Advertisement
Research ArticleGASTROINTESTINAL, HEPATIC, PULMONARY, AND RENAL

Region-Dependent Modulation of Intestinal Permeability by Drug Efflux Transporters: In Vitro Studies in mdr1a(−/−) Mouse Intestine

R. H. Stephens, J. Tanianis-Hughes, N. B. Higgs, M. Humphrey and G. Warhurst
Journal of Pharmacology and Experimental Therapeutics December 2002, 303 (3) 1095-1101; DOI: https://doi.org/10.1124/jpet.102.041236

Abstract

Information on the extent to which xenobiotics interact with P-glycoprotein (PGP) during transit through the intestine is crucial in determining the influence of PGP on oral drug absorption. We have recently described a novel use of isolated ileum from PGP-deficient mdr1a(−/−) mice to resolve PGP- and non-PGP-dependent drug efflux and provide a definitive measure of intrinsic drug permeability without recourse to inhibitors (Stephens et al., 2002). The present study uses this approach to investigate the impact of PGP on intestinal permeability of paclitaxel and digoxin in different regions of the mouse intestine (jejunum, ileum, and proximal and distal colon). Absorption of paclitaxel and digoxin in tissues from wild-type mice was low and showed little regional variation. In contrast, absorption of both drugs was markedly higher in mdr1a(−/−) intestine, although the increase was highly region-dependent, with the ileum and distal colon showing the greatest effect and much smaller changes in the jejunum and proximal colon. These effects were accompanied by the abolition of paclitaxel and digoxin secretion in mdr1a(−/−) mice, suggesting that regional variations in intestinal permeability are masked by differential PGP expression, confirmed by immunoblotting studies. Propranolol permeability, which is not influenced by PGP, showed similar regional variation in both wild-type and mdr1a(−/−) tissues, suggesting that differences are at the level of transcellular permeability. These data suggest that the ileum and the distal colon are regions of relatively high transcellular permeability for xenobiotics that are compensated by enhanced expression of PGP.

The xenobiotic transporter P-glycoprotein (PGP), the protein product of the MDR1 gene, was initially associated with the development of multidrug resistance in tumor cells (Hunter and Hirst, 1997). It is now understood that this ABC protein forms an integral part of the intestinal barrier together with several other polyspecific efflux transporters, including members of the multidrug resistance-related protein (MRP) family and breast cancer resistance protein (BCRP) (Hunter and Hirst, 1997; Makhey et al., 1998; Gotoh et al., 2000;Jonker et al., 2000; Litman et al., 2001). The ability of PGP to interact with a broad range of natural and synthetic xenobiotics, drugs, and naturally occurring toxins or food constituents (Hunter and Hirst, 1997; Walle and Walle, 1999; Litman et al., 2001) strongly suggests that its ability to limit oral drug absorption is part of a broader protective role as a modulator of intestinal permeability. Indeed, growing awareness of the wider physiological context in which PGP resides has linked this and other efflux transport proteins to the regulation of electrolyte transport, apoptosis, and inflammation (Johnstone et al., 2000).

Clearer definition of these physiological and pharmacological roles, especially those directly related to xenobiotic transport, requires a better understanding of the extent to which compounds interact with PGP as they move sequentially along the cephalocaudal axis of the gut. This will depend on how transporter activity varies along the gut and whether non-PGP transporters contribute to efflux. MRP and BCRP can interact with PGP substrates and appear to be expressed heterogeneously in the intestine (Jonker et al., 2000; Mottino et al., 2000; Litman et al., 2001; Maliepaard et al., 2001). Several other variables including membrane composition and passive epithelial permeability may also be important in determining which xenobiotics are transported effectively (Brasitus and Schachter, 1984; Fagerholm et al., 1997; Ungell et al., 1998).

The way in which the activity of PGP and other efflux transporters varies along the intestine remains poorly defined. Several studies have noted significant regional differences in PGP expression at the mRNA level with expression increasing from the small intestine to the colon (Fojo et al., 1987; Fricker et al., 1996; Li et al., 1999). However, other studies have reported a different distribution (Cordon-Cardo et al., 1990; Chianale et al., 1995), which may in part be due to species differences in transporter expression. It is also noteworthy that little work has been done to examine regional variation of any transporter within the large intestine, a physiologically important site for xenobiotic efflux (Penny and Campbell, 1994).

We have recently described a new in vitro approach using isolated ileum from wild-type mice and mdr1a(−/−) mice, which express no functional PGP in the intestine (Stephens et al., 2002). This model provides functional analysis of PGP without recourse to inhibitors and allows a compound's true passive permeability across the gut epithelium to be defined. In addition, the contribution of non-PGP transporters can be more readily identified. Using an extension of this approach, the present study has sought to produce a functional profile of PGP activity and expression along the cephalocaudal axis of the mouse small and large intestine. We have also sought to investigate aspects of the broader role that PGP might play in controlling the passive permeability of the gut epithelium to xenobiotics.

Materials and Methods

Materials.

[G-3H]Digoxin andl-[4-3H]propranolol were purchased from PerkinElmer Life Sciences (Hounslow, Middlesex, UK), and [G-3H]paclitaxel was obtained from Moravek Biochemicals, Inc. (Brea, CA).d-[1-14C]Mannitol was obtained from Amersham Biosciences UK, Ltd. (Little Chalfont, Buckinghamshire, UK). MK571 (3-([{3-(2-[7-chloro-2-quinolinyl]ethenyl)phenyl}-{(3-dimethyl)-amino3-oxopropyl)-thio}-methyl]thio}-methyl]thio)propionic acid) was purchased from Affiniti Research Products Ltd. (Exeter, UK). All other compounds were obtained from Sigma-Aldrich Chemical Co. Ltd. (Poole, UK). Wild-type FVB [mdr1a(+/+)] mice were obtained from local barrier maintained stock. Transgenic mdr1a(−/−) mice were obtained from Taconic M&B A/S (Bomholtgärd, Denmark).

Animals and Tissues.

