(137e) Endothelial Cell Communication During Vasculogenesis in Vitro | AIChE

(137e) Endothelial Cell Communication During Vasculogenesis in Vitro

Authors 

Diaz-Santana, A. - Presenter, Cornell University
Cross, V. - Presenter, Cornell University
Zheng, Y. - Presenter, Cornell University
Choi, N. W. - Presenter, Cornell University
Wu, M. - Presenter, Cornell University
Stroock, A. D. - Presenter, Cornell University


Vasculogenesis ? the self-organization of vascular networks ? represents an important step in development in vivo and could provide a basis for the functional vascularization of tissue scaffolds for tissue engineering in vitro. A characteristic of self-organizing systems is that the behavior of individual units dictates global organization. Gaining an understanding of the cellular dynamics, cell-cell communication, and driving factors that lead to cell-cell interactions would represent a significant step in manipulating vascularization of tissue scaffolds. Models on two-dimensional substrates have been used to understand the role of particular biochemical factors and mechanics in vasculogenesis. However, these systems fail to recapitulate both the three-dimensional (3D), soft environments that cells experience in many tissues and the formation of open lumens. In 3D environments, cells can use matrix fibers for support, migration and potentially for communication with other cells in their vicinity (tractional forces from cell motility). In addition, secretion of soluble factors (ie. growth factors) can establish local gradients that may induce cell communication at a distance. We aim to distinguish the prominence of these different potential modes of communication between human umbilical vein endothelial cells (HUVECs) undergoing vasculogenesis in 3D collagenous matrices. HUVECs seeded homogeneously in an isotropic matrix of type 1 collagen extend protrusions. When cells are sufficiently close to one another, these protrusions eventually form permanent connections between neighboring cells. We report that this protrusion activity, rather than cell migration, is the main mechanism for network formation in a 3D collagen scaffold. In addition, the random search of isolated cells is biased by the presence of neighboring cells or impermeable, solid boundaries but not by permeable boundaries. These effects suggest that signaling via soluble chemical factors dominates the communication between the cells. We examine the possibility that, in vasculogenesis, this cell search is guided by local chemical gradients. In particular, we focus on the response to growth factors (GFs) and fibronectin (Fn), a matrix component that is localized by HUVECs and plays an important role in binding GFs. We begin with a brief presentation of our technical approach: To create a purely diffusive linear gradient of a soluble factor in a 3D environment, we use microfluidic systems whose lithographic features are defined in collagen at high mass fraction. A 3-channel structure provides a source, a sink, and a culture region (center channel loaded with HUVECs in collagen of low mass fraction). We characterize the mass transfer in this device and demonstrate the ability to create steady linear gradients of macromolecules and small molecules in the culture region. We proceed to describe our experiments: We work at low seeding densities of HUVECs to allow for the observation of isolated cells exposed to well-defined gradients. We perform time-lapse study of the protrusive and migratory activity of HUVECs. We characterize protrusive activity by performing statistical analysis on the angular frequency of protrusions in order to assess the presence or absence of bias and its correlation to the imposed gradients. Exploiting these methods, we report degree of bias as a function of the gradients and of the absolute concentration of vascular endothelial growth factor (VEGF) and Hedgehog, and as a function of exogenous Fn delivered both uniformly and in controlled gradients. We compare these observations to those for pairs of cells in the absence of global gradients and to a simple mathematical model of the mass transfer, binding kinetics, and cellular response. We outline our perspective on cellular communication in vasculogenesis that emerges from these studies. Finally, we conclude by pointing toward potential significance to in vivo vasculogenesis and to the engineering of functional microvascular structure in vitro.