Effects of Bombesin Antagonists on the Growth of Small Cell Lung Cancer Cells in Vitro Judith E

[CANCER RESEARCH 48,4783-4789, September 1, 1988! Effects of Bombesin Antagonists on the Growth of Small Cell Lung Cancer Cells in Vitro Judith E. I.avion,' Denis B. Scanion,2 Claire Soveny, and George Morstyn Ludwig Institute for Cancer Research, Melbourne Tumour Biology Branch, Victoria, Australia, 3050 [J. E. L., D. B. S., G. M.J; and Prince Henry's Hospital, Melbourne, Australia 3000 Â¡C.SJ

ABSTRACT stimulated growth may be useful for SCLC treatment (8). However, therapy with monoclonal antibodies would need to Small cell lung cancer (SCLC) produces several neuroendocrine peptides, including gastrin-releasing peptide (GRP), the mammalian equiv be prolonged and potential problems include the development of neutralizing antibodies, serum sickness, and other hypersen- alent of bombesin. There is some evidence to support the suggestion that GRP is an autocrine regulator of SCLC growth. Therefore, we have sitivity reactions, and poor penetration of the antibodies within tested the effect of bombesin and two antagonists of bombesin on SCLC the tumor. An alternative to anti-GRP antibodies for the treat cell growth in a serum-free liquid tissue culture system. The antagonists ment of SCLC might be GRP antagonists. Several analogues used were analogues of substance P: spantide and (D-Arg1, D-Pro2, D- of substance P have been shown to inhibit the action of bom Trp '', Leu") substance P. The cell lines used in this study all produced besin in various in vitro biological assays (12-15). The present GRP-related peptides and one line had demonstrable GRP receptors. study examines the effect of GRP antagonists on SCLC cell Exogenous bombesin did not cause any stimulation of growth in the liquid growth. The two antagonists used were spantide, which inhibits culture assay. The bombesin antagonists inhibited SCLC cell growth, but binding of bombesin to brain slices (16), and DAPTL-SP, which apparently not via the bombesin receptor. The bombesin used was bio inhibits bombesin-induced proliferation of Swiss 3T3 cells (13, logically active because it stimulated the proliferation of Swiss 3T3 fibroblasts. The antagonists caused inhibition of this bombesin-induced 15) and amylase secretion from pancreatic acinar cells (12). proliferation, which was reversed by addition of excess bombesin. In The biological activities of the peptides used were confirmed by addition, the antagonists and substance P alone stimulated proliferation testing their activity on Swiss 3T3 fibroblasts. of 3T3 cells, indicating that they may interact with another growth factor receptor on 3T3 cells. We conclude that growth of SCLC cells is not dependent on bombesin under all in vitro culture conditions because MATERIALS AND METHODS bombesin failed to stimulate growth in liquid cultures and the growth Cell Lines. Murine Swiss 3T3 fibroblasts were maintained by twice inhibition caused by bombesin antagonists was probably not mediated by weekly passage in DME containing 10% PCS, and antibiotics. Cultures the bombesin receptor. were incubated in 10% CCÂ«2inair. The establishment and characteristics of the SCLC cell lines used in INTRODUCTION this study have been described previously (17) with the exception of SCLC3 comprises 20% of lung cancers. Although mixed LICM 222 which is a recently established line. LICM 222 was estab lished from a bone marrow biopsy from a female patient at relapse. tumors with adeno and squamous elements occur, SCLC ap Culture supernatant was positive for neuron-specific enolase. All the pears to be a distinct entity. SCLC cells exhibit many neuroen cell lines appear to be of the classic type as defined by Carney et al. docrine characteristics, including the presence of neuron-spe (18). Cell lines were maintained in HITES medium (19) containing cific enolase and dense-core granules, as well as the secretion 2.5% PCS, L-glutamine and antibiotics, with weekly passaging, and of a variety of peptide hormones including ACTH, calcitonin, were incubated in 5% CO: in air. Powdered media and PCS were obtained from Flow Laboratories, North Ryde, Australia. Both cell vasopressin, lipotropin, parathormone, glucagon, somatostatin, types were grown in 75-cm2 tissue culture flasks (Nunc, Denmark). and neurotensin (1-3). Most SCLC cell lines of the classic type Peptides. Peptides were synthesized on an Applied Biosystems (Fos produce GRP, the mammalian equivalent of bombesin (2-7). ter City, Ã‡A)430A peptide synthesizer and were purified by high- This has led to the suggestion that SCLC is an autocrine tumor, performance liquid chromatography. The peptides were characterized dependent on GRP for growth (8). The evidence supporting and analyzed quantitatively by amino acid analysis. The amino acid this hypothesis is: most SCLC cell lines produce GRP; about sequences of the peptides that were synthesized are given in Table 1. 20% of lines tested bind detectable GRP/bombesin (9); some Small quantities of peptides were purchased from Peninsula Labora SCLC cell lines have been reported to respond to bombesin in tories (Belmont, CA) and their activity was compared with that of the vitro (8, 10, 11); and a monoclonal antibombesin antibody peptides we synthesized. In all cases the activities were similar. The inhibits the growth of some SCLC cell lines in vitro and in nude GRP used in this study was purchased from Peninsula. Peptides were mice (8). dissolved and diluted in normal saline containing 4 HIMHC1 and 0.1% If SCLC is dependent on GRP, then inhibition of GRP- BSA (purified from plasma). Both the 3T3 cells and SCLC cells gave no response to the diluent. Received 7/28/87; revised 12/29/87, 5/12/88; accepted 6/1/88. Assays for Bombesin Production. SCLC cell lines were cultured at The costs of publication of this article were defrayed in part by the payment approximately 106/ml and the culture supernatant was collected after of page charges. This article must therefore be hereby marked advertisement in 4 days for assay. Cell extracts were prepared from 0.7 to 9 x IO7cells accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ' To whom requests for reprints should be addressed, at Ludwig Institute for as described by Cuttitta et al. (8). Cell extracts and supernatants were Cancer Research, Melbourne Tumor Biology Branch, Royal Melbourne Hospital, assayed for bombesin-like immunoreactivity by a radioimmunoassay Victoria 3050, Australia. described previously (20). In some cases, culture supernatants were 2 Present address: Auspep, Victorian College of Pharmacy, Parkville. Australia, concentrated by lyophilization and reconstitution in a reduced volume 3052. 'The Abbreviations used are: SCLC, small cell lung cancer; GRP, gastrin- prior to assay. Several dilutions of each sample were tested to confirm releasing peptide; DAPTL-SP, (D-Arg1, D-Pro2, D-Trp7'', Leu") substance P; that the assay was linear for the supernatants and extracts. PCS, fetal calf serum; HITES medium. RPMI 1640 medium containing insulin Proliferation Assays. Swiss 3T3 cells were seeded into 96-well micro- (5 Mg/ml), transferrin (10 Mg/ml), hydrocortisone (10 nM), estradiol (10 nM), and liter plates (Nunc) at IO4 cells/well. Cultures were used after 6 or 7 sodium selenite (30 nM); BSA, bovine serum albumin; ACTH, adrenocorticotropic hormone, DME, Dulbecco's modified Eagle's medium; PBS. phosphate buffered days incubation when they were confluent and relatively quiescent. The saline; ICM, level producing 50% inhibition; BN, bombesin; ITS, insulin, trans cells were washed twice with DME without FCS and responses were ferrin, and sodium selenite in the same ratio as in HITES medium. tested in the same medium. Peptides were added at culture initiation 4783