Nonfasting male mice (10–20 weeks, 20–36 g) were killed by cervical dislocation. Intestinal tissue from the region of interest was then immediately removed and the lumen flushed with ice-cold, bicarbonate-buffered Ringer solution containing 146 mM−1 Na+, 4.2 mM−1 K+, 1.2 mM−1 Ca2+, 1.2 mM−1 Mg2+, 125.8 mM−1 Cl−, 26.6 mM−1HCO3 −, 1.2 mM−1HPO4 2−, 0.2 mM−1H2PO4 −, and 10 mM d-glucose, which had been equilibrated to pH 7.4 by bubbling with 5% CO2/95% O2.

Intestinal regions used for permeability studies were defined as follows. For small intestine, a segment beginning 13 cm distal to the ligament of Treitz and extending distally for approximately 13 cm was labeled as distal jejunum. A further segment beginning 26 cm distal to the ligament of Treitz and extending approximately 13 cm to the ileocecal junction was labeled as ileum. Permeability studies were routinely performed on four pieces of tissue from each of the two regions of small intestine. For colon, the tissue from the ileocecal junction to the rectum was divided into two equal halves of approximately 4 cm in length representing proximal and distal colon. In this case, permeability was measured routinely in two pieces of tissue from each of the two sections of colon (see Fig.1).

Figure 1
Figure 1

Intestinal regions used in this study. A diagram of the isolated mouse intestine shows approximate juxtaposition of the distal jejunum, ileum, and proximal and distal colon. Diagram is not to scale, although relative sizes are approximately correct.

Tissue segments were mounted intact in modified Ussing chambers (0.52 cm2 cross-sectional area) without removal of the serosal muscle layer (Stephens et al., 2002). Mounting was completed within 20 min of removal from the animal. All procedures involving animals conformed to current UK Home Office legislation.

Permeability Studies.

Drug transport across intestinal tissues was measured by methods similar to those described previously (Collett et al., 1999; Stephens et al., 2002). Intestinal mucosa was bathed on the mucosal (apical) and serosal (basolateral) surfaces with 5 ml of bicarbonate-buffered Ringer, pH 7.4 (as above). Spontaneous tissue open-circuit potential difference, short-circuit current (ISC), and transepithelial electrical resistance (RT) were monitored periodically throughout the experiment; otherwise, tissues were maintained under open-circuit conditions. A 30-min equilibration period was allowed prior to beginning permeability measurements to allow stabilization of electrical parameters. Unidirectional apical (A) to basolateral (B) and B-A permeability of digoxin, paclitaxel, or propranolol was measured following addition of radiolabeled (0.2 μCi · ml−1, 7.4 kBq · ml−1), and unlabeled drug (20–100 μM) to A or B chambers. Drugs were added as stock solutions in dimethyl sulfoxide giving a final solvent concentration of 0.02 to 0.3%. In the case of digoxin, unlabeled drug was added to an aliquot of transport buffer from the “donor” chamber, which was vortexed for 1 min and returned to the chamber to ensure thorough mixing. For all three compounds, 1-ml samples were removed from the “receiver” chamber att = 0 and after each of six 40-min periods and replaced with fresh transport buffer. Samples (100 μl) were also taken from the donor chamber at the beginning of the first period and at the end of the experiment to monitor any changes in donor drug concentrations during the experiment and to safeguard mass balance.

Samples were analyzed by liquid scintillation counting. Results are expressed as transepithelial apparent permeability (Papp) in centimeters per second, given by:Embedded Image where dQ/dt is the rate of appearance of compound in the receiver chamber, C is the substrate concentration in the donor chamber, and A is the cross-sectional area of the tissue (0.52 cm2). dQ/dt is derived from the rate of appearance of the radioisotope-labeled form of the compound and the relative initial concentrations of labeled and unlabeled compound in the donor chamber. Papp values are either unidirectional, i.e., absorptive, (A-B) or secretory (B-A) or net (PappB-A − PappA-B). If the ratio of PappB-A/PappA-B is greater than 1, this indicates that net secretion (B-A asymmetry or efflux) is taking place. Where values for net Papp are shown, positive values (i.e., B-A > A-B) indicate secretion, whereas negative values represent net absorption in the A-B direction (i.e., B-A < A-B).

The results as expressed are the mean of the Pappvalues over the indicated number of flux periods, usually three. In the case of all three drugs (digoxin, paclitaxel, and propranolol), the flux was shown to remain linear over the time course of the experiment. In some experiments using mdr1a(+/+) tissues, the PGP inhibitor quinidine (200 μM) was added to both apical and basolateral chambers after the third 40-min flux period. The data from three to six tissue sections in which Papp has been measured in each of the A-B or B-A directions are then pooled.

Integrity of the paracellular route in mouse tissues was assessed using mannitol. Following tissue equilibration, 100 μM mannitol, containing 0.2 μCi · ml−1[14C]mannitol, was added to the apical chamber, and samples were taken from the basolateral chamber at 60-min intervals. Papp was calculated as described above.

Viability of Mouse Tissues.

Changes in transepithelial resistance, RT, were used as the primary indicator of tissue viability, and RT values were monitored throughout the course of the experiment. Distal colon exhibited the highest RT [mdr1a(+/+), 124.2 ± 5.; mdr1a(−/−, 130.5 ± 7.6 Ω · cm2], whereas the other three segments were lower and similar to each other [mdr1a(+/+), DJ: 64.9 ± 2.7, I: 72.7 ± 1.5, PC: 86.84 ± 2.9 Ω · cm2; mdr1a(−/−), DJ: 86.3 ± 4.3, I: 75.0 ± 1.6, PC: 81.3 ± 3.0 Ω · cm2]. RT values did not vary significantly during the experiment, remaining within 15% of the starting value, and tissues were excluded where RT fell by more than this. As a further test of tissue viability, the cAMP agonist, forskolin (1 μM), was added basolaterally at the conclusion of each experiment. This elicited a sharp and sustained rise in Isc (>25% increase) in viable tissues, indicating stimulation of electrogenic Cl− secretion (Warhurst et al., 1996). Any tissue segments that failed to respond to forskolin at the end of the experiment were discarded.

