Article Text

Download PDFPDF

Ontogeny of P-Glycoprotein in Mouse Intestine, Liver, and Kidney
  1. Burhan Mahmood,
  2. Monica J. Daood,
  3. Claudia Hart,
  4. Thor W.R. Hansen,
  5. Jon F. Watchko
  1. From the Department of Pediatrics (B.M., M.J.D., C.H., J.F.W.), Magee-Womens Research Institute, University of Pittsburgh School of Medicine, Pa, Norway
  2. Department of Pediatrics (T.W.H.), Rikshopitalet, University of Oslo, Norway.
  1. Address correspondence to: J.F. Watchko, MD, Department of Pediatrics, Magee-Womens Hospital, 300 Halket Street, Pittsburgh, PA 15213. E-mail jwatchko{at}mail.magee.edu
  2. This work was supported by grants HD-35847 and NS-38993 from the National Institutes of Health, the Irene McClenahan Young Investigator Research Fund of the Magee-Womens Health Research Foundation, the Mario Lemieux Centers for Patient Care and Research, and the 25 Club of Magee-Womens Hospital.

Abstract

Background P-glycoprotein (Pgp) is an ATP-dependent, integral plasma-membrane efflux pump that is constitutively expressed on (i) adult apical brush-border epithelial cells of the intestine, (ii) the bile canalicular face of hepatocytes, and (iii) the brush border epithelium of renal proximal tubules. This Pgp tissue distribution and localization affects the absorption, distribution, metabolism, and excretion of Pgp substrates. Little is known regarding the ontogeny of Pgp expression in these tissues.

Methods Postnatal expression of Pgp on brush border membranes of small intestine, liver, and kidney as a function of maturity from birth through adulthood was determined using Western immunoblotting and immunohistochemical techniques. Tissue was isolated from FVB mice at four different ages: day of life 0 (D0), day of life 7 (D7), day of life 21 (D21), and adult (Ad). The relative expression of Pgp protein on Western immunoblots was assessed by scanning densitometry and indexed as a percentage (mean±SEM) of the adult levels.

Results On Western immunoblots, Pgp expression was limited at birth (19±6% of Ad) and increased significantly with maturation in intestine (ANOVA, P <0.005). In contrast, hepatic (113±12% of Ad) and renal (96±15% of Ad) Pgp expression were at adult levels at birth. The tissue-specific developmental pattern of Pgp expression was confirmed by immunohistochemistry.

Conclusions We conclude that Pgp is expressed in a tissue-specific and developmentally regulated fashion and speculate that developmental modulation of intestine-Pgp expression may affect the oral bioavailability of Pgp substrates.

Keywords
  • Development
  • pharmacology
  • drug transporters

Statistics from Altmetric.com

Keywords

INTRODUCTION

P-glycoprotein (Pgp) is a member of the ATP-binding cassette superfamily of transport proteins that utilize ATP to translocate a wide range of substrates across biological membranes.1Pgp is encoded for by a family of genes referred to as the multidrug resistant or MDR genes,2,3whose products function as multidrug efflux pumps that limit the cellular influx and retention of numerous lipophilic compounds.3,4The intron-exon structures of the mouse mdr1a and human MDR1 genes are virtually identical,3,5,6and both encode the Pgp isoform that is expressed in gut, liver, and kidney.7,8More specifically, mdr1 Pgp is expressed in abundance on adult (i) apical brush-border intestinal epithelium, (ii) the bile canalicular face of hepatocytes, and (iii) the brush border epithelium of renal proximal tubules.7,8This Pgp tissue distribution and cellular localization has been shown to affect the absorption, distribution, metabolism, and elimination of Pgp substrates, including xenobiotics and clinically important drugs (eg, digoxin, verapamil, dexamethasone) in adults.9-12Moreover, intestinal Pgp has been shown to limit the absorption of xenobiotics via a countertransport (Phase III) mechanism and may act synergistically with intestinal cytochrome P450 isoenzymes to decrease the oral bioavailability of compounds. In the liver and kidney, Pgp acts to enhance the renal and biliary excretion of drugs. Altered expression of Pgp in these tissues may (i) predispose or protect against susceptibility to exogenous and endogenous toxins and (ii) impact the pharmacokinetics and pharmacodynamics of Pgp substrates. In this context, little is know regarding the ontogeny of Pgp in the intestine, liver, and kidney. Previous studies have demonstrated marked increases in mdr1 Pgp expression in the central nervous system during maturation,13,14which suggests an important regulatory role of ontogenetic factors in mdr1 Pgp expression. In the current study we tested the hypothesis in mice that intestinal, hepatic, and renal Pgp expression are limited in newborns and increase with postnatal maturation from birth through adulthood.