Table 1 Amino acid sequences of peptides PeptideBombesinGRP.4-27Substance sequence"pGlu-Gln-Arg-Leu-Gly-Asn-Gln-Trp-Ala-Val-Gly-His-Leu-Met-NHzacid

Met-Tyr-Pro-Arg-Gly-Asn-His-Trp-AIa-Val-Gly-His-Uu-Met-NH2 P Arg-Pro-Lys-Pro-Gln-Gln-Phe-Phe-Gly-Leu-Met-NHi Spantide ARG-Pro-Lys-Pro-Gln-Gln-TRP-Phe-TRP-Lcu-Leu-NH2 (DAPTL)-SPAmino ARGTRO-Lys-Pro-Gln-Gln-TRP-Phe-TRP-Leu-tiu-NHj " D-amino acids are in capital letters. Amino acid substitutions in the substance P analogues are underlined. and DNA synthesis was measured by pulsing the cultures with [3H- not an antagonist (12, 16, 36). Fig. 1 shows that 3T3 cells /nefAy/lthymidine (0.5 Â¿iCi/well;specific activity 5 Ci/mmol, Amer- responded to bombesin by increased [3H]thymidine uptake, with sham. Inc., UK) for 4-6 h after 20-24 h incubation. Cells were harvested a half-maximal response occurring at 2 nM bombesin. The by trypsinizing, then using a Skatron cell harvester (Skatron, Norway). response of 3T3 cells to substance P and the antagonists in the In several experiments this procedure was compared with TCA precip absence of bombesin is shown in Fig. 2 (top). It was found that itation and the results were similar. A control group in which the response to 10% PCS was measured was included in every experiment. the antagonists at greater than 25 nM and substance P at greater Because the incorporation of [3H-m?rA>7]thymidine in DM E alone was than l Â¿/Mstimulated DNA synthesis. The maximum response relatively high, we subcloned our 3T3 cells and obtained a clone with to substance P was similar in magnitude to the maximum 10-fold lower background [3H-mefA/y/]thymidine incorporation, but response to bombesin, whereas the response to the antagonists this clone gave similar responses to peptides to the uncloned line. All was always lower. Fig. 2 (bottom) shows the effects of the experiments reported here were performed on the uncloned cells. antagonists and substance P on the response to a slightly SCLC cells were used 2 to 4 days after passage. Single cell suspen suboptimal concentration of bombesin (3 x 10~9 M). Both sions were prepared by washing the cells twice with PBS then pipeting antagonists caused maximum inhibition of the bombesin re them in PBS containing 0.2 g/liter glucose, 0.2 g/liter EDTA, and 14 sponse at 12.5 Â¿A!,butat higher concentrations the inhibition mM lignocaine hydrochloride (21) (Astra Pharmaceuticals, North Ryde, NSW, Australia), at 37'C until the suspension appeared to be uniform was overcome by a partial agonist effect, particularly with (2-4 min). The cells were washed three times and resuspended in DAPTL-SP. Substance P did not inhibit the bombesin response HITES without FCS. Cultures were set up at 3 x IO4cells/well in 96- and caused a slight increase in the response at concentrations well microtitre plates. Peptides were added at initiation of culture and above 1 p\i. Thus, bombesin and the antagonists were biolog proliferation was measured by pulsing the cultures on Day 3 of incu ically active although unexpectedly substance P and its ana bation for 6 h with [3H-mefA>'/]thymidine then harvesting the cultures logues showed stimulatory activity at high concentration. The on a cell harvester. antagonistic effect of spantide and DAPTL-SP could be re Binding Assays. 3T3 cells were seeded into 24-well tissue culture versed by the addition of excess bombesin (Fig. 3), indicating plates (Nunc) at 5 x 104/well and used after 6 or 7 days incubation. that the effect was probably mediated by the bombesin receptor. Cultures were washed twice with DME without PCS and binding was carried out in 0.2 ml DME containing 0.1% BSA and 25 mM 4-(2- Binding of GRP to 3T3 Cells. To further support the conclu hydroxyethyl)-l-piperazineethanesulfonic acid (pH 7.4). 125I-Labeled sion that the antagonists were acting by inhibiting binding to GRP ([125I]GRP)was purchased from Amersham (specific activity, 2000 the bombesin receptor, binding studies were carried out with Ci/mmol) and 10* cpm/well (0.1-0.3 nM) was used. Initial kinetic [125I]GRP. Kinetic studies at 37Â°Cestablished that maximum experiments carried out at 37Â°Cand at room temperature (18-20Â°C) binding was achieved after 15-min incubation, but this level of established that optimal binding was achieved by incubating assays for binding started to decrease by 30-45 min, making experiments 2 h at room temperature. The cells were then washed four times with technically difficult. To slow down the loss of binding, experi ice-cold PBS/0.1% BSA, solubilized with 0.5 ml 0.1 M NaOH and ments were carried out at room temperature. Under these transferred to tubes for counting. The wells were washed once with 0.5 conditions a plateau level of binding similar to the maximum ml distilled water and the washings were added to the appropriate tubes. binding at 37Â°Cwas reached by l h and remained stable for a SCLC cells were treated with lignocaine as described above to prepare further 2 h. Thus a 2-h incubation period at room temperature single cell suspensions. Binding assays were carried out in 1.5 ml was used for equilibrium binding. Substance P, the antagonists Eppendorf tubes using HITES medium containing 0.1% BSA and 25 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid (pH 7.4). One and GRP-related peptides were tested for their ability to inhibit x IO6cells in 0.2 ml/tube and 105cpm [125I]GRP/tube were used. Tubes [125I]GRP binding. Data from these experiments are summa were rotated for 30 min at room temperature as this was found to give rized in Table 2. Half-maximal inhibition of binding by BN/ optimal binding. Cells were centrifuged and the pellet resuspended in GRP peptides was achieved at concentrations similar to those 0.2 ml binding medium, then layered over 0.1 ml mineral oil (three parts di-/i-butyl phthalate to one part dinonyl phthalate; BDH, Poole, 30 UK) in a microcentrifuge tube (Sarstedt, W. Germany), centrifuged and the tip of the tube cut off and counted. Analysis of Data. The statistical significance of differences between groups was determined by using the t test for unpaired data. All experiments were performed at least three times with similar results to those shown in this report. Binding data were analyzed by computer using the "Ligand" program of Munson and Rodbard (22).

RESULTS Confirmation of Biological Activity of Peptides. To show that the peptides used in this study were biologically active, Swiss Ht- 0.1 0.3 1 3 10 30 3T3 fibroblasts were stimulated with bombesin and the effect Bombesinconcentration (nM) of the antagonists on the response was determined. Substance Fig. 1. The response of Swiss 3T3 cells to bombesin. The assay was carried P was included as a control for nonspecific effects because it is out as described in "Materials and Methods." Points, mean Â±SD of triplicates. 4784

Table 2 Inhibition ofGRP binding to Swiss 3T3 cells by various peptides Peptide 1CÂ»"(nM) GRP 2.0 Â±1.4 BN 2.7 Â±1.7 GRP.4-27 2.2 Â±0.9 DAPTL-SP 7.700 Â±1,800 Spantide 6,700 Â±900 Substance P 5,100 Â±1,100 " The mean Â±SD of three experiments is given.

Table 3 Production ofBLl by SCLC cell lines BL1" present in supernatant' extract' o CelllineLICM (fmol/ml)23 (fmol/106cells)22 44 (0-80) 120 L1CM 56 174(26-310) 215 LICM 107 50(19-100) 172 LICM 222Culture 23CCell 30 " BLI, bombesin-like immunoreactivity assayed by radioimmunoassay (20). * Mean of four assays (range). r One assay only.