Immunoblot Analysis of P-Glycoprotein Expression.

Epithelial cells were isolated from the same four regions of mouse intestine described above using the procedures below, which gave maximum cell yield. Segments of small and large intestine were flushed twice with ice-cold Ca2+- and Mg2+-free bicarbonate-buffered Ringer solution [121 mM−1 NaCl, 25 mM−1NaHCO3, 1.6 mM−1KHCO3, 1.2 mM−1K2HPO4, 0.2 mM−1KH2PO4, 10 mM−1d-glucose, 0.05 dithiothreitol, pH 7.4]. Segments were everted over a glass rod and incubated for 30 min (small bowel) or 60 min (large bowel) in Ca2+- and Mg2+-free Ringer as above, continuously gassed with 5% CO2/95% O2, at 4°C. Tissue was then transferred to Ca2+- and Mg2+-free bicarbonate-buffered Ringer solution (as above) containing 5 mM EDTA (jejunum, ileum) or 10 mM EDTA (colon) and incubated for 15 min at 4°C. Following this, segments were attached to a Vibromixer (Chemap AG, Volketswil, Switzerland) and vibrated at 60 Hz for 2 to 5 min in 30-s bursts, replacing the EDTA solution after every two or three bursts. Light microscopy of the bathing solution fractions revealed that this procedure isolated primarily villus cells, mainly as intact units. Once fractions began to contain significant numbers of crypts and little or no villus material was evident, the bathing solution was warmed to room temperature and the surface of the tissue was subjected to gentle mechanical disruption by rubbing it with a fine glass rod. This resulted in copious production of free crypts. Once more, sequential fractions of bathing solution were examined by light microscopy to determine the point at which little or no additional material could be removed. At this point, all the collected fractions were pooled, washed three times by centrifugation at 2000g/4°C in complete bicarbonate-buffered Ringer solution without dithiothreitol (seeAnimals and Tissues) and stored at −70°C until required.

Epithelial preparations were suspended in lysis buffer (120 mM NaCl, 5 mM HEPES, pH 7.5, 1% Triton X-100, 2 mM EDTA, 25 mM NaF, 1 mM NaVO4) containing a cocktail of protease inhibitors (50 μl/ml buffer; Sigma-Aldrich, catalog number P8340). The cell suspension was left on ice for 30 min, sonicated for 3 s, and centrifuged at 13,000g for 30 min at 4°C. The resulting supernatant was retained and the protein concentration determined by the bicinchoninic acid method (Pierce, Rockford, IL). Aliquots (10 μg) of supernatant protein were fractionated on 10% SDS-polyacrylamide gel electrophoresis. Gels were transferred to Hybond ECL (enhanced chemiluminescence) nitrocellulose membrane (Amersham Biosciences UK, Ltd). Membranes were soaked in TBS-T [50 mM Tris-HCl, pH 7.9, 150 mM NaCl, and 0.05% (v/v) Tween 20] containing 2% (w/v) milk protein at 4°C overnight. Membranes were subsequently washed in TBS-T and incubated for 2 h with a mixture of monoclonal anti-PGP antibody (C219, 1:400 dilution in TBS-T; Alexis Corp., San Diego, CA) and monoclonal anti-β-actin antibody (clone AC15, 1: 5000 dilution in TBS-T; Sigma-Aldrich). The secondary antibody was goat anti-mouse IgG horseradish peroxidase-conjugated antibody (Bio-Rad Laboratories Ltd., Hemel Hempstead, UK) used at a dilution of 1:5000. Blots were developed using the ECL system (Amersham Biosciences UK, Ltd.).

Statistical Methods.

Values are expressed as mean ± S.E.M. (n). Statistical analyses were carried out using Arcus Quickstat (Research Solutions, Cambridge, UK) Papp values and the effects of inhibitors on substrate fluxes were tested for significance using one-way ANOVA with an appropriate post hoc test (Bonferroni) with a significance level of 5%.

Results

PGP Ablation Unmasks Differences in Regional Permeability of Paclitaxel and Digoxin along Mouse Intestine.

Initial studies used two compounds known to interact with PGP, paclitaxel and digoxin, to investigate the effects of PGP on drug permeability in different regions of the mouse intestine. In mdr1a(+/+) mice, paclitaxel exhibited a low absorptive permeability (A-B) along the intestine with relatively small regional variations, although ileal permeability was significantly higher (2.26 ± 0.2 × 10−6 cm · s−1,n = 5) than any other region (p < 0.05) with proximal colon having the lowest permeability (1.08 ± 0.2 × 10−6 cm · s−1, n = 5) (Fig.2A). Tissues from mdr1a(−/−) mice showed significantly higher paclitaxel absorptive permeability throughout the intestine, although the degree of increase varied markedly from region to region (Fig. 2A). Large increases in paclitaxel absorption were observed in the ileum and distal colon with permeability ≈5-fold higher than in the equivalent tissues from mdr1a(+/+) animals. Increases in permeability in distal jejunum and proximal colon were more modest, being only 2-fold higher. Comparison of paclitaxel's efflux in mdr1a(+/+) and mdr1a(−/−) tissues show an opposite profile (Fig. 2B). mdr1a(+/+) tissues showed high levels of paclitaxel secretion, which varied markedly along the intestine, being greatest in ileum and distal colon and lowest in distal jejunum and proximal colon, corresponding with regions of high A-B permeability in mdr1a(−/−) intestine. In contrast, paclitaxel secretion was completely abolished in mdr1a(−/−) tissues from all intestinal regions (Fig. 2B). These data confirm PGP as the sole mediator of paclitaxel secretion in small and large intestine tissues. Removal of PGP reveals marked, region-dependent differences in intestinal permeability of this compound.