METHODS

Studies were conducted on wild type FVB mice (Taconic, Germantown, NY) at four different ages (n=6 at each age), including day of delivery (D0), day of life 7 (D7), day of life 21 (D21), and adulthood. Mice were sacrificed by lethal intraperitoneal injection of pentobarbital. Small intestine, liver, and kidney were immediately removed and stored at -80°C for future study. To obtain a sufficient amount of Pgp, tissues from D0 and D7 littermates were pooled and counted as one sample. Membrane vesicle preparations were used to provide high-yield protein fractions as described below. The study protocol was approved by the Magee-Womens Hospital and Research Institute Institutional Animal Care and Use Committee.

Preparation of Intestinal and Renal Brush-Border Membrane-Enriched Fractions

Enriched fractions of intestine and renal brush-border membrane vesicles were prepared for Western immunoblots using high-yield magnesium precipitation methods.15-17All procedures were performed at 4°C. Intestines were homogenized using a Polytron homogenizer (Brinkmann Instruments, Westbury, NY) in 15 to 20 volumes of a buffer consisting of 60 mM mannitol, 1 mM EGTA, 2.4 mM tris -HCl (pH 7.1), 2 mM phenylmethylsufonyl fluoride (PMSF), and a protease inhibitor cocktail for inhibition of serine, cysteine, aspartic and metalloproteases (Sigma, St Louis, Mo). Kidneys were homogenized as above in 10 volumes of the same buffer but at five times its concentration, except for the protease inhibitors (including PMSF), which remained the same. To each kidney homogenate, cold distilled water was added at 1.4 times the volume of the homogenate. To all homogenates 10 to 12 mM solid MgCl2 was added, and extracts incubated on ice for 20 minutes with occasional mixing. Intestinal extracts were centrifuged at 3000 g and kidney extracts at 1500 g for 15 minutes. The intestinal supernatants were centrifuged at 27,000 g and the kidney supernatant at 15,000 g for 30 minutes. The pellets were resuspended in 60 mM mannitol, 5 mM EGTA, 12 mM tris -HCl (pH 7.1), and protease inhibitors for the intestines and the kidney pellets resuspended in 150 mM mannitol, 2.5 mM EGTA, 6 mM tris -HCl (pH 7.4). All were homogenized using a Potter-Elvejhem homogenizer (Fisher Scientific, Pittsburgh, Pa) with a teflon pestle. Again, 10 to 12 mM solid MgCl2 was added to each homogenate and left on ice for 20 minutes, then centrifuged exactly as above for each tissue, both at the lower speed and the supernatants at the higher speed. The kidney pellets were resuspended in their original homogenating buffer and stored at -80°C, whereas the intestinal pellets were resuspended in 300 mM mannitol and 10 mM potassium phosphate (pH 7.4), homogenized again with a Potter-Elvejhem, and centrifuged at 47,000 g for 30 minutes. These pellets were finally resuspended in 60 mM mannitol, 5 mM EGTA, 12 mM tris -HCl (pH 7.1) and stored at -80°C.