O 6 12.5 25 50 100 O 0.1 0.3 1 3 10 30 Peptide concentration l/iM)

Fig. 2. The elicci of bombesin and antagonists on Swiss 3T3 cell proliferation. Top, effects of the antagonists or substance P in the absence of bombesin; bottom, effects of the antagonists or substance P on the response to a suboptimal concentration of bombesin (3 x 10"*M).The results were calculated by subtracting the background response in the absence of any peptide from all data points, then using the formula: 50 107 222 % Bombesin response = â€”â€”Jcpm (antagonist 5- + bombesin) : x 100% cpm bombesin Fig. 4. Binding of '"I-labeled GRP to SCLC cell lines. A, test of four different cell lines; â€¢.binding of '"I-GRP alone; D, binding of '"(-labeled GRP in the Points, mean Â±SD of triplicates. D, DAPTL-SP; A, Spantide; O, Substance P. *, presence of IO"* M unlabeled GRP; bars, mean Â±SD of triplicates; B, Scatchard inhibition is statistically significant !/' â€¢0.01). plot of binding to LICM 107 cells. Unlabeled GRP was titrated in the presence of a constant amount of ['"IJ-GRP.

DNA synthesis (Fig. 2). Substance P also inhibited GRP bind ing, with an IC5o value of 5 MM. Production of Bombesin by SCLC Cell Lines. The four cell lines used in this study were tested for production of bombesin- related peptides by radioimmunoassay of culture supernatants and cell extracts (Table 3). The level of bombesin-like immu noreactivity in culture supernatants varied considerably between assays, probably because of differences in culture conditions during the 2-year period in which the assays were carried out. The mean bombesin levels correlated with the amount found in the cell extracts. Detection of Bombesin Receptors on SCLC Cell Lines. Kinetic studies of [I25I]GRP binding to SCLC cells gave similar results to the 3T3 cell kinetic studies. Thus, binding was again per formed at room temperature, although in this case a plateau Peptide DAPTL-SP Spantide Substance P level of binding was reached after 30 min. Cell lines were tested Fig. 3. Reversibility of antagonist inhibition of the response of Swiss 3T3 cells to bombesin. Cells were cultured with 3 x 10"' M bombesin (LT),3 x 10"' M for binding of [I25I]GRP in the absence or presence of excess bombesin plus antagonist or substance P (â€¢).or 3 x 10 " M bombesin plus unlabeled GRPi4-27 (Fig. 4/4). Only one of the SCLC cell lines antagonist or substance P IÃŸ)-DAPTL-SP was used at 25 /IM, spantide at 50 /IM, and substance P at 100 MM.Hun. the mean Â±SD of triplicates. *, inhibition is (LICM 107) showed detectable binding above the nonspecific statistically significant (/'< 0.01). background seen in the presence of excess cold GRP. Fig. 4B shows the Scatchard analysis of data from a lit rat ion of unla beled GRP in the presence of constant amounts of [I25I]GRP. causing stimulation of DNA synthesis (Fig. 1). Scatchard analy The graph is a straight line indicating that there is a single sis of the data gave a KD value for GRP of 1.0 Â±0.3 nM and binding site for GRP. The KD value for GRP binding to LICM 23,000 Â±6,000 receptors per cell (means Â±SD of three deter 107 was 0.32 Â±0.01 nM and the number of receptors per cell minations). The ICsovalues for the two antagonists were similar was 2900 Â±900 (mean Â±SD values from three experiments). to the concentrations causing inhibition of bombesin-stimulated Effect of Bombesin on SCLC Cell Growth. Single cell suspen- 4785

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1988 American Association for Cancer Research. BOMBESIN ANTAGONISTS ON GROWTH OF SMALL CELL LUNG CANCER sions were prepared from recently passaged stock cultures. Cells density, which gave much lower [3H]thymidine incorporation were cultured for 3 days and proliferation was assessed by on a per cell basis but bombesin had no effect under these measuring [3H]thymidine incorporation on Day 3. Fig. 5A conditions. The response was also tested on Days 1 and 2 of shows the response of SCLC cell lines to ITS and PCS under culture and with 3 additional cell lines, again with no effect these culture conditions. Both additives were titrated and the being observed. However, these cell lines make their own GRP optimal response is shown. The maximum response to ITS was and thus may not require exogenous GRP/bombesin under our obtained with 25% of the amount in HITES medium and double culture conditions. We therefore tested the effect of the bom the normal concentration was slightly inhibitory. The maxi besin antagonists in the absence of added bombesin. mum response to PCS was usually obtained at a concentration Effect of Bombesin Antagonists on SCLC Cell Growth. Single of 2.5%. Five and 10% PCS were inhibitory. The optimal cell suspensions in HITES medium were cultured with the responses ranged from 2- to 5-fold above the background [3H]- antagonists or substance P and [3H]thymidine incorporation thymidine incorporation in RPMI with no additives. When was measured on Day 3 of culture (Fig. 7). Both antagonists PCS was added to cells grown in HITES medium, low concen inhibited proliferation but at higher concentrations than were trations of PCS induced an additional 30 to 220% increase in required in the 3T3 cell assays. Substance P had no effect, [3H]thymidine incorporation over HITES alone, but higher suggesting that the inhibition was a specific effect rather than PCS concentrations were inhibitory (Fig. 5B). Thus, the data a toxic effect of higher peptide concentration. The reversibility in Fig. 5 have established that a response to growth factors can of the inhibition was tested by adding excess amounts of bom be detected in liquid cultures. besin before the addition of antagonists at various concentra The effect of bombesin was then tested on cells grown in tions. The results with LICM 107 are shown in Fig. 8. Bombesin HITES medium without serum, under the same conditions. was not able to reverse the effect of the antagonists even when Fig. 6 shows that bombesin at concentrations of 10~'Â°-10~6M only partial inhibition was seen. The other three cell lines gave had no significant effect on the proliferation of any of the small similar results. It appears then that the inhibition was not cell lines. In other experiments, the bombesin concentration mediated by the bombesin receptor. This was confirmed by was increased to 10~" M and half-log dilution steps were used: testing the peptides on an irrelevant cell line (A549, a pulmo again no effect was observed. Cells were cultured at lower cell nary adenocarcinoma) which does not produce bombesin or have bombesin receptors (11). The growth of A549 cells was inhibited by the antagonists with similar dose-response curves to those obtained with the SCLC cell lines (Table 4). Thus the 500 antagonists are able to affect cells that do not have bombesin Â£ WO receptors.

