Recent therapeutic success of large-molecule biologics has led to intense interest in assays to measure with precision their transport across the vascular endothelium and into the target tissue. Most current in vitro endothelial models show unrealistically large permeability coefficients due to a non-physiological paracellular transport. Thus, more advanced systems are required to better recapitulate and discern the important contribution of transcellular transport (transcytosis), particularly of pharmaceutically-relevant proteins. Here, a robust platform technology for the measurement of transport through a human endothelium is presented, which utilizes in vitro microvascular networks (MVNs). The self-assembled MVNs recapitulate the morphology and junctional complexity of in vivo capillaries, and express key endothelial vesicular transport proteins. This results in measured permeabilities to large molecules comparable to those observed in vivo, which are orders of magnitude lower than those measured in transwells. The permeability of albumin and immunoglobulin G (IgG), biopharmaceutically-relevant proteins, is shown to occur primarily via transcytosis, with passage of IgG regulated by the receptor FcRn. The physiological relevance of the MVNs make it a valuable tool to assess the distribution of biopharmaceuticals into tissues, and may be used to prioritize candidate molecules from this increasingly important class of therapeutics.

The blood-brain barrier (BBB) greatly restricts the entry of biological and engineered therapeutic molecules into the brain. Due to challenges in translating results from animal models to the clinic, relevant in vitro human BBB models are needed to assess pathophysiological molecular transport mechanisms and enable the design of targeted therapies for neurological disorders. This protocol describes an in vitro model of the human BBB self-assembled within microfluidic devices from stem-cell-derived or primary brain endothelial cells, and primary brain pericytes and astrocytes. This protocol requires 1.5 d for device fabrication, 7 d for device culture and up to 5 d for downstream imaging, protein and gene expression analyses. Methodologies to measure the permeability of any molecule in the BBB model, which take 30 min per device, are also included. Compared with standard 2D assays, the BBB model features relevant cellular organization and morphological characteristics, as well as values of molecular permeability within the range expected in vivo. These properties, coupled with a functional brain endothelial expression profile and the capability to easily test several repeats with low reagent consumption, make this BBB model highly suitable for widespread use in academic and industrial laboratories.

Fibrils composed of tau protein are a pathological hallmark of several neurodegenerative disorders including Alzheimer’s disease (AD). Here we show that when recombinant tau protein is seeded with paired helical filaments (PHFs) isolated from AD brain, the amyloid formed shares many of the structural features of AD PHFs. In contrast, tau amyloids formed with heparin as an inducing agent—a common biochemical model of tau misfolding—are structurally distinct from brain-derived PHFs. Using ultrastructural analysis by electron microscopy, circular dichroism, and chemical denaturation, we found that AD seeded recombinant tau fibrils were not significantly different than tau fibrils isolated from AD brain tissue. Tau fibrils produced by incubating recombinant tau with heparin had significantly narrower fibrils with a longer periodicity, higher chemical stability, and distinct secondary structure compared to AD PHFs. The addition of heparin to the reaction of recombinant tau and AD PHFs also corrupted the templating process, resulting in a mixture of fibril conformations. Our results suggest that AD-isolated PHFs act as a conformational template for the formation of recombinant tau fibrils. Therefore, the use of AD PHFs as seeds to stimulate recombinant tau amyloid formation produces synthetic tau fibers that that closely resemble those associated with AD pathology and provides a biochemical model of tau misfolding that may be of improved utility for structural studies and drug screening. These results also demonstrate that posttranslational modifications such as phosphorylation are not a prerequisite for the propagation of the tau fibril conformation found in AD.