Pharmaceutical Research, Vol. 16, No. 10, 1999
`
`Research Paper
`
`Inhibition of P-Glycoprotein by D-a-
`Tocophery! Polyethylene Glycol
`1000 Succinate (TPGS)
`
`Jay M. Dintaman! and Jeffrey A. Silverman!”
`
`Received March 5, 1999; accepted June 30, 1999
`
`Purpose. To investigate whether d-ca-tocopheryl polyethylene glycol
`1000 succinate (TPGS) functions as an inhibitor of P-glycoprotein (P-
`gp), the multidrug resistance transporter.
`Methods. Two assays were used to measure the function of TPGS on
`P-gp function. First, we examined the ability of TPGS to modulate
`the cytotoxicity of established, cytotoxic, P-glycoprotein substrates.
`Parental NIH 3T3 cells and NIH 3T3cells transfected with the human
`MDR1 cDNA (G185) were exposed to doxorubicin, paclitaxel, colchi-
`cine, vinblastine and 5-fluorouracil (SFU) in the presence or absence
`of TPGS. Cytotoxicity was assessed with the MTT assay. Second,
`polarized transport of the P-gp substrates rhodamine 123 (R123), pacli-
`taxel and vinblastine was measured using the humanintestinal HCT-
`8 and Caco-2 cell
`lines grown in Transwell dishes. Drug flux was
`measuredbyliquid scintillation counting or fluorescence spectroscopy
`of the media.
`Results. G185 cells were 27-135 fold moreresistant to the cytotoxic
`drugs doxorubicin,vinblastine, colchicine and paclitaxelthan the paren-
`tal NIH 373 cells. In contrast SFU, which is not a P-gp substrate, is
`equally cytotoxic to parental and G185 cells. Co-administration of
`TPGSenhanced the cytotoxicity of doxorubicin,vinblastine, paclitaxel,
`and colchicine in the G185 cells to levels comparable to the parental
`cells. TPGS did not increase the cytotoxicity of 5FU in the G185 cells.
`Using a polarized epithelial cell transport assay, TPGS blocked P-gp
`mediated transport of R123 andpaclitaxel in a dose responsive manner.
`Conclusions. These data demonstrate that TPGS acts as a reversal
`agent for P-glycoprotein mediated multidrug resistance and inhibits P-
`gp mediated drug transport. These results suggest that enhanced oral
`bioavailability of drugs co-administered with TPGS may,in part, be
`due to inhibition of P-glycoprotein in the intestine.
`KEY WORDS:P-glycoprotein; TPGS; drug transport; bioavailability.
`
`INTRODUCTION
`
`The multidrug transporter, P-glycoprotein (P-gp), is a 170
`kDa membraneprotein which functions as an ATP-dependent
`drug efflux pump. One activity of this protein is to lower
`the intracellular concentration of drugs thereby reducing the
`cytotoxic activity of anticancer drugs. Increased expression of
`this protein has been observed in human tumors andis often
`associated with failure of chemotherapy due to drug resistance
`(1-5). P-gp removes a large number of chemically unrelated
`drugs extending over many therapeutic indications such as anti-
`cancer drugs, steroids, antihistamines, antibiotics, calcium
`channel blockers and anti-HIV peptidomimetics (2,4,5).
`
`
`
`‘ Division of Drug Transport, AvMax,Inc. Berkeley, California 94710.
`7To whom correspondence
`should
`be
`addressed.
`(e-mail:
`jeffrey.silverman@ att.net)
`ABBREVIATIONS: P-gp, P-glycoprotein; R 123, Rhodamine 123;
`CsA, Cyclosporine A; TPGS, d-c-tocopheryl polyethylene glycol
`1000 succinate.
`
`The P-gp drug transporter is encoded by one gene, MDR1,
`in humans whereas in rodents two genes, mdria and mdrib
`encode highly similar drug transporters (6,7). P-gp is primarily
`expressed on the luminalsurface ofepithelial cells from several
`tissues including the intestine, liver, kidney, and the endothelial
`cells comprising the blood-brain and blood-testes barriers (8—
`10). The ability of this protein to export toxic compounds
`combined with this localization led to the hypothesis that a
`physiological function of the MDR1 encoded P-gp may be as a
`protective barrier or export mechanism for xenobiotics. Indeed,
`recent investigations with knockout mice in which the mdrla
`gene wasdisrupted have confirmed such a protective role for
`P-gp (11-14). Exposure of mdr/a deficient mice to vinblastine
`or ivermectin results in significantly higher tissue and plasma
`levels compared to wild-type animals. Moreover, these com-
`pounds are toxic in the knockout mice at doses which are
`innocuous to heterozygous and wild-type mice. These experi-
`ments further suggesteda role for P-gp in the blood brain barrier
`since the ivermectin accumulated in the brain of the mdrla
`deficient animals but not animals with an intact mdrla gene.
