Contribution of organic anion transporting polypeptides to bile acid uptake in the Caco-2 cell monolayer and gastrointestinal tract

TL;DR

This study shows that bile acid uptake in Caco-2 cells involves multiple transporters, including ASBT and OATPs, suggesting a more complex mechanism than previously thought.

Contribution

The study identifies OATP1A2 and OATP1B3 as additional contributors to bile acid uptake in Caco-2 cells, beyond the known ASBT transporter.

Findings

Bile acid uptake in Caco-2 cells involves ASBT and OATP transporters.

OATP1B3 is localized to the apical membrane and contributes to TC uptake.

OATP1B3 expression in colon-derived organoids suggests a role in colonic bile acid transport.

Abstract

The apical sodium-dependent bile acid transporter (ASBT, encoded by SLC10A2) is the sole transporter responsible for bile acid absorption in the human intestinal tract and is a therapeutic target for dyslipidemia. Here, we investigated the mechanisms of bile acid uptake in Caco-2 cells, a widely used in vitro model of intestinal absorption. Initial uptake assays revealed biphasic kinetics in bile acid uptake, suggesting the involvement of multiple transporters. At low substrate concentrations, uptake in Caco-2 cells showed marked sodium dependence, and the kinetic parameters were consistent with those observed in taurocholic acid (TC) uptake by ASBT-expressing cells, identifying ASBT-mediated TC uptake as one component of uptake. The second component showed pronounced pH dependence in Caco-2 monolayers, prompting us to focus on organic anion-transporting polypeptides (OATPs; SLCO family), which are pH-dependent transporters of amphiphilic substrates and mediate bile acid transport in other organs. To identify transporters beyond ASBT, we also examined human OATPs. OATP1A2 and OATP1B3 were identified as candidate transporters contributing to TC uptake in Caco-2 cell monolayers, and OATP1B3 was localized to the apical membrane. These findings indicate that bile acid uptake in Caco-2 cells is mediated by multiple transporters beyond ASBT, underscoring the need for cautious interpretation of bile acid transport studies using this model. Expression of OATP1B3 on the apical side of human colon–derived organoid monolayers, which exhibited TC uptake activity, suggests a potential role in colonic bile acid transport.

Figures

Full text

The enterohepatic circulation of bile acid and the intestinal bile acid transport system have gained increasing attention as potential therapeutic targets for metabolic syndrome (1). Bile acids, amphipathic compounds synthesized from cholesterol in the liver, primarily consist of cholic acid (CA) and chenodeoxycholic acid in humans. Most of these bile acids are conjugated with either taurine or glycine, which enhances their water solubility and reduces their toxicity to the liver (2, 3). After being secreted into the small intestine, bile acids facilitate the absorption of nutrients, and over 95% of them are reabsorbed in the terminal ileum to return to the liver for reuse, a process known as the enterohepatic circulation of bile acids (4). Inhibiting bile acid reabsorption from the ileum is known to reduce plasma cholesterol levels; this is because inhibited bile acid reabsorption in the ileum increases the amount of bile acid excreted into feces, enhances de novo synthesis from cholesterol in the liver, and lowers cholesterol secretion into the blood (5).

On the other hand, bile acid that enters the large intestine is deconjugated by the gut microbiota and becomes secondary bile acid, such as deoxycholic acid (DCA) and lithocholic acid (6). As these secondary bile acids exhibit cell toxicity, they are associated with intestinal inflammation and/or cancer progression in the large intestine (7). These secondary bile acids are absorbed mainly via passive diffusion from the large intestine (8), whereas bile acid uptake from the small intestine is mediated by transporter proteins.

A pivotal transporter involved in the reabsorption of intestinal bile acids is the apical sodium-dependent bile acid transporter (ASBT), encoded by SLC10A2. ASBT is primarily expressed in the apical membranes of enterocytes, where it facilitates sodium-dependent bile acid transport. ASBT consists of seven transmembrane domains (9) and operates by transporting bile acids alongside two sodium ions, thereby triggering conformational changes necessary for substrate transport. Upon absorption from the apical membrane by ASBT, bile acids bind to the ileal bile acid binding protein for transport to the basolateral membrane and eventual delivery to the portal vein via the organic solute transporter α/β (10). Given that the initial step in this reabsorption process, mediated by ASBT, is rate-limiting, it represents a promising therapeutic target for metabolic disorders (11). Drugs that target ASBT have been developed and are currently being used in clinical settings (12). Our group screened for ASBT inhibitors in food ingredients, identified a type of black tea polyphenol, theaflavin, and reported on its inhibitory mechanism (13).

Caco-2 cell monolayer is the most extensively used in vitro intestinal permeable cell model due to its ability to form well-differentiated and polarized cell monolayers as surrogates for the human intestinal epithelium. Caco-2 monolayers possess important structural and functional characteristics of intestinal enterocytes, making them an attractive cell line for investigating drug and food constituent absorption. Our group also used Caco-2 monolayer membranes to analyze the intestinal absorption mechanisms of bile acids (13), cholesterol (14, 15), flavonoids (16, 17, 18, 19), and so on. However, the Caco-2 cell monolayer has several limitations, such as it shows an expression pattern different from that of actual intestinal epithelial cells due to its origin in adenocarcinoma.

Previous studies on bile acid reabsorption have been conducted using Caco-2 cells. For example, when the intestinal permeability of liraglutide nanoparticles was examined, it was reported that increasing the binding between the nanoparticles and bile acids increases their affinity for ASBT and intestinal absorption (20). Cereal-derived mold toxins have been reported to reduce taurocholic acid (TC) permeation in Caco-2 cell monolayers by decreasing ASBT mRNA expression via the MAPK pathway and inhibiting protein synthesis on ribosomes (21). Apple juice (22) and tea catechins (23) have also been reported to decrease ASBT expression and bile acid uptake.

ASBT is a sodium-dependent transporter (24), and to the best of our knowledge, it is pH-independent. ASBT is a dominant bile acid transporter in the human ileum (25) and Caco-2 cells (26). Our previous study also demonstrated that most TC uptake into Caco-2 cells was sodium-dependent when tested at low TC concentration (13), indicating that the absorption was due to ASBT. However, sodium-independent bile acid uptake was also observed in Caco-2 cells (27). In addition, sodium-independent bile acid uptake was found in rodents’ intestines (28); however, to the best of our knowledge, no further studies have been conducted. Among the experiments using Caco-2 to investigate bile acid uptake, if other transporters exist, there is a risk of misinterpreting the results. It also needs to be clarified whether there are bile acid transporters in the human digestive tract other than ASBT or whether it is different from Caco-2. Although bile acid uptake primarily occurs in the ileum, some bile acids also reach the colon. It is not fully understood whether there are bile acid transporters in the large intestine.

In this study, we used TC as a representative bile acid and investigated its uptake mechanism in Caco-2 cells by analyzing its sodium, pH, and concentration dependence along with kinetic analysis. Furthermore, to assess whether transporters other than ASBT could mediate bile acid uptake, we screened transporter-expressing cells and examined gene and protein expression using Caco-2, human gastrointestinal tract total RNA, and organoid. In addition, TC permeability assays were performed to evaluate the transepithelial transport capacity of TC in the organoid model.

