Elena Daoutsali

Alexander disease (AxD) is a rare genetic leukodystrophy caused by dominant mutations in the gene coding for the astrocytic intermediate filament protein Glial Fibrillary Acidic Protein (GFAP). Understanding disease mechanisms is complicated by the fact that mutant GFAP promotes further GFAP accumulation, making it difficult to distinguish mutation-specific effects from those caused by elevated GFAP levels. Here, we performed mass spectrometry-based single cell proteomics to investigate proteomic alterations associated with increasing levels of wildtype GFAP and a common AxD-causing mutant GFAP R239C. By analyzing individual cells spanning a broad range of GFAP expression levels, we separated dosage-dependent effects from mutation-specific responses. Cells expressing high levels of wildtype GFAP increasingly resembled R239C-expressing cells at the proteomic level, supporting a role for GFAP accumulation in disease pathogenesis. In contrast, mutant GFAP specifically correlated with increased mitochondrial antioxidant enzymes and remodeling of the proteasome tailored toward ATP- and ubiquitin-independent degradation of oxidized proteins. Integration with proteomics data from patient-derived cortical organoids revealed increased abundance of nuclear-encoded mitochondrial proteins and respiratory chain assembly factors without corresponding changes in mitochondrially encoded proteins, suggesting impaired electron transport chain (ETC) assembly. Consistent with this, AxD cells displayed reduced oxygen consumption, increased glycolytic activity, enhanced utilization of the cytosolic part of the glycerol-3-phosphate (G3P) shuttle, and accumulation of triacylglycerides and lipid droplets. We propose that ETC dysfunction promotes partial uncoupling of the G3P shuttle, allowing sustained regeneration of cytosolic NAD⁺ to support glycolysis while diverting G3P toward lipid synthesis and storage. By regenerating NAD⁺ through the G3P shuttle rather than lactate fermentation, AxD astrocytes may sustain glycolytic flux while preserving lactate production, an important metabolic substrate for neighboring neurons. Together, our findings identify ETC dysfunction as a central feature of AxD and suggest that metabolic adaptation through the G3P shuttle links mitochondrial dysfunction to glycolytic reprogramming and lipid accumulation. More broadly, these results provide new insight into how astrocyte dysfunction drives molecular pathways implicated in neurodegeneration.

Antisense technology demonstrates significant potential for addressing inherited brain diseases, with over a dozen products already available and numerous others in the development pipeline. The versatility of differentiating induced pluripotent stem cells (iPSCs) into nearly all neural cell types proves invaluable for comprehending the mechanisms behind neurological diseases, replicating cellular phenotypes, and advancing the testing and development of new therapies, including antisense oligonucleotide therapeutics. While delivering antisense oligonucleotides (ASOs) to human iPSC-based neuronal models has posed challenges, this study explores various delivery methods, including lipid-based transfection, gymnotic uptake, Ca⁽²⁺⁾-enhanced medium (CEM)-based delivery, and electroporation, in 2D and 3D hiPSC-derived neuronal models. This study reveals that CEM-based delivery exhibits efficiency and low toxicity in both 2D neuronal cultures and 3D brain organoids. Furthermore, the findings indicate that CEM is slightly more effective in neurons than in astrocytes, suggesting promising avenues for further exploration and optimization of preclinical ASO strategies in the treatment of neurological disorders.