`The knockout mice displayed ivermectin toxicity at doses 50
`to 100 fold less than wild-type mice.
`Additional data have supported a role of P-gp in the intes-
`tine as both a barrier to absorption as well as a mechanism of
`disposition of drugs such as vinblastine, etoposide, paclitaxel
`and digoxin. For example, Su and Huang observed that inhibi-
`tion of P-gp increased bioavailability of digoxin by increasing
`absorption as well as reducing excretion (15). A similar phe-
`nomenonwas observed with etoposide (16). P-glycoprotein has
`recently been suggested to becritical in oral drug absorption
`(17-19). In concert with the drug metabolizing enzyme CYP3A,
`P-gp maylimit oral drug bioavailability in the gut by controlling
`drugtransport from the intestinal lumen and byaffecting access
`to CYP3A (19).
`Vitamin E TPGS,d-a-tocopheryl polyethylene glycol 1000
`succinate,is a derivative of vitamin E consisting of a hydrophilic
`polar head group (tocopherol succinate) and a lipophilic alkyl
`tail (polyethylene glycol) resulting in amphiphillic properties
`(Eastman Kodak,
`technical bulletin EFC-226). TPGS has a
`relatively low critical micelle concentration, 0.02 wt%, and acts
`to solubilize lipophilic compounds. Bordreaux et al. reported
`a two-fold increase in cyclosporine CsA area underthe plasma
`concentration-time-curve (AUC) when co-administered with
`LiquiE, a glycerol and water solution of TPGS (20). Sokol et
`al. similarly observed increases up to 71% in CsA AUC in
`subjects who received concomitant TPGS(21). Both Sokol and
`Bordreaux suggested that the increased drug absorption was
`due to enhanced micelle formation, resulting in improved CsA
`solubilization. Chang et al.
`later reported a 61% increase in
`CsA AUC when dosed with 20-25% of the TPGS previously
`used in the Sokol or Bordreaux studies (22). Chang etal. also
`suggested that TPGS may interact with P-gp in the intestine to
`increase CsA absorption.
`In the current investigation we examine the effect of TPGS
`on P-gp mediated drug resistance and transport of established
`P-gp substrates. If this agent functions as a P-gp reversal agent
`then perhapsits effect on drug absorptionis, in part, mediated
`by inhibition of active drug efflux in the intestine. Our data
`show TPGSto be an effective inhibitor of P-gp mediated drug
`
`0724-874 1/99/ 1000- 1550$16.00/0 © 1999 Plenum Publishing Corporation
`
`1550
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0001
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0001
`
`

`

`TPGSInhibition of P-Glycoprotein
`
`resistance and transport at concentrations well below the
`reported critical micelle concentration and suggests that its
`reversal activity is due to an effect on transport activity.
`
`MATERIALS AND METHODS
`
`Cell Culture
`
`The NIH3T3 Swiss mouse embryocell line was obtained
`from American Type Culture Collection (ATCC, Rockville,
`MD)and was grown in Dulbecco’s Modified Eagles Medium
`(Biowhittaker, Walkersville, MD) supplemented with 4.5 g/L
`glucose, 10% fetal bovine serum (Hyclone Laboratories, Logan,
`UT), 2 mM L-glutamine (Advanced Biotechnologies Incorpo-
`rated (ABI), Columbia, MD), and 0.01 mg/ml gentamicin (ABI).
`The drug resistant, NIH-MDR-G185, cell line, derived by trans-
`fection of the human MDRI gene into NIH3T3 cells (23),
`was obtained from M. M. Gottesman (NCI, NIH) and was
`maintained in similar medium supplemented with 60 ng/ml of
`colchicine (Sigma, St. Louis, MO). HCT-8 cells (ATCC), iso-
`lated from a human ileocecal adenocarcinoma cell line, were
`grown in RPMI-1640 medium (Biowhittaker) supplemented
`with 10% horse serum (Biowhittaker), 1 mM sodium pyruvate
`(Gibco BRL, Grand Island NY) and 0.01 mg/m! gentamicin.
`Caco-2 cells (ATCC), derived from a human colonic adenocarci-
`noma, were grown in Eagle’s MEM (Biowhittaker) supple-
`mented with 10% fetal bovine serum, and 0.01 mg/ml
`gentamicin. All cells were maintained in a humidified atmo-
`sphere with 5% CO, at 37°C.
`
`Cytotoxicity Assay
`
`Cells were plated at a density of 2.5-3.0 < 10° cells/well
`in 96-well microtiter plates (PGC, Gaithersburg, MD) and were
`exposed to 1-5000 nM of doxorubicin, vinblastine, colchicine,
`paclitaxel, 0.1-25 nM 5-fluorouracil (Sigma) and 0.00 1—.005%
`TPGS(Eastman, Kingsport, TN) for 72 hours. To ensure solubi-
`lization of the TPGS, a 1% solution of TPGSin ethanol was
`prepared fresh for each experiment and diluted further in cell
`culture medium to the indicated concentrations. Cell viability
`wasdetermined with the colorimetric MTT (3-(4,5-dimethy|Ithi-
`azol-2-yl)-2,5-dipheny] tetrazolium, Sigma) assay as previously
`described (24,25) and the absorbance was measured with a
`Dynex MRX Microplate Reader (Chantilly, VA) at 570 nm.
