optimal structure requirements for pluronic block...

Optimal Structure Requirements for Pluronic Block Copolymersin Modifying P-glycoprotein Drug Efflux Transporter Activity inBovine Brain Microvessel Endothelial Cells

ELENA V. BATRAKOVA, SHU LI, VALERY YU ALAKHOV, DONALD W. MILLER, and ALEXANDER V. KABANOV

Department of Pharmaceutical Sciences, University of Nebraska Medical Center, Omaha, Nebraska (E.V.K., S.L., S.V.V., D.W.M., A.V.K.); andSupratek Pharma Inc., Laval, Quebec, Canada (V.Y.A.)

Received August 27, 2002; accepted October 14, 2002

ABSTRACTPluronic block copolymer P85 was shown to inhibit the P-glycoprotein (Pgp) drug efflux system and to increase the per-meability of a broad spectrum of drugs in the blood-brainbarrier (BBB). However, there is an entire series of Pluronicsvarying in lengths of propylene oxide and ethylene oxide andoverall lipophilicity. This study identifies those structural char-acteristics of Pluronics required for maximal impact on drugefflux transporter activity in bovine brain microvessel endothe-lial cells (BBMECs). Using a wide range of block copolymers,differing in hydrophilic-lipophilic balance (HLB), this studyshows that lipophilic Pluronics with intermediate length of pro-pylene oxide block (from 30 to 60 units) and HLB �20 are themost effective at inhibiting Pgp efflux in BBMECs. The methods

used included 1) cellular accumulation studies with the Pgpsubstrate rhodamine 123 in BBMECs to assess Pgp activity; 2)luciferin/luciferase ATP assay to evaluate changes in cellularATP; 3) 1,6-diphenyl-1,3,5-hexatriene membrane microviscos-ity studies to determine alterations in membrane fluidity; and 4)Pgp ATPase assays using human Pgp-expressing membranes.Pluronics with intermediate lipophilic properties showed thestrongest fluidization effect on the cell membranes along withthe most efficient reduction of intracellular ATP synthesis inBBMEC monolayers. The relationship between the structure ofPluronic block copolymers and their biological response-mod-ifying effects in BBMECs are useful for determining formula-tions with maximal efficacy for increasing BBB permeability.

Drug delivery systems based on synthetic polymers haveattracted significant attention during the last two decades(Yokoyama, 1992; Alakhov and Kabanov, 1998; Kwon andOkano, 1999). An important and promising example of thesesystems is Pluronic block copolymers, which were shown toenhance drug performance by acting as biological response-modifying agents. Specifically, they sensitize multidrug-re-sistant cells by inhibiting drug efflux transporters (Venne etal., 1996). Moreover, Pluronic formulations were shown toenhance transgene expression in the body (Lemieux et al.,2000). One particular composition, Pluronic block copolymerP85 (P85), was found to improve the transport of select sol-utes across the BBB in vitro (Batrakova et al., 1998, 1999)and in vivo (Kabanov et al., 1989; Batrakova et al., 2001b).Using monolayers of polarized bovine brain microvessel en-

dothelial cells (BBMECs) as an in vitro model of the BBB itwas demonstrated that coadministration of P85 significantlyincreased the transport of various Pgp substrates acrossbrain microvessels by inhibiting the Pgp efflux transportsystem (Batrakova et al., 1998, 1999). Similar responseswere observed in vivo because administration of P85 signif-icantly enhanced the brain penetration of the Pgp substratedigoxin in wild-type mice expressing functional Pgp to levelsthat were similar to those observed in Pgp-deficient knockoutmice (Batrakova et al., 2001b). The mechanism of the effect ofthis copolymer on the drug efflux system in the BBB has beendiscussed recently (Batrakova et al., 2001a). It was foundthat the P85 treatment caused membrane fluidization lead-ing to decreases in Pgp ATPase activity (Batrakova et al.,2001a). Furthermore, P85 caused significant depletion of cel-lular ATP in BBMEC monolayers that affected the ATP-dependent Pgp efflux transport system and increased drugdiffusion across the blood-brain barrier cell monolayers. Con-sequently, both membrane fluidization (inhibiting Pgp AT-Pase activity) and energy depletion (decreasing the ATP pool

This study was supported by National Institutes of Health Grants NS36229(to A.V.K.) and A617294 (to D.W.M.).

Article, publication date, and citation information can be found athttp://jpet.aspetjournals.org.

DOI: 10.1124/jpet.102.043307.

ABBREVIATIONS: BBB, blood-brain barrier; BBMEC, bovine brain microvessel endothelial cell; EO, ethylene oxide; PO, propylene oxide; CMC,critical micelle concentration; HLB, hydrophilic-lipophilic balance; FITC, fluorescein isothiocyanate; HUVEC, human umbilical vein endothelial cell;R123, rhodamine 123; PBS, phosphate-buffered saline; Pgp, P-glycoprotein; MES, 4-morpholineethanesulfonic acid; DPH, 1,6-diphenyl-1,3,5-hexatriene; TMA, 1-[4-(trimethylamino)phenyl]-6-phenylhexa-1,3,5-triene.

0022-3565/03/3042-845–854$7.00THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Vol. 304, No. 2Copyright © 2003 by The American Society for Pharmacology and Experimental Therapeutics 43307/1041304JPET 304:845–854, 2003 Printed in U.S.A.

845

at ASPE

T Journals on July 15, 2018

jpet.aspetjournals.orgD

ownloaded from

http://jpet.aspetjournals.org/

available for Pgp) were found to be critical factors contribut-ing to the activity of the block copolymer in the BBB.

Pluronic block copolymers consist of ethylene oxide (EO)and propylene oxide (PO) blocks arranged in a basic A-B-Astructure: EOx-POy-EOx. Due to their amphiphilic nature,Pluronic block copolymers are able to self-assemble into mi-celles in aqueous solutions above critical micelle concentra-tion (CMC). Below the CMC, Pluronic copolymers exist insolution in the form of a molecular dispersion of individualblock copolymer molecules termed “unimers” (Alexandridiset al., 1994). Variations in the number of hydrophilic EOunits (x) and lipophilic PO units (y) result in copolymers withdifferent molecular mass and distinct hydrophilic-lipophilicbalance (HLB). Copolymers with a short hydrophilic poly-EOblock or/and an extended lipophilic poly-PO block (such asPluronic L121 and L101) are highly lipophilic and are char-acterized by a relatively low CMC and low HLB. In contrast,copolymers with an extended hydrophilic poly-EO block or/and short lipophilic poly-PO block (such as Pluronic F108 andF88) are hydrophilic and are characterized by relatively highCMC and high HLB. Pluronic compositions such as P85 orP103 are intermediate in their lipophilicity and have CMCand HLB values that fall between the two extremes identi-fied above.