Figure 2
Figure 2

Regional differences in paclitaxel Pappin isolated intestine segments. A, unidirectional absorptive (A-B) Papp of 20 μM paclitaxel in isolated segments of the distal jejunum (DJ), ileum (I), proximal colon (PC), and distal colon (DC) of mdr1a(+/+) mice and mdr1a(−/−) mice in Ussing chambers. ★,p < 0.001 with respect to distal jejunum, †,p < 0.01 with respect to proximal colon, ANOVA. Absorptive Papp was significantly higher in mdr1a(−/−) than in mdr1a(+/+) tissues in all regions (p < 0.05, ANOVA). B, net secretory Papp of 20 μM paclitaxel in the same intestinal regions as in A. ★, p < 0.001 with respect to distal jejunum, †, p < 0.01 with respect to proximal colon, ANOVA. Net secretory Papp was significantly lower in mdr1a(−/−) than in mdr1a(+/+) intestine in all regions (p < 0.01, ANOVA). Values are mean Papp ± S.E.M. in centimeters per second,n = 3 to 5 tissues measured A-B and 3 to 4 tissues measured B-A in each region; total animals, 10 mdr1a(+/+) and 11 mdr1a(−/−).

If the observed variations in intestinal paclitaxel permeability between mdr1a(+/+) and mdr1a(−/−) mice are due specifically to the removal of PGP, we reasoned that it should be possible to reproduce the region-dependent increases in paclitaxel absorption in mdr1a(+/+) tissues using the PGP inhibitor quinidine. Figure3 shows the effect of 200 μM quinidine on the absorptive permeability of paclitaxel in mdr1a(+/+) tissues. Although there are quantitative differences between the inhibitor-induced increases in permeability and those seen in mdr1a(−/−) mice, the region-specific increases in paclitaxel permeability observed in mdr1a(−/−) tissues are clearly reproduced by treatment with quinidine.

Figure 3
Figure 3

Regional differences in the effect of quinidine on paclitaxel absorptive Papp in isolated intestine segments. Shown is the unidirectional absorptive (A-B) Papp of 20 μM paclitaxel in isolated segments of distal jejunum (DJ), ileum (I), proximal colon (PC), and distal colon (DC) of mdr1a(+/+) mice in Ussing chambers, in the presence and absence of 200 μM quinidine. A-B Papp for the corresponding regions of intestine from mdr1a(−/−) mice is shown for comparison (data from Fig. 2). Absorptive paclitaxel Papp was significantly increased by quinidine in all regions with respect to control (p< 0.05, ANOVA). ★, p < 0.01 with respect to distal jejunum; †, p < 0.01 with respect to proximal colon, ANOVA. Values are mean Papp ± S.E.M. in centimeters per second, n = 3 to 5 tissues measured A-B and 3 to 4 tissues measured B-A in each region; total animals, 10 mdr1a(+/+) and 11 mdr1a(−/−).

To determine whether the differences in permeability unmasked by PGP ablation are also seen with other drugs, regional permeability of the cardiac glycoside, digoxin, was studied in mdr1a(+/+) and mdr1a(−/−) tissues (Fig. 4A). Digoxin permeability decreased from small intestine to colon in mdr1a(+/+) animals [distal jejunum, 2.84 ± 0.3; ileum, 3.06 ± 0.6; proximal colon, 1.73 ± 0.7, and distal colon, 1.64 ± 0.4 × 10−6 cm · s−1(n = 3–4 in each group)]. Digoxin absorptive permeability was significantly higher in mdr1a(−/−) tissues with a regional profile similar to that seen with paclitaxel, with the greatest increase occurring in ileum and distal colon. It was noticeable, however, that increases in digoxin absorptive permeability (ranging from 1.5–3 fold) in mdr1a(−/−) tissues were lower than those seen with paclitaxel (2- to 5-fold). The abolition of digoxin secretion in mdr1a(−/−) tissues (Fig. 4B) confirms the importance of this transporter in modulating digoxin absorption along the gut. Once again, the distribution of secretory activity corresponded with regions of high A-B permeability and mirrored that seen with paclitaxel.

Figure 4
Figure 4

Regional differences in digoxin Papp in isolated intestine segments. A, unidirectional absorptive (A-B) Papp of 40 μM digoxin in isolated segments of the distal jejunum (DJ), ileum (I), proximal colon (PC), and distal colon (DC) of mdr1a(+/+) mice and mdr1a(−/−) mice in Ussing chambers. As was the case with paclitaxel, absorptive Papp was significantly higher in mdr1a(−/−) than in mdr1a(+/+) tissues in all regions (p < 0.05, ANOVA). ★, p < 0.01 with respect to distal jejunum. B, net secretory Pappof 40 μM digoxin in the same intestinal regions as in A. ★,p < 0.01 with respect to distal jejunum; †,p < 0.01 with respect to proximal colon, ANOVA. Net secretory Papp was significantly lower in mdr1a(−/−) than in mdr1a(+/+) intestine in all regions (p < 0.01, ANOVA). Values are mean Papp ± S.E.M. in centimeters per second, n = 3 to 4 tissues measured A-B and 3 to 4 tissues measured B-A in each region; total animals, 11 mdr1a(+/+) and 11 mdr1a(−/−).

Molecular Expression of PGP in Mouse Intestine.

To complement the functional studies described above, the molecular expression of PGP was measured in epithelial cell preparations from the four regions by Western blotting. Figure 5 shows regional expression of PGP for epithelial preparations from mdr1a(+/+) mice standardized against β-actin. In the small intestine, the ileum exhibits the highest PGP expression while in the large intestine maximum expression occurs in the distal colon. Therefore, the distribution of PGP expression is consistent with the functional data on paclitaxel and digoxin permeability.

Figure 5
Figure 5

Regional differences in P-glycoprotein expression determined by immunoblotting of purified intestinal epithelial fractions. PGP (170-kDa band) showed markedly greater band intensity in ileum (I) compared with distal jejunum (DJ) and in distal colon (DC) compared with proximal colon (PC). The band intensity for β-actin was relatively constant across all samples (42-kDa band). Data shown are representative of epithelial preparations from three animals.