`This assay is based on the reduction of MTT by mitochondria
`in viable cells to water insoluble formazan. The data presented
`are the mean SD ofat least 3 independent experiments, each
`performed in quadruplicate.
`
`Rhodamine 123 Transport
`
`Rhodamine 123 (R123; Sigma) transport was examined
`as previously described (26,27) using both HCT-8 and Caco-
`2 cells. Briefly, cells were grown in 6 well Corning Transwell
`dishes (HCT-8) or collagen coated Transwell dishes (Caco-2)
`until a tight monolayer was formed as measured by transepithe-
`lial electrical resistance or lucifer yellow impermeability. The
`integrity of the monolayers following the transport experiments
`was similarly evaluated. Typical TEER values were > 300
`Ohms/cm?. R123 was added ata final concentration of 13 4M
`to the basal or apical compartments and 200 wl samples were
`taken at the indicated times from the opposite chamber. TPGS
`
`1551
`
`was addedas an inhibitor to both compartments. Fluorescence
`of R123 in the media samples was measured using a Biotek
`FL500 Fluorescence Plate Reader (Winooski, VT) with an exci-
`tation wavelength of 485 nm and an emission wavelength of
`530 nm. All experiments were performedin triplicate; the data
`presented are the mean +SD andare representative of multi-
`ple experiments.
`
`Paclitaxel, Vinblastine and Cyclosporine Transport
`
`Inhibition of (H] paclitaxel (Moravek Biochemical, Brea,
`CA), PH] vinblastine (Amersham, Arlington Heights, IL), and
`[7H] cyclosporine (CsA; Amersham)efflux by TPGS was exam-
`ined in a manner similar to R123. The transported drug, 0.1
`wM (0.25 pCi/ml), was added to either the basal or apical
`compartment and 200 yl aliquots were taken at the indicated
`times from the opposite chamber. Radioactivity was measured
`by liquid scintillation counting.
`
`Western Blot Analysis
`
`Western blot analysis was performed as previously
`described 28. Briefly, crude cell membranes were isolated by
`lysing the cells in 10 mM Tris-HCl, pH 7.5; 10 mM NaCl;
`1
`mM MgCl, supplemented with pepstatin (1.5 wg/ml), leupeptin
`(1.5 jxg/ml) and 0.2 mM pefabloc. Cells were homogenized
`with 20 strokes of Dounce “B” (tight) pestle (Wheaton, Mill-
`ville, NJ), nuclei and cell debris were removed by centrifugation
`for 10 minutes at 400 X g. The supernatants were then centri-
`fuged at 100,000 X g for 30 minutes at 4°C andthepellets
`were resuspended in lysis buffer and stored at —80°C. 20 wg
`samples were fractionated in 8% polyacrylamide-SDSgel and
`transferred to 0.45 jxm nitrocellulose membrane. The mem-
`branes were blocked in PBS-T (0.1% Tween-20 in PBS) con-
`taining 5% skim milk for | hour and then probed with 1 wg/
`ml of C219 antibody (Signet Laboratories, Dedham, MA)in
`PBS overnight. The membranes were visualized by enhanced
`chemiluminescence according to the manufacturer’s instruc-
`tions (Pierce, Rockford, IL).
`
`RESULTS
`
`Western Blot Analysis
`
`Wefirst measured the relative levels of P-gp expression
`in the NIH3T3 and G185 cell lines by western blot analysis
`using the C219 antibody, which recognizes all P-gp isoforms
`(29). Consistent with previous data, high P-gp expression was
`observed in the G185 cells relative to that in the parental NIH
`3T3 cells (Fig. 1). We also examined P-gp expression in two
`humanintestinal carcinomacell lines, Caco-2 and HCT-8 which
`have been previously used for investigation of drug transport
`and to have polarized expression of P-gp (26). We observed
`that each of these intestinal cell
`lines have moderate P-gp
`expression albeit lower than the G185 cells (Fig. 1).
`
`Cytotoxicity Experiments
`
`The interaction of TPGS with P-gp wasinitially examined
`with cytotoxicity assays using parental NIH3T3 and MDRI
`transfected G185 cells to cytotoxic anticancer drugs. Consistent
`with previous reports (23,30), G185 cells were more resistant
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0002
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0002
`
`

`

`1552
`
`Dintaman and Silverman
`
` iene
`
`_ .
`Fig. 1. Western blot analysis of P-glycoprotein expression. Twenty
`microgramsof total cell membrane proteins were separated by SDS-
`PAGE,transferred to PVDFfilters which were subsequently probed
`with the C219 antibody and visualized using chemiluminescence as
`described in Materials and Methods. Lane 1, NIH-3T3 G185; lane 2,
`NIH-3T3; lane 3, HCT-8; lane 4, Caco-2.
