01921nas a2200301   4500000000100000000000100001008004100002260001200043653002700055653002400082653001200106100002000118700001700138700001700155700001500172700001900187700002600206700001200232700002200244700003100266700001600297245011700313856005500430300001200485490000600497520110200503022001401605       2025                            d  c2025-0410aBiomedical Engineering10aNeurological models10aTissues1 aJames G. Shamul1 aZhiyuan Wang1 aHyeyeon Gong1 aWenquan Ou1 aAlisa M. White1 aDiogo P. Moniz-Garcia1 aShuo Gu1 aAlisa Morss Clyne1 aAlfredo Quiñones-Hinojosa1 aXiaoming He00aMeta-analysis of the make-up and properties of in vitro models of the healthy and diseased blood–brain barrier  uhttps://www.nature.com/articles/s41551-024-01250-2  a566-5980 v93 aIn vitro models of the human blood–brain barrier (BBB) are increasingly used to develop therapeutics that can cross the BBB for treating diseases of the central nervous system. Here we report a meta-analysis of the make-up and properties of transwell and microfluidic models of the healthy BBB and of BBBs in glioblastoma, Alzheimer’s disease, Parkinson’s disease and inflammatory diseases. We found that the type of model, the culture method (static or dynamic), the cell types and cell ratios, and the biomaterials employed as extracellular matrix are all crucial to recapitulate the low permeability and high expression of tight-junction proteins of the BBB, and to obtain high trans-endothelial electrical resistance. Specifically, for models of the healthy BBB, the inclusion of endothelial cells and pericytes as well as physiological shear stresses (~10–20 dyne cm–2) are necessary, and when astrocytes are added, astrocytes or pericytes should outnumber endothelial cells. We expect this meta-analysis to facilitate the design of increasingly physiological models of the BBB.  a2157-846X