Influence of basal media composition on barrier fidelity within human pluripotent stem cell-derived blood-brain barrier models.

It is increasingly recognized that brain microvascular endothelial cells (BMECs), the principal component of the blood-brain barrier (BBB), are highly sensitive to soluble cues from both the bloodstream and the brain. This concept extends in vitro, where the extracellular milieu can also influence BBB properties in cultured cells. However, the extent to which baseline culture conditions can affect BBB properties in vitro remains unclear, which has implications for model variability and reproducibility, as well as downstream assessments of molecular transport and disease phenotypes. Here, we explore this concept by examining BBB properties within human-induced pluripotent stem cell (iPSC)-derived BMEC-like cells cultured under serum-free conditions in DMEM/F12 and Neurobasal media, which have fully defined compositions. We demonstrate notable differences in both passive and active BBB properties as a function of basal media composition. Further, RNA sequencing and phosphoproteome analyses revealed alterations to various signaling pathways in response to basal media differences. Overall, our results demonstrate that baseline culture conditions can have a profound influence on the performance of in vitro BBB models, and these effects should be considered when designing experiments that utilize such models for basic research and preclinical assays.

A Simplified, Fully Defined Differentiation Scheme for Producing Blood-Brain Barrier Endothelial Cells from Human iPSCs.

Human induced pluripotent stem cell (iPSC)-derived developmental lineages are key tools for in vitro mechanistic interrogations, drug discovery, and disease modeling. iPSCs have previously been differentiated to endothelial cells with blood-brain barrier (BBB) properties, as defined by high transendothelial electrical resistance (TEER), low passive permeability, and active transporter functions. Typical protocols use undefined components, which impart unacceptable variability on the differentiation process. We demonstrate that replacement of serum with fully defined components, from common medium supplements to a simple mixture of insulin, transferrin, and selenium, yields BBB endothelium with TEER in the range of 2,000-8,000 Ω × cm2 across multiple iPSC lines, with appropriate marker expression and active transporters. The use of a fully defined medium vastly improves the consistency of differentiation, and co-culture of BBB endothelium with iPSC-derived astrocytes produces a robust in vitro neurovascular model. This defined differentiation scheme should broadly enable the use of human BBB endothelium for diverse applications.