Franny Jongbloed

68 CHAPTER 3 an upregulation of pathways involved in growth inhibition, tumor suppression and cell death was seen. Strikingly, the response of fasted tumor tissue appeared partially oppositely regulated as compared to fasted liver tissue since upregulation of various cell cycle, apoptosis and cytokine pathways in fasted tumor tissue was predominantly found. These results are partially in line with the recently described DSS 8,10,12 . In addition, the decreased activation of Hsp90 in tumor tissue of fasted mice after irinotecan administration might point towards an additional inhibitory effect of irinotecan after fasting 8,11,12 .Thus, although the transcriptomic response of fasted tumor tissue differs significantly from that of AL tumor tissue, this does not affect the antitumor activity of irinotecan 7,10 . Due to the heterogeneity within the tumor groups, any common pathway responses should be interpreted with care as more studies are needed to fully discern between partly orchestrated -common and heterogenic- individual responses to chemotherapeutic insults in tumor tissue after dietary intervention. Previously, we found decreased levels of the active metabolite of irinotecan, SN-38, in plasma as well as liver, but not in tumor tissue of fasted mice 10 . The relative upregulation of the carboxylesterases in this study would suggest an increased hydrolysis of irinotecan into SN-38 7,24 . Upregulation of Ugt1a1 and ABC transporter Abcc1 is indicative of increased transport of SN-38 out of the cell 25-27 . An increased production of SN-38 in the liver and more efficient excretion by transporters may explain the lowered SN-38 levels in the liver and increased plasma levels we found previously 10 . Ugt1a1 is involved in conversion of SN-38 into its inactive from SN-38G, which reduces the levels of the active drug in the intestine. Since one of the main side effects of irinotecan, diarrhea, is thought to be due to high concentrations of intra-luminal SN-38, this inactivation of SN-38 by Ugt1a1 could be responsible for the absence of diarrhea in fasted mice 28-30 . Further studies should examine in depth the activity of the enzymes involved in irinotecan drug metabolism and at various time points after exposure in order to understand their role in the reduced side effects induced by fasting. The changes occurring in the drug metabolism of irinotecan as seen in liver tissue were far less prominent in tumor tissue without obvious changes due to fasting. This is consistent with the previous findings that irinotecan and its metabolite levels were not affected by fasting in tumor tissue. The inhibition of mouse double minute 2 homolog ( Mdm2 ) in the tumor fasting vehicle group is partially maintained in the fasting irinotecan group, which may point towards an activation of tumor suppressor p53 by fasting 31 . Further research is needed in order to understand the etiology of fasting on the effects of chemotherapy on tumor tissue. Since pharmacokinetic studies showed that the half-life of SN-38 is approximately 12 hours 10 , we choose to explore the transcriptional changes at one time point, using one dosage of irinotecan in the therapeutic window in which the effects of both irinotecan and fasting would be maximal. Analyses of multiple time points and with different dosages of irinotecan