Expression of V1A and GRP receptors leads to cellular transformation and increased sensitivity to substance-P analogue-induced growth inhibition

Small-cell lung cancer (SCLC) is a particularly aggressive cancer, which metastasises early. Despite initial sensitivity to radio- and chemo-therapy, it invariably relapses, so that the 2-year survival remains less than 5%. Neuropeptides particularly arginine vasopressin (AVP) and gastrin-releasing peptide (GRP) act as autocrine and paracrine growth factors and the expression of these and their receptors are a hallmark of the disease. Substance-P analogues including [D-Arg1,D-Phe5,D-Trp7,9,Leu11]-substance-P (SP-D) and [Arg6,D-Trp7,9,NmePhe8]-substance-P (6–11) (SP-G) inhibit the growth of SCLC cells by modulating neuropeptide signalling. We show that GRP and V1A receptors expression leads to the development of a transformed phenotype. Addition of neuropeptide provides some protection from etoposide-induced cytotoxicity. Receptor expression also leads to an increased sensitivity to substance-P analogue-induced growth inhibition. We show that SP-D and SP-G act as biased agonists at GRP and V1A receptors causing blockade of Gq-mediated Ca2+ release while directing signalling to activate ERK via a pertussis toxin-sensitive pathway. This is the first description of biased agonism at V1A receptors. This unique pharmacology governs the antiproliferative properties of these agents and highlights their potential therapeutic potential for the treatment of SCLC and particularly in tumours, which have developed resistance to chemotherapy.

