Search

The liver is one of the main organs involved in the metabolism of xenobiotics and a key organ in toxicity studies. Prior to accessing the hepatocytes, xenobiotics pass through the hepatic sinusoid formed by liver sinusoidal endothelial cells (LSECs). The LSECs barrier regulates the kinetics and concentrations of the xenobiotics before their metabolic processing by the hepatocytes. To mimic this physiological situation, we developed an in vitro model reproducing an LSECs barrier in coculture with a hepatocyte biochip, using a fluidic platform. This technology made dynamic coculture and tissue crosstalk possible. SK-HEP-1 and HepG2/C3a cells were used as LSECs and as hepatocyte models, respectively. We confirmed the LSECs phenotype by measuring PECAM-1 and stabilin-2 expression levels and the barrier’s permeability/transport properties with various molecules. The tightness of the SK-HEP-1 barrier was enhanced in the dynamic coculture. The morphology, albumin secretion, and gene expression levels of markers of HepG2/C3a were not modified by coculture with the LSECs barrier. Using paracetamol, a well-known hepatotoxic drug, to study tissue crosstalk, there was a reduction in the expression levels of the LSECs markers stabilin-2 and PECAM-1, and a modification of those of CLEC4M and KDR. No HepG2/C3a toxicity was observed. The metabolisation of paracetamol by HepG2/C3a monocultures and cocultures was confirmed. Although primary cells are required to propose a fully relevant model, the present approach highlights the potential of our system for investigating xenobiotic metabolism and toxicity.

The 3Rs guidelines recommend replacing animal testing with alternative models. One of the solutions proposed is organ-on-chip technology in which liver-on-chip is one of the most promising alternatives for drug screening and toxicological assays. The main challenge is to achieve the relevant in vivo-like functionalities of the liver tissue in an optimized cellular microenvironment. Here, we investigated the development of hepatic cells under dynamic conditions inside a 3D hydroscaffold embedded in a microfluidic device. The hydroscaffold is made of hyaluronic acid and composed of liver extracellular matrix components (galactosamine, collagen I/IV) with RGDS (Arg-Gly-Asp-Ser) sites for cell adhesion. The HepG2/C3A cell line was cultured under a flow rate of 10 µL/min for 21 days. After seeding, the cells formed aggregates and proliferated, forming 3D spheroids. The cell viability, functionality, and spheroid integrity were investigated and compared to static cultures. The results showed a 3D aggregate organization of the cells up to large spheroid formations, high viability and albumin production, and an enhancement of HepG2 cell functionalities. Overall, these results highlighted the role of the liver-on-chip model coupled with a hydroscaffold in the enhancement of cell functions and its potential for engineering a relevant liver model for drug screening and disease study.

International audience; AbstractIn this article, we propose a hierarchical deep learning framework for the nuclei 3D reconstruction from a stack of microscopic images representing 3D cancer cell culture. The framework goes through three successive stages namely: at the slice level of the stack (i) the spheroid detection and (ii) their nuclei segmentation then at the stack level (iii) the nuclei 3D reconstruction. For this purpose, we prepared a dataset of bright-field microscopic images acquired from 3D cultures of HeLa cells and manually annotated by the experts for both tasks (spheroids detection and nuclei segmentation). Two CNN models namely, YOLOv5x and U-Net-VGG19 have been trained and validated on our dataset for the detection and the segmentation tasks, respectively. For the 3D reconstruction task, the delaunay triangulation technique has been adopted by exploiting point cloud clusters that represent the segmented nuclei in the stack. Our framework offers to the biologists an efficient assisting tool for quantifying the number of spheroids and analyzing the morphology of their nuclei. The conducted experiments on our generated dataset show the promising results obtained by our framework with notably an average precision of 0.892 and 0.76 on the spheroids detection and nuclei segmentation respectively. Moreover, our 3D reconstruction technique shows visually a consistant representation of nuclei in term of volumetery and shape

