Characterization of 8 Opioid Receptors in Lung Cancer Using a Novel Nonpeptidic Ligand1

Home , BW373U86, Naltriben

[CANCER RESEARCH 56. 1965-1701. April I. 19%] Characterization of 8 Opioid Receptors in Lung Cancer Using a Novel Nonpeptidic Ligand1

Michael J. Campa, Gilbert Schreiber, Gerold Bepler, Michael J. Bishop, Robert W. McNutt, Kwen-Jen Chang, and Edward F. Patz, Jr.2

Department of Radioing)- Â¡M.J. C., C. S., E. F. P.I, Division of Pulmonary Medicine 1C. S.I. Department of Medicine [C. B.], and Department of Anesthesiology IK-J. CJ. Duke University Medical Center. Durham. North Carolina 27710: and Division of Organic Chemistry. Burroughs Wellcome Co.. Research Triangle Park, North Carolina 27709 Â¡M.J. B., R. W. M.]

ABSTRACT proach to lung cancer detection and staging that exploits fundamental biochemical differences between normal and neoplastic cells. The Cancer cells are often characterized by the presence of membrane most commonly used radiopharmaceutical. FDG, takes advantage of receptors not normally associated with nontransformed cells from the increased glucose uptake and metabolism by transformed cells (8, 9). same tissue type. Recent studies have demonstrated increased expression Once FDG is inside the cell, its efflux is minimized by phosphoryl- of high-affinity binding sites for opioid receptor-selective ligands in lung cancer cell lines relative to normal lung tissue. We investigated the binding ation by hexokinase. and further glucose metabolism is prevented by of a nonpeptidic <>opioid receptor ligand in small cell lung cancer (SCLC) the absence of the hydroxyl moiety at the C2 position. The overall and non-small cell lung cancer (NSCLC) cells with the aim of developing effect is that FDG is trapped within the cell (10). the ligand as a novel lung cancer imaging agent. The ligand, I'll| (+ )-4- Although tumor tissue exhibits levels of glucose utilization and, |(a-Ã„)-a-((2S,5Ã„)-4-allyl-2,5-dimethyl-l-piperazinyl)-3-hydroxybenzyl)- NJV- diethylbenzamide (['H]( + )BW373U86), bound with high-affinity [Kd hence. FDG trapping sufficiently high to allow accumulation of FDG far surpassing that of the surrounding normal tissue, some benign (dissociation constant) = 0.066 Â±0.012 nM] to membranes prepared from lesions, such as active tuberculosis, histoplasmosis, and sarcoidosis, six different SCLC cell lines but not to those from seven NSCLC cell lines, including one mesothelioma. The number of binding sites varied from 10 can also show elevated glucose uptake (11). In a study recently to 300 fmol/mg membrane protein. Competition binding studies demon conducted at Duke University Medical Center, we evaluated FDG- strated displacement of [JH]( + )BW373U86 binding by the 5-selective PET imaging in patients with a variety of focal pulmonary abnormal antagonists naltriben and 7-benzylidenenaItrexone but not with the Â¡L- ities and found FDG-PET to be highly sensitive for bronchogenic and K-selective antagonists D-Phe-Cys-Tyr-D-Trp-Orn-Thr-Pen-Thr-NH2 carcinoma (11). However. â€”¿11%ofthe benign lesions in the study andfrans-(Â±)-3,4-dichloro-A'-methyl-A42-(l-pyrrolidinyl)cyclohexyl]benzeneacetamide methanesulfonate. Mean apparent A',s for naltriben and were falsely labeled as malignant by PET. This lack of specificity led us to investigate additional common 7-benzylidenenaltrexone in membranes from two SCLC cell lines were features of lung cancer cells that are not characteristic of normal lung 0.17 and 3.9 nM, respectively, but were >10 /UMfor the ft and K ligands. The nonselective antagonist naloxone displaced [3H]( + )BW373U86 bind tissue. We focused our effort on abnormally expressed cell membrane ing with an apparent KÂ¡ofâ€”¿29nM.On the basis of these data, we believe receptors to determine the feasibility of using a radiolabeled receptor the lung cancer receptor to be similar, if not identical, to the human brain ligand as a lung cancer imaging agent. Ã±opioid receptor. The lack of high-affinity [3H]( + 1BW373U86 binding in Within the past several years, opioid receptors have surfaced as a normal mouse lung membranes suggests a potential role for this ligand as potential focal point for novel lung cancer treatment modalities (12). a novel therapeutic or imaging agent. It is conceivable that these receptors will be useful in alternative imaging modalities as well. Early evidence for the presence of opioid receptors in lung cancer cells was provided by the demonstration of INTRODUCTION high-affinity [Kd (dissociation constant) = 1 nM] saturable binding of Bronchogenic carcinoma is a major health problem and a financial the opioid agonist [3H]etorphine in membranes prepared from the burden to the healthcare system. Accurate diagnosis and staging are SCLC cell lines NCI-H146 and NCI-H187 (13). More recently, high- essential for therapeutic decisions and prognostic information. Devel affinity binding sites for opioid receptor ligands, ranging in density opment of a noninvasive diagnostic technique capable of detecting from 7 to 450 fmol binding sites/mg protein, were found in a variety and locating lung cancer precisely, particularly at an early stage, of SCLC and NSCLC cell lines (12). Studies with ligands selective for would have great clinical and economic impact. the various opioid receptor types demonstrated the presence of p. and Current noninvasive modalities, including conventional chest radi 8 opioid receptors in both SCLC and NSCLC cell lines: SCLC cells ography, computed tomography, and magnetic resonance imaging, possess K opioid receptors as well (12). cannot accurately distinguish benign from malignant pulmonary, pleu Given the apparent ubiquity and abundance of opioid receptors in ral, or lymph node abnormalities (1-7). A new imaging technique lung cancer cells, we chose to investigate the development of a highly applied to thoracic diseases. PET,3 is a promising noninvasive ap- selective lung cancer imaging agent based on this family of receptors. The ligand on which we have focused our investigation is a recently Received 10/30/95: accepted 1/25/96. described nonpeptidic ligand that is selective for S opioid receptors, The costs of publication of this article were defrayed in part by the payment of page the benzhydrylpiperazine BW373U86 (14). A tritium-labeled version charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. of racemic BW373U86 was prepared with two tritium atoms in the 1Tritiation for this study was performed at the National Tritium Labeling Facility. allyl side chain, and the active (+ ) enantiomer was isolated on a chiral Lawrence Berkeley Laboratory. University of California, which is supported by NIH National Center for Research Resources Grant RR01237. high-performance liquid chromatography column. In the study re 2 To whom requests for reprints should be addressed, at Duke University Medical ported here, we investigated the binding of [3H]( + )BW373U86 in Center. Box 3808. Durham. NC 27710. ^ The abbreviations used are: PET. positron emission tomography; ( + JBW373U86. lung cancer cells. The possibility of using this receptor/ligand pair as (+)-4-[(a-R)-a-((25.5R)-4-allyl-2.5-dimethyl-l-piperazinyl)-3-hydroxybenzyl)-/VJV- a target for novel imaging modalities is discussed. diethylbenzamide; FDG. | '*F]-2-fluoro-2-deoxy D-glucose; SCLC. small cell lung cancer; NSCLC. non-small cell lung cancer; BNTX. 7-benzylidenenaltrexone; NTB. naltriben methanesulfonate: CTOP. t>Phe-Cys-Tyr-B-Trp-Orn-Thr-Pen-Thr-NH2; U-50,488, trans- sulfonate; RBM. rat brain membrane; fluoropropyl-( + )BW373U86, 4-((a-/?)-a-((2S.5/f )- (Â±)-3,4-dichloro-W-methyl-A'-[2-(l-pyrrolidinyl)cyclohexyl]benzeneacetamide methane- 2.5-dimethyl-4-( 3-fluoropropyl )-1-piperazinyl)-3-hydroxybenzyl)-W,'V-diethylbenzamide. 1695

