ICANCER RESEARCH 53. 977-984. March I. 1993]

Functional Imaging of Multidrug-resistant P-Glycoprotein with anOrganotechnetium Complex1

David Piwnica-Worms,2 Mary L. Chiù, Mark Budding, James F. Kronauge, Robert A. Kramer, and James M. Croop

Department of Radiologo Brigham ami Women's Hospital fD. P-W.. M. L C., J. F. K.J. and Division of PédiatrieHematologv/Oncologv, Dana-Farber Cancer Institute.Tile Children's Hospital ¡M.B., J. M. C.¡.Harvard Medical School, Boston, Massachusetts 02115, and Lederle Lítboratories, American Cyanatnid, Pearl River, M'u- York 10965

IR. A. K.I

ABSTRACT

The multidrug-resistant P-glycoprotein (Pgp), a M, 170,000 plasma

membrane protein encoded by the mammalian multidrug resistance gene(MI)Kl ), appears to function as an energy-dependent efflux pump. Many

of the drugs that interact with Pgp are lipophilic and cationic at physiological pH. We tested the hypothesis that the synthetic -/-emitting organ-

otechnetium complex, hexakis(2-methoxyisobutylisonitrile)technetium(I)(["•"TcjSESTAMIBI), a lipophilic cationic radiopharmaceutical, could be

a suitable Pgp transport substrate capable of functional imaging of theMDR phenotype. The cellular pharmacological profile of ["mTc]SESTA-

MIBI transport was examined in Chinese hamster V79 lung fibroblastsand the 77A and LZ derivative cell lines which express modestly low,intermediate, and very high levels of Pgp, respectively. Steady-state contents of ["-TclSESTAMIBI in V79, 77A, and LZ cells were 10.0 ±0.5

(SEM) (n = 9), 3.6 ±0.5 (n = 8), and 0.4 ±0.02 (n = 9) fmol-(mgprotein)"1 (nM0)"', respectively, consistent with enhanced extrusion of the

imaging agent by Pgp-enriched cells. Maximal doses i 100 UM) of themultidrug-resistant reversal agents verapamil and cyclosporin A enhanced["•"TcJSESTAMIBIaccumulation in V79, 77A, and LZ cells by approxi

mately 10-, 25-, and 200-fold, respectively. The median effective concen

tration values for tracer accumulation in the presence of verapamil in V79,77A, and LZ cells were 4, 100, and 200 UM,and those for cyclosporin Awere 0.9, 3, and >25 MM,respectively. Pgp-mediated ["""TcJSESTAMIBI

transport occurred against its electrochemical gradient and was found tobe ATP dependent displaying an apparent A",,,of 50 UM.Carrier-added

[*9Tc]SESTAMIBI was 11- to 13-fold less toxic in multidrug-resistant cells,and inhibited photolabeling of Pgp by [I25l]iodoaryl azidoprazosin ¡na

concentration-dependent manner; half-maximal displacement was observed at approximately 100- to 1000-fold molar excess ["TcJSESTAMI-BI. Exploiting the favorable 7 emission properties of 9*mTc, functional

expression of Pgp was successfully imaged in human tumor xenographs innude mice with pharmacologically inert tracer quantities of [WmTc]SES-

TAMIBI. Functional imaging with these Organotechnetium complexesmay provide a novel mechanism to rapidly characterize Pgp expression inhuman tumors in vivo, target reversal agents in vivo, and ultimately provide a means to direct patients to specific cancer therapies.

INTRODUCTION

Overexpression of the mammalian multidrug resistance gene(MDR1) is responsible for resistance to a broad spectrum of diversecytotoxic agents (1-3). Multidrug-resistant cell lines overexpressingPgp3 are resistant to a structurally and functionally diverse group of

chemotherapeutic compounds which include anthracyclines, Vincaalkaloids, and actinomycin D (1, 2,4). Recent investigations have alsofocused on an expanding number of agents including calcium channelblockers, calmodulin antagonists, quinine derivatives, synthetic iso-

Received 10/1/92; accepted 1/19/93.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1This work was supported by N1H Grants (HL42966, CA011227. and CA48162) andby American Cancer Society Grant JFRA-42Õ)to J. M. C.

2 Established Investigator of the American Heart Association. To whom requests forreprints should be addressed, at Department of Radiology. Brigham and Women's Hos

pital, 75 Francis Street. Boston. MA 02115.3 The abbreviations used are: Pgp, P-glycoprotein; MEBSS. modified Earle's balanced

salt solution; CCCP, carbonyl cyanide-m-chlorophenylhydrazone; IAP, [125I|iodoaryl azidoprazosin; SRB, sulforhodamine B; MeGlc. 3-O-|'H]methylglucose; EC5(>,median ef

fective concentration; SESTAMIBI, hexakis(2-methoxyisobutyli.sonitrile)technetium(l).

prenoids, tamoxifen, and cyclosporins which are capable of reversingthe multidrug-resistant phenotype. presumably through competitiveinteractions or steric changes of Pgp (5-9). The drugs possess diverse

chemical features, but many tend to be hydrophobic and positivelycharged at neutral pH (5). These general features and the broad ligand-

binding properties of Pgp raised the possibility that the syntheticy-emitting Organotechnetium complex, [""'Tc]SESTAMIBI, a lipo

philic cationic radiopharmaceutical useful in cardiac imaging, wouldalso be a suitable transport substrate for Pgp.

Chemical analysis of ground state ["TcJSESTAMIBI reveals a

stable monovalent cation with a central Tc(I) core surrounded by sixidentical MIBI ligands coordinated through the isonitrile carbons in anoctahedral geometry (10). In effect, the alkyl isonitrile ligands encasethe metal in a sphere of lipophilicity while enabling delocalization ofthe cationic charge. The complex contains no ionizable functionalgroups and is extremely stable in vivo without significant metabolism(11). Prior biophysical studies have established that [99mTc]SESTA-

MIBI is a Nernstian probe of membrane potential (12, 13). Net accumulation and unidirectional uptake rates of [""Tc]SESTAMIBI are

thermodynamically driven by negative mitochondrial inner matrixpotentials (Ai|/) and plasma membrane potentials (£,„)(13), therebyconcentrating the agent within cells in a manner similar to that of otherlipophilic cationic probes of membrane potential (14, 15). Approvedfor use clinically as a myocardial perfusion imaging agent (16), recentstudies also show tracer accumulation in various tumors in vivo (17-

19) and human carcinoma cell lines in vitro (20). It is possible that thevariable accumulation identified in these tissues may be in part attributable to extrusion of [99mTc]SESTAMIBI by Pgp. We examinedthe hypothesis that [99nTc]SESTAMIBI may interact with Pgp as a

novel organometallic substrate by characterizing the tracer accumulation and inhibition profile of [99mTc]SESTAMIBI in multidrug-resistant cell lines. The results indicate that [99mTc]SESTAMIBI is a

transport substrate recognized by Pgp and further demonstrate thefeasibility of imaging Pgp function in vivo by using a nude mousetumor model.