`
`Table 1. ECsg in NIH3T3 and NIH3T3-G185 Cells
`NIH3T3- NIH37T3-
`G185
`G185
`(1 uM (SpM_ NIH3T3-G185
`NIH3T3-
`NIH3T3 Gi85 Verapamil) CsA)
`(.0025% TPGS)
`
`35
`35
`35
`950
`35
`Doxorubicin
`40
`6
`20
`270
`2
`Vinblastine
`1070
`40
`100
`>5000
`60
`Paclitaxel
`45
`100
`ND
`1000
`30
`Colchicine
`
`Note: NIH 3T3 and G185cells were treated with 0-5000 nM doxorubi-
`
`cin, vinblastine, paclitaxel or colchicine in the absence or presence of
`1 wM verapamil, 5 4M CsA or 0.0025% TPGS. The concentration of
`the drug that reduces cell viability by 50% (ECs)) was determined
`using the MTT cytotoxicity assay as described in Materials and Meth-
`ods. Each experiment was performed in quadruplicate and repeated in
`at least 3 independent experiments. ND, not determined.
`
`to doxorubicin, paclitaxel, vinblastine and colchicine compared
`to parental NIH3T3 cells (Fig. 2). ECs) values were 27 to 135
`fold higher in G185 cells relative to the parental NIH3T3cells
`(Table 1). Established P-gp reversal agents, such as cyclosporine
`A (CsA) and verapamil, reduced the resistance to doxorubicin
`cytotoxicity in G185 cells to levels comparable to parental
`NIH3T3cells (Fig. 3). The reversal effect of CsA on doxorubi-
`cin, vinblastine, taxol and colchicine mediated toxicity in paren-
`tal NIH3T3 cells was modest as previously reported. This is
`
`consistent with their low level of P-gp expression (data not
`shown, (27,30). Co-administration of CsA or verapamil caused
`a similar reversal of G185 resistance to vinblastine, paclitaxel,
`and colchicine (Table 1 and data not shown).
`The effect of TPGS on P-gp mediated drug resistance was
`investigated by treating G185 cells with doxorubicin, vinblas-
`tine, paclitaxel, and colchicine concomitantly with varying
`doses of TPGS. The presence of TPGS increased drug sensitiv-
`ity of the G185 cells to doxorubicin in a dose dependent manner
`
`
`
`
`Colchicine
`
`+ ¢- 373 Control
`—«~G185 Control
`
`Colchicine (nM)
`
`Paclitaxel
`
`- *- 373 Control
`[6185Contra
`
`1.4
`5"
`
`23aSo
`
`s>
`B=
`&a
`4
`
`a
`12
`
`2f
`
`> 2
`
`os
`=a
`3 04
`a
`
`os
`
`- -373 Control.
`—»~—G185 Control
`
`1
`
`40
`
`100
`
`+000
`
`10000
`
`Doxorubicin (nM) =
`
`3S
`
`> £
`
`&ao
`a
`
`08
`
`06
`0.4
`02
`0
`
`Doxorubicin
`
`
`Vinblastine
`_——
`
`> o- 3T3 Control
`—2—G185 Control
`
`1
`
`10
`
`100
`
`4000
`
`10000
`
`1.4
`c
`z ,
`3S os
`
`5 2
`
`os
`a3 04
`a 0.2
`
`Vinblastine (nM)
`
`1
`
`10
`
`100
`
`4000
`
`10000
`
`Fig. 2. Cytotoxicity of doxorubicin (A), colchicine (B), vinblastine (C), and taxol (D) in parental NIH-3T3 cells (circles), and
`MDRI-transfected NIH-3T3 G185 cells (squares). Cells were treated with the indicated concentrations of drugs and the viability
`was measured by the MTT assay as described in Materials and Methods. Data are expressed relative to untreated controlcells.
`Each experiment was performed in quadruplicate and the data presented represent the mean + SD offour independent experiments.
`
`Paclitaxel (nM)
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0003
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0003
`
`

`

`TPGSInhibition of P-Glycoprotein
`
`1.6
`
`e+ 373 Control
`——G185 Coatrol
`= > Gt8S SO pM CoA
`—O—GI185 5.0 uM Verap
`
`Viability 1
`Relative
`
`10
`
`400
`
`1000
`
`10000
`
`Doxorubicin (nM)
`Fig. 3. Effect of CsA and Verapamil on the cytotoxicity of doxorubicin.
`Parental NIH-3T3 (diamonds) and G185 (squares) cells were exposed
`to the indicated concentrations of doxorubicin in the absence or pres-
`ence of CsA, 5.0 41M, or verapamil, 5.0 .M.. Data are expressed
`relative to untreated control cells. Each experiment was performed in
`quadruplicate and the data presented represent the mean + SDof four
`independent experiments.
