Alteration of pancreatic carcinoma and promyeloblastic cell adhesion in liver microvasculature by co-culture of hepatocytes, hepatic stellate cells and endothelial cells in a physiologically-relevant model.

In vitro models of the liver microvasculature, especially with respect to cancer cell extravasation, should include not only endothelial and cancer cells but also surrounding cells to mimic the physiological situation. To this end, in the present study, we established a physiologically-relevant hierarchical co-culture model by stacking layers of primary rat hepatocytes (Hep), hepatic stellate cells embedded in collagen gel (LX-2) and endothelial cells (HUVECs) on a specially designed oxygen-permeable polydimethylsiloxane PDMS bottom plate. The model was used to investigate the role and contribution of each of the three cell types in pancreatic cancer and promyeloblast cell adhesion. In particular, we showed an increase in albumin production by the primary hepatocytes and in the consumption of the produced vascular endothelial growth factors (VEGFs). Furthermore, in co-culture, the HUVECs exhibited a mature vascular endothelial and non-inflamed phenotype, as evidenced by Stabilin-1, lymphatic vessel endothelial hyaluronan receptor-1 (LYVE-1), intercellular adhesion molecule (ICAM-1), and vascular adhesion protein-1 (VAP-1) expression. The HUVECs were also successfully activated with an inflammatory cytokine and their ICAM-1 response was found to be higher in monoculture compared to co-culture. Additionally, the adhesion of MiaPaCa-2 pancreatic cancer cells and HL60 promyeloblasts was tested in both cases (i.e.: activation or not by an inflammatory cytokine). It has been found that their adhesion was always reduced in the co-culture model. These results highlight the importance of integrating hepatic stellate cells in the design of biomimetic models of the hepatic endothelial barrier.