Solved in vivo/in vitro discrepancy in methapyrilene induced gene alterations in rat liver and in collagen sandwich cultured primary rat hepatocytes

Recent studies have presented evidence that in vivo obtained gene expression data canbe used for carcinogen classification, for instance to differentiate between genotoxic andnon-genotoxic liver carcinogens. This led several groups to study RNA-expressionpatterns also in cultured hepatocytes. However, these experiments resulted in conflictingdata to the in vivo situation. An example are the genes Gsk3â and Myd116 that have beenreported to be deregulated by methapyrilene (MPy) in hepatocytes in vitro but not in ratliver in vivo . To address this discrepancy, we first optimized an in vitro system with culturedhepatocytes. Testing various culture conditions (collagen coated dishes vs. Matrigel vs.sandwich cultures) and medium additives, we finally decided to use an in vitro system withchemically defined serum free culture medium and hepatocytes between two collagenlayers (sandwich culture). Interestingly, the reason for the in vivo / in vitro discrepancy couldclearly be identified. MPy in rats has a very short half-life of approx. 2.8 h. Therefore, weobserved an induction of Gsk3â and Myd116 after 13 and 18 h that decreased to basallevels after 24 h. Obviously, MPy caused only a transient induction. However, in vitro apermanent induction was obtained due to the constant concentration of MPy in the culturemedium. In conclusion the reported difference between in vivo and in vitro data was aconsequence of pharmacokinetics and not of qualitatively different behaviour ofhepatocytes in vitro and in vivo .