Antagonists of bombesin/gastrin-releasing Peptide inhibit growth of jar human choriocarcinoma cells and production of cyclic-amp in-vitro

The effects of bombesin/GRP antagonists RC-3095 and RC-3940-II on the in vitro proliferation of JAR human choriocarcinoma cells were evaluated. Antagonists RC-3095 and RC-3940-II effectively inhibited growth of cultured JAR cells, inducing a dose- and time-dependent decrease in the number of treated cells. RC-3940-II was more potent than RC-3095 in inhibiting the growth of JAR cells. Addition of RC-3940-II to JAR cell cultures significantly inhibited the cell proliferation at concentrations as low as 1 nM, while 10 nM RC-3095 was required for a similar effect. Receptor binding studies demonstrated the presence of a single class of binding sites for bombesin on JAR cells. RC-3940-II displaced [I-125]Tyr(4)-bombesin bound to the receptors. When JAR cells were cultured in the presence of 10 nM RC-3095 or RC-3940-II for 72 h, cAMP levels in the incubation medium were decreased by 70-80%, compared to the controls. These results suggest that bombesin/GRP antagonists RC-3095 and RC-3940-II inhibit the proliferation of JAR human chorionic adenocarcinoma cells in vitro and that these effects may involve intracellular cAMP pathway.     

Systematic development of bombesin/gastrin-releasing peptide antagonists.

Several families of very potent bombesin (Bn) receptor antagonist analogues have recently been developed and their biological potencies evaluated in a number of in vitro systems including guinea pig and rat pancreatic acini and Swiss 3T3 cells. These studies showed that analogues can exhibit diverse properties ranging from full antagonists, partial agonists, or full agonists depending on the assay system and animal species employed. We have developed two classes of more potent, shorter chain antagonists based on [psi CH2NH(13-14)]Bn(6-14) and desMet14Bn(6-13)NH2 structures. [D-Phe6 psi Leu13-Leu14] Bn(6-14)NH2 was a potent antagonist (Ki 6nM) in Swiss 3T3 cells and guinea pig acini but exhibited 10% partial agonist activity and lower binding affinity (Ki 60 nM) in rat acini. The partial agonism could be eliminated by using p-Cl-Phe or D-Phe at the C-terminus and partially eliminated using D-4-Cl-Phe in position 6. With the antagonist [D-Phe6]Bn(6-13)NH2 (Ki 96 nM), alkyl substituents on the amide group increased affinity 25-fold with the propylamide being the most potent peptide (Ki 4 nM) in 3T3 cells or guinea pig acini. It did, however, have high 40% partial agonist activity in rat acini. Alkyl esters or hydrazide derivatives were, in contrast, pure antagonists in all systems tested with [D-Phe6]Bn(6-13)OMe having the highest affinity in all systems and also excellent in vivo properties. All of the potent antagonists examined had little affinity for neuromedin B--preferring bombesin receptors, which had entirely new ligand structure-activity relationships.(ABSTRACT TRUNCATED AT 250 WORDS)     

Stimulation by bombesin and inhibition by bombesin/gastrin-releasing peptide antagonist RC-3095 of growth of human breast cancer cell lines.

Recently, it was reported that bombesin/gastrin-releasing peptide (GRP) have mitogenic effects on some human breast cancer cell lines. In this study, we investigated the effects of bombesin/GRP and its receptor antagonist (RC-3095) on the proliferation of three breast cancer cell lines, MDA-MB-231, MCF-7 MIII, and MCF-7. Stimulation by bombesin and inhibition by RC-3095 of cell growth were found in MDA-MB-231 and MCF-7 MIII cells cultured in phenol red-free medium with 5% heat-inactivated and dextran-coated charcoal-treated fetal bovine serum (DCC-FBS). A stimulatory effect by bombesin was not observed in the presence of untreated FBS. [3H]Thymidine incorporation into DNA in these cells was suppressed by RC-3095. MCF-7 cells failed to respond to bombesin and RC-3095 in the presence of either FBS or DCC-FBS. GRP-like immunoreactivity was found in the cell extracts and FBS, but it was undetectable in DCC-FBS. It appears that the stimulatory effect of bombesin on cell proliferation of MCF-7 MIII and MDA-MB-231 cell lines could be obtained because of reduction in the levels of some serum factors in DCC-FBS. These results suggest that bombesin/GRP can act as growth factors through bombesin/GRP receptors in some human breast cancers.     