0.010

Fig. I. Determination of ['H]( + )BW373U86 0.008^ specific activity. Competition binding experiments in RBMs were carried out using a diffÃ©rentcon- cenlration of |'H|( + )BW373U86 for each of four experiments. Each reaction tube contained 0.006 - 1 ml total volume. /Cs,,s for nonradiolubeled ( + 1BW373U86 were determined from Dixon plots of the data (larne crtiph). The units of the X-axis 1/B (Â¿)are nM <+ 1BW373U86 and that of the X-axis ( l/ÃŸ)are I/specific cpm. Specifically bound cpms 0.004 J were calculated by subtracting the cpm bound in the presence of 1 Â¡Ã•M(+ )BW373U86 from those hound in its absence. Insel, plot of the cpin/ml I 'H |<+ iBW.17.1UK6used for eachof the four com petition binding experiments ivrvu.v the /f'Vis (in 0.002 DM) obtained from the Dixon plots. The specific activity of the |'H|( + )BW.17.1U86 is calculated 2000 6000 10000 troni the slope o! the regression line. cpm/ml 0.000 i.O 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1

MATERIALS ANI) METHODS jug/ml each aprotinin and leupeptin. The paniculate fraction was pelleted by eentrifugation at 4(),()(K) x Â¡>for 30 min al 4"C and resuspended by homoge- Cell Lines. The SCLC cell lines used were SCLC-22H. SCLC-16HC, ni/ation in homogeni/ation buffer containing 5 IHM Tris-HCI (pH 7.5). After a SCLC-86MI, SCLC-24H. DMS-79, and SW2I0.5; the NSCLC cell lines were 30-min incubation on ice, the suspension was homogeni/.ed using an Ultra EPLC-32MI. ADLC-5M2, NCI-H23, NCI-HI355, NCI-H324. LCLC-I03H, Tttrrax tissue disperser (IKA-Works, Inc., Cincinnati. OH); unbroken cells and ami the mesothclioma cell line MSTO-21 IH. The establishment and specific debris were removed by centrifugalion at 5(X) x g for 10 min at 4Â°C. characteristics of these cell lines have been described (15). All cell lines were maintained at 37Â°Cin medium consisting of 90% RPMI 1640. 10% fetal Membranes remaining in the supernatant fraction were recovered by cenlril- ugalion at 40.(MX) x % as before, resuspended in homogeni/.ation buffer, and bovine serum. 25 unils penicillin/ml, 25 ^ig streptomycin/ml, and K) IHM stored at â€”¿70Â°Cuntil use. Protein contenÃ was estimated by a dye-binding HKPKS (pH 7.4) in a humidified atmosphere of 95% air/5%- CO,. They were assay (Bio-Rad, Richmond, CA) using BSA as a standard. free of mycoplasma contamination. Separation of ['II](Â±)B\V.V7311H6 Knuntiomvrs. |'M|(Â±)BW373U86 Mcmhranc-ltindini; Assays. Binding of [ 'M|( + 1BW373UN6 in lung can cer cells was investigated by incubating cell membranes ( I()()-2(M) /xg protein/ was prepared by reduction of the propargyl analogue of ( Â±1BW373U86 under tube) in the presence of |'H|( + )BW373U86, followed by the separation of tritium gas. Tritialion was performed at the National Tritium Labeling Facility (Lawrence Berkeley Laboratory. University of California). A complete de hound from free ligand by vacuum filtration through Whatman CiF/C glass scription of the synthesis of ('H|( Â±(BW373U86 will be published elsewhere.4 fiber tillers (Whatman Paper Ltd., Kent, Kngland) and three washes of 5 ml The active ( + ) enantiomer was purified by high-performance liquid chroma- each of ice-cold 50 mM Tris-HCI (pH 7.5). Filtration was accomplished using tography using a Cyclobond I column (Advanced Separation Technologies, a Brandel cell harvester (model M-4K; (Â¡ailhershurg, MD). All binding assays Whippany. NJ) and isocratic elulion with 35% methanol:65% 0.1 M ammo were carried out in a total volume of I ml homogeni/ation buffer supplemented with 2 mg/ml BSA. |'H|(+ IBW373U86 and various nonradiolaheled ligands nium acÃ©tale(v/v). Active fractions were identified by binding to RBM. a tissue possessing high levels of the S opioid receptor, as described below. were present at the concentration! indicated in each experiment. Specific Specific Activity Determination. Specific activity was determined based binding was determined by subtracting the bound cpm measured in the pres on the method of Blanchard ci Â¡il.(Idi. Binding experiments were performed ence of I fiM NTB from lhat in its absence. Using this assay, binding using RBM to determine the level of the Iritialed compound required to equilibrium was achieved after 90 min at room temperature. Therefore, all saturate the binding sites in KM) /u.g RBM protein. Competition binding incubations were carried out for 2 h at room temperature to ensure that experiments were then carried out using nonradiolabeled ( + 1BW373U86 to equilibrium was reached. Radioactivity associated with the membranes was displace the binding of rH|( + )BW373U86 in RBM using four different suhsaturating concentrations of the triliated ligand ( 17). /C',,,s were determined quanlitated by liquid scintillation spectrophotometry. Kluoropropyl Analogue of ( + IBW373U86. We prepared a lluoropropyl from linear regression analysis of Dixon plots [ \/li -- !//<â€ž+\KBâ€žâ€¢¿ICW)â€¢¿L, analogue of the compound to assess (he effects of fluorinalion ( "*F is (he most where // and /<<,arethe quantities of bound ligand at ligand concentrations /, commonly used positron emitter for imaging agents) on the affinity of and /ero ligand. respectively| for each displacement curve. The /C\,,s were ( + IBW373U86 for its binding site. ( + )BW373U86 was de-allylated using a then plotted against the cpm/tuhc used for each displacement experiment, and Pd(0) catalyst and 2-mercaploben/.oic acid according to the method of the specific activity was determined from the slope of the regression line. Lemaire-Audoire el al. ( 18) to give 4-((Â«A)-Â«-((2.V.5R )-2,5-dimethyl-1 -pipera- Isolation of Cell Membranes. Cultured SCLC cells, which grow as float /.inyl)-3-hydroxyhen/yl)-A/jV-diethylhcn/amide. Before alkylating the pipcra- ing aggregates, were harvested by centrifugalion. The NSCLC cells were /.ine nitrogen, the phenol was protected as the fc/Y-butyldimethylsilyl ether, detached from the surface of the culture flask in PBS supplemented with 1 mM KDTA, transferred to 50-ml tubes, and pelleted by centril'ugation. Cell pellets, using r-butylchlorodimelhylsilane and imida/ole in /V./V-dimcthylformamidc. Fluoroalkylalion was then achieved with a slight excess of l-bromo-3-fluoro- or freshly dissected rat brain (as a positive control) or mouse lung, were DonnÃ©ehomogcni/.edin ice-cold homogeni/alion buffer consisting of 50 HIM propane and sodium carbonate in telrahydroturan at reflux overnight. The Tris-HCI (pH 7.5). l IHM KDTA, 50 /Â¿g/mlsoybean trypsin inhibitor, and 10 i-butyldimethylsilyl protecting group was removed by treatment with tetraelh- ylammonium fluoride in acetonilrile at room temperature for 30 min. The desired product was purified via silica gel chromatography with 5'/i ethanol in 4 M. J. Cumpa. K-J. Chang, R. W. McNuli. J. A. Hill, and K. K. Pat/. The character istics of the binding of the nonpeptidic Aagonist |'ll|( l )BW.T7.1llK6ti>A-opioid receptors dichloromcthanc and characlcri/ed by 'H nuclear magnetic resonance and in rat brain membranes, manuscript in preparation. mass spectrometry. 1696

