Border control at the vessel wall: The endothelium as customs official for leukocyte crossing (LSBR 1649)
Project leader: Prof. Dr. Jaap D. van Buul, Sanquin Research
PhD student: Bram van Steen (Sept. 2017 – Sept. 2021)

A significant development in this project was the vessel on a chip. This has resulted in several fruitful collaborations, not only national but also international. The work was published in J Cell Science (, with an additional editorial interview with the shared-first two authors. Using this technology, we have identified a hitherto undiscovered migratory signaling pathway for neutrophils that are deficient for the actin-regulating ArpC1B. Thus, using new technology, new pathways are identified that could potentially lead to new therapeutic interventions.

In addition, we identified the full ICAM-1 adhesome: the ICAM-1 adhesome contains all proteins that could potentially be recruited to ICAM-1 upon clustering. Clustering occurs under inflammatory conditions induced by leukocytes that attach to the endothelium via beta2-integrins. We used anti-ICAM-1 antibody-coated beads to mimic this interaction and used mass spectrometry to identify all potential interactors. Out of all the hits, we focused on two candidates SNAP23 and CD44.
SNAP23 is involved in the transcellular migration of CD8 T lymphocytes across the endothelium. This means that these cells go straight through the endothelium and do not use the so- called paracellular pathway, i.e., the cell-cell contacts. The effector T cells are activated by local recruited specific set of chemokines that prompt the T cell to migrate, not in horizontal movement but in vertical movement, thus leading to migration right through the endothelium, i.e., transcellular migration. This work has been published in Cell Reports (
Thanks to the ICAM-1 adhesome project, CD44, a glycoprotein known to interact with sugar structures such as the glycocalyx on the vessel wall, was found to be recruited by ICAM-1 after clustering. Reducing the expression of CD44 using knockdown techniques resulted in a dysfunctional ICAM-1 molecule, which was no longer able to efficiently bind leukocyte integrins and therefore was unable to support transendothelial migration of leukocytes. In addition, we were able to show that the chemokine presentation by hyaluronan, one of the components of the glycocalyx, was impaired in the absence of CD44. This study sheds new light on the role of CD44 and the glycocalyx in transendothelial migration of leukocytes.

Finally, the work from this project has led to a re-evaluation of how the endothelial cells are connected. We found that endothelial cells partially overlap at junctional regions, i.e., the plasma membrane extends beyond the VE-cadherin boundaries. This phenotype was confirmed in vivo using confetti knock-in endothelium-specific mouse models, giving each endothelial cell a different color. This allowed us to identify the endothelial membrane overlaps. We discovered these overlaps in the liver, lungs, and skin. Further analysis showed that these overlaps were characterized by strong PECAM-1/CD31 staining rather than VE-cadherin. Moreover, these overlaps were transmigration hotspots for neutrophils: the greater the overlap, the more neutrophils preferred such sites to cross the endothelium. Using advanced real-time microscopy, i.e., lattice light-sheet microscopy in close collaboration with the Janelia Research Institute in Ashburn, USA, we found that the endothelium generated transmigration membrane tunnels to allow leukocytes to pass through.
This work demonstrates that the current dogma of how leukocytes cross the endothelium by temporarily disconnecting two adjacent endothelial cells may be incorrect or incomplete in each case and need to be modified. This will lead to a new concept and realization of how leukocytes manage to cross the endothelium while maximizing the integrity of the endothelial barrier. This work is being followed up by another PhD student in the group.