Liver Membrane Preparation

Crude liver membrane preparations were made using 100 to 150 mg of tissue for each age group.18Livers were homogenized using a Dounce homogenizer in 7 mL phosphate-buffered saline (PBS) (pH 7.4) that contained 2 mM PMSF and a protease inhibitor cocktail. The homogenates were centrifuged at 4000 g for 10 minutes. The pellet was washed with an additional 5 mL of PBS and centrifuged again as above. The washed pellet was extracted in six volumes of a buffer containing 10 mM tris -HCl (pH 8.6), 1.5 mM MgCl2, 140 mM NaCl, 1% (v/v) NP-40, 2 mM PMSF, and protease inhibitor cocktail. The extraction was left on ice for 10 minutes with occasional vigorous vortexing before centrifugation at 31,000 g for 20 minutes. The supernatant containing the membranes was quick frozen in liquid nitrogen and stored at -80°C.

Western Immunoblots

Twenty micrograms of intestinal, hepatic, and renal membrane proteins (as determined by the Lowry method19) were loaded onto 7.5% acrylamide/ bis -acrylamide (29.2:0.8) gels. Electrophoresis was performed using a vertical slab-gel unit (Integrated Separation Systems, Natick, Mass) at 100 V for 90 minutes. The proteins were transferred electrophoretically to nitrocellulose at 100 V for 2 hours. Blots were blocked for 2 hours at room temperature with 5% non-fat powdered milk. Primary monoclonal antibodies to Pgp, including C219 (1:100) or C494 (1:100), were applied to the blots and incubated at 4°C overnight. The C219 antibody is directed to internal epitopes of Pgp near the ATP-binding cassette in the C- and N-terminal halves of the protein,20whereas C494 is directed to distinct internal epitope proximal to the ATP-binding site near the N-terminal half of Pgp.21The membrane was washed five times (5 minutes each) in TBS with 0.5% Tween 20. The nitrocellulose membrane was then incubated with anti-mouse Ig secondary antibody (Fab) (1:1250) for 90 minutes at room temperature. Membranes underwent another five 5-minute washes in TBS with 0.5% Tween 20. The Pgp band was developed using chemiluminescence with Renaissance Kit (New England Nuclear, Boston Mass). Chinese hamster ovary cells that overexpress Pgp were used as a positive control (generously supplied by Dr. Alan Senior of the University of Rochester Medical Center, Rochester, NY),22and intestinal, hepatic, and renal tissue from mdr1a null, mutant Pgp-deficient mice (Taconic, Germantown, NY) were used as negative controls.6Relative changes in Pgp protein expression across age groups were assessed using a high performance CCD camera with Harmony software (Magee-Womens Research Institute Core Imaging Facility). Pgp-band densitometry was standardized for any potential interlane protein-loading variation by dividing these values by the density of an internal control protein (β-actin). Relative Pgp expression was indexed as the percentage of that expressed in the adult, normalized to β-actin.

Immunohistochemistry

Five-micrometer frozen tissue sections of liver, kidney, and small intestine (n=4 for each tissue at each age group) were cut at -20°C, air dried, fixed in cold acetone, then rehydrated in PBS (pH 7.6). Staining was performed using the sandwich technique with aminoethylcarbazole as the chromogen, as described by Toth and colleagues for the detection of Pgp.23Endogenous peroxidase activity was quenched by incubating sections in 0.3% hydrogen peroxide. To prevent nonspecific protein binding, tissue sections were incubated in PBS (pH 7.4) with 0.03% casein and 0.05% Tween-20. The monoclonal Pgp-antibody C219 (1:50; Signet Labs, Dedham, Mass) diluted in PBS (pH 7.6) and 0.05% Tween-20 served as the primary antibody. The diluent alone served as the negative control. Tissue sections were incubated with primary antibody overnight at 4°C in a humidified chamber. A secondary antibody of peroxidase-conjugated rabbit anti-mouse IgG (Jackson ImmunoResearch, West Grove, Pa) was applied at a concentration of 1:75. Sections were then incubated with mouse monoclonal peroxidase-antiperoxidase complex (Sternberger Immunochemical, Lutherville, Md) at a 1:150 dilution. After rinsing, the first secondary antibody was again applied. After a final rinse, sections were stained with aminoethylcarbazole substrate (Vector Labs, Burlingame, Calif) and counterstained with hematoxylin. All negative controls showed absence of staining.