MATERIALS AND METHODS

Cells. Chinese hamster V79 lung fibroblasts and the Adriamycin-selected

77A and LZ derivative cell lines were grown by minor modification of previously published methods (21). Briefly, cells were plated in l(M)-mm Petridishes containing seven 25-mm glass coverslips on the bottom and were grownto confluence in a-minimal essential medium (GIBCO, Grand Island, NY)supplemented with i.-glutamine (1%), penicillin/streptomycin (1%), and fetal

calf serum (10%) in the presence of 0. 0.1, and 8 |jg/ml Adriamycin, respectively. For serial passage, cells were dispersed by incubating in 0.25% trypsinsolution for 1-2 min (25°C)prior to plating. Human Alex and drug-resistantAlex/A.5 cells (22) were passaged serially in Dulbecco's modified Eagle's

medium supplemented as above in the presence of 0 and 0.5 ug/ml Adriamycin,respectively. Human CEM cells and the multidrug-resistant derivative line

CEM/VBL (9) were grown in supplemented RPMI medium in the presence of0 and 0.3 ug/ml vinblastine, respectively.

Preparation of ['""TclSESTAMIBI and ["TcjSESTAMIBl. Synthesisof the radiolabeled compound ["'""TcJSESTAMIBI was performed with a

one-step kit formulation (Cardiolite; gift of Du Pont. Medical Products Divi

sion, Billerica, MA) containing solid stannous chloride (0.075 mg) as a reducing agent for the technetium and excess MIBI as the Cu(MIBI)4BF4 salt (1.0

977

Research. on September 10, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

IMAGING OF MULTIDRUG-RESISTANT Pgp

mg) (12). [""'"TclTcOj (20-30 mCi; 5-10 pmol/mCi) in 1-2 ml saline (0.15 M

NaCll obtained from a commercial molybdenum/technetium generator (DuPont Pharma. Billerica, MA I was added to the kit reaction vial, heated at 100°C

for 15 min. and allowed to cool to room temperature, producing an almostquantitative yield of the |w'"Tc]MIBI<,4 complex. Excess reducing agent and

starting materials were separated from the radiolabeled component as follows:the contents of the reaction vial were loaded by syringe onto a reversed-phaseC|K-Sep-Pak cartridge (Waters Associates, Miltbrd. MA) prewet with 3 ml

90% ethanol followed by 5 ml distilled water. Hydrophilic impurities wereeluted from the cartridge by washing with 10ml saline (0.15 M).and the desired[WmTclSESTAMIBI was collected by elution with ethanol:saline (2 ml: 9:1,v/v). Final total ''""Tc activity in the 2-ml effluent (stock) was assayed in a

standard dose calibrator (CRC-12. Capintec. Ramsey, NJ). Radiochemicalpurity was found to be greater than 97% by thin-layer chromatography (alu

minum oxide plates; ¡.T. Baker, Phillipsburg. NJ) with the use of ethanol

(absolute) as the moble phase.Carrier-added |""Tc]SESTAMIBI was prepared from 8-10 mg ( 10-12 umol)

of solid [Tc](MlBD,, chloride powder dissolved in 0.5 ml of 95% ethanol

solution as described (II. 12).Solutions and Reagents. Control solution for transport experiments was

MEBSS containing (HIM): 145 Na\ 5.4 K +, 1.2 Ca2 +, 0.8 Mg2+, 152 C]-, 0.8H2PO4-. 0.8 SOj2-. 5.6 dextrose, 4.0 4-(2-hydroxyethyl)-l-piperazineethane-

sulfonic acid, and \c7r bovine calf serum (v/v), pH 7.4 ±0.05. A 130 niM

potassium/20 m\i chloride solution was made by equimolar substitution of

potassium methanesulfonate for NaCI as described (12).Verapamil. vinblastine. colchicine. daunorubicin, Adriamycin. quinidine.

CCCP. methotrexate. and valinomycin (Sigma Chemical Co.. St. Louis. MO),but not cyclosporin A (Sando/ Pharmaceuticals), were dissolved in dimethylsulfoxide before they were added to solutions. Final concentrations of dimethylsulfoxide (drug carrier) and ethanol ((*"""Tc|SESTAMIBI eluate) were typi

cally <0.5%. which have been found to have no effect on [WmTc]SESTAMIBI

net uptake in cultured cells ( 12). lAPwas obtained from New England Nuclear.

Cell Kinetic Studies. Coverslips with confluent cells were removed fromculture media and preequilibrated for 40-60 s in control buffer. Uptake andretention experiments were initiated by immersion of coverslips in 60-mm

glass Pyrex dishes containing 4 ml loading solution consisting of MEBSS with0.1-0.6 nw [Tc-SESTAMIBI] (25-100 uCi/ml). Cells on coverslips were removed at various times, rinsed three times in 25 ml ice-cold (2°C)isotope-free

solution for 8 s each to clear extracellular spaces, and placed in 35-mm plastic

Petri dishes (12). Aliquots of the loading buffer and stock solutions were placed

in glass test tubes for standardizing cellular data with extracellular concentration (nM(l) of [WmTc]SESTAMIBI. Cells on coverslips, stock solutions, and

extracellular buffer samples were assayed for y activity in a well-type sodium

iodide gamma counter (Omega 1: Canberra. Meridan. CT). after which cellswere extracted in 1% sodium dodecyl sulfate with 10 msi sodium borate for

protein assay by the method of Lowry, using bovine serum albumin as theprotein standard. Appropriate geometric controls allowed normalization of yactivity assayed in Petri dishes to that assayed in glass test tubes. Knowledgeof the elution history of the generator and activity of stock solutions alloweduse of generator equilibrium equations to calculate the absolute concentrationof total Tc-SESTAMIBI in the solutions (13). This allowed data to be presentedas fmoHmg proteinl^'InM,,)"1. Pharmacological ECiu values were estimated

by computer fit of concentration-effect curves.For one series of experiments. |2<"TI]thallous chloride in saline (Du Pont

Pharma) was added to buffer for cell tracer studies (20-25 uCi/ml: 15-30

pmol/mCi) and analy/ed identically.•¿�Cell Survival Studies and Median Lethal Dose Determinations. Sur

vival of parental and multidrug-resistant cell lines exposed to carrier-added[TclSESTAMIBI was assayed in 96-well microtiter plates. Cells (4.000-

2().(KX))were plated with increasing concentrations of [TcJSESTAMIBI andincubated for 72 h at 37°Cin triplicate in 3 separate experiments. Multidrug-

resistanl cells were cultured in drug-free media for 72-96 h prior to culture in

[TcJSESTAMIBI. Cell survival was assayed by using SRB (22). Cells werefixed in 10% trichloroacetic acid for 60 min at 4°C.washed 5 times with tap

water, and stained with 0.4% SRB in 1% acetic acid for 15 min at roomtemperature. Excess SRB was removed with four 1% acetic acid washes andthe stain was redissolved in IO ITIMunbuffered Tris base. Quantitation wascarried out by using an enzyme-linked immunosorbent assay plate reader at a

wavelength of 550 nm. Survival was expressed as the percentage of surviving

cells relative to growth in minimal essential media without [Tc]SESTAMIBI.

Median lethal dose determinations were obtained by interpolation from the cellsurvival curves.