Propranolol Exhibits Regional Differences in Absorptive Permeability in Both mdr1a(+/+) and mdr1a(−/−) Tissues.

The above data indicate that PGP is the only efflux transporter involved in modulating paclitaxel and digoxin permeability along the mouse intestine. As a result, the absorptive permeability of these compounds measured in mdr1a(−/−) tissues should be indicative of their true passive permeability across the gut epithelium. One explanation for the regional differences in digoxin and paclitaxel permeability in mdr1a(−/−) tissues is that this represents inherent differences in transcellular permeability along the gut epithelium, which are unmasked when PGP is removed. If this is the case, high-permeability compounds, such as propranolol, that cross the epithelium by a passive transcellular process and are unaffected by PGP (Stephens et al., 2002) would be expected to exhibit a similar pattern of regionality in both mdr1a(+/+) and mdr1a(−/−) tissues. Figure6A shows that the absorptive permeability of propranolol exhibits regional differences in absorptive permeability in both mouse strains that are qualitatively similar to those seen with paclitaxel and digoxin in mdr1a(−/−) tissues. Interestingly, absorptive propranolol permeability was significantly higher in mdr1a(+/+) than in mdr1a(−/−) intestine in all regions except the distal colon (p < 0.001, Fig. 6A).

Figure 6
Figure 6

Regional differences in absorptive Pappof propranolol and mannitol. A, propranolol absorptive (A-B) Papp measured across isolated segments of distal jejunum (DJ), ileum (I), proximal (PC), and distal colon (DC) from mdr1a(+/+) and mdr1a(−/−) mice in modified Ussing chambers. ★,p < 0.05 with respect to distal jejunum; †,p < 0.01 with respect to proximal colon, ANOVA. Propranolol Papp was significantly higher in mdr1a(+/+) intestine than in mdr1a(−/−) intestine (p < 0.001, ANOVA), in all regions except distal colon. Values are mean ± S.E.M. in centimeters per second, n = 4 to 8 tissues in each region from 15 mdr1a(+/+) and 14 mdr1a(−/−) animals. B, mannitol absorptive (A-B) Papp measured across isolated segments of distal jejunum (DJ), ileum (I), proximal colon (PC), and distal colon (DC) from mdr1a(+/+) and mdr1a(−/−) mice in modified Ussing chambers. ★, p < 0.05; †,p < 0.01 with respect to distal jejunum, ANOVA. Mannitol Papp was not significantly different between mdr1a(+/+) and mdr1a(−/−) intestine in any region. Values are mean ± S.E.M. in centimeters per second, n = 4 to 8 tissues in each region from 13 mdr1a(+/+) and 12 mdr1a(−/−) animals.

In contrast, paracellular permeability, assessed using mannitol, did not show this profile of regional variation (Fig. 6B), nor was there any evidence for significant differences in mannitol permeability between mdr1a(+/+) and mdr1a(−/−) tissues. Mannitol permeability was significantly higher in jejunum than in ileum in both mouse strains (p < 0.05), but with no further reduction in permeability in colonic segments. The fact that mannitol and propranolol have clearly different permeability profiles supports the contention that the regional variation in propranolol permeability, and by inference that of digoxin and paclitaxel, is specific to the transcellular route.

Discussion

Information on the extent of interaction of drug molecules with PGP and other efflux transporters as they move sequentially along the cephalocaudal axis of the intestine is crucial to understanding the likely impact of these transporters on drug absorption. This is particularly true of controlled release drugs with target sites in the lower ileum or colon. We investigated the role of PGP in regulating passive transcellular permeability along the mouse intestine in vitro. The approach used has been to compare the regional permeability of two compounds we know to be “pure” PGP substrates (Litman et al., 2001; Stephens et al., 2002) in intestinal tissues from mdr1a(+/+) (WT) and mdr1a(−/−) mice. This approach offers several advantages over existing in vitro systems: the influence of PGP on drug permeability can be characterized without recourse to inhibitors, the role of non-PGP transporters can be identified, and mdr1a(−/−) tissues can provide a definitive measure of a compound's passive A-B permeability in the absence of PGP-mediated secretion (Stephens et al., 2002). The increase in absorptive permeability when moving from WT to mdr1a(−/−) tissues may be used as an indicator of the level of expression and functional distribution of PGP in WT tissues. In WT intestine, PGP effectively restricts paclitaxel and digoxin permeability resulting in low levels of absorption that vary little from jejunum to distal colon. However, similar studies in tissues from mdr1a(−/−) animals unmask a marked increase in the absorptive permeability of both drugs which is highly region-dependent with the rank order ileum > distal colon > jejunum > proximal colon. Such findings are consistent with regional differences in passive transcellular permeability of the intestinal epithelium that are presumably masked by the parallel expression of PGP in WT tissues.

The regional distribution of PGP activity observed here conflicts with the generally accepted view that expression of PGP increases progressively from duodenum to colon, at least at the level of mRNA (Fojo et al., 1987; Fricker et al., 1996). However, other studies have observed an apparent expression peak in the ileum (Trezise et al., 1992; Chianale et al., 1995) but did not suggest a second PGP expression peak in the distal colon like that reported here. There have been relatively few studies of regional PGP transporter activity, arguably a more reliable measure. In vitro measurements of drug efflux along the rat intestine have suggested peak PGP activity in ileum with lower levels in upper small intestine and colon (Chianale et al., 1995;Yumoto et al., 1999; Kunta et al., 2000; Stephens et al., 2001). The present study, utilizing a more unequivocal measure of PGP-mediated drug transport using mdr1a(−/−) mouse intestine, offers clear evidence that PGP-mediated activity increases markedly from jejunum to ileum. As a result, when mdr1a(−/−) mouse intestine is studied, which expresses no PGP, the ileum is almost 3 times more permeable to paclitaxel than the jejunum, whereas in WT intestine, paclitaxel permeability is similar in both regions. Subdividing the colon into proximal and distal segments also highlighted a significant, and previously unsuspected, heterogeneity in PGP activity in this region. The transition from ileum to proximal colon is accompanied by a fall in PGP activity to a level similar to that observed in jejunum. However, there was a clear increase in PGP activity in distal colon, borne out by molecular evidence. The effect was particularly noticeable with paclitaxel, with a 3-fold increase in absorptive permeability compared with proximal colon in mdr1a(−/−) tissues. This finding may have implications for delivery systems that aim to target drug release to the distal intestine (Yang et al., 2002), although this would clearly depend on similar regional differences in PGP expression and drug permeability occurring in other species, particularly human. It is interesting in this respect that coadministration of the PGP inhibitor GF120918 (N-{4-[2-(1,2,3,4-tetrahydro-6,7-dimethoxy-2-isoquinolinyl)ethyl]phenyl}-9,10-dihydro-5-methoxy-9-oxo-4-acridine carboxamide) caused a 3-fold increase in the apical to basolateral absorption of somatostatin receptor peptidomimetics in rat distal colon in vitro but had no effect in proximal colon (Emery et al., 2002).