Statistical Analysis

Statistical methods (Minitab Statistical Software, PC Version 8.0, Minitab, State College, Pa)24included an analysis of variance (ANOVA) for comparisons of Pgp expression on Western immunoblots across age groups. In the event of a significant ANOVA, post-hoc analysis was performed using the Mann-Whitney U test. Statistical significance was expressed as P <0.05. Data are reported as a mean±SEM.

RESULTS

Representative Western immunoblots of adult wild type FVB and mdr1a null, mutant transgenic Pgp-deficient mouse intestine, liver, and kidney, as well as Chinese hamster ovary cells that overexpress Pgp, are shown in Figure 1. These immunoblots demonstrate that adult FVB intestine, liver, and kidney express Pgp, and that the monoclonal Pgp antibodies used do not cross-react with non-mdr1a Pgp proteins in these FVB line tissues.

Figure 1.

Western immunoblot for P-glycoprotein (Pgp) of wild type FVB mouse intestine, liver, kidney, and respective mdr1a null, mutant [-/-] Pgp-deficient mouse tissue. Chinese hamster ovary (CHO) cell samples that overexpress Pgp are shown as positive controls. Pgp was detected in FVB samples and CHO cells, but not in mdr1a null, mutant Pgp-deficient tissue.

Ontogeny of Small Intestinal P-Glycoprotein Expression

Representative Western immunoblots of intestine brush-border membrane Pgp at D0, D7, D21, and the adult age groups are shown in Figure 2. Limited but consistently detectable Pgp expression was noted at D0 and increased markedly with postnatal maturation. Relative Pgp expression on Western immunoblot of intestine as a function of developmental age group is shown in Figure 3a. Pgp protein was consistently evident at D0 and increased 5-fold with maturation (P <0.005). More specifically, there was significantly more Pgp expressed on D7, D21, and in the adult as compared with D0 (P <0.05). In addition, there was also significantly more Pgp expressed on D21 and in the adult as compared with D7 (P <0.05). Pgp protein expression at D21 and in the adult was not statistically different. Immunohistochemistry of small intestine consistently (n=4) confirmed this developmental pattern of Pgp expression with negative to weak staining at D0, weak staining at D7, moderate staining at D21, and moderate to strong staining in the adult sections (Figure 4). Moreover, at days 0 and 7, granular cytoplasmic staining predominated with brush-border membrane localization becoming more prominent and intense at day 21 and in adulthood.

Figure 2.

Representative Western immunoblots of FVB mouse intestine, liver, and kidney at day 0 (D0), day 7 (D7), and day 21 (D21) of postnatal life, and in adults (AD). β-actin immunoblot is shown as interlane standard. Chinese hamster ovary (CHO) cell samples are shown as positive controls.

Figure 3.

Relative levels of P-glycoprotein in FVB intestine (Panel A, solid bars), liver (Panel B, hatched bars), and kidney (Panel C, open bars) as a function of postnatal age, expressed as percentages of those observed in the adult. *, P * <0.05 vs. D21 and Adult; #, P * <0.05 vs. D7.

Figure 4.

Immunohistochemistry of small intestine. Panels show day 0, 7, and 21, and adult tissue; all photos were taken at x1000 original magnification. P-glyco-protein immunoreactivity was negative to weak at day 0, weak at day 7, moderate at day 21, and moderate to strong in the adult. At D0 and D7, granular cytoplasmic staining predominated, with brush-border membrane localization becoming more prominent and intense at D21 and in the adult.

