Sigma-I and Sigma-2 Receptors Are Expressed in a Wide Variety of Human and Rodent Tumor Cell Lines

[CANCERRESEARCH55, 408-413, January 15, 19951 Sigma-i and Sigma-2 Receptors Are Expressed in a Wide Variety of Human and Rodent Tumor Cell Lines

Bertold J. Vilner, Christy S. John, and Wayne D. Bowen' Unit on Receptor Biochemistry and Pharmacology, Laboratory of Medicinal Chemistry, National Institute of Diabetes Digestive and Kidney Diseases, National institutes of Health, Bethesda, Maryland 20892 [B. J. V.. W. D. B.], and Radiopharmaceutical Chemistry Section, Department of Radiology, The George Washington University Medical Center, Washington, DC 20037 (C. S .1.1

ABSTRACT Sigma receptors occur in at least two classes which are distinguish able both pharmacologically and by molecular properties (1, 8â€”10). Thirteen tumor-derived cell lines of human and nonhuman origin and The prototypic sigma ligands, haloperidol, DTG,2 and (+)-3-PPP do from various tissues were examined for the presence and density of not strongly differentiate the sites. However, sigma-i sites are readily sigma-i and sigma-2 receptors. Sigma-i receptors of a crude membrane fraction were labeled using [3H](+)-pentazocine, and sigma-2 receptors distinguished from sigma-2 sites on the basis of their affinity for were labeled with [3H11,3-di-o-tolylguanidine ([3HJDTG); in the presence benzomorphan-type opiates such as pentazocine and SKF 10,047. or absence ofdextrallorphan. [3H](+)-Pentazocine-bindingsites were het Sigma-i receptors exhibit high affinity for (+)-benzomorphans and erogeneous. In rodent cell lines (e.g., C6 glioma, N1E-115 neuroblastoma, lower affinity for the corresponding (â€”)-enantiomer.Sigma-2 recep and NG1OSâ€”15neuroblastoma x glioma hybrid), human T47D breast tors show the opposite enantioselectivity, having very low affinity for ductal careinoma, human NCI-H727 lung carcinold, and hwnan A375 (+)-benzomorphans. (â€”)-Benzomorphans do not readily differentiate melanoma, [3H](+)-pentazocine bound to high- and low-affinity sites with the two sites. Other differentiating features of these two sites are as Kdl 0.67-7.0 UM,B@1 = 25.5-108 fmol/mg protein, K@2= 127-600 follows: (a) the apparent molecular weight of sigma-i receptors is fiM, and B@12 = 942-5431 fmol/mg protein. However, [311](+)-pentazo 25,000,comparedto 18,000â€”21,500forsigma-2,(b) sigma-i recep cine bound to a single site in other cell lines. In human U-13SMG glib tors appear to be sensitive to guanine nucleotides, while sigma-2 blastoma, SK-N-SH neuroblastoma, and LNCaP.FGC prostate, Kd 28-61 asi andB@ = 975-1196fmol/mgprotein,whereasin ThP-1 receptors are not, and (c) sigma-i receptors are allosterically modu leukemia Kd 146 ass and B_11,, = 1411 fmol/mg protein The sigma-i-like lated by ropizine and phenytoin, while sigma-2 sites are not. Sigma-i nature of[3H](+)-pentazocine-biading sites was confirmed by competition receptors are selectively labeled using the (+)-benzomorphan, studies which revealed high affinity for haloperidol and enantloselectlvlty [3H](+)-pentazocine (1 1). No selective probe yet exists for sigma-2 for (+)-pentazocine over (â€”)-pentazocine.Interestingly,human MCF-7 sites, but they can be labeled using [3H]DTG in the presence of breast adenocarcinoma showed little or no specific binding of I@Hl(+) dextrallorphan to mask labeling of sigma-i sites (1 1, 12, 13, 14). pentazocine, suggesting the absence of sigma-i receptors in this cell line. Cell lines expressing neurotransmitter and hormone receptors All cell lines examined expressed a high density of slgma-2 receptors with have been quite useful in elucidating receptor function at the Kd values for [3H]DTG ranging from 20 to 101 nsi and B@ values of 491 biochemical and cellular levels. Sigma receptors have previously to 7324 fmol/mg protein. Competition studies indiCated possible hetero been demonstrated in tumor-derived cell lines from rodents. Some geneity of sigma-2 receptors. While sites labeled by [311]DTGin all cell lines tested exhibited affinity for haloperidol and prefrrence for (â€”)- of theseincludePC12pheochromocytomacells(8),NCB-20cells pentazocine over the (+)-enantiomer, human cell lines generally showed (15, 16), and C6 glioma (13). In fact, PC12 cells, C6 glioma, and 4-to7-foldloweraffinityforhaloperidolandapproximately10-foldlower several other rodent-derived cell lines were instrumental in our affinity for (â€”)-pentazocinecompared with the rodent cell lines. The high initial demonstration of the existence of sigma-2 receptors (8, 13). density of sigma-i and sigma 2-binding sites in these cell lines suggests Here we examine the existence of sigma receptor subtypes in important cellular functions in cancer, as well as potential diagnostic additional cell lines of nonhuman origin, as well as in human tumor utility for tumor-imaging agents which target sigma sites. These cell lines cell lines derived from various tissues and organs. may be useful as model systems In which to study the functions of sigma sites in normal tissues, as well as their possible role in tumor blologj@. MATERIALS AND METHODS