100 22H 16HC 80-

600

10 30 20 60 100 1.0 1.5 2.0 0.5 1.0 1.5 2.0 2.5

300 SW210.5 O) 200- T3 4000 C 800- co 2000 - y 100 400- 'o (D CL 0 100 200 300 0 40 80 0.0 0.5 1.0 1.5 2.0 1.0 1.5 2.0

24H

800-

400-

20 40 60 80 1.0 1.5 2.0 0.0 0.5 1.0 1.5 2.0

nM [3H](+)BW373U86

Fig. 2. Saturation binding studies in SCLC cell lines. Saturation binding studies were carried out as described in "Materials and Methods" using membranes isolated from the SCLC cell lines shown above. Insel. Scatchard plots of the data from each of the larger graphs; X-axes, bound |3H]( + )BW373U86 (fmol/mg membrane protein); and K-axes. bound:free |-HJ( + )BW373U86 (fmol/mg membrane protein/nM). The amount of nonsaturable binding was similar in the membranes from all SCLC cell lines tested.

1697

Table 1 Summary of Â¡3HI(+)BW373US6 binding in SCLC cell membranes following competing ligands were chosen based on their selectivity Apparent Kd and ÃŸmaxvalueswere calculated for [-'H]( + )BW373U86 specific binding for each of the opioid receptor subtypes: BNTX (8), NTB (S), CTOP via Scatchard analysis. Each value listed in the table is the mean of two determinations using independent membrane preparations. The mean and SD of the Kds for all cell lines (/LI),naloxone (primarily /u.but nonselective at higher concentrations), are shown at the bottom of the table. and U-50,488 (K) (19, 21, and 22). The /A- and Â«-selective ligands CTOP and U-50,488 were unable to displace [3H](+)BW373U86 Cellline22HI6HC86M124HDMS-79SW2I0.5Average(nM)0.0630.0680.0840.0580.0490.0720.066(0.012)Smax46.8142.588.677.610.0300.0 even at a concentration of 10 /IM, whereas the o-selective ligands BNTX and NTB both displayed high affinity for the [3H]( + )BW373U86 binding site. The intermediate apparent K, displayed by naloxone is consistent with its broad selectivity. The a and l-(l-phenylcyclohexyl)piperidine receptor ligands ( + )-/V-allyl- (SD)Kd normetazocine hydrochloride and (5/f,105)-( + )-5-methyl-10,l 1-di- max= fmol binding sites/mg protein. hydro-5//-dibenzo[a,d]cyclohepten-5,10-imine hydrogen malÃ©atealso failed to compete for [3H]( + )BW373U86 binding (data not shown), indicating a lack of contribution to [3H]( + )BW373U86 binding by Table 2 Competition binding studies in the cell lines SCLC-I6HC and SCLC-22H these sites. Other investigators have shown both sites to be present on Competition binding assays used increasing concentrations of the nonradiolabeled compounds listed above to assess the selectivity of the l'H]( + )BW373U86 binding site the surface of SCLC cells (23). in lung cancer cells for various opioid ligands. K^ were calculated from the /C5(,s using In contrast to a previously published report demonstrating the the Cheng-Prusoff equation (36). The /C5os were estimated from the displacement curves. presence of Ã±opioid receptors in NSCLC cells ( 12). we did not detect NLX, naloxone. any high-affinity, NTB-sensitive [1H|( + )BW373U86 binding in (nM)22H0.21 Competing membranes isolated from the NSCLC cell lines EPLC-32M1, ADLC- ligandNTB 5M2, NCI-H23, NC1-H324, NCI-HI355, and LCLC-103H or from the mesothelioma cell line MSTO-211H (data not shown). Experi BNTX 4.4 ments carried out in NCI-HI355 and NCI-H23 membranes using up to NLX 24 34 > 10,000" > 10,000" 500 nM [3H]( + )BW373U86 and nonradiolabeled ( + )BW373U86 for CTOP MKI6HC0.123.4 U-50.488SelectivityÂ«28,H.fÃ¬.K > 10,000"K, > 10,000" " Less than 40% inhibition of binding was observed with IO JJ.Mof these ligands.