Inhibition of growth of human malignant glioblastoma in nude mice by antagonists of bombesin/gastrin-releasing peptide

The effects of antagonists of bombesin/gastrin-releasing peptide (GRP) on the growth of human malignant glioblastoma cell line U-87MG xenografted into nude mice were evaluated. Nude mice bearing s.c. implanted U-87MG tumors were treated with bombesin/GRP antagonists RC-3095 and RC-3940-II. RC-3095 and RC-3940-II administered s.c. at a dose of 20 micrograms/day for 4 weeks decreased the volume of U-87MG xenografts by 60 and 74%, respectively, compared with controls. RT-PCR analysis showed that U-87MG xenografts expressed mRNA for bombesin receptor subtype (BRS)-1 (GRP receptor) and BRS-2 (neuromedin-B receptor), but the mRNA for GRP ligand was not detected in U-87MG cells suggesting that GRP may stimulate the growth of U-87MG glioblastomas by a paracrine mechanism. The levels of mRNA for c-fos oncogene were decreased by 30-40% in U-87MG tumors treated with RC-3095 or RC-3940-II. In U-373MG glioblastoma cells, which also express BRS-1, and U-87MG cells, cultured in vitro, GRP(14-27) induced the expression of c-fos mRNA, and some c-jun mRNA, in a time-dependent manner with the maximal effect occurring 2 h after the stimulation and a return to basal levels after 8 h. Antagonist RC-3940-II inhibited the stimulation of c-fos by GRP(14-27). Our results indicate that antagonists of bombesin/GRP inhibit the growth of U-87MG glioblastomas by a mechanism that may involve the downregulation of c-fos oncogene.     

Characterization of a bombesin/gastrin-releasing peptide receptor on a human gastric-cancer cell line

This study examined the expression of receptors of the bombesin (BBS) family in human gastric-cancer cell lines. Of 5 cell lines screened, only one, St42, demonstrated specific binding sites for ¹²⁵I-Tyr⁴-BBS, which have been further characterized. This binding was saturable, and temperature- and time-dependent. Scatchard analysis of displacement data performed at 37°C revealed 2 binding sites: a high-affinity, low-capacity site (K_D = 0.13 nM, B_max = 1500 sites/cell) and a lower-affinity, higher-capacity site (K_D = 11 nM, B_max = 35,000 sites/cell); the latter was lost when internalization of peptide was prevented, suggesting that it may be an artefact. Displacement assays with gastrin-releasing peptide (GRP) and neuromedin B (NMB) revealed that the receptor was of the GRP-preferring sub-type (GRP IC₅₀ = 0.35 nM; NMB IC₅₀ = 112 nM). Co-valent cross-linking of ¹²⁵I-Tyr⁴-BBS to the receptor demonstrated the presence of a single band corresponding to a molecular weight of 37 to 44 kDa on SDS-PAGE, similar to that of the cloned GRP receptor protein core. G-protein linkage of this receptor was demonstrated by selective inhibition of ¹²⁵I-Tyr⁴-BBS binding by guanosine nucleotides. The binding of BBS to the receptor resulted in a rise in intracellular calcium. Three of four structurally distinct BBS antagonists bound to the receptor with high affinity, but [DPhe¹², Leu¹⁴]-bombesin did not cause any displacement of ¹²⁵I-Tyr⁴-BBS even at 10 mM. The functional significance of GRP receptors on human gastric-cancer cells is as yet unknown, but further studies may determine whether such receptors have importance in the therapy of gastric cancer.     

