Fluids and Barriers of the CNS (Feb 2018)

Functional brain-specific microvessels from iPSC-derived human brain microvascular endothelial cells: the role of matrix composition on monolayer formation

  • Moriah E. Katt,
  • Raleigh M. Linville,
  • Lakyn N. Mayo,
  • Zinnia S. Xu,
  • Peter C. Searson

DOI
https://doi.org/10.1186/s12987-018-0092-7
Journal volume & issue
Vol. 15, no. 1
pp. 1 – 12

Abstract

Read online

Abstract Background Transwell-based models of the blood–brain barrier (BBB) incorporating monolayers of human brain microvascular endothelial cells (dhBMECs) derived from induced pluripotent stem cells show many of the key features of the BBB, including expression of transporters and efflux pumps, expression of tight junction proteins, and physiological values of transendothelial electrical resistance. The fabrication of 3D BBB models using dhBMECs has so far been unsuccessful due to the poor adhesion and survival of these cells on matrix materials commonly used in tissue engineering. Methods To address this issue, we systematically screened a wide range of matrix materials (collagen I, hyaluronic acid, and fibrin), compositions (laminin/entactin), protein coatings (fibronectin, laminin, collagen IV, perlecan, and agrin), and soluble factors (ROCK inhibitor and cyclic adenosine monophosphate) in 2D culture to assess cell adhesion, spreading, and barrier function. Results Cell coverage increased with stiffness of collagen I gels coated with collagen IV and fibronectin. On 7 mg mL−1 collagen I gels coated with basement membrane proteins (fibronectin, collagen IV, and laminin), cell coverage was high but did not reliably reach confluence. The transendothelial electrical resistance (TEER) on collagen I gels coated with basement membrane proteins was lower than on coated transwell membranes. Agrin, a heparin sulfate proteoglycan found in basement membranes of the brain, promoted monolayer formation but resulted in a significant decrease in transendothelial electrical resistance (TEER). However, the addition of ROCK inhibitor, cAMP, or cross-linking the gels to increase stiffness, resulted in a significant improvement of TEER values and enabled the formation of confluent monolayers. Conclusions Having identified matrix compositions that promote monolayer formation and barrier function, we successfully fabricated dhBMEC microvessels in cross-linked collagen I gels coated with fibronectin and collagen IV, and treated with ROCK inhibitor and cAMP. We measured apparent permeability values for Lucifer yellow, comparable to values obtained in the transwell assay. During these experiments we observed no focal leaks, suggesting the formation of tight junctions that effectively block paracellular transport.

Keywords