Sharing sensitive data in life sciences: an overview of centralized and federated approaches.

Biomedical data are generated and collected from various sources, including medical imaging, laboratory tests and genome sequencing. Sharing these data for research can help address unmet health needs, contribute to scientific breakthroughs, accelerate the development of more effective treatments and inform public health policy. Due to the potential sensitivity of such data, however, privacy concerns have led to policies that restrict data sharing. In addition, sharing sensitive data requires a secure and robust infrastructure with appropriate storage solutions. Here, we examine and compare the centralized and federated data sharing models through the prism of five large-scale and real-world use cases of strategic significance within the European data sharing landscape: the French Health Data Hub, the BBMRI-ERIC Colorectal Cancer Cohort, the federated European Genome-phenome Archive, the Observational Medical Outcomes Partnership/OHDSI network and the EBRAINS Medical Informatics Platform. Our analysis indicates that centralized models facilitate data linkage, harmonization and interoperability, while federated models facilitate scaling up and legal compliance, as the data typically reside on the data generator's premises, allowing for better control of how data are shared. This comparative study thus offers guidance on the selection of the most appropriate sharing strategy for sensitive datasets and provides key insights for informed decision-making in data sharing efforts.

Mutations in ATP6AP2 cause autophagic liver disease in humans.

The biogenesis of the proton pump V-ATPase commences with the assembly of the proton pore sector V0 in the endoplasmic reticulum (ER). This process occurs under the control of a group of assembly factors whose mutations have recently been shown to cause glycosylation disorders with overlapping phenotypes in humans. Using whole exome sequencing, we demonstrate that mutations of the accessory V-ATPase subunit ATP6AP2 cause a similar disease characterized by hepatosteatosis, lipid abnormalities, immunodeficiency and cognitive impairment. ATP6AP2 interacts with members of the V0 assembly complex, and its ER localization is crucial for V-ATPase activity. Moreover, ATP6AP2 mutations can cause developmental defects and steatotic phenotypes when introduced into Drosophila. Altogether, our data suggest that these phenotypes are the result of a pathogenetic cascade that includes impaired V-ATPase assembly, defective lysosomal acidification, reduced MTOR signaling and autophagic misregulation.

Mutations in the X-linked ATP6AP2 cause a glycosylation disorder with autophagic defects.

The biogenesis of the multi-subunit vacuolar-type H+-ATPase (V-ATPase) is initiated in the endoplasmic reticulum with the assembly of the proton pore V0, which is controlled by a group of assembly factors. Here, we identify two hemizygous missense mutations in the extracellular domain of the accessory V-ATPase subunit ATP6AP2 (also known as the [pro]renin receptor) responsible for a glycosylation disorder with liver disease, immunodeficiency, cutis laxa, and psychomotor impairment. We show that ATP6AP2 deficiency in the mouse liver caused hypoglycosylation of serum proteins and autophagy defects. The introduction of one of the missense mutations into Drosophila led to reduced survival and altered lipid metabolism. We further demonstrate that in the liver-like fat body, the autophagic dysregulation was associated with defects in lysosomal acidification and mammalian target of rapamycin (mTOR) signaling. Finally, both ATP6AP2 mutations impaired protein stability and the interaction with ATP6AP1, a member of the V0 assembly complex. Collectively, our data suggest that the missense mutations in ATP6AP2 lead to impaired V-ATPase assembly and subsequent defects in glycosylation and autophagy.

WDR81 mutations cause extreme microcephaly and impair mitotic progression in human fibroblasts and Drosophila neural stem cells.

Microlissencephaly is a rare brain malformation characterized by congenital microcephaly and lissencephaly. Microlissencephaly is suspected to result from abnormalities in the proliferation or survival of neural progenitors. Despite the recent identification of six genes involved in microlissencephaly, the pathophysiological basis of this condition remains poorly understood. We performed trio-based whole exome sequencing in seven subjects from five non-consanguineous families who presented with either microcephaly or microlissencephaly. This led to the identification of compound heterozygous mutations in WDR81, a gene previously associated with cerebellar ataxia, intellectual disability and quadrupedal locomotion. Patient phenotypes ranged from severe microcephaly with extremely reduced gyration with pontocerebellar hypoplasia to moderate microcephaly with cerebellar atrophy. In patient fibroblast cells, WDR81 mutations were associated with increased mitotic index and delayed prometaphase/metaphase transition. Similarly, in vivo, we showed that knockdown of the WDR81 orthologue in Drosophila led to increased mitotic index of neural stem cells with delayed mitotic progression. In summary, we highlight the broad phenotypic spectrum of WDR81-related brain malformations, which include microcephaly with moderate to extremely reduced gyration and cerebellar anomalies. Our results suggest that WDR81 might have a role in mitosis that is conserved between Drosophila and humans.

New insights into centrosome imaging in Drosophila and mouse neuroepithelial tissues.