Selective stimulation of small cell lung cancer clonal growth by bombesin and gastrin-releasing peptide

Human small cell lung cancers (SCLC) produce and secrete the regulatory peptide bombesin (BN) or its mammalian counterpart gastrin-releasing peptide (GRP). In addition, several SCLC tumor lines have been shown to express high affinity receptors for BN/GRP. On the basis of these findings, we investigated the effect of exogenously added BN and GRP on the soft agarose colony growth of a panel of human cell lines. In serum-free defined medium, colony formation of 9 of 10 SCLC cell lines was stimulated up to 150-fold by BN or GRP, with peak colony stimulation observed at 50 nM BN. In contrast, no stimulatory effect of BN was observed on nine non-SCLC cell lines. Although no stimulation of colony growth by BN was seen in serum-supplemented medium, addition of BN to the serum-free medium increased cloning efficiency to that achieved by serum in most of the SCLC cell lines. GRP 1-27, the active mammalian analogue of Bn, stimulated colony growth of SCLC cells similar to the manner of BN, while the physiologically inactive BN analogue, des-Leu (13)-Met (14)-BN, had no effect on colony growth. No correlation was observed in SCLC cell lines between the response of these cells to exogenous BN and the amount of cellular BN/GRP produced or the presence of BN receptors. These data suggest that BN/GRP may in some instances function as an autocrine growth factor for SCLC and indicate new ways for modulating SCLC growth in patients with this tumor.     

Expression of matrix metalloproteinase-9 and bombesin/gastrin-releasing peptide in human prostate cancers and their lymph node metastases

Neuroendocrine differentiation and subsequent excretion of neuropeptides have been demonstrated to be associated with progression of human prostate cancer. Among neuropeptides found to exist in the prostate, bombesin/gastrin-releasing peptide has been shown to upregulate matrix metalloproteinase-9 (MMP-9) in human prostate cancer cell lines. Expression levels of bombesin, MMP-9, and neuron-specific enolase were examined by immunohistochemistry in 41 cases of clinically organ-confined prostate cancers including 9 with microscopic lymph node metastases. Twenty-seven (64%) of the 41 radical prostatectomy specimens were positive for both MMP-9 and bombesin. Expression of these molecules was observed in almost the same population of the cancer cells. The remaining 14 cases were negative for both MMP-9 and bombesin. High-grade tumors (Gleason sum ≥7) were more likely to express MMP-9 and bombesin (21/24 : 88%) than low-grade tumors (Gleason sum ≥6) (7/17 : 41%). In eight of the nine cases with pathological lymph node metastases, expression of MMP-9 and bombesin was also noted in metastatic sites. Neuron-specific enolase was positive in 16 cases (39%) and not always associated with the expression of bombesin. Expression of bombesin and expression of MMP-9 are common in human prostate cancers and may be related to an aggressive phenotype.     

Effect of bombesin,gastrin-releasing peptide (GRP)(14–27) and bombesin/GRP receptor antagonist RC-3095 on growth of nitrosamine-induced pancreatic cancers in hamsters

Female Syrian golden hamsters with N-nitroso-bis (2-oxopropyl) amine (BOP)-induced pancreatic cancers were treated for 2 months with bombesin/gastrin-releasing peptide (GRP) antagonist D-Tpi⁶, Leu¹³Ψ(CH₂NH)Leu¹⁴ bombesin(6-14) (RC-3095). Bombesin and GRP(14–27) were also administered alone and in combination with the antagonist RC-3095. RC-3095 exerted a dose-dependent inhibitory effect on growth of pancreatic cancers. The number of animals with pancreatic cancers was significantly lower in the group treated with 60 μg/day of RC-3095 and the weight of tumorous pancreata was reduced. Administration of bombesin or GRP alone did not stimulate the growth of pancreatic tumors and, in fact, had a slightly suppressive effect on cancers which was significant only in Experiment I. Bombesin and GRP (14–27) given together with RC-3095 did not nullify the inhibitory effect of the antagonist on pancreatic cancer growth. Actually, a greater inhibition of pancreatic tumors was observed after administration of RC-3095 together with bombesin or GRP, than with RC-3095 alone. The mechanism of action of bombesin, GRP, and bombesin antagonists on pancreatic cancers appears to be complex. The inhibitory effect of bombesin antagonists on pancreatic cancer growth was accompanied by a decrease in the binding capacity of EGF receptors in tumor membranes. Administration of bombesin also caused a down-regulation of EGF receptors and the greatest decrease in binding capacity of EGF receptors was observed after treatment with RC-3095 in combination with GRP. Inhibition of pancreatic cancer can thus be tentatively explained by some common pathways in the action of bombesin, GRP and their antagonists, that could be mediated by interference with EGF-receptor mechanisms.     

