P-glycoprotein–mediated transport of berberine across Caco-2 cell monolayers

Download P-glycoprotein–mediated transport of berberine across Caco-2 cell monolayers

Post on 15-Jun-2016




0 download


<ul><li><p>P-GlycoproteinMediated Transport of Berberine acrossCaco-2 Cell Monolayers</p><p>HAN-JOO MAENG, HO-JUNG YOO, IN-WHA KIM, IM-SOOK SONG, SUK-JAE CHUNG, CHANG-KOO SHIM</p><p>Department of Pharmaceutics, College of Pharmacy, Seoul National University, Seoul 151-742, Korea</p><p>Received 7 March 2002; revised 30 May 2002; accepted 6 July 2002</p><p>ABSTRACT: The objective of this study was to investigate the mechanisms by whichberberine is transported in the secretory and absorptive directions across Caco-2 cellmonolayers. The basolateral-to-apical (B-A) flux was 30-fold greater than the apical-to-basolateral flux and temperature dependent (i.e., drastic decrease at 48C compared with378C). The above results suggest the involvement of a carrier-mediated active transportmechanism for the B-A transport of berberine. However, no significant concentrationdependency for the permeability (Papp) of berberinewas observed forB-A transport over aconcentration range of 5300mM, indicating that theKmvalue of berberine for the carriersystem is greater than 300 mM. Well-documented P-glycoprotein (P-gp) substrates suchas verapamil, daunomycin, and rhodamine123 inhibited the B-A flux of berberine,whereas tetraethylammonium and taurocholate did not, suggesting that P-gp is involv-ed in the transport. For the case of daunomycin, the B-A flux, but not the apical-to-basolateral flux, was significantly increased after pretreatment of the cell monolayerswith berberine. In addition, the uptake of 1 mM daunomycin into Caco-2 cells was de-creased as a result of this pretreatment. These results suggest that the repeatedadministration of berberinemay up-regulate P-gp functions in Caco-2 cells. If this occursin the gastrointestinal epithelial cells, the repeated administration of berberine mayreduce the gastrointestinal absorption of P-gp substrates including chemotherapeuticagents such as daunomycin. 2002 Wiley-Liss, Inc. and the American PharmaceuticalAssociation J Pharm Sci 91:26142621, 2002</p><p>Keywords: berberine; P-gp; Caco-2 cell monolayers; transport</p><p>INTRODUCTION</p><p>Berberine (Fig. 1) is an isoquinoline1 that has avaried pharmacology including anti-microbial,2</p><p>anti-motility,3 and anti-secretory activities.46 Ithas been used for more than 2000 years in tradi-tional Eastern medicine as an effective medi-cation for the treatment of gastro-enteritis andsecretory diarrhea.1 It is also effective in theprevention and treatment of diarrhea in animalmodels.79</p><p>It has recently been reported that berberineis readily extruded from microbial cells by the</p><p>multidrug resistance pump of cell membranes,which has a role in protecting the cells from anti-microbials.10 In vitro interactions of berberinewith P-glycoprotein (P-gp) have also been report-ed in hepatoma cells.11 These reports stronglysuggest that berberine is a substrate for P-gp,a 170 kDa membrane glycoprotein encoded bythe multidrug resistance gene (e.g., MDR1 inhumans).</p><p>P-gp confersmultidrug resistance by enhancingthe active efflux of chemotherapeutic agents fromcells,12 and is involved in the transport of a varietyof structurally and pharmacologically unrelatedhydrophobic drugs, including vinca alkaloids,anthracyclines, cyclosporin A, digoxin, and gluco-corticoids.1316 It is expressed in normal tissuessuch as the brush-border membrane of renalproximal tubules, the bile canalicular membrane</p><p>2614 JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 12, DECEMBER 2002</p><p>Correspondence to: Chang-Koo Shim (Telephone: 82-2-880-7873; Fax: 82-2-885-8429.; E-mail: shimck@plaza.snu.ac.kr)</p><p>Journal of Pharmaceutical Sciences, Vol. 91, 26142621 (2002) 2002 