Publikationen

The development of complex multicellular human in vitro systems holds great promise for modelling disease and advancing drug discovery and tissue engineering1. In the liver, despite the identification of key signalling pathways involved in hepatic regeneration2,3, in vitro expansion of human hepatocytes directly from fresh patient tissue has not yet been achieved, limiting the possibility of modelling liver composite structures in vitro. Here we first developed human hepatocyte organoids (h-HepOrgs) from 28 different patients. Patient-derived hepatocyte organoids sustained long-term expansion of hepatocytes in vitro and maintained patient-specific gene expression and bile canaliculus features and function of the in vivo tissue. After transplantation, expanded h-HepOrgs rescued the phenotype of a mouse model of liver disease. By combining h-HepOrgs with portal mesenchyme and our previously published cholangiocyte organoids4-6, we generated patient-specific periportal liver assembloids that retain the histological arrangement, gene expression and cell interactions of periportal liver tissue, with cholangiocytes and mesenchyme embedded in the hepatocyte parenchyma. We leveraged this platform to model aspects of biliary fibrosis. Our human periportal liver assembloid system represents a novel in vitro platform to investigate human liver pathophysiology, accelerate drug development, enable early diagnosis and advance personalized medicine.

In recent years, there has been growing interest in jointly analyzing a foreground dataset, representing an experimental group, and a background dataset, representing a control group. The goal of such contrastive investigations is to identify salient features in the experimental group relative to the control. Independent component analysis (ICA) is a powerful tool for learning independent patterns in a dataset. We generalize it to contrastive ICA (cICA). For this purpose, we devise a linear algebra-based tensor decomposition algorithm, which is more expressive but just as efficient and identifiable as other linear algebra-based algorithms. We establish the identifiability of cICA and demonstrate its performance in finding patterns and visualizing data, using synthetic, semisynthetic, and real-world datasets, comparing the approach to existing methods.

Parkinson's disease (PD) is histopathologically defined by the presence of Lewy bodies, which are intracellular proteinaceous inclusions that contain mainly aggregated alpha-synuclein (aSyn). It is believed that oligomeric intermediates between monomeric aSyn and large aggregates are neurotoxic, which would lead to the demise of dopaminergic neurons. Therefore, novel therapies preventing aSyn-induced cell death need to be developed. Therefore, we performed a genome-wide siRNA screening in an aSyn-induced dopaminergic cell death model and found the knockdown of three transforming growth factor-beta (TGFb) pathway-related genes to be protective. Hence, we hypothesized that a reduction in TGFb signaling would protect dopaminergic neurons from aSyn-induced toxicity. Thus, we validated the results of the genome-wide knockdown screening with the use of two different types of siRNAs. We confirmed that the knockdown of Activin receptor-like kinase 5 (ALK5) and Mothers against decapentaplegic homolog 2 (SMAD2), two genes of the TGFb pathway, protected dopaminergic neurons from aSyn-induced toxicity. An increase in TGFb signaling by treatment with TGFb ligands further exacerbated aSyn-induced toxicity, whereas this effect was mitigated by knockdown of ALK5, SMAD2, or Dynein light chain roadblock type-1 (DYNLRB1). Moreover, TGFb ligand treatment induced an up-regulation of SNCA mRNA expression in aSyn-overexpressing cells. Interestingly, consistent with the literature, we identified an up-regulation of the genes of the TGFb pathway in aSyn-overexpressing cells. Altogether, we identified a potential protective role by interference with the TGFb pathway against aSyn-induced toxicity. These findings provide a rationale for the development of novel strategies against PD.

Actins are cytoskeletal proteins that are essential for multiple cellular processes. Mutations in the ACTB and ACTG1 genes, encoding the ubiquitous beta- and gamma-cytoskeletal actin isoforms, respectively, cause a broad spectrum of neurodevelopmental disorders, with microcephaly as the most frequent one. To investigate the pathogenesis underlying this cortical malformation, we studied patient-derived cerebral organoids from induced pluripotent stem cells of individuals with the Baraitser-Winter-CerebroFrontoFacial syndrome (BWCFF-S) carrying an ACTB/ACTG1 missense mutation. These organoids were reduced in size, showing a thinner ventricular zone (VZ) due to reduced VZ progenitor abundance. Strikingly, VZ progenitors in BWCFF-S cerebral organoids displayed a shift in the orientation of their cleavage plane from a predominantly vertical to a majoritarian horizontal orientation. The latter cleavage plane orientation is incompatible with increasing VZ progenitor abundance and instead promotes basal progenitor generation. Various cytoskeletal and morphological irregularities of BWCFF-S VZ progenitors, notably in the apical region, seemingly contribute to this change in cleavage plane orientation. Our results provide insight into the cell biological basis of the microcephaly associated with BWCFF-S caused by actin mutations.