`
`(Fig. 4). Treatment ofthe drug resistant G185 cells with TPGS
`lowered the ECs) concentrations for doxorubicin, vinblastine,
`paclitaxel and colchicine (Table 1). TPGS, 0.0025%, sensitized
`the G185 cells to all four of these P-gp substrate cytotoxic
`drugs to levels comparable to the parental NIH 3T3 cells. The
`highest dose of TPGS, 0.005%, resulted in decreased viability
`of both NIH3T3 and G185 cells and is likely due to toxicity
`associated with the high concentration of TPGS. At concentra-
`tions below 0.005% TPGSitself did not affect cell viability.
`These data suggest that TPGS modulates drug resistance by
`inhibiting P-gp activity in cells which over-express
`the
`MDRI1 gene.
`
`5-Fluorouracil Cytotoxicity
`
`Treatment of parental NIH 3T3 and G185 cells with 5-
`fluorouracil (SFU), a chemotherapeutic agent not transported
`
`1553
`
`by P-gp, results in a similar level of cytotoxicity in both cell
`lines (Fig. 5A) (1). Furthermore, co-incubation of SFU with
`CsA had no effect on the cytotoxicity of 5FU in either G185
`or NIH3T3 cells (Fig. 5B). Similarly, co-incubation of TPGS
`with 5FU did not increase the cytotoxicity of 5FU in either of
`these cell lines (Fig. SC).
`
`Rhodamine 123 Transport
`
`The fluorescent dye R123, an established substrate of P-
`glycoprotein (32,33), was used to examinethe ability of TPGS
`to block P-gp mediated transport. HCT-8 and Caco-2 cells have
`previously demonstrated directional transport of established P-
`gp substrates such as vinblastine, paclitaxel, CsA and R123 in
`the basolateral to apical direction (26,34—36). Expression of P-
`gp in these cells was confirmed by western blot analysis using
`the C219 antibody (Fig. 1). R123 was transported approximately
`7 and 9 fold greater flux in the basolateral to apical direction
`in HCT-8 and Caco-2 cells, respectively (Fig. 6). Consistent
`with this transport being mediated by P-gp, R123 flux was
`inhibited approximately 80% by co-incubation with 5 uM CsA.
`Similarly, 0.001—0.0025% TPGS blocked the basolateral
`to
`apical transport of R123 in a dose responsive manner further
`suggesting that TPGS inhibits transport mediated by P-gp
`(Fig. 6).
`
`Paclitaxel Transport
`
`The ability of TPGS to inhibit P-gp was confirmed by
`measuring polarized transport of paclitaxel. [7H] Paclitaxel is
`a good substrate for P-gp with approximately 14 and 40 fold
`greater transport from the basolateral to the apical compartment
`in HCT-8 and Caco-2 cells, respectively (Fig. 7). Addition of
`5 uM CsAblocked the polarized flux of paclitaxel by 80-90%.
`Similarly, co-incubation with TPGS resulted in a dose depen-
`dent decrease in paclitaxel transport (Fig. 7). The ICs) of TPGS
`for inhibition of paclitaxel transport is approximately 0.001%
`(v/v) in HCT-8 cells and 0.005% in Caco-2 cells. Polarized
`
`
`
`
`
`
`—e- -NIH3T3
`—s— G18S
`+-@- G185+0.001% TPGS
`—*- -G185+0.0025% TPGS
`—0— G185+0.005% TPGS
`
`
`
`
`
`
`
`RelativeViability
`
`~ oe _—
`
`1
`
`10
`
`100
`
`1000
`
`10000
`
`Doxorubicin (nM)
`
`Fig. 4. TPGS reversal of P-gp mediated resistance to doxorubicin. Parental NYH-3T3 (diamonds) and
`G185 cells were exposedto the indicated concentrations of doxorubicin with 0% TPGS(squares), 0.001%
`TPGS(circles), 0.0025% TPGS(triangles) or 0.005% TPGS (open circles, ©). Data are expressed relative
`to untreated control cells. Each experiment was performed in quadruplicate andthe data presented represent
`the mean +SD of four independent experiments.