Cell Water Determination. Intracellular water space of V79 cells wasmeasured by a modification of the method of Kletzein et al. (23). Cells oncoverslips were incubated in glucose-free normal potassium or high potassium-

low chloride MEBSS buffer containing 2. 5. or 10 HIMMeGIc (1 uCi/ml; 65Ci/mmol) for l h at 37°C.Control experiments in cultured cells have shown

complete equilibration of the nonmetabolizable hexose across cell membranesby 40 min (24). Cells were rinsed three times in 25-ml volumes of ice-cold

buffer to clear extracellular MeGIc. extracted in 1% sodium dodecyl sulfate forprotein determination as previously described, and assayed for 'H activity by

standard liquid scintillation techniques, along with a 200-ul sample of the

extracellular solution. MeGIc uptake/mg protein increased linearly with extracellular [MeGIc] (r > 0.99) and intersected the ordinate at a point not signif

icantly different from the origin (P > 0.25).ATP Content. Cellular ATP content established during a 15-min incubation

in glucose-free MEBSS containing 1 g/ml bovine serum albumin and various

sublethal concentrations of the uncoupler CCCP in the presence or absence ofverapamil (10 UM) was assayed fluorometrically by a standard hexokinasereaction on cells that had simultaneously been loaded as above with [Wl"Tc]-

SESTAMIB1. After tracer loading and rinsing in ice-cold isotope-free buffer,

preparations were extracted in perchloric acid, neutralized with potassiumcarbonate, immediately assayed for -y activity, and frozen (-20°C) as previ

ously described (12). Cell protein was determined by Lowry assay. Previouslyfrozen cell extracts were then thawed within 5 days of the experiment andassayed for ATP content fluorometrically (SFM 25: Kontron Instruments,Zurich. Switzerland). ATP is expressed as nmoHmg protein)"'.

Photolabeling of Pgp. Plasma membrane-enriched fractions from parental(V79 and CEM) and multidrug-resistant (LZ and CEM/VBL) cultured cellswere isolated with a high speed spin (100.000 X #) after dounce homogeni-zation in 10 imi 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid buffered

0.25 Msucrose (pH 7.3) and removal of nuclei with a low speed spin (600 X#). Fifty ug of membranes in a final volume of 50 ul were incubated with 2.5nm ['25I|iodoaryl azidoprazosin for 60 min at 25°C in the dark (25, 26).

[Tc]SESTAMIBI or verapamil were included at the concentrations shown.The sample was irradiated with an UV lamp (UVP, Model UVL-56. 366 nmwavelength) for 20 min at 25°Cand fractionated on a 10% polyacrylamide gel.

The gel was fixed with 10% acetic acid, rinsed in 2% glycerol. dried, andexposed to XAR-5 film with an intensifying screen at -70°C for 12^18 h.

Western Blot Analysis. Pgp was delected in enriched membrane preparations from drug-sensitive and multidrug-resistant cells and tumors by Westernblotting with the use of a 1:150 dilution of the anti-Pgp mouse monoclonal

antibody C219 (Centocor Corp.). Membrane preparations were fractionated ona 7% sodium dodecyl sulfate-polyacrylamide gel. and transferred to nitrocellulose where immune complexes were revealed with goat anti-mouse antibody

coupled to alkaline phosphatase and developed as described (27).Scintigraphy. All studies were approved by institutional animal welfare

committees. Athymic BALB/c-/i«/«Hmice were inoculated s.c. with 10'' drug-

sensitive KB cells in the right flank and drug-resistant KB-8-5 cells (28) in the

left flank, and tumors were allowed to grow to approximately 1.5 g. Animalswere anesthetized with sodium pentobarbital i.p. (50 ing/kg) and positionedover a gamma camera (GE Starcam; LEAP collimatori. A bolus of [WmTc]-SESTAMIBI (0.25 mCi) (to characterize Pgp phenotype) mixed with [2"'T1]-

thallous chloride (0.25 mCi) (to document perfusion) (29) was then injected viaa tail vein. Sequential planar images were collected for 60-s intervals for 1 h.Energy discrimination was provided by 20% windows centered over the 140-KeV photopeak of """Te and the 68-92 KeV emission spectra of 2"'T1. Each

image was corrected on line for camera nonuniformity with a 300-million

count flood and stored at a digital resolution of 64 x 64. Images were correctedfor emission spillover into each window and radioactive decay. No attenuationor scatter correction was used. Accumulation of tracers into internal organssuch as heart, liver, kidneys, and bladder was masked to highlight the s.c.tumors in false color images.

Following imaging, mice were sacrificed by lethal pentobarbital injection,and tumors and major organs were removed, trimmed of connective tissue, andweighed in tared vials. Tissues were then assayed for WmTc and :'"TI y activity

in a well-type gamma counter as above, using 20% energy discrimination

windows, and then data were corrected for emission spillover and radioactivedecay.

978

Research. on September 10, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

IMAGING Or MU1.TIDRUG-RI-SISTANT I'gp

Statistics. Values are presented as mean ±SEM. Statistical significancewas determined by one-way analysis of variance, or unpaired Student's i test

as indicated in the text. P < 0.05 was considered significant.

RESULTS

[WnlTc]SESTAMIBI Transport Assays. Chinese hamster V79

lung fibroblasts and the Adriamycin-selected 77A and LZ derivative

cell lines express modestly low. intermediate, and very high levels ofthe Pgp. respectively (Fig. 1). V79, 77A, and LZ cells in monolayerculture exposed to tracer [w"Tc]SESTAMIBI accumulated the radi-

olabel to a plateau within 15 min (Fig. 2). Steady-state contents of[""TcJSESTAMIBI in V79. 77A, and LZ cells were 10.0 ±0.5

(n = 9), 3.6 ±0.5 (n = 8), and 0.4 ±0.02 (/; = 9) fmoHmgproteinr'-(nM,,r', respectively, consistent with either reduced influx

or enhanced extrusion of the imaging agent by Pgp-enriched cells.To ascertain if [''""Te|SESTAMIBI cell content could be increased

by inhibition of Pgp. the effect of several known multidrug resistancereversing agents were determined over a wide range of concentrations.As shown in Fig. 3. verapamil and cyclosporin A were each found tobe potent enhancing agents of |''''"'Tc|SESTAMIBI net uptake. Maximal doses (>100 JJM)of these reversing agents enhanced [w'"Tc|-

SESTAMIBI accumulation in V79, 77A, and LZ cells by approximately 10-, 25-, and 200-fold, respectively. The EC,,, values for["'"TcjSESTAMIBI accumulation in the presence of verapamil in

V79, 77A, and LZ cells were 4, 100. and 200 UM,respectively. TheEC5() values for cyclosporin A were 0.9, 3, and >25 UM,respectively.The reversal curves were proportionally shifted to higher concentrations in the increasingly drug-resistant 77A and LZ cells. Cytotoxicdrugs included in the multidrug-resistant phenotype such as vinblas-

tine, colchicine. and daunorubicin, as well as the reversing agentquinidine (30), also enhanced [l>''"'Tc]SESTAMIBI net uptake in a

concentration-dependent manner. By comparison, methotrexate, a cy-totoxic agent to which drug-resistant cells remain sensitive (5, 30),produced no enhancement of |9<'"'Tc]SESTAMIBI (data not shown).

