UC Berkeley

In vitro cell culture models have proven to be essential in early drug discovery and drug development to predict in vivo drug responses. More specifically, in vitro models of the small intestine are crucial tools for the prediction of drug absorption. The Caco-2

monolayer transwell model has been widely employed to assess drug absorption across the intestine. However, it is now well-established that 3D in vitro models capture tissue-specifc architecture and interactions with the extracellular matrix and therefore better recapitulate the complex in vivo environment. We develop a novel Sacricial Micromolding technique to culture geometrically controlled

self-assembling lumenized and polarized Caco-2 spheroids in a three-dimensional matrix. Compared to other techniques, this Sacricial Micromolding technique allows for precise control of size, shape as well as location of spheroids within a gel matrix. In addition, it does

not expose the spheroids to unnecessary sheer stress that could alter the cellular phenotype. We further characterize the intestinal spheroids for correct polarization, barrier properties and changes in gene expression and transporter function compared to 2D monolayers. We report that the spheroids display reproducible intestinal features and functions that are more representative of the in vivo small intestine than the widely used 2D transwell model. We also show that Caco-2 cell maturation and differentiation into the intestinal epithelial phenotype occur faster in spheroids and that they are viable for a longer period of time.

To overcome the limitation of lacking access to the luminal space of these spheroids, we exploit our three-dimensional patterning expertise as well as the directed epithelial self-organization to develop a technique to non-intrusively incorporate polymeric microparticles into the reconstituted intestinal spheroids. These polymeric microparticles can act as carriers or sensors to instruct or characterize tissue biology for in vitro assays. We use this technique to further study how microparticles of different shape and size as well as different compositions affect the self-organization and lumenization process of these tissues. We also develop a novel pH sensitive microsensor that can measure the luminal pH of reconstituted epithelial microtissues. We found that spherical microparticles that are 30 μm in diameter do not affect the self-organization process and are efficient at localizing to the lumen. Interestingly, we found that adhesive microparticles like polystyrene perturbed the self-organization and lumen formation and led to tissue inversion. Finally, we used this finding to invert the polarity of the spheroids by culturing them around Matrigel beads allowing superficial access to the apical membrane and making our intestinal spheroid model more physiological.

These studies offer a novel approach for investigating luminal microenvironments and drug-delivery across epithelial barriers without affecting the self-organization of the tissues. We believe that using Sacricial Micromolding we developed a robust and reproducible in vitro intestinal model that could serve as a valuable system to expedite drug screening as well as to study intestinal transporter function.