Results

Sodium dependence and concentration dependence of TC uptake in Caco-2 cells

ASBT is a known sodium-dependent transporter expressed in Caco-2 cells (29). Therefore, we evaluated the sodium dependence of TC uptake in Caco-2 cells (Fig. 1, A and B). TC uptake in a sodium-containing buffer exhibited a linear increase up to 15 min, with a continued increase up to 120 min. In contrast, TC uptake in a sodium-free buffer reached a plateau at 5 min and remained stable for 60 min. In addition, TC uptake in the sodium-containing buffer was significantly higher than that in the sodium-free buffer at all time points measured. To further characterize TC uptake into Caco-2 cells, we assessed its concentration dependence (Fig. 1C). TC uptake in Caco-2 cells (pH 7.4) displayed a saturable process. The results of the concentration-dependent uptake assay were transformed into an Eadie–Hofstee plot, which is a linearized expression of the Michaelis–Menten equation. When V/[S] was plotted on the x-axis and V on the y-axis, this plot showed the Vmax and Km as the x-intercept and slope, respectively. The TC uptake plotted using the Eadie–Hofstee plot for TC uptake exhibited biphasic kinetics (Fig. 1D), suggesting the involvement of more than one transporter in the uptake process. In addition, Akaike’s information criteria (AIC) for the model assuming two saturable components (equation B shown in Experimental procedures) was −122, whereas that for the model assuming one saturable component (equation A) was 210; therefore, equation B was adopted for calculating the kinetic parameters. The saturable components were categorized into two groups: transporter 1 and transporter 2. The kinetic parameters for transporter 1 were Vmax = 36.0 ± 31.8 pmol/mg protein/min and Km = 13.3 ± 10.7 μM (Table 1). For transporter 2, the parameters were Vmax = 293.8 ± 234.5 pmol/mg protein/min and Km = 338.3 ± 517.4 μM. We provisionally assigned the saturable component with the lower Km as transporter 1 and the one with the higher Km as transporter 2. A low Km indicates high substrate affinity, whereas a high Vmax indicates high transport capacity. Thus, transporter 1 exhibits higher affinity than transporter 2, whereas transporter 2 has a greater transport capacity than transporter 1. The Km values estimated for transporter 2 were relatively large with substantial variance, suggesting possible involvement of multiple transporters. The TC uptake and kinetic analysis (Fig. 1C) revealed that transporter 1 contributed predominantly to the total uptake at lower TC concentrations. However, as the TC concentration increased, the contribution of transporter 2 escalated dramatically, surpassing that of transporter 1, particularly above 25 μM (Fig. 1, B and C).Figure 1**Taurocholic acid (TC) uptake in Caco-2 cell monolayer.**A, experimental procedure overview is shown in panel A. Caco-2 cell monolayer cultured for 1 week were exposed to uptake solutions containing 36 nM [^3^H]-TC and unlabeled TC for a specified duration. B, sodium dependence of TC uptake. Caco-2 monolayers were treated with TC solution (36 nM [^3^H]-TC and 5 μM TC) with (closed circle) or without (open circle) sodium for various time intervals (1, 5, 15, 30, 45, 60, 90, and 120 min) at 37 °C. Sodium-dependent uptake assays were performed using Hank's Balanced Salt Solution with pH adjusted with Hepes (HBS-Hepes) (pH 7.4), while sodium-independent uptake assays used buffers where NaCl was replaced with choline chloride, with pH adjusted to 7.4 using KOH. C, concentration dependence of TC uptake in Caco-2 cell monolayer at pH 7.4. Cells were exposed to uptake solutions containing 36 nM [3H]-TC and varying concentrations (0, 2.5, 5, 10, 15, 25, 50, 75, 100, 125, 150, 175, and 200 μM) of unlabeled TC for 15 min at 37 °C. Assays were conducted in HBS-Hepes (pH 7.4). The black line represents the total uptake, which were separated into two transporter uptakes according to the Michaelis–Menten equation (the red line and the blue line). D, Eadie–Hofstee plot of the uptake. The equation for linear analysis of Eadie–Hofstee Plot is the one given in Experimental procedures. The values are expressed as mean ± SD (n = 3).Table 1. Kinetic parameters for taurocholic acid (TC) uptake in Caco-2 cells at pH 7.4ParameterpH 7.4Transporter 1Transporter 2Vmax (pmol/mg protein/min)36.0 ± 31.8293.8 ± 234.5Km (μM)13.3 ± 10.7338.3 ± 517.4Vmax/Km (pmol/mg protein/min/μM)2.70.87Expressed as mean ± SD (n = 3). Transporter 1 and transporter 2 represent the high-affinity and low-affinity components, respectively (refer to Fig. 1, B and C).

pH dependence of TC uptake by Caco-2 cells cultured for 1 and 2 weeks

To further characterize transporter 1 and transporter 2 in Caco-2 cells, we examined the pH dependence of TC uptake in Caco-2 cells. Some solute carrier (SLC) transporters proton-coupled or increased substrate transport activity at acidic pH (30, 31). TC uptake at pH 6.0 was notably higher than at pH 7.4 across most of the concentrations tested (p < 0.05) (Fig. 2A). The concentration-dependent transport displayed a saturable process (Fig. 2B). Transforming the measured V values from the concentration-dependent uptake assay into an Eadie–Hofstee plot revealed biphasic kinetics in the same way as that for pH 7.4 (Fig. 2C). In addition, the AIC value for the model assuming two saturable components (equation B shown in Experimental procedures) was −119, whereas that for the model assuming one saturable component (equation A) was 217; therefore, equation B was adopted for calculating the kinetic parameters. For transporter 1, the kinetic parameters were determined as Vmax = 48.6 ± 37.1 pmol/mg protein/min and Km = 13.2 ± 9.8 μM. Conversely, for transporter 2, the parameters were Vmax = 968.4 ± 1753 pmol/mg protein/min and Km = 809.3 ± 1913 μM (Table 2). Here, we provisionally designated the saturable component with the lower Km as transporter 1 and the one with the higher Km as transporter 2. Accordingly, transporter 1 exhibited higher affinity, whereas transporter 2 demonstrated greater capacity for TC uptake. Notably, transporter 2 displayed nearly double the Vmax at pH 6.0 compared with that at pH 7.4, and Vmax/Km ratios were elevated at pH 6.0. This finding indicates that transporter 2 is pH dependent. The same uptake test was also performed with Caco-2 cells cultured for 2 weeks, which should enhance differentiation of the cells. The TC uptake rate decreased compared with that observed at 1 week under all pH conditions (Fig. 2, A and D). The rate of TC uptake was pH 7.4 > pH 6.0 for low TC concentrations and pH 6.0 > pH 7.4 for high TC concentrations (Fig. 2D), with difference in uptake rate by pH was less than 1-week culture (Fig. 2, A and D). However, the Eadie–Hofstee plot also showed biphasic kinetics at both pH (Fig. 2, E and F), indicating the involvement of more than two transporters with 2 weeks cultured in the same manner of 1 week. However, due to the poor fit to the Michaelis–Menten equation in 2 weeks–cultured Caco-2 cell monolayer, the kinetic parameters of each transporter could not be calculated using nonlinear least squares methods.Figure 2**pH dependence of TC uptake in Caco-2 cell monolayers cultured for 1 and 2 weeks.**A, concentration dependence of TC uptake compared between pH 7.4 and pH 6.0. Caco-2 cell monolayers were exposed to uptake solutions for 15 min at 37 °C in Hank's Balanced Salt Solution with pH adjusted with Mes (HBS-Mes) (pH 6.0). The uptake at pH 7.4 was extracted from Figure 1C. The area enclosed by the dotted box is a magnified view of the graph with low TC concentration (0–25 μM). Statistical differences between pH 7.4 and pH 6.0 were assessed using the Student’s t test (∗p < 0.05; ∗∗p < 0.01). B, kinetic analysis of TC uptake in Caco-2 cell monolayer at pH 6.0. C, Eadie–Hofstee plot of the uptake. D, concentration dependence of TC uptake at pH 7.4 and pH 6.0 in Caco-2 cells cultured for 2 weeks. The area enclosed by the dotted box is a magnified view of the graph with low TC concentration (0–25 μM). E, Eadie–Hofstee plot of TC uptake in Caco-2 cell monolayers cultured for 2 weeks at pH 7.4 and (F) pH 6.0. All values are expressed as mean ± SD, All values are expressed as mean ± SD (n = 3). The statistical differences between pH 7.4 and pH 6.0 (A and C) were assessed using unpaired t test (∗p < 0.05). TC, taurocholic acid.Table 2. Kinetic parameters for taurocholic acid (TC) uptake in Caco-2 cells at pH 6.0ParameterpH 6.0Transporter 1Transporter 2Vmax (pmol/mg protein/min)48.6 ± 37.1968.4 ± 1753Km (μM)13.2 ± 9.8809.3 ± 1913Vmax/Km (pmol/mg protein/min/μM)3.71.2Expressed as mean ± SD (n = 3). Transporter 1 and transporter 2 represent the high-affinity components and low-affinity components, respectively (refer to Fig. 2, B and C).

