Use of fluorescent probes to monitor the efflux transporters P- Glycoprotein and MRP1 in BeWo cells.**

Rhodamine, Fluorescein and the Acetoxy Methyl (AM) derivative (calcein AM) of the fluorescent indicator calcein were used to evaluate the functional role of the efflux transport system in BeWo cells. Uptake studies with P-glycoprotein or MRP1 inhibitors show significant enhancement of accumulation of calcein. Since both calcein and calcein AM are extruded by MRP1, whereas only Calcein AM is a substrate of P-glycoprotein, the effect of modulators on the transport of calcein AM is indicative of the functional activity of both P-glycoprotein and MRP1. However, studies with rhodamine (substrate of P-glycoprotein) did not reveal the expected functional activity in the presence of the same modulators. Uptake of Fluorescein (substrate of MRP1) increased in the presence of inhibitors of MRP1, but was unaffected by P-glycoprotein inhibitors. It was found that cyclosporinA, verapamil, indomethacin, probenecid, sodium orthovanadate and vinblastine affected the uptake, efflux and transport of calcein. All of the modulators except sodium orthovanadate and sodium azide increased the uptake of calcein. The transport was polarized with greater permeability from the apical to the basolateral direction. The inhibition of MRP 1 and Pgp resulted in a decrease of apical to basolateral transport. However, none were able to decrease transport from basolateral to the apical side significantly. The efflux of calcein was decreased by all the compounds tested, indicating an inhibition of MRP 1. Sodium orthovanadate also decreased the efflux but did not affect transport. Therefore, we believe that calcein AM can be used as an indicator of Pgp and MRP1 functional activity in BeWo cells. Fluorescein uptake was only affected by MRP1 inhibitors. P-glycoprotein inhibitors Verapamil and C219 had no effect on the uptake of fluorescein, suggesting that it is an effective indicator of MRP1 functionality. However the transport of fluorescein did not show polarization. This may be due to fluorescein showing a permeability ( ~ lxl0~cms"') ten times greater than that of calcein. Finally, rhodamine cannot be used as a functional indicatior of P-glycoprotein activity in BeWo cells due to the presence of other transporters which interfere with rhodamine uptake and transport.


INTRODUCTION
The BeWo cell line derived from a human choriocarcinoma 1 produces confluent monolayers displaying properties of normal trophoblasts.However, primary cultures of human trophoblasts do not form monolayers 2 .Therefore, the use of the BeWo cell line serves as a model for the study of transport of compounds across the maternal-fetal barrier.Although tHe placenta transports many substances by passive diffusion, the active efflux transporters P-glycoprotein (Pgp) 3 , Multidrug Resistance Associated Protein (MRP1) 4,5 and Breast Cancer Resistance Protein (BCRP) 6 have now been found to be present in Bewo cells and the placenta.Efflux pump activity affects efficient drug delivery and acts as a mechanism of drug resistance 7 .
The placental transporters are of vital importance during pregnancy.These transporters ensure that the nutrients are provided to the fetus, and metabolites are efficiently effluxed from the uterus.Drug transporters similarly protect the fetus from adverse effects by effluxing drugs present in the maternal circulation.Thus, the function of drug transporters and substances capable of modulating these transporters need to be carefully studied.cellular level is very limited.The current study examines the functional activity of both Pgp and MRP1 by comparing the effect of Pgp and MRP 1 inhibitors on the uptake, efflux and transport of acetoxy methyl derivative of calcein (calcein AM) (chemical structures in page 336) in BeWo cells.We have used calcein AM, a non fluorescent hydrophobic compound which is hydroiyzed to the fluorescent anion calcein by cytoplasmic esterases.Calcein AM is the nonfluorescent acetoxy methyl ester (AM) of the fluorescent compound, calcein.Calcein is a substrate of MRP 1, but not of Pgp.However, the ester calcein AM is a substrate of both Pgp and MRP1 8 .Pgp extrudes calcein AM from the plasma membrane, reducing accumulation of calcein in the cytosol.The amount of Pgp activity is inversely proportional to the accumulation of intracellular calcein fluorescence.Rhodamine is generally a good marker for functional Pgp activity 9 .However, we were unable to obtain significant increases in uptake or see a polarization in transport of rhodamine.Rhodamine is a cationic dye, unlike the anionic compounds, fluorescein and calcein (chemical structures in page 336).On the other hand, uptake of fluorescein yielded the expected results with MRP1 inhibitors and is an effective tool for assessment of MRP 1 functionality.