Although most experiments examining the effects of Plu-ronic on BBB permeability were performed with PluronicP85, there is an entire series of Pluronic block copolymerswith differing molecular properties. The availability of a widevariety of Pluronic compositions provides a unique opportu-nity to identify those structural features that are importantfor the effects of the block copolymer on drug efflux trans-porter activity in BBB. In the current study, a series ofPluronic block copolymers with a wide range of HLB wereused to identify 1) those composition with the best trans-porter activity profiles in brain endothelial cells, and 2) theimpact of copolymer composition on critical factors (i.e., en-ergy depletion and membrane fluidization) known to influ-ence Pgp transporter activity in BBMEC monolayers.

Materials and MethodsPreparation of Block Copolymer Solutions. The list of the

block copolymers used in this work and their molecular characteris-tics is presented in Table 1. All Pluronic block copolymers used werekindly provided by BASF Corp. (Parispany, NJ). Aqueous solutions

of Pluronic block copolymer were prepared in assay buffer containing122 mM sodium chloride, 25 mM sodium bicarbonate, 10 mM glu-cose, 10 mM HEPES, 3 mM potassium chloride, 1.2 mM magnesiumsulfate, 1.4 mM calcium chloride, and 0.4 mM potassium phosphatedibasic, pH 7.4. Pluronic solutions were incubated for at least 1 h at37°C before using. For the microscopy studies Pluronics P85, L35,F108, and L121 were labeled by fluorescein isothiocyanate (FITC)attached to the one of the block copolymer free ends as describedpreviously (Beauchamp et al., 1983). Briefly, Pluronic block copoly-mers were activated with 1,1�carbonildiimidazole, and then modifiedwith excess of ethylenediamine and purified by gel filtration. Amino-modified copolymers were conjugated with fluorescein isothiocya-nate according to manufacturer’s protocol. A gel-permeation chroma-tography was used to ensure that FITC was attached to the blockcopolymer. Two-step separation was performed: 1) rough separationon Sephadex PD-10 column (Sigma-Aldrich, St. Louis, MO) in ace-tonitrile/water (1:1) phase, and 2) precise separation on SephadexLH-20 (Sigma-Aldrich) column in ethanol/water (1:1) phase. FITC-labeled Pluronic was eluted in the first distant peak, and detected byspectrofluorimeter (for FITC) and iodine test (for Pluronic). Secondpeak of nonconjugated FITC and low molecular products was elutedwith a substantial delay. The yield of FITC-Pluronic conjugation wasca. 70%.

Cell Isolation and Culture. BBMECs were isolated from freshcow brains using a combination of enzymatic digestion and densitycentrifugation as described previously (Miller et al., 1992). The cellswere maintained in minimal essential medium/F-12 culture mediumsupplemented with 10% horse serum, 100 �g/ml heparin sulfate, 2.5�g/ml amphotericin B, and 50 �g/ml gentamicin. Tissue culturemedia were obtained from Invitrogen (Carlsbad, CA), and serum andmedium supplements were purchased from Sigma-Aldrich. IsolatedBBMECs were seeded at a density of 50,000 cells/cm2 in 24-wellplates and were used for the ATP assay and R123 accumulationstudies after reaching confluence (typically within 14 days).

Characterization of Pgp Expression in BBMEC Monolay-ers. Identification of Pgp was done using immunoblot techniquedescribed previously (Miller et al., 1996). The monoclonal antibodiesto Pgp, C219 (DAKO, Carpinteria, CA), and �-actin, anti-�-1-chickenIntegrin (Sigma-Aldrich), were used at 1:100 and 1:200 dilutions,respectively. The secondary horseradish peroxide anti-mouse Ig an-tibodies (1:1500 dilution) were purchased from Amersham Bio-sciences, Inc. (Cleveland, OH). The specific protein bands were visu-alized using a chemiluminescence kit (Pierce Chemical, Rockford,IL). The level of Pgp expression was quantified by densitometry(Nucleo Vision; Nucleo Tech, Curitiba-Pr., Brazil). To correct forloading differences, the level of the protein was normalized to con-stitutively expressed �-actin. The relative amount of the protein inthe Pgp-overexpressing human oral epidermal carcinoma (KBv) de-

TABLE 1Physicochemical characteristics of Pluronic block copolymers

Copolymer mol.wt. Average no. of EO units (x)a Average no. of PO units (y)a HLBb Lot no. CMC, %wtc

F88 11,400 207.27 39.31 28 WPAS-575B 0.28F108 14,600 265.45 50.34 27 WPON-522C 0.032F38 4,700 37.56 17.1 25 WPDR-504B N.A.F127 12,600 200.4 65.2 22 WPMN-581B 0.004L35 1,900 21.59 16.38 19 WPMQ-592D 1P85 4,600 52.27 39.66 16 WPOP-587A 0.03L64 2,900 26.36 30.00 15 WPAQ-561B 0.14P105 6,500 73.86 56.03 15 WPER-598D 0.004L43 1,850 12.61 22.33 12 WPMS-508B 0.4P103 4,950 33.75 59.74 9 WPWQ-557B 0.003L81 2,750 6.25 42.67 2 WSOO-83457 0.0063L101 3,800 8.64 58.97 1 WPHP-547B 0.0008L121 4,400 10.00 68.28 1 WPAO-550B 0.0004

N.A., not applicable.a The average numbers of EO and PO units were calculated using the average molecular weighs (mol.wt.) provided by the manufacturer.b HLB of the copolymers were determined by the manufacturer.c CMC values were determined previously using pyrene probe (Kozlov et al., 2000).

846 Batrakova et al.

at ASPE



ownloaded from


rived by selection with vinblastine were used as a positive control,and human umbilical vein endothelial cells (HUVECs) was used as anegative control.

R123 Accumulation Studies. R123 accumulation in BBMECswas studied as described previously (Miller et al., 1997). Briefly,confluent cell monolayers were preincubated with the assay bufferfor 30 min at 37°C, and then the assay buffer was removed and cellmonolayers were exposed to 3.2 �M R123 in either assay buffer orPluronic solutions at different concentrations for 90 min. After theincubation the dye solutions were removed, the cell monolayers werewashed three times with ice-cold PBS, and solubilized in TritonX-100 (1.0%). Aliquots were removed for determination of the cellu-lar dye using an RF5000 fluorescent spectrophotometer (�ex � 505nm, �em � 540 nm) (Shimadzu, Kyoto, Japan) and cellular proteinusing the Pierce bicinchoninic acid assay. All experiments werecarried out in quadruplicate.

