*3.3. Di*ff*erential Expression of CES1 and CES2 in EpiIntestinal Microtissues and Caco-2 cells*

Ishiguro et al. reported that Caco-2 cells, albeit originated from human colorectal carcinoma, resemble rather hepatocytes with regard to expression of the isoforms CES1 and CES2 [4]: Whereas CES2 is predominantly expressed in human intestine, CES1 is the major esterase in human hepatocytes and Caco-2 cells. In order to profile EpiIntestinal microtissues in this regard, we investigated the metabolism of dabigatran etexilate, the same substrate used by Ishiguro et al., in human hepatocytes (huHEP), cryopreserved human intestinal mucosa (HIM), Caco-2 cells and EpiIntestinal microtissues. The metabolic pathway for dabigatran etexilate is depicted in Figure 3a: the formation of the intermediate metabolite BIBR 1087 from the double prodrug dabigatran etexilate and the formation of the active drug from the intermediate metabolite BIBR 951 is catalyzed by CES1. The formation of BIBR 951 from dabigatran etexilate and the formation of the active drug BIBR 953 from BIBR 1087 is catalyzed by CES2. In tissues and cells with predominant expression of CES1 (liver e.g.,), BIBR 1087 should be the major metabolite; in tissues and cells with predominant expression of CES2, BIBR 951 should be the major metabolite (e.g., intestine). As shown in Figure 3b, in human hepatocytes and human intestinal mucosa, the expected metabolite pattern was observed—BIBR 1087 as the main metabolite in hepatocytes, and BIBR 951 as the main metabolite in human intestinal mucosa. Consistent with the data by Ishiguro et al., BIBR 1087 was found to be the main metabolite in Caco-2 cells. Interestingly, the metabolite pattern of dabigatran etexilate in EpiIntestinal microtissues resembles none of the other three models. The active drug BIBR 953 was found as the main metabolite in EpiIntestinal microtissues, suggesting similar CES1 and CES2 enzyme activities in this model.

(**a**)

**Figure 3.** *Cont.*

**Figure 3.** Metabolic pathways and metabolite pattern of dabigatran etexilate. (**a**) Metabolic pathways of dabigatran involving CES1 and CES2. (**b**) Detection of metabolites of dabigatran etexilate in human hepatocytes (huHEP), human intestinal mucosa (HIM), Caco-2 cells, and EpiIntestinal microtissues. Dabigatran was incubated with hepatocytes and intestinal mucosa in suspension, or given to the apical compartment of Caco-2 cells and EpiIntestinal microtissues and incubated at 37 ◦C. At the indicated timepoints, samples were taken from the suspension of huHEP and HIM or from the basal compartments of Caco-2 cells and EpiIntestinal microtissues. The metabolites were quantified using LC-MS/MS. Data shown as the mean and SD of triplicates.

#### *3.4. UGT and SULT Activities in EpiIntestinal Microtissues*

β Extensive glucuronidation in the intestine is one of the major reasons that hamper the oral availability of drugs. Raloxifene and ezetimibe are two examples showing extensive intestinal glucuronidation in human [28–30]. When given to the apical compartment of EpiIntestinal microtissues, two putative glucuronides for both drugs were detected by mass scan (Table 5). We could confirm that these are glucuronides of both drugs by digestion with β-glucuronidase (data not shown). For ezetimibe, glucuronidation was the only metabolism found in EpiIntestinal microtissues. In case of raloxifene, we found an unexpectedly high amount of sulfation products of raloxifene (Table 5). To verify the physiological relevance of this finding, we performed metabolite identification for both drugs also with cryopreserved human intestinal mucosa. As shown in Table 5, the metabolite patterns of both drugs are similar in EpiIntestinal and in HIM: glucuronidation only for ezetimibe; glucuronidation and sulfation for raloxifene.


**Table 5.** EpiIntestinal microtissues and HIM were incubated with 10 µM ezetimibe or raloxsifene.


n.d.: Not detectable.