The higher levels of absorption of paclitaxel and digoxin observed in mdr1a(−/−) intestine are expected given that these compounds are subject to PGP-mediated efflux. The complete abolition of efflux of both compounds in mdr1a(−/−) intestine shows that their permeability is unaffected by other intestinal transporters and, as a result, permeability measured in mdr1a(−/−) conditions should be a measure of their “true” passive permeability in intestine. The regional variation in permeability under these conditions suggests that there are inherent differences in transcellular permeability in the gut, which are masked by the action of efflux transporters. If this is the case, a compound that is absorbed by a passive transcellular route but with intestinal permeability that is not significantly influenced by efflux transporters should exhibit a similar regional profile in both mdr1a(−/−) and WT tissues. Propranolol meets these criteria(Ungell et al., 1998; Letrent et al., 1999) and does indeed show regional differences in permeability in WT tissues qualitatively similar to those seen for paclitaxel and digoxin in mdr1a(−/−) tissues. The high permeability of propranolol compared with the other compounds may explain why the regional differences are more modest, since it would be expected to be nearer the maximal absorptive capacity of the tissue. Although the underlying reason for observed differences in passive membrane permeability in different regions is unclear, it could reflect previously described regional variations in membrane lipid composition and fluidity (Brasitus and Dudeja, 1985; Meddings 1989), which may alter intestinal permeability to lipophilic compounds. Recent evidence suggests that the level of P-glycoprotein may itself influence the distribution of cholesterol within the plasma membrane (Garrigues et al., 2002). Clearly, further studies on the lipid content of intestinal regions in different species are needed.

There was no evidence that changes in paracellular permeability played a role in these effects since mannitol displayed a regional permeability profile completely different from that of propranolol, digoxin, and paclitaxel. Mannitol permeability was significantly higher in jejunal tissues, with little difference between ileum and the two segments of the colon, and values from wild-type and mdr1a(−/−) tissues were virtually identical. The absolute values reported in this study are similar to those described by Ungell et al. (1998) in isolated rat tissues, although the decrease in permeability from jejunum to distal colon was greater than that seen in the present study. Evidence suggests that the profile of mannitol permeability along the intestine can vary considerably between species (van Meeteren et al., 1998; Nejdfors et al., 2000).

Is there a physiological rationale for the observed differences in PGP activity along the gut? PGP is considered to play an important role in reducing the absorption of potentially toxic xenobiotics such as those present in the gut lumen either as a result of microbial action in the gut or by ingestion, e.g., in food (Blackmore et al., 2001). With regard to the small intestine, the jejunum has a relatively short transit time (e.g., ≈40 min in the rat; Kayne et al., 1993), favoring absorption of lipophilic xenobiotics with high membrane permeability. Such compounds may partition into the membrane at such a fast rate that PGP-mediated efflux is balanced by passive influx and net absorption is unaffected (Litman et al., 2001) In contrast, the transit time in the ileum is considerably longer (≈140 min in the rat; Kayne et al., 1993), and xenobiotics reaching this region will tend to be of lower permeability, making them more susceptible to PGP-mediated efflux. Given that many natural xenobiotics are likely to fall into this second category, it could be argued that there is physiological merit in having a higher PGP activity in the ileum. The reason for colonic heterogeneity in PGP activity is not clear, although the ability of distal colon to dehydrate the luminal contents against a high luminal hydraulic resistance (Naftalin et al., 1999; Naftalin and Pedley, 1999) may be a factor. This would be expected to concentrate xenobiotics reaching the distal colon and may explain the observation of increased PGP activity in this region. Although more information is needed on the sites of absorption of natural xenobiotics, and the role of other transporters such as MRP and BCRP, it is interesting to speculate that the regional variations in PGP distribution are integral to the protective function of PGP in the murine gut barrier.

In conclusion, by comparing the permeability of “pure” PGP substrates in tissues from normal and mdr1a(−/−) mice, we have demonstrated significant regional variations in transporter activity within both small and large intestine with peaks of activity in distal small intestine and distal colon. Ablation of PGP reveals inherent differences in transcellular permeability along the gut, which are masked under normal circumstances by the action of efflux transporters. These data may provide a basis for a better understanding of the likely impact of PGP on the absorption of the drugs as they move along the intestine but will also be relevant in defining the physiological role that PGP plays in protecting the gut epithelium from luminal xenobiotics.

Footnotes

    • Received July 9, 2002.
    • Accepted August 27, 2002.

  • This work was supported by Pfizer Global Research and Development.