INTRODUCTION Cell CUltUre. Cells were plated at a density of 2 X 10â€•cellsf7S-cm2 plastic flask (Coming Co., Corning, NY) and cultured using standard procedures at Because of their high affinity for most typical neuroleptic drugs, 37Â°Cina humidified atmosphere of 5% CO@I95%air. Human glioblastoma sigma receptors have received much attention as potential alternative U-138 MG, melanoma A375 (amelanotic), neuroblastoma SK-N-SH, and targets for the development of antipsychotic drugs. However, their breast adenocarcinoma MCF-7 were cultured in DMEM supplemented with high affinity for other classes of compounds, including some anticon 10% FBS. Human breast ductal carcinoma T47D, lung carcinoma NCI-H727, vulsants, antitussives, and neuroprotective agents, have suggested leukemia ThP-1, and metastatic prostate adenocarcinoma LNCaP.FGC were other possible roles of sigma sites (1â€”7).Details of the functional roles cultured in RPMI with 10% FBS. Rat glioma C6 and mouse neuroblastoma of sigma receptors have remained elusive, although they have been NB41A3 were grown in RPMI 1640 medium supplemented with 10% FBS. implicated in motor function, neurotransmitter synthesis and release, NG1O8â€”15mouseneuroblastoma X rat glioma hybrid cells were grown in digestive function, regulation of smooth muscle contraction, and DMEM supplemented with 10% FBS, 1% nonessential amino acids, 1% L-glutamine, and 100 units penicillin and 100 @agstreptomycin/mi medium. neurodegeneration. Mouse neuroblastomas N1E-115 and S-20Y were grown in DMEM supple mented with 10% FBS. Media were changed 3 times a week. Received 7/19/94; accepted 11/7/94. Membrane Preparation. Cells were allowed to grow to confluency. After The costs of publication of this article were defrayed in part by the payment of page decantation of the medium, the cells were washed in situ with HBSS. The cells charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. were then detached by scraping with a cell scraper (Costar Corporation, I To whomrequestsforreprintsshouldbeaddressed,atUnit onReceptorBiochemistry Cambridge, MA) in HBSS. Suspended cells were centrifuged for 5â€”7mm at and Pharmacology, NlDDKThiboratory of Medicinal Chemistry, Bldg. 8, Rm. B1-23, 8 1000 rpm in a cold table-top centrifuge. The cell pellet was resuspended in Center Dr. MSC 0815, Bethesda, MD 20892â€”0815. ice-cold 10 mMTris-HC1,pH 7.4, containing 0.32 Msucrose to a concentration 2 The abbreviations used are: DTG, 1,3-di-o-tolylguanidine; (+)-3-PPP, (+)-3-(3- hydroxyphenyl)-N-(1-propyl)piperidine;FBS, fetal bovine serum; SKF 10,047, N-allyl of 1 X iO7-i0'@cells/mIand homogenized by 10â€”12hand-drivenstrokes in a normetazocine. Potter-Elvehjem homogenizer (Teflon pestle). The homogenate was then cen 408