120 RESULTS

Specific Activity Determination. Separation of the enantiomers from the racemic mixture [3H](Â±)BW373U86 yielded a single peak exhibiting saturable binding to RBMs (data not shown). Data from competition binding assays using nonradiolabeled (+ )BW373U86 against four different subsaturating concentrations of the tritiated compound yielded linear Dixon plots (Fig. 1). The specific activity of [3H]( + )BW373U86 was calculated by plotting the /C50s obtained from the Dixon plots against the amount of tritiated ligand used in each experiment. This value, 48.4 cpm/fmol, was utilized for calcu .001 1 10 100 1000 10000 lating the concentration of [3H]( + )BW373U86 in all subsequent nM Competitor experiments. Fig. 3. Displacement of [3H]( + )BW373U86 binding in SCLC-I6HC membranes by Binding of [JH](+)BW373U86 in SCLC Cell Membranes. antagonists exhibiting selectivity for o, /i, and K opioid receptors. Binding of [3H]( + )BW373U86 to membranes isolated from SCLC-16HC was measured in the Binding studies using nonradiolabeled ( + )BW373U86 to determine absence or presence of the ligands shown in the figure. The data are expressed as the specific binding were characterized by incomplete saturation and percentage of specific binding observed in the absence of competing ligand. Specific binding was determined using 1 JAMNTB as described in "Materials and Methods." biphasic Scatchard plots. This suggested the presence of two distinct binding sites: a high-affinity/low-capacity site and a low-affinity/ high-capacity site (data not shown). Binding of [3H]( + )BW373U86 to the high-affinity site could be blocked with the highly selective 8 20 o DMS79 opioid receptor antagonist NTB (19), whereas the low-affinity site was much less sensitive; 10 /AMof the antagonist resulted in only a is â€¢¿lung 20% decrease in binding. This implies that the high-affinity site was either identical to the brain 0 opioid receptor characterized previously le (20) or a closely related variant. We therefore used 1 /AMnonradio labeled NTB to determine specific binding in all subsequent experi ments to examine this high-affinity S-like site without interference from the low-affinity site. Binding experiments in multiple SCLC cell lines consistently showed high-affinity, saturable binding with receptor density ranging -5 from 10 to 300 fmol binding sites/mg protein (Fig. 2, Table 1). The similarity in binding affinities of [3H]( + )BW373U86 suggested the 0.0 0.5 1.0 1.5 2.0 nM [3H](+)BW373U86 presence of the same receptor in each of the cell lines. We therefore Fig. 4. Binding of [â€¢1H](+)BW373U86 in mouse lung membranes. Saturation binding examined the ability of various nonradiolabeled opioid ligands to experiments were carried out in membranes isolated from healthy mice, and specific displace [3H](+)BW373U86 binding in two of these cell lines (Table binding was determined using I U.MNTB, as described in "Materials and Methods. The 2). The data for the cell line SCLC-16HC are shown in Fig. 3. The binding data from the SCLC cell line DMS-79 are shown for comparison. 1698

Ã‡ Fig. 5. Displacement of |'H|( +(BW373U86 binding hy fluoropro- T3 pyl-( + )BW373U86. Binding of ['H]( + )BW373U86 lo membranes iso CO lated from SCLC-16HC was measured in the absence or presence of increasing concentrations of fluoropropyl-( + (BW373U86. The data are u expressed as the percentage of specific binding observed in the absence o of competing ligand and are the means of the data from three experi ments: bars. SD. Specific binding was determined using l /UMNTB as described in "Materials and Methods." Inset, structure of fluoropropyl- to ( + 1BW373U86.

.01 .1 1 10 100 1000 10000 nM Fluoropropyl-(+)BW373U86 determination of specific binding provided evidence for low-affinity lung cancer cell-specific membrane receptors as targets for imaging. binding only. The present study focused on the binding of the highly selective S Binding of [^HK+JBWSTSUSoinMouse Lung Membranes. We opioid receptor agonist (+ )BW373U86 to membranes of SCLC cells. carried out saturation binding experiments in freshly isolated normal The pharmacological effects of opioids, typified by the alkaloid mouse lung membranes to determine whether the [3H](+ )BW373U86 morphine, are mediated through interaction with specific membrane binding site is expressed in normal lung tissue. High-affinity binding