I4
C sucrose was from Perkin Elmer, Boston, MA,USA.
Cell culture: The BeWo cell line was originally derived from a human choriocarcinoma 1 .The BeWo clone (b30) was obtained from Dr. Alan Schwartz (Washington University, St. Louis, MO, USA).The cells were continuously cultured in Dulbecco's Modified Eagle's Medium (DMEM) with 10% heat inactivated FBS containing 0.37% sodium bicarbonate, 2 mM L-glutamine, 0.1 mM non-essential amino acids and 1% antibiotics (10,000 U/mL penicillin and 1 Omg/mL streptomycin).The cells were maintained in 175-cm 2 Falcon flasks at pH 7.4 under 5% C0 2 and 95% humidity at 37 °C.After 4-5 of culture, the cells were ready for passage.The cells were harvested by exposure (5 min) to a trypsin-EDTA solution (0.25%o trypsin and 0.2% Ethylene diamine tetra-acefic acid (EDTA) in Hank's Balance Salt solution or HBSS).Cells were passed onto 12 mm Transwell polyester membranes March 2007   previously coated with type IV human placental collagen (0.75 fAg/mL ?70 uL/insert:Sigma) for monolayer cultures.The seeding density was 50,000 cells per well.Uptake and efflux experiments were performed in 12 well culture plates precoated with Poly-D-Lysine (5 u.g/mL solution, 100 uL per well), sterilized, and coated with flbronectin (50 ug/mL, 4 drops per well) Cells were seeded at a density of 50,000 cells per well.The culture medium was changed every other day.Cells formed monolayers or were confluent in 4-5.For consistency, experiments were performed on the 5 th day post seeding.
Uptake studies: Accumulation of rhodamine 123 (0.5, 1 and 5 jiM), Calcein (luM CAM) and fluorescein (1 and 100 mM) was investigated in BeWo cells.On the 5 th day post seeding of culture plates, the growth medium was aspirated off and the cells were rinsed three times with pre-warmed (37 °C) HBSS.The monolayers were then equilibrated for 30 min in 1 mL HBSS at 37 °C.HBSS was aspirated and the fluorescent indicator was added without inhibitor (for control), while appropriate inhibitor was added to the indicator solution for the experimental wells (lmL/well).The wells were incubated for lh with gentle shaking.Thereafter, the wells were aspirated and washed 3 times with ice cold HBSS.The cells were lysed with lysing solution (2% v/v Triton X-100 for calcein accumulation, and 0.5% v/v Triton X-100 in 0.2N NaOH for rhodamine and fluorescein accumulation).The lysates were then assayed using a fluorescent spectrophotometer (SLM instruments, Inc.).Fluorescence was measured at excitation 485nm and emission 530 nm and quantified against a standard curve of RM23, calcein or sodium fluorescein.The protein content of each monolayer was determined using the BCA protein assay reagent kit.The results are expressed as total fluorescent probe accumulation (nmol or pmol) per mg of protein.
Efflux experiments: Cell seeding and growth conditions were identical to those utilized in accumulation experiments.Three methods were utilized for efflux assay.The first involved accumulation of fluorescent probe in the presence of the inhibitor used for efflux.Accumulation of probe for the other two methods was in the absence of the modulator.Accumulation was performed as mentioned in uptake studies.After one hour of uptake of probe, the monolayers were washed with ice cold HBSS as before, and then pre-warmed (37 °C) efflux solutions (1 mL per well) were added.In the first two methods, the efflux solutions were HBSS with or without (control) the appropriate modulator.Efflux was monitored for a time of 2 h.In the third method, efflux for the first hour was in the absence of any modulator.Modulator was added at the 59 th minute and then efflux monitored for a further two hours.At suitable time intervals, 100 ^L aliquots were removed for fluorescent assay and replenished with 100 uL of the respective efflux solution.The fluorescent probe effluxed was quantitated against a standard curve.At the end of the two hour efflux period, the monolayers were washed with ice cold HBSS and lysed in lysing solution, and the cell associated fluorescence and protein content were measured.Results are expressed as pmol fluorescent probe effluxed per mg protein, and pmol cell associated fluorescent probe per mg protein.
Transport studies: BeWo cells were seeded onto human placental collagen pre-coated polyester transwells.The cells were allowed to attach for 24 h, then the medium in the apical and basolateral compartments were removed and new media added every two days.Transport was performed on the 5 th or 6 th day post seeding.The Trans Epithelial Electrical Resistance (TEER) for BeWo monolayers were measured and found to be very low (90 -150 Qcm 2 ) 10 .Therefore, the integrity of the membrarf&s were assessed by performing a 14 C-sucrose flux for lh after the fluorescent probe transport assay.Vectorial transport of calcein AM, fluorescein or rhodamine was investigated in the presence or absence of the inhibitor.The receiver compartment contained HBSS and the inhibitor, while the donor contained the fluorescent probe and the inhibitor (the control did not have the inhibitor in both compartments).The volume on the apical side was 1.0 \iL and the basolateral compartment had 2.0 uL.At various time intervals, a 100 uL sample was removed from the receiver compartment and replaced immediately with an equal volume of the receiver solution.Transport was performed for 2h.The monolayers were then rinsed in ice-cold HBSS and sucrose flux assay performed.
Data Analysis: Apparent permeability coefficients (P ) were calculated according to the following equation: Where V is the volume of the receiver chamber, A is the area of the monolayer, C o is the initial donor concentration, and dQ/dt is the flux of the probe.