2 300 Inhibition of GRP Binding to LICM 107 by Bombesin Antag | onists. The effect of substance P and the bombesin antagonists 3 200 on the binding of [125I]GRP to LICM 107 cells was determined. a? Binding was inhibited in a dose-dependent manner and the IC5o 100

0 ' 0.1 0.5 2.5 12.5

Cell line PCS concentratori (%v/v) Fig. 5. The response of SCLC cell lines to HITES ingredients and FCS. Responses were calculated as a percentage of the control [3H]thymidine incorpo ration. Data, mean Â±SDof triplicates; A, cell suspensions were cultured in RPMI (O), or RPMI containing various concentrations of ITS or FCS. The optimal responses to ITS (â€¢)andPCS (â€¢)areshown. Controls (RPMI alone) were LICM 44; 2760 cpm: LICM 56, 910 cpm; LICM 107, 18,400 cpm; LICM 222, 3,390 cpm. B, responses of SCLC cells cultured in HITES medium to titrations of FCS. O, LICM 44; â€¢,LICM 56; D, LICM 107; â€¢,LICM 222. Controls (HITES alone) were LICM 44, 3672 cpm; LICM 567, 1968 cpm; LICM 107,10,539 cpm, LICM 222. 2430 cpm. Titrations have been repeated at least three times with similar results.

80 100 0 20 100 Pepfide concentration ( -10 -9 -8 -7 -6 Fig. 7. The effects of bombesin antagonists and substance P on the prolifera Bombesin concentration (log M) tion of SCLC cell lines. Proliferation was assessed by measuring incorporation of Fig. 6. The effect of bombesin on the proliferation of SCLC cell lines. [3H]thymidine. The control response in the absence of added peptide was taken Bombesin was titrated from 0.1 nM to l UM.O, LICM 44; â€¢,LICM 56; D, LICM as 100%. Cells were cultured with DAPTL-SP (D). spantide (A), or substance P 107; â€¢LICM 222. Poi'ntt, mean Â±SD of triplicates. (O), /"omis, mean Â±SD of triplicates. 4786