The effects of Pluronic compositions on Pgp activity were ex-pressed as the R123 enhancement factor (maximal R123 accumula-tion levels in the presence of Pluronic versus those observed in thecontrol groups in the absence of the block copolymer).

ATP Assay. To examine the effects of Pluronics on ATP intracel-lular levels the confluent BBMEC monolayers were pretreated withassay buffer for 30 min after which the cells were incubated withvarious concentrations of Pluronic solutions for 2 h. After treatment,the cells were washed two times with ice-cold PBS, solubilized inTriton X-100 (1.0%), and immediately frozen for subsequent ATPquantification (conducted within 24 h after the sample collection).Cellular ATP was determined using a luciferin/luciferase assay(Garewal et al., 1986). For this purpose, 100-�l aliquots of cell lysatewere mixed with 100 �l of ATP assay mix (FL-AAM; Sigma-Aldrich).Light emission was measured with a luminometer (model 20/20;Turner Designs, Inc., Sunnyvale, CA). Raw data were collected asrelative light units integrated over 20 s for samples and converted toATP concentrations with the aid of a standard calibration curveobtained using ATP standard (FL-AAS; Sigma-Aldrich). CellularATP levels were normalized for protein content and each data pointrepresented the mean � S.E.M. of a minimum of four replicates.

Pgp ATPase Assay. Membranes from Pgp-overexpressing cellswere used to evaluate effects of P85 on Pgp ATPase activity (BDGentest, Woburn, MA). A 0.06-ml reaction mixture containing 40 �gof membranes, 20 �l of the various Pluronic compositions or assaybuffer, and 3 to 5 mM MgATP, in a 50 mM Tris-MES buffer contain-ing 2 mM EGTA, 50 mM KCl, 2 mM dithiothreitol, and 5 mM sodiumazide, pH 6.8. The membrane samples were incubated at 37°C for 20min. An identical reaction mixture containing 100 �M sodium or-thovanadate was assayed in parallel. Orthovanadate inhibits Pgp bytrapping MgADP in the nucleotide-binding site. Thus, ATPase activ-ity measured in the presence of orthovanadate represents non-PgpATPase activity and can be subtracted from the activity generatedwithout orthovanadate to yield vanadate-sensitive ATPase activity(Pgp ATPase activity). The reaction was stopped by the addition of 30�l of 10% SDS with Antifoam A. Aliquots (200 �l) of ammoniummolybdate in 15 mM zinc acetate/10% ascorbic acid (1:4) were addedto each sample and incubated for an additional 20 min at 37°C. Theliberation of inorganic phosphate was detected by its absorbance at630 nm and quantitated by comparing the absorbance to a phosphatestandard curve (Druekes et al., 1995; Shepard et al., 1998).

DPH and 1-[4-(Trimethylamino)phenyl]-6-phenylhexa-1,3,5-triene (TMA-DPH) Labeling of BBMECs. DPH was used asa probe to examine the fluidity properties of the hydrocarbon regionof the cell membranes. DPH is a hydrophobic fluorescent compoundthat spontaneously incorporates in the hydrocarbon regions of lipidmembranes (Shinizky and Inbar, 1967; Laat et al., 1977). Transfer ofDPH from the aqueous environment into the cell membranes resultsin a drastic increase of the fluorescence emission for this probe.Furthermore, once the probe is incorporated into lipid membranes,its fluorescence polarization is strongly dependent on the microen-vironment, with decreases in membrane microviscosity resulting in

increased fluorescent polarization. It should be noted that DPHbinds not only with the plasma membranes but also with othermembranes within the cells, thus the polarization value obtainedreflects the overall membrane microviscosity of the cells (Pagano etal., 1977). Consequently, once the polarization changes are observed,it is difficult to discriminate which membranes (i.e., plasma or in-tracellular organelle) are affected. To examine effects of Pluronics onplasma membranes of BBMECs, the cationic analog of DPH, TMA-DPH, was used. The cationic charge of this probe ensures thatTMA-DPH is anchored at the lipid-water interface, whereas the DPHmoiety is intercalated between the upper portions of the lipid milieu(Prendergast et al., 1981). For these studies, the BBMEC suspensionwas washed twice with PBS and incubated with 2 �M DPH (Sigma-Aldrich) dispersion for 1 h at 37°C. For TMA-DPH studies, cells wereincubated with 2 �M TMA-DPH (Molecular Probes, Eugene, OR) for10 min at 37°C. Then, the cells were washed twice with PBS toremove extracellular probe, and resuspended in an appropriate vol-ume of PBS. To evaluate the kinetic effects of Pluronics in BBMECs,four different Pluronic compositions representing each group wereadded at concentrations producing maximal inhibition of the Pgpefflux system in BBMECs, and changes in fluorescent polarizationwere recorded.

Fluorescence Polarization Measurements. Fluorescence in-tensities were measured with a Hitachi F5000 spectrophotometerequipped with a polarizer set. This instrument detects fluorescenceintensity (I) with the relative position of the polarizer and analyzer(parallel, I�, or perpendicular, I�) and fluorescence anisotropy r, wascalculated according to eq.1:

r � �I� � I��/ � �I� � 2I�� (1)

An excitation wavelength of 365 nm and an emission wavelength of425 nm were used for both probes. Cell suspensions were gentlymixed before each reading. In all cases, corrections for stray lightand intrinsic fluorescence were made by subtracting the values for I�

and I� of unlabeled samples from those of identical but labeledsamples.

Microviscosities (�) were derived as described previously for DPH(Shinizky and Inbar, 1967; Laat et al., 1977) and TMA-DPH (Cha-zotte, 1994) by the method based on the Perrin equation (eq. 2) forrotational depolarization of a nonspherical fluorophore:

r0/r � 1 � C�r�T� /� (2)

where r0 and r are limiting and measured fluorescence anisotropies,T is the absolute temperature, and � is the exited state lifetime. Thevalue of r0 used for both probes was 0.362; � values were 10 and 7 nsfor DPH and TMA-DPH, respectively. C(r) is a molecular shapeparameter equal to 8.6 � 105 poise � deg�1 s�1 and 15.3 � 105 poise �

deg�1 s�1 for DPH and TMA-DPH, respectively.Fluorescent Microscopy. BBMECs grown on chamber slides

(Fisher Scientific Co., Fair Lawn, NJ) were incubated with 0.1%F108-FITC, P85-FITC, L35-FITC, and L121-FITC in assay buffer for2 h at 37°C. After this period, the loading solutions were removed,and the cell monolayers were washed three times with ice-cold PBScontaining 1% bovine serum albumin and examined using an ACAS-570 (Meridian Instruments, Okimos, MI) confocal laser microscope.