Kinetic studies analysis:
The rate of uptake was measured in uptake studies performed with rhodamine, fluorescein or with fluorescein in the presence of 10 \iM MK 571 at concentrations ranging from 1-150 uM.The fluorescent probes were allowed to be accumulated by the cells for 5, 10,15 and 30 min and quenched at the designated times with ice cold HBSS.The cells were lysed and assayed as described above in uptake studies.
The results were analyzed by Michaelis -Menten kinetics using the equation: m the Sigma plot programme.

Calcein accumulation and efflux
The accumulation of calcein in BeWo cells after uptake of luM Calcein AM for lh our was carried out in the presence and absence of selected agents (Figure 1).The Pgp inhibitors cyclosporinA and verapamil, and the Pgp/ MRP 1 inhibitor, vinblastine and the MRP 1 inhibitors indomethacin and probenecid (at 500 uM only) all significantly increased calcein/ calcein AM uptake.However, sodium orthovanadate and sodium azide which inhibit ATPase activity, decreased the accumulation of calcein in the cells.C219, a mouse monoclonal antibody to MDR1 (Pgp) also increased the uptake in a manner similar to cyclosporinA.Hence, inhibitors of Pgp and MRPl increase the cellular uptake of calcein AM.Monitoring the efflux of calcein from loaded cells revealed the function of the efflux protein MRPl in releasing calcein into the external medium.In two different experiments, calcein AM was loaded into the cells in the presence (Figure 2a) or in the absence (Figure 2b) of the inhibitors.The efflux was then monitored for two hours in the presence of the respective inhibitor.When the cells were loaded in the presence of the respective Pgp inhibitors, cyclosporin A showed a small increase, but verapamil (at 5 and 50 uM) did not greatly affect the amount of calcein effluxed from the cells.However, MRPl inhibitors vinblastine and indomethacin decreased efflux during the two hour efflux period (Figure 2a).The intracellular calcein retention post efflux showed the same trend of increased calcein concentrations as seen for uptake in the presence of all the inhibitors except vanadate which had a very low intracellular calcein concentration.The high intracellular calcein concentration observed with loading cells in the presence of cyclosporin A may account for the increased "efflux" if passive diffusion comes into play.This is a likely explanation as evidenced in Figure 2b where cyclosporin A did not greatly affect the efflux when cells were loaded without the inhibitor to ensure equal intracellular calcein concentrations.The efflux of calcein when cells were allowed to take up calcein AM in the absence of inhibitors, but effluxed in the presence of inhibitors, is shown in Figure 2b.Orthovanadate and the MRP inhibitor indomethacin decreased the efflux of calcein as expected.To further clarify the effect on efflux, a third experiment was performed wherein the cells were loaded in the absence of inhibitors, then allowed to efflux in the absence of inhibitors, until inhibitors were added at the 59 th minute.Thereafter efflux was monitored in the presence of inhibitor for two hours (Figure 2c).This strategy resulted in a distinct decrease in efflux after addition of inhibitors for all compounds tested (except 1 u,M vinblastine).Thus, the effect of these inhibitors on the efflux protein MRPl can be observed in this experiment.