trifuged at 31,000 x g for 15 mm at 4Â°Cand the supematant discarded. The RESULTS final pellet was resuspended in ice-cold 10 mMTris-HC1,pH 7.4, to a protein concentration of 15â€”20mg/mI,and aliquots were stored at â€”80Â°Cuntiluse. Tables 1 and 2 show the results of Scatchard analysis of [3H](+)- Protein concentration was determined by the method of Lowry et al (17) using pentazocine binding to sigma-i sites and [3HJDTG binding (in the BSA as standard. presence or absence of dextrallorphan) to sigma-2 sites in several Radioligand Binding. Sigma-i receptors were labeled using [3H](+)- tumor cell lines of human and rodent origin. All cell lines examined pentazocine (51.7 Ci/mmol) (11). The indicated concentration of radioligand were found to express a high density of sigma-2 receptors labeled by was incubated in a final volume of0.25 ml of 50 mM Tris-HC1, pH 8.0, for 120 [3H]DTG. In each case, plots were linear, suggesting labeling of a mm at 25 or 37Â°Cusing200 @gofmembraneprotein. Nonspecific binding was determined in the presence of 10 p.M (+)-pentazocine (a similar level of single population of sites. The Kd values for [3H]DTG were generally nonspecific binding was observed using 10 @.aMhaloperidol).Assays were similar across cell lines (20â€”46nM)and were close to those obtained terminated by addition of 5 ml of ice-cold 10 [email protected],pH 7.4, and for sigma-2 sites in PCi2 cells, rat liver, and rat kidney (8, 14), filtration through glass fiber ifiters using a Brandel (Gaithersburg, MD) cell although the Kd value in C6 glioma, NB41A3 neuroblastoma, harvester. Filters were then washed twice with 5 ml of ice-cold buffer. Filters NG1O8â€”15hybrid,andhumanU-138MG glioblastomaweresome were soaked in 0.5% polyethyleneimine for at least 30 mm at 25Â°Cpriorto use. what higher (62â€”101nM) than in the other cell lines. Bmas values The filters were counted in CytoScint (ICN, Costa Mesa, CA) after an over ranged from 491 fmol/mg protein in ThP-1 leukemia to 7324 fmol/mg night extraction of counts. Sigma-2 receptors were labeled using [3H]DTG protein in NB41A3 neuroblastoma. (39.1 Cimmmol)in the presence of 1 @iMdextrallorphan(11, 13). Nonspecffic binding was determined in the presence of 10 @Mhaloperidol. Other Cells also expressed binding sites for [3H](+)-pentazocine. How manipulations were as described above. ever, unlike the single site for [3H]DTG binding, human T47D breast Chemicals and Cell Lines. The NG1O8-15 neuroblastoma X glioma ductal carcinoma, NCI-H727 lung carcinoid, A375 melanoma, and all hybrid cell line was the gift of Dr. R. J. Weber (National Institute of Diabetes rodent cell lines tested exhibited two apparent binding sites for and Digestive and Kidney Diseases, Bethesda, MD). N1E-115 neuroblastoma [3H](+)-pentazocine. In these cells, the [3H](+)-pentazocine Scat cells were obtained from Dr. Elliott Richelson (Mayo Clinic, Jacksonville, FL). chard plots were markedly curvilinear. The Kd of the high-affinity site Neuroblastoma S-20Y was obtained from Dr. Enrique Silva (Beth Israel ranged from 0.67â€”7.0nM,whereas the low-affinity site had Kd values Hospital, Boston, MA). NCI-H727 lung carcinoid was the gift of Dr. Terry of 127â€”360nM, with NG1O8â€”15showing a low-affinity site of Kd 600 Moody (National CancerInstitute,Bethesda,MD). All othercell lines were purchased from American Type Culture Collection (Rockville, MD). nM. In all cases in which two [3H](+)-pentazocine-binding sites were [3H](+)-Pentazocine (51.7 Ci/mmol) was synthesized by B. de Costa (Na apparent, the lower affinity site was present in much higher abundance tional Institute of Diabetes and Digestive and Kidney Diseases), as described compared to the higher affinity site, ranging from 16-fold higher in previously (11). [3HIDTG (39.1 Ci/mmol) was purchased from DuPont/New human T47D breast ductal carcinoma to 95-fold higher in human EnglandNuclear(Boston,MA).Tris-HC1,polyethyleneimine,haloperidol,and A375 amelanotic melanoma. However, in most cases the total density tissue culture reagents (RPM! 1640, HBSS) were purchased from Sigma of sigma-i binding sites (high-affinity Bm,,@+ low-affinity Bmax)was Chemical Company (St. Louis, MO). DMEM was purchased from GIBCO comparable to the Bmaz of sigma-2 sites. BRL (Grand Island, NY). FBS was purchased from Advanced Biotechnolo There were exceptions to the two-site labeling of [3H](+)-penta gies, Inc. (Columbia, MD). Enantiomers of pentazocine and SKF 10,047 were provided by Dr. Kenner C. Rice (National Institute of Diabetes and Digestive zocine. MCF-7 breast adenocarcinoma exhibited little or no signifi and Kidney Diseases, Bethesda, MD). DTG was purchased from Aldrich cant specific binding of [3H](+)-pentazocine, indicating the absence Chemical Company (Milwaukee, WI). (+)-3-PPP was purchased from of sigma-i receptors. ThP-1 leukemia, U-138MG glioblastoma, SK Research Biochemicals Inc. (Natick, MA). N-SH neuroblastoma, and LNCaP.FOC prostate each exhibited a high

Table 1 Binding parameters of sigma-i and sigma-2 receptors in membranes of various human tumor cell lines Sigma-i and sigma-2 binding assays were carried out under the conditions described in â€œMaterialsandMethodsâ€•at37Â°C.[3H](+)-Pentazocineconcentration was varied in 15 concentrations over a range of 0.1â€”500nt.i.[3HJDTG(in the presence of 1 @u.idextrallorphan)was incubated in 15 concentrations at 1â€”400m.i.A combination oflabeled and unlabeled ligand was used to achieve concentrations >50 nai for [3H](+)-pentazocine and >15 nr@ifor[3H]DTG. Data were analyzed using the iterative curve-fitting program BDATA (EMF Software,Baltimore,MD).Valuesaretheaveragesof2â€”4experimentsÂ±salbi.Eachexperimentwascarriedoutduplicate.Cell in dextrallorphan)MCF-7line Sigma-i([3HJ(+)-pentazocine) Sigma-2([3H]DTG+ breast adenocarcinoma No specific binding K,, = 24.54 Â±5.57 @ 734T47D 2071 Â± breast ductal carcinoma Kdl = 6.62 Â±1.03 j@d= 19.95 Â±3.53 Bm,@,1108Â±M.6 B@,,1221Â±264 K,,2=261Â±41.48 164NCI-H727 Bm,,,,2 1690 Â± lung carcinoid K,,l = 3.81 Â±1.80 K,, 44.44 Â±1.78 B,,,1@,1=25.51 Â±7.46 B,@ = 2835 Â±467 K,,2 = 127 Â±8.15 313A375 Bmu2 2099 Â± melanoma (amelanotic) Kdl = 0.67 Â±0.23 Kd = 34.19 Â±2.18 B@j,1 33.99 Â±12.77 B,@ 3403 Â±348 Kd2= 129Â±12.05 339ThP-1 B,..@2= 3223 Â±

leukemia Kd 146 Â±52.59 K,, 39.5 Â±6.72 B,,,,,,=491Â±22U-138MG B,,,,,,,=1411Â±102 glioblastoma Kd = 60.88 Â±13.21 Kd 80.83 Â±5.85 B,,,,@,,=3136Â±871SK-N-SH B,,.@=1115Â±244 neuroblastoma Kd = 27.99 Â±5.78 K,, = 32.35 Â±3.26 104LNCaP.FGC B,@ = 975 Â±225 B@ = 944 Â± prostate Kd = 38.44 Â±17.78 Kd 39.00 Â±0.40 B,..@=1196Â±490 B,,,@=727Â±5.67