receptors (27). The existence of three distinct types of opioid receptor, was essentially undetectable in this tissue. For comparison, the referred to as p., o, and K,was established in studies carried out nearly amount of NTB-sensitive [3H](+ )BW373U86 binding in mouse lung 20 years ago (28-30). A fourth type of receptor, the cr receptor, was membranes is shown with that observed in membranes isolated from originally placed into the same family as the /i, o, and K opioid the SCLC cell line DMS-79 (Fig. 4). receptors. Additional studies have led to the acceptance of a classifi Competition Binding Studies of Fluoropropyl-( + )BW373U86. cation scheme in which the a receptor is placed in a receptor family We synthesized a fluorinated version of (+ )BW373U86 and tested its distinct from that of the classic opioid receptors (31). ability to displace the tritiated molecule to examine the possibility of Recent studies have demonstrated the presence of opioid receptors using (+ )BW373U86 as a PET imaging agent. Fluoropropyl- (+ )BW373U86 displaced [3H](+ )BW373U86 binding with an appar in lung cancer cell lines (13, 14). Although their significance is not known, the receptors appear to be functional in that incubation of lung ent K, of -0.12 Â±0.04nM(Fig. 5). This represents an affinity for the [3H](+ )BW373U86 binding site <2-fold lower than that of the un cancer cells in medium containing opioid ligands results in a decrease in the intracellular level of cAMP (13). This effect is attenuated by modified molecule determined by saturation binding studies pretreatment of the cells with pertussis toxin, implicating the involve (ATd= 0.066 Â±0.012 nM). ment of trimeric G-proteins in the response. In addition, opioid agonists including morphine, [o-Ala2, D-Leu^enkephalin, and DISCUSSION U-50,488 inhibit cell proliferation (32). Lung cancer accounts for ~25% of all cancer deaths in the United With the aim of developing a highly specific and novel lung cancer imaging agent, we characterized the binding in lung cancer cells of the States. More than 170.000 new cases of lung cancer are diagnosed nonpeptidic Ã´opioid receptor ligand [3Hj(+ )BW373U86. This com annually. In addition, the overall 5-year survival of ~14% has not changed over the past several decades (24, 25). Diagnosis of the pound is highly selective for S opioid receptors and functions as an disease depends on invasive studies such as bronchoscopy, percuta agonist in both in vitro and in vivo paradigms (15). In displacement neous needle biopsy, thoracoscopy, or open lung biopsy. These pro experiments using RBMs. the most effective competitors for binding cedures have associated risks, are expensive, and are not always to (+ )BW373U86 binding sites are antagonists of proven Ã´receptor diagnostic. Clinical scries have demonstrated that >50% of resected selectivity (15). In addition. Chang et til. (14) demonstrated that pulmonary nodules are benign: thus, a number of unnecessary proce (Â±)BW373U86could inhibit electrically induced muscle contraction dures are performed for radiologically indeterminate lesions (26). in isolated mouse vas deferens with an ED.,,, of 0.2 Â±0.02 nM. More recently, PET has emerged as a promising noninvasive means Inhibition of muscle contraction in mouse vas deferens is mediated for examining thoracic abnormalities; it provides not only anatomic primarily by 8 opioid receptors, whereas that in guinea pig ileum, a but physiological information. This is the first time that detection, tissue in which (Â±)BW373U86exhibits only weak activity, is medi staging, and follow-up of patients with bronchogenic carcinoma has ated chiefly by p. receptors (30). Furthermore, the activity of been accomplished by exploiting fundamental biochemical differ (Â±)BW373U86in mouse vas deferens can be blocked by the 5 opioid ences between normal and neoplastic cells. Although current imaging antagonist naltrindole (15). Although these studies used the racemic agents appear sensitive for malignant lesions (at least lesions >1 cm mixture of the compound, the (â€”¿)enantiomer exhibits an affinity for in diameter), they are not specific. Thus, in an attempt to improve the 8 receptor â€”¿500-foldlowerthan that of the (+ ) enantiomer (data specificity, we have begun to investigate the possibility of utilizing not shown). Therefore, the activity observed in these studies was most 1699