Ontogeny of Hepatic P-Glycoprotein Expression

Representative Western immunoblots of hepatic brush-border membrane Pgp at D0, D7, D21, and the adult age groups are shown in Figure 2. Abundant Pgp expression was noted at D0 and was comparable to that seen during postnatal maturation and in the adult. Relative Pgp expression on Western immunoblot of liver as a function of developmental age group is shown in Figure 3b. Pgp protein expression at D0, D7, D21, and in the adult were not statistically different. Immunohistochemistry confirmed this pattern of Pgp expression with moderate to strong Pgp immunoreactivity across all study age groups (Figure 5). Pgp was prominently immunolocalized to the bile canalicular surface of the hepatocytes with occasional granular cytoplasmic staining in all age groups.

Figure 5.

Immunohistochemistry of liver. Panels show day 0, 7, and 21, and adult tissue; all photos were taken at x1000 original magnification. P-glycoprotein immunoreactivity was moderate to strong across all study age groups. P-glycoprotein was prominently localized to the bile canalicular surface of the hepatocytes, and occasional granular cytoplasmic staining occurred in all age groups.

Ontogeny of Renal P-Glycoprotein Expression

Representative Western immunoblots of renal brush-border membrane Pgp at D0, D7, D21, and the adult age groups are shown in Figure 2. Abundant Pgp expression was noted at D0 and was comparable to that seen during postnatal maturation and in the adult. Relative Pgp expression on Western immunoblot of the kidney as a function of developmental age group is shown in Figure 3c. Pgp protein expression at D0, D7, D21, and in the adult were not statistically different. Immunohistochemistry confirmed this pattern of Pgp expression with moderate to strong Pgp immunoreactivity across all study age groups (Figure 6). Pgp immunostaining was principally localized to the proximal tubule brush-border epithelium at D21 and in the adult, whereas D0 and D7 kidney manifested more prominent granular cytoplasmic staining.

Figure 6.

Immunohistochemistry of kidney. Panels show day 0, 7, and 21, and adult tissue; all photos were taken at x1000 original magnification. P-glyco-protein immunoreactivity was moderate to strong across all study age groups. P-glycoprotein immunostaining was principally localized to the proximal tubule brush-border epithelium at D21 and in the adult, whereas D0 and D7 kidney manifest more prominent granular cytoplasmic staining.

DISCUSSION

The current study assessed the postnatal expression of Pgp in mouse intestine, liver, and kidney using Western immunoblotting and immunohistochemical techniques. These tissues were selected for study because their Pgp expression affects the absorption, distribution, metabolism, and excretion of drugs. Our results indicate that Pgp is expressed in a tissue-specific and developmentally regulated fashion in mice. More specifically, a marked postnatal increase in intestinal Pgp expression was observed, whereas hepatic and renal Pgp were expressed at adult levels from birth onwards. We also observed that the electrophoretic mobility of intestinal Pgp on polyacrylamide gels varied as a function of age (Figure 2). The electrophoretic mobility of Pgp on polyacrylamide gels is dependent on the conditions used, the degree of Pgp glycosylation, the species and tissue studied, and whether the Pgp is tissue- or cell-derived.25,26The latter, in all likelihood, accounts for the difference in electrophoretic mobility between Chinese hamster ovary cells and tissue Pgp, whereas we speculate that the developmental changes in intestine-Pgp electrophoretic mobility are secondary to differences in Pgp glycosylation.

Previous studies have reported marked changes in tissue mdr1 gene expression during embryogenesis27and postnatal maturation,13,14,28which demonstrates the importance of ontogenetic regulation of tissue-Pgp expression. In the current study, we observed a significant five-fold increase in intestinal Pgp expression on Western immunoblots from birth through adulthood. Immunohistochemistry of small intestine confirmed this developmental pattern of Pgp expression. These novel observations are consistent with (i) the absence of MDR gene expression in human fetal intestine until late in gestation, if at all,27as well as with (ii) evidence that other membrane transport systems are under developmental control.28-30Regarding the latter, the organic cation-transport system is not fully mature at birth in several species, and the species' capabilities to secrete organic cations gradually increase to adult capacity during postnatal development.28-30Because many Pgp substrates are organic cations, it is not surprising that Pgp expression is also subject to postnatal modulation.