Lung cancer is the most common fatal malignancy in the developed world. Small-cell lung cancer (SCLC), which constitutes 25% of the total, is a particularly aggressive form of lung cancer. It metastasises early and over 90% of patients have widespread metastasis at presentation precluding curative surgery. Despite initial sensitivity to radio-and chemo-therapy, SCLC almost invariably relapses, so that the 2-year survival remains less than 5% (Smyth et al, 1986). Small-cell lung cancer is a paradigm for neuropeptide-driven tumourigenesis. Small-cell lung cancer cells proliferate in response to a range of neuropeptide growth factors and in many cases these neuropeptides are involved in autocrine and paracrine growth loops, which fuel unrestrained proliferation (Moody et al, 1981;Sethi and Rozengurt, 1991;Sethi et al, 1992;North, 2000). This is thought to be part of a special process of oncogenic transformation called Selective Tumour gene Expression of Peptides essential for Survival (STEPS) (North, 2000). The expression of neuropeptides and their receptors in tumour cells may render them sensitive to pharmacological therapeutic intervention and may be of diagnostic importance for the early detection of lung cancer and for the selection of appropriate treatment. Much of the recent studies in SCLC have focused on the roles of the specific mitogens gastrin-releasing peptide (GRP, the mammalian homologue of bombesin) and arginine vasopressin (AVP).
Gastrin-releasing peptide receptors are frequently aberrantly expressed in human neuroendocrine tumours of the breast, prostate, lung and colon (Carroll et al, 1999;Jensen et al, 2001) where they have both mitogenic and morphogenic roles and confer a survival advantage in proliferating cancer cells. It has been suggested that the detection of these markers in the peripheral blood of patients may be useful as early markers of SCLC. Scintigraphy with the labelled peptide ligand 99mTc-bombesin/ GRP has been shown to detect prostate cancer and invasion of pelvic lymph nodes and should be applicable to other endocrine tumours particularly SCLC (De Vincentis et al, 2004). Monoclonal antibodies have been developed against circulating bombesin and one such antibody 2A11 has been shown to inhibit the growth of SCLC in vitro and as xenografts in nude mice (Chaudhry et al, 1999); however, it has limited efficacy in human trials.
The presence of AVP and V 1A receptors on SCLC tumours has been extensively studied and the potential presence of an autocrine growth loop has been established. Small-cell lung cancer patients frequently display symptoms of inappropriate antidiuretic hormone secretion such as hyponatremia and urinary hyperosmolality (Johnson et al, 1997). Independent studies have shown the expression of V 1A receptors in five out of five SCLC lines and zero out of four non-small-cell lung cancer (NSCLC) lines (Ocejo-Garcia et al, 2001) while we showed expression of V 1A receptors in four out of four SCLC lines (Waters et al, 2003). Expression of V 1A receptors and vasopressin is the most useful diagnostic tool for differentiating SCLC from NSCLC and other cancers (Coulson et al, 2003). Therapy that targets V 1A receptors may be of potential benefit for SCLC. Small-cell lung cancer cells can also express the AVP gene as provasopressin, which remains attached to the cell membrane and conceivably contributes to the autocrine-driven mitogenesis (Friedmann et al, 1994). Antibodies recognising this cell surface antigen have been developed as a potential diagnostic and therapeutic tool that targets SCLC tumours in vivo (Keegan et al, 2002).
However, the main focus of research in this drug development strategy has been in the development of broad-spectrum neuropeptide antagonists. Synthetic analogues of substance-P for example, [Arg 6 ,D-Trp 7,9 ,N me Phe 8 ]-substance-P (6 -11) (SP-G) were initially identified as antagonists of substance-P-mediated cellular effects and were subsequently found to also antagonise the cellular effects of bombesin (Jensen et al, 1984). SP-G and its analogue SP-D inhibit calcium mobilisation stimulated by bombesin and vasopressin in SCLC cells (Woll and Rozengurt, 1988;Langdon et al, 1992;Sethi et al, 1992). They also inhibit mitogenesis by the same neuropeptides in both Swiss 3T3 cells and SCLC cells (Woll and Rozengurt, 1988;Sethi et al, 1992;Seckl et al, 1995) In addition to the in vitro growth inhibitory effects of substance-P analogues, these compounds inhibit the growth of tumours in xenograft models in nude mice  and are more effective than specific neuropeptide receptor antagonists. SP-G has recently completed a phase I clinical trial where it shows minimal toxicity at therapeutic plasma concentrations (Clive et al, 2001). However, its mechanism of action is still being investigated.
Our previous work has shown that SP-G and SP-D do not act as simple competitive antagonists of GRP receptors but rather act as biased agonists, inhibiting GRP-stimulated PLC activation via G q while directly stimulating JNK and ERK via G 12 and G i , respectively (Tallet et al, 1996;Jarpe et al, 1998;MacKinnon et al, 1999MacKinnon et al, , 2001 The prolonged stimulation of JNK and ERK coupled with an inhibition of intracellular Ca 2 þ is fundamental for the antiproliferative and proapoptotic effects of these agents (MacKinnon et al, 2001, Waters et al, 2003. It is not known whether these compounds interact with vasopressin receptors in the same fashion.
It was the objective of the present study to examine the effect of GRPr and V 1A r expression, the two most common neuropeptide receptors present on SCLC, on growth and transformation of the epithelial cell line, CHO-K1. We show that expression of these receptors leads to an increase in basal and anchorage-independent growth in low serum, and an increased sensitivity to substance-P analogues. As well as directing GRP receptor signalling, substance-P analogues also act as biased agonists at V 1A vasopressin receptors. These combined effects will be useful for SCLC therapy and particularly in targeting well-differentiated tumours, which have developed resistance to chemotherapy.

Cell culture and transfection
CHO-K1 cells were maintained in DMEM supplemented with 10% (v v À1 ) foetal bovine serum (heat-inactivated at 571C for 1 h) 50 U ml À1 penicillin, 50 mg ml À1 streptomycin and 5 mg ml À1 Lglutamine in a humidified atmosphere of 5% CO 2 :95% air at 371C. CHO-K1 cells were transfected with full-length human GRP receptor or human V 1A receptor using lipofectamine plus (Invitrogen) as per the manufacturer's instructions. Stable cell cultures were maintained in the presence of 400 mg ml À1 G418sulphate.

Receptor binding
Receptor binding was carried out in confluent and quiescent cultures of CHO-K1-V 1A r and GRPr cells in a binding medium containing DMEM, 1 mg ml À1 bovine serum albumin and radioligand (1.0 nM GRP containing 2nCi [ 125 I]-GRP or 0.6 nM AVP containing 45 nCi [ 125 I]-AVP. Incubations were carried out at 371C for 30 min in the presence of inhibitors as indicated. Nonspecific binding was defined in the presence of 1 mM GRP or AVP, respectively. The reaction was stopped on ice and unbound ligand was removed by washing Â 3 with ice-cold phosphate-buffered saline. After solubilisation in 0.1 M NaOH, 2% Na 2 C0 3 containing 1% SDS, bound ligand was estimated by liquid scintillation counting. The binding parameters K d and B max were calculated from competition binding isotherms with unlabelled ligand (DeBlasi et al, 1989). The IC 50 (concentration of drug displacing 50% specific binding) was converted to the inhibitory constant (Cheng and Prusoff, 1973).