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1996 American Association for Cancer Research. BINDING OF AN OPIOID LIGAND IN SCLC CELLS likely that of the enantiomer used in our study. Taken together, these an imaging agent in tissue surrounding a tumor tends to erode the data attest to the high degree of S selectivity exhibited by signal:noise ratio, the large difference in binding affinity between the ( + )BW373U86. S opioid site and the low-affinity site detected in normal lung is such We detected high-affinity binding of [3H]( + )BW373U86 in mem that binding to the low-affinity site can be avoided. Utilization of branes isolated from six different SCLC cell lines. Competition bind [I8F]( + )BW373U86 in vivo at or near its apparent K, of 0.12 nM ing studies showed that the binding site also has high affinity for would minimize binding to the low-affinity site. We are currently antagonists of proven Ãópioid receptor selectivity. The relative af conducting biodistribution studies in nude mice bearing tumors of finities of BNTX and NTB are similar to those reported by Knapp et lung origin to assess the ability of [3H]( + )BW373U86 to localize to al. (21). In their study, NTB was shown to displace the binding of tumor tissue in a more physiologically relevant model. [3H]naltrindole in COS-7 cells expressing the cloned human S opioid In summary, our study demonstrates high-affinity binding of receptor with an /C50 ~ 10-fold lower than that for BNTX. [3H]( + )BW373U86 to membranes from SCLC cells. Our data Although the prevalence of the Ãópioid receptor in SCLC must strongly suggest that the binding site is identical to the 8 opioid await further testing of cultured cell lines, the fact that six out of six receptor. This liganoVreceptor pair may prove to be useful in the cell lines exhibited high-affinity [3H]( + )BW373U86 binding suggests diagnosis and staging of SCLC. that this receptor may be a universal trait of this tumor. In contrast, we did not find any high-affinity [3H]( + )BW373U86 binding in mem branes from seven different NSCLC cell lines, including one mesothe- REFERENCES lioma. The presence of the 8 receptor in NSCLC cells, however, has Daly. B. D. T.. Paling. L. J.. Pugalch. R. D.. Jung-Legg. Y.. Gale. M. E.. Bite. G., and been reported by others (13). In the experiments of Maneckjee and Snider. G. L. Computed tomography: an effective technique for mediastinal staging Minna (12), Ã´receptors were identified by displacement of [3H][o- in lung cancer. J. Thorac. Cardiovasc. Surg., 88: 486-494. 1984. Baron. R. L.. Leviti. R. G.. Sagel. S. S.. White. M. J.. Roper. C. L., and Marharger. Ala2, D-Leu5]enkephalin binding by the enkephalin analogue [o-Pen2, J. P. Computed tomography in the preoperative evaluation of hronchogenic carci D-Pen5]enkephalin. In the present study, we utilized the binding of the noma. Radiology. 145: 727-732. 1982. Brion. J. P., Depauw. L., KÃ¼hn,G., de Fracquen. P.. Friberg. J.. Rocmans. P.. and S-selective compound [3H]( + )BW373U86 and its displacement by Struyven. J. Role of computed tomography and mediastinoscopy in preoperative NTB and BNTX to identify 8 receptors. The discrepancy in binding staging of lung carcinoma. J. Comput. Assisted Tomogr.. 9: 480-484. 1985. Heelan. R. T.. Martini. N.. Westcott. J. W.. Bains, M. S.. Watson. R. C. Caravelli, data between the two laboratories could possibly be explained by a J. F.. Berkmen. Y. M., Henschke. C. I. McCormack. P. M.. and McCaughan. B. C. difference in selectivity between the two methods used. The existence Carcinomatous involvement of the hilum and mediastinum: computed tomographic and magnetic resonance evaluation. Radiology. 156: 111-115. 1985. of two subtypes of Ãópioid receptor, termed 8, and S2, has been Poon. P. Y.. Bronskill. M. J., Henkelman. R. M.. Rideout. D. F.. Shulman. H. S.. postulated based on in vitro (33) and in vivo (34) experiments. Our Weisbrod. G. L.. Steinhardt. M. I.. Dunlap. H. J.. Ginsberg. R. J., Feld, R.. Sutcliffe, finding that NTB displaces [3H]( + )BW373U86 binding in SCLC cell S. B., and Williams. J. I. Mediastinal lymph node mÃ©tastasesfrom bronchogenic carcinoma: detection with MR imaging and CT. Radiology. lf>2: 651-656. 1987. membranes more effectively than does BNTX suggests that the opioid Musset. D., Grenier. P.. Carette, M. F.. Frija, G.. Hauuy. M. P.. Desbleds. M. T., receptor in these cells is the 82 subtype. In addition, the high affinity Girard. P.. Bigot. J. M., and Lallemand. D. Primary lung cancer staging: prospective of [3H]( + )BW373U86 binding to these membranes may imply a comparative study of MR imaging with CT. Radiology. 