Further evidence for Pgp-mediated |WmTclSESTAMIBI transport

was provided by inhibition of efflux with verapamil, a Pgp-reversing

agent. V79 cells were loaded to steady state with the tracer and thenincubated in isotope-free buffer in the presence or absence of verapamil. Compared to control, the initial efflux rate of [9ymTc]SESTA-

MIBI (determined at 2 min) was ~2.5 times slower in the presence ofverapamil (10 UM)(Fig. 4). [WmTc]SESTAMIBI efflux would not be

expected to be completely inhibited, since passive transmembrane

1 2 3- 207

- 111

- 71

- 44

Fig. 1. Expression of Pgp in V79. 77A, and LZ cells (Lanes I, 2, and J, respectively)as determined by Western blot of plasma membrane preparations. Arrow. M, 170,(KK).

s

I20.006"o

ouH 2

HViWo

0

0 20 30

Time (min)Fig. 2. Characterization of ['""'TcISESTAMIBI transpon hy drug-sensitive and drug-

resistant cells. Net accumulation of tracer in V79, 77A, and LZ cells expressing modestlylow. intermediate, and high levels of Pgp. respectively. Points, mean of 3—4determinations: SEM did not exceed I5r/f of mean values.

diffusion of the tracer down its concentration gradient would occur inparallel with Pgp-mediated efflux (24).

Initial uptake rates of |WmTc)SESTAMIBI were determined in V79

and 77A cells in the presence and absence of verapamil. In theseexperiments, initial uptake rates were used as estimates of unidirectional influx of r'"Tc|SESTAMIBI. When Pgp-mediated efflux was

inhibited with verapamil, initial uptake rates in drug-sensitive V79and multidrug-resistant 77A cells were indistinguishable (Fig. 5: 0. IOSversus 0.113 fmoHmg proteinr'-(nMor'-(sr'; p > 0.5). In the

presence or absence of verapamil. initial uptake rates in V79 cellswere also identical. However, in the absence of verapamil. there wasearly evidence of Pgp-mediated efflux of [''''"'TclSESTAMIBI in 77A

cells manifested by flattening of the accumulation curve within 15-20s. Thus, steady-state differences between drug-sensitive and drug-

resistant cells were unlikely to be a result of differences in influx ofthe agent.

Energy-dependent Efflux. The energetic requirement for [WmTcl-

SESTAMIBI efflux was established by quantitative compartmentalanalysis in V79 cells which express modestly low levels of Pgp. Cellswere incubated in 130 ITIMK(, 20 IDMCl„buffer plus valinomycin(1 ug/ml). thereby depolarizing both At|/ and E,,, of cells toward zeroand eliminating the inward driving force for 99nlTc-SESTAMIBI ac

cumulation against its concentration gradient (12, 13). Under theseconditions, a steady-state content of 2.2 ±0.1 fmol-(mg protein)"'-(riM,,)"1 (/i = 9) was obtained. While cell water was 8.7 ±0.4(n = 10) ul (mg protein)"1 in control buffer, it was 7.4 ±0.6 (n = 10)Hi (mg protein)"' in high K(, buffer, yielding a ['"""Tc]SESTAMIBI

in:out ratio of 0.30 ±0.02. This was significantly less than theequilibrium ratio of 1.0 expected for passive diffusion and equilibration of [l'v"'Tc]SESTAMIBI with intracellular water spaces under

these conditions. Exclusion of [Wl"TcjSESTAMIBI from the cytosol

of V79 cells at near zero-potential conditions implied the presence ofan energy-dependent efflux pump mechanism. Addition of a 100 UMconcentration of the reversing agent quinidine increased the |Wl"Tc|-

SESTAMIBI in:out ratio to 0.75 ±0.14 (P < 0.05), consistent withinhibition of active transport by the Pgp. This observation is furtherhighlighted by the low levels of agent accumulation in the Pgp-

979

Research. on September 10, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

IMAGINO OF MULTIDRUG-RESISTANT Pgp

oL*CL

O

SìoU

O 0.1 l 10 100 1000

[Verapamil] (uM)

« 20 -

[Cyclosporin A] ( uM )Fig. 3. Effect of multidrug resistance reversing agents on ["""TcJSESTAMlBI accu

mulation. Cells were incubated for 15 min in buffer containing [Wn1Tc)SESTAMIBI and

the indicated concentrations of verapamil (AI or cyclosporin A (B). Closed symbols.accumulation in the absence of drug. Points, mean of 3^4 determinations: SEM did not

exceed 15% of mean values.

enriched 77A and LZ cells (Fig. 2), which point toward Pgp-mediatedexclusion of ["'"TcjSESTAMIBI from their cytosols (assuming sim

ilar cell water), even in the presence of the strong inward driving forceof normal membrane potentials.

|w"Tc]SESTAMIBI content was ATP dependent as demonstrated

by analysis of verapamil-enhanced accumulation of the tracer in relation to cytosolic ATP concentration. Enhancement of [w"Tc]SES-

TAMIB1 accumulation by verapamil was used to represent the transport mediated by Pgp. A range of intracellular ATP values wasestablished by exposing V79 cells to graded sublethal concentrationsof the mitochondria! uncoupler CCCP (O.I to 10 UM)for 15 min (12,31). Control experiments confirmed that a steady-state [w"'Tc]SES-

TAMIBI level was achieved within 15 min and maintained for at least45 min under these conditions. [WmTc]SESTAMIBI and ATP contents

were concurrently measured in each preparation in the presence orabsence of verapamil (10 UM).Eadie-Hofstee analysis of verapamil-enhanced [""TcJSESTAMIBI accumulation as a function of ATP

concentration (based on the cell water determination of 8.7 ul (mgprotein)"'] was consistent with Michaelis-Menton kinetics displaying

an apparent Km of 50 p\i for ATP-dependent transport (Fig. 6). These

data establish a pharmacological and biophysical profile consistent

8o.00

"o

oo*—ICQ

IV)Wco

è

i V79 control0+10 JO.Mverapamil

.1

5 10

Time ( min)Fig. 4. Pgp-mediated efflux of ["""'Tc|SESTAMlBl. V79 cells were equilibrated (15

min) in loading buffer containing |'"""Tc|SESTAMIBI and then transferred to isotope-free

solution in the absence or presence of 10 fiM verapamil for the times indicated. Cell-associated activity during washout is plotted. Points, mean of 3^4 determinations; SEMdid not exceed \5CAof mean values. Note semilogarithmic scale.

c'S

Sex00

"o

e

Sìoo

I—-I

CO

•¿�V79o V79 + 50 uM verapamil•¿�77Aa 77A + 50 uM verapamil

S3Vi

Fig. 5. Initial uptake rales of (""'"TclSESTAMlBl in V79 and 77A cells in the absence

(•. •¿�)and presence (O, D) of verapamil (50 JÃŒM).Eaeh point represents a singledetermination. Solili lines, linear regressions of the data with the following slopes: V79,0.110 fmoHmg protein)''-(nMj-''(sec)-'; V79 + verapamil, 0.108; 77A. 0.002; 77A +

verapamil, 0.113, respectively.