Wiley-Liss, Inc. and the American Pharmaceutical Association</p></li><li><p>of hepatocytes, intestinal epithelial cells, capillaryendothelial cells of brain and testis, adrenal glandand placental trophoblasts, aswell as inmultidrug-resistant tumor cells.17,18 Thus, it contributes tothe excretion of xenobiotics and endogenous com-pounds into the urine, bile, and intestinal lumen,as well as the formation of biological barriersagainst such compounds in various tissues ororgans including brain, placenta, and tumor cells.For example, intestinal P-gp is thought to limit theabsorption of drugs after oral administration.1922</p><p>If berberine is a substrate of P-gp in the body, itwould be expected to influence the pharmacoki-netics of various substances that are transportedvia the P-gp system. The objective of this study,therefore, was to investigate the mechanisms bywhich berberine is transported in secretory andabsorptive directions across a biological mem-brane. Of particular interest was its possible in-teractionwith the P-gp system on cell membranes.Caco-2 cells were used as a representativemodel of biological membranes, on which P-gp isexpressed.20,23,24</p><p>MATERIALS AND METHODS</p><p>Materials</p><p>Berberine chloride (Sigma Chemical Co., St.Louis, MO), [14C]mannitol (50 mCi), [3H]taurocho-late (250 mCi), [3H]daunomycin (250 mCi) (all fromNew England Nuclear, Boston, MA), fetal bovineserum (Hyclone Laboratories, Logan, UT), tryp-sin-EDTA (Gibco Laboratories, Gaithersburg,MD), and Dulbeccos modified Eagles medium,nonessential amino acid solution, penicillin-streptomycin, Hanks balanced salt solution</p><p>(HBSS), HEPES (all from Sigma Chemical Co.)were used as purchased. All other reagents wereof analytical grade.</p><p>Cell Culture</p><p>The human colon adenocarcinoma cell line, Caco-2 (American Type Culture Collection, Rockville,MD), was grown in the form of monolayers inDulbeccos modified Eagles medium, 10% fetalbovine serum, 1% nonessential amino acid solu-tion, 100 U/mL penicillin, and 0.1 mg/mL strepto-mycin at 378C in an atmosphere of 5% CO2 and90% relative humidity. Stock cultures were grownin 75-cm2 tissue culture flasks and were split 1:3at 80 to 90% confluency using 0.02% EDTA and0.05% trypsin. The Caco-2 cells from passagenumbers 46 to 55 were seeded on the permeablepolycarbonate inserts (1-cm2, 0.4-mm pore size;Corning Costar Co., Cambridge, MA) in 12 Trans-well plates (Corning Costar Co.) at a density of2.5 105 to 3.0 105 cells/cm2. The inserts werefed with the incubation medium at 2-day intervalsfor the first week and then at daily intervals,usually for 2 weeks, until they were used for thetransport experiments.</p><p>In experiments to investigate the effect of ber-berine pretreatment on the transport of dauno-mycin, the cell monolayer grown on the insert wasfurther fed with the incubation medium contain-ing berberine (30 mM) for specified time periods(i.e., 1, 2, 3, 5, 7, or 10 days).</p><p>The integrity of the cell monolayers wasevaluated by means of transepithelial electricalresistance (TEER) measurements using anEVOMTM epithelial volt/ohm-meter (World Preci-sion Instruments, Sarasota, FL). When the TEERvalue reached 300700 Ocm2, the cell insertswere used for the transport experiments. Thetransport of [14C]mannitol (5.4 mM)was</p></li><li><p>aspiration.25 In each transport experiment, threeinserts were used. To measure the transport ofberberine from the apical-to-basolateral (A-B)side, 0.5 mL of transport medium (pH 7.4) con-taining 1% (v/v) dimethylsulfoxide and the drugwas added to the apical side, and 1.5 mL of thetransport medium without the drug to the basalside. The inserts were moved to wells containingfresh transport medium at 30-min intervals for2 h. A 300-mL aliquot of the medium, taken at eachtime point, was assayed for berberine by high-performance liquid chromatography (HPLC).