  • DOI: 10.1124/jpet.102.041236

Abbreviations

PGP
P-glycoprotein
mdr1
multidrug resistance protein 1
MRP
multidrug resistance-associated protein
BCRP
breast cancer resistance protein
ISC
short-circuit current
RT
transepithelial electrical resistance
A-B
apical-to-basolateral
B-A
basolateral-to-apical
Papp
apparent permeability
TBS-T
Tris-buffered saline/Tween 20
ANOVA
analysis of variance
WT
wild-type

References

  1. ↵
    1. Blackmore CG,
    2. Naughton PA, and
    3. van Heen HW
    (2001) Multidrug transporters in prokaryotic and eukaryotic cells: physiological functions and transport mechanisms. Mol Membr Biol 18:97–103.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Brasitus TA and
    2. Dudeja PK
    (1985) Regional differences in the lipid composition and fluidity of rat colonic brush-border membranes. Biochim Biophys Acta 819:10–17.
    OpenUrlPubMed
  3. ↵
    1. Brasitus TA and
    2. Schachter D
    (1984) Lipid composition and fluidity of rat enterocyte basolateral membranes. Regional differences. Biochim Biophys Acta 774:138–146.
    OpenUrlPubMed
  4. ↵
    1. Chianale J,
    2. Vollrath V,
    3. Wielandt AM,
    4. Miranda S,
    5. Gonzalez R,
    6. Fresno AM,
    7. Quintana C,
    8. Gonzalez S,
    9. Andrade L, and
    10. Guzman S
    (1995) Differences between nuclear run-off and mRNA levels for multidrug resistance gene expression in the cephalocaudal axis of the mouse intestine. Biochim Biophys Acta 1264:369–376.
    OpenUrlPubMed
  5. ↵
    1. Collett A,
    2. Higgs NB,
    3. Sims E,
    4. Rowland M, and
    5. Warhurst G
    (1999) Modulation of the permeability of H2 receptor antagonists cimetidine and ranitidine by P-glycoprotein in rat intestine and the human colonic cell line Caco-2. J Pharmacol Exp Ther 288:171–178.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Cordon-Cardo C,
    2. O'Brien JP,
    3. Boccia J,
    4. Casals D,
    5. Bertino JR, and
    6. Melamed MR
    (1990) Expression of the multidrug resistance gene product (P-glycoprotein) in human normal and tumor tissues. J Histochem Cytochem 38:1277–1287.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Emery PT,
    2. Higgs NB,
    3. Warhurst AC,
    4. Carlson GL, and
    5. Warhurst G
    (2002) Anti-secretory properties of non-peptide somatostatin receptor agonists in isolated rat colon: luminal activity and possible interaction with p-glycoprotein. Br J Pharmacol 135:1443–1448.
    OpenUrlCrossRefPubMed
  8. ↵
    1. Fagerholm U,
    2. Lindahl A, and
    3. Lennernas H
    (1997) Regional intestinal permeability in rats of compounds with different physicochemical properties and transport mechanisms. J Pharm Pharmacol 49:687–690.
    OpenUrlPubMed
  9. ↵
    1. Fojo AT,
    2. Ueda K,
    3. Slamon DJ,
    4. Poplack DG,
    5. Gottesman MM, and
    6. Pastan I
    (1987) Expression of a multidrug-resistance gene in human tumors and tissues. Proc Natl Acad Sci USA 84:265–269.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Fricker G,
    2. Drewe J,
    3. Huwyler J,
    4. Gutmann H, and
    5. Beglinger C
    (1996) Relevance of p-glycoprotein for the enteral absorption of cyclosporin A: in vitro-in vivo correlation. Br J Pharmacol 118:1841–1847.
    OpenUrlPubMed
  11. ↵
    1. Garrigues A,
    2. Escargueil AE, and
    3. Orlowski S
    (2002) The multidrug transporter, P-glycoprotein actively mediates cholesterol redistribution in the cell membrane. Proc Natl Acad Sci USA 99:10347–10352.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Gotoh Y,
    2. Suzuki H,
    3. Kinoshita S,
    4. Hirohashi T,
    5. Kato Y, and
    6. Sugiyama Y
    (2000) Involvement of an organic anion transporter (canalicular multispecific organic anion transporter/multidrug resistance-associated protein 2) in gastrointestinal secretion of glutathione conjugates in rats. J Pharmacol Exp Ther 292:433–439.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Hunter J and
    2. Hirst BH
    (1997) Intestinal secretion of drugs. The role of P-glycoprotein and related drug efflux systems in limiting oral drug absorption. Adv Drug Delivery Rev 25:129–157.
    OpenUrlCrossRef
  14. ↵
    1. Johnstone RW,
    2. Ruefli AA, and
    3. Smyth MJ
    (2000) Multiple physiological functions for multidrug transporter P-glycoprotein? Trends Biochem Sci 25:1–6.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Jonker JW,
    2. Smit JW,
    3. Brinkhuis RF,
    4. Maliepaard M,
    5. Beijnen JH,
    6. Schellens JHM, and
    7. Schinkel AH
    (2000) Role of breast cancer resistance protein in the bioavailability and fetal penetration of topotecan. J Natl Cancer Inst 92:1651–1656.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Kayne LH,
    2. D'Argenio DZ,
    3. Meyer JH,
    4. Hu MS,
    5. Jamgotchian N, and
    6. Lee DB
    (1993) Analysis of segmental phosphate absorption in intact rats. A compartmental analysis approach. J Clin Investig 91:915–922.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Kunta J,
    2. Yan J,
    3. Makhey VD, and
    4. Sinko PJ
    (2000) Active efflux kinetics of etoposide from rabbit small intestine and colon. Biopharm Drug Dispos 21:83–93.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Letrent SP,
    2. Polli JW,
    3. Humphreys JE,
    4. Pollack GM,
    5. Brouwer KR, and
    6. Brouwer KL
    (1999) P-glycoprotein-mediated transport of morphine in brain capillary endothelial cells. Biochem Pharmacol 58:951–957.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Li M,
    2. Hurren R,
    3. Zastawny RL,
    4. Ling V, and
    5. Buick RN