Cytotoxicity Assay. To examine the possible cytotoxic effect ofstudied block copolymers, BBMEC were seeded in 96-well plates at adensity of 5000 cells/well and allowed to reattach overnight. Then,the cells were exposed to various concentrations of Pluronic solutionsfor 2 h at 37°C. After this treatment cells were washed three timesand cultured for 3 days in the media. The cytotoxic effects weredetermined using a standard 3-(4,5-dimethylthiazol-2-yl)-2,5-diphe-nyltetrazolium assay (Ferrari et al., 1990). All experiments wererepeated eight times. No cytotoxic effects of Pluronic block copoly-mers were observed over entire range of concentrations used in thisstudy.

Optimization of Pluronic Compositions in BBMECs 847

at ASPE



ownloaded from


Statistical Analysis. All statistical tests were performed by Mi-crosoft Excel 97 SR-1 program using the two-tailed heteroscedastic ttests. A minimum p value of 0.05 was estimated as the significancelevel for all tests. S.E.M. values for R123 accumulation levels, mi-croviscosity, and ATP measurements were less than 10% of themean.

ResultsEffects of Pluronics on R123 Accumulation in

BBMECs. Measurements of cellular accumulation of R123, asubstrate of Pgp, have been commonly used for evaluatingthe functional activity of Pgp in cells (Jancis et al., 1993; Leeet al., 1994, Fontaine et al., 1996). In the preliminary study,we examined the level of Pgp expression in BBMEC mono-layers by Western blot. The determined amount of the pro-tein was normalized to the amount of constitutively ex-pressed �-actin. Pgp-overexpressing human oral epidermalcarcinoma (KBv) cells derived by selection with vinblastinewere used as a positive control for the drug transporterexpression, whereas the HUVECs were used as a negativecontrol. The relative amounts of Pgp in KBv, BBMEC, andHUVEC cells were 1.33, 0.81, and 0.07, respectively, whichconfirms substantial Pgp expression in BBMECs.

To estimate the ability of various Pluronics to inhibit thePgp efflux system in BBMEC monolayers, R123 accumula-tion studies in the presence and absence of 12 different Plu-ronic compositions were performed.

The concentration-dependent effects of four selected Plu-ronic compositions on R123 accumulation in BBMEC mono-layers are presented in Fig. 1A. As it seen in the figure, onepattern observed for all the block copolymers is that accumu-lation of R123 reaches maximal levels at or near the respec-tive copolymer CMC, and then decreases at concentrationsabove the CMC. This result is consistent with previous re-ports suggesting that the unimers of Pluronic (i.e., singlemolecular chains of block copolymer) are responsible for theinhibition of the Pgp efflux system in these cells (Batrakovaet al., 1998, 2001a). The effect of high Pluronic concentra-tions is believed to be due to incorporation of the probe in themicelles, resulting in a decrease in the amount of free probeavailable for diffusion into the cells.

The effects of the various Pluronic compositions on R123accumulation were plotted as a function of length of PO block(Fig. 1B). This parameter was chosen as an index of polymerlipophilicity. As seen in Fig. 1B, the effects of the variousPluronic block copolymers on R123 accumulation in BBMECmonolayers were dependent on the composition of the poly-mer. The hydrophilic copolymers with HLB ranging from 20to 29 (group I) had little affect on Pgp functional activity. Incontrast, the lipophilic copolymers (HLB �20) with interme-diate-length PO blocks (i.e., 30–60 units) (group II) werevery effective at inhibiting of Pgp activity in BBMEC mono-layers (Fig. 1B). Those lipophilic copolymers with the POblocks less than 30 (group III a) and longer than 60 (group IIIb) had little if any effect on R123 accumulation. These groupsare presented in Fig. 2 showing a grid of Pluronic blockcopolymers with different number of PO blocks (NPO) andHLB. This figure also depicts selective copolymer represen-tative of each group.

Effects of Pluronics on the Total Membrane Micro-viscosity in BBMECs. It has been shown recently that

Pluronic P85 treatment causes significant changes in mem-branes microviscosity and that these effects correlate withthe degree of inhibition of the Pgp efflux system in BBMECs(Batrakova et al., 2001a). To examine the effects of Pluronicstructure on membrane microviscosity, the interactions ofvarious Pluronic compositions with BBMEC membraneswere studied using the DPH fluorescence polarizationmethod. This compound has been used extensively as a mem-brane probe for assaying the microenvironment in the hydro-carbon regions of the lipid bilayer (Laat et al., 1977). It hasbeen shown that DPH binds with plasma membranes as wellas with other membranes within the cells, and thus the dataobtained with DPH reflect the net polarization value of allcell membranes (Pagano et al., 1977). We examined the time-

Fig. 1. A, concentration effects of various Pluronics on R123 accumula-tion in BBMECs. B, relationship between the R123 accumulation en-hancement factor and the length of lipophilic PO segment (NPO) in Plu-ronic block copolymers. BBMEC monolayers were exposed for 60 min to3.2 �M R123 in the assay buffer containing different concentrations ofPluronic copolymers: F108 (filled diamonds), L121 (filled triangles), P85(filled circles), and L35 (crosses). Thereafter, R123 accumulation en-hancement factors were determined as the ratios of R123 levels in thecells exposed to the dye in Pluronic solution (at the most effective con-centration) and assay buffer. Arrows correspond to CMC of each Pluronic.


at ASPE



ownloaded from


dependent changes in fluorescence polarization of DPH inBBMECs after exposure to the various Pluronic block copol-ymers. All Pluronics were added at concentrations producingmaximal inhibition of the Pgp efflux system in BBMECs. Themicroviscosity values were calculated from the polarizationmeasurements, as described under Materials and Methods.The changes in the total microviscosity of BBMECs exposedto representative Pluronic compositions from each group areshown in Fig. 3A.

There are two distinct effects caused by Pluronics inBBMEC cellular membranes. The hydrophilic block copoly-mer F88 (group I) and the lipophilic block copolymer with ashort lipophilic PO block L35 (group III a) caused solidifica-tion, i.e., increasing of the membrane microviscosity (Fig.3A), suggesting that molecules of these Pluronics adhere onthe cellular surface and limit the lateral mobility of themembrane lipids. In contrast, the lipophilic copolymers witheither long PO block L121 (group III b) or intermediate POblock P85 (group II) decreased the microviscosity in BBMECs(Fig. 3A), indicating their incorporation into the lipid bilayerand subsequent increase in membrane fluidization. Allchanges in microviscosity for each Pluronic were observedwithin the first 20 to 40 min after addition of the blockcopolymer to the cell suspension. After that time period, themicroviscosity leveled off and remained constant throughoutthe duration of the experiment.