160: 607-611, 1986. selectivity toward this subtype. [o-Pen2, D-Pen5]enkephalin, on the Staples. C. A.. MÃ¼ller.N.L., Miller. R. R.. Evans. K. G., and Nelems, B. Mediastinal nodes in bronchogenic carcinoma: comparison between CT und mediastinoscopy. other hand, has greater selectivity toward the 8, subtype. A prevalence Radiology, 167: 367-372, 1988. Warburg. O. Versuche an Ã¼berlebendemKarzinomgewebe (Methoden). Biochem. Z.. of the 8, subtype in NSCLC cells would therefore result in minimal 142: 317-333. 1923. binding of [3H]( + )BW373U86. Additional studies are needed to Flier, J. S.. Mueckler. M. M.. Usher. P.. and Lodish, H. F. Elevated levels of glucose elucidate the precise nature of the opioid receptors present in these transport and Iransporter messenger RNA are induced by ras or xrc oncogenes. Science (Washington DC). 2J5: 1492-1495, 1987. cell types. Gallagher. B. M.. Fowler. J. S.. Gutterson. N. !.. MacGregor. R. R.. Wan. C. N.. and In addition to this high-affinity binding site, we also identified a Wolf, A. P. Metabolic trapping as a principle of radiopharmaceuticul design: some low-affinity, NTB-insensitive site in SCLC, NSCLC. and normal lung factors responsible for the biodistribution of ['*F]2-deoxy-2-fluoro-i>-glucose. J. NucÃ.Med.. 19: 1154-1161, 1978. cell membranes from mouse. Displacement experiments in which the Patz, E. F.. Lowe, V. J.. Hoffman. J. M.. Paine. S. S.. Burrowes, P., Coleman. R. E., high-affinity site was masked by NTB revealed an /C50 for and Goodman. P. C. Focal pulmonary abnormalities: evaluation with F-18 fluorode- (+ )BW373U86 at the low-affinity site of -100 nM. However, at oxyglucose PET scanning. Radiology. 188: 487-490, 1993. 12. Maneckjee, R.. and Minna. J. D. Opioid and nicotine receptors affect growth regu tempts to saturate the [3H]( + )BW373U86 binding sites in two lation of human lung cancer cell lines. Proc. Nati. Acad. Sci. USA, 87: 3294-3298, NSCLC cell lines failed even at 500 nM ligand. This suggests that the 1990. true affinity of [3H]( + )BW373U86 for the low-affinity site is greater 13. Roth. K. A., and Barchas. J. D. Small cell carcinoma cell lines contain opioid peptides and receptors. Cancer (Phila.), 57: 769-773, 1986. than that estimated from the /C50. Furthermore, binding to the low- 14. Chang. K. J.. Rigdon. G. C.. Howard. J. L.. and McNutt. R. W. A novel, polenl and affinity site was insensitive to heating at 68Â°C,whereas this same selective nonpeptidic 8 opioid receptor agonist BW373U86. J. Pharmacol. Exp. Ther.. 267- 852-857. 1993. treatment completely eliminated binding to the high-affinity site (data 15. Bepler. G., and Koehler. A. Multiple chromosomal aberrations and 11p allelotyping not shown). This implies that binding to the low-affinity site may not in lung cancer cell lines. Cancer Genel. Cytogenet.. 84: 25-30. 1995. 16. Blanchard. S. G.. Lee. P. H. K.. Pugh. W. W., Hong. J. S.. and Chang. K-J. be mediated by a conventional membrane receptor. Characterization of the binding of a morphine (Â¿Â¿)receptor-specificligand: Tyr-Pro- Although low levels of [3H]etorphine binding to lung membranes NMePhe-D-Pro-NH2 [3H]-PL17. Mol. Pharmacol., 31: 326-333. 1987. have been reported (35), we were unable to detect high-affinity 17. Bishop. M. J.. and McNutt. R. W. An efficient synthesis of the hen/.hydrylpiperazine [3H](+)BW373U86 binding in membranes from normal lung. This &opioid agonist ( + 1BW373U86. Bioorg. & Med. Chem. Lett. 5: 1311-1314. 1995. 18. Lemaire-Audoire. S., Savignac. M.. and GenÃªt. J. P. Selective deprotection of allyl implies that (3H]( + )BW373U86 may be useful to differentiate normal amines using palladium. Tetrahedron Lett., 36: 1267-1270. 1995. from malignant lung tissue, a suggestion underscored by the magni 19. Portoghese. P. S.. Nagase. H.. Maloney-Huss. K. E., Lin, C. E., and Takemori, A. E. Role of spacer and address components in peptidomimetic Ãópioid receptor antag tude of the difference in [3H]( + )BW373U86 binding between normal onists related to naltrindole. J. Med. Chem.. 34: 1715-1720, 1991. lung cell membranes and those from DMS-79 cells, the SCLC cell line 20. Knapp, R. J.. Malalynska. E.. Fang. L., Li. X.. Babin. E.. Nguyen. M., Santoro, G., Varga. E. V.. Hruhy. V. J.. Roeske. W. R.. and Yamamura. H. I. Identification of a exhibiting the lowest receptor density. The feasibility of this com human Ãópioid receptor: cloning and expression. Pharmacol. Lett.. 54: 463-469. pound as an imaging agent is supported by the retention of binding 1994. affinity of fluoropropyl-( + )BW373U86 for the [3H]( + )BW373U86 Portoghese. P. S.. Sultana. M.. Nagase. H.. and Takemori. A. E. A highly selective 5, binding site, implying that [18F]fluorination would not lead to an opioid receptor antagonist: 7-benzylidenenaltrexone. Eur. J. Pharmacol.. 21H: 195- 196. 1992. unacceptable loss of binding. Although the binding or accumulation of 22. Gulya. K.. PellÃ³n. J. T.. Hruby. V. J.. and Yamamura, H. I. Cyclic somatostatin 1700