980

Research. on September 10, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

IMAGING OF MUI.TIDRUG-RESISTANT Pgp

with the hypothesis that |'''""Tc]SESTAMIBI is a transport substrate

recognized by Pgp.Cytotoxicity. To functionally assay carrier-added |1)9Tc]SESTA-

MIBI toxicity in relation to Pgp expression, survival of both hamsterand human parental and multidrug-resistant cells was determined in

the presence of the agent. fTclSESTAMIBI can produce acute cellular toxicity, probably mediated by mitochondria! depolarization anduncoupling (13, 15), at concentrations exceeding 75 UM(>105-fold

higher than the tracer concentrations used clinically). Carrier-addedquantities of p^TclSESTAMIBI were synthesized and hamster V79

and LZ cells and the human Alexander hepatocellular carcinoma cellline (Alex) and a multidrug-resistant derivative (Alex/A.5) were cul

tured in increasing concentrations of the compound. As shown inFig. 7, multidrug-resistant LZ and Alex/A.5 cells were 11- and 13-foldmore resistant to |')lTclSESTAMIBI compared to their drug-sensitive

parental cell lines.Inhibition of Photolabeling. The interaction of [WmTc]SESTA-

MIBI with Pgp was directly examined by displacement of the Pgpphotoaffinity probe IAP from plasma membrane preparations derivedfrom drug-resistant cells. Carrier-added [l'''Tc]SESTAMIBI was incu

bated in reaction buffer containing enriched membrane preparationsfrom drug-resistant LZ cells or human leukemia CEM/VBL cells with

the photoaffinity probe. Fig. 8 demonstrates that [TcJSESTAMIBIinhibited photolabeling of Pgp in a concentration-dependent manner.Half-maximal effect was observed at approximately 100- to 1000-foldmolar excess [9Tc|SESTAMIBI, similar to other Pgp modulators

(8, 26), including verapamil (Fig. 8).Functional Imaging of Pgp. The use of -/-emitting [''''"'TclSES-

TAM1BI to image Pgp function in human tumors in vivo was demonstrated with a nude mouse tumor model. Bilateral tumor xenographswere produced in opposite flanks of Athymic BALB/c-m</m/ micewith drug-sensitive KB human carcinoma cells and mdrl mRNAexpressing multidrug-resistant KB-8-5 cells (28). Studies in vitro havereported that KB-8-5 cells are 3.2-fold resistant to Adriamycin compared to parental KB cells (28). A relative resistance of 2-2.5 forKB-8-5 tumors in vivo was documented for this study by growthanalysis as follows. Tumor si/.e (mg) at 2 weeks in BALB/c-rtK/nu

200 400

[ATP] (jiM)

600

Fig. 6. ATP dependence of verapamil-induced [WmTc|SESTAMIBI accumulation inV79 cells. Cell ATP and ['"'"TclSESTAMIBI content were determined simultaneously on

each preparation following a 15-min incubation with various CCCP concentrations in thepresence or absence of verapamil <K) UM).The [''4niTc]SESTAMIBI concentration in this

experiment was 0.8 nw. Points, mean ±SEM of 3 determinations each. Inset, Eadie-Hofstee plot of the data. Solid line, linear regression which yields an apparent Km = 50MM.

120

100-

80-13'S 603

on

* 401

20-

0

OV79•¿�LZ

A

.1 1 10 100[99TC-SESTAMIBI] (\iM)

1000

vi

120

100-

80-

60-

40-

20-

0

B

o Alex•¿�Alex/A.5

.1 1 10 100[99TC-SESTAMIBI] (\iM)

1000

Fig. 7. Cell survival studies and LDy, determination. Survival of parental V79 andmultidrug resistant L7. cells (A ) and parental Alex and multidrug-resistunt Alex/A.5 cells(ß)in increasing concentrations of carrier-added fTclSESTAMIBI.

mice treated at Adriamycin doses of 8 and 12 mg/kg, respectively(percentage of control): KB, 25 and 13%; KB-8-5. 51 and 32%. Thus,

low levels of mdrl mRNA expression were sufficient to confer resistance both in vitro and in vivo. Further evidence supporting maintain-ance of the MDR phenotype by KB-8-5 cells in vivo was provided by

demonstrating that MDR reversal agents (e.g., cyclosporin A) increased activity of Adriamycin in KB-8-5. but not in parental KBtumors.4

Simultaneous images of tumor Pgp function with |WmTc]SESTA-MIBI and tumor perfusion with [2("Tljthallous chloride (29) were

generated in this mouse tumor model with dual-isotope planar scin-

tigraphy. Images from a representative mouse are shown in Fig. 9. ThePgp-enriched KB-8-5 tumor accumulated less ["mTc]SESTAMIBI

compared to the parental tumor in the opposite flank (Fig. 9A, arrow),producing a readily visualized difference. Tumor perfusion, however,as reflected by 2("T1 imaging was similar for both tumors (Fig. 9ß,

arrows). Control studies in vitro with drug-resistant LZ cells confirmed that 2<"T1was not a Pgp transport substrate (net 20IT1 accu-

4 R. A. Kramer, unpublished observations.

Research. on September 10, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

IMAGING OF MULTIDRUG-RESISTANT Pgp

Fig. X. Inhibition by [Tc]SESTAMIBI ofI'-MIIAP binding to Pgp. (Al Urne I. V79, and

Lane 2. LX, membranes pholoaffinity labeled withIAP alone. Arm»: Pgp labeling at M, 170,(XK).LZmembranes labeled with IAP in the presence of0.25, 2.5, 25. and 250 MMcarrier-added |'"Tc|SES-

TAMIB1 (Lanes 3-6) or verapamil (Lanes 7-10),respectively. (B) Lime I. CEM, and ¡¿me2, CEM/VBL, membranes labeled with IAP alone (M,I70.IKK). arrow). CEM/VBL membranes labeledwith IAP in the presence of 0.25. 2.5, 25, 250 UMcarrier-added |Tc|SESTAMIBl (Lanes 3-6) orverapamil (Lines 7-lii). respectively.

12 3456789 10

M««

123456789 10

B

initiation at 30 min |fmol-(mg protein) ' (nM,,) '): control, 135.4 ±

15.9; + 50 UMverapamil. 138.6 ±4.1 (n = 4); P not significant].The imaging results were confirmed with quantitative biodistribu-

tion analysis of the tracers in the tumors and major organs. The s.c.implanted tumors were modestly perfused as demonstrated by 60-mintumor/liver ratios for 2I"T1 (per g tissue) of 0.17 ±0.03 (n = 8).

Perfusion was not significantly different between Pgp-enriched andparental tumors (-"'Tl/g tumor: P not significant), and therefore de

livery of diffusible tracers was similar. In accord with the scintigraphicimages, analysis of paired tumor xenographs in four animals demonstrated r""Tc]SESTAMIBI/20'Tl ratios 35 ±4% lower in Pgp-en

riched tumors compared to parental tumors implanted in the opposite

flank. These biodistribution data confirmed enhanced Pgp-mediatedtransport of [w'"Tc]SESTAMIBI out of multidrug-resistant tumors

DISCUSSION

Pgp expression in human tumors has been characterized by use ofimmunohistochemical techniques. RNA expression analysis, and flowcytometry of fluorescent substrates (32-38). However, characteriza

tion of tumor specimens from patients with these techniques requiresserial tissue biopsies, and the usefulness of these approaches has beenlimited by the labor intensity of current protocols, the invasiveness of

Fig. 9. Scinligraphic images magnified X 2.6 of Pgp expression (A land perfusion (fl) in human carcinoma xenographs in nude mice. Anterior planar images at 60 min (mouse supinefacing reader) show low |"""Tc|SESTAMIBI accumulation in the Pgp-enriched KB-8-5 tumor (A: left flank) compared to the parental KB tumor (A: right flank, arrow) despite similar-'"TI perfusion to each (B; arrows),

982

Research. on September 10, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

IMAGINO OF MUI.TIDRUG.RESISTANT Pgp

biopsies, sampling errors from heterogeneous tumors, and the sensitivity and specificity required of RNA and antibody probes (1, 39).Because an ;';/ vivo screen would ease detection of the Pgp and aid

development of new drugs targeted to inhibit the Pgp, the goal of ourstudy was to establish a scintigraphic substrate for the MDR1 geneproduct that could be used as a noninvasive probe of tissue expression.

