Rifampicin Induces Gene, Protein, and Activity of P-Glycoprotein (ABCB1) in Human Precision-Cut Intestinal Slices

Abstract

P-glycoprotein (ABCB1), an ATP-binding cassette efflux transporter, limits intestinal absorption of its substrates and is a common site of drug-drug interactions. Drug-mediated induction of intestinal ABCB1 is a clinically relevant phenomenon associated with significantly decreased drug bioavailability. Currently, there are no well-established human models for evaluating its induction, so drug regulatory authorities provide no recommendations for in vitro/ex vivo testing drugs' ABCB1-inducing activity. Human precision-cut intestinal slices (hPCISs) contain cells in their natural environment and express physiological levels of nuclear factors required for ABCB1 induction. We found that hPCISs incubated in William's Medium E for 48 h maintained intact morphology, ATP content, and ABCB1 efflux activity. Here, we asked whether rifampicin (a model ligand of pregnane X receptor, PXR), at 30 μM, induces functional expression of ABCB1 in hPCISs over 24- and 48-h incubation (the time to allow complete induction to occur). Rifampicin significantly increased gene expression, protein levels, and efflux activity of ABCB1. Moreover, we described dynamic changes in ABCB1 transcript levels in hPCISs over 48 h incubation. We also observed that peaks of induction are achieved among donors at different times, and the extent of ABCB1 gene induction is proportional to PXR mRNA levels in the intestine. In conclusion, we showed that hPCISs incubated in conditions comparable to those used for inhibition studies can be used to evaluate drugs' ABCB1-inducing potency in the human intestine. Thus, hPCISs may be valuable experimental tools that can be prospectively used in complex experimental evaluation of drug-drug interactions.

Keywords: P-glycoprotein (ABCB1 protein); absorption; human precision-cut intestinal slices; induction; pregnane X receptor; rifampicin.

FIGURE 1

Morphology and ABCB1 expression in non-sliced intestine and hPCISs. Intact brush border epithelium and significant ABCB1 positivity (brown color) were observed in the non-sliced tissue (A) in fresh hPCISs, time = 0 h (B), after 24 h (C), and in hPCISs incubated for 48 h (D). The slides were counterstained with hematoxylin (scale bar = 100 µm).

FIGURE 2

ATP contents in freshly prepared hPCISs (0 h) and after incubation in WME for 24 and 48 h with or without RIF (30 µM). Presented data are medians calculated from eight donors with interquartile ranges. Dots represent individual donor’s values. No significant differences between RIF-free and RIF-treated samples were detected by the nonparametric Kruskal–Wallis analysis followed by Dunn’s test.

FIGURE 3

Accumulation of RHD123 analyzed in fresh hPCISs and hPCISs after incubation for 24 or 48 h. Each line represents samples collected from one donor. Corresponding calculated medians did not differ according to Kruskal–Wallis tests (n = 8) in the absence of CP100356 (A). The addition of 2 µM CP100356 significantly increased the accumulation of RHD123 in hPCISs treated at all three time points (n = 8). Data are presented as a pairwise comparison showing individual donor values (control/CP100356-treated) together with indications of the significance of between-treatment differences according to the Wilcoxon signed-rank test (*, p < 0.05; **, p < 0.01) (B).

FIGURE 4

ABCB1, CYP3A4, PXR, and RXRA gene expression in hPCIS determined by qRT-PCR. Fold changes in ABCB1 and CYP3A4 mRNA levels in hPCISs incubated with RIF (30 µM) for 4, 8, 12, 16, 20, 24, and 48 h (A), relative to expression in RIF-free controls, showing that RIF significantly increased mRNA levels of both genes in samples from all donors at all tested time points (one-sample t-tests, three slices from one donor: *, p < 0.05; **, p < 0.01; ***, p < 0.001). PXR (B) and RXRA (C) were demonstrated to be expressed in hPCISs incubated with RIF (30 µM) at all tested time points. Amounts of PXR and RXRA transcripts are presented in arbitrary units (a.u.), calculated as 2^−∆Ct × 10⁶. (D) The matrix of Pearson’s coefficients of correlation was used to assess associations among all targeted transcripts in RIF-treated hPCISs (positive in blue and negative in red, with color intensity proportional to the magnitude of the coefficients). Asterisks indicate significant differences: *, p < 0.05; **, p < 0.01; ***, p < 0.001.

FIGURE 5

ABCB1 levels (indicated by brown staining) in RIF-free samples incubated for 24 h (A) and 48 h (C) and in counterparts treated with 30 µM RIF for 24 h (B) and 48 h (D). The slides were counterstained with hematoxylin (scale bar = 100 µm). Panels (A–D) represent only a partial cutout of the whole section to demonstrate immunohistochemical staining. Images of the entire sections were used for the analysis. (E) Fold changes in numbers of strong ABCB1-positive pixels between images of RIF-treated and RIF-free hPCISs from donors after incubation for indicated times according to subsequent algorithmic analysis. Asterisks indicate significant differences between them according to one-sample t-tests (n = 3): *, p < 0.05; **, p < 0.01; ***, p < 0.001.

FIGURE 6

Levels of RHD123 accumulation in hPCISs incubated with 30 µM RIF were significantly lower than those in RIF-free samples after both 24- and 48-h incubations (Wilcoxon signed-rank tests: *, p < 0.05, and **, p < 0.01). Data are presented as pairwise comparisons between RIF-free (dot) and RIF-treated (square) samples. Each line represents an individual donor.