Effects of somatostatin analogue RC-160 and bombesin/gastrin-releasing peptide antagonists on the growth of human small-cell and non-small-cell lung carcinomas in nude mice.

We investigated the effects of our synthetic bombesin/gastrin-releasing peptide (GRP) antagonists and somatostatin analogue RC-160 on the growth of human small-cell lung carcinoma (SCLC) and non-small-cell lung carcinoma (non-SCLC) lines in nude mice. Athymic nude mice bearing xenografts of the SCLC NCl-H69 line or non-SCLC NCl-H157 line were treated for 5 and 4 weeks, respectively, with somatostatin analogue RC-160 or various bombesin/GRP antagonists. RC-160, administered s.c. peritumorally at a dose of 100 micrograms per animal per day, inhibited the growth of H69 SCLC xenografts as shown by more than 70% reduction in tumour volumes and weights, as compared with the control group. Bombesin/GRP antagonists, RC-3440, RC-3095 and RC-3950-II, given s.c. peritumorally at a dose of 20 micrograms per animal per day, also inhibited the growth of H69 SCLC tumours. RC-3950-II had the greatest inhibitory effect and decreased tumour volume and weights by more than 80%. The growth of H-157 non-SCLC xenografts was significantly reduced by treatment with RC-160, but not with bombesin/GRP antagonist RC-3095. In mice bearing either tumour model, administration of RC-160 significantly decreased serum growth hormone and gastrin levels. Specific high-affinity receptors for bombesin and somatostatin were found on membranes of SCLC H69 tumours, but not on non-SCLC H157 tumours. Receptor analyses demonstrated high-affinity binding sites for epidermal growth factor (EGF) and insulin-like growth factor I (IGF-I) on the membranes of H69 and H157 tumours. EGF receptors were down-regulated on H69 tumours after treatment with RC-160 and bombesin/GRP antagonists. The concentration of binding sites for EGF and IGF-I on the H157 tumours was decreased after treatment with RC-160, but bombesin/GRP antagonist RC-3095 had no effect. These results demonstrate that bombesin/GRP antagonists inhibit the growth of H-69 SCLC, but not of H-157 non-SCLC xenografts in nude mice, whereas somatostatin analogue RC-160 is effective in both tumour models. This raises the possibility that these peptide analogues could be used selectively in the treatment of various subclasses of lung cancer.     

Antagonists of growth hormone-releasing hormone (GH-RH) inhibit IGF-II production and growth of HT-29 human colon cancers