Growth assays
Liquid growth Exponentially growing CHO-K1 cells were trypsinised and suspended in DMEM with 5% FCS at a density of 5 Â 10 4 cells per plate in the presence or absence of mediators in triplicate. Cells were grown for 1 -9 days and cell number determined using a Coulter Counter (model Z1, Coulter).
Clonogenic assay CHO-K1 (2x10 4 ) viable cells were mixed with DMEM containing 0.3% agarose in the presence or absence of mediators and layered over a solid base of 0.5% agarose in DMEM in 35 mm plastic dishes. The cultures were incubated at 371C for 1 -10 days, and then stained with 1 mg ml À1 nitro-blue tetrazolium (NBT, Sigma) overnight at 371C. Colonies from 10 separate fields were counted using a microscope with a Â 4 objective. Cloning efficiency is calculated as the percentage of original number of seeded cells forming colonies of 46 cells.
Aggregation assay CHO-K1 cells were suspended in DMEM in the presence of 5% FCS and seeded into low adhesion tissue culture plates on top of a layer (1 ml) of 0.5% agar. Under these conditions, the cells did not adhere. Cells were maintained in culture for 7 days briefly trypsinised to disaggregate clusters and viable cells counted.

Determination of intracellular Ca 2 þ concentration
CHO-K1 cells expressing the GRP or V 1A receptor were grown to confluence on 10 cm plates and quiesced overnight in DMEM containing 0.1%. FCS. Cells were trypsinised and loaded with fura-2-tetraacetoxymethylester AME (FURA-2-AM, 1 mM) in calciumfree Hank's balanced salt solution for 10 min at 371C. The cells were pelleted and resuspended in 2 ml of Hank's balanced salt solution containing 1.8 mM CaCl 2 . Fluorescence was recorded in a fluorescence spectrophotometer (Perkin Elmer). Alternate dual wavelength excitation at 380 and 410 nm allowed ratiometric analysis of bound and unbound FURA-2AM when measured at 505 nm. [Ca 2 þ ] was calculated according to the equation where F is the ratio of the unknown sample, F max is the ratio after the addition of 0.1% Triton X-100 and F min is the ratio after Ca 2 þ chelation with 10 mM EGTA. K is the dissociation constant for Fura-2, which is 224 nM.

Western blotting
Quiescent cell cultures in six-well plates were treated as described in figure legends and lysed at 41C in 0.25 ml lysis buffer containing; 25 mM HEPES pH 7.4, 0.3 M NaCl, 1.5 mM MgCl 2 , 0.2 mM EDTA, 0.5% Triton X-100, 20 mM b-glycerphosphate, 0.5 mM dithiothreitol, 1 mM sodium orthovanadate and protease inhibitors (Boehringer Mannheim, Sussex, UK; prepared as per the manufacturer's instructions). Lysates were clarified by centrifugation, equilibrated for protein using BCA protein assay reagent (Perbio Science, Cheshire, UK) and denatured by boiling (5 min) in SDS -PAGE loading buffer. In all, 20 ml lysate/lane was resolved on 12% SDS -PAGE gels and electroblotted onto nitrocellulose membranes.