The microviscosity data for each Pluronic were used forcalculating the total microviscosity factor (value of total cel-lular membrane microviscosity in the control groups in theabsence of the block copolymer versus those observed in thepresence of the Pluronic). After that, the microviscosity factorvalues were plotted versus length of lipophilic PO block foreach Pluronic (Fig. 3B). As seen in the figure, the graph isqualitatively similar to those results obtained with the R123enhancement factor (Fig. 1B). There are two separate phe-nomena: a bell-shaped curve corresponding to the lipophilicPluronics and a linear dependence corresponding to the hy-drophilic Pluronics. The similarities between the dependenceof the rhodamine enhancement factor versus length of POblock and dependence of the microviscosity factor versuslength of PO block suggest that there is a strong relationshipbetween the effects of Pluronics on the membrane microvis-

cosity and their ability to inhibit the Pgp efflux system inBBMECs.

Effects of Pluronics on Pgp ATPase Activity. Theeffect of various Pluronic compositions on Pgp ATPase activ-ity was also examined (Fig. 4). All Pluronics were added atconcentration producing maximal inhibition of the Pgp effluxsystem in BBMECs. The effects of four different Pluroniccompositions representing each group on Pgp ATPase activ-ity are shown in Fig. 4A.

Similar to the microviscosity data (Fig. 3A) there were twodistinct effects on Pgp ATPase activity caused by Pluronics.The hydrophilic block copolymer F108 (group I) and the li-pophilic block copolymer with a short lipophilic PO block L35(group III a) increased the Pgp ATPase activity. In contrast,the lipophilic copolymers with the long PO block L121 (group

Fig. 2. A grid of Pluronics indicating four groups determined based on theactivity of these copolymers displayed in BBMEC monolayers.

Fig. 3. A, kinetic effects of various Pluronics on the total microviscosity inBBMECs. B, relationship between the total microviscosity factor and thelength of lipophilic PO segment (NPO) in Pluronic block copolymers. Cellswere treated with F88 (filled diamonds), L121 (filled triangles), P85(filled circles), and L35 (crosses) at 37°C. Copolymers were used at theconcentrations that caused the most efficient inhibition of the Pgp effluxsystem in BBMECs.


at ASPE



ownloaded from


III b) or intermediate PO block P85 (group II) decreased thePgp ATPase activity in Pgp-overexpressing membranes.With regard to the data described in the previous section, itindicates that membrane solidification, i.e., decreasing themobility of membrane lipids caused by the hydrophilic Plu-ronics, results in an increase in Pgp ATPase activity. Incontrast, membrane fluidization, caused by the lipophilicblock copolymers, results in a decrease in Pgp ATPase activ-ity.

Despite these similarities, there is a major difference be-tween the effects of various Pluronics on the membrane mi-croviscosity and Pgp ATPase activity. Basically, the mostsignificant membrane fluidization was caused by the inter-mediate lipophilic Pluronics (group II) (Fig. 3A), although themost efficient inhibition of Pgp ATPase activity was causedby the extremely lipophilic Pluronics (group III b) (Fig. 4A).This is more clearly observed by plotting the Pgp ATPase

activity factor (values of Pgp ATPase activity in the controlmembranes in the absence of the block copolymer versusthose observed in the presence of the Pluronic) versus afunction of length of lipophilic PO block for each Pluronic(Fig. 4B).

As is seen in the figure, the S-shape curve for the PgpATPase activity factor (Fig. 4B) differs from the bell-shapecurves observed for the microviscosity and R123 enhance-ment factors (Figs. 1B and 3B) corresponding to the lipophilicPluronics. These differences may be due to differences inexperimental protocols. Whole cells were used for the micro-viscosity experiments, whereas cell membranes were used forthe Pgp ATPase activity studies. Therefore, the changes inthe total membrane microviscosity caused by Pluronicsmight depend on transport of the block copolymer inside thecells. In contrast, changes in the Pgp ATPase activity in thePgp membranes should not be affected by this factor. Toprove this suggestion, the effects of various Pluronics on theplasma membrane microviscosity of BBMECs were exam-ined.

Effects of Pluronics on the Microviscosity of PlasmaMembranes in BBMECs. Changes in the dynamics of theplasma membrane in BBMECs caused by various Pluronicswere studied by a fluorescence polarization method usingTMA-DPH. This cationic probe at early time points interactswith head groups of phospholipids and intercalates into theouter surface leaflet of the plasma membrane of cells, but notinto the intracellular compartments (Prendergast et al.,1981). All Pluronics were used at the concentration thatcaused the most efficient inhibition of the Pgp efflux systemin BBMECs.

Similar to the results obtained from the DPH polarizationmeasurements, two distinct effects of block copolymers onTMA-DPH polarization in BBMECs were observed: theplasma membrane solidification caused by the Pluronics F88and L35 (groups I and III a) and the plasma membranefluidization caused by the block copolymers P85 and L121(groups II and III b) (Fig. 5A). The changes in the plasmamembrane microviscosity in Pluronic-treated BBMECs oc-curred faster (within the first 3–10 min after addition of theblock copolymer to the cell suspension) (Fig. 5A) than thechanges in the total membrane microviscosity (Fig. 3A). Theshorter time frame with TMA-DPH suggests that the inser-tion of Pluronics into the plasma membrane and the effect onthe mobility of plasma membrane lipids occurs more quicklythan transport of Pluronics into the cells and the resultingeffects on total membrane microviscosity.

The plasma membrane microviscosity factor for each Plu-ronic composition was calculated, and the resulting datawere plotted as a function of length of PO block for eachPluronic (Fig. 5B). An S-shaped curve corresponding to thelipophilic Pluronics and a linear dependence correspondingto the hydrophilic Pluronics were observed. It is noteworthy,that the S-shaped dependence of plasma membrane micro-viscosity factor (Fig. 5B) is analogous to the S-shaped depen-dence of the Pgp ATPase activity factor (Fig. 4B).