409

Table 2 Binding parameters of sigma-i and sigma-2 receptors in membranes of various rodent tumor cell lines Membranes were incubated using the conditions described in â€œMaterialsandMethodsâ€•andcaption to Table 1, at 25Â°C([3HJ(+)-pentazocine,13â€”17concentrations;[3HIDTO, 14 concentrations). Data were analyzed using the iterative curve-fitting program BDATA (EMF Software, Baltimore, MD). Values for 13HJ(+)-pentazocine are the averages of 3â€”5 experiments Â±5EM,eachcarried out in duplicate. Data for [3HJDTGwere taken from Vilner and Bowen (13) and are the averages of 3â€”5experimentsÂ±sat,each earned out in duplicate. [3HJDTGwas incubated in the absence of dextrallorphan in these experiments. However, as shown in Table 4 and discussed in the text, 1 ps. dextrallorphan had no effect on the level of specific [3HIDTG binding, anddextrallorphan.Cell [3HJDTG labeled sites with a clear sigma-2 profile in rodent-derived cell lines in the absence of ([3HIDTG)C6 line Sigma-i ([3H1(+)-Pentazocine) Sigma-2 glioma (rat) K,,l 5.7 Â±1.3 Kd 101 Â±7.0 Bm,,,,1 42.0 Â±17.4 Bm,@ 5507 Â±537 K,,2 = 287 Â±65 136NB41A3 Bm@,,2 942 Â± neuroblastoma (mouse) Kdl = 3.0 Â±1.4 Kd 62.3 Â±6.4 Bm,,@,1 76.5 Â±15.2 B@y._@7324 Â±670 Kd2=327Â±92 B@pj,2M31Â±17N1E-115 neuroblastoma (mouse) K,,! = 3.0 Â±0.3 Kd = 25.5 Â±0.4 B@1=41.5Â±fl.3 B,,@=3344Â±301 K,,2 = 360 Â±145 @ 134NG1O8-15 1678 Â±

mouse neuroblastoma X rat glioma K,,I = 7.0 Â±1.2 K,, 75.0 Â±0.7 B@pj,1 94.8 Â±1.6 B@ 3134 Â±229 Kd2=600Â±197 1687S-20Y Bmax2 4553 Â±

neuroblastoma (mouse) Not done K,, = 46.0 Â±4.9 Bmu 2456 Â±135

density of an apparent single population of [3H](+)-pentazocine the conditions of the competition assay, [3H](+)-pentazocine binding sites. Based on the Kd values, the site in ThP-1 leukemia predominantly labeled a single site with the properties of sigma-i appeared to correspond to the lower affinity [3H](+)-pentazocine receptors. Furthermore, leukemia and glioblastoma, which showed binding site in cell lines such as C6 glioma and NCI-H727 lung only low- and moderate-affinity [3H@(+)-pentazocine-binding sites, carcinoid. Interestingly, the [3H](+)-pentazocine-binding site cx respectively, also showed sigma-i-like profiles. This strongly pressed in human U-138MG glioblastoma, SK-N-SH neuroblastoma, suggests that these lower affinity sites are in fact sigma-i receptors. and LNCaP.FGC prostate had Kd values intermediate between the Table 4 shows that sites labeled by [3H]DTG in all cell lines high- and low-affinity sites and, thus, may represent an additional exhibited a sigma-2-like profile, as indicated by the high affinity for class of binding site. DTG, low affinity for (+)-pentazocine, and by enantioselectivity for In order to verify that [3H@(+)-pentazocine and [3H]DTG labeled (â€”)-benzomorphans over the corresponding (+)-isomer. However, sigma-i and sigma-2 sites as characterized in other tissues, pharma although [3H]DTG labeled an apparent single population of binding cological profiles were determined in selected cell lines. The results sites in each individual cell line (as indicated by linear Scatchard are shown in Table 3 for sigma-i and Table 4 for sigma-2. Sites plots), a comparison of pharmacological profiles across cell lines labeled by [3H](+)-pentazocine in C6 glioma cells exhibited the revealed evidence for sigma-2 heterogeneity. All cell lines examined pharmacological profile characteristic of sigma-i sites, with high exhibited similar high affinity for DTG, moderate affinity for (+)-3- affinity for haloperidol, DTG, and (+)-3-PPP and enantioselectivity PPP and fluphenazine,and low affinity for (+)-pentazocine. How for (+)-benzomorphans over the corresponding (â€”)-enantiomer. The ever, sigma-2-like sites of the human cell lines generally exhibited same was true for ductal breast, leukemia, and glioblastoma, for which about 4- to 7-fold lower affinity for haloperidol and about 10-fold the rank order of potency was: haloperidol > (+)-pentazocine > (â€”)- lower affinity for (â€”)-pentazocine compared to the sigma-2 sites of pentazocine. The competition curves for all ligands were monophasic. the rodent cell lines. Furthermore, although human cell line [3H]DTG This indicated, that despite the presence of multiple sites as demon sites exhibited the characteristic preference for (â€”)-pentazocine over strated by saturation analysis in ductal breast and C6 glioma, under (+)-pentazocine, the degree of enantioselectivity was much reduced

linesMembranes Table 3 Pharmacological profile ofsites labeled by (3HJ(+)-pentazocine in representative cell were incubated with test ligand and 30 [email protected](+)-pentazocinein 50 mai Tris-Ha, pH 8.0, for 120 mm at 37Â°Cusingthe procedures described in â€œMaterialsand 2).TwelveMethods.â€•Thehigh concentration of [3H](+)-pentazocinewas used to assure a high level of specific binding because of the low number of high-affinity sites (see Tables 1 and (SanDiego,concentrations of test ligand were used ranging from 0.05â€”10,[email protected]â€”100,000nai.Data were analyzed using the iterative curve-fitting program GraphPAD InPiot CA) or CDATA (EMF Software, Baltimore, MD). Data were best fit to a one-site model. The K@valueswere calculated from IC,@,valuesusing the Cheng-Prusoff equation (18)thehigher and K,, values shown in Tables 1 and 2. In cell lines in which two [3H@(+)-pentazocine-binding sites were present (Co glioma and T47D breast ductal carcinoma), the Kd of Valuesareaffinity site was used. This was justified since the (+)-pentazocine IC50approximated the higher affinity Kd, and the competition curves had Hill coefficients near unity. the averages of 2â€”3experiments Â±SEM.Each experiment was carried out in duplicate.K, (nM)Ligand vs. [3H1(+)-pentazocine