Receptor binding
To assess receptor expression, stable cultures of CHO-K1-GRP and CHO-K1-V 1A cells were incubated with the appropriate radioligand, and receptor number and affinity were measured from competition binding curves with unlabelled ligand. Figure 1 shows that GRP receptors were expressed with K d ¼ 2.5570.84 nM and B max of 11517326 sites/cell (n ¼ 4). V 1A receptors were expressed at a slightly lower level with a B max ¼ 450780 sites/cell and K d ¼ 2.9870.71 nM (n ¼ 4, Figure 1B). The substance-P analogues SP-G and SP-D inhibited [ 125 I]-GRP binding with affinities in the micromolar range (SP- . This data show that SP-G is relatively (six-fold) more selective for the V 1A receptor whereas SP-D is GRP receptor selective (13-fold). Effects of V 1A and GRP receptor expression on cell growth in liquid culture and semisolid media Figure 2A shows that in liquid culture, in the presence of 5% FCS, the GRP and the V 1A expressing cells proliferated at a similar rate to untransfected CHOs, but on day 6 after seeding, the GRP cells had reached a density of 4.5 Â 10 6 cells per 100 mm dish. The cells did not appear to be contact inhibited and exhibited a rounded morphology. Cells over grew one another and significant numbers of cells were observed to be detached and growing in clusters ( Figure 2B). Trypan blue exclusion revealed that the cells were 495% viable. A similar, though less pronounced, effect was observed with the V 1A -transfected cells, which reached a density of 3.1 Â 10 6 cells per 100 mm dish. The parental cell line exhibited normal contact inhibition and had flat fibroblast morphology and did not show significant detachment. The end point density of these cells was 2.1 Â 10 6 cells per 100 mm dish. These observations suggested that transfection of GRP and V 1A receptors caused CHO-K1 cells to lose contact inhibition and adopt a more transformed phenotype. These findings led us to examine their growth in semisolid media, which is widely regarded as being a hallmark of the transformed phenotype. Cells were trypsinised and suspended in 0.3% agarose in DMEM (1 Â 10 4 cells ml À1 ) layered over a 0.5% layer of agarose in DMEM. Figure 2C shows that at day 9 after seeding, clusters of cells (46 cells) were visible with MTT staining. The cloning efficiency was significantly higher in the GRPexpressing cells (14.171.5%, Po0.05, ANOVA) and the V 1A receptor-expressing cells (8.571.0%, Po0.05, ANOVA) than the vector control cells in the presence of 5% FCS (5.570.02%). In a separate series of experiments, the effects of SP-D and SP-G were examined ( Figure 3A). At 30 mM, both analogues significantly reduced colony formation in the GRP and V 1A receptor expressing cells but had no effect on the cloning efficiency of vector control CHO-K1 cells. The effect of receptor expression was also examined in an aggregation assay where cells are prevented from adhering and form aggregates in suspension. After 6 days, cells are briefly trypsinised and viable cells counted. These experiments showed that expression of the GRPr and V 1A r increased anchorageindependent growth compared to wild-type cells or vector control cells. Similarly SP-G and SP-D were able to inhibit cell growth in receptor expressing cells ( Figure 3B). Together, these data suggest that expression of the GRP and V 1A receptors induce a more transformed phenotype in CHO cells and induce sensitivity to substance-P analogue-induced growth inhibition.

Chemosensitivity
The response to etoposide in control and receptor-transfected cells was measured by MTT accumulation. Figure 4 shows that after 48 h in culture in the absence of serum, etoposide produced a dosedependent inhibition of proliferation in all cell types (IC 50 ¼ 12.473.1,8.172.3 and 14.274.0 mg ml À1 in CHO-WT, CHO-GRPr and CHO-V 1A r cells, respectively). Incubation with 50 nM of either bombesin or AVP added to CHO-GRPr or CHO-V 1A r produced a small but significant protection from etoposide, which was not observed when both neuropeptides were added to wild-type cells (IC 50 ¼ 13.0 and 26.9 mg ml À1 in control and AVPtreated V 1A r-expressing cells, respectively; and 6.30 and 12.7 mg ml À1 in bombesin-treated GRPr-expressing cells, respectively). At 40 mg ml À1 etoposide, vasopressin-treated V 1A r-expressing cells accumulated 92% more MTT than control cells (Po0.01). In GRPr-expressing cells, bombesin treatment caused an increase of 52% MTT accumulation compared to untreated cells (Po0.01). These results suggest that neuropeptide receptor stimulation may contribute to an increase in chemoresistance.