Insulin-like growth factors (IGFs) I and II are implicated in progression of various tumours including colorectal carcinomas. To interfere with the production of IGFs, we treated male nude mice bearing xenografts of HT-29 human colon cancer with various potent growth hormone-releasing hormone (GH-RH) antagonists. Twice daily injections of antagonist MZ-4-71, 10 microg intraperitoneally or 5 microg subcutaneously (s.c.) resulted in a significant 43-45% inhibition of tumour growth. Longer acting GH-RH antagonists, MZ-5-156 and JV-1-36 given once daily at doses of 20 microg s.c. produced a 43-58% decrease in volume and weight of cancers. Histological analyses of HT-29 cancers demonstrated that both a decreased cell proliferation and an increased apoptosis contributed to tumour inhibition. GH-RH antagonists did not change serum IGF-I or IGF-II levels, but significantly decreased IGF-II concentration and reduced mRNA expression for IGF-II in tumours. In vitro studies showed that HT-29 cells produced and secreted IGF-II into the medium, and addition of MZ-5-156 dose-dependently decreased IGF-II production by about 40% as well as proliferation of HT-29 cells. Our studies demonstrate that GH-RH antagonists inhibit growth of HT-29 human colon cancers in vivo and in vitro. The effect of GH-RH antagonists may be mediated through a reduced production and secretion of IGF-II by cancer cells.     

Inhibitory effect of bombesin/gastrin-releasing peptide antagonist RC-3095 and high dose of somatostatin analogue RC-160 on nitrosamine-induced pancreatic cancers in hamsters.

Female Syrian golden hamsters with N-nitrosobis(2-oxopropyl)amine-induced pancreatic cancers were treated for 2 months with new pseudononapeptide bombesin receptor antagonist [D-Tpi6,Leu13 psi (CH2NH)-Leu14]bombesin(6-14)(RC-3095), administered s.c. with implanted osmotic minipumps releasing 20 micrograms/day of the analogue. The results were compared to those obtained by treatment with somatostatin analogue RC-160 (35 micrograms/day and 150 micrograms/day) or [D-Trp6]luteinizing hormone-releasing hormone (25 micrograms/day), which inhibited the growth of pancreatic cancers in our previous studies. A new acetylated somatostatin analogue [formula: see text] (30 micrograms/day) also was used for comparison of therapeutic response. All peptide analogues induced tumor inhibition by at least one of the measured parameters. Bombesin antagonist RC-3095 and high dose of RC-160 (150 micrograms/day) had the greatest inhibitory effect on pancreatic cancers: A significant decrease in the number of animals with tumors, reduced pancreatic weight, 87-89% inhibition of tumorous pancreas weight, and a significant diminution in the number of tumor nodules and argyrophilic nucleolar organizer region count in tumor cell nuclei were observed in the groups treated with these regimens. We were able to detect receptors for bombesin in membranes of N-nitrosobis(2-oxopropyl)amine-induced pancreatic tumors and these receptors were not down-regulated after treatment with the bombesin antagonist. In hamsters treated with bombesin antagonists, tumor inhibition might be explained by a significant decrease in the binding capacity of epidermal growth factor receptors in pancreatic cancers. The acetylated somatostatin analogue RC-160-II had a similar inhibitory effect on the tumors as the original analogue RC-160. Our results suggest that the increase in the dose of RC-160 improves the therapeutic response, and this finding should be taken into account in clinical use of this somatostatin analogue. In view of its strong inhibitory effect on experimental pancreatic tumors, the bombesin antagonist RC-3095 might be considered as a possible new agent for the therapy of human exocrine pancreatic cancer.     

Antagonists of growth hormone releasing hormone and bombesin inhibit the expression of EGF/HER receptor family in H-69 small cell lung carcinoma

Effects of in vivo treatment with antagonists of growth hormone-releasing hormone (GHRH), JV-1-65 and MZ-J-7-110, and bombesin/gastrin-releasing peptide antagonist RC-3940-II, on the EGF receptor (EGFR) family, were investigated in H-69 SCLC. Tumors were analyzed by RT-PCR, immunoblotting and binding assays. Treatment with these analogs reduced the binding capacity of EGFR by 18-64%, and inhibited the mRNA expression for EGFR, HER-2 and -3 by 27-75.4, 17-26.3, and 13.8-46.6%, respectively. The antagonists also decreased the protein levels for EGFR by 21-34%, HER-2 by 36-68% and HER-3 by 43-49%. This is the first demonstration that antiproliferative effects of GHRH antagonists are associated with a downregulation of EGF/HER receptors.     