Characterization of [99mTc]SESTAMIBI Interaction with Pgp.Several lines of evidence supported the hypothesis that [w"Tc|SES-

TAMIBI was a transport substrate recognized by Pgp: (a) drug-sensitive and multidrug-resistant cells demonstrated steady-state accumu

lation of the agent in reverse rank order of their expression of Pgp: (/?)known multidrug-resistant-reversing drugs such as verapamil and cy-closporin A enhanced (''''"TcjSESTAMIBI levels in a concentration-

dependent manner with ECM>values comparable to published values(5): (r) unidirectional efflux of ["""TclSESTAMIBI was inhibited by

the Pgp modulator verapamil; (</) initial uptake rates were identical indrug-sensitive V79 and drug-resistant 77A cells when Pgp was inhib

ited with verapamil. This further indicated that the lower net uptake inPgp-enriched cells was a result of enhanced extrusion of the agent, not

a difference in influx; (e) analysis of in:out ratios established that thefreely diffusible |'''""Tc]SESTAMIBI could be excluded from the cy-

tosol against its concentration gradient. Thermodynamically this required the presence of an active transporter such as Pgp to mediatedrug efflux; (/) verapamil-enhanced accumulation of (Wl"Tc]SESTA-

MIBI was ATP dependent, displaying Michaelis-Menton kinetics with

an apparent Km of 50 UM,similar to that for the ATP dependence ofPgp-mediated vinblastinc transport (7); (#) multidrug-resistant cellswere 11- to 13-fold more resistant to the toxic effects of carrier-added[•"TcjSESTAMIBIcompared to parental cells; and (/;) |""Tc]SESTA-

MIBI inhibited photolabeling of Pgp by [I25I]IAP.[''""TclSESTAMIBI is a membrane potential-dependent probe sim

ilar to tetraphenyIphosphonium (12. 13, 24). The opposing interactionof membrane potential-dependent influx and Pgp-mediated efflux of[""TcjSESTAMIBI could be demonstrated in Pgp-enriched cells. Forexample, initial uptake rates of [''''"'TclSESTAMIBI in drug-sensitive

V79 and multidrug-resistant 77A cells were identical in the presence

of verapamil, indicating that the plasma membrane potentials weresimilar in these cells, and therefore could not account for the transportdifferences in each cell line. However, the robust capacity of Pgp toextrude [WnTclSESTAMIBI from 77A cells in the absence of verap

amil was evident by flattening of the uptake curves after only 15 s.Given the disparate conclusions reported for the effects of Pgp expression on drug influx (5), these data highlight the potential forunderestimation of true initial uptake rates of drugs in control bufferwith multidrug-resistant cells, with the use of techniques using de

layed sampling times.The maximal accumulation levels of [""TcjSESTAMIBI during

Pgp inhibition approached similar levels in all hamster cell lines. Thisindicated that in the absence of Pgp-mediated efflux, the volume ofdistribution for |''l'"'Tc)SESTAMIBI (reflecting the sum driving forces

of mitochondria! and plasma membrane potentials acting in series oninner matrix and cytosolic spaces) was not significantly differentbetween drug-sensitive and multidrug-resistant cells. Thus, the differ

ences observed in the absence of reversal agents were best attributedto expression of Pgp in the plasma membranes.

Interestingly, the |''''"'Tc]SESTAMIBI reversal curves were propor

tionally shifted to higher concentrations of reversal agent (higher ECS()values) in the increasingly drug-resistant 77A and LZ cells. Due to the

high levels of Pgp expression, these data may reflect a decrease inactual cell content of the competing reversal agent in resistant cells atconstant extracellular concentrations of each compound. Hypotheti-cally. however, a proportionally higher affinity for [WmTc]SESTA-

MIBI or lower affinity for the reversing agents in the Pgp-enriched

cell lines could not be excluded, perhaps mediated by posttranslational

modification on Pgp or by Pgp-induced changes in cytosolic pH (40).

The shifted inhibition curves also indicated that the optimal reversalconcentration for each individual agent may in general depend on thelevel of Pgp expression (5).

Identification of |"'""Tc|SESTAMIBI as a transport substrate for

Pgp further broadens the structural and functional criteria for putativerecognition domains of the transporter. While [w"Tc]SESTAMIBI is

biophysically similar to other lipophilic cationic probes of membranepotential recently reported as Pgp substrates (41), this organotechne-

tium complex is not planar and contains no basic nitrogen atoms,titratable protons, or phenyl groups (II, 24). The only functionality isthe external mcthoxy groups symmetrically distributed over the molecular surface.

Utility of a -/-emitting Pgp Substrate. This study demonstrated

the feasibility of scintigraphically imaging Pgp expression in vivo with[99l"Tc]SESTAMIBI and points to applications in functional imaging

with this transport substrate. This radiopharmaceutical combines aproven safety profile in humans at tracer doses (PM) (16) with theadvantages inherent to WmTc, the most commonly used radioisótopo

in planar and single photon emission computed tomography medicalimaging. ''''"'Tc provides a nearly ideal 7 emission energy (140 KeV),

short half-life (6.02 h), high photon flux, favorable dosimetry. lowcost, and extremely high specific activity (IO9 Ci/mol) and is readily

available from generator systems (42). Advantage can also be taken ofthe rapid blood clearance ( 16). low levels of nonspecific binding (12,13), and lack of metabolism of t"mTc]SESTAMIBI (II, 43), thereby

allowing early imaging and a higher biologically relevant signal compared to other labeled compounds.

This study used 2I"TI. the classic perfusion tracer (29). to document

independently equivalent initial delivery of the diffusible tracers todrug-sensitive and drug-resistant tumors. Initial tissue distribution of20IT1 has been well documented to correlate with regional perfusion

(29), while net cell uptake and retention are largely dependent onactive transport by the Na/K-ATPase (44. 45). Several human tumorshave been reported to retain -("T1 (45-^9); however, our data indicatethat 2<"TI is not recognized as a transport substrate by Pgp. Therefore,this enables dual isotope imaging, whereby the ratio of [Wl"Tc|-SESTAMIBI to 2("TI (or any other perfusion tracer) may be used tonormalize for differences in initial delivery of [9'''"Tc|SESTAMIBI to

the tumors.The ability to functionally assay IH vivo the Pgp transporter nonin-

vasively with a pharmacologically inert tracer substrate provides asignificant new tool for advancing the clinical understanding of multidrug-resistant phenotype in cancer patients. This approach may ul

timately be used to guide chemotherapeutic protocols, assist clinicaltrials or screening of new Pgp reversing agents, and direct tumorbiopsies. In vitro testing of cytotoxic and reversing agents in multi-drug-resistant cells may also be assisted by application of a simpleradiotracer method. Scintigraphic identification of Pgp-enriched tis

sues IHvivo necessarily requires detection of diminished accumulationof [l"""Tc]SESTAMIBI. a result of outward transport of this substrate.