Background: The manual segmentation of cellular structures on Z-stack microscopic images is time-consuming and often inaccurate, highlighting the need to develop auto-segmentation tools to facilitate this process. Objective: This study aimed to compare the performance of three different machine learning archi-tectures, including random forest (RF), AdaBoost, and multi-layer perceptron (MLP), for the auto-segmentation of nuclei in proliferating cervical cancer cells on Z-Stack cellular microscopy prolif-eration images provided by the HCS Pharma. The impact of using post-processing techniques, such as the StarDist plugin and majority voting, was also evaluated. Methods: The RF, AdaBoost, and MLP algorithms were used to auto-segment the nuclei of cervi-cal cancer cells on microscopic images at different Z-stack positions. Post-processing techniques were then applied to each algorithm. The performance of all algorithms was compared by an expert to globally generated ground truth by calculating the accuracy detection rate, the Dice coefficient, and the Jaccard index. Results: RF achieved the best accuracy, followed by the AdaBoost and then the MLP. All algo-rithms achieved good pixel classifications except in regions whereby the nuclei overlapped. The majority voting and StarDist plugin improved the accuracy of the segmentation but did not resolve the nuclei overlap issue. The Z-Stack analysis revealed similar segmentation results to the Z-stack layer used to train the image. However, a worse performance was noted for segmentations per-formed on different Z-stack positions, which were not used to train the algorithms. Conclusion: All machine learning architectures provided a good segmentation of nuclei in cervical cancer cells but did not resolve the problem of overlapping nuclei and Z-stack segmentation. Fur-ther research should therefore evaluate the combined segmentation techniques and deep learning architectures to resolve these issues.

Treatment for CNS related diseases are limited by the difficulty of the drugs to cross the blood-brain barrier (BBB). The functionalization of polymeric nanoparticles (NPs) coated with the surfactants polysorbate 80 (PS80) and poloxamer 188 (P188), have shown promising results as drugs carriers are able to cross the BBB on animal models. In this study, poly(lactide-co-glycolide) (PLGA) NPs coated with PS80 and P188, labelled with a fluorescent dye were tested on human pre-clinical in vitro model to evaluate and compare their uptake profiles, mechanisms of transport and crossing over human brain-like endothelial cells (BLECs) mimicking the human BBB. In addition, these NPs were produced using a method facilitating their reproducible production at high scale, the MicroJet reactor® technology. Results showed that both formulations were biocompatible and able to be internalized within the BLECs in different uptake profiles depending on their coating: P188 NP showed higher internalization capacity than PS80 NP. Both NPs uptakes were ATP-dependent, following more than one endocytosis pathway with colocalization in the early endosomes, ending with a NPs release in the brain compartment. Thus, both surfactant-coated PLGA NPs are interesting formulations for delivery to the brain through the BBB, presenting different uptake profiles.

International audience; Pancreatic cancer (PC) is a deadly cancer for which no diagnostic or prevention plan currently exists since this cancer is asymptomatic in its early development. The very specific and complex tumor microenvironment of PC is responsible for the unsuccessful delivery of therapeutics molecules to tumor cells. For high throughput testing of new molecules, there is a crucial need for more accurate preclinical model. For the first time, we integrated in a microfluidic system 5 major features of PC microenvironment: i) type-I collagen (CI), ii) hyaluronic acid (HA), iii) pancreatic stellate cells (SC), iv) interstitial fluid flow and v) nutrient and oxygen gradients. We showed long 3D culture (30 days). Furthermore, we implemented, thanks to the perfusion capability, the actual FOLFIRINOX protocol to assess the IC50 and we showed that dynamic conditions induce increased chemoresistance compared to 3D static.