Antisense oligonucleotides to gastrin inhibit growth of human pancreatic cancer

Human pancreatic cancer is stimulated by the autocrine production of gastrin. In this study, the effects of administration of antisense oligonucleotides to gastrin on growth of pancreatic cancer were evaluated in vitro and in vivo. Log phase BxPC-3 human pancreatic cancer cells in culture were exposed to increasing concentrations (0.5-10 microM) of a synthetic 20-mer antisense phosphorothioate oligonucleotide to gastrin for 48 h and growth was assessed by the cellular proliferation assay. Growth was inhibited up to 88% by anti-gastrin oligonucleotides in a dose-related fashion compared to cells treated with diluent or a randomized sequence with the same composition as the anti-gastrin oligonucleotide. In vivo nude mice bearing BxPC-3 xenografts were treated daily for 14 days with a 0.1-ml intratumoral injection of either anti-gastrin (5 microM), the scrambled sequence control phosphorothioate oligonucleotide (5 microM), or buffer. Tumors from the anti-gastrin-treated mice were significantly smaller in volume and weight and had less gastrin detected by radioimmunoassay than either controls. These results support the role of gastrin as a stimulatory peptide for growth of human pancreatic cancer. Antisense oligonucleotide to gastrin may have a role in the future treatment of patients with pancreatic cancer.     

Antagonists of growth hormone-releasing hormone inhibit the growth of U-87MG human glioblastoma in nude mice

Antagonists of growth hormone-releasing hormone(GH-RH)inhibit the growth of various cancers by mechanisms that involve the suppression of the insulin-like growth factor (IGF)-I and/or IGF-II. In view of the importance of the IGF system in glioma tumorigenesis, the effects of GH-RH antagonists MZ-5-156 and JV-1-36 were investigated in nude mice bearing subcutaneous and orthotopic xenografts of U-87MG human glioblastomas. After 4 weeks of therapy with MZ-5-156 or JV-1 -36 at the dose of 20 microg/day per animal, the final volume of subcutaneous U-87MG tumors was significantly (P < .01) decreased by 84% and 76%, respectively, as compared with controls. Treatment with GH-RH antagonists also reduced tumor weight and the levels of mRNA for IGF receptor type I (IGFR-I). A reduction in the mRNA levels for IGF-II was found in tumors of mice treated with MZ-5-156. Treatment with MZ-5-156 or JV-1 -36 also extended the survival of nude mice implanted orthotopically with U-87MG glioblastomas by 81% (P < .005) and 18%, respectively, as compared with the controls. Exposure in vitro to GH-RH antagonists MZ-5-156 or JV-1 -36 at 1 microM concentration for 24 hours decreased the tumorigenicity of U-87MG cells in nude mice by 10% to 30% and extended the latency period for the development of subcutaneous palpable tumors by 31% to 56%, as compared with the controls. Exposure of U-87MG cells to GH-RH antagonists in vitro also resulted in a time-dependent increase in the mRNA levels of IGFR-II or a decrease in the mRNA levels of IGFR-I. mRNA for GH-RH was detected in U-87MG cells and xenografts implying that GH-RH may play a role in the pathogenesis of this tumor. Our results suggest that GH-RH antagonists MZ-5-156 and JV-1-36 inhibit the growth of U-87MG human glioblastoma by mechanisms that involve the suppression of IGF system. Antagonistic analogs of GH-RH merit further development for the treatment of malignant glioblastoma.     