</p><p>For measurement of the basolateral-to-apical(B-A) transport of berberine, 1.5 mL of the trans-port medium containing the drug was added to thebasolateral side, and 0.5 mL of the transportmedium without the drug to the apical side. Theinserts were then incubated at 378C, and a 300-mLaliquotwas removed from the apical side at 30-minintervals for 2 h and replaced with 300 mL of thefresh transport medium. A 300-mL aliquot of eachsample was assayed for berberine by HPLC.</p><p>In experiments to investigate the effect of vari-ous compounds on the B-A transport of 20 mM ber-berine, the inhibitors were added to the transportmedium on the basal side of the cell monolayers.</p><p>Berberine levels were determined by the HPLCprocedure described below, whereas [14C]manni-tol, [3H]taurocholate, and [3H]daunomycin weredetermined by liquid scintillation counting using aWallac model 1409 instrument (Wallac, Gaithers-burg, MD).</p><p>HPLC Assay of Berberine</p><p>Themethod of Chen and Chang26 was used for thedetermination of berberine, with minor modifica-tions. To summarize, the HPLC system consistedof a Hitachi L-7110 pump (Hitachi, Japan), aShimadzu SPD-6AV UV-Vis detector (Shimadzu,Japan), a Hitachi L-7200 autosampler (Hitachi),and a Shimadzu C-R6A integrator (Shimadzu).C18 column was used. The mobile phase wasprepared by mixing 60% acetonitrile in a 0.1%phosphoric acid solution and adjusting the pH to6.0 using concentrated ammonia solution. Theeluent resulted in sharp, well-resolved peaks. Thechromatogram was monitored by UV detection ata maximum wavelength of 267 nm. Calibrationcurves for berberine were linear over the con-centration range of 25 1000 ng/mL with acorrelation coefficient of 0.999, but the limit ofdetection which exhibits </p></li><li><p>berberine over the range of 5 300 mM (Fig. 3 forthe B-A transport, data not shown for the A-Btransport), exhibiting no significant concentra-tion dependency in the permeability (Papp) of thetransport in either direction. This indicates thatthe Km of berberine for the responsible carriersystem in Caco-2 cells is &gt;300 mM. Because ofprobable cytotoxicity problems,26,28,29 the trans-port rates for concentrations &gt;300 mM were notpursued. The transport was found to be tempera-ture dependent (i.e., drastic decrease in the flux at48C compared with 378C) for the B-A flux of</p><p>berberine (62.4 1.6 versus 5.7 0.4 pmole/min,meanSD, n 3). Collectively, the above resultssuggest the involvement of a carrier-mediatedactive transport mechanism in the B-A transportof berberine in Caco-2 cells.</p><p>Effect of Transport Inhibitorson Berberine Transport</p><p>It is well known that P-gp is expressed in theapical membrane of mucosal cells in the intestineand often pumps (i.e., secretion) a variety of xeno-biotics into the lumen, leading to a limited netintestinal absorption of such compounds. Thus,the relationship between the active transport ofberberine in the efflux (secretary) direction acrossthe Caco-2 cell monolayers and P-gp was thenexamined. For this purpose, the flux of berberine(20 mM) was examined in the absence and pre-sence of representative P-gp inhibitors such asverapamil (1 mM), daunomycin (1 mM), andrhodamine 123 (1 mM) on the basal side. Thepresence of the above P-gp inhibitors had nosignificant influence on the TEER values ofCaco-2 cell monolayers, but substantially inhib-ited the B-A flux of berberine (Fig. 4), suggestingthat berberine is transported across Caco-2 cellmonolayers via assistance from P-gp, and that acarrier-mediated active mechanism is operative.However, the presence of tetraethylammonium(1 mM) or taurocholate (1 mM) on the basal side</p><p>Figure 2. Time course for the apical-to-basolateral(*) and basolateral-to-apical (*) transport of 100 mMberberine (meanSD, n 3) across Caco-2 cell mono-layers in HBSS at 378C.