The SH-SY5Y cell line is commonly used for the assessment of neurotoxicity in drug discovery. These neuroblastoma-derived cells can be differentiated into neurons using many methods. The present study has compared 24 of these differentiation methods on SH-SY5Y cells. After morphologic selection of the three most differentiating media (retinoic acid in 10% fetal bovine serum (FBS), staurosporine in 1% FBS medium, and cyclic adenosine monophosphate (cAMP) in B21-supplemented neurobasal medium), cells were analyzed for pan-neuronal and specific neuronal protein expression by fluorescent automated imaging. The response of SH-SY5Y to a set of compounds of known toxicity was examined in these culture conditions performed in 2D, and also in a 3D hyaluronic acid-based hydroscaffold™ which mimics the extracellular matrix. The extent of neuronal markers expression and the sensitivity to neurotoxic compounds varied according to the differentiation medium. The cAMP B21-supplemented neurobasal medium led to the higher neuronal differentiation, and the higher sensitivity to neurotoxic compounds. The culture in 3D modified the neurotoxic response, through a lower sensitivity of cells compared to the 2D culture. The in vitro differentiation environment influences the neurotoxic response of SH-SY5Y cells and thus should be considered carefully in research as well as in drug discovery.

International audience; Central nervous system (CNS) diseases are one of the top causes of death worldwide. As there is a difficulty of drug penetration into the brain due to the blood–brain barrier (BBB), many CNS drugs treatments fail in clinical trials. Hence, there is a need to develop effective CNS drugs following strategies for delivery to the brain by better selecting them as early as possible during the drug discovery process. The use of in vitro BBB models has proved useful to evaluate the impact of drugs/compounds toxicity, BBB permeation rates and molecular transport mechanisms within the brain cells in academic research and early-stage drug discovery. However, these studies that require biological material (animal brain or human cells) are time-consuming and involve costly amounts of materials and plastic wastes due to the format of the models. Hence, to adapt to the high yields needed in early-stage drug discoveries for compound screenings, a patented well-established human in vitro BBB model was miniaturized and automated into a 96-well format. This replicate met all the BBB model reliability criteria to get predictive results, allowing a significant reduction in biological materials, waste and a higher screening capacity for being extensively used during early-stage drug discovery studies.

L’imagerie cellulaire a haut contenu, plus connu sous le nom de High Content Screening (HCS) fait reference a l’utilisation de microscopes et de l’analyse d’images automatises pour visualiser et mesurer quantitativement les caracteristiques des cellules a grande echelle. Une plate-forme typique dediee a l’HCS comprend la manipulation automatisee des liquides pour le traitement des cellules, la fixation et le marquage des cellules, des microscopes automatises pour la prise d’image, des analyses automatisees des images et de donnees statistiques. Contrairement aux autres methodes a haut debit (High Throughout Screening [HTS]), l’approche HCS permet la mise en œuvre de criblage phenotypique robuste, reposant sur l’etude des modifications phenotypes des cellules, incluant la morphologie des cellules et/ou des organelles comme le noyau ou les mitochondries, leur taille, mais aussi l’expression de proteines d’interet et le comportement des cellules (migration, proliferation…). Le HCS permet les mesures phenotypiques dans les cellules individuelles et les analyses de reponse heterogenes, fournissant des informations plus approfondies sur les processus biologiques. De nos jours, les maladies neurologiques demeurent un grand defi en termes d’options therapeutiques, pouvant mettre en evidence la complexite des mecanismes cellulaires et moleculaires impliques dans leur initiation et leur progression. Alors que les approches traditionnelles pour identifier les voies de la maladie reposent souvent sur la connaissance du defaut genetique responsable, les criblages HCS offrent une methode non biaisee basee sur des defauts morphologiques ou fonctionnels de cellules ou de tissus malades. De plus, ce type d’approche est particulierement pertinent pour le repositionnement des medicaments dans les maladies rares, en limitant la recherche a des composes qui ont deja ete testes chez l’homme, reduisant ainsi la necessite de tests precliniques approfondis. Dans cette presentation, nous discuterons des principaux aspects techniques de HCS, ses applications pour la decouverte de therapies et de mecanismes moleculaires impliques dans les processus pathologiques, en se concentrant sur les troubles du systeme nerveux central. Nous presenterons egalement certains des defis actuels dans le domaine et les solutions possibles offertes par des systemes de culture cellulaire innovants.