GRP concentration range and only at very high concentrations (3 fiM): thus the significance of these observations is not clear. Bombesin has been reported to stimulate the growth of most SCLC lines tested in an agarose cloning assay with a peak response at a concentration of 50 nM (11). There are several possible reasons for obtaining different results from different assay systems. It is possible that culture in agarose is less optimal for cell growth than liquid culture and therefore the effect of bombesin is more easily seen. Another possibility is that the cells in liquid culture produce more GRP than those in agarose, and thus are less dependent on exogenous bombesin. A third possibility is that different subpopulations of cells grow in liquid and semisolid cultures. For example, the proliferating cells detected in the relatively short liquid culture assay may only be those already in cycle when the cultures are set up, and these cells may not be dependent on bombesin. In the longer, semisolid agarose assays, resting cells may be stimulated by bombesin to enter the cell cycle and form colonies. We have tested the growth of these cell lines in semisolid agarose cul 0.01 0.1 1.0 Bombesin concentration (juM) tures, but the results have been inconsistent with the lines at times showing some response to bombesin and at other times, Fig. 8. The effect of bombesin on the inhibition of LICM 107 proliferation by bombesin antagonists. Cells were cultured with DAPTL-SP or spantide in no response. The basis for this inconsistency is not clear and amounts that caused partial inhibition of proliferation. Bombesin was added at does not relate to cell viability or phase of growth of the culture culture initiation, before the antagonists were added. , control response with no added peptide; points, mean Â±SD of triplicates. from which the cells were cloned. It is clear that the factors affecting the growth of SCLC cells are not completely under Table 4 Effect of bombesin antagonists on the growth ofA549 cells stood because the cells will grow well in some but not all Cells were cultured at 5 x 103/well in RPMI with 10% FCS. [3H]lhymidine experiments, as reported by others (23) and also noted in our was added for 6 h on day 3 of culture. studies. Percentage of control [3H]thy- Binding studies with [I25I]GRP established that one out of midine incorporation" Antagonist con the four SCLC cell lines tested had detectable GRP receptors centration ((.M)12.5 and thus should have been able to respond to bombesin. The Â±12 Â±12 binding data are consistent with the results of Moody et al. (9) 25 110Â± 17 104 +657 who found that 19% of cell lines tested bound detectable iodi- 50 17Â±3 Â±13 nated Tyr4-bombesin. It is not clear whether the negative cell 100DAPTL-SP99 17 Â±3Spantide101 9.2 Â±2 Â°Control incorporation was 51,735 Â±2,865 cpm. Values are the means Â±SD lines have GRP receptors or not. Their receptors may have been down-regulated by the cells' own GRP, but this seems of triplicates. unlikely because the one cell line with detectable receptors Table 5 Inhibition ofCRP binding to LICM 107 cells by various peptides (LICM 107) produced as much GRP as the other cell lines. Peptide 1CÂ»(nM)Â° Alternatively, receptor numbers may be too low to be detected GRP 0.48 Â±0.05 by this method. SCLC cell lines that respond to bombesin but DAPTL-SP 4,800 Â±300 do not have detectable receptors have been described by others Spantide 4,340 + 480 Substance P 13,300 + 3,200 (23), indicating that some cell lines have very low receptor Â°Themean + SD of three experiments is given. numbers. The 3T3 cells bound [I25I]GRP with a similar receptor affinity to that reported by others for iodinated GRP or Tyr4-bombesin values are summarized in Table 5. The bombesin antagonists (15,16,24-27). The number of receptors per cell was somewhat inhibited binding to LICM 107 at similar concentrations to lower than the 105receptors reported by Zachary and Rozengurt those inhibiting binding to 3T3 cells (compare Tables 2 and 5). (15). Both antagonists inhibited [125I]GRP binding and the IC5o It is noteworthy that inhibition of proliferation required at least 10-fold higher antagonist concentrations than inhibition of values were similar to the concentrations causing inhibition of binding. DNA synthesis. Substance P also inhibited the binding of GRP. This finding was not expected, since it has been reported that substance P at 30 ^M does not interact with bombesin receptors DISCUSSION on pancreatic acinar cells from guinea pig (12) and the IC5o for The major findings of this study are that SCLC cell lines do substance P inhibition of [l25I]Tyr4-bombesin binding to rat not respond to exogenous bombesin in a liquid tissue culture brain was greater than 10 /AI (16). Zachary and Rozengurt did system and although their growth is inhibited by bombesin not examine the effect of substance P on [125I]GRP binding to antagonists, this inhibition is probably not mediated by the 3T3 cells at greater than 1 /Â¿M(15). We infer from these results bombesin receptor. The liquid culture assay was able to detect that either the receptor on murine 3T3 cells is different from responses to growth factors, as shown by the responses to ITS the receptor on rat brain and guinea pig pancreatic cells or the and FCS. Weber et al. have previously reported that GRP binding of GRP is in some way different from that of Tyr4- stimulated the growth of two SCLC cell lines in a liquid culture bombesin. system similar to the one used in this study, although serum- Although both 3T3 cells and SCLC cells express GRP recep supplemented RPMI was used instead of serum-free HITES tors, only SCLC cells produce GRP and thus the effects of medium (10). They found growth stimulation over a narrow antagonists on the two cell types may differ. Cuttitta et al. have 4787