Remaining to be ascertained is the ability to detect heterogeneoustumor expression with this constraint. In addition, although there israpid excretion of [>Wl"Tc]SESTAlVIIBIby liver, kidney, and bowel

(16), perhaps reflecting a normal transport function of Pgp. the highlevels of initial tracer uptake in these tissues will require developmentof optimal imaging times and judicious use of single photon emissioncomputed tomography to visualize abdominal tumors. In this regard,we envision future research targeted toward development of agents ofthis class with high affinity binding, rather than transport properties,to enable direct imaging of high Pgp expression. The relatively simplechemical structure of these organotechnetium complexes would allowfacile chemical modification for this purpose. An in vivo functional

Research. on September 10, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

IMAGING OF MULTIDRUG-RES1STANT Pgp

assay for Pgp should extend our understanding of this unique membrane transport protein in human physiology and cancer biology.

ACKNOWLEDGMENTS

The authors thank Drs. Robert Arced, Lee Herman, and Ting Wu forcritically reviewing the manuscript; Georgia Washington for her secretarialassistance; and Kate Sobey and Diane Johnson for their assistance with scin-

tigraphic image acquisition and analysis.

REFERENCES1. Gottesman. M. M., and Pastan, I. The mulinimi- Iransporter, a double-edged sword.

J. Biol. Chem., 263: I2163-I2I66. 1988.2. Endicott, J. A., and Ling. V. The biochemistry of P-glycoprotein-mediated multidrug

resistance. Annu. Rev. Biochem.. 58: 137-171. 1989.

3. Croop, J. M.. Gros. P., and Housman, D. E. Genetics of multidrug resistance. J. Clin.Invest., 81: 1303-1309, 1988.

4. Roninson, I. B. (ed.). Molecular and Cellular Biology of Multidrug Resistance inTumor Cells. New York: Plenum Publishing Corp.. 1991.

5. Ford, J. M., and Hail, W. N. Pharmacology of drugs that alter multidrug resistance incancer. Pharmacol. Rev.. 42: 155-199. 1990.

6. Safa, A. R., Glover, C. J.. Meyers. M. B.. Biedler. J. L.. and Felsted. R. L. Vinblastinephotoatfinity labeling of a high molecular weight surface membrane glycoproteinspecific for multidrag-resistant cells. J. Biol. Chem.. 261: 6137-6140. 1986.

7. Horio, M.. Gottesman, M. M.. and Pastan. 1. ATP-dependent transport of vinblaslinein vesicles from human multidrug-resistant cells. Proc. Nati. Acad. Sci. USA. 85:3580-3584. 1988.

8. Yang, C. P., Mellado. W., and Horwitz. S. B. Azidopine photoaffinity labelling ofmultidrug resistance-associated glycoproteins. Biochem. Pharmacol., 37: 1417-1421.

1988.9. Qian, X. D., and Beck, W. T. Progesterone photoaffinity labels P-glycoprotein in

multidrug-resistant human leukemic lymphoblasts. J. Biol. Chem.. 265: 18753-

18756. 1990.10. Jones. A. G.. Abrams. M. J.. Davison, A.. Brodack. J. W., Toothaker. A. K.. Adelstein.

S. J., and Kassis, A. I. Biological studies of a new class of technetium complexes: thehexakis(alkylisonitrile) technetium(l) cations. Int. J. Nucí.Med. Biol.. //. 225-234,

1984.11. Kronauge, J. F.. Kawamura. M.. Lepisto, E.. Holman. B. L, Davison. A., Jones, A. G..

Costello, C. E., and Zeng, C-H. Metabolic studies of the myocardial perfusion agentTc-(MIBl). In: M. Nicolini, G. Bandoli, and U. Mazzi (eds.), Technetium and Rhenium in Chemistry and Nuclear Medicine, pp. 677-682. Verona. Italy: Cortina Inter

national. 1990.12. Piwnica-Worms. D., Kronauge. J. F., and Chiù,M. L. Uptake and retention of hexakis

(2-methoxy isobutyl isonitrile) technetium(l) in cultured chick myocardial cells:mitochondrial and plasma membrane potential dependence. Circulation, 82: 1826-18.18, 1990.

13. Piwnica-Worms. D.. Kronauge. J. F.. and Chiù,M. L. Enhancement by tetraphenylb-orate of WmTc-MIBI uptake kinetics and accumulation in cultured chick heart cells.J. Nucí.Med.. 32: 1992-1999, 1991.

14. Waggoner. A. S. Dye indicators of membrane potential. Annu. Rev. Biophys. Bioeng.,8: 47-68, 1979.

15. Richie. R. J. A critical assessment of the use of lipophilic cations as membranepotential probes. Prog. Biophys. Mol. Bio]., 43: 1-32. 1984.

16. Wacker. F. J. T.. Berman. D. S.. Maddahi, J.. Watson, D. D., Beller. G. A.. Strauss. H.W.. Boucher, C. A., Picard, M., Holman, B. L., and Fridrich, R. Tc-99m-hexakis2-methoxy isobutylisonitrile: human biodistribution, dosimetry. safety, and preliminary comparison to thallium-201 for myocardial perfusion imaging. J. Nucí.Med.. 30:301-311, 1989.

17. Hassan, I. M.. Sahweil. A., Constantinides. C., Mahmoud, A.. Nair, M., Omar, Y. T,and Abdel-Dayem. H. M. Uptake and kinetics of Tc-99m hexakis 2-methoxy isobutylisonitrile in benign and malignant lesions of the lungs. Clin. Nucí.Med., 14: 333-340,

1989.18. Piwnica-Worms. D.. and Holman, B. L. Noncardiac applications of hexakis (alkyl-

isonitrile) technetium-99m complexes. J. Nucí.Med., 31: 1166-1167, 1990.19. Caner, B.. Kitapel. M., Unlu. M.. Erbengi, G.. Calikoglu, T, Gogus. T, and Bekdik,

C. Tc-99m-MIBI uptake in benign and malignant bone lesions: a comparative studywith Tc-99m-MDP. J. Nucí.Med., 33: 319-324, 1992.

20. Delmon-Moingeon, L. I., Piwnica-Worms, D., Van den Abbeele, A. D., Holman. B. L.,Davison. A., and Jones. A. G. Uptake of the cation hexakis(2-methoxyisobutylisoni-trilc)-tcchnetium-99m by human carcinoma cell lines in vitro. Cancer Res., 50:2198-2202, 1990.

21. Howell, N., Belli. T. A.. Zaczkiewics, L. T, and Belli, J. A. High-level, unstableAdriamycin resistance in Chinese hamster mutant cell line with double minute chromosomes. Cancer Res.. 44: 4023-Õ029, 1984.

22. Mazzanti, R., Gatmaitan, Z.. Croop, J., Shu, H., and Arias, I. Quantitative imageanalysis of rhodamine 123 transport by Adriamycin-sensitive and resistant NIH 3T3and human hepatocellular carcinoma (Alexander! cells. J. Cell Pharmacol., /: 50-56,

1990.23. Kletzien. R. F.. Panza. M. W., Becker, J. E., and Potter, V. R. A method using

3-O-methyl-D-glucose and phloretin for the determination of intracellular water spaceof cells in monolayer culture. Anal. Biochem.. 68: 537-544. 1975.

24. Chiù, M. L., Kronauge. J. F.. and Piwnica-Worms, D. Effect of mitochondrial andplasma membrane potentials on accumulation of hexakis(2-methoxyisobutylisoni-trile) technetium(I) in cultured mouse fibroblasts. J. Nucí.Med.. 31: 1646-1653,1990.

