Koserstr. 20
14195 Berlin
+49 30 838 75784
anatomie@vetmed.fu-berlin.de
Question
Publications utilizing cell culture models have risen exponentially within the last two decades, driven in part by the desire to reduce animal testing but also by the possibility of understanding more about the formation of tissues and organs. The problems involved in obtaining viable tissues for studies of Epithelial transport are pronounced. In particular when studying the rumen of cattle and sheep, problems arise in obtaining the animals and extracting viable tissues. There is thus a need for establishing and characterizing a ruminal epithelial model. However, the generation of such a model may suffer from the difficulty of separating epithelial cells from the underlying structures, i.e. the ruminal lamina propria mucosae with blood and lymphatic vessels and particularly active fibroblasts. It is therefore an additional aim of the present study to asses to what extent such cultures are contaminated by fibroblasts.
Methods
Performed essentially as previously described [1, 2], papillae of the ventral sac of sheep rumen were removed and keratinocytes isolated via stepwise trypsinization, discarding the first two fractions and then cultivating the subsequent five fractions. The cells were characterized immunohistochemically using antibodies against fibronectin, cytokeratin AE1 and AE3, and MNF. Fibroblasts from a previous isolation from the bovine claw were stained in parallel. After ~3 passages for expansion, keratinocytes were seeded on filters (polycarbonate membrane, pore size 0.4 µm). The model epithelia were investigated via measurement of the transepithelial electrical resistance and transmission electron microscopy.
Results
Approximately 6 days after isolation of cells from ruminal papillae, two types of cells could be visually identified: round and spindle shaped cells. Both types of cells showed staining for the cytokeratin markers and for fibronectin. Conversely, control fibroblasts only stained for fibronectin, but not for the cytokeratin markers. Cells grown 12 days on filters developed resistances of 400 to 900 Ω ∙ cm2. Transmission electron microscopy showed a multilayered structure with cell-cell contacts similar to the tight junctions seen in native ruminal epithelium of sheep.
Conclusions
We confirm that it is possible to generate multi-layered model epithelia with barrier properties that indicate functional cellular tight junctions from cells isolated from the ruminal epithelium. While contamination by fibroblasts seemed to be minor, development of further methods is required for the reliable differentiation of fibroblasts from keratinocytes.