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1988 American Association for Cancer Research. BOMBESIN ANTAGONISTS ON GROWTH OF SMALL CELL LUNG CANCER described a monoclonal antibombesin antibody that inhibits the possible that substance P and its analogues interact with a growth of SCLC cell lines in vitro and in vivo (8). The antibody growth factor receptor in addition to the bombesin receptor, inhibited in vitro growth in the absence of exogenous bombesin and that this interaction is responsible for the growth stimula and also inhibited a cell line that gave a marginal response to tion seen at high concentrations. bombesin, therefore it was reasonable to expect that the bom Overall our results show that small cells are not dependent besin antagonists would be effective at inhibiting growth, even on bombesin in a liquid culture assay. It is not clear which in though we could not demonstrate a response to bombesin by vitro growth conditions best mimic in vivo conditions and the LICM cell lines. However, the inhibition observed was not therefore whether in vivo GRP is an essential autocrine growth reversible by excess bombesin and thus was probably not me factor for SCLC cells. It has also not yet been demonstrated diated by the bombesin receptor. Binding of GRP to LICM 107 that the GRP secreted by SCLC cells is biologically active since was inhibited by the antagonists with an ICso of 4-5 /XM,which in our study and in previous studies (1-7, 9) it has been is similar to the data obtained with Swiss 3T3 cells. This is at quantitated by biochemical or immunological not biological least 10-fold lower than the amount required to give 50% methods. inhibition of proliferation. Such a large discrepancy has not been found in other situations where antagonists have inhibited ACKNOWLEDGMENTS a response to BN (12, 13, 15) and thus it also suggests that the inhibition of proliferation may not be mediated by the BN We thank J. Hansky and A. W. Burgess for helpful comments on receptor. It is possible that the antagonists were acting via a this work, A. Rubenfeld, J. Bishop, and M. Schwartz for tumor speci different growth factor receptor. DAPTL-SP can interact with mens, and J. Gardner and D. McPhee for technical assistance. at least one other growth factor since it has been shown to inhibit the response of Swiss 3T3 cells to vasopressin as well as REFERENCES bombesin (13). Another possibility is that autocrine stimulation of SCLC cells occurs intracellularly as described recently for v- 1. Sorenson, G. D., Pettengill, O. S., Brinck-Johnsen, T., Cate, C. C., and Maurer, L. H. Hormone production by cultures of small cell carcinoma of sis-transformed cells (28). This phenomenon would explain the the lung. Cancer (Phila.), 47:1289-1296, 1981. lack of specific effect of the external bombesin antagonists. 2. Wood, S. M., Wood, J. R., Ghatei, M. A., Lee, Y. C, O'Shaughnessy, D., and Bloom, S. R. Bombesin, somatostatin and neurotensin-like immuno- However, if this were the case, the antibombesin antibody would reactivity in bronchial carcinoma. J. Clin. Endocrino!. Metab., 53: 1310- also be ineffective. Thus there is no evidence that SCLC cell 1312, 1981. lines are dependent on bombesin under the liquid culture con 3. Luster, W., Gropp, C., Kern, H. F., and Havemann, K. Lung tumour cell lines synthesizing peptide hormones established from tumours of four his ditions used in our study. tolÃ³gica!types: characterization of the cell lines and analysis of their peptide The effects of bombesin appear to be dependent on culture hormone production. Br. J. Cancer, 51: 865-875, 1985. 4. Moody, T. W., Pert, C. P., Gazdar, A. F., Carney, D. N., and Minna, J. D. conditions and possibly the cell line tested and cannot neces High levels of intracellular bombesin characterize human small-cell lung sarily be extrapolated to in vivo growth conditions. Based on carcinoma. Science (Wash. DC), 214:1246-1248, 1981. these results, we suggest that the antibody-mediated inhibition 5. Erisman, M. D., Linnoila, R. I., Hernandez, O., DiAugustine, R. P., and Lazarus, L. H. Human lung small-cell carcinoma contains bombesin. Proc. of SCLC cell growth in nude mice seen by Cuttitta et al. (8) Nati. Acad. Sci. USA, 79: 2379-2383, 1982. may be due to indirect effects that result from GRP being 6. Sorenson, G. D., Bloom, S. R., Ghatei, M. A., Del Prete, S. A., Cate, C. C., directly associated with the cell membrane rather than receptor and Pettengill, O. S. Bombesin production by human small cell carcinoma bound. An example of a membrane-bound growth factor is of the lung. Regul. Pept., 4: 59-66, 1982. 