GlioblastomaHaloperidol C6 Glioma T47D Breast ThP-1 Leukemia U-138MG 1.32 Â±0.29 2.18 Â±0.70 9.32 Â±0.46 52.05 Â±5.11 @ @4T@(+)-3-PPPDTG 62.60 Â±4.12 NDâ€• NDFluphenazine 114Â±43 ND ND ND(+)-Pentazocine 175 Â±34 ND ND 19(â€”)-Pentazocine 6.66 Â±0.54 8.80 Â±1.43 53.16 Â±9.48 302 Â± 352(+)-SKF 3731 Â±1.56 1070 Â±297 968 Â±85 5692 Â± ND(-)-SKF10,047 409 Â±11 ND ND NDa 10,047 2811 Â±246 ND ND ND, not done. 410

Table 4 Pharmacological profile of sites labeled by (3HJDTG in representative cell lines Competition assays were carried out and data analyzed as described in the caption to Table 3. [3HJDTG was used at a concentration of 5 nt@ialone or 10 nM in the presence of 1 @LMdextrallorphan. Results are the averages of 3â€”6 experiments (rodent cell lines) or 2â€”3 experiments (human cell lines) Â±su.i. Each experiment was carried out in duplicate. Data for Co glioma, NB41A3 neuroblastoma, and NG1O8-15hybrid were taken from Vilner and Bowen (13) and were determined without the presence of dextrallorphan. The data were not affected by the absence or presence of dextrallorphan, since experiments carried out with Co glioma, T47D breast carcinoma, and SK-N-SH neuroblastoma in the presence and absence of 1 p.Mdextrallorphan revealed no masking of [3H]DTG binding and maintenance of the differences between human and rodent cell lines (see text). Compounds producing negligible inhibition at 10,000 rat as determined in C6 glioma were: apomorphine, (+)-10,1 1-dihydro-5-methyl-5H-dibenzo[a,djcyclohepten-5,1O-imine (MK-801), naloxone, naltrexone D-A1a2,N-MePhe4,G1y5-ol-enkephalin D-Ser@,Leu5,Thr6-enkephalin morphine y-amino butyric acid and glutamate atropine produced about 50% inhibition at 10,000 nM K1 vs. [3HJDTG (nM)

NeuroblastomaHaloperidol39.3LigandC6 GliomaNB41A3NeuroblastomaNG1OS-15HybridMCF-7 BreastT47D BreastThP-1 LeukemiaU-138MGGlioblastomaSK-N-SH 57DTG37.5 Â±12.888.8 Â±6.550.0 Â±12.6212 Â±20262 Â±37261 Â±4.3346 Â±72369 Â± 8.5(+)-3-PPP286 Â±11.049.3 Â±8.038.6 Â±3.132.1 Â±5.343.6 Â±3.3NDaND58.4 Â± 186Fluphenazine294Â±28527Â±127457Â±73791Â±124701Â±22NDND954Â±104(+)-Pentazocine1723Â±332431Â±3233343Â±11613172Â±403739Â±5522116Â±2637684Â±17539447Â±2900(â€”)-Pentazocine320Â±25375 Â±40493 Â±70645 Â±71552 Â±52NDND1469 Â±

Â±63(+)-SKF Â±20162 Â±2942.5 Â±1.71291 Â±469655 Â±761192 Â±1133970 Â±21082225 4329(â€”)-SKF1O,0476122Â±5294074Â±3403513Â±131612,059Â±1894NDNDND41,461Â±5801a10,04762,882 Â±722591,822 Â±419019,337 Â±684437,302 Â±5861NDNDND39,740 Â±