The thymus is the primary site of T cell development, central to the establishment of self-tolerance and adaptive immune function. In mammals, the thymus undergoes age-related involution, resulting in a global decline in immune function. The thymus has some regenerative ability that relies on pre-existing thymic remnants but is insufficient to prevent involution. Here, we show that the juvenile axolotl (Ambystoma mexicanum) is able to regenerate its thymus de novo after complete removal, constituting an exception among vertebrates. Using single-cell transcriptomics and genetic and transplantation approaches, we demonstrate that de novo thymus regeneration results in the restoration of morphology, cell-type diversity, and function. FOXN1, although it has a conserved role in thymus organogenesis, is dispensable for the initiation of thymic regeneration. In contrast, we identify midkine signaling as a possible early driver of de novo thymus regeneration. This study demonstrates an instance of organ-level regeneration of the lymphoid system, which could guide future clinical strategies seeking to promote thymus regrowth.

The trophectoderm (TE), the first lineage specified during mammalian development, initiates implantation and gives rise to placental trophoblasts. While animal models have elucidated key conserved signaling pathways involved in early TE specification, including bone morphogenetic protein (BMP), WNT, and HIPPO, species-specific differences during early development emphasize the need for human-specific models. We previously identified VGLL1, a coactivator of TEAD transcription factors, as a human-specific placental marker. In this study, we employed a pluripotent stem cell (PSC)-based model of TE induction by BMP4 to investigate chromatin remodeling and transcriptional dynamics during TE formation. BMP4-induced chromatin accessibility changes promoted a trophoblast gene expression program, while mesoderm lineage markers were only transiently expressed upon canonical WNT activation. We found that VGLL1 was expressed downstream of key TE transcription factors (GATA2/3, TFAP2A/C) but was essential for establishment of full trophoblast identity by up-regulating the epidermal growth factor receptor (EGFR) and reinforcing GATA3 expression through positive feedback. Notably, VGLL1 enhanced canonical WNT signaling via direct regulation of WNT receptors and effectors. We also identified KDM6B, a histone demethylase that removes H3K27me3 repressive marks, as a direct VGLL1 target. KDM6B facilitated activation of bivalent promoters associated with TE markers, linking epigenetic regulation to lineage identity. Our findings establish a mechanistic framework positioning VGLL1 as a central regulator that integrates HIPPO, BMP, and WNT signaling pathways to drive establishment of human TE.

Biomolecular condensates have been implicated in key cellular processes such as gene regulation, stress response, and signaling, and dysregulation of condensates has been linked to neurodegeneration and other diseases. Computational algorithms that predict protein condensation can aid systematic characterization of biomolecular condensates at the proteome scale. However, many experimental labs may lack the computational background or resources to run sophisticated prediction tools locally.

Many subcellular compartments are biomolecular condensates made of multiple components, often including several distinct proteins and nucleic acids. However, current tools to measure condensate composition are limited and cannot capture this complexity quantitatively because they either require fluorescent labels, which can perturb composition, or can distinguish only one or two components. Here we describe a label-free method based on quantitative phase imaging and analysis of tie-lines and refractive index to measure the composition of reconstituted condensates with multiple components. We first validate the method empirically in binary mixtures, revealing sequence-encoded density variation and complex ageing dynamics for condensates composed of full-length proteins. We then use analysis of tie-lines and refractive index to simultaneously resolve the concentrations of five macromolecular solutes in multicomponent condensates containing RNA and constructs of multiple RNA-binding proteins. Our measurements reveal an unexpected decoupling of density and composition, highlighting the need to determine molecular stoichiometry in multicomponent condensates. We foresee this approach enabling the study of compositional regulation of condensate properties and function.

We express the maximum likelihood (ML) degrees of a family of toric varieties in terms of Möbius invariants of matroids. The family of interest are those parametrized by monomial maps given by Lawrence lifts of totally unimodular matrices with even circuits. Specifying these matrices to be vertex-edge incidence matrices of bipartite graphs gives the ML degrees of some hierarchical models and three dimensional quasi-independence models. Included in this list are the no-three-way interaction models with one binary random variable, for which we give closed formulae.

Mammalian brain development involves coordinated progenitor amplification, migration, and differentiation. Apical progenitors (APs) establish the ventricular zone (VZ) and are key determinants of brain size and complexity. Rho-family guanosine triphosphatases (Rho-GTPases) regulate cytoskeletal dynamics and AP behavior, but how specific Rho GTPase-activating proteins (Rho-GAPs) influence progenitor identity and VZ architecture remains unclear. Using human forebrain organoids, we investigate the function of the Rho GAP ARHGAP11A in human corticogenesis. CRISPR-Cas9-mediated ARHGAP11A knockout reveals its essential role in maintaining VZ integrity. Loss of ARHGAP11A impairs neuroepithelial organization and randomizes mitotic cleavage-plane orientation via the RHOA-ROCK-actin axis. This leads to premature AP delamination, AP depletion, and reduced cell density and glial numbers. Pharmacological inhibition of RHOA or ROCK rescues these defects, highlighting ARHGAP11A's role in cytoskeletal remodeling to maintain cortical progenitors. Our findings establish ARHGAP11A as a critical regulator of AP identity and VZ integrity, with broad implications for human corticogenesis.