Organ-specific carboxylesterase profiling identifies the small intestine and kidney as major contributors of activation of the anticancer prodrug CPT-11

Abstract

The activation of the anticancer prodrug CPT-11, to its active metabolite SN-38, is primarily mediated by carboxylesterases (CE). In humans, three CEs have been identified, of which human liver CE (hCE1; CES1) and human intestinal CE (hiCE; CES2) demonstrate significant ability to hydrolyze the drug. However, while the kinetic parameters of CPT-11 hydrolysis have been measured, the actual contribution of each enzyme to activate the drug in biological samples has not been addressed. Hence, we have used a combination of specific CE inhibition and conventional chromatographic techniques to determine the amounts, and hydrolytic activity, of CEs present within human liver, kidney, intestinal and lung specimens. These studies confirm that hiCE demonstrates the most efficient kinetic parameters for CPT-11 activation, however, due to the high levels of hCE1 that are expressed in liver, the latter enzyme can contribute up to 50% of the total of drug hydrolysis in this tissue. Conversely, in human duodenum, jejunum, ileum and kidney, where hCE1 expression is very low, greater than 99% of the conversion of CPT-11 to SN-38 was mediated by hiCE. Furthermore, analysis of lung microsomal extracts indicated that CPT-11 activation was more proficient in samples obtained from smokers. Overall, our studies demonstrate that hCE1 plays a significant role in CPT-11 hydrolysis even though it is up to 100-fold less efficient at drug activation than hiCE, and that drug activation in the intestine and kidney are likely major contributors to SN-38 production in vivo.

Introduction

The anticancer drug CPT-11 (irinotecan, 7-ethyl-10-[4-(1-piperidino)-1-piperidino]carbonyloxycamptothecin), is a prodrug that is activated by esterases to yield SN-38 (7-ethyl-10-hydroxycamptothecin), a potent topoisomerase I poison [1]. The majority of biochemical studies have demonstrated that this is achieved by carboxylesterases (CE) [2], [3], [4], [5], [6], however butyrylcholinesterases (BChE) can also effect this process, albeit with poor efficiency [7], [8], [9], [10]. In humans, three CEs have so far been identified. The human liver CE, hCE1 (CES1), is predominantly expressed in the liver and demonstrates a preference for small, non-bulky substrates [11], [12], [13]. The human intestinal CE, hiCE (CES2), is expressed in the gut and the liver, and can hydrolyze much larger, more complex molecules. This is likely due to flexible domains present within the active site of the enzyme that allows for accommodation of these esters [14], [15]. The human brain CE, hBr3 (CES3), is believed to be expressed in the epithelia that form part of the blood brain barrier [16], although this protein has not been exhaustively tested for its substrate specificity [17]. However, all of these enzymes have been compared for their ability to activate CPT-11 [4], [5], [15], [17].

Results from these studies indicate that the hiCE is 64- to 100-fold more efficient than hCE1 at CPT-11 hydrolysis, with hBr3 being 20-fold poorer than the latter enzyme. Hence, due to the poor kinetic parameters for hBr3 with the drug (∼2000-fold less efficient than hiCE), and its very limited pattern of expression, it is unlikely that this CE significantly contributes to drug activation in vivo. Based upon this biochemical and enzyme kinetic evidence, we and others have assumed that hiCE would be the major esterase responsible for CPT-11 hydrolysis in cancer patients [4], [5], [17]. We hypothesized therefore that using selective hiCE inhibitors [18], [19], it would be possible to determine the amount of this enzyme present in biological samples using a simple substrate such as o-NPA. Simply, the difference in the enzyme activity assays in the presence and absence of the inhibitor should represent the amount of hiCE in the preparation. This could then be used as a measure of the ability of the sample to hydrolyze CPT-11. Such an approach would obviate the need for expensive and time consuming assays (HPLC with fluorescence detection) to monitor drug hydrolysis.

The studies described here sought to validate this approach by examining the ability of selective hiCE inhibitors [18], [19] to prevent the conversion of CPT-11 to SN-38 in a series of human microsomal samples. However, we were unable to dramatically reduce drug activation in these specimens using these specific inhibitors, suggesting that other proteins within the extracts could mediate the hydrolysis of CPT-11. Therefore, we have used a combination of chromatography and biochemical assays using CE inhibitors (both specific and non-specific), to determine the contribution of other enzymes towards CPT-11 activation. These studies demonstrate that hCE1, while demonstrating poor kinetic parameters for this substrate, can significantly contribute to drug hydrolysis. Furthermore, our studies identify the kidney as a source of CPT-11 activation and demonstrate that drug hydrolysis is more proficient in lung tissue isolated from smokers.