Microfluidic Devices for Angiogenesis - Mechanical and Chemical Signaling in Angiogenesis - page 108

subjected to fluid shear stress of 1.2 Pa over the apical cell surface. It remains to
resolve this apparently contradictory behavior.
In these studies the effects of hemodynamic shear stress parallel to the mono-
layer surface was investigated. But in the vascular system, flows also occur across
the vascular endothelium (transendothelial flow) due to differences in hydrostatic
plus oncotic pressure between the blood and the interstitial space. These flows
across the arterial wall have been estimated to be on the order of 0.01 lm/s in
normal arteries, but could easily rise much higher either when endothelial cell
turnover occurs [ 95 ], as a result of vessel wall trauma, or in vascular beds that that
more permeable than normal, such as in the vascular networks that form in the
vicinity of malignant tumors [ 96 ].
4.3 Effects of Transendothelial Flow
Two groups have examined the effects of interstitial flow across an endothelial
monolayer on angiogenesis in microfluidic systems. One used a system in which a
gel was formed in a PDMS chamber, on one side of which an endothelial monolayer
was grown, across which flow could be directed from the apical surface into the gel
[ 97 ]. These experiments showed that flow enhances the rate of vascular ingrowth by
a Src-mediated process at mean flow velocities ranging up to 50 lm/min. Vascular
sprouts appeared to emerge from localized islands of activated Src. More recent
studies, however, raise questions about this observation. Using a microfluidic
system containing a gel positioned between two media channels so that flow could
be directed either in the apical-to-basal or basal-to-apical direction, Song and Munn
observed a different response [ 98 ]. Whether or not shear stress was applied to the
apical surface, and with or without a VEGF gradient, they found that apical-to-basal
flow tended to reduce the tendency for angiogenesis as compared to the case with no
flow, or with flow from the basal-to-apical side. The mechanisms or signaling
pathways activated by this process were not investigated, and this remains an
interesting area for further investigation. It is worth nothing, however, that in other
studies interstitial flow has been shown to elicit a biological response from smooth
muscle cells and fibroblasts [ 95 ], and that this effect could be inhibited by the
introduction of heparanase, presumably removing the cell-secreted glycocalyx in
these collagen I gels. In addition, transendothelial flow has been shown to cause a
reduction in monolayer hydraulic conductivity and protein permeability [ 99 ]. The
link to angiogenesis is that sprouting is often accompanied by a loosening of the
cell-cell junctions, which would also cause an increase in permeability.
In a separate study [ 35 ] it was observed that interstitial flows of *90 lm/min past
endothelial cells seeded inside the gel helped to promote the formation of an inter-
connected network, and also induced the cells to form a monolayer on the upstream face
of the gel region. These results, taken together, suggest that flows of a magnitude that
would be expected in vivo, either parallel to or perpendicular to an endothelial mono-
layer, can have a dramatic influence on angiogenesis. Studies are just now appearing in