Kinetic analysis of TC uptake by ASBT

To validate the kinetic analysis in Caco-2 cells and attribute transporter 1 and transporter 2, we initially investigated the TC uptake mechanism mediated by ASBT in COS7 cells overexpressing ASBT. The TC uptake in COS7 cells demonstrated a saturable process (Fig. 3, A and B). Kinetic analysis was performed using equation A, which is shown in Experimental procedures for monophasic uptake. The kinetic parameters at pH 7.4 were determined as Vmax = 184.9 ± 14.9 pmol/mg protein/min and Km = 15.0 ± 4.3 μM. Conversely, at pH 6.0, the parameters were Vmax = 69.7 ± 8.6 pmol/mg protein/min and Km = 24.4 ± 9.3 μM (Table 3). Considering that the Vmax and Km values were consistent with those of the transporter 1 of Caco-2 cell monolayer (Tables 1 and 2), we inferred that transporter 1 corresponded to ASBT.Figure 3**Kinetic analysis of TC uptake in COS7 cells overexpressing ASBT.**A, an overview of the experimental procedure. COS7 cells were transfected with a vector containing ASBT (SLC10A2) cDNA and subsequently subjected to TC uptake assays. The cells were exposed to uptake solutions (36 nM [^3^H]-TC and varying concentrations of unlabeled TC: 0, 5, 10, 25, 50, 75, and 100 μM) for 15 min at 37 °C. B, assays were conducted in either HBS-Hepes (pH 7.4, closed circle) or HBS-Mes (pH 6.0, open circle). Kinetic analysis was performed using the Michaelis–Menten equation; values are expressed as mean ± SD (n = 3). ASBT, apical sodium-dependent bile acid transporter; cDNA, complementary DNA; TC, taurocholic acid.Table 3. Kinetic parameters for taurocholic acid (TC) uptake in ASBT-expressing cellsParameterpH 7.4pH 6.0Vmax (pmol/mg protein/min)184.9 ± 14.969.7 ± 8.6Km (μM)15.0 ± 4.324.4 ± 9.3Vmax/Km (pmol/mg protein/min/μM)12.32.6ASBT, apical sodium-dependent bile acid transporter.Expressed as mean ± SD (n = 3).

Screening for organic anion transporting polypeptides that mediate bile acid uptake

Since transporter 2 in Caco-2 cell monolayer showed lower affinity and greater pH dependence than transporter 1 (Figs. 1, C and D, 2, B and C), we considered this transporter to belong to the organic anion-transporting polypeptide (OATP) family. Some OATPs mediate bile acid uptake in other organ cells such as the liver and placenta (10, 32), and they generally exhibit increased transport activity at acidic pH (33). Therefore, we conducted screened human OATPs for their TC uptake. The transfected cells were evaluated by assessing the uptake of known OATP substrates reported in the literature, such as estrone-3-sulfate (E3S) or prostaglandin E2 (PGE_2_). E3S uptake in OATP3A1-and OATP4C1-transfected cells did not differ from that in mock-transfected cells (Fig. 4, G and I), and E3S or PGE2 uptake in OATP5A1-and OATP6A1-transfected cells was not examined due to insufficient literature supporting their uptake; nevertheless, their gene expression in the transfected cells was confirmed via Reverse transcription-quantitive PCR (RT-qPCR) (Fig. S1).Figure 4**Screening for OATP-overexpressing HEK293T cells that mediate TC uptake.**A, An overview of the experimental procedure. HEK293T cells were transfected with vectors containing cDNA of each of the nine OATPs or mock vectors. B–K, For validation of functional expression, estrone-3 sulfate (E3S), known as typical substrate, uptake assays were conducted for all OATPs, except OATP2A1, OATP5A1, and OATP6A1, while prostaglandin E2 (PGE_2_) uptake assays were conducted for OATP2A1. For TC uptake assays, HEK293T cells transfected with each vector were exposed to uptake solutions (36 nM [^3^H]-TC and 5 μM unlabeled TC) for 15 min at 37 °C. Assays were performed in either HBS-Hepes (pH 7.4) or HBS-Mes (pH 6.0). Values are presented as mean ± SD (n = 3–5). Statistical differences between mock-transfected cells (M) and OATP-transfected cells (O) were assessed using unpaired t test (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001). The difference in TC uptake between pH 7.4 and pH 6.0 was evaluated by comparing relative TC uptake in OATPs cDNA-containing cells against mock-transfected cells for each pH, with statistical differences between pHs assessed using unpaired t test (##p < 0.01; ####p < 0.0001). cDNA, complementary DNA; OATP, organic anion-transporting polypeptide; TC, taurocholic acid.

TC uptake in cells transfected with OATP1A2, OATP1B1, and OATP1B3 was significantly higher than that in mock-transfected cells at both pH 7.4 and pH 6.0 (Fig. 4, B–D). HEK293T cells transfected with OATP2B1 exhibited greater TC uptake than mock-transfected cells only at pH 6.0 (Fig. 4F). In addition, TC uptake in cells transfected with OATP1A2, OATP1B1, and OATP1B3 increased significantly at pH 6.0 compared with that at pH 7.4 (Figure B–D). In particular, OATP1A2 showed high pH dependence.

Gene and protein expression analysis in Caco-2 cells

To further ascertain the involvement of OATP family members in TC uptake in Caco-2 cells, we assessed the gene and protein expression of OATPs and ASBT. First, the differentiation status of the created Caco-2 monolayer was confirmed. Ezrin, claudin-1 (CLD1), intestinal alkaline phosphatase (ALP-I), and sucrase-isomaltase (SI), the apical and differentiation markers of the Caco-2 cell monolayer (34, 35), were subjected to RT-qPCR analysis. The expression levels of Ezrin, CLD1, and SI in Caco-2 cells cultured for 1 to 3 weeks did not differ significantly from each other (Fig. 5, A–D). The expression level of ALP-I increased from week 2 onward relative to week 1 (Fig. 5C), whereas the activity of the ALP-I enzyme remained unchanged during this period (Fig. 5E). Consequently, we concluded that the cells were sufficiently differentiated and suitable for use in the uptake assays, even under the 1-week plate-culture conditions employed in this study. ASBT expression continued to increase with an increase in the days of culture (Fig. 5F), whereas OATP1A2 and OATP1B3 expression did not differ between the weeks (Fig. 5, G and H). OATP1B1 was not detected.Figure 5Expression of ASBT and OATP families in Caco-2 cell monolayer. Gene expression of Caco-2 cell monolayer cultured for 1 to 3 weeks at an initial density of 1 × 10^5^ cells/well on 24-well plates and subsequently subjected to RT-qPCR analysis. Gene expression of apical marker and differentiation marker (A) E**zrin (B) CLD1. (C) ALP-I, (D) SI were analyzed. E, ALP-I enzyme activity was measured. Gene expression of transporters were also measured; (F) ASBT, (G) OATP1A2 and (E) OATP1B3. All gene expression values are shown as relative expression to β-actin. I, representative blots of western blot analysis of ASBT, (J) OATP1A2, and (K) OATP1B3. Graphs on the right side of the image shows relative expression against β-actin. The values are presented as mean ± SD (n = 3). Statistic differences of each gene expression and protein expression cultured for 1 to 3 weeks were evaluated using one-way ANOVA followed by Tukey’s test, and different letters indicate p < 0.05 between the groups. ALP-I, intestinal alkaline phosphatase; ASBT, apical sodium-dependent bile acid transporter; CLD1, claudin-1; OATP, organic anion-transporting polypeptide.

