells.Results: We assembled a cohort of 37 individuals with heterozygous CUL3 variants presenting a syndromic NDD characterized by intellectual disability with or without autistic features. Of these, 35 have loss‐of‐function (LoF) and 2 have missense variants. CUL3 LoF variants in patients may affect protein stability leading to perturbations in protein homeostasis, as evidenced by decreased ubiquitin‐protein conjugates in vitro. Notably, we show that 4E‐BP1 (EIF4EBP1), a prominent substrate of CUL3, fails to be targeted for proteasomal degradation in patient‐derived cells.Interpretation: Our study further refines the clinical and mutational spectrum of CUL3‐associated NDDs, expands the spectrum of cullin RING E3 ligase‐associated neuropsychiatric disorders, and suggests haploinsufficiency via LoF variants is the predominant pathogenic mechanism. ANN NEUROL 2025;97:76–89

Neurodevelopmental proteasomopathies represent a distinctive category of neurodevelopmental disorders (NDD) characterized by genetic variations within the 26S proteasome, a protein complex governing eukaryotic cellular protein homeostasis. In our comprehensive study, we identified 23 unique variants in PSMC5 , which encodes the AAA-ATPase proteasome subunit PSMC5/Rpt6, causing syndromic NDD in 38 unrelated individuals. Overexpression of PSMC5 variants altered human hippocampal neuron morphology, while PSMC5 knockdown led to impaired reversal learning in flies and loss of excitatory synapses in rat hippocampal neurons. PSMC5 loss-of-function resulted in abnormal protein aggregation, profoundly impacting innate immune signaling, mitophagy rates, and lipid metabolism in affected individuals. Importantly, targeting key components of the integrated stress response, such as PKR and GCN2 kinases, ameliorated immune dysregulations in cells from affected individuals. These findings significantly advance our understanding of the molecular mechanisms underlying neurodevelopmental proteasomopathies, provide links to research in neurodegenerative diseases, and open up potential therapeutic avenues.

A critical step in preserving protein homeostasis by the ubiquitin-proteasome system (UPS) is the recognition, binding, unfolding, and translocation of protein substrates by AAA-ATPase proteasome subunits for degradation by 26S proteasomes. Here, we identified fourteen different de novo missense variants in the PSMC3 gene encoding the AAA-ATPase proteasome subunit Rpt5 in twenty-two unrelated heterozygous subjects with an autosomal dominant form of neurodevelopmental delay and intellectual disability. Indeed, depletion of PSMC3 impaired reversal learning capabilities in a Drosophila model. The PSMC3 variants cause proteasome dysfunction in patient-derived cells by disruption of substrate translocation, proteotoxic stress and proteostatic imbalances, as well as alterations in proteins controlling developmental and innate immune programs. Molecular analysis confirmed the induction of cellular stress responses and dysregulated mitophagy along with an elevated type I interferon (IFN) signature. Our data define PSMC3 variants as the genetic cause of proteotoxic stress alerting the innate immune system to mount a type I IFN response and link neurodevelopmental syndromes to interferonopathies.

The melanoma antigens MELOE-1 and MELOE-2 are encoded by a messenger, called meloe, overexpressed in melanomas compared with other tumour cell types and healthy tissues. They are both able to elicit melanoma-specific T cell responses in melanoma patients, and MELOE-1-specific CD8 T cells have been involved in melanoma immunosurveillance. With the aim to develop immunotherapies targeting this antigen, we investigated the transcriptional mechanisms leading to the preferential expression of meloe messenger in the melanocytic lineage. We defined the minimal promoter region of meloe gene and identified binding motifs for a set of transcription factors. Using mutagenesis, co-transfection experiments and chromatin immunoprecipitation, we showed that transcription factors involved in meloe promoter activity in melanomas were the melanocytic specific SOX9 and SOX10 proteins together with the activated P-CREB protein. Furthermore, we showed that meloe promoter was hypomethylated in melanomas and melanocytes, and hypermethylated in colon cancer cell lines and mesotheliomas, thus explaining the absence of P-CREB binding in these cell lines. This was a second key to explain the overerexpression of meloe messenger in the melanocytic lineage. To our knowledge, such a dual transcriptional control conferring tissue-specificity has never been described for the expression of tumour antigens.

Our previous studies on melanoma antigens identified two new polypeptides, named MELOE-1 and MELOE-2, that are involved in immunosurveillance. Intriguingly, these antigens are coded by distinct open reading frames (ORF) of the meloe mRNA which is significantly expressed only in the melanocytic lineage. In addition, MELOE-1 and -2 specific T cell clones recognized melanoma cells but very poorly normal melanocytes suggesting differential translation of meloe in normal vs tumor cells. This prompted us to elucidate the mechanisms of translation of these antigens in melanoma cells. We first demonstrated that no splicing event or cryptic promoter could generate shorter meloe transcripts containing only one of the two ORFs. Triggering meloe RNA degradation with a siRNA close to the ORF coding for MELOE-2 abrogated expression of both MELOE-1 and MELOE-2, thus confirming that the two ORFs are always associated. Next we showed, in a bicistronic reporter system, that IRES activities could be detected upstream of MELOE-1 and MELOE-2 and finally confirmed their translation from full length meloe cDNA in melanoma cells with eGFP constructs. In conclusion, meloe is a polycistronic mRNA that generates both MELOE-1 and MELOE-2 antigens through IRES-dependent translation in melanoma cells and that may explain their tumor specificity.

MELOE-1 is an overexpressed melanoma antigen containing a HLA-A2 restricted epitope, involved in melanoma immunosurveillance of patients adoptively transferred with tumour infiltrating lymphocytes (TIL). The use of the full-length antigen (46 aa) for anti-melanoma vaccination could be considered, subject to the presence of Th epitopes all along MELOE-1 sequence. Thus, in this study we evaluated in vitro the immunoprevalence of the different regions of MELOE-1 (i.e. their ability to induce CD4 T cell responses in vitro from PBMC). Stimulation of PBMC from healthy subjects with MELOE-1 induced the amplification of CD4 T cells specific for various regions of the protein in multiple HLA contexts, for each tested donor. We confirmed these results in a panel of melanoma patients, and documented that MELOE-1 specific CD4 T cells, were mainly Th1 cells, presumably favourable to the amplification of CD8 specific T cells. Using autologous DC, we further showed that these class II epitopes could be naturally processed from MELOE-1 whole protein and identified minimal epitopes derived from each region of MELOE-1, and presented in four distinct HLA contexts. In conclusion, vaccination with MELOE-1 whole polypeptide should induce specific Th1 CD4 responses in a majority of melanoma patients, stimulating the amplification of CD8 effector cells, reactive against melanoma cells.