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0004
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0004
`
`

`

`Dintaman and Silverman
`
`—8— Control B:A
`—O— Control A:B
`++ de CSA BIA
`—— 0.001% TPGS B:A
`— + ~ 0.0025% TPGS B:A
`
`A Caco-2
`$
`8
`7
`
`3a
`
`£5
`
`6
`ow
`&
`Cc
`© 4
`Fe
`ao
`3
`
`Sm
`
`A
`
`“
`
`5-Flucrouracil
`
`bn G85
`
`[Te NIH373
`
`1554
`
`£a3> >c
`
`s
`xc]oe
`x
`
`
`
`
`
`
`
`2
`S14
`
`
`
`0
`5-Fluorouracil (nM)
`
`0
`
`1
`
`4
`
`5
`
`6
`
`2 3
`Time, hrs
`
`—#— Control B:A
`~—~&— Control! A:B
`— @- 0.005% TPGS B:A
`
`B Het-8
`
`:-@--0.001%TPGS B:A
`
`~#-- 0.0025%TPGS B:A
`
`Time, hrs
`Fig. 6. Rhodamine 123 transport in Caco-2 and HCT-8 cells. Caco-2
`(A) and HCT-8 (B) were grown on Transwell dishes as described in
`Materials and Methods. Rhodamine 123, 13 «1M, was added to the
`apical or basolateral compartment in the absence or presence of CsA,
`5 uM, or 0.0025, 0.005, 0.001% TPGS and media aliquots were taken
`from the opposite chamber at the indicated times. The data presented
`are the mean +SD oftriplicate wells and are representative of at least
`three independent experiments.
`
`10 wM 5Fiuorouracil with CsA
`
`ONIH3T3
`mG185
`
`
`
`
`06
`6
`
`0
`
`CsA (4M)
`
`7.5 1M 5-Fluorouracil with TPGS
`
`ae
`wGI85s
`|
`ee
`
`0.001
`
`9.0025
`wt% TPGS
`
`0.006
`
` o
`
`oo B
`OB
`
`2”
`Bos
`2 06
`£ o4
`Sos o2
`oA
`Qa
`
`09 c
`os
`07
`
`2
`3z
`>oe>
`
`=2o~
`
`Fig. 5. Cytotoxicity of 5-fluorouracil to parental NIH-3T3, and drug
`resistent NIH-3T3 G185cell lines in the presence and absence of CsA
`and TPGS. A. NIH-3T3 (diamonds) and NIH-3T3 G185 (squares)cells
`were exposedto the indicated concentrations of 5-fluorouracil. Viability
`was measured by the MTTassayas described in Materials and Methods.
`Data are expressed relative to untreated control cells. Each experiment
`was performed in quadruplicate and the data presented represent the
`mean +SD of three independent experiments. B. Parental NIH-3T3
`(open bars) and G185 (closed bars) cells were exposed to 10 4M of
`SFU with the indicated concentrations of CsA. C. Parental NIH-3T3
`(open bars) and G185 (closed bars) cells were exposed to 7.5 1M of
`SFU and the indicated concentrations of TPGS.
`
`transport of [3H] vinblastine and [7H] CsA were also inhibited
`by addition of TPGS (data not shown). These data, combined
`with the cytotoxicity and R123 transport data suggest that TPGS
`is an effective P-gp reversal agent.
`
`DISCUSSION
`
`A majoreffort has been undertaken by many laboratories
`to identify inhibitors of P-glycoprotein to increase the efficacy
`of cancer treatment and to enhance the absorption of orally
`administered drugs. The data presented here support the hypoth-
`esis that TPGS functions as one such P-gp inhibitor. TPGS
`
`increased the sensitivity of P-gp expressing cells to several
`widely used cytotoxic drugs which are well established P-gp
`substrates. TPGS also effectively blocked polarized transport
`of R123 and paclitaxel in an epithelial cel! transport assay. The
`reduction of directional transport provides strong evidence for
`TPGS functioning as an inhibitor of P-gp. Conversely, no effect
`was observed with SFU, a cytotoxic drug not associated with
`P-gp mediated drug resistance or transport. 5FU is not transpor-
`ted by the P-gp pumpthus, its cytotoxicity is unaffected by the
`addition of established P-gp inhibitors such as quinine, quini-
`dine or verapamil (31,37). In the experiments presented here
`neither TPGS nor CsA impacted the cytotoxicity of SFU in
`either the NIH 3T3 or G185cells.
`Previously it has been suggested that co-administration of
`TPGSwith CsA enhancedabsorption of the immunosuppressant
`due to micelle formation (21). Concentrations of TPGS adminis-
`tered in the current work are well below the critical micelle
`concentration, 0.02 wt% in waterat 37°C,thereforeit is unlikely
`that micelle formation is responsible for the observed effects.
`In fact, the ICsrequired to inhibit R123 and paclitaxel transport
`across HCT-8 or Caco-2 cell monolayers is 20 fold less than.