The centrosome is the main microtubule-organizing center in animal cells. It participates in the assembly of a bipolar spindle that ensures accurate segregation of chromosomes during mitosis. Recently, mutations in centrosome genes have been identified in patients affected by neurodevelopmental disorders. In fact, the etiology of several neurodevelopmental pathologies seems to be linked to defects in the assembly of the mitotic spindle in the neural stem cell compartment during neurogenesis. Therefore, getting better insights into the structure and function/dysfunction of the mitotic spindle apparatus in an intact tissue environment is of utmost importance. However, imaging nanometer-scale structures like centrosomes and microtubule bundles within the depth of a tissue is still challenging. Here we describe two procedures to acquire high-resolution images on fixed tissues and to perform live imaging of microtubule-based structures in the neuroepithelia of the Drosophila brain and of the mouse neocortex. We take advantage of the accumulation of centrosomes and mitotic figures at the apical surface of these polarized tissues to improve the quality of staining and imaging. Both Drosophila and mouse models with centrosome dysfunction showed abnormalities in the neuroepithelium reminiscent of the ones described in brains of human patients. These observations have highlighted their value as model organisms to study the etiology of human neurodevelopmental pathologies.

The microcephaly protein Asp regulates neuroepithelium morphogenesis by controlling the spatial distribution of myosin II.

Mutations in ASPM are the most frequent cause of microcephaly, a disorder characterized by reduced brain size at birth. ASPM is recognized as a major regulator of brain size, yet its role during neural development remains poorly understood. Moreover, the role of ASPM proteins in invertebrate brain morphogenesis has never been investigated. Here, we characterized the function of the Drosophila ASPM orthologue, Asp, and found that asp mutants present severe defects in brain size and neuroepithelium morphogenesis. We show that size reduction depends on the mitotic function of Asp, whereas regulation of tissue shape depends on an uncharacterized function. Asp interacts with myosin II regulating its polarized distribution along the apico-basal axis. In the absence of Asp, mislocalization of myosin II results in interkinetic nuclear migration and tissue architecture defects. We propose that Asp regulates neuroepithelium morphogenesis through myosin-II-mediated structural and mechanical processes to maintain force balance and tissue cohesiveness.

Centrosome amplification causes microcephaly.

Centrosome amplification is a hallmark of human tumours. In flies, extra centrosomes cause spindle position defects that result in the expansion of the neural stem cell (NSC) pool and consequently in tumour formation. Here we investigated the consequences of centrosome amplification during mouse brain development and homeostasis. We show that centrosome amplification causes microcephaly due to inefficient clustering mechanisms, where NSCs divide in a multipolar fashion producing aneuploid cells that enter apoptosis. Importantly, we show that apoptosis inhibition causes the accumulation of highly aneuploid cells that lose their proliferative capacity and differentiate, thus depleting the pool of progenitors. Even if these conditions are not sufficient to halt brain development, they cause premature death due to tissue degeneration. Our results support an alternative concept to explain the etiology of microcephaly and show that centrosome amplification and aneuploidy can result in tissue degeneration rather than overproliferation and cancer.

A DNAJB chaperone subfamily with HDAC-dependent activities suppresses toxic protein aggregation.

Misfolding and aggregation are associated with cytotoxicity in several protein folding diseases. A large network of molecular chaperones ensures protein quality control. Here, we show that within the Hsp70, Hsp110, and Hsp40 (DNAJ) chaperone families, members of a subclass of the DNAJB family (particularly DNAJB6b and DNAJB8) are superior suppressors of aggregation and toxicity of disease-associated polyglutamine proteins. The antiaggregation activity is largely independent of the N-terminal Hsp70-interacting J-domain. Rather, a C-terminal serine-rich (SSF-SST) region and the C-terminal tail are essential. The SSF-SST region is involved in substrate binding, formation of polydisperse oligomeric complexes, and interaction with histone deacetylases (HDAC4, HDAC6, SIRT2). Inhibiting HDAC4 reduced DNAJB8 function. DNAJB8 is (de)acetylated at two conserved C-terminal lysines that are not involved in substrate binding, but do play a role in suppressing protein aggregation. Combined, our data provide a functional link between HDACs and DNAJs in suppressing cytotoxic protein aggregation.

Polarised asymmetric inheritance of accumulated protein damage in higher eukaryotes.

Disease-associated misfolded proteins or proteins damaged due to cellular stress are generally disposed via the cellular protein quality-control system. However, under saturating conditions, misfolded proteins will aggregate. In higher eukaryotes, these aggregates can be transported to accumulate in aggresomes at the microtubule organizing center. The fate of cells that contain aggresomes is currently unknown. Here we report that cells that have formed aggresomes can undergo normal mitosis. As a result, the aggregated proteins are asymmetrically distributed to one of the daughter cells, leaving the other daughter free of accumulated protein damage. Using both epithelial crypts of the small intestine of patients with a protein folding disease and Drosophila melanogaster neural precursor cells as models, we found that the inheritance of protein aggregates during mitosis occurs with a fixed polarity indicative of a mechanism to preserve the long-lived progeny.

Retinoic acid-induced differentiation into astrocytes and glutamatergic neurons is associated with expression of functional and activable phospholipase D.

Phospholipase D (PLD) activity in mammalian cells has been associated with cell proliferation and differentiation. Here, we investigated the expression of PLD during differentiation of pluripotent embryonal carcinoma cells (P19) into astrocytes and neurons. Retinoic acid (RA)-induced differentiation increased PLD1 and PLD2 mRNA levels and PLD activity that was responsive to phorbol myristate acetate. Various agonists of membrane receptors activated PLD in RA-differentiated cells. Glutamate was a potent activator of PLD in neurons but not in astrocytes, whereas noradrenaline and carbachol increased PLD activity only in astrocytes. P19 neurons but not astrocytes released glutamate in response to a depolarizing stimulus, confirming the glutamatergic phenotype of these neurons. These results indicate upregulation of PLD gene expression associated with RA-induced neural differentiation.