    (1999) Regulation and expression of multidrug resistance (MDR) transcripts in the intestinal epithelium. Br J Cancer 80:1123–1131.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Litman T,
    2. Druley TE,
    3. Stein WD, and
    4. Bates SE
    (2001) From MDR to MXR: new understanding of multidrug resistance systems, their properties and clinical significance. Cell Mol Life Sci 58:931–959.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Makhey VD,
    2. Gua A,
    3. Norris DA,
    4. Hu P,
    5. Yan J, and
    6. Sinko P
    (1998) Characterization and regional intestinal kinetics of drug efflux in rat and human intestine and Caco-2 cells. Pharm Res (NY) 15:1160–1167.
    OpenUrl
  22. ↵
    1. Maliepaard M,
    2. Scheffer GL,
    3. Faneyte IF,
    4. van Gastelen MA,
    5. Pijnenborg ACLM,
    6. Schinkel AH,
    7. van de Vijver MJ,
    8. Scheper RJ, and
    9. Schellens JHM
    (2001) Subcellular localization and distribution of the breast cancer resistance protein transporter in normal human tissues. Cancer Res 61:3458–3464.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Meddings JB
    (1989) Lipid permeability of the intestinal microvillus membrane may be modulated by membrane fluidity in the rat. Biochim Biophys Acta 984:158–166.
    OpenUrlPubMed
  24. ↵
    1. Mottino AD,
    2. Hoffman T,
    3. Jennes L, and
    4. Vore M
    (2000) Expression and localization of multidrug resistant protein mrp2 in rat small intestine. J Pharmacol Exp Ther 293:717–723.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Naftalin RJ and
    2. Pedley KC
    (1999) Regional crypt function in rat large intestine in relation to fluid absorption and growth of the pericryptal sheath. J Physiol 514:211–227.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Naftalin RJ,
    2. Zammit PS, and
    3. Pedley KC
    (1999) Regional differences in rat large intestinal crypt function in relation to dehydrating capacity in vivo. J Physiol 514:201–210.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Nejdfors P,
    2. Ekelund M,
    3. Jeppsson B, and
    4. Westrom BR
    (2000) Mucosal in vitro permeability in the intestinal tract of the pig, the rat and man: species- and region-related differences. Scand J Gastroenterol 35:501–507.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Penny JI and
    2. Campbell FC
    (1994) Active transport of benzo[a]pyrene in apical membrane vesicles from normal human epithelium. Biochim Biophys Acta 1226:232–236.
    OpenUrlPubMed
  29. ↵
    1. Stephens RH,
    2. O'Neill CA,
    3. Bennett J,
    4. Humphries M,
    5. Henry B,
    6. Rowland M, and
    7. Warhurst G
    (2002) Use of the isolated wild type and mdr1a(−/−) mouse intestine in the study of drug transport mechanisms in vitro. Br J Pharmacol 135:2038–2046.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Stephens RH,
    2. O'Neill CA,
    3. Warhurst A,
    4. Carlson GL,
    5. Rowland M, and
    6. Warhurst G
    (2001) Kinetic profiling of P-glycoprotein mediated drug efflux in rat and human intestinal epithelia. J Pharmacol Exp Ther 296:584–591.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Trezise AE,
    2. Romano PR,
    3. Gill DR,
    4. Hyde SC,
    5. Sepulveda FV,
    6. Buchwald M, and
    7. Higgins CF
    (1992) The multidrug resistance and cystic fibrosis genes have complementary patterns of epithelial expression. EMBO (Eur Mol Biol Organ) J 11:4291–4303.
    OpenUrlPubMed
  32. ↵
    1. Ungell AL,
    2. Nylander S,
    3. Bergstrand S,
    4. Sjoberg A, and
    5. Lennernas H
    (1998) Membrane transport of drugs in different regions of the intestinal tract of the rat. J Pharm Sci 87:360–366.
    OpenUrlCrossRefPubMed
  33. ↵
    1. van Meeteren ME,
    2. van Bergeijk JD,
    3. van Dijk AP,
    4. Tak CJ,
    5. Meijssen MA, and
    6. Zijlstra FJ
    (1998) Intestinal permeability and contractility in murine colitis. Mediat Inflamm 7:163–168.
    OpenUrl
  34. ↵
    1. Walle UK and
    2. Walle T
    (1999) Transport of the cooked-food mutagen 2-amino-1-methyl-6-phenylimidazo-[4,5-b]pyridine (PhIP) across the human intestinal Caco-2 cell monolayer: role of efflux pumps. Carcinogenesis 20:2153–2157.
    OpenUrlCrossRefPubMed
  35. ↵
    1. Warhurst G,
    2. Higgs NB,
    3. Fakhoury H,
    4. Warhurst AC,
    5. Garde J, and
    6. Coy DH
    (1996) Somatostatin receptor subtype 2 mediates somatostatin inhibition of ion secretion in rat distal colon. Gastroenterology 111:325–333.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Yang L,
    2. Chu JS, and
    3. Fix JA
    (2002) Colon-specific drug delivery: new approaches and in vitro/in vivo evaluation. Int J Pharm 235:1–15.
    OpenUrlCrossRefPubMed
  37. ↵
    1. Yumoto R,
    2. Murakami T,
    3. Nakamoto Y,
    4. Hasegawa R,
    5. Nagai J, and
    6. Takano M
    (1999) Transport of rhodamine 123, a P-glycoprotein substrate, across rat intestine and Caco-2 cell monolayers in the presence of cytochrome P-450 3A-related compounds. J Pharmacol Exp Ther 289:149–155.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Download PDF
Article Alerts
Email Article
Citation Tools

Research ArticleGASTROINTESTINAL, HEPATIC, PULMONARY, AND RENAL

Region-Dependent Modulation of Intestinal Permeability by Drug Efflux Transporters: In Vitro Studies in mdr1a(−/−) Mouse Intestine

R. H. Stephens, J. Tanianis-Hughes, N. B. Higgs, M. Humphrey and G. Warhurst
Journal of Pharmacology and Experimental Therapeutics December 1, 2002, 303 (3) 1095-1101; DOI: https://doi.org/10.1124/jpet.102.041236
Reddit logo Twitter logo Facebook logo Mendeley logo

Jump to section

Advertisement