Effects of Pluronics on the Intracellular ATP Contentin BBMECs. The effects of various Pluronic compositions onintracellular ATP content were measured by luciferin-lucif-erase assay (Garewal et al., 1986). The concentration-depen-dent effect of four different Pluronic compositions from eachmajor group on cellular ATP is shown in Fig. 6A. The BBMEC

Fig. 4. A, effect of various Pluronics on the Pgp ATPase activity in humanPgp-expressing membranes. B, relationship between the Pgp ATPaseactivity factor and the length of lipophilic PO segment (NPO) in Pluronicblock copolymers. Membranes were exposed to various Pluronic solutionsat the most effective concentrations (in respect to the inhibition of Pgpefflux system), and Pgp ATPase activity was calculated as describedunder Materials and Methods.


at ASPE



ownloaded from


monolayers treated with the hydrophilic Pluronic F88 (group I)and lipophilic block copolymer with a short lipophilic PO blockL35 (group III a) increased intracellular ATP content. In con-trast, incubation of the cell monolayers with the lipophilic Plu-ronic L121 (group III b) caused substantial energy depletion inBBMECs. Taking into account that hydrophilic Pluronics cor-responding to group I and III a caused a significant membranesolidification (Fig. 3A), whereas the lipophilic Pluronics (groupsII and III b) caused membrane fluidization, it suggests therelationship between the status of the membranes and theintracellular ATP level in BBMECs. Generally, Pluronics thatincrease the membrane microviscosity elevate the ATP content,and vise versa, block copolymers that decrease membrane mi-croviscosity case energy depletion in the blood-brain barriercells. The reason of this phenomenon is unknown. IntermediatePluronic P85 (group II) reduced the intracellular ATP level,

identically to lipophilic L121, but at significantly higher extent(less than 10% of Pluronic nontreated control cells). This resultis consistent with our previous observation showing the mostefficient membrane fluidization by the lipophilic block copoly-mers with the intermediate length of PO block.

Finally, the ATP depletion factor (ATP intracellular levelsin the absence of Pluronic versus those observed in the pres-ence of Pluronic) was calculated and plotted versus thelength of PO block for the each copolymer (Fig. 6B). There aretwo separate dependences: a bell-shaped curve correspondingto the lipophilic Pluronics and a linear dependence corre-sponding to the hydrophilic block copolymers. The lipophilicPluronics with the intermediate length of PO block causedthe most significant energy depletion in BBMECs.

Confocal Microscopy Studies of FITC-labeled PluronicsTransport into BBMEC Pluronic F108, L35, P85, and L121were labeled with FITC, as described previously (Beauchamp

Fig. 5. A, kinetic effects of various Pluronics on the plasma microviscosityin BBMECs. B, relationship between the plasma microviscosity factorand the length of lipophilic PO segment (NPO) in Pluronic block copoly-mers. Cells were treated with F88 (filled diamonds), L121 (filled trian-gles), P85 (filled circles), and L35 (crosses) at 37°C. Copolymers were usedat the concentrations that caused the most efficient inhibition of the Pgpefflux system in BBMECs.

Fig. 6. A, concentration effects of various Pluronics on the ATP intracel-lular content in BBMECs. B, relationship between the ATP depletionfactor and the length of lipophilic PO segment (NPO) in Pluronic blockcopolymers. Cells were treated with F88 (filled diamonds), L121 (filledtriangles), P85 (filled circles), and L35 (crosses) for 2 h at 37°C. After that,the ATP intracellular levels were calculated as described under Materialsand Methods.


at ASPE



ownloaded from


et al., 1983) and used to determine the cellular distribution inBBMEC monolayers. Figure 7 shows the confocal fluores-cence photomicrographs of BBMECs after a 2-h (37°C) expo-sure to FITC-labeled Pluronics. As is seen in the figure, theFITC-labeled hydrophilic Pluronic F108 (group I) shows apoor internalization into the cells, with most of the internal-ized Pluronic confined to what is presumed to be endocyticcompartments (Fig. 7A). In contrast, lipophilic Pluronic L35with short PO block (group III a) and intermediate PluronicP85 (group II) accumulated throughout the cells, includingthe cytoplasm, the cellular organelles, and to some extent thenuclei (Fig. 7, B and C). The cellular accumulation of highlylipophilic Pluronic L121 with the long PO block (group III b)was dramatically different from that observed with PluronicsL35 and P85 (Fig. 7D). The fluorescent microphotographshows L121 localized presumably in the endocytic compart-ments. This suggests that the highly lipophilic Pluronic L121could not cross out the BBMEC membranes, probably, due toits strong interaction with the lipid bilayer. Thus, L121 ef-fectively decreases the microviscosity of the plasma mem-branes (Fig. 5B), but not the intracellular membranes. Thesedata provide an explanation for the effects of various Pluron-ics on the total membrane microviscosity in BBMEC. Be-cause DPH polarization reflects changes in the microenviron-ment of all membranes, the total effect of L121 on the netmicroviscosity in BBMECs is less than the effect of interme-diate Pluronics (group II), which explains the bell-shapeddependence for the total microviscosity on the length of POblock (Fig. 3A).

DiscussionThe rationale for this work was to examine how Pluronic

block copolymers, with different molecular structure, inter-

act with the BBB cells and to find those Pluronic composi-tions having the maximal inhibitory effects on Pgp effluxtransport system in BBMECs. Data from 12 Pluronic copol-ymer compositions were examined. Using lipophilicity as adescriptive index, the 12 polymer compositions were subdi-vided into four groups [hydrophilic copolymers (HLB 20)(I); lipophilic copolymers (HLB �20) with intermediatelength of PO block ranging from ca. 30 to 60 PO repeatingunits (II); and lipophilic copolymers (HLB �20) with shorter(III a) and with longer PO blocks (III b)] and examined fortheir ability to alter the membrane microviscosity, Pgp AT-Pase activity, and ATP intracellular levels. These parame-ters were selected based on previous studies showing that theeffects of Pluronic P85 on Pgp activity are correlated withmembrane microviscosity, Pgp ATPase activity, and ATPdepletion in BBB cells.

Hydrophilic block copolymers F38, F88, F108, and F127(group I) showed no or little inhibition of the Pgp effluxsystem in BBMEC monolayers. Confocal microphotographsshowed a poor cellular internalization of hydrophilic Pluron-ics, with intracellular accumulation mainly restricted pre-sumably to endocytic compartments. Molecules of these blockcopolymers adhere to the surface plasma membrane of thecells and limit the lateral mobility of membrane lipids, caus-ing membrane solidification. The increased membrane micro-viscosity could be a reason for the elevated Pgp ATPaseactivity observed with these block copolymers. It has beenfound recently that membrane fluidization by various agents,including nonionic surfactants, abolishes Pgp ATPase activ-ity (Regev et al., 1999). Moreover, a mutational analysis ofPgp showed that interaction between the two ATP bindingsites in the efflux protein is essential for the ATP hydrolysis(Ambudkar et al., 1999). Therefore, we suggest that mem-brane solidification caused by hydrophilic Pluronics may in-crease interaction between the main functional domains ofPgp and enhance the Pgp ATPase activity.