In contrast to later expression in the intestine, the levels of hepatic and renal Pgp expression at birth were no different from those of their adult counterparts. Consistent with these findings are the observations of Kalken and colleagues, who reported that kidney and liver demonstrate earlier and more consistent Pgp expression than other tissues during human gestation (7-38 weeks of gestation).27Our observations in FVB mice, however, contrast with a previous study in CF-1 mice that reported undetectable hepatic and renal Pgp expression at birth and significant postnatal increases in renal and hepatic Pgp expression.31The marked difference in hepatic and renal Pgp expression patterns between these murine lines may relate to the strain of mouse used. In this regard, a subpopulation of CF-1 mice is Pgp deficient: a naturally occurring equivalent of the mdr1a null, mutant transgenic Pgp knock-out.31Indeed, investigators have determined that among randomly selected CF-1 mice there is a 25% incidence of P-glycoprotein deficiency in mice.31This phenomenon was reported after the CF-1 Pgp ontogeny investigation and may have confounded the results of that study. Alternatively, species and strain differences in Pgp expression during maturation may exist. This issue merits further investigation as part of the attempt to unravel the mechanisms underlying the ontogenic regulation of tissue-Pgp expression.

Strictly speaking, the factors that modulate the developmental pattern of Pgp expression in the intestine are unknown, but they may include both endogenous and exogenous regulators. The expression of mdr1 is modulated by corticosteroids32and other hormones whose activity may change in the perinatal and postnatal periods.32Ligand activators of mdr1 gene expression include numerous chemosensitizers and naturally occurring cytotoxic compounds33that the newborn may be exposed to during postnatal development. Recent studies suggest that other environmental factors, particularly diet, may also affect Pgp expression.34,35It is of interest in this regard that intestinal Pgp expression increases in pups feeding on breast milk, and it reaches adult levels at the time of weaning (ie, D21). Breast milk contains various biologically active peptide growth factors (eg, insulin like growth factor I, epidermal growth factor) and hormones that are known to stimulate gastrointestinal development36,37as well as enhance Pgp expression and function in tumor cell lines in vitro.38,39Whether these or other factors in breast milk modulate intestinal Pgp expression is unknown, but the temporal relationship between the attainment of adult levels of intestinal Pgp and weaning suggests that this issue warrants further study.

The functional impact of developmental alterations in tissue-Pgp expression remains unclear. Adult levels of hepatic and renal Pgp at birth suggest that the newborn mouse will manifest Pgp-mediated hepatobiliary and renal clearance of Pgp substrates at adult capacity. A potential caveat to this assertion is that Pgp immunolocalization was often of a granular cytoplasmic nature at D0 and/or D7 in kidney and liver. This pattern may indicate that Pgp has yet to localize at its epithelial site of action, which suggests that the transport function of Pgp may not be mature at these ages. There are, however, no functional data to assess the implications of this Pgp immunolocalization pattern. In contrast, the clearly limited intestine-Pgp expression in the neonatal mouse may enhance the oral bioavailability of Pgp substrates. Indeed, tissue concentrations (including intestine) and plasma levels of orally administered Pgp substrates in mdr1a[-/-] null, mutant Pgp-deficient mice are increased several fold, consistent with the animals' increased net uptake from the gastrointestinal tract, decreased elimination, or both.6,40The functional impact of Pgp deficiency on oral drug bioavailability is an area of current investigation in several laboratories.41,42

In summary, the current study is the first to characterize the postnatal expression of mdr1 Pgp across intestine, liver, and kidney, tissues important in the absorption, distribution, metabolism, and excretion of xenobiotics and several clinically important drugs. We conclude that Pgp is expressed in a tissue-specific and developmentally regulated fashion and speculate that developmental modulation of intestine-Pgp expression may affect the oral bioavailability of Pgp substrates.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.

Supplementary material

Related Data

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.