This Frontiers review analyzes the rapidly growing microfluidic strategies that have been employed in attempts to create physio relevant 'organ-on-chip' models using primary tissue removed from a body (human or animal). Tissue harvested immediately from an organism, and cultured under artificial conditions is referred to as ex vivo tissue. The use of primary (organotypic) tissue offers unique benefits over traditional cell culture experiments, and microfluidic technology can be used to further exploit these advantages. Defining the utility of particular models, determining necessary constituents for acceptable modeling of in vivo physiology, and describing the role of microfluidic systems in tissue modeling processes is paramount to the future of organotypic models ex vivo. Virtually all tissues within the body are characterized by a large diversity of cellular composition, morphology, and blood supply (e.g., nutrient needs including oxygen). Microfluidic technology can provide a means to help maintain tissue in more physiologically relevant environments, for tissue relevant time-frames (e.g., matching the natural rates of cell turnover), and at in vivo oxygen tensions that can be controlled within modern microfluidic culture systems. Models for ex vivo tissues continue to emerge and grow in efficacy as mimics of in vivo physiology. This review addresses developments in microfluidic devices for the study of tissues ex vivo that can serve as an important bridge to translational value.

A microfluidic organotypic device that maintains mouse colon explants for up to 72 h in a physiologically relevant environment is reported. The device is easy to assemble and maintains physiologically accurate oxygen concentrations across the tissue.

Organotypic tissue slices provide seminatural, three-dimensional microenvironments for use in ex vivo study of specific organs and have advanced investigative capabilities compared with isolated cell cultures. Several characteristics of the gastrointestinal tract have made in vitro models for studying the intestine challenging, such as maintaining the intricate structure of microvilli, the intrinsic enteric nervous system, Peyer's patches, the microbiome, and the active contraction of gut muscles. In the present study, an organotypic intestinal slice model was developed that allows for functional investigation across regions of the intestine. Intestinal tissue slices were maintained ex vivo for several days in a physiologically relevant environment that preserved normal enterocyte structure, intact and proliferating crypt cells, submucosal organization, and muscle wall composure. Cell death was measured by a membrane-impermeable DNA binding indicator, ethidium homodimer, and less than 5% of cells were labeled in all regions of the villi and crypt epithelia at 24 h ex vivo. This tissue slice model demonstrated intact myenteric and submucosal neuronal plexuses and functional interstitial cells of Cajal to the extent that nonstimulated, segmental contractions occurred for up to 48 h ex vivo. To detect changes in physiological responses, slices were also assessed for segmental contractions in the presence and absence of antibiotic treatment, which resulted in slices with lesser or greater amounts of commensal bacteria, respectively. Segmental contractions were significantly greater in slices without antibiotics and increased native microbiota. This model renders mechanisms of neuroimmune-microbiome interactions in a complex gut environment available to direct observation and controlled perturbation.