Effect of inhibitors on uptake and efflux of Calcein
Efflux of Calcein with inhibitors  Effect of modulators on uptake of rhodamine.
Rhodamine accumulation by BeWo cells was measured at 0.5 uM, luM and 5 uM rhodamine concentrations in the presence of several modulators of both Pgp and MRP 1.
The inhibitors used were cyclosporin A (5 and 10 uM ), 50 ^M verapamil, 100 uM indomethacin, 100 \iM probenecid, 100 uM sodium orthovanadate, 10 ^M vinblastine, 500 uM triethylamine hydrochloride, 100 uM dipyridamole, 100 uM Quinidine, 100 jxM MPP + , 0.1% sodium azide and C219.Pre incubation with azide and consequent treatment with cyclosporin A and triethylamine was also performed.None of the modulators tested above were able to significantly increase the uptake of rhodamine into BeWo cells.The same was observed when a combination of two inhibitors from MPPVTEA/cyclosporin A were used.(All results not shown).
Effect of modulators on uptake of fluorescein.
Accumulation of fluorescein was measured at two concentrations.At a fluorescein concentration of 100 [iM, modulators had no effect on uptake (results not shown).This could be due to saturation of the efflux pump at high concentration.When uptake was performed using 1 uM fluorescein, MRPl inhibitors were able to increase the uptake of fluorescein, but Pgp in inhibitors had no effect (Figure 3).Again, the ATPase inhibitors, azide and orthovanadate did not increase uptake.

Kinetics of rhodamine and fluorescein uptake.
Effect of rhodamine concentration on uptake was measured and Km was obtained as 229.7 uM (Figure 4a).Vmax was found to be 0.49 nmol/mg protein/min.Kinetics of fluorescein uptake indicated a Km of 247.9 ^iM and a  Vmax of 1.58 nmol/mg protein/min (Figure 4b).An uptake study of fluorescein at varying concentrations with and without the presence of the inhibitor MK571 indicated that uptake in the first 10 minutes was not affected by the inhibitor.However low fluorescein concentrations (1 and 10 £iM) MK 571 increased the overall uptake in 30 minutes (Figure 5).At 100 uM concentration, the inhibitor did not increase uptake as was seen here as well as in previous studies 3. From this data, the rate of passive influx was calculated to be 0.05 and efflux to be 0.07 nmol/mg protein/ min at l^M, and 0.18 and 0.33 nmol/mg protein/min respectively at 10 ^M fluorescein concentration.At these concentrations the rate of influx is less than the rate of efflux which accounts for the MDR effect.The rate of influx atlOO^M concentration, was 1.71 nmol/mg protein/ min, which is much greater than the rate of efflux which was 0.18 nmol/mg protein/min.A notable observation is the low uptake concentration of ~0.2 nmol/mg protein for uptake at 1 ^,M rhodamine (Figure not shown) in comparison to -1.0 nmol/mg protein for uptake at 1 ^M calceinAM (Figure 1) and fluorescein (Figure 3) in agreement with the 3 fold difference in their Vmax values.

Transport of rhodamine and fluorescein.
Transport of 5 \iM rhodamine was performed and values of 1.1 -1.8x 10" 5 cms 1 were obtained for apical to basolateral direction.The values for the basolateral to apical direction ranged from 1.0 -1.8 x 10" 5 cms -1 .Thus there was no polarized transport.These values did not change significantly when transport was performed in the presence of the inhibitors cyclosporinA or verapamil and remained in the range 1.4 -1.6x10" 5 cms 1 for transport in both directions.
Fluorescein transport for the apical to basolateral direction was almost similar to that of the basolateral to apical direction.The inhibitors cyclosporinA, indomethacin and MK 571 resulted in a small decrease in the permeability ratio, which was not significant (Table 1).The intracellular fluorescein accumulation was found to be two fold greater for apical loading than for basolateral loading as observed in the case of calcein accumulation during transport (Table 2).These transport values are comparable to transport values of 1.0-1.5 x 10 5 cms 1 obtained for rhodamine and sucrose.The transport of fluorescein across membranes coated with collagen only and devoid of BeWo cells was 2.8 x 10~5 cms 1 in both directions.