25. Goldstein. L. J.. Galski, H., Fojo, H., Willingham. M.. Lai, S. L., Gazdar, A.. Pirker.R., Green, A., Crist, W.. and Brodeur, G. M. Expression of a multidrug resistance genein human cancers. J. Nail. Cancer Inst., 81: 116-124, 1989.

26. Greenberger, L. M., Yang. C. P.. Gindin, E., and Horwitz, S. B. Photoaffinity probesfor the Ofi-adrenergic receptor and the calcium channel bind to a common domain inP-glycoprotein. J. Biol. Chem.. 265: 4394-^401, 1990.

27. Devault, A., and Gros, P. Two members of the mouse mar gene family confermultidrug resistance with overlapping but distinct drug specificities. Mol. Cell. Biol.,10: 1652-1663. 1990.

28. Shen, D. W., Fojo, A., Chin. J. E.. Roninson, I. B.. Richer!, N., Pastan, I., andGottesman. M. M. Human multidrug-resistant cell lines: increased mdrl expressioncan precede gene amplification. Science (Washington DC), 232: 643-645, 1986.

29. Strauss. H. W.. and Boucher, C. A. Myocardial perfusion studies: lessons from adecade of clinical use. Radiology, 760: 577-584, 1986.

30. Pastan, 1., and Gottesman. M. M. Multiple-drug resistance in human cancer. N. Engl.J. Med., 316: 1388-1393, 1987.

31 Wittenberg, B. A., and Wittenberg. J. B. Oxygen pressure gradients in isolated cardiacmyocytes. J. Biol. Chem.. 260: 6548-6554, 1985.

32. Fojo, A. T, Ueda, K., Slamon, D. J.. Poplack. D. G., Gottesman. M. M.. and Pastan.I. Expression of a multidrug-resistance gene in human tumors and tissues. Proc. Nati.Acad. Sci. USA, 84: 265-269, 1987.

33. Ross, D. D., Thompson, B. W.. Ordonez. J. V, and Joneckis, C. C. Improvement offlow-cytometric detection of multidrug-resistant cells by cell-volume normalizationof intracellular daunorubicin content. Cytometry, 10: 185-191, 1989.

34. Lampidis. T. J., Castello, C., del Giglio, A.. Pressman. B. C.. Viallel, P., Trevorrow.K. W., Valet, G. K., Tapiero, H.. and Savaraj. N. Relevance of the chemical charge ofrhodamine dyes to multiple drug resistance. Biochem. Pharmacol., 38: 4267—4271.1989.

35. Dalton, W. S., Grogan, T. M., Rybski, J. A.. Scheper, R. J.. Richter, L., Kailey. J.,Broxterman. H. J.. Pinedo, H. M., and Salmon, S. E. Immunohistochemical detectionand quantitation of P-glycoprotein in multiple drug-resistant human myeloma cells:association with level of drug resistance and drug accumulation. Blood, 73: 747-752.1989.

36. Chan, H., Thomer. P.. Haddad, G., and Ling, V. Immunohistochemical detection ofP-glycoprotein: prognostic correlation in soft tissue sarcoma of childhood. J. Clin.Oncol.. 8: 689-704, 1990.

37. Van der Valk, P., van Kalken. C. K.. Ketelaars. H., Boxterman, H. J., Scheffer. G.,Kuiper. C. M.. Tsuruo, T. Lankelma. J.. Meijer. C. J., Pinedo, H., and Scheper, R.Distribution of multi-drug resistance-associated P-glycoprotein in normal and neo-plastic human tissues. Annal. Oncol.. /: 56-64. 1990.

38. Kessel, D., Beck. W. T.. Kukuruga. D.. and Schulz. V Characterization of multidrugresistance by fluorescent dyes. Cancer Res., 51: 4665^670, 1991.

39. Noonan, K. E., Beck, C.. Holzmayer, T. A.. Chin. J. E., Wunder, J. S., Andrulis, I. L.,Gazdar, A. F.. Willman, C. L., Griffith, B., and Von Hoff. D. Quantitative analysis ofMDR1 (multidrug resistance) gene expression in human tumors by polymerase chainreaction. Proc. Nati. Acad. Sci. USA, 87: 7160-7164. 1990.

40. Roepe, P. D., Carlson, D., Scott, H., and Wei, L-Y. Evidence that MDR P-glycoproteinis not an active drug transporter, but indirectly alters steady-state level of chemother-apeutic drug efflux by altering intracellular pH. J. Gen. Physiol., 100: 52a, 1992.

41. Gros, P., Talbol, F., Tang-Wai. D., Bibi. E.. and Kaback. H. R. Lipophilic cations: agroup of model substrates for the multidrug-resistance transporter. Biochemistry, 31:1992-1998. 1992.

42. Deutsch, E., Bushong, W., and Claven, K. A. Heart imaging with cationic complexesof technetium. Science (Washington DC), 214: 85-86, 1981.

43. Carvalho, P. A., Chiù,M. L., Kronauge. J. F.. Kawamura, M., Jones. A. G.. Holman,B. L., and Piwnica-Worms, D. Subcellular distribution and analysis of technetium-99m-MIBI in isolated perfused rat heart. J. Nucí.Med., 33: 1516-1521, 1992.

44. McCall. D.. Zimmer, L. J.. and Katz. A. M. Kinetics of thallium exchange in culturedrat myocardial cells. Circ. Res., 56: 370-376, 1985.

45. Waxman, A. D.. Ramanna. L., Memsic. L. D.. Foster, C. E., Silberman. A. W.,Gleischman. S. H., Brenner. R. J.. Brachman. M. B., Kuhar, C. J.. and Yadegar, J.Thallium scintigraphy in the evaluation of mass abnormalities of the breast. J. Nucí.Med.. 34: 18-23, 1993.

46. Kim. K. T. Black, K. L., Marciano, D., Mazziotta, J. C.. Guze. B. H.. Grafton. S.,Hawkins, R. A., and Becker. D. P. Thallium-201 SPECT imaging of brain tumors:methods and results. J. Nucí.Med.. 31: 965-969, 1990.

47. Kallinowski, F, Wilkerson, R., Moore, R., Strauss, W., and Vaupel. P. Vascularity,perfusion rate and local tissue oxygénationof tumors derived from ra.t-transformedfibroblasts. Int. J. Cancer. 48: 121-127, 1991.

48. Aklolun. C., Bayhan. H.. and Kir, M. Clinical experience with Tc-99m MIB1 imagingin patients with malignant tumors. Preliminary results and comparison with Tl-201.Clin. Nucí.Med.. 17: 171-176, 1992.

49. Caluser, C., Macapinlac. H.. Healey, J., Ghavimi, F., Meyers, P., Wollner, N.,Kalaigian, J.. Kostakoglu, L., Abdel-Dayem, H. M.. and Yeh, S. D. The relationshipbetween thallium uptake, blood flow, and blood pool activity in bone and soft tissuetumors. Clin. Nucí.Med., 17: 565-572, 1992.

984

Research. on September 10, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

1993;53:977-984. Cancer Res   David Piwnica-Worms, Mary L. Chiu, Mark Budding, et al.   an Organotechnetium ComplexFunctional Imaging of Multidrug-resistant P-Glycoprotein with

  Updated version

  http://cancerres.aacrjournals.org/content/53/5/977

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/53/5/977To request permission to re-use all or part of this article, use this link

Research. on September 10, 2020. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from