The protein expression of ASBT, OATP1A2, and OATP1B3 was confirmed through western blotting (Fig. 5, I–K). The ASBT protein expression did not reach significant difference. OATP1A2 expression increased after 2 weeks and dropped after 3 weeks. OATP1B3 was highly expressed in 1-week culture, but its expression diminished over time.

The localization of ASBT and OATP1B3 in the intestinal epithelium was evaluated via immunocytochemistry in Caco-2 cell monolayers, both cultured for 1 and 2 weeks (Fig. 6, A and B). Spatial expression sites were verified by taking XYZ axes projection images using confocal microscopy. The transporters (ASBT and OATP1B3), ezrin (an apical marker of intestinal epithelial cells), and 4′,6-diamidino-2-phenylindole (DAPI) (nuclei) are represented in green, red, and blue, respectively. Ezrin expression was observed only on the apical side. For both ASBT and OATP1B3, overlap of the fluorescent signal with that of ezrin was observed in the merged images (shown by arrow heads), which indicated that these transporters are expressed on the apical side at both 1 and 2 weeks of culture.Figure 6Localization of ASBT and OATP1B3 in Caco-2 cell monolayer. Orthogonal sections form z-stack confocal images of (A) ASBT and (B) OATP1B3, with apical marker ezrin in Caco-2 cell monolayers via immunofluorescence analysis. Caco-2 cells were seeded on a chamber slide and cultured for 1 and 2 weeks. Cells were then fluorescently labeled with antibodies against ASBT (A) or OATP1B3 (F) and Ezerin, and nuclei were stained with DAPI. Negative control (NTC) image shows Caco-2 cells incubated without primary antibodies. Merged images of the three channels are shown in the right in each row. The inset panel “a” is an XY axes projection, with green and red lines indicating the location of optical sectioning for the XZ (inset b) and YZ (inset c) axes projections, respectively. AP and BL denote the apical and basolateral sides of the Caco-2 cell monolayer, respectively. Arrowhead in the merged image indicates the overlap of ASBT or OATP1B3 (green) with ezrin (red) signals. All scale bars represent 10 μm. ASBT, apical sodium-dependent bile acid transporter; DAPI, 4′,6-diamidino-2-phenylindole; OATP, organic anion-transporting polypeptide.

Gene and protein expression analysis in the human intestine and organoid

Next, the mRNA expression of OATPs, which were found to transport TC via screening, was measured in the human digestive tract. The mRNA expression of OATP1A2, OATP1B1, OATP1B3, and ASBT was analyzed in human total RNA samples. ASBT exhibited high expression levels in the ileum and colon, whereas OATP1A2 expression was very limited (Fig. 7, A and B). OATP1B3 expression was higher in the colon (Fig. 7C). To further analyze protein and mRNA expression within the same samples, organoids derived from a healthy human individual were used. The mRNA expression of ASBT was higher in the ileum, whereas that of OATP1B3 was higher in the colon (Fig. 7, D and F). OATP1A2 was not detected at the mRNA level (data not shown); thus, it was excluded from the further analysis. Western blot analysis confirmed the protein expression of ASBT in the ileum and that of OATP1B3 in both the ileum and colon, with higher expression in the colon than in the ileum (Fig. 6, E and G). Furthermore, a human colon organoid monolayer (TEER > 135 Ω cm^2^) was established and analyzed for OATP1B3’s localization and TC transport activity. ZO-1 and F-actin labeled with phalloidin were used as apical membrane markers, and OATP1B3 was colocalized with them at the apical side of the monolayer (Fig. 7, H and I).Figure 7Expression analysis of ASBT, OATP1A2, and OATP1B3 in the human intestine and human intestine-derived organoid and TC permeability assay using organoids. Gene expression of (A) ASBT, (B) OATP1A2 and (C) OATP3 were analyzed in human ileum and colon total RNA extracted from normal tissue that were obtained from a commercial source. Gene and protein expression analysis in human organoid derived from healthy man. D, ASBT gene and (E) protein expression. F, OATP1B3 gene and (G) protein expression. H, human colon organoid monolayers were fluorescently labeled with antibodies against ZO-1 (white signal), phalloidin (green signal), and OATP1B3 (red), and nuclei were stained with Hoechst 33342 (cyan). ZO-1 and phalloidin served as apical markers. Merged images of the four channels are shown on the right-hand side. The inset panel “a” is a projection on the X and Y axes, with yellow lines indicating the location of optical sectioning for the projections on the X and Z axes (inset b) and Y and Z axes (inset c), respectively. AP and BL denote the apical and basolateral sides of the organoid monolayer, respectively. All scale bars represent 10 μm. I, negative control of the organoid monolayer incubated without primary antibody for OATP1B3. J, overview of the assay testing TC transport in the organoid monolayer, (K) time-dependent transport of TC, and (L) TC amount transported to the basolateral chamber and inside the cell after 120 min. Values are expressed as mean ± SD, if appropriate (A–C; n = 2–3 and D–G; n = 1, J, K; n = 5). Due to the limited sample size, no statistical analysis was performed for A–G. ASBT, apical sodium-dependent bile acid transporter; OATP, organic anion-transporting polypeptide; TC, taurocholic acid.

Permeation of taurocholic acid across human colon organoid monolayers

A TC permeability assay was conducted to investigate whether human colon organoid monolayers exhibit TC transport activity (Fig. 7, J–L). With increasing incubation time, the amount of TC permeating from the apical to the basal side increased, and TC was also detected within the cells, indicating that the organoid monolayer cellular uptake of TC. Collectively, these results suggest that the colonic organoid monolayer possesses TC permeation activity. Given its apical localization, OATP1B3 likely contributes to TC permeation across the monolayer, without excluding contributions from ASBT-independent diffusion or other SLC transporters.

Discussion

For a long time, little research has been conducted on transporters other than ASBT that are involved in bile acid absorption in the digestive tract. Our study aimed to investigate the mechanism of bile acid uptake in Caco-2 cells, serving as a model for intestinal absorption, and revealed that the OATP family, particularly OATP1A2 and OATP1B3, may also play a role in bile acid uptake in Caco-2 cells, especially in 1 to 2 weeks of culture. In addition, using human colon organoids, our study demonstrated that OATP1B3 is expressed in human colon organoids and suggested that a mechanism for bile acid (TC) absorption may exist not only in the ileum but also in the large intestine. Our approach and results are summarized in Figure 8.Figure 8Summary of screening approach and results. The upper image illustrates the screening approach used in this study. The bottom figures summarize the results.

We began investigating TC uptake mechanism into Caco-2 cell monolayers. Notably, time-dependent TC uptake increased only when measured in the sodium-containing buffer (Fig. 1B). Given that ASBT is a sodium-dependent transporter, this suggests ASBT’s dominance in TC uptake under these conditions, consistent with previous research (26). However, this is limited to conditions of extremely low concentration. Although assessing the concentration dependence of TC uptake, the Eadie–Hofstee plot displayed biphasic kinetics, implying the involvement of multiple transporters (Fig. 1D). This prompted us to further characterize the uptake of TC in a buffer at a different pH.