`the critical micelle concentration. Further, 0.001 wt% TPGS
`also significantly reversed the multidrug resistant phenotype of
`the NIH3T3-G185 cell line to doxorubicin, vinblastine, taxol
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0005
`
`
`
`0.00 «45.00=17.80«626006=— 6.00 7.60 «010.00 12.50 20.00 «2250 26.00
`
`
`
`
`
`
`
`
`Transported
`%R123
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0005
`
`

`

`TPGSInhibition of P-Glycoprotein
`
`1555
`
`--@--.005% TPGSBtoA
`
`1.87(aPedtawiAtwe A
`
`receiving TPGShad previously experienced chronic cholestasis
`1.6||—-t—Paclitaxel B to A
`—~@—CSABtOA
`resulting in decreased bile flow suggesting poor solubilization
`1.4
`— dh— .0025% TPGS BtoA
`of the lipophilic CsA. It was hypothesized that TPGS functioned
`as a bile substitute and solubilized the CsA through micelle
`formation, thus facilitating the absorption of the drug through
`the intestinal lumen. Similarly, Pan et al. reported a 28 and
`32% decrease in CsA daily dose when co-administered with
`Liqui-E, a water soluble form of TPGS and a 26% decrease in
`daily CsA cost (44). Using normal healthy volunteers, Chang
`et al. observed a 60% rise in CsA area under the curve (AUC)
`in subjects receiving a TPGS-CsA cocktail. Decreased oral
`clearance and volume of distribution were also observed in
`those subjects. These authors proposed that the large, amphi-
`pathic TPGS mayalso be acting as an inhibitor of P-glycopro-
`tein to enhance absorption and decrease transport back into the
`intestinal lumen. The current data support the hypothesis that
`one mechanism through which TPGS may enhance oral bio-
`availability is via inhibition of P-gp. Clearly further study on
`the effect of TPGS on oral drug delivery is required to confirm
`sucharole.
`
`Time, hrs
`
`B
`
`[—€3-—-Pacitaxal Ato B
`—t—Pacitaxel BtoA
`—-*—CsABIOA
`— A— 0025 % TPGS BtoA
`= + @ + +.005 % TPGS Bto A
`— HE - 01% TPGS BtoA
`
`
`
`35
`
`Paclitaxel,
`
`pmol
`Paclitaxel,
`pmol
`
`Time, hrs
`
`REFERENCES
`
`2.
`
`7.
`
`1. L. J. Goldstein. MDR/ gene expression in solid tumors. Eur. J.
`Cancer. 32A:1039—1050 (1996).
`J. A. Endicott and V. Ling. The biochemistry of P-glycoprotein
`mediated multidrug resistance. Ann. Rev. Biochem. 58:137-171
`(1989).
`3. M. M. Gottesman. How cancer cells evade chemotherapy: Six-
`Fig. 7. Paclitaxel transport in HCT-8 and Caco-2 cells. Cells were
`teenth Richard and Hinda Rosenthal foundation award lecture.
`grown on in Transwell dishes as described in Materials and Methods.
`Cancer Res. 53:747-754 (1993).
`(7HjPaclitaxel, 0.1 41M (0.2 4.Ci/ml) wasplacedin the basolateral (B)
`4. M. M. Gottesman and I. Pastan. The multidrug transporter, a
`or apical (A) chamber; media aliquots were takenat the indicated times
`double edged sword. J. Biol. Chem. 263:12163—12166 (1988).
`and radioactivity was measuredby liquidscintillation counting. Control
`5. M.M. Gottesman andI. Pastan. Biochemistry of multidrug resis-
`Caco-2 (A) or HCT-8 (B) cells B to A (closed squares), control A to
`tance mediated by the multidrug transporter. Ann. Rev. Biochem.
`62:385-427 (1993).
`B (open squares), 5 ~M CsA B to A (diamonds), 0.0025% TPGS B
`6. S.S. Thorgeirsson, J. A. Silverman, T. W. Gant, and P. A. Marino.
`to A (triangles), 0.005% TPGSBto A (circles).
`Multidrug resistance gene family and chemical carcinogens. Phar-
`macol. Ther. 49:283-292 (1991).
`P. Borst, A. H. Schinkel, J. J. M. Smit, E. Wagenaar, L. Van
`Deemter, A. J. Smith, W. H. M. Eijdems, F. Baas, and G. J. R.
`Zaman. Classical and novel forms of multidrug resistance and
`the physiological functions of P-glycoproteins in mammals. Phar-
`macol. Ther. 60:289-—299 (1993).
`8. F. Thiebaut, T. Tsuruo, H. Hamada, M. M. Gottesman,I. Pastan,
`and M. C. Willingham. Cellular localization of the multidrug
`resistance gene product P-glycoprotein in normal humantissues.
`Proc. Natl. Acad. Sci. USA 84:7735-7738 (1987).
`9. F. Thiebaut, T. Tsuruo, H. Hamada, M. M. Gottesman,I. Pastan,
`and M. C. Willingham. Immunohistochemicallocalization in nor-
`mal tissues of different epitopes in the multidrug transport protein
`P170: Evidencefor localization in brain capillaries and crossreac-
`tivity on one antibody with a muscle protein. J. Histochem. Cyto-
`chem. 37:159-164 (1989).
`10. B.L. Lum and M. P. Gosland. MDRexpression in normal tissues.
`Hematol. Oncol. North Am. 9:319-336 (1995).