Role for cyclic AMP in the postreceptor control of cytokine-stimulated stromal cell growth factor production

We examined the role of augmented formation of intracellular cyclic AMP (cAMP) in the mediation of stromal cell growth factor production that occurs constitutively or upon cytokine stimulation. Clonal murine marrow adherent cell lines were stimulated under serum-free conditions by interleukin-1 (IL-1) or lipopolysaccharide (LPS) and one (+/+ -1.LDA11) was found to produce low quantities of granulocyte macrophage colony-stimulating factor (GM-CSF). GM-CSF identity was confirmed by the ability of supernatants from stromal cells to promote proliferation of the factor-dependent cell line FDC-P1, neutralization of this activity by antiserum to GM-CSF, and by Northern blot analysis. However, optimal concentrations of IL-1 and tumor necrosis factor-alpha (TNF-alpha), in combination, led to synergistic (greater than 5-fold higher quantity) GM-CSF production compared with either stimulus alone in the +/+ -1. LDA11 cell line, capable of GM-CSF production after only single stimulation with IL-1 or LPS. In addition, synergistic stimulation by IL-1 and TNF-alpha led to equivalent high amounts of GM-CSF in another cell line incapable of GM-CSF production after induction with only IL-1 or LPS. Any of several means to raise intracellular cAMP levels, including addition of 8-bromo-cyclic AMP (8Br cAMP) (0.25-1mM), pertussis toxin (20-100 ng/ml), or addition of prostaglandin E1 (PGE1) (1 microM), failed to stimulate GM-CSF production alone and strongly inhibited GM-CSF production in stromal cells stimulated by IL-1, LPS, or the synergistic combination of IL-1 and TNF-alpha. In addition, PGE1 and pertussis intoxication were agonists of adenylate cyclase in membranes of marrow adherent cells, whereas IL-1 and LPS were not. The role for regulators of intracellular cAMP was specific because any of the cAMP agonists alone, or in the presence of cytokine stimulators of stromal cells, strongly enhanced IL-6 production, an event known to be cAMP-responsive. Thus, acute formation of intracellular cAMP is a negative regulator of stromal cell GM-CSF production mediated by cytokines, but positively regulates IL-6 production and may be an important determinant of cytokine-directed marrow microenvironmental function. These findings on the requirement for augmentation versus inhibition of cytokine-mediated production of hemopoietic growth factors might be applied to an analysis of marrow stromal cell heterogeneity.     

Antagonists of growth hormone-releasing hormone (GH-RH) inhibit in vivo proliferation of experimental pancreatic cancers and decrease IGF-II levels in tumours

Insulin-like growth factors (IGF-I and IGF-II) are implicated in the pathogenesis of pancreatic carcinoma. Antagonists of growth hormone-releasing hormone (GH-RH) suppress the GH-RH-GH-IGF-I axis and also act directly on tumours to reduce production of IGF-I or II. The aim of this study was to investigate the effects of two potent GH-RH antagonists in two experimental models of pancreatic cancer. Syrian golden hamsters with nitrosamine-induced pancreatic tumours were treated with 10 micrograms/day of GH-RH antagonist MZ-4-71 for 60 days. The therapy reduced the number of tumorous animals, decreased the weight of tumorous pancreata by 55%, and lowered AgNOR numbers in tumour cells. In two other experiments, GH-RH antagonists MZ-4-71 and MZ-5-156 significantly inhibited growth of SW-1990 human pancreatic cancers xenografted into nude mice, as shown by a reduction in tumour volume and tumour weights, and a decrease in AgNORs in cancer cells. IGF-I levels in serum and in pancreatic cancer tissue remained unchanged after therapy, suggesting that an effect on IGF-I is not involved in tumour inhibition. In contrast, IGF-II concentrations in tumours were significantly reduced by 50-60% after treatment with the GH-RH antagonists as compared with controls. In vitro studies showed that the concentration of IGF-II in the culture medium was increased after seeding of SW-1990 cells, indicating that this pancreatic cancer cell line produced and released IGF-II. This finding was also supported by the expression of IGF-II mRNA in the SW-1990 cells. Addition of 3 x 10(-6) M of GH-RH antagonist MZ-5-156 to the reduced-serum medium decreased cell proliferation, IGF-II mRNA expression in the cells and IGF-II concentration in the medium. Our findings indicate that inhibitory effects of GH-RH antagonists on the growth of experimental pancreatic cancers, may result from a decrease in the production and concentration of IGF-II in the tumours.