Interestingly, the total uptake was higher at pH 6.0 compared with pH 7.4, and the kinetic parameters for a transporter with lower affinity (transporter 2) increased its capacity at pH 6.0 (Fig. 2, A–C, and Table 2), suggesting the involvement of more than two transporters in TC uptake in Caco-2 cells, with transporter two exhibiting pH-dependent characteristics. Notably, transporter 2 may represent two or more transporters with similar Km values, making their distinction challenging via the Eadie–Hosftee plot. In addition, the sodium dependence study in Figure 1B, conducted at a low concentration (5 μM), suggests ASBT, likely corresponding to transporter 1, contributes prominently to TC uptake at low substrate concentrations. To further analyze the uptake, we used Caco-2 cells cultured for 2 weeks. Caco-2 cells are known to differentiate into intestinal epithelial-like cells following continued culture in a confluent state (36). Compared with the cells cultured for 1 week, the V of TC uptake was comparable at pH 7.4 and pH 6.0 (Fig. 2A vs Fig. 2D). Combined with the Eadie–Hofstee plot analysis (Fig. 2, E and F), these data suggest that although the 2-week culture increased the contribution of a high-affinity transporter (transporter 1), a low-affinity transporter (transporter 2) also continued to function. On the other hand, 2 weeks of incubation significantly increased the overall protein content of the cells, whereas the increase in TC uptake was little (data not shown), resulting in a significant decrease in apparent TC uptake and Vmax. In addition, these data did not fit to any reasonable kinetic models, and kinetic parameters were not calculated for cells cultured for 2 weeks. The parameter of transporter 2 could include passive diffusion of TC into Caco-2 cells. Recently, De Bruijn et al. reported that in Caco-2 cells, the transport of glycocholic acid (GC), a glycine-conjugated form of CA, was mainly mediated by ASBT, although a small fraction (16%) permeated to the basolateral side via passive diffusion (37). Since TC is water-soluble and passive diffusion should be limited (38, 39, 40), it is difficult to assume that all of the TC uptake mediated by transporter 2 actually was due to passive diffusion.

To identify transporters other than ASBT (transporter 1) in Caco-2 cells, we initially verified whether transporter 1 was ASBT. The kinetic parameters of transporter 1 in Caco-2 cells were compared with those of ASBT-expressing COS7 cells and Caco-2 cells at both pH 7.4 and 6.0 (Fig. 3B). The Km for ASBT was similar to that for transporter 1 in Caco-2 cells, and the kinetic parameters were consistent with values reported elsewhere (41, 42), indicating that transporter 1 was indeed ASBT.

Next, we aimed to identify a low-affinity transporter (transporter 2) in Caco-2 cells, which exhibited increased transport activity at lower pH (Table 1 and 2). In addition to ASBT, members of the OATP family, encoded by SLC22 or SLCO, play a role in bile acid transport across various organs such as the liver and placenta (10, 32). In humans, this transporter family comprises 11 members, all facilitating sodium-independent and anion exchange transport, possibly involving substrates like glutathione or bicarbonate (33, 43). Notably, most OATPs exhibit pH-dependent transport due to a pH-sensitive histidine in their third transmembrane domain (43). Although the substrates of OATPs vary among family members, some share common substrates such as estrone and bile acids (44). Although certain OATP family members are expressed in the human intestine, their specific role and contribution to gastrointestinal bile acid uptake remains unclear. Given that most OATPs are pH dependent (43), some are involved in bile acid uptake in the human liver (10), and certain OATPs are expressed in the intestine (45, 46, 47), we examined members of the OATP family. Among the 10 OATP-expressing cells successfully generated, OATP1A2, OATP1B1, OATP1B3, and OATP2B1 showed TC uptake (Fig. 4, B–K), indicating their potential as TC transporters in Caco-2 cells. The findings for OATP1A2, OATP1B1, OATP1B3, and OATP2B1 were consistent with those of previous studies (43, 48, 49) However, due to the low TC uptake capacity of OATP2B1 (data not shown), kinetic analysis was not feasible, which led us to focus on OATP1A2, OATP1B1, and OATP1B3 as candidates. OATP1B1 and 1B3 are well-known bile acid transporters on the basolateral membrane of the liver (50), whereas OATP1A2 is reported to mediate bile acid transport in the placenta (51). It is therefore consistent that these transporters would mediate bile acid transport if expressed in Caco-2 cells. Thus, we then analyzed their expression in Caco-2 cells at the mRNA level. OATP1B1 did not show mRNA expression, whereas OATP1A2 and OATP1B3 were expressed at a comparable level against β-actin at week 1, and ASBT expression increased over time. As gene expression levels do not necessarily correlate with protein expression levels (52), protein expression was also examined. ASBT protein levels did not significantly change over time (Fig. 5I). In contrast, compared with 1 week of culture, OATP1A2 expression increased, whereas OATP1B3 expression decreased after 2 weeks of culture (Fig. 5, J and K). ASBT protein levels did not significantly change over time (Fig. 5I). In contrast, OATP1A2 expression increased compared to after 1 week of culture while OATP1B3 expression decreased after 2 weeks of culture (Fig. 5, J and K). Notably, the overall pH dependence of TC uptake decreased (Fig. 2, A and D). Since OATP1A2 has greater pH dependence in TC uptake (Fig. 4B) and OATP1B3 activity was similar at pH 6.0 or pH 7.4, OATP1B3 likely contributes more than OATP1A2 to TC uptake in Caco-2 cells. The large apparent Km values with substantial variance observed for transporter 2 suggest that this component reflects heterogeneous contributions from multiple transporters—potentially including OATP1A2 and OATP1B3—and may incorporate a minor contribution from residual passive diffusion. Because many OATPs exhibit transport properties closer to facilitated diffusion than strictly saturable carrier-mediated transport (53), the concentration-dependent transport profile of transporter 2—including its shallow saturation and large apparent Km value— is consistent with established OATP-mediated transport characteristics. Accordingly, the kinetic parameters estimated for transporter 2 should be interpreted as apparent values rather than those of a single, well-defined transporter entity.

To confirm the transporters’ localization in Caco-2 cells, we focused on OATP1B3, which can transport bile acid at both pH 7.4 and pH 6.0. The immunofluorescence image of ezrin and ASBT confirmed that the Caco-2 cells used in this study were polarized and that ASBT is expressed on the apical membrane side (Fig. 6A). OATP1B3 also colocalized with ezrin (Fig. 6B), supporting that OATP1B3 contributes to TC uptake into Caco-2 cells. This study did not examine the potential involvement of other SLC transporters, such as organic anion transporters, and therefore, we cannot exclude the possibility that transporters other than OATPs may contribute to TC uptake in Caco-2 cells. However, this study focused on the role of OATPs, and our findings suggest that ASBT, OATP1B3, and possibly OATP1A2 are transporters of bile acids in Caco-2 cells. Our findings that OATPs also mediate TC uptake in Caco-2 cells were in part consistent with those of Couto et al. In the study by Couto et al., Na^+^-independent TC transport was reduced by tumor necrosis factor α (TNFα) and interferon γ (IFNγ) in Caco-2 cells cultured for 10 to 15 days (27), whereas Na^+^-dependent uptake (ASBT-mediated) was not affected by IFNγ treatment. Although no direct evidence was presented, this reduction in Na^+^-independent TC transport may be due to OATPs. Indeed, some research suggests that OATP1B3 expression was downregulated by these cytokines in the liver or bile duct (54, 55). There also are some studies that contradict our study. Hidalgo et al. reported the involvement of only one sodium-dependent transporter in TC transport in Caco-2 cells (26). They used, by which OATP expression was shown to have decreased. They used Transwell inserts with Caco-2 cells cultured for 3 weeks, which differs from our conditions. Based on our results, it is suggested that OATP expressions decrease after the culture period of 3 weeks (Fig. 5, I–K), which may partly explain why their findings differed from ours. Several studies have used Caco-2 cells cultured for around 14 days on plates to study the functions of ASBT (36, 56). However, in our investigation of the mechanisms underlying bile acid uptake kinetics using plate-cultured Caco-2 cells, we determined that 2 weeks of culture represented the maximum duration for reliable uptake analysis. At the 2-week time point, OATP1A2 and OATP1B3 were significantly involved in bile acid transport. To isolate ASBT-specific transport activity, the use of a Na^+^-free buffer is recommended to suppress nonspecific uptake mediated by OATPs.