Background: Spinocerebellar ataxia type 1 (SCA1) is a neurodegenerative disease caused by a polyglutamine expansion in the ataxin‐1 protein resulting in neuropathology including mutant ataxin‐1 protein aggregation, aberrant neurodevelopment, and mitochondrial dysfunction.Objectives: Identify SCA1‐relevant phenotypes in patient‐specific fibroblasts and SCA1 induced pluripotent stem cells (iPSCs) neuronal cultures.Methods: SCA1 iPSCs were generated and differentiated into neuronal cultures. Protein aggregation and neuronal morphology were evaluated using fluorescent microscopy. Mitochondrial respiration was measured using the Seahorse Analyzer. The multi‐electrode array (MEA) was used to identify network activity. Finally, gene expression changes were studied using RNA‐seq to identify disease‐specific mechanisms.Results: Bioenergetics deficits in patient‐derived fibroblasts and SCA1 neuronal cultures showed altered oxygen consumption rate, suggesting involvement of mitochondrial dysfunction in SCA1. In SCA1 hiPSC‐derived neuronal cells, nuclear and cytoplasmic aggregates were identified similar in localization as aggregates in SCA1 postmortem brain tissue. SCA1 hiPSC‐derived neuronal cells showed reduced dendrite length and number of branching points while MEA recordings identified delayed development in network activity in SCA1 hiPSC‐derived neuronal cells. Transcriptome analysis identified 1050 differentially expressed genes in SCA1 hiPSC‐derived neuronal cells associated with synapse organization and neuron projection guidance, where a subgroup of 151 genes was highly associated with SCA1 phenotypes and linked to SCA1 relevant signaling pathways.Conclusions: Patient‐derived cells recapitulate key pathological features of SCA1 pathogenesis providing a valuable tool for the identification of novel disease‐specific processes. This model can be used for high throughput screenings to identify compounds, which may prevent or rescue neurodegeneration in this devastating disease. © 2023 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Introduction: ADutch-type cerebral amyloid angiopathy (D-CAA) is a hereditary brain disorder caused by a point mutation in the amyloid precursor protein (APP) gene. The mutation is located within the amyloid beta (Aβ) domain of APP and leads to Aβ peptide accumulation in and around the cerebral vasculature. There lack of disease models to study the cellular and molecular pathological mechanisms of D-CAA together with the absence of a disease phenotype in vitro in overexpression cell models, as well as the limited availability of D-CAA animal models indicates the need for a D-CAA patient-derived model.Methods: We generated cerebral organoids from four D-CAA patients and four controls, cultured them up to 110 days and performed immunofluorescent and targeted gene expression analyses at two time points (D52 and D110).Results: D-CAA cerebral organoids exhibited Aβ accumulations, showed enhanced neuronal and astrocytic gene expression and TGFβ pathway de-regulation.Conclusions: These results illustrate the potential of cerebral organoids as in vitro disease model of D-CAA that can be used to understand disease mechanisms of D-CAA and can serve as therapeutic intervention platform for various Aβ-related disorders.

Dutch-type cerebral amyloid angiopathy (D-CAA) is a monogenic form of cerebral amyloid angiopathy and is inherited in an autosomal dominant manner. The disease is caused by a point mutation in exon 17 of the amyloid precursor protein ( APP ) gene that leads to an amino acid substitution at codon 693. The mutation is located within the amyloid beta (Aβ) domain of APP, and leads to accumulation of toxic Aβ peptide in and around the cerebral vasculature. We have designed an antisense oligonucleotide (AON) approach that results in skipping of exon 17, generating a shorter APP isoform that lacks part of the Aβ domain and the D-CAA mutation. We demonstrate efficient AON-induced skipping of exon 17 at RNA level and the occurrence of a shorter APP protein isoform in three different cell types. This resulted in a reduction of Aβ40 in neuronally differentiated, patient-derived induced pluripotent stem cells. AON-treated wild-type mice showed successful exon skipping on RNA and protein levels throughout the brain. These results illustrate APP splice modulation as a promising therapeutic approach for D-CAA.

For most protein-coding genes, the genetic code is dispersed over a number of exons, which are interspersed by longer non-coding sequences, the introns. Translation of the messenger RNA to protein requires removal of the introns and ligation of the exons, a procedure called precursor mRNA (pre-mRNA) splicing. Soon after the discovery of splicing, it was found that defects in pre-mRNA splicing are a common underlying factor of genetic and acquired diseases. Despite the complexity of splicing, it can be intentionally modulated in different ways. This chapter gives an overview of antisense oligonucleotide-mediated splicing modulation and its progress towards therapeutics. It outlines how this approach can be exploited to restore cryptic splicing, restore open reading frames, establish exon inclusion, alter the ratio of alternatively spliced transcripts and induce the production of less toxic or non-toxic proteins. For each application, specific diseases for which antisense oligonucleotide-mediated splicing modulation has been tested are used to illustrate the approach.