octapeptide analogues with high affinity and selectivity toward /Â¿opioid receptors. 30. Lord. J. A.. Waterfield. A. A.. Hughes. J.. and Kosterlitz. H. W. Endogenous opioid Life Sci., 38: 2221-2229, 1986. peptides: multiple agonists and receptors. Nature (Lond.l. 267: 495-499. 1977. 23. Maneckjee, R., and Minna. J. D. Biologically active MK-801 and SKF-10.047 31. Quirion. R., Chicheportiche. R.. Contreras. P. C.. Johnson, K. M.. Lodge, D.. Tarn, binding sites distinct from those in rat brain are expressed on human lung cancer cells. S. W., Woods, J. H., and Zukin. S. R. Classification and nomenclature of phencyc- Mol. Biol. Cell, 3: 613-619. lidine and a receptor sites. Trends Neurosci., 10: 444-446, 1987. 24. Osteen. R. T. Cancer Manual. Ed. 8. Boslon: American Cancer Society, 1990. 32. Maneckjee. R.. and Minna. J. D. Opioid and nicotine receptors affect growth regu 25. Boring. C. C. Squires, T. S., and Tong, T. Cancer statistics: 1992. CA Cancer J. Clin., lation of human lung cancer cell lines. Proc. Nati. Acad. Sci. USA. 87: 3294-3298. 42: 19-38, 1992. 1990. 26. Mack. M. J.. Hazelrigg. S. R.. Landreneau. R. J.. and Acuff. T. E. Thoracoscopy for 33. Bu/as. B.. Izenwasser, S.. Portoghese. P. S.. and Cox. B. M. Evidence for 6 opioid the diagnosis of the indeterminate solitary pulmonary. Ann. Thorac. Surg., 36: 825-830, 1993. receptor subtypes regulating adenylyl cyclase activity in rat brain. Life Sci.. 54: 27. Chang. K-J. Opioid receptor: multiplicity and sequelae of ligand-receptor interac 101-106, 1994. tions. In:. P. M. Conn (ed.). The Receptors. Vol. 1. pp. 1-81. Orlando, Academic 34. Sofuoglu. M., Portoghese. P. S., and Takemori. A. E. 7-Ben?ylidenenaltrexone Press, 1984. (BNTX). A selective 6, opioid receptor antagonist in the mouse spinal cord. Life Sci.. 28. Martin. W. R.. Eades, C. G.. Thompson. J. A.. Huppler, R. E.. and Gilbert. P. E. The 52: 769-775, 1993. effects of morphine- and nalorphine-like drugs in the nondependent and morphine- 35. Tang, J., Chou, J.. Zhang. A. Z.. Yang, H-YT, and Costa, E. Met5-Enkephalin-Argcyclazocine-dependent chronic spinal the concentration of inhibitor which causes 50"7f inhibition (1C,,,) of an enzymatic dog. J. Pharmacol. Exp. Then. IVX: 66-82. 1976. reaction. Biochem. Pharmacol.. 22: 3099-3108. 1973.

1701

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1996 American Association for Cancer Research. Characterization of δ Opioid Receptors in Lung Cancer Using a Novel Nonpeptidic Ligand

Michael J. Campa, Gilbert Schreiber, Gerold Bepler, et al.

Cancer Res 1996;56:1695-1701.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/56/7/1695

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/56/7/1695. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.