Caco-2 cells have long served as an important model for studying the human gut and enterocyte (36) to assess the absorption and transport mechanism of various foods and drugs; however, a considerable number of studies argue that these cancer-derived cells have different features compared with normal human intestine or intestinal epithelium cells (36). For example, a drug metabolizing enzyme cytochrome P450 is expressed much less in the Caco-2 cell compared with the human small intestine (57). However, several limitations have been pointed out. First, expression levels of ASBT in Caco-2 cells are relatively low compared to those in the human intestine. The present study revealed that OATP1B3 and OATP1A2, which are not commonly considered as TC transporters in human ileum, were expressed in Caco-2 cells and mediated TC transport. As these findings indicate that Caco-2 cells do not necessarily exhibit the same physiological characteristics as human ileum, these differences should be considered when utilizing Caco-2 cells. Therefore, to investigate whether the expression patterns observed in Caco-2 cells reflect those in the human intestinal tract, we evaluated gene expression levels of ASBT, OATP1A2, and OATP1B3 in the human ileum and colon (Fig. 7, A–C). In human total RNA samples obtained from a commercial source, all of them were expressed, but OATP1A2 was relatively low. In the ileum, OATP1B3 expression was lower than ASBT expression (Fig. 7, A and C), although this finding does not completely exclude the possibility of a compensatory function of OATP1B3 alongside ASBT. Conversely, in the colon, OATP1B3 expression was comparable to ASBT expression. To further analyze protein expressions, we used organoid since obtaining protein samples from human was impossible. In organoid, ASBT was highly expressed in the ileum, which was consistent with a previous human study (58), and the protein expression had a similar trend. Interestingly, OATP 1B3’s expression was greater in the colon than in the ileum at both the mRNA and protein levels, and it localized at the apical side of the plasma membrane (Fig. 7, D–H). Since the organoid monolayer exhibited TC transport activity (Fig. 7, J–L), OATP1B3 expressed in organoids is thought to contribute to TC transport.

Although OATP1B3 is expressed in the human colon and human organoid and it has the capacity to mediate TC transport, its role and the degree of contribution are unknown. Bile acid reabsorption has been observed in the human colon (59); OATP1B3 may be a compensatory transporter for ASBT in colon. In other words, bile acids are absorbed by the high-affinity transporter ASBT in the ileum first, and OATP1B3 takes up the remaining bile acids that have flowed into the colon. The pH of the colon is low due to the action of intestinal bacteria, so it is reasonable that OATP, which works at low pH, is responsible for this function.

The significance of OATP1B3 may lie in its involvement in regulating lipid absorption and metabolism. Previous studies have reported that the inhibition of the bile acid transporter ASBT in metabolic syndrome via the promotion of bile acid excretion reduces systemic cholesterol levels. Our findings indicate that OATP1B3 may be a target molecule for metabolic disorders, although its extent of action may not be the same as that of ASBT. In addition, bile acid contributes to the regulation of lipid metabolism through bile acid–activated nuclear receptors; therefore, OATP1B3-mediated bile acid uptake could also modulate metabolism. The effect could be both positive and negative, depending on the bile acid species in the intestine and those absorbed from OATP1B3 or ASBT. Bile acids are potent ligands for the farnesoid X receptor (FXR), whose activation by bile acid species in the intestine induces the secretion of FGF19 (Fgf15 in mice) via the short heterodimer partner. FGF19/Fgf15 is transmitted to the liver to bind to the receptor FGFR4/β-Kloth, where it suppresses de novo bile acid synthesis through the ERK1/2 pathway (60). Consequently, lipid absorption is reduced due to the decreased bile acid pool and alterations in bile acid composition (61). The degree of FXR activation varies among bile acid species, with CA acting as an antagonist, and chenodeoxycholic acid (CDCA) and DCA serving as agonists (62, 63). Because OATP1B3 transports not only TC but also unconjugated bile acids (such as CA), conjugated primary bile acids, and conjugated secondary bile acids (49), bile acid uptake via OATP1B3 may also contribute to the FXR-mediated regulation of lipid metabolism.

Second, OATP1B3 may be significant due to its association with colorectal cancer. RNA-seq analysis indicated upregulation of bile acid–related genes, including OATP1B3, in the colon of patients with colon adenocarcinoma (64), which may be related to cancer progression. Conjugated bile acids, such as taurocholic acid, undergo conversion to secondary bile acid species in the colon that are potentially cytotoxic. In our study, ASBT and OATP1B3 were detected in Caco-2 cells, and OATP1B3 was also detected at the protein level in the colon. Studies on the role of these bile acid transporters in the colon are limited, but they may protect against conditions such as colonic inflammation and cancer (7), where secondary bile acid species are at risk. Moreover, intestinal inflammation is known to suppress ASBT expression and/or activity (27, 65), enhance bile acid influx into the colon, and increase secondary bile acid production by the intestinal microbiota. Furthermore, the higher concentration of primary bile acids, such as taurocholic acid, in feces in patients with ulcerative colitis than in healthy controls (66) suggest that ASBT function in the ileum is impaired, which can result in primary bile acid influx into the large intestine. Therefore, OATP1B3 may compensate for the reduced function of ASBT, especially in colitis.

In conclusion, this study demonstrates the role of OATP1A2 and OATP1B3 alongside ASBT in mediating bile acid uptake in human colonic adenocarcinoma Caco-2 cells and OATP1B3’s expression and function in the colon organoid. Further research is needed to clarify specific regions of expression and assess the physiological significance of transporters in the colon. When analyzing bile acid uptake mechanisms using Caco-2 cells, experiments should be designed to account for the contribution of OATPs in addition to ASBT.

Experimental procedures

Materials

[^3^H]-Taurocholic acid (6.7 Ci/mmol) and [^3^H]-prostaglandin E_2_ (185.4 Ci/mmol) were procured from PerkinElmer, while [^3^H]-Estrone 3-sulfate (40 Ci/mmol) was obtained from American Radiolabeled Chemicals. Sodium taurocholate and PGE_2_ were sourced from Fujifilm Wako Pure Chemicals, with unlabeled estrone 3-sulfate purchased from Cayman Chemical Company. Dulbecco’s modified Eagle medium (DMEM) (high glucose) was acquired from Nacalai Tesque, and fetal bovine serum (FBS) was obtained from Life Technologies. The Caco-2 cells and COS-7 cells used in this study were provided by RIKEN BRC through the national bio-resource project of MEXT/AMED, Japan. HEK293T cells were generously provided by Dr H. Matsunami (Duke University). Vectors containing transporter complementary DNA (cDNA) from the mammalian gene collection were purchased from various sources: RIKEN BRC for OATP1A2, OATP2A1, OATP4A1, and OATP4C1; Horizon Discovery for OATP1B1, OATP2B1, OATP3A1, and OATP5A1; and Vector Builder for OATP1A2 and OATP1B3. Human ileum total RNA isolated from normal tissue was obtained from OriGene Technologies (#CR559756) and BioChain (#R1234101). Similarly, all human colon total RNAs were obtained from BioChain (ascending colon #R1234091; transverse colon #R1234096; descending colon #R1234092; sigmoid colon #R1234095). All other chemicals used in this study were of reagent grade and commercially available.