Hydrophilic Pluronic compositions also increase intracel-lular ATP levels in BBMECs. The mechanism of this effectremains unclear and needs further investigation. Taken to-gether, Pluronics in group I have an extended hydrophilicethylene oxide block, do not incorporate into lipid bilayers,undergo limited transport into the cells, and as a result, havenegligible effect on Pgp efflux activity in BBMECs (Fig. 8).

Pluronic compositions in group II consist of lipophilic co-polymers with intermediate length of PO block (PluronicsL64, P85, L81, and P105). Members of this group rapidlyadhered to the cell membranes and incorporated into them,resulting in an increased fluidization. This is also supportedby the confocal microphotographs demonstrating that groupII Pluronics spread throughout the cells into the cytoplasm,cellular organelles, and even some extend into the nuclei.Our recent studies indicated colocalization of Pluronic P85with a mitochondrial marker in BBMEC monolayers (Batra-kova et al., 2001a), indicating possible interactions of P85and mitochondria membranes. The membrane distribution ofP85 and related polymers has a 2-fold effect; causing 1) adecrease of Pgp ATPase activity due to changes in the lipidmicroenvironment of Pgp, and 2) an inhibition of ATP syn-thesis due to changes in the electron transport in the mito-chondria membranes. The evidence supporting the latter isthat intermediate Pluronic P105 decreased the activity of theelectron transport chains in the mitochondria from HL-60

Fig. 7. Intracellular localization of FITC-F108 (A), FITC-P85 (B), FITC-L35 (C), and FITC-L121 (D) in BBMEC monolayers by confocal micros-copy. Cells were exposed to various FITC-labeled Pluronics for 2 h,washed with BSA/PBS solution, and examined by fluorescent confocalmicroscopy.


at ASPE



ownloaded from


cells (Rapoport et al., 200). Finally, the reduced ability of Pgpto consume intracellular ATP, along with energy depletion,leads to a dramatic inhibition of Pgp efflux transport system,facilitating drug transport to the brain (Batrakova et al.,2001a) (Fig. 8). There may be other mechanisms by whichPluronics increase drug transport into the cells. For example,it was shown recently that intermediate Pluronics enhancedthe transport of doxorubicin by accelerating the processes ofsolute diffusion within lipid bilayers (Erukova et al., 2000).However, the inhibition of efflux transport systems is be-lieved to be the most important in the facilitation of drugtransport across the BBB.

Lipophilic copolymers with short PO blocks Pluronics L35and L43 (group III a) could be placed between the hydrophilicPluronics (group I) and the intermediate lipophilic Pluronics(group II) with respect to their effect on Pgp activity inBBMEC monolayers. Similar to the hydrophilic block copol-ymers, molecules of these Pluronics adhere on the surfacemembranes of BBMECs causing the membrane solidificationand increasing Pgp ATPase activity. However, in contrast tothe hydrophilic block copolymers and similar to the interme-diate lipophilic Pluronics, they easily spread throughout thecells into cytoplasm and reach intracellular compartments,including nuclei. In spite of their effective transport into theBBMECs, they practically do not affect ATP content in thecells. The lack of effect on ATP intracellular levels is likelydue to absence of membrane fluidization, particularly, inmitochondria membranes.

Finally, extremely lipophilic copolymers with long POblocks Pluronics L121 and L101 (group III b) are the mostmembranotropic block copolymers. They cause the highestfluidization effect on plasma membranes and the most effi-cient inhibition of Pgp ATPase activity in Pgp-containingmembranes. Because of such high membranotropic proper-ties these block copolymers anchor in the plasma membranesand remain there for an extended period of time. As a result,they are less efficiently transported into the intracellular

compartments than intermediate Pluronics (group II) (Fig.8). Therefore, the extremely lipophilic Pluronics cause lessenergy depletion and, consequently, have less effect on Pgpefflux system in blood-brain barrier cells than the interme-diate block copolymers. An additional consideration with thevery lipophilic Pluronic compositions is the low CMC. It hasbeen shown previously that the effect of Pluronics is medi-ated by the copolymer single chain unimers, rather than bythe micelles (Miller et al., 1997). Extremely lipophilic Plu-ronics tend to form micelles at low concentrations of thecopolymer in water solutions. Thus, the micelle formationdecreases the ability of Pluronic molecules to enter the cellsand reduces the influence of the copolymer on all systems inthe barrier cells.

All in all, a delicate balance between hydrophilic and li-pophilic components in the Pluronic molecule should be ac-complished to provide the best interactions and the mostsignificant impact of the block copolymer on the endothelialcell transport.

Fig. 8. Scheme describing interactions between the blood-brain barriercells and four major groups of Pluronics.

Fig. 9. Relationship between the R123 enhancement factor and totalmicroviscosity factor (A) or ATP depletion factor (B) for different Pluron-ics.


at ASPE



ownloaded from


Present studies strongly indicate the key roles of mem-brane fluidization and energy depletion caused by Pluronicson the inhibition of the Pgp efflux system in BBMECs. Tomake this statement clear the dependence of R123 accumu-lation factor versus total microviscosity factor (Fig. 9A) andR123 accumulation factor versus ATP depletion factor (Fig.9B) were plotted for all studied block copolymers. As is seenin the figure, both dependences are linear. This supports theimportance of membrane fluidization and energy depletion,in the effects of Pluronics on Pgp efflux transport activity inBBMECs. Overall, Pluronics with the intermediate hydro-philic-lipophilic properties are believed to have a remarkablepotential use for the delivery of therapeutic agents to thebrain.

Acknowledgments

We thank Janice Taylor (Confocal Laser Scanning MicroscopeCore Facility, University of Nebraska Medical Center), which issupported by the Nebraska research Initiative, for providing assis-tance with confocal microscopy.

ReferencesAlakhov VY and Kabanov AV (1998) Block copolymeric biotransport carriers as

versative vehicles for drug delivery. Expert Opin Investig Drugs 7:1453–1473.Alexandridis P, Holzwarth JF, and Hatton TA (1994) Micellization of poly(ethylene

oxide)-poly(propylene oxide)-poly(ethylene oxide) triblock copolymers in aqueoussolutions: thermodynamics of copolymer association. Macromolecules 27:2414–2425.

Ambudkar SV, Dey S, Hrycyna CA, Ramachandra M, Pastan I, and Gottesman MM(1999) Biochemical, cellular and pharmacological aspects of the multidrug trans-porter. Annu Rev Pharmacol Toxicol 39:361–398.

Batrakova EV, Han HY, Miller DW, and Kabanov AV (1998) Effects of Pluronic P85unimers and micelles on drug permeability in polarized BBMEC and Caco-2 cells.Pharm Res 15:1525–1532.