Transport of calcein across monolayers.
Calcein showed polarized transport in the apical (AP) to basolateral (BL) direction.Measurement is made of the calcein that is transported across the monolayer.In all cases, the intracellular calcein accumulated for the AP to  BL direction was 3-6 times higher than the intracellular calcein retained for the BL to AP direction (Table 2).This was much greater than the difference observed with fluorescein.
In the case of orthovanadate, the BL loading intracellular concentration was 10 times less than the AP loading intracellular concentration.The permeability values obtained indicate that BL-AP permeability is not affected in all cases except with orthovanadate (Table 2).The AP-BL permeability is decreased in all cases.However the ratio of basolateral to apical transport did not reach unity in any of these cases.Thus, transport is complicated due to the fact that both MRP 1 and Pgp act on the transport of calcein and efflux of calcein AM.The result observed with sodium orthovanadate was a decrease in the Basolateral to Apical transport ratio, which was not seen with any of the other inhibitors. 14C Sucrose transport was measured for membrane integrity.The permeability value for paracellular transport of sucrose was 1.5 x 10" 5 cms 1 .The permeability values obtained for calcein transport was approximately 10 fold lower than those of fluorescein and rhodamine transport.Rhodamine contains a positive charge whereas calcein th and fluorescein are negatively charged.Calcein AM is a neutral molecule.Several workers have reported the use of rhodamine as a marker for the study of uptake and lg transport in many cell systems 911 .Since we were unable to LP observe the expected results with rhodamine, calcein AM ty and fluorescein were selected as the probes.We found ot that the predicted results were obtainable with both these -)• probes in uptake studies: MRPl and Pgp inhibitors S-affected the uptake of calcein AM (Figure 1) and only id MRPl inhibitors had an effect on the uptake of fluorescein is (Figure 3  modulators influence the uptake of the fluorescent >rt substrate 14 .We have shown that at 1 and 10 \iM fluorescein concentrations, the rate of efflux is greater than the passive influx (Figure 5), and that at 100 jxM the rate of efflux was negligible compared to influx which explains the observation of the MDR effect at 1 \iM and not at 100 ^M.Fluorescein uptake was increased by the MRPl inhibitors probenecid, indomethacin, MK571 and vinblastine.Pgp in inhibitors monoclonal antibody C219 and verapamil did not affect uptake of fluorescein.However, the ability of cyclosporin A to inhibit MRPl is demonstrated by the increased uptake of fluorescein (Figure 3).We found that fluorescein had a Km of 247.9 u,M comparable to that of 229.7uM found for rhodamine and 268 uM for calcein 15 .
The efflux of calcein was inhibited by several MRPl inhibitors.When the cells were loaded with calceinAM in the presence of inhibitors of Pgp or MRP 1, the intracellular concentration of accumulated calcein was increased and the efflux was affected by the MRPl inhibitors only.The Km for calcein AM is .05uM for MRPl and 0.12 uM for Pgp 15 .However, the Km of MRPl for calcein is 268 uM.This means that calcein is less efficiently transported than calceinAM.The absence of any effect on the permeability of calcien in the BL to AP direction may be due to the very low intracellular calcein concentration and the high Km of calcein.Sodium orthovandate is a Pgp inhibitor.It was also shown to bind and inhibit MRP 1, but is not transported by MRPl 16 .The low uptake of calcein, fluorescein and rhodamine in the presence of sodium orthovandate and sodium azide may be related to the fact that sodium orthovanadate suppresses rhodamine 123 accumulation in primary rat hepatocyte cultures expressing low levels of pgp 17 .However when the cells were pretreated with sodium azide prior to uptake, and uptake performed in the absence of azide, there was no observed suppression of uptake of rhodamine (results not shown).This leads us to assume that the presence of azide (or vanadate) during the entire duration of uptake or transport may lead to toxicity of cells causing a leakiness in the cells.This also explains the decreased uptake of calcein, and the 10 fold decrease in residual calcein retained in cells during transport (Table 2).
Transport of calcein was monitored by addition of calcein AM to the donor compartment and assaying the receiver for appearance of calcein.From the data presented in table 2, it appears that the transport from the basolateral to apical compartment is unaffected by modulators.The greater permeability from apical to basolateral direction may be due to the increased intracellular calcein concentration compared to the low intracellular concentration observed with basolateral to apical transport.It is known that MRPl is located on the basolateral membranes of hepatocytes 18 , choroid plexis 19 Since uptake and transport of rhodamine was not affected by any of the inhibitors used, several different concentrations of rhodamine were tried.However, even at 1 uM concentration, there was no increased uptake in the presence of the modulators tested (results not shown).It was not possible to measure fluorescence at concentrations below 1 uM and a kinetic study of uptake was attempted to determine the Km value.A Km value of 229.7 uM was obtained 17 , have reported that the apparent Km measured as the dye concentration required for half maximal stimulation of ATPase was found to be 40 [xM.Rhodamine 123 has been shown to have a fast transmembrane movement rate, and lowest dye binding rate when compared to other rhodamine dyes 21 .The lack of the MDR effect with rhodamine may be attributed to a low efflux rate at concentrations below 40 uM compared to influx rates at the concentrations tested.Since both fluorescein and calcein are anions, they are substrates of MRPl which functions as a transporter of amphiphilic anions and natural product anti cancer drugs.Rhodamine is cationic and is not a substrate of the MRP 1.In the absence of rhodamine, calcein AM is the only fluorescent probe that can be used to assess the functionality of Pgp.We have successfully used radiolabeled substrates of Pglycoprotein to demonstrate that modulators of Pgp do effect the uptake and transport of the radiolabeled substrate (results not shown).In order to understand the anomaly in the rhodamine uptake, a vinblastine resistant cell line with increased expression of Pgp may be used as a model.The increased expression of Pgp has shown large differences in uptake in vinblastine resistant Caco-2 cells 22 .
Although the existence of Pgp in the placenta is well known, an understanding of the functional activity at the Corresponding author Part of this work was presented at the Peradeniya University Research Sessions 2005, Peradeniya.