</p><p>Figure 3. The basolateral-to-apical (B-A) flux ofberberine (meanSD, n 3) across Caco-2 cell mono-layers as a function of berberine concentration (5, 10, 20,50, 100, and 300 mM). Each point represents themeanSD for three experiments. The inset indicatesan Eadie-Hofstee plot of the transport rates (V) andberberine concentration (C).</p><p>Figure 4. Effect of P-gp inhibitors (verapamil, rhoda-mine123, and daunomycin, 1 mM each), an organicanion (taurocholate, 1 mM), and an organic cation(tetraethylammonium, TEA, 1 mM) in the basal sideon thebasolateral-to-apical (B-A)fluxof 20mMberberine(meanSD, n 3) across Caco-2 cell monolayers. BDL,below detection limit.</p><p>BERBERINE TRANSPORT ACROSS CACO-2 MONOLAYERS 2617</p><p>JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 91, NO. 12, DECEMBER 2002</p></li><li><p>had no effect on the B-A transport of berberine(20 mM) (Fig. 4), suggesting that the transportsystem for berberine is independent of the othertransport systems that are involved in the trans-port of, for example, organic cations and bileacids.</p><p>If P-gp is responsible for the efflux of berberine,then the presence of P-gp inhibitors might in-crease the absorptive flux of berberine. However,the presence of 1 mM verapamil in the apical sidehad no effect on the A-B transport of 100 mMberberine, suggesting that the inhibitory effect ofverapamil on the A-B transport of berberine isoperative only in the presence of verapamil in thebasal side.</p><p>Effect of Berberine Pretreatment onDaunomycin Transport</p><p>A variety of factors including heat shock, differ-entiation inducers, carcinogens, and anticancerdrugs often induce the expression of P-gp in tumorcells through transcriptional regulation.30 A simi-lar induction of P-gp has been reported for variousP-gp substrates. Thus, the effect of berberinepretreatment on P-gp function in Caco-2 cellmonolayers was examined. The effect of the pre-treatment period on daunomycin transport wasinitially examined for 30 mM berberine. For thispurpose, Caco-2 cell monolayers were pretreatedwith 30 mM berberine for specified time periods(i.e., 1, 2, 3, 5, 7, or 10 days) as described inMethods. The monolayers were then washedthree times with 2 mL of the fresh transportmedium to remove residual berberine in the cells,and the transport of 1 mM [3H]daunomycin inboth directions was measured. As indicated byFigure 5, no change was observed in the A-B fluxfor daunomycin regardless of the pretreatmentperiod used, whereas a significant increase in B-A flux was observed within 3 days of pretreat-ments which approached a plateau thereafter for10 days. Thus, the pretreatment period was set at7 days in subsequent experiments.</p><p>The effect of berberine concentration (1, 10, and30 mM) on the directional transport of daunomycinwas then examined. As shown by Figure 6, the B-Aflux of 1 mM daunomycin was increased by the7-day pretreatment as the concentration of ber-berine in the pretreatment medium increased.This increase was clearly evident for the pretreat-ment, evenwith 1 mMberberine.However, such aneffect was not observed for the A-B flux of 1 mMdaunomycin (Fig. 6). To determine whether the</p><p>increase in the B-A flux of daunomycin as theresult of berberine pretreatment is associatedwithcell integrity, the effects of a 7-day pretreatmentwith berberine (1, 10, 30 mM) on mannitol leakage(A-B direction) and TEER values were measured(Fig. 7). No significant changes in these character-istics were detected (data not shown), consistentwith the previous study that reported the non-toxicity of berberine pretreatment in termsof HepG-2 cell viability.11 This indicates that theincrease in the transport of daunomycin is notassociated with the reduced integrity of the cellmonolayer.</p><p>Figure 5. Effect of the period of pretreatment withberberine (30 mM) on the vectorial transport of 1 mMdaunomycin (meanSD,n 3) acrossCaco-2 cellmono-l...</p></li></ul>


View more >