(443d) Adipocyte-Preadipocyte Interactions in Adipose Tissue Development

Motivation: A better understanding of the mechanism of post-natal adipose tissue development is of paramount importance in devising new therapeutic strategies for the treatment and prevention of obesity and related diseases. In rodents, expansion of adipose tissue mass during adulthood not only results from increased adipocyte size (hypertrophy), but also from increased adipocyte number (hyperplasia). Adipocyte hyperplasia has been observed in high fat-fed rats, obese Zucker rats treated with troglitazone, transgenic mice overexpressing GLUT4 selectively in adipose tissue, and estrogen-receptor- knockout mice. The occurrence of adipocyte hyperplasia in humans remains controversial. Early observation by Hirsch and Batchelor in the 1970s have suggested that new adipocyte formation occurs in the adipose tissue once mature adipocytes have reached a critical size. This critical cell size hypothesis has been difficult to test directly, despite accumulating evidence that the microenvironment established by hypertrophic adipocytes plays an important role in the proliferation and differentiation of locally resident progenitor cells. One significant challenge has been establishing an appropriate in vitro model of adipose tissue development. To date, in vitro studies on adipose tissue development have relied on batch cultures of randomly mixed populations of preadipocytes and adipocytes. In this paper, we characterize a novel engineered adipose tissue model with spatially defined composition of preadipocytes and adipocytes.

Methods: A micro-fluidic culture chamber with internal posts was fabricated using standard soft-lithographic methods. The posts divided the device into a middle cell culture chamber and flanking medium channels. Spacing between posts was varied from 10 to 100 um. The width of the culture chamber was varied from 0.4 to 3 mm. Steady medium flow through the flanking channels was driven by gravity. Velocity and concentration profiles through the device were modeled using coupled convection-diffusion equations and simulated with the COMSOL software package. For flow characterization experiments, the culture chamber was filled with a collagen gel through injection of the uncrossed collagen solution followed by a temperature shift. For differentiation experiments, 3T3-L1 preadipocytes were injected into collagen coated culture chamber as suspensions. The preadipocytes were grown to confluence and then differentiated using an adipogenic hormone cocktail. For selective differentiation of a subset of the preadipocytes, a differentiation cocktail gradient was established across the culture chamber by varying the medium compositions in the two flanking flow channels.

Results: Flow visualization experiments with a colored dye solution demonstrated that a stable chemical gradient could be established within the micro-fluidic device across the cell culture chamber. The observed shape of the gradient was in excellent agreement with simulation results. We next determined whether the gradient-generating culture devices could support 3T3-L1 preadipocyte expansion and differentiation into mature adipocytes. After 16 days of culture, during which both medium channels were infused with an adipogenic hormone cocktail, more than 90% of the cells contained visible lipid droplets. These cells also exhibited the characteristic round morphology of terminally differentiated adipocytes. The next set of experiments investigated whether a differentiation cocktail gradient could be utilized to generate defined zones of preadipocytes and adipocytes within the single culture chamber. After 16 days, during which only one medium channel was infused with the hormone cocktail, the culture underwent asymmetrical differentiation. Phase contrast microscopy and AdipoRed staining showed a well-defined gradient of differentiated adipocytes and lipid droplet sizes across the culture chamber, demonstrating proof-of-concept for chemical gradient-based selective differentiation. On-going experiments address the role of cell size in the recruitment, proliferation, and differentiation of new adipocytes by characterizing the microenvironment of hypertrophic adipocytes. In these experiments, co-cultures of preadipocytes and adipocytes with different compositions and contacting patterns were generated by varying the shape of the differentiation cocktail gradient. Prospectively, the in vitro model presented in this paper could enables studies on the role of cell-cell interactions in adult adipose tissue expansion.