`11. A. H. Schinkel, J. J. M. Smit, O. van Tellingen, J. H. Beijnen,
`E. Wagenaar, L. van Deemter, C. A. A. M. Mol, M. A. van der
`Valk, E. C. Robanus-Maandag,H. P. J. te Riele, A. J. M. Berns,
`and P. Borst. Disruption of the mouse mdr/a P-glycoprotein gene
`leads to a deficiency in the blood-brain barrier and to increased
`sensitivity to drugs. Cell 77:491~502 (1994).
`J. van Asperen, O. van Tellingen, A. Sparreboom, A. H. Schinkel,
`P. Borst, W. J. Nooijen, and J. H. Beijnen. Enhancedoral bioavail-
`ability of paclitaxel in mice treated with the P-glycoprotein blocker
`SDZ PSC 833. Br J. Cancer 76:1181—1183 (1997).
`J. van Asperen, A. H. Schinkel, J. H. Beijnen, W. J. Nooijen, P.
`Borst, and O. van Tellingen. Altered pharmacokinetics of vinblas-
`tine in mdrla P-glycoprotein-deficient mice. J. Natl. Cancer Inst.
`88:994—999 (1996).
`
`and colchicine,all established P-gp substrates. These data sug-
`gest that TPGS micelle formation in the intestinal lumen may
`not be the sole factor behind the increase in CsA absorption
`previously observed (20-22).
`Several other surfactants, e.g. polysorbates, Cremophor
`EL, and Solutol 15, have been observed to be inhibitors of P-
`gp (38-41). These compoundsare frequently added to pharma-
`ceutical formulations to enhance solubility. These agents may
`also function to inhibit P-gp to addto their effect of enhancing
`drug absorption. Indeed the plasma concentrations of Cremo-
`phor ELin patients administered paclitaxel, which is formulated
`with this surfactant, reach levels sufficient to inhibit P-gp in
`vitro (42). The efficacy of this drug may, in part, be due to the
`activity of the Cremophor EL. Pluronic P85 has also recently
`been observed to block P-gp mediated rhodamine 123 efflux
`in Caco-2 and bovine brain microvessel endothelial cells (43).
`These data suggested that this agent may be useful for formula-
`tions to enhance brain and oral absorption.
`TPGS has been used to enhance the bioavailability of
`CsAin liver transplant patients with the effects of significantly
`improving absorption and reducing daily drug cost. Sokol et
`al. and Boudreauxet al. reported increases in CsA absorption
`in pediatric transplant recipients treated with oral TPGS, 12.5
`IU/kg and 10 U/kg, respectively. The majority of patients
`
`12.
`
`13.
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0006
`
`Apotex v. Cellgene - IPR2023-00512
`Petitioner Apotex Exhibit 1020-0006
`
`

`

`1556
`
`Dintaman and Silverman
`
`14.
`
`A. H. Schinkel, U. Mayer, E. Wagenaar, C. A. A. M. Mol, L.
`van Deemter, J. J. M. Smit, M. A. van der Valk, A. C. Voordouw,
`H. Spits, O. van Tellingen, J. M. J. M. Zijlmans, W. E. Fibbe,
`and P. Borst. Normal viability and altered pharmacokinetics in
`mice lacking mdr -type (drug transporting) P-glycoproteins. Proc.
`Natl. Acad. Sci. USA 94:4028—4033 (1997).
`. §S.-F Su and J. -D. Huang. Inhibition of the intestinal digoxin
`absorption and exsorption by quinidine. Drug Metab. Dispos.
`24:142-147 (1996).
`B. -L. Leu and J. -ding Huang. Inhibition of intestinal P-glycopro-
`tein and effects on etoposide absorption. Cancer Chemother. Phar-
`macol. 35:432—436 (1995).
`V. J. Wacher, C. -Y. Wu, and L. Z. Benet. Overlapping substrate
`specificities and tissue distribution of cytochrome P450 3A and
`P-glycoprotein: Implications for drug delivery and activity in
`cancer chemotherapy. Mol. Carcin. 13:129-134 (1995).
`. V.J. Wacher, L. Salphati, and L. Z. Benet. Active secretion and
`enterocytic drug metabolism barriers to drug absorption. Adv.
`Drug Del. Rev. 20:99-112 (1996).
`V. J. Wacher, J. A. Silverman, Y. Zhang, and L. Z. Benet. Role
`of P-glycoprotein and cytochrome P450 3A in limiting oral
`absorption of peptides and peptidomimetics. J. Pharm. Sci.
`87:1322—1330 (1998).
`J. P. Boudreaux, D. H. Hayes, S. Mizrahi, P. Maggiore, J. Blazek,
`and D. Dick. Use of water-soluble liquid vitamin E to enhance
`cyclosporine absorption in children after liver transplant. Trans-
`plant. Proc. 25:1875 (1993).
`R. J. Sokol, K. E. Johnson, F. M. Karrer, M. R. Narkewicz, D.
`Smith, and I. Kam. Improvement of cyclosporin absorption in
`children after liver transplantation by means of water-soluble
`vitamin E. Lancet. 338:212-215 (19