Stephanie van Hoppe

141 The impact of OATPs on disposition and toxicity of antitumor drugs; insights from KO and humanized mice in wild-type mice, by both transgenic OATP1B1 and OATP1B3 (Iusuf et al., 2014). This indicates that the function of the removed Oatp1a/1b genes responsible for keeping liver Ces1 expression low could be taken over by both human OATP1B1 and human OATP1B3. Still, the exact mechanism underlying these Ces-regulatory functions of Oatp/OATP1 proteins is as yet unclear. An obvious possibility seems some detoxifying function of these proteins, normally involved in removing one or more Ces-inducing compounds from the body. Such compounds might be of endogenous or exogenous origin (dietary, or metabolites or even primary products formed by the intestinal microflora), or possibly both. Their prolonged retention in the body might increase the overall liver exposure and thus Ces induction in the knockout strain. Further support for the involvement of a detoxifying role of the Oatp1a/1b proteins in Ces gene expression comes from the finding that the knockout of a number of other broad- specificity detoxifying proteins in FVB mice, including the multidrug efflux transporters Abcb1a/1b and Abcg2, and the multidrug metabolizing Cyp3a complex, cause similar levels of upregulation of hepatic Ces1b, Ces1c, Ces1d, and Ces1e expression (Lagas et al., 2012; Tang et al., 2015; Tang et al., 2014). Alternatively, a function of the OATPs in liver uptake of Ces-repressing compounds is also a possibility. If these compounds then get efficiently cleared through alternative routes, maybe renal or metabolic, the effective liver exposure to these compounds might be reduced in the Oatp1a/1b knockout mice. However, for the moment these options remain speculative. 2 . 2 . 2 . I r i no te can s t ud i e s i n OAT P1B1 and OAT P1B3 human i ze d mi ce Whatever the mechanism of Ces upregulation, the increased plasma levels of Ces1c in Oatp1a/1b(-/-) mice might confound interpretation of the irinotecan and SN-38 disposition and toxicity results in these mice. The increased conversion of irinotecan to SN-38 could reduce plasma irinotecan levels, and increase SN-38 levels independent of direct Oatp-mediated transport. For this reason, among others, i.v. irinotecan studies were also performed in the OATP1B1 and OATP1B3 humanized mice, as these have a mostly normalized plasma Ces1c activity. The results obtained for irinotecan and SN-38 were strikingly different: the humanizedmice showed higher plasma levels of irinotecan compared to wild-type mice, but similar liver concentrations, resulting in markedly reduced liver-to-plasma ratios of irinotecan. As these differences cannot be attributed to Ces upregulation, the results suggest that the transgenic OATP1B1 and OATP1B3 are far less efficient (if at all) in taking up irinotecan into the liver than the mouse Oatp1a/1b proteins (which are absent from the humanized mice). This also explains the higher plasma levels of irinotecan in the humanized mice compared to wild-type mice. In contrast, for SN-38 after i.v. irinotecan administration we found that the plasma levels in the transgenic mice were markedly reduced compared to Oatp1a/1b(-/-) mice,