Batrakova EV, Li S, Miller DW, and Kabanov AV (1999) Pluronic P85 increasespermeability of a broad spectrum of drugs in polarized BBMEC and Caco-2 cellmonolayers. Pharm Res 16:1368–1374.

Batrakova EV, Li S, Vinogradov SV, Alakhov VY, Miller DW, and Kabanov AV(2001a) Mechanism of pluronic effect on P-glycoprotein efflux system in bloodbrain barrier: contributions of energy depletion and membrane fluidization.J Pharmacol Exp Ther 299:483–493.

Batrakova EV, Miller DW, Li S, Alakhov VY, Kabanov AV, and Elmquist WF (2001b)Pluronic P85 enhances the delivery of digoxin to the brain: in vitro and in vivostudies. J Pharmacol Exp Ther 296:556–562.

Beauchamp CO, Gonias SL, Menapace DP, and Pizzo SV (1983) A new procedure forthe synthesis of polyethylene glycol-protein adducts; effects on function, receptorrecognition, and clearance of superoxide dismutase, lactoferrin and alpha 2-mac-roglobulin. Anal Biochem 131:25–33.

Chazotte B (1994) Comparisons of the relative effects of polyhydroxyl compounds onlocal versus long-range motions in the mitochondrial inner membrane. Fluores-cence recovery after photobleaching, fluorescence lifetime and fluorescence anisot-ropy studies. Biochim Biophys Acta 1194:315–328.

Druekes P, Schinzel R, and Palm D (1995) Photometric microtiter assay of inorganicphosphate in the presence of acid-labile organic phosphates. Anal Biochem 230:173–177.

Erukova V, Yu, Krylova OO, Antonenko, Yu N, and Melik-Nubarov NS (2000) Effect

of ethylene oxide and propylene oxide block copolymers on the permeability ofbilayer lipid membranes to small solutes including doxorubicin. Biochim BiophysActa 1468:73–86.

Ferrari M, Fornasiero MC, and Isetta AM (1990) MTT calorimetric assay for testingmacrophage cytotoxic activity in vitro. J Immunol Methods 131:165–172.

Fontaine M, Elmquist WF, and Miller DW (1996) Use of rhodamine 123 to examinethe functional activity of P-glycoprotein in primary cultured brain microvesselendothelial cell monolayers. Life Sci 5:1521–1531.

Garewal HS, Ahmenn FR, Shifman RB, and Celniker A (1986) ATP assay: ability todistinguish cytostatic from cytocidal anticancer drug effects. J Natl Cancer Inst77:1039–1045.

Jancis EM, Carbone R, Loechner KJ, and Dannies PS (1993) Estradiol induction ofrhodamine 123 efflux and the multidrug resistance pump in rat pituitary tumorcells. Estradiol induction of rhodamine 123 efflux and multidrug resistance pumpin rat pituitary tumor cells. Mol Pharmacol 43:51–56.

Kabanov AV, Chekhonin VP, Alakhov VY, Batrakova EV, Labedev AS, Melik-Nubarov NS, Arzhakov SA, Lavashov AV, Morozov GV, Severin ES, et al. (1989)The neuroleptic activity of haloperidol increases after its solubilization in surfac-tant micelles. FEBS Lett 258:343–345.

Kwon GS and Okano T (1999) Soluble self-assembled block copolymers for drugdelivery. Pharm Res 16:597–600.

Laat SW, Saag PT, and Shinitzky M (1977) Microviscosity modulation during the cellcycle of neuroblastoma cells. Proc Natl Acad Sci USA 10:4458–4461.

Lee JS, Paull K, Alvarez M, Hose C, Monks A, Grever M, Fojo AT, and Bates SE(1994) Rhodamine efflux patterns predict P-glycoprotein substrates in the nationalcancer institute drug screen. Rhodamine efflux patterns predict P-glycoproteinsubstrates in the National Cancer Institute drug screen. Mol Pharmacol 46:627–638.

Lemieux P, Guerin N, Paradis G, Proulx R, Chistyakova L, Kabanov A, and AlakhovV (2000) A combination of poloxamers increases gene expression of plasmid DNAin skeletal muscle Gene Ther 7:986–991.

Miller DW, Audus KL, and Borchardt RT (1992) Application of cultured bovine brainendothelial cells of the brain microvasiculature in the study of the blood-brainbarrier. J Tissue Cult Methods 14:217–224.

Miller DW, Batrakova EV, Waltner TO, Alakhov VY, and Kabanov AV (1997)Interactions of Pluronic block copolymers with brain microvessel endothelial cells:evidence of two potential pathways for drug absorption. Bioconj Chem 8:649–657.

Miller DW, Fontain M, Kolar C, and Lawson T (1996) The expression of multidrugresistance-associated protein (MRP) in pancreatic adenocarcinoma cell lines. Can-cer Lett 107:301–306.

Pagano R, Ozato K, and Ruysschaert J-M (1977) Intracellular distribution of li-pophilic fluorescent probes in mammalian cells. Biochem-Biophys Acta 465:661–666.

Prendergast FG, Haugland RP, and Callahan PJ (1981) 1-[4-(Trimethylamino)phe-nyl]-6-phenylhexa-1,3,5-triene: synthesis, fluorescence properties and use as afluorescence probe of lipid bilayers. Biochemistry 20:7333–7338.

Regev R, Assaraf YG, and Eytan GD (1999) Membrane fluidization by ether, otheranesthetics and certain agents abolishes P-glycoprotein ATPase activity and mod-ulates efflux from multidrug-resistant cells. Eur J Biochem 259:18–24.

Shepard RL, Winter MA, Hsaio SC, Pearce HL, Beck WT, and Dantzig AH (1998)Effect of Modulators on the ATPase activity and vanadate nucleotide trapping ofhuman P-glycoprotein. Biochem Pharmacol 56:719–727.

Shinizky M and Inbar M (1967) Microviscosity parameters and protein mobility inbiological membranes. Biochim Biophys Acta 433:133–149.

Venne A, Li S, Mandeville R, Kabanov AV, and Alakhov VY (1996) Hypersensitizingeffect of Pluronic L61 on cytotoxic activity, transport and subcellular distributionof doxorubicin in multiple drug-resistant cells. Cancer Res 56:3626–3629.

Yokoyama M (1992) Block copolymers as drug carriers. Crit Rev Ther Drug CarrierSyst 9:213–248.

Address correspondence to: Dr. Alexander V. Kabanov, Department of Phar-maceutical Sciences, University of Nebraska Medical Center, 986025 NebraskaMedical Center, Omaha, NE 68198-6025. E-mail: [email protected]


at ASPE



ownloaded from


optimal structure requirements for pluronic block...

Documents