ND, not done. compared to the rodent lines because of the lower affinity for (â€”)- low-affinity sites in most of the cell lines examined. This was not pentazocine. Similar observations can be made for the enantiomers of observed in any other tissues examined (1 1, 14). The high-affinity site SKF 10,047 when rodent and human cell lines are compared. (Kd 1â€”7nM)had a Kd value similar to sigma-i sites found in other We investigated the possibility that the differences between the tissues (1 i, 14). The lower affinity site (Kd 125â€”360flM)was much human cell lines and rodent cell lines could be due in some way to the more abundant, with Bmaxvalues 16- to 95-fold higher than that of the fact that assays with the rodent lines were carried out without 1 p@M high-affinity site. One possible explanation for the presence of high dextrallorphan present, whereas all assays with human cell lines were and low-affinity [3H](+)-pentazocine-binding sites is that the lower carried out in the presence of 1 @tMdextrallorphan. Since possible affinity site represents labeling of sigma-2 receptors, which occur in interference from sigma-i receptors might be different in cell lines high abundance in the cell lines. However, the Kd values in the having different complements of [3H](+)-pentazocine-binding sites, a 125â€”360aM range would be too low for sigma-2 sites, since (+)- comparison was done using C6 glioma, T47D breast carcinoma, and pentazocine exhibits sigma-2 binding constants in the micromolar SK-N-SH neuroblastoma. These cell lines were selected since they range (8, ii, 14). In addition, linear plots are observed with [3H](+)- represent rat and human cell lines which have high- and low-affinity pentazocine in rat liver and kidney membranes, where the density of [3H](+)-pentazocine-binding sites (C6 glioma and T47D breast sigma-2 sites is 3-fold higher than sigma-i (14). Moreover, the human carcinoma) and a human line (SK-N-SH neuroblastoma) with a single, leukemia cell line expressed a single [3HJ(+)-pentazocine-binding moderate-affinity [3H}(+)-pentazocine-binding site. Haloperidol, site with a Kd value (146 nM) comparable to the lower affinity site in (â€”)-pentazocine, and DTG were competed versus 10 n@i[3H]DTG in cells which expressed two sites. This may suggest that the low- and the absence and presence of 1 @.LMdextrallorphan. First, in each high-affinity sites are distinct entities which can exist in different cells individual cell line, there was no difference in the level of specific and do not represent interconvertible receptor states. Also, guanine [3H]DTGbindingin the absenceandpresenceof 1 p.Mdextrallorphan. nucleotides [GTP and Gpp(NH)p] at concentrations up to 1 mM had no This shows that [3H]DTG produced no detectable sigma-i receptor significant effect on [3H](+)-pentazocine binding to C6 glioma cell labeling under these conditions, presumably because of the very high membranes (not shown), suggesting that the high- and low-affinity density of sigma-2 sites relative to high-affinity sigma-i sites. Also, in sites on these cells are not the result of receptor interactions with each cell line, haloperidol, (â€”)-pentazocine, and DTG had the same 0-proteins. Finally, the intermediate affinity (Kd 30â€”60nM) of sites K1 values whether or not 1 p.M dextrallorphan was present. Impor in human neuroblastoma, glioblastoma, and prostate cell lines may tantly, the differences between C6 glioma and the two human cell suggest yet a third type of [3H](+)-pentazocine-binding site. lines were preserved, whether or not dextrallorphan was present, with Despite the apparent heterogeneity indicated by saturation anal haloperidol and (â€”)-pentazocine having considerably lower affinity in ysis, competition analysis in four cell lines representing cells with the human lines and DTG having the same affinity in all three cell lines@ Thus, the apparent sigma-2 heterogeneity indicated by the subtle differ the various complements of [3H](+)-pentazocine-binding sites encesin pharmacologicalprofileis notdueto the methodof assay. (both high and low affinity, low affinity only, and intermediate affinity) revealed a sigma-i-like profile in each (Table 3). Based on the current data, it cannot be said with certainty that the DISCUSSION [3HJ(+)-pentazocine-binding sites having different Kd values rep The data presented here show that tumor cell lines of various tissue resent different subclasses of sigma- 1 receptor, since other factors origin and species express sigma-i or sigma-2 receptors in high could account for this. The nature of the apparent heterogeneity of density. Most cell lines examined expressed both subtypes. However, sigma-i receptors or [3H](+)-pentazocine-binding sites is it is quite noteworthy that MCF-7 breast adenocarcinoma failed to currently under more detailed investigation. exhibit any specific [3H](+)-pentazocine binding and, as such, is the Sigma-2 sites also appeared to show heterogeneity. The pharmaco first cell or tissue examined to date which does not possess sigma-i logical profile observed in the rodent-derived cell lines was similar to receptors. This cell line should be useful in functional studies of sigma-2 receptors found in other tissues (i 1, 14). Whereas all cell sigma-2 receptors, since ligands selective for the sigma-2 subtype lines showed similar affinity for DTG, (+)-3-PPP, fluphenazine, and over sigma-i are not yet widely available. (+)-pentazocine, the human-derived lines showed a significantly Interestingly, both [3H](+)-pentazocine- and [3H]DTG-binding lower affinity for haloperidol and (â€”)-pentazocine compared to the sites were heterogeneous. [3H](+)-Pentazocine labeled high- and rodent cell lines. Also, the human cell lines were characterized by a 411

