*3.2. Drug-Metabolising Enzymes (DME) in EpiIntestinal Microtissues*

We studied the effect of DME on the GI first-pass availability with two model drugs: Midazolam for CYP3A4 [24] and astemizole for CYP2J2 [25,26]. As shown in Figure 1, midazolam showed a lower availability in EpiIntestinal microtissues than in Caco-2 (recovery in basal/receiver compartment after 24 h: 46.7% vs. 81.0%) when added to the apical compartment of both models. The addition of the covalent CYP3A4 inhibitor CYP3cide [27] increased the availability of midazolam in EpiIntestinal microtissues (46.7% vs. 75.9%), but had almost no effect on the availability in Caco-2 (81.0% vs. 87.9%). Moreover, substantial amount of 1-hydroxymidazolam, the CYP3A4-selective metabolite of midazolam, was detected in EpiIntestinal microtissues and was suppressed by the addition of the selective inhibitor, whereas only a negligible amount of the metabolite was detected in Caco-2, consistent with the low level of CYP3A4 expression in this cell line.

**Figure 1.** Apical-to-basal transport of midazolam in EpiIntestinal microtissues and Caco-2 cells. Midazolam (10 µM) was added to the apical compartment of EpiIntestinal microtissues (**left panels**) or Caco-2 cells (**right panels**) grown on Transwell inserts and incubated at 37 ◦C. At the time points as indicated, samples were taken from the basal (receiver) compartment. Midazolam (**upper panels**) and 1-Hydroxymidazolam (**lower panels**) were quantified in the samples via HPLC-MS/MS. The incubation was carried out in the absence or in the presence of the selective covalent CYP3A inhibitor CYP3acide (1 µM). Data are shown as mean values of triplicates. Error bars show standard deviations.

In contrast to CYP3A4, CYP2J2 activities were readily detected both in EpiIntestinal microtissues and in Caco-2 (Figure 2). In both models, co-incubation with the CYP2J2 inhibitor Ebastein (50 µM) [26] increased the availability of astemizole in the receiver compartment.

β

**Figure 2.** Apical-to-basal transport of astemizole in EpiIntestinal microtissues and Caco-2 cells. Astemizole (10 µM) was added to the apical compartment of EpiIntestinal microtissues (**left**) or Caco-2 cells (**right**) grown on Transwell inserts and incubated at 37 ◦C. At the indicated timepoints, samples were taken from the basal (receiver) compartment. Astemizole was quantified in the samples via HPLC-MS/MS. The incubation was carried out in the absence or in the presence of the competitive CYP2J2 inhibitor Ebastine (50 µM). Data are shown as mean values of triplicates. Error bars show standard deviations.

Encouraged by the results regarding CYP3A4 and CYP2J2 activities, we screened the activities of DME in EpiIntestinal microtissues more systematically: we quantified the formation of selective metabolite of the respective enzymes at different substrate concentrations (Figure S1). The substrate concentration approaching enzyme saturation (marked dots in Figure S1) was subsequently used to compare the activities in EpiIntestinal microtissues and Caco-2 cells. As shown in Table 4, the most prominent difference in DME activities between both models are the CYP3A4 activities, which could be demonstrated by two different substrates (testosterone and midazolam). In addition, CYP1A2 is the only enzyme with much lower activities in EpiIntestinal than in Caco-2 cells. Furthermore, we measured all metabolites also in the cell lysate. Some of the metabolites showed substantial intracellular accumulation. In order to monitor batch variability with regard to DME activities, experiments shown in Table 4 were repeated with microtissues from a different batch with comparable results (not shown).


**Table 4.** Determination of activities of DMEs in EpiIntestinal microtissues and Caco-2 cells. Enzyme activities in EpiIntestinal microtissues and Caco-2 cells in 96-well Transwell plates were measured with the respective substrates shown in Figure S1 and at the marked concentration. Data shown as mean and SD from triplicates.

\* Determined as the rate of metabolite formation in supernatant; BLQ: Below limit of quantification.