Cell culture of Caco-2 cells

Caco-2 cells were cultured in DMEM supplemented with 10% FBS, 1% nonessential amino acids, 100 U/ml penicillin, 0.1 mg/ml streptomycin, and 50 μg/L gentamicin. The cells were seeded in Cellmatrix Type I-A (Nitta Gelatin) collagen-coated 24-well plates at a density of 1.0 × 10^5^ cells/well and incubated at 37 °C in a humidified atmosphere with 5% CO_2_. The culture medium was replaced every 2 to 3 days, and uptake assays were conducted on day 7 of culture.

Organoid culture

A human colorectal organoid was established from the healthy part of a surgical specimen collected from a patient with colorectal cancer at Gunma University Hospital after obtaining their informed consent and study approval from the ethical committees (HS2022–054). We cultured healthy human colonic organoids as described previously (67). In brief, colonic epithelial fragments were washed vigorously with ice-cold PBS and were treated with 2.5 mM EDTA for 45 min at 4 °C with gentle rocking to release crypts. Isolated crypts were suspended in Matrigel (Corning), dispensed onto 24-well plates as 50 μl droplets, and overlaid with the medium described below. Advanced Dulbecco’s modified Eagle’s medium/F12 (Thermo Fisher Scientific) was supplemented with penicillin and streptomycin (Thermo Fisher Scientific), 2 mM GlutaMAX (Thermo Fisher Scientific), 10 mM Hepes (Thermo Fisher Scientific), 1 × B-27 supplement (Thermo Fisher Scientific), 1 mM N-acetylcysteine (Sigma-Aldrich), 50% Wnt3a conditioned medium (homemade prepared as described in a previous article) (68), 50 ng ml^−1^ recombinant mouse EGF (Thermo Fisher Scientific), 100 ng ml^−1^ recombinant mouse Noggin (PeproTech), 1 μg ml^−1^ recombinant human R-spondin-1 (R&D), 500 nM A83-01 (Tocris), 100 ng ml^−1^ recombinant human IGF-1 (BioLegend), and 50 ng ml^−1^ recombinant human FGF-basic (FGF-2) (PeproTech).

Plasmid preparation

The vectors containing the cDNA of various transporters, including SLC10A2 (GeneBank code: BC130523), OATP1A2 (GeneBank code: NM_134431.5), OATP1B1 (GeneBank code: BC114376), OATP1B3 (GeneBank code: NM_019844.3), OATP2A1 (GeneBank code: BC041140), OATP2B1 (GeneBank code: BC041095), OATP3A1 (GeneBank code: BC000585), OATP4A1 (GeneBank code: BC015727), and OATP5A1 (GeneBank code: BC137424), were subcloned into the pEF6/V5-His-A vector (Life Technologies) using the In-Fusion HD Cloning Kit (Takara) following the manufacturer’s instructions. The coding sequence of OATP6A1 (GeneBank code: NM_00114594.2) was synthesized and cloned between 5′ BamHⅠ and 3′ EcoRⅠ site of pEF6/V5-His-A vector (outsourced to Azenta). Verification of all constructs was performed through DNA sequencing.

Cell culture and transient transfection of COS-7 and HEK293T cells

COS-7 cells and HEK293T cells were cultured in DMEM supplemented with 10% FBS, 100 U/ml penicillin, and 0.1 mg/ml streptomycin at 37 °C in a humidified atmosphere with 5% CO_2_. Cells were seeded in 24-well plates at a density of 0.5 × 10^5^ cells/well. For HEK293T cells, the plates were coated with Cellmatrix Type I-A collagen (Nitta Gelatin). Transient transfection was performed 24 h after seeding using Lipofectamine 3000 (Thermo Fisher Scientific) according to the manufacturer’s instructions, with cells transfected with either 0.5 μg of transporter cDNA in the pEF6/V5-HisA vector or mock DNA. Twenty-four hours post transfection, the cells were subjected to uptake assays.

Preparation of the uptake assay solutions

For both sodium-dependent and sodium-independent uptake assays, we prepared uptake assay solutions with a fixed concentration of [^3^H]-TC (36 nM) and variable concentrations of unlabeled sodium taurocholate. In sodium-dependent uptake assays, the concentration of unlabeled sodium taurocholate was maintained at 5 μM. For concentration-dependent uptake assays, concentrations of unlabeled sodium taurocholate were adjusted to 0, 2.5, 5, 10, 15, 25, 50, 75, 100, 125, 150, 175, and 200 μM. These substrates were diluted in Hank’s balanced salt solution supplemented with Hepes (pH 7.4) or Mes (pH 6.0). NaCl-free buffers were prepared for sodium-dependent uptake assays by replacing NaCl with choline chloride. The buffers used in this study are listed in Table S1.

For all TC uptake assays conducted in COS-7 and HEK293T cells transfected with ASBT or OATP, we used assay solutions with a fixed concentration of [^3^H]-TC (36 nM). In concentration-dependent uptake assays, we adjusted the concentrations of unlabeled sodium taurocholate to 0, 5, 10, 25, 50, 75, and 100 μM. For screening purposes to identify OATPs involved in TC uptake, we maintained the concentration of unlabeled sodium taurocholate at 5 μM. These substrates were diluted in Hank’s balanced salt solution supplemented with Hepes (pH 7.4) or Mes (pH 6.0) (Table S1).

For OATP2A1, the uptake assays for PGE_2_ and E3S were prepared as follows: the PGE_2_ solution comprised 1.5 nM [^3^H]-PGE_2_ and 1 μM unlabeled PGE_2_; for OATP1A2, the E3S solution consisted of 28 nM [^3^H]-E3S and 272 nM unlabeled E3S; for the other OATPs, the E3S solution contained 20 nM [^3^H]-E3S. All substrates were diluted in Hank’s balanced salt solution supplemented with Mes (pH 6.0).

Uptake assay in Caco-2 cells and COS-7 cells

The uptake assays were conducted following the procedures outlined in our previous publication (13). In brief, cells were washed twice with the appropriate buffers at 37 °C and incubated for a minimum of 15 min to ensure cellular equilibration with the buffer conditions. The uptake assay commenced with the addition of 400 μl of each uptake assay solution to the cells, followed by further incubation at 37 °C for specified durations: 15 min for all bile acid uptake assays and 10 min for PGE_2_ and E3S uptake assays. Termination of the uptake assay involved removing the solutions and performing two additional washes of the cells with ice-cold buffer corresponding to the buffer used in each uptake solution preparation. Subsequently, the cells were lysed in 0.1 N ice-cold NaOH solution, kept on ice for 1 h, and the radioactivity was quantified using a liquid scintillation counter Tri-Carb4810 TR (PerkinElmer).

Kinetic analysis

The kinetic parameters for transport activity were determined by fitting the data with nonlinear least squares using GraphPad Prism 10.1.2 (Dotmatics). Equation 1 was used for monophasic uptake, whereas Equation 2 was used for biphasic uptake. The selection of the appropriate equation for fitting was guided by the Eadie–Hofstee plot and AIC (69).

$$[eqn]$$ $$[eqn]$$

where V, S, Km, and Vmax represent the initial uptake rate, substrate concentration, Michaelis constant, and maximum uptake rate, respectively. The subscript numbers 1 and 2 indicate the saturable components (transporters) of high- and low-affinity, respectively. Total protein concentration in Caco-2 cells and COS-7 cells was determined using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific) according to the manufacturer’s instructions.