lower difference in affinity between the (+)- and (â€”)-enantiomers of effects were also seen in U-i38MG glioblastoma, MCF-7 breast benzomorphans. The lower affinity of human lines for haloperidol is adenocarcinoma, and A375 melanoma.3 These results suggest that particularly significant, since high haloperidol affinity (<50 nM) is a sigma receptors play some important role in the maintenance of hallmark of both sigma-i and sigma-2 receptors (1, 10). cellular viability and the possible utility of sigma ligands as antitumor The apparent sigma-2 heterogeneity was not due to the method agents. In addition, we have shown that the novel iodinated probes of assay, since similar results were obtained in the absence and N-[2-(piperidinylamino)ethyl@-4-iodobenzamide (IPAB) and (N- presence of 1 pM dextrallorphan when a direct comparison was benzylpiperidin-4-yl)-4-iodobenzamide (4-IBP) bind with high affin made in C6 glioma, T47D breast carcinoma, and SK-N-SH neu ity and selectivity to sigma sites of guinea pig brain or rat liver, and roblastoma using 10 [email protected][3H]DTG. Furthermore, the sigma-2 het the 1@I-labeled derivatives bound with nanomolar affinity to sigma erogeneity does not appear to be related to the types of [3HJ(+)- sites of human melanoma and MCF-7 breast carcinoma cells, respec pentazocine-binding sites present in the cells. For example, C6 tively (24, 25). Furthermore, [131I]PAB produced clear scintigraphic glioma and T47D breast carcinoma have the same complement of images of the tumor in nude mice with human malignant melanoma [3H](+)-pentazocine sites and yet show the differences in profile, xenografts (24). Thus, sigma sites may be useful as markers in the whereas T47D breast carcinoma, SK-N-SH neuroblastoma, and noninvasive detection and visualization of a wide variety of tumors ThP-1 leukemia all have different complements of [3H](+)-penta using single proton emission computed tomography and positron zocine-binding sites and yet show similar profiles. In addition, emission tomography technology. MCF-7 breast cells show the profile difference when compared to We have demonstrated the presence of sigma-i and sigma-2 the rodent cell lines, despite the fact that these cells do not possess receptor subtypes in several tumor cell lines from both rodent and any specific [3H](+)-pentazocine-binding activity. human. Further investigation will be needed in order to elucidate The sigma-2-like site of the human cell lines shares some, but not the function of sigma-binding sites and their ligands in these cells. all, characteristics with a â€œlow-affinityâ€•sigma-likesite previously The use of cell lines expressing sigma- 1 and sigma-2 receptors will described in NCB-20 cells (16). This site exhibited affinity for halo also facilitate ongoing studies of sigma receptors in the brain and peridol and a slight preference for (â€”)-benzomorphans over (+)- peripheral tissues. benzomorphans, suggesting a relationship with the sigma-2 receptors found in other tissues. However, the lower affinity for haloperidol (K, ACKNOWLEDGMENTS 508nM),(â€”)-pentazocine(K1i532 nM),(+)-3-PPP(K, 8246nr@t),and several other sigma ligands demonstrated this site to be distinct. It is We wish to thankDes. TerryMoody, ElliottRichelson,EnriqueSilva, and not clear whether the apparent sigma-2 heterogeneity among cell lines R. J. Weberfor theirgifts of NCI-H727lung carcinoid,N1E-115 neuroblas described in the present study represents species differences, sigma-2 toma, S-20Y neuroblastoma, and NG1O8â€”15hybrid cells, respectively. receptor subtypes, or some other phenomenon. This will require further study using additional tissues and ligands, particularly REFERENCES neuroleptics and enantiomeric benzomorphans. 1. Walker, J. M., Bowen, W. D., Walker, F. 0. Matsumoto, R. R., de Costa, B. R., and Among the cell lines examined in the current study, the number Rice, K. C. Sigma receptors: biology and function. Pharmacol. Rev., 42: 355â€”402, of sigma receptors expressed per cell is very high. For example, C6 1990. glioma, NG1O8â€”15 neuroblastoma-glioma hybrid, and N1E-1 15 2. Itzhak Y., and Stein, I. Sigma binding sites in the brain: an emerging concept for multiple sites and their relevance for psychiatric disorders. Life Sci., 47: 1073â€”1081, neuroblastoma express about 0.3, 1.3, and 1.4 million sigma-2 1990. receptors per cell, respectively. It is likely that this level of 3. Su, T-P. Sigma receptors: putative links between nervous, endocrine, and immune @ expression is quite different from that of normal tissue. The systems. Eur. J. Biochem., 200: 633â€”642,1990. 4. Su, T-P. Delineating biochemical and functional properties of sigma receptors: values observed in membranes from normal tissues such as brain, emerging concepts. Cot. Rev. NeurobioL, 7: 187â€”203,1993. liver, and kidney (8, i i, 14) represent the density of sites in a 5. Ferris, C. D., Hirsh, D. J., Brooks, B. P. and Snyder, S. H. Sigma receptors: from molecule to man. J. Neurochem., 57: 729â€”737,1991. heterogeneous cell population, some of which may have high and 6. Bowen, W. D. Biochemical pharmacology of sigma receptors. In: T. W. Stone (ed), some of which may have low sigma receptor density. As such, the Aspects of Synaptic Transmission, Vol. II: Acetylcholine, Sigma Receptors, CCK and Bmax values from normal tissues represent an average receptor Eicosanoids, Neurotoxins, pp. 113â€”136.London:Taylor and Francis, 1993. 7. Debonnel, G. Current hypotheses on sigma receptors and their physiological role: density, and it is therefore not possible to accurately estimate the possible implications in psychiatry. J. Psychiatr. Neurosci., 18: 157â€”172,1993. density on any given cell type within the tissue. However, when 8. Hellewell, S. B., and Bowen, W. D. A sigma-like binding site in rat pheochromocy toma (PC12) cells: decreased affithty for (+)-benzomorphans and lower molecular this is taken into account, high-affinity sigma-i sites would occur weight suggest a different sigma receptor form from that of guinea pig brain. Brain in much lower density, whereas sigma-2 sites would occur in much Rca., 527: 244â€”253,1990. higher density in these cell lines compared to normal tissues. This 9. Rothman, R. B., Reid, A., Mahboubi, A., Kim, C-H., de Costa, B. R., Jacobson, A. E., and Rice, K. C. Labeling by 13H]1,3-di(2-tolyl)guanidineoftwo high affinity binding would be consistent with the results of others who showed that sites in guinea pig brain: evidence for allosteric regulation by calcium channel sigma receptors labeled with [3H]DTG are overexpressed in human antagonists and psuedoallosteric modulation by sigma ligands. Mol. Pharmacol., 39: solid tumors from various body sites relative to surrounding nor 222-232 1991. 10. Quirion, R., Bowen, W. D., Itzhak, Y., Junien, I. L, Musacchio, I. M., Rothman, mal tissue (19, 20). The high level of sigma receptor expression in R. B., Su, T-P., Tam, S. W., and Taylor, D. P. A proposal for the classification of tumor-derived cell lines and solid tumors suggests the likelihood sigma binding sites. Trends Pharmacol. Sci., 13: 85â€”86,1992. 11. Bowen, W. D., de Costa, B. R., Hellewell, S. B., Walker, J. M., and Rice, K. C. that these receptors subserve an important function in cancer cells. [3H](+)-Pentazocine: a potent and highly selective benzomorphan-based probe for There are other implications of the high sigma receptor density for sigma-l receptors. Mol. Neurophannacol., 3: 117â€”126,1993. tumor biology. We have shown that sigma ligands produce marked 12. Di Paolo, L, Carroll, F. I., Abraham, P., Bai, X., Parham, K., Mascarella, S. W., Than& X., Wallace, P., Walker, J. M., and Bowen, W. D. N-Substituted derivatives changes in the morphology and viability of tumor-derived cell lines of normetazocine: differentiation of sigma-i and sigma-2 receptors. Soc. Neurosci. with potencies which approximately parallel their rank order of affin Abstr., 17: 814, 1991. ity at sigma receptors (21â€”23).The neuroleptics haloperidol, reduced 13. Vilner, B. J., and Bowen, W. D. Characterization of sigma-like binding sites of NB41A3, S-20Y, and N1E-115 neuroblastomas, Co glioma, and NG1O8-15 neuro haloperidol, and fluphenazine as well as the novel sigma ligands blastoma-glioma hybrid cells: Further evidence for sigma-2 receptors. In: J. M. BD737, BD1008, SH344, and JL-II-147 produced morphology Kamenka and E. F. Domino (eds.), Multiple Sigma and PCP Receptor Uganda: changes and cytotoxic effects in several rodent cell lines and also in the human neuroblastomas SK-N-SH and SH-SY5Y (23). Cytotoxic 3 w. D. Bowen and B. J. Vilner, unpublished observations. 412

