P43 Adhesion of human bone marrow-derived mesenchymal stem cells to liver sinusoidal endothelium is both CD29 and CD44 dependant, whilst initial tethering of MSCs to endothelium involves CD29 but not CD44. (16th November 2010)

• Record Type: Journal Article
• Authors: Aldridge, V; Garg, A; Youster, J; Newsome, P
• Is Part Of: Gut. Volume 59(2010)Supplement 2
• Page Start: A28
• Page End: A28
• Publication Date: 2010-11-16
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gut.2010.223362.69
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: Whilst MSCs have been postulated to have a range of functions within the liver little is known about the factors regulating their migration, adhesion and localisation within liver tissue. Modulation of this process may have important therapeutic implications. Aim: To define the adhesion molecule profile on hMSCs and the molecular regulation of their interaction with liver sinusoidal endothelium. Method: Adhesion molecule expression of human bone marrow-derived mesenchymal stem cells (hBM-MSCs) was comprehensively profiled by flow cytometry before and after TNFα stimulation. Static adhesion assays were performed to assess binding to hBM-MSCs (±adhesion molecule blockade) to both extra-cellular matrix (ECM) proteins and sections of liver tissue. This was further assessed in flow assays over ECM and liver sinusoidal endothelial cells. MSC migration towards chemokine ligands was studied in Boyden assays. Results: hBM-MSCs expressed known markers (CD90/CD44/CD105) and demonstrated tri-lineage differentiation. Profiling of the adhesion molecules expressed by hBM-MSCs using flow cytometry found that they express high levels of integrin subunits (CD29 (β1), CD49c (α3), CD49d (α4)) and chemokine receptors CCR4 and CCR5. Furthermore, TNF stimulation increased expression of chemokine receptors including CCR4 and CCR5 by approximately 20%. In static binding asays, MSCs bound to immobilised ligands such as VCAM-1 and fibronectin, whose expression is increased in injured liver. In static liver tissue binding assays CD29 and CD44 blockade reduced binding to the liver sinusoids (30.9%±9 and 31%±7, respectively, both p<0.05), whilst only CD29 blockade significantly reduced binding to the parenchyma (36.2%±10 (p<0.05). Using a novel flow adhesion assay which re-creates venous blood flow in the liver sinusoids we observed tethering, but not rolling of hBM-MSCs when flowed over liver sinusoidal endothelium (LSEC). Interestingly we found that hBM-MSCs do not bind to LSEC under flow. Tethering of MSCs was ablated by blocking of CD29 on MSCs. Furthermore, using a modified stop-flow protocol we identified a role for CD29 and CD44 in firm adhesion of MSC to LSEC. hBM-MSCs migrated towards ligands for CCR4 and CXCR4 in Boyden chamber assays. Conclusion: Binding of hBM-MSCs to liver sinusoidal endothelium occurs via tethering and firm adhesion without prior rolling. Tethering is mediated by CD29, whilst firm adhesion of hBM-MSCs to LSEC involves CD29 and CD44 interactions. The ability of MSCs to migrate in response to CCR4 and CXCR4 ligands, along with studies showing upregulation of these ligands in injured liver tissue suggests that these receptors may be involved in controlling MSC migration to and/or within injured liver.