Search

Malignant growth of small-cell lung carcinoma is promoted by various neuroendocrine autocrine/paracrine loops. Therefore, to interfere with this mitogenic process, it is crucial to elucidate the mechanisms involved. It is known that the oxytocin (OT) and vasopressin (VP) genes, normally transcriptionally restricted in their expression, are activated in small-cell lung cancer (SCLC), concomitantly with expression of their receptors (OTR, V1aR, V1bR/V3R and V2R). The aim of the present study was to characterize, in concentrations close to physiological and pharmacological conditions, intracellular signalling events triggered by OT and VP binding to their specific receptors in SCLC cells and to identify factors mediating OT- and VP-induced mitogenic effects on SCLC. Known agonists for OTR ([Thr4,Gly7]OT) and V1aR (F180), in addition to OT and VP, were able to elicit increases in cytosolic Ca2+ levels and this effect could be blocked using an OTR antagonist (OVTA) or a V1aR antagonist (SR49059) respectively. There was no activation of the cAMP pathway detected after VP, dDAVP (a V2R agonist), or OT treatment. Stimulation of SCLC cells with OT and VP led to an increase of extracellular signal-regulated kinase (ERK) 1/2 phosphorylation, maximal at 5 min, and the subsequent phosphorylation of its downstream target p90 ribosomal S6 kinase (p90RSK). Pre-incubation with OVTA and SR49059, and with inhibitors of phospholipase C (PLC), protein kinase C (PKC), mitogen-activated protein kinase/ERK kinase (MEK) 1/2 and a Ca2+ chelator significantly reduced OT- and VP-induced ERK1/2 phosphorylations. OVTA, SR49059 as well as MEK1/2 and PKC inhibitors also downregulated OT- and VP-induced p90RSK phosphorylation. In [3H]thymidine-uptake experiments, we subsequently observed that PLC, Ca2+, PKC and ERK1/2 are absolutely required for the OT- and VP-stimulated SCLC cellular growth process. In conclusion, the results presented here indicate that OT- and VP-induced mitogenic effects on SCLC are respectively mediated by OTR and V1aR signalling and that this mitogenic signalling passes through the phosphorylation of ERK1/2 and p90RSK in a PLC-, Ca2+-, PKC- and MEK1/2-dependent pathway.

The vasopressin (VP) gene is largely expressed in hypothalamic neurons, where the resultant pro-VP protein is enzymatically cleaved into its peptide hormone components, which include the neuropeptide VP, VP-associated neurophysin, and VP-associated glycopeptide (VAG). Small cell lung cancer (SCLC) tumors also express the VP gene, but the tumor pro-VP protein can remain intact and localize to the cell surface membrane. Previous studies have shown that polyclonal antibodies directed against different regions of the pro-VP protein bind specifically to the surface of cultured SCLC cells and recognize proteins of approximately 20 and approximately 40 kDa in cultured SCLC whole-cell lysate. Thus, these proteins have been designated neurophysin-related cell surface antigen (NRSA). A monoclonal antibody (mAb) designated MAG-1 was raised in this laboratory using a synthetic peptide representing the COOH-terminal sequence of VAG. The MAG-1 mAb recognizes NRSA in SCLC cell and tissue lysates by Western analysis, whereas immunofluorescent cytometric and microscopic analyses indicate that MAG-1 reacts specifically with NRSA on the surface of viable SCLC cells of both the classical and the variant subtype. Immunohistochemical analysis demonstrates that MAG-1 reacts with human SCLC tumor, but not with normal pulmonary epithelial cells in lung tissue. Additionally, a MAG-1 Fab fragment was generated that was also able to recognize NRSA. This is the first study to demonstrate that a mAb directed to the VAG region of the pro-VP protein has the potential for development into an in vivo diagnostic and therapeutic tool that targets plasma membrane-incorporated NRSA.

It is proposed that neuropeptide production by tumours is an important part of a special process of oncogenic transformation rather than a pre-existing condition of progenitor cells; this concept is called Selective Tumour gene Expression of Peptides essential for Survival (STEPS). All small-cell lung cancers and breast cancers evidently express the vasopressin gene, and this gene seems to be structurally normal in all but exceptional cases. Vasopressin gene expression in cancer cells leads to the production of both normal and abnormal forms of tumour vasopressin mRNA and proteins. Although the necessary post-translational processing enzymes are expressed in these cells, most processing seems to be extragranular, and most of the protein products become components of the plasma membrane. Small-cell lung cancer and breast cancer cells also express normal genes for all vasopressin receptors and produce normal vasopressin receptor mRNAs and V1a and V1b receptor proteins, and the vasopressin-activated calcium mobilising (VACM) protein; plus both normal and abnormal forms of the V2 receptor. Through these receptors, vasopressin exercises multifaceted effects on tumour growth and metabolism. A normal protein vasopressin gene promoter seems to be present in small-cell lung cancer cells, and this promoter contains all of the transcriptional elements known to be involved in gene regulation within hypothalamic neurones. Since these elements largely account for regulation of tumour gene expression observed in vitro, it is likely that as yet unknown factors are selectively produced by tumours in vivo to account for the observed seemingly autonomous or unregulated production of hormone in tumour patients. Promoter elements thought to be responsible for selective vasopressin gene expression in small-cell lung cancer probably include an E-box and a neurone restrictive silencer element close to the transcription start site. It is possible that transcription factors acting at these same elements can explain selective vasopressin expression, not only in small-cell tumours, but also in all other tumours such as breast cancer. By extrapolation, similar mechanisms might also be responsible for the expression of additional features that characterize the 'neuroendocrine' profile of these cancers.

Vasopressin is one of several small neuropeptides that are reported to be autocrine growth factors for small cell carcinoma of the lung (SCCL). It has been assumed that this peptide exercises its mitogenic influences through the vasopressin V1a receptor, and we have previously demonstrated that this receptor is expressed by classical and variant SCCL. Activation of the vasopressin V1a receptor produces changes in phospholipases C, D, and A2, in protein kinase C, and in Ca2+ mobilization. This study demonstrