Recapitulating Transport Properties of the Human Blood Brain Barrier in an Isogenic Microphysiological Model

<u>Introduction</u>. There has been a recent and dramatic interest in the development of models capable of recapitulating the unique characteristics of the human blood-brain barrier. This comes both from the development of several new antibody drugs for neurodegenerative diseases such as Alzheimer’s disease and from the desire to utilize the membrane-bound transporters unique to the brain endothelium for targeted delivery of drugs to treat cerebral diseases such as glioblastoma. To meet the need to understand the mechanisms of transport and use that understanding to develop new drugs and modes of delivery to the brain, researchers and the pharmaceutical and biotech industries have accelerated efforts to develop in vitro models of the BBB. Numerous models now exist ranging from 2D monolayer systems with a single (endothelial) cell type to more complex 3D models consisting of multiple cell types felt to be instrumental in determining the transport properties of the BBB. Among the approaches used in the complex model, ones that utilize the natural self-assembly properties of the BBB cell types (endothelial, pericyte, and astrocyte) appear capable of growing in vitro models with in vivo-like morphology and function that closely parallel those of the human brain.<br/>While these models have demonstrated their potential, they suffer from relatively high variability due both to the natural biological diversity of the self-organized vascular networks and to the lack of a consistent, long-term cell source. This has led to several groups seeking to develop models that can be generated entirely from induced pluripotent cells that have the combined advantages of being derivable from patient-specific cells and, in principle, constitute an inexhaustible supply thereby representing a long-term consistent source for models. <br/> <br/><u>Methods and Results</u>. Here we report on one such model that combines iPSC-derived endothelial cells, pericytes and astrocytes from a single cell line (Alstem iPS11) widely available to both the research community and industry. Published protocols are used to derive pericytes and astrocytes, and a protocol for endothelial cells using inducible ETV2 has been adapted from prior publication [ref]. Using methods developed in our lab for generating a 3D BBB model [Hajal, et al., Nat Prot, 2022], the tri-ulture system is developed in a perfusable microfluidic platform using cell ratios that have been optimized to closely match those found in the human brain. Both the morphology (characterized by quantitative analysis of the network and interactions between the different cell types) and function (reflected by vascular permeability) were quantitatively assessed for direct comparisons to in vivo data. Transcriptional analysis and immunohistochemistry are used to determine the expression levels of key brain-specific markers, including the various transporter proteins found in the endothelial cells, and compared to human cell lines. In particular, genes associated with brain endothelial phenotype appear to be highly expressed whereas there is an absence of those genes that tend to be expressed in cells with an epithelial phenotype. Vascular permeability for a 10 kDa dextran (2.0 x 10^-7 cm/sec) matches in vivo measurements in rats. High resolution 3D imaging and multi-photon metabolic imaging demonstrate tight endothelial cell junctions marked by ZO-1 and claudin-5 along with a uniform distribution of vasculature throughout the full 500 mm height of the vascularized gel region.<br/> <br/><u>Conclusions</u>. An isogenic model of the human blood-brain barrier can be produced from a tri-culture of endothelial cells, pericytes and astrocytes with physiological morphology and barrier function similar to in vivo. This system offers long-term consistency for use in models of neurovascular diseases such as cerebral amyloid angiopathy and for the screening of molecular therapeutics that utilize transporter proteins specific to the brain.