Figure 1 :
Figure 1: Uptake of 1 uM Calcein AM in the presence of various inhibitors over a period of 1 hour All data points were performed in quadruplicate.The concentrations of inhibitors are in uM Azide used was 0.1% Antibody C219 was diluted 1 40 Stars indicate significant increase in uptake by students T test, (p < 0 05 compared with control).
Cells loaded in the presence of inhibitor and effluxed in the presence of inhibitor.Cells loaded in the absence of inhibitor and effluxed in the presence of inhibitor.Effect of inhibitors on efflux of calcein Cells loaded in the absence of inhibitor, effluxed for 1 hour without inhibitor and then added inhibitor and effluxed a further two hours.Arrow indicates point of addition of inhibitors.

Figures 2 :
Figures 2: Efflux of Calcein decreased on addition of Pgp and MRPl inhibitors.The average of 4 experimental wells were taken for each data point.

Figures 3 :Figures 4a :Figures 5 :Fluorescein
Figures 3: Effect of inhibitors on fluorescein uptake.Each data point is an average of four wells.Stars indicate significant increase in uptake by students T test, (p <0.05 compared with control).
March 2007Journal of the National Science Foundation of Sri Lanka 35(1) DISCUSSIONdi lc Fluorescent probes are useful indicators for assessing p whether compounds are inhibitors or modulators of Pgp.rr This study was performed as a comparison of the transport si fluorescent indicators in BeWo cells.Li s a,b).When 100 uM fluorescein was used, there ct was no effect on the uptake by any of the modulators.An ie earlier study 3 using 100 uM fluorescein was interpreted a as a sign of absence of MRPl expression in BeWo cells.:n However, the presence of MRPl mRNA in BeWo cells se has been confirmed 512 .This lack of effect by modulators ie observed earlier was likely due to a saturation of the efflux se protein (MRP 1) at high fluorescein substrate or concentration.It has been shown that the rate of Pgp m mediated efflux must be greater than the rate of transmembrane passive influx to observe the effect of ~ modulators 13 .The efficiency of substrate extrusion depends on slow passive transmembrane movement.At low concentration (luJVl fluorescein), the passive ig permeability effect is minimal, and therefore the p.
Karunaratne &, Kenneth L. Audus and kidney cells 20 .The ability of MRPl inhibitors to decrease the AP to BL permeability supports the basolateral location of MRPl.

Table 2 :
Effect of inhibitors on the permeability of calcein, and the residual intracellular concentration.

Table 1 :
Effect of inhibitors on the transport of fluorescein.