Development of the organoid monolayer

The organoid monolayer was developed based on a previously reported method (70). In brief, organoids were detached by TrypLeTM Express (Thermo Fisher Scientific) and pipetted to single cells. Then, 1.0 × 10^6^ cells were seeded on a 24-well Transwell insert (pore size, 0.4 μm, Griner Bio-One) coated with Matrigel and cultured in 200 μl of the same medium used for the organoid culture, but supplemented with 10% FBS (Biosera) and lacking both IGF-1 and FGF-2. In total, 500 μl of the same medium was added to the basolateral chamber. The medium was replaced daily for 4 days until the formation of a confluent monolayer. Monolayers with TEER >135 Ω cm^2^ were used for immunocytochemistry and the TC transport assay.

Assay of TC transport in organoid monolayer

The organoid monolayer was washed twice with Hank's Balanced Salt Solution (HBSS) (pH 7.4, 37 °C) and preincubated for 15 min. Prior to the assay, 500 μl of HBSS was added to the basolateral chamber. The TC transport assay was initiated by adding 200 μl of TC solution (i.e., [3H]-TC 360 nM and TC 5 μM) to the apical-side chamber and incubating it for 120 min. At 5, 10, 15, 30, 45, 60, 90, and 120 min, 200 μl of HBSS was sampled from the basal-side chamber, and an equal volume of HBSS (37 °C) was immediately added. At 120 min, organoids were washed twice with ice-cold HBSS, and 0.1 N NaOH was added to dissolve the organoids on ice for 1 h. Radioactivity was quantified using a liquid scintillation counter, Tri-Carb 4810 TR (PerkinElmer), and the cumulative total amount of TC transported to the basolateral compartment was calculated.

RT-qPCR analysis

Total RNA was extracted from Caco-2 cells, HEK293T cells, and organoids transiently transfected with transporter DNA or mock DNA using TRIzol Reagent (Thermo Fisher Scientific). For human colon samples, total RNA isolation was performed using the RNeasy Mini Kit (QIAGEN). Subsequently, reverse transcription was conducted using the PrimeScript RT Master mix (Takara). RT-qPCR analysis was conducted using a Thermal Cycler Dice Real-Time PCR system (Takara) with TB Green Premix Ex Taq II and gene-specific primers listed in Table S2. Gene expression levels were normalized to either β-actin or GAPDH, and relative expression was determined using the 2^−ΔΔCt^ method.

Western blotting

Caco-2 cells were cultured in a 100-mm dish at a density of 2.0 × 10^5^ cells/ml and maintained for 1 or 2 weeks. Subsequently, cell lysis was performed using radioimmunoprecipitation assay (RIPA) buffer (Fujifilm Wako Pure Chemicals) containing 1% protease inhibitor cocktail (Sigma-Aldrich). For organoid, the protein was extracted using TRIzol Reagent (Thermo Fisher Scientific) from organic phase according to the manufactures’ instructions. The protein concentration was determined using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). Samples containing 20 μg of protein were subjected to SDS-PAGE, followed by transfer onto iBlot 3 Transfer Stacks Midi PVDF membranes using the iBlot Western Blot Transfer System (Thermo Fisher Scientific). After blocking with Bullet Blocking One for western blotting for 5 min at 20 °C (Nacalai tesque), membranes were incubated with primary antibodies at 4 °C overnight. Then, the membranes were washed with TBS-0.1% Tween, followed by incubation with secondary antibodies at 4 °C overnight. After detecting ASBT, the antibody was stripped with EzReprob (ATTO) according to the manufacturer’s instructions, blocked, and reproved with OATP1A2 or OATP1B3 antibody. After detections, the antibody was stripped, and the membrane was blocked again for incubation with the β-actin antibody. Details of all antibodies and their respective working dilutions are listed in Table S3. Protein bands were visualized using the ImageQuant LAS 4000 system (Fujifilm) with ECL Prime Western Blotting Detection Reagents (Cytiva), and each band was quantified using ImageJ.

Immunocytochemistry

Caco-2 cells were seeded at 2.0 × 10^5^ cells/ml in collagen-coated 4-well slide chambers and cultured for 1 or 2 weeks. For immunostaining, cells grown on slide chambers were washed with PBS and then fixed with ice-cold methanol for 10 min at −20 °C. Cells were then permeabilized by 0.1% Triton X-100/PBS for 15 min at room temperature and blocked with EzBlock BSA (ATTO) containing 0.1% Triton X-100/PBS for 1 h at room temperature. Cells were incubated with either primary antibodies against ASBT and ezrin or those against OATP1B3 and ezrin at 4 °C overnight. After being washed with 0.1% Triton X-100/PBS, cells were incubated with secondary antibodies for 90 min at room temperature. Nuclei were stained using DAPI (Thermo Fisher Scientific). Cells were mounted with ProLong Diamond (Thermo Fisher Scientific) and visualized using a confocal laser scanning microscope (LSM5 PASCAL, Zeiss).

Organoids were washed four times with 0.1% Triton X-100/PBS and fixed with 4% paraformaldehyde in PBS (Fujifilm Wako Pure Chemicals) for 15 min at room temperature. Then, they were washed again three times with 0.1% Triton X-100/PBS and blocked with 1% Block Ace Powder (Snow Brand Milk Products Co., Ltd) for 1 h at room temperature. After another wash with 0.1% Triton X-100/PBS, the organoids were incubated with OATP1B3 and ZO-1 primary antibodies diluted in 1% Block Ace powder in 0.1% Triton X 100/PBS at 4 °C overnight. After a final washing with 0.1% Triton X-100/PBS, they were incubated with secondary antibodies for 2 h at room temperature in the dark. Nuclei were stained using DAPI (Invitrogen), and F-actin, which serves as an apical membrane marker (71), was stained with phalloidin (Invitrogen). Transwell inserts were cut out and mounted on a slide glass with SlowFade Diamond Antifade Mountant (Invitrogen) and visualized using a confocal laser scanning microscope (Fluoview FV3000, Olympus). Details of all antibodies used and their respective working dilutions are listed in Table S4.

ALP activity assay

Caco-2 cells were seeded on the Cellmatrix Type I-A (Nitta Gelatin) collagen-coated 96-well plate at a density of 0.15 × 10^5^ cells/well and cultured for 1, 2, and 3 weeks, with medium changed every 2 or 3 days. The number of seeded cells corresponded to that seeded on a 24 well plate (1 × 10^5^ cells/well). After the incubation period, the ALP activity assay was performed using the TRACP & ALP Assay Kit (Takara Bio) with Bacteria ALP (Takara Bio, #2120A) as a standard, according to the manufacturer’s instructions. ALP activity was normalized to the protein content and expressed as unit/μl/mg protein. Enzyme activity was defined as one unit when 1 μmol of p-nitrophenol was liberated per minute at 25 °C and pH 8.0. Protein content was determined using the Pierce BCA assay Kit (Thermo Fisher Scientific).

Statistics

The data are presented as mean ± standard deviation (SD) derived from at least three independent replications. Statistical analyses were conducted using unpaired t test if there were only two groups and using one-way ANOVA followed by Tukey’s test if there were three groups (GraphPad Prism 10.4., GraphPad Software). Sample size and statistical analysis methods used for each experiment are detailed in each figure legend. A significance threshold of p < 0.05 was applied to determine statistical significance.

Data availability

All data are contained within the manuscript and Supporting information.

Supporting information

This article contains supporting information.

Conflict of interest

The authors declare that they have no conflicts of interest with the contents of this article.