3D model of mouse embryonic pancreas and endocrine compartment using stem cell-derived mesoderm and pancreatic progenitors.

The developing mouse pancreas is surrounded by mesoderm compartments providing signals that induce pancreas formation. Most pancreatic organoid protocols lack this mesoderm niche and only partially capture the pancreatic cell repertoire. This work aims to generate pancreatic aggregates by differentiating mouse embryonic stem cells (mESCs) into mesoderm progenitors (MPs) and pancreas progenitors (PPs), without using Matrigel. First, mESCs were differentiated into epiblast stem cells (EpiSCs) to enhance the PP differentiation rate. Next, PPs and MPs aggregated together giving rise to various pancreatic cell types, including endocrine, acinar, and ductal cells, and to endothelial cells. Single-cell RNA sequencing analysis revealed a larger endocrine population within the PP + MP aggregates, as compared to PPs alone or PPs in Matrigel aggregates. The PP + MP aggregate gene expression signatures and its endocrine population percentage closely resembled those of the endocrine population found in the mouse embryonic pancreas, which holds promise for studying pancreas development.

Targeting cholesterol impairs cell invasion of all breast cancer types.

BACKGROUND: Breast cancer clinical outcome relies on its intrinsic molecular subtype and mortality is almost exclusively due to metastasis, whose mechanism remains unclear. We recently revealed the specific contribution of plasma membrane cholesterol to the invasion of malignant MCF10CAIa but not premalignant MCF10AT and normal MCF10A cell lines in 2D, through invadopodia formation and extracellular matrix (ECM) degradation. In the present study, we address the impact of breast cancer subtypes, mutations and aggressiveness on cholesterol implication in breast cancer cell invasion and 3D spheroid invasion and growth. METHODS: We used nine breast cancer cell lines grouped in four subtypes matching breast tumor classification. Four of these cell lines were also used to generate 3D spheroids. These cell lines were compared for cell invasion in 2D and 3D, spheroid growth in 3D, gelatin degradation, cortactin expression, activation and subcellular distribution as well as cell surface cholesterol distribution and lipid droplets. The effect of plasma membrane cholesterol depletion on all these parameters was determined in parallel and systematically compared with the impact of global matrix metalloproteinase (MMP) inhibition. RESULTS: The six invasive cell lines in 2D were sensitive to partial cholesterol depletion, independently of their subtype, aggressiveness or mutation. Nevertheless, the effect was stronger in the three cell lines able to degrade gelatin. 3D spheroid invasion was also reduced after cholesterol depletion in all breast cancer subtypes tested. Notably, targeting cholesterol was more powerful than MMP inhibition in reducing invasion in both 2D and 3D culture models. Moreover, cholesterol depletion in the six invasive cell lines impaired cortactin distribution in the perinuclear region where invadopodia localized. Breast cancer cell line aggressiveness relied on cholesterol-enriched domains at the ECM-free side and intracellular lipid droplets. Furthermore, the three gelatin-degrading cell lines were characterized by increased cholesterol-enriched submicrometric domains at their ECM-contact side. CONCLUSION: Together, our data suggest cell surface cholesterol combined with lipid droplet labeling as a breast cancer cell aggressiveness marker. They also open the way to test other cholesterol-targeting drugs in more complex models to further evaluate whether cholesterol could represent a strategy in breast cancer therapy.

Piezo1 Is Required for Myoblast Migration and Involves Polarized Clustering in Association with Cholesterol and GM1 Ganglioside.

A specific plasma membrane distribution of the mechanosensitive ion channel Piezo1 is required for cell migration, but the mechanism remains elusive. Here, we addressed this question using WT and Piezo1-silenced C2C12 mouse myoblasts and WT and Piezo1-KO human kidney HEK293T cells. We showed that cell migration in a cell-free area and through a porous membrane decreased upon Piezo1 silencing or deletion, but increased upon Piezo1 activation by Yoda1, whereas migration towards a chemoattractant gradient was reduced by Yoda1. Piezo1 organized into clusters, which were preferentially enriched at the front. This polarization was stimulated by Yoda1, accompanied by Ca2+ polarization, and abrogated by partial cholesterol depletion. Piezo1 clusters partially colocalized with cholesterol- and GM1 ganglioside-enriched domains, the proportion of which was increased by Yoda1. Mechanistically, Piezo1 activation induced a differential mobile fraction of GM1 associated with domains and the bulk membrane. Conversely, cholesterol depletion abrogated the differential mobile fraction of Piezo1 associated with clusters and the bulk membrane. In conclusion, we revealed, for the first time, the differential implication of Piezo1 depending on the migration mode and the interplay between GM1/cholesterol-enriched domains at the front during migration in a cell-free area. These domains could provide the optimal biophysical properties for Piezo1 activity and/or spatial dissociation from the PMCA calcium efflux pump.

Piezo1 Regulation Involves Lipid Domains and the Cytoskeleton and Is Favored by the Stomatocyte-Discocyte-Echinocyte Transformation.

Piezo1 is a mechanosensitive ion channel required for various biological processes, but its regulation remains poorly understood. Here, we used erythrocytes to address this question since they display Piezo1 clusters, a strong and dynamic cytoskeleton and three types of submicrometric lipid domains, respectively enriched in cholesterol, GM1 ganglioside/cholesterol and sphingomyelin/cholesterol. We revealed that Piezo1 clusters were present in both the rim and the dimple erythrocyte regions. Upon Piezo1 chemical activation by Yoda1, the Piezo1 cluster proportion mainly increased in the dimple area. This increase was accompanied by Ca2+ influx and a rise in echinocytes, in GM1/cholesterol-enriched domains in the dimple and in cholesterol-enriched domains in the rim. Conversely, the effects of Piezo1 activation were abrogated upon membrane cholesterol depletion. Furthermore, upon Piezo1-independent Ca2+ influx, the above changes were not observed. In healthy donors with a high echinocyte proportion, Ca2+ influx, lipid domains and Piezo1 fluorescence were high even at resting state, whereas the cytoskeleton membrane occupancy was lower. Accordingly, upon decreases in cytoskeleton membrane occupancy and stiffness in erythrocytes from patients with hereditary spherocytosis, Piezo1 fluorescence was increased. Altogether, we showed that Piezo1 was differentially controlled by lipid domains and the cytoskeleton and was favored by the stomatocyte-discocyte-echinocyte transformation.