Mechanisms for Neuromodulation and Neuroprotection?, pp. 341â€”353.Ann Arbor, 20. Bem, W. T., Thomas, G. E., Mamone, J. Y., Homan, S. M., Levy, B. K., Johnson, Ml: NPP Books, 1992. F. E., and Coscia, C. J. Overexpression of sigma receptors in nonneural human 14. Helleweil, S. B., Bruce, A., Feinstein, G., Orringer, J., Williams, W., and Bowen, Wmors@Cancer Res., 51: 6@58-6@62, 1991. w. D. Ratliverandkidneycontainhighdensitiesofsigma-iandsigma-2receptors: 21. Bowen, W. D., Vilner, B. J., and de Costa, B. R. Sigma receptor-mediated changes in characterization by ligand binding and photoaffinity labeling. Eur. J. Pharmacol., morphology, division, and viability of Co glioma cells. Soc. Neurosci. Abstr., 19: Mol.Pharmacol.Sect.,268:9â€”18,1994. 420, 1993. 15. Kushner, L, Zukin, S. R., and Zukin, R. S. Characterization of opioid, sigma, and 22. Vilner, B. J., and Bowen, W. D. Sigma receptor-active neuroleptica are cytotoxic to phencyclidine receptors in the neuroblastoma-brain hybrid cell line NCB-20. Mol. Co glioma cells in culture. Eur. J. Pharmacol., 244: 199â€”201,1993. PharmacoL, 34: 689â€”694,1988. 23. Vilner, B. J., de Costa, B. R., and Bowen, W. D. Cytotoxic effects of sigma ligands: 16. Wu, X-Z., Bell, J. A., Spivak, C. E., London, E. D., and Su, T-P. Electrophysiological sigma receptor-mediated alterations in cellular morphology and viability. J. Neurosci., and binding studies on intact NCB-20 cells suggest presence of low affinity sigma 15: 117â€”134,1995. receptor. J. Pharmacol. Exp. Ther., 257: 351â€”359,1991. 24. John, C. S., Bowen, W. D., Saga, 1., Kinuyam S., Vilner, B. J., Baumgold, J., Paik, 17. Lowry, 0. H., Rosebrough, N. J., Fan, A. L, and Randall, R. J. Protein measurement C. H., Reba, R. C., Neumann, R. D., Varma, V. M., and McAfee, J. G. A new with the Folin phenol reagent. J. Biol. Chem., 193: 265â€”271,1951. malignant melanoma imaging agent: synthesis, characterization, in vitro binding and 18. Cheng, Y-C., and Prusoff, W. H. Relationship between the inhibition constant (Ki) biodistribution of iodinc-125-(2-piperidinylaminoethyl)4-iodobenzamide. J. Nucl. and the concentration of inhibitor which causes 50% inhibition (150)of an enzymatic Med., 34: 2169â€”2175,1993. reaction. Biochem. Pharmacol., 22: 3099â€”3108,1973. 25. John, C. S., Vilner, B. J., and Bowen, W. D. Synthesis and characterization of 19. Thomas, G. E., Szucs, M., Mamone, J. Y., Bem, W. T., Rush, M. D., Johnson, F. E., [â€˜@I](N-benzylpipeddin-4-yl@iodobenzamide,4-IBP,a new sigma receptor radio and Coscia, C. J. Sigma and opioid receptors in human brain tumors. Life Sci., 46: pharmaceutical: high affinity binding to MCF-7 breast tumor cells. J. Med. Chem., 1279â€”1286,1990. 37: 1737â€”1739,1994.

4i3

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1995 American Association for Cancer Research. Sigma-1 and Sigma-2 Receptors Are Expressed in a Wide Variety of Human and Rodent Tumor Cell Lines

Bertold J. Vilner, Christy S. John and Wayne D. Bowen

Cancer Res 1995;55:408-413.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/55/2/408

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/55/2/408. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.