Intracellular [Ca 2 þ ] i
The mobilization of calcium from intracellular stores is one of the earliest events triggered by neuropeptide receptor activation of G aq leading to PLC activation and subsequent generation of IP 3 . In untransfected CHO-K1 cells, neither GRP nor AVP produced a significant change in [Ca 2 þ ] i (data not shown). In the CHO-GRP cells, GRP produced a concentration-dependent increase in [Ca 2 þ ] i (EC 50 ¼ 2.0070.4 nM, n ¼ 5, Figure 5A). SP-D and SP-G produced no change in [Ca 2 þ ] i on their own but inhibited GRPinduced [Ca 2 þ ] i elevation. Figure 5A and B show that SP-D and SP-G inhibited GRP-induced [Ca 2 þ ] i with resultant pA 2 values of 7.21 and 5.72 for SP-D and SP-G respectively. In V 1A -expressing cells, AVP increased calcium mobilisation with an EC 50 ¼ 3.170.60 nM (n ¼ 4, Figure 5C and D). As in the GRPexpressing cells, SP-D and SP-G did not mobilise intracellular calcium directly but inhibited the response to AVP ( Figure 5C and D). SP-G was more efficacious in inhibiting the vasopressin response than the GRP response. PA 2 values for AVP antagonism were 7.23 and 6.53 for SP-D and SP-G, respectively. Together, these data suggest that when stably transfected into CHO cells the GRP and the V 1A receptor effectively couple to G aq to increase intracellular calcium and that the substance-P analogues are effective antagonists of this response.

ERK activation
AVP and bombesin have been shown to activate extracellular regulated protein kinases (ERK 1 and 2) in a variety of cell types expressing the receptors. Much of these data suggest a G q -and PKC-dependent mechanism although these observations may be cell type-dependent (Della Rocca et al, 1999;Sinnett-Smith et al, 2000). Previous studies have shown that substance-P analogues reversibly inhibit ERK activation by neuropeptides in Swiss 3T3 cells; however, our studies suggest that in rat-1 fibroblasts and in human small-cell lung carcinoma cells (SCLC), in the absence of bombesin, substance-P analogues can activate ERK and c-jun kinase (JNK) in a GRP receptor-dependent manner (MacKinnon et al, 2001). In this study, ERK activation was measured in CHO cells expressing GRP or V 1A receptors in response to the appropriate neuropeptide or substance-P analogue. Activation was estimated in immunoblots from stimulated lysates using a phosphorylation state-specific antibody. Lysates from quiesced cells stimulated for 10 min with either neuropeptide or SP-D or SP-G were separated on 12% gels and blotted for pERK2 or total ERK1/2. Figure 6A shows that bombesin stimulated ERK phosphorylation in GRP-expressing CHO cells. Quantification of five separate experiments showed that bombesin produced a maximal 3.5-fold increase in ERK phosphorylation at 1 nM with an EC 50 of 0.56 nM. In V 1A -expressing cells, AVP produced an increase in ERK phosphorylation of 4.2-fold with an EC 50 of 0.72 nM, which correlated with its affinity for V 1A receptors measured in binding studies and for the stimulation of intracellular calcium. Neither AVP nor GRP produced any stimulation of ERK phosphorylation in untransfected CHO cells (data not shown).
SP-D and SP-G stimulated ERK activation in GRP receptor expressing CHO cells ( Figure 6B). Activation of ERK was evident at 3 mM for SP-D, but the maximal stimulation was less than that observed with bombesin (2.8-fold for SP-D and 2.1-fold for SP-G at 50 mM). The substance-P analogues also stimulated ERK in V 1Aexpressing cells. However, in these cells SP-G was more efficacious than SP-D giving a 4.3-fold stimulation at 50 mM compared to a 2fold stimulation by the same concentration of SP-D. The observed increase in ERK phosphorylation was not due to a change in total ERK expression, as ERK1/2 immunoreactivity did not change with drug treatment. These data show that at concentrations, which block neuropeptide-stimulated calcium release, substance-P analogues are agonists for ERK activation in GRP and V 1A expressing CHO cells.

Mechanisms of ERK phosphorylation
The mechanisms of GRP-and V 1A receptor-mediated ERK activation are not fully understood, although it has been shown in some cell types that activation is mediated by G q -induced PKC activation and subsequent activation of raf. Other GPCRs have been shown to activate ERK in a G i and ras-dependent mechanism, which may or may not involve activation of PI3 K (Della Rocca et al, 1999). To examine the possible recruitment of G i , CHO cells were incubated overnight in quiescent media containing 100 ng ml À1 pertussis toxin. Figure 7 shows that in GRP receptor expressing cells, PTX produced a small but nonsignificant inhibition of bombesinstimulated ERK activation whereas the response to 10 mM SP-D was completely abolished (Po0.01, ANOVA). In V 1A -expressing cells, the response to AVP was not inhibited by PTX pretreatment but the response to SP-G (20 mM) was completely inhibited (Po0.01, ANOVA). In both cell types, ERK activation by 1 mM LPA was completely inhibited by PTX treatment (data not shown). This differential sensitivity to pertussis toxin indicates that dual G i /G qcoupled mechanisms activate the ERK 1/2 cascade via GRP and V 1A receptors expressed in CHO cells.