Caco-2 cells, widely used to study carrier mediated uptake and efflux mechanisms, are known to have different properties when cultured under different conditions. In this study, Caco-2 cells from 10 different laboratories were compared in terms of mRNA expression levels of 72 drug and nutrient transporters, and 17 other target genes, including drug metabolising enzymes, using real-time PCR. The rank order of the top five expressed genes was: HPT1>GLUT3>GLUT5>GST1A>OATP-B. Rank correlation showed that for most of the samples, the gene ranking was not significantly different. Functionality of transporters and the permeability of passive transport markers metoprolol (transcellular) and atenolol (paracellular) were also compared. MDR1 and PepT1 function was investigated using talinolol and Gly-Sar transport, respectively. Sulfobromophthalein (BSP) was used as a marker for MRP2 and OATP-B functionality. Atenolol permeability was more variable across laboratories than metoprolol permeability. Talinolol efflux was observed by all the laboratories, whereas only five laboratories observed significant apical uptake of Gly-Sar. Three laboratories observed significant efflux of BSP. MDR1 expression significantly correlated to the efflux ratio and net active efflux of talinolol. PepT1 mRNA levels showed significant correlation to the uptake ratio and net active uptake of Gly-Sar. MRP2 and OATP-B showed no correlation to BSP transport parameters. Heterogeneity in transporter activity may thus be due to differences in transporter expression as shown for PepT1 and MDR1 which in turn is determined by the culture conditions. Absolute expression of genes was variable indicating that small differences in culture conditions have a significant impact on gene expression, although the overall expression patterns were similar.

The most commonly used method to assess intestinal permeability is the measurement of compound flux across a Caco-2 cells monolayer by using Hanks balanced salt solution (HBSS)-like buffers. Nevertheless, lipophilic acid drugs are poorly or not at all soluble in these types of buffers and their adsorption on the transwell plate is commonly observed. To reduce adsorption and increase solubility, permeability assays need to be developed in conditions other than classic conditions for lipophilic compounds. The best model to increase recovery of lipophilic compounds was determined as fasted state simulated intestinal fluid (FaSSIF) in the apical compartment and HBSS with 1% bovine serum albumin (BSA) in basolateral compartment. This model allows a correlation between absorption on Caco-2 cells and absorbed fraction in humans. For 35 compounds, only 2 outliers were observed in the Caco-2 assay using the FaSSIF model. These two outliers were the same outlier compounds as those observed with a classic Caco-2 method. Furthermore, a permeability assay of Pgp substrates evidenced efflux transport in both models and addition of a Pgp inhibitor suppressed Pgp efflux transport. FaSSIF in the apical compartment and HBSS with 1% BSA in the basolateral compartment is the model of choice to predict in vivo absorption for lipophilic acid drugs.

International audience; Expression of drug transporters corresponds to a crucial parameter in intestinal Caco-2 cells widely used for investigating drug absorption. In order to characterize it in an accurate, reproducible and comparative manner, we analyzed mRNA levels of 19 influx and efflux drug transporters through real-time quantitative polymerase chain reaction assays combined with the use of a total RNA reference standard. Profiles of transporter expression were found to be significantly correlated in two independent Caco-2 cell clones and in human small intestine, which may support the use of Caco-2 cells for investigating intestinal drug transport. Several transporters were nevertheless quantitatively expressed at higher (MRP2, MRP3, MRP4, MRP5, MRP6, OATP-A, OATP-B, OCT1 and MCT1) or lower (BCRP) levels in Caco-2 cells comparatively to small intestine. Moreover, MDR1, MRP2, OATP-A and PEPT1 mRNA relative expression were found to differ in the two analyzed Caco-2 cell clones by at least a twofold factor, highlighting that some variations in transporter expression may occur in Caco-2 cells depending on cell origin, and therefore underlining the interest of carefully characterizing transporter levels in any Caco-2 cell clone before its use for drug transport assays.