7. Yamaguchi, K., Abe, K., Kameya, T., Adachi, I., Taguchi, S., Otsubo, K., membrane-bound interleukin 1, which has been described on and Yanaihara, N. Production and molecular size heterogeneity of immu- noreactive gasirin releasing peptide in fetal and adult lungs and primary lung macrophages that secrete interleukin 1 (29). Indirect effects tumours. Cancer Res., 43: 3932-3939, 1983. such as complement-dependent cell lysis antibody-dependent 8. Cuttitta, F., Carney, D. N., Mulshine, J., Moody, T. W., Fedorko, J., Fischler, killer cell activity or phagocytosis have been observed with the A., and Minna, J. D. Bombesin-like peptides can function as autocrine growth in vivo use of other monoclonal antitumor antibodies (30-32). factors in human small-cell lung cancer. Nature (Lond.), 316:823-826,1985. 9. Moody, T. W., Bertness, V., and Carney, D. N. Bombesin-like peptides and Previous studies on factors affecting the growth of SCLC receptors in human tumor cell lines. Peptides, 4: 683-686, 1983. have not distinguished between growth factor responsiveness 10. Weber, S., Zuckerman, J. E., Bostwick, D. G., Bensch, K. G., Sikic, B. I., and Radin, T. A. Gastrin releasing peptide is a selective mitogen for small and growth factor dependence. It is possible that SCLC cells cell lung carcinoma in vitro. 3. Clin. Invest., 75: 306-309, 1985. can respond to bombesin without being dependent on it. The 11. Carney, D. N., Cuttitta, F., Moody, T. W., and Minna, J. D. Selective stimulation of small cell lung cancer clonal growth by bombesin and gastrin- recently reported results of Carney et al. (Il) show that the releasing peptide. Cancer Res., 47: 821-825, 1987. response to bombesin was not seen in the presence of serum, 12. Jensen, R. T., Jones, S. W., Folkers, K., and Gardner, J. D. A synthetic indicating that other growth factors can substitute for bom peptide that is a bombesin receptor antagonist. Nature (Lond.), 309: 61-63, besin. Fetal calf serum has been shown to contain GRP-related 1984. 13. Corps, A. N., Rees, L. H., and Brown, K. D. A peptide that inhibits the peptides, but the concentrations were far below those required mitogenic stimulation of Swiss 3T3 cells by bombesin or vasopressin. Biochem. J. 231: 781-784, 1985. for the stimulation of SCLC cells (33). The data of Cuttitta et 14. Mizrahi, J., Dion, S., D'Orleans-Juste, P., and Regoli, D. Activities and al. (8) showing that an antibombesin antibody inhibited in vitro antagonism of bombesin on urinary smooth muscles. Eur. J. Pharmacol., growth indicates that small cells may be dependent on GRP Ill: 339-345, 1985. 15. Zachary, I., and Rozengurt, E. High-affinity receptors for peptides of the under some conditions. It would be interesting to know whether bombesin family in Swiss 3T3 cells. Proc. Nati. Acad. Sci. USA, 82: 7616- the addition of other growth factors that these cells have been 7620, 1985. reported to respond to, e.g., ACTH (34) or vasopressin (35), 16. Yachnis, A. T., Crawley, J. N., Jensen, R. T., McGrane, M. M., and Moody, T. W. The antagonism of bombesin in the CNS by substance P analogues. would circumvent the requirement for bombesin. Life Sci., 35: 1963-1969, 1984. An unexpected finding was the partial agonist activity of the 17. Morstyn, G., Brown, J., Novak, U., Gardner, J., Bishop, J., and Garson, M. Heterogeneous cytogenetic abnormalities in small cell lung cancer cell lines. bombesin antagonists and the stimulatory effect of substance P Cancer Res., 47: 3322-3327, 1987. on Swiss 3T3 cells. Substance P and DAPTL-SP by themselves 18. Carney, D. N., Gazdar, A. F., Bepler, G., Guccion, J. G., Marangos, P. J., have been reported to have no effect on these cells (13, 36). Moody, T. W., Zweig, M. H., and Minna, J. D. Establishment and identifi cation of small cell lung cancer cell lines having classic and variant features. Different lines of Swiss 3T3 cells may bear different growth Cancer Res., 45: 2913-2923, 1985. factor receptors and may therefore behave differently. It is 19. Carney, D. N., Bunn, P. A., Gazdar, A. F., Pagan, J. A., and Minna, J. D. 4788