DISCUSSION
This study utilised a model CHO cell system for neuropeptide receptor expression as these cells have an acceptable null background for these receptors. The results from this work shows that (1) GRP and V 1A receptor expression leads to the development of a transformed phenotype in CHO-K1 cells.
(2) Receptor expressing cells showed some increased resistance to the chemotherapeutic agent etoposide and increased sensitivity to the substance-P analogues, SP-D and SP-G. (3) SP-D and SP-G act as biased agonists at GRP and V 1A receptors. This is the first demonstration of biased agonism at vasopressin receptors. (4) This pharmacological activity is crucial for the antiproliferative effect of these agents, which may be of particular benefit in more differentiated cancers that have developed resistance to chemotherapy.
Introduction of the neuropeptide receptors into CHO cells lead to an increase in proliferation and an increase in the ability of cells to grow as colonies in semisolid medium. The cells also showed a greater propensity to grow as nonadherent clusters. Colony growth was also evident in low serum suggesting that expression of these receptors allows for anchorage-and serum-independent growth consistent with transformation. Moreover, it suggests that these 10 −11 10 −10 10 −9 10 −8 10 −7 10 −6 10 −11 10 −10 10 −9 10 −8 10 −7 10 −6 10 −11 10 −10 10 −9 10 −8 10 −7 10 −6 10 −11 10 −10 10 −9 10 −8 10 − receptors show some constitutive activity, as wild-type receptor expression alone was sufficient to increase transformation in the absence of exogenously added neuropeptide. The ability to observe constitutive GPCR activity is not only determined by the intrinsic molecular properties of the GPCR studied but also by the specific cellular background. Apparent constitutive activity of a wild-type GPCR may vary substantially in various systems. Differences in GPCR expression level, GPCR desensitisation, G-protein complement and GPCR/G-protein stochiometry can all contribute to such differences (Seifert and Wenzel-Seifert, 2002) This could explain why GRP receptors expressed in NCM460 cells, a nonmalignant human colon epithelial cell line, are constitutively active (Ferris et al, 1997) but are not in other cell types (Benya et al, 1994). Our findings are unlikely to be due to an overexpression artefact as the receptors in this study were expressed at levels not greater than those measured from a variety of other cancers (100 -50 000 sites/ cell, Jensen et al, 2001) So at these moderate levels, wild-type receptor was sufficient to activate downstream signals that promote cell survival under anchorage-independent conditions. The cells were, however, still able to respond to agonist, which resulted in a small but significant increase in resistance to the cytotoxic effects of etoposide. It is unclear how significant this change in sensitivity relates to the development of chemoresistance in vivo, but highlights an important area for future investigation.
Our previous work shows that sensitivity to AVP and GRP increases during the progression to chemoresistance in a set of SCLC cell lines (Waters et al, 2003). This would be analogous to the development of autocrine and paracrine growth loops in tumour development in vivo, with tumour cells becoming more dependent on neuropeptide growth factors for survival and subsequently developing chemoresistance. The mechanisms governing the survival advantage gained in neuropeptide receptor expressing cells are poorly understood. During colon cancer progression, increased expression of GRP receptors leads to tumour cell differentiation, increased motility and adhesion to extracellular matrix via an increased activation of focal adhesion kinase (FAK) (Glover et al, 2004). Previous reports have shown that GRP receptors can activate insulin-like growth factor-1 receptors (IGF-IR) leading to Akt activation in prostate cancer cells (Sumitomo et al, 2001), a mechanism which may underlie bombesin-mediated cell survival. In another study, human chorionic gonadotrophin (hCG) increased ovarian cell survival by up-regulating IGF-1 (Kuroda et al, 1998). Conversely, specific bombesin antagonists have been shown to inhibit growth and decrease EGF receptor expression in SCLC cells (Koppan et al, 1998) and breast cancer cells (Bajo et al, 2002). Bombesin antagonists in combination with GHRH antagonists have also been shown to down-regulate IGF-1, IGF-II, IGF receptors, GRP and EGF receptors in SCLC xenografts (Kanashiro et al, 2003). There is therefore potential for neuropeptides to regulate growth factor signalling pathways that control survival. The present study shows that expression of GRP and V 1A receptors in CHO cells also leads to an increased sensitivity to substance-P analogues. This was demonstrated by an