Confounding effect of UGT1A1 interindividual variability on inhibition of its in vitro activity by kinase inhibitors.

Abstract

Background and ObjectiveInhibition of UGT1A1 by kinase inhibitors (KIs) has been shown to cause mild to serve hyperbilirubinemia in cancer patients1. UGT1A1 is highly polymorphic and displays variable expression. The specific aims of this study were i) to stratify human liver microsomes (HLMs) based on UGT1A1 abundance, activity and genotype, ii) to screen 10 clinically used KIs for their UGT1A1 inhibition potential, and iii) to determine inhibition potential (IC50) of the top three potent KIs in high and low UGT1A1 expressing HLM samples.MethodsTargeted genotyping of liver tissue donors (N=445) was performed; and UGT1A1 protein abundance and activity in HLM samples were quantified using a validated proteomics method and using bilirubin or estradiol as probe substrates, respectively. The mono and di‐bilirubin glucuronides and estradiol 3‐β‐D‐glucuronide (E3βG) were monitored by LC‐MS/MS. Association analysis of UGT1A1 genotype with protein abundance or activity was performed. Screening of UGT1A1 inhibition by KIs (at 1 and 20 uM) was performed using a high UGT1A1 expressing (46.7 ± 4.6 pmol/mg) HLM sample. IC50 values of KIs for UGT1A1 were determined using above mentioned high and a low (10.1 ± 1.6 pmol/mg) expressing HLM samples.ResultsUGT1A1 abundance and activity exhibited age‐dependent maturation with 8‐fold increase from neonates to adults. UGT1A1 SNPs rs10929302, rs887829, rs111741722 and rs4124874, are associated with decreased UGT1A1 abundance and activity, whereas rs8175347 was only associated with decreased abundance. Glucuronidation of both substrates was inhibited by the KIs in a substrate‐independent manner in the following order (% inhibition of E3βG activity at 1 and 20 μM in parenthesis): dasatanib (46 and 91)> sorafenib (11 and 74)> temsirolimus (50 and 71)> sirolimus (0 and 63) > sunitinib (21 and 38) > nilotinib (22 and 34) > gefitinib (0 and 29) > pazopanib = imatinib = lapatinib (0). The top three KIs showed differential IC50 values for E3βG inhibition in high and low UGT1A1 expressers, i.e., dasatanib (2.0, and 4.0), sirolimus (IC50= 7.6, and 32.6), and sorafenib (0.95, and 18.0), respectively.DiscussionUGT1A1 abundance is highly variable with significant effect of age and genotype. Inter‐individual variability in UGT1A1 plays a confounding role in predicting drug‐endobiotic or drug‐drug interactions (DDIs) mediated by this enzyme. Identification of genetic or non‐genetic factors associated with UGT1A1 activity will allow better prediction of UGT1A1 mediated drug metabolism and toxicity.Support or Funding InformationThis work was primarily supported by the National Institutes of Health [Grant no. R01 HD081299]. NA and AB contributed equally to this work. This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.