Selective growth in serum-free hormone supplemented medium of tumor 28. Keating, M. T., Williams, L. T. Autocrine stimulation of intracellular PDGF cells obtained by biopsy from patients with small cell carcinoma of the lung. receptors in v-sis-transformed cells. Science (Wash. DC), 239: 914-916, Proc. Nati. Acad. Sci. USA, 78: 3185-3189, 1981. 1988. 20. Soveny, C.. and Hansky, J. Bombesin-like immunoreactivity in rat brain and 29. Kurt-Jones, E. A., Beller, D. I.. Mizel, S. B., and Unanue, E. R. Identification gastrointestinal tract: influence of diet and starvation. Regul. Pep.. 3: 325- of a membrane-associated interleukin-1 in macrophages. Proc. Nat). Acad. 331, 1982. Sci. USA, 82: 1204-1208, 1985. 21. Rabinovitch, M.. and DeStefano, M. J. Use of the local anesthetic lidocaine 30. Bernstein, I. D., Tarn, M. R., and Nowinski, R. C. Mouse leukemia: therapy for cell harvesting and subcultivation. In Vitro (Rockville), //: 379-381, with monoclonal antibodies against a thymus differentiation antigen. Science 1975. (Wash. DC), 207: 68-71. 1980. 22. Munson. P. J., and Rodbard, D. L1GAND: a versatile computerized approach 31. Herlyn, D., Steplewski, /... Herlyn. M., and Koprowski, H. Inhibition of for characterization of ligand-binding systems. Anal. Biochem., 707: 220- growth of colorÃ©ela!carcinoma in nude mice by monoclonal antibody. Cancer Res., Â¥0:717-721. 1980. 239, 1980. 23. Bepler, G.. Carney, D. N., Gazdar, A. F., and Minna, J. D. In vitro growth 32. Masui, H.. Kawamoto, T., Sato, J. D., Wolf. B., Sato, G., and Mendelsohn, J. Growth inhibition of human tumour cells in athymic mice by anti- inhibition of human small cell lung cancer by physalaemin. Cancer Res., 47: 2371-2375, 1987. epidermal growth factor receptor monoclonal antibodies. Cancer Res., 44: 1002-1007, 1984. 24. Jensen, R. T., Moody, T., Pert, C., Rivier, J. E., and Gardner, 3. D. 33. Wiedermann, C. J., Goldman, M. E., and Pert, C. B. Chromatographie Interaction of borni/esili and litorin with specific membrane receptors on analysis of gastrin-releasing peptide in heat-inactivated fetal calf serum. Cell pancreatic acinar cells. Proc. Nati. Acad. Sci. USA, 75: 6139-6143, 1978. Tissue Kinet., 19:467-470, 1986. 25. Moody, T. W., Pert. C. B., Rivier, J., and M. R. Brown. Bombesin: Specific 34. Gropp, C., Luster, W., and K. Havemann. In vitro studies in small cell lung binding to rat brain membranes. Proc. Nati. Acad. Sci. USA, 75:5372-5376, cancer cell lines. Recent Res. Cancer Res., 97: 55-64. 1985. 1978. 35. Carney, D. N., Bepler, G., and Gazdar, A. F. The serum-free establishment 26. Westendorf, J. M., and Schonbrunn, A. Characterization of bombesin recep and in vitro growth properties of classic and variant small cell lung cancer tors in a rat pituitary cell line. J. Biol. Chem., 25Â«:7527-7535, 1983. cell lines. Recent Res. Cancer Res., 99: 157-166. 1985. 27. Moody, T. W.. Carney, D. N., Cuttitta, F., Quattrocchi, K., and Minna, J. 36. Rozengurt, E., and Sinnett-Smith, J. Bombesin stimulation of DNA synthesis D. High affinity receptors for bombesin/GRP-like peptides on human small and cell division in cultures of Swiss 3T3 cells. Proc. Nati. Acad. Sci. USA, cell lung cancer. Life Sci., 37:105-113, 1985. 80: 2936-2940, 1983.

4789

Judith E. Layton, Denis B. Scanlon, Claire Soveny, et al.

Cancer Res 1988;48:4783-4789.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/48/17/4783

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/48/17/4783. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.