(430b) A Micro-Fluidic Adipose Reactor for 3-D Localization of Heterogeneous Cell Components

Authors: 
Wood, A. M. - Presenter, Tufts University
Lai, N. - Presenter, Tufts University
Shi, H. - Presenter, Tufts University
Schmidtke, D. - Presenter, University of Oklahoma


The differentiation of preadipocytes into adipocytes, or adipogenesis, is a key component of the hyperplastic growth of adipose tissue. Excessive expansion of adipose tissue mass, or adiposity, is the principal driver for obesity and its related diseases, including type 2 diabetes. In vitro studies of adipogenesis have traditionally relied on chemical induction of primary or transformed preadipocytes in 2-D culture settings. A general limitation of these conventional cultures is that they fail to capture tissue-inherent features, such as 3-D cell-cell and cell-extracellular matrix (ECM) interactions, which molecular evidences point to as critical to the regulation of adipogenesis in vivo. In particular, questions remain over the roles of the numerous soluble factors secreted by mature adipocytes in the recruitment and differentiation of locally present preadipocytes. A number of these factors, including adiponectin and resistin, have been recognized as important metabolic hormones and potential therapeutic targets for obesity. Very recently, a small number of in vitro adipose tissue model systems have been described that utilize synthetic polymers as cell culture scaffolds. However, adipogenic induction has remained spatially uncontrolled, leading to mixed co-cultures, and hampering detailed investigations on preadipocyte-adipocyte communication.

Here, we describe the design and implementation of a novel, micro-fluidic reactor that supports a spatially defined, 3-D co-culture of preadipocytes and adipocytes. In prior work, we had shown that parallel streams of low Re number flows could be used to establish a stable chemical solution gradient across a culture volume. In this work, we incorporated the chemical solution gradient feature into a ladder chamber to afford micron-scale localization of chemical triggers for adipogenic differentiation. The chambers were fabricated using standard soft lithography and rapid prototyping methods. Localization of preadipocytes to the bridge channels of the ladder chamber were achieved by selective delivery and subsequent cross-linking of cell suspensions in a natural ECM (collagen) pre-polymer solution.

Bright-field microscopy confirmed the presence of stable gradients across the bridge channels following the infusion of a dye solution in one of the two main flow channels. The transient and steady-state shapes of the gradients were functions of both the infusion rates and the relative dimensions of the main and bridge channels, which were varied from 1 ? 2:0.27 ? 30:5 for H:L:W, respectively. Compared to the conventional 2-D planar cultures, collagen entrapped 3-D cultures of adipocytes exhibited increased glucose consumption, lipid loading, and attained cell volumes comparable to mature adipocytes found in vivo. Furthermore, the 3-D ECM shielded the entrapped cells from direct exposure to fluid shear, extending the stability of the culture by a factor of two (2) compared to conventional cultures (from 2 to 4 weeks). Finally, differentiation of preadipocytes was selectively localized to the region proximal to the main flow channel carrying a standard differentiation cocktail. Prospectively, the spatial control achieved in our adipose reactor design should enable controlled studies on the roles of both direct (physical) and indirect (soluble factor mediated) cell-cell interactions in adipose tissue growth.