inhibition of clonal growth by substance-P analogues only in receptortransfected cells. We previously showed that substance-P analogues are more effective at inhibiting growth of chemoresistant tumour cells expressing high GRP receptor levels and can sensitise cells to etoposide-mediated cell death (Waters et al, 2003). Moreover, diverse tumour types expressing GRP receptors have a greater sensitivity to SP-D (Waters et al, 2003). Together, these data suggest that the increase in neuropeptide receptor expression that occurs as cells become more differentiated may contribute to the development of chemoresistance and may be exploited as a target for drug therapy. The potential role of neuropeptide receptors in the development of chemoresistance is an interesting finding that requires further study.
The mechanism of action of substance-P analogues was studied in receptor-transfected cells. Upon neuropeptide receptor transfection, the cells responded to the appropriate neuropeptide as expected for these predominately G q -coupled receptors. The substance-P analogues behaved as antagonists for Ca 2 þ release with potencies that reflected their affinities in receptor binding experiments, with SP-G showing some selectivity for V 1A receptors. This may be useful in tailoring therapy to a specific tumour phenotype in as much as SP-G may be more beneficial in V 1A receptor expressing tumours while SP-D may be better for tumours, which express both GRP and V 1A receptors. Both neuropeptides and the substance-P analogues activated ERK in CHO cells expressing the GRP receptor and the V 1A receptor. No ERK activation was observed in untransfected CHO cells suggesting that the activation is receptor mediated. These observations suggest that substance-P analogues act as dual efficacy receptor agonists directing receptor signalling via the GRP and the V 1A receptor towards ERK activation while blocking liberation of intracellular calcium. Moreover, the substance-P analogues activate ERK via G i as the response is pertussis toxin sensitive and more prolonged than that observed with neuropeptide (data not shown). These data support a multitrack signalling complex leading from neuropeptide receptor to ERK activation; where one track is mediated by G q and most likely involves PKC, whereas another feeds into a G i -mediated pathway (Wetzker and Bohmer, 2003). Although by far the majority of evidence suggests a preference for interacting with G q , some previous work has shown V 1A receptors coupling to G ia3 in quiescent fibroblasts and in hepatocytes (Strakova et al, 1997;Abel et al, 2000) There is also evidence that GRP receptors couple to G i in pancreatic acinar cells (Profrock et al, 1992), and that G i mediates bombesin-stimulated chemotaxis in monocytes (Djanani et al, 2003). Our data suggest that in the presence of substance-P analogues, the neuropeptide receptors exist in a conformation that is unable to activate G q but which is able to interact with G i and activate ERK. These analogues may act at a common region of the GPCRs perhaps at the receptor/ G-protein interface, which favours coupling to G i . Although activation of ERK is normally associated with a mitogenic response as is seen with the neuropeptides, deregulated ERK activation has also been shown to cause disruption of the cell cycle, growth arrest and apoptosis. In SCLC-activated raf-1, the upstream regulator of ERK causes growth arrest (Ravi et al, 1998). It is our hypothesis that deregulated ERK coupled with a decrease in intracellular Ca 2 þ are important factors in the growth inhibitory/apoptosis inducing activity of substance-P analogues.
In summary, the results presented here show that neuropeptide receptor expression leads to the development of a transformed phenotype in CHO epithelial cells. These cells display increased sensitivity to substance-P analogues that parallels the situation observed in SCLC cells, which develop chemoresistance in vivo. These analogues activate GRP and V 1A receptors in such a way as to cause activation of only a subset of possible downstream signals. They are competitive antagonists for PI hydrolysis mediated via G q but have agonist activity that results in activation of ERK via G i . This presents a novel pharmacological mechanism whereby substance-P analogues utilise the normally mitogenic neuropeptide receptors to transduce an antiproliferative and apoptogenic cell signal. Moreover, the concentration of SPG used in this study was detected in patient plasma following administration with no noticeable toxicity (Clive et al, 2001), suggesting therapeutic doses can be achieved clinically. Development of these agents may provide a unique opportunity to treat SCLC and may be of particular benefit for tumours that have developed resistance to chemotherapy.