Main text

Cohen syndrome (CS) is a rare congenital, neurodevelopmental disorder that manifests clinical features such as obesity, hypotonia, intellectual disabilities, microcephaly, and distinctive facial features [1]. Many people with CS have autosomal recessive mutations in the VPS13B gene (also known as COH1), which harbors up to 62 exons, and an open reading frame encoding 4022 amino acids and is generally involved in vesicle-mediated trafficking and intracellular protein transport [2]. While VPS13B mutant CS mouse models exhibited specific clinical symptoms in CS [3], there is a growing need for pathophysiological studies in human cells derived from CS patient-specific induced pluripotent stem cells (iPSCs). Human skin fibroblasts from CS patients, carrying frameshift or nonsense mutations in VPS13B, display a fragmented Golgi complex that leads to disruption of Golgi integrity [4].

To uncover pathological phenotypes at the ultrastructural level in CS patients with VPS13B novel mutations, we performed electron microscopic (EM) analysis on control and CS fibroblasts [5]. There were no gross defects regarding the morphology of the nuclei, ER, or mitochondria (including the Golgi-complex at the ultrastructural level) in CS fibroblasts, however, autophagic vacuoles (AVs), which are a hallmark of autophagy, accumulated tremendously in the cytosol of CS fibroblasts when compared to control fibroblasts (Fig. 1a). The number of early autophagosomes or autolysosomes was significantly increased in CS fibroblasts compared to control fibroblasts (Fig. 1b). To further examine autophagic activity in control and CS fibroblasts, autophagic flux assays were performed in the presence or absence of lysosomal inhibitors (100 nM chloroquine or 100 nM BafA1) for 24 h with or without rapamycin (100 nM). In the presence of lysosomal inhibitors, more accumulation of LC3B-II was found in CS fibroblasts compared to control fibroblasts, indicating that CS fibroblasts exhibit upregulated autophagic flux (Fig. 1c-d).

Fig. 1
figure 1

Autophagic flux is upregulated in Cohen syndrome (CS) fibroblasts and CS iPSC-derived neurons. a Electron microscopic images of control and CS fibroblasts. b Quantitative analysis of EM images. The bar graph indicates the means ± SEM. *, p < 0.05 (Student’s t-test). c Western blot analysis showing autophagic flux in the presence or absence of a lysosomal inhibitor (bafilomycin A1 [BafA1], 100 nM) with or without rapamycin (100 nM, 4 h) using anti-LC3 or -GAPDH antibodies. d Relative expression of LC3-II normalized by GAPDH. The bar graph indicates the means ± SEM from three independent experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001 (one-way ANOVA followed by Tukey’s Multiple Comparison Test). e Electron microscopic images of synaptic regions in control and CS iPSC-derived neurons. f Western blot analysis showing autophagic flux in the presence or absence of a lysosomal inhibitor (chloroquine [CQ], 100 nM) using anti-LC3 or -β-actin antibodies. g Relative expression of LC3-II normalized by β-actin. The bar graph indicates the means ± SEM from three independent experiments. ***, p < 0.001 (one-way ANOVA followed by Tukey’s Multiple Comparison Test). h Western blot analysis showing autophagic flux in the presence or absence of a lysosomal inhibitor (BafA1, 100 nM) using anti-LC3 or -β-actin antibodies with WT or VPS13B KO HeLa cell (#1, or #2) lysates. i Quantification of band intensity of LC3-II or β-actin. The bar graph indicates the means ± SEM from three independent experiments. **, p < 0.01; ***, p < 0.001 (one-way ANOVA followed by Tukey’s Multiple Comparison Test). j The mRNA level of ATG4C determined by quantitative RT-PCR. The bar graph indicates the means ± SEM from three independent experiments. **, p < 0.01; ***, p < 0.001 (one-way ANOVA followed by Tukey’s Multiple Comparison Test). k Quantitative analysis of ATG4C expression in CS-derived neurons. (Inset) Western blot images of ATG4C expression in CS-derived neurons. The bar graph indicates the means ± SEM from five independent experiments. *, p < 0.05; **, p < 0.01 (one-way ANOVA followed by Tukey’s Multiple Comparison Test)

Full size image

The autophagy pathway is a conserved cellular degradation and recycling system that relies on lysosomal degradation of superfluous or cytotoxic components and damaged organelles engulfed by autophagic vacuoles in a selective or non-selective manner [6, 7]. Increasing evidence suggests that neuronal presynaptic autophagy in the axon terminal is essential for synaptic maintenance and plasticity [8]. To investigate whether the autophagy pathway is upregulated in CS iPSC-derived neurons, which previously showed synaptic protein (e.g., SV2B) alteration, 4-week old CS neurons were analyzed at the ultrastructural level using EM. Similar to EM results from CS fibroblasts, AVs were accumulated in presynaptic terminals of CS neurons compared to control neurons (Fig. 1e). Autophagic flux was significantly increased in CS neurons compared to control neurons (Fig. 1f-g). These data suggest that autophagy was also upregulated in CS neurons.

Following the previous finding, an examination of how autophagy is upregulated in CS patient-derived fibroblasts and neurons was conducted; previous findings showed that the mRNA level of VPS13B was significantly reduced (approximately 50%) in CS fibroblasts with heterogenous missense mutations (c.T1239G [p.Y413X] and c.G10333A [p.V3445M]) compared to control fibroblasts (under revision). Therefore, to clarify the role of VPS13B in autophagy, we used the CRISPR/Cas9 genome editing system to disrupt the gene encoding human VPS13B in HeLa cells (CRISPR guide RNA sequence [gRNAs] #1: 5′-GGTAATTACCATCAATACTA-3′; gRNAs #2: 5′- AATTGAGGATTCATGTACCA-3′) and validated knockout of VPS13B by genomic PCR. Figure 1h-i shows the level of LC3-II in VPS13B KO cells was higher in the presence of lysosomal inhibitors than in wild-type HeLa cells, indicating that the loss of VPS13B is associated with upregulated autophagic flux, suggesting its potential role in basal autophagy.

To discover autophagy-associated genes responsible for the upregulation of autophagy activity in CS neurons, a transcriptomic analysis for autophagy pathway-related genes in CS neurons and control neurons was implemented. In CS neurons, Gene Ontology analysis showed that two processes (autophagosome organization and macroautophagy) were significant. MFN2, ATG9B, ATP13A2, KIAA1324, MAP1LC3C, and ATG4C genes were upregulated in CS neurons compared to control neurons. To provide further confirmation, quantitative RT-PCR, and western blot analyses were performed from control and CS neurons and revealed that mRNA and protein levels of ATG4C in CS neurons were higher compared to control neurons (Fig. 1j-k). ATG4 is a cysteine protease required for LC3/gamma-aminobutyric acid receptor-associated protein (GABARAP) processing, which allows for the latter to be conjugated to phosphatidylethanolamine on autophagosomal membranes; a key step in autophagosome biogenesis and recycling [9].

Further examination was undertaken to see whether the gene expression of LC3B or GABARAP was affected in CS neurons. However, there was no significant difference in mRNA expression of LC3B or GABARAP between control and CS neurons (data not shown). The autophagic flux may be increased, at least partially, by the upregulation of ATG4C in CS neurons; however, VPS13B’s association with the upregulation of autophagy and CS pathogenesis should be further investigated. The VPS13 family is associated with various human diseases [10]; in particular, a loss of VPS13A function disrupts autophagic flux and leads to chorea-acanthocytosis (ChAc), a rare neurodegenerative disease with no known cure. The absence of VPS13A or VPS13B causes defective or increased autophagic flux; for this reason, autophagy might be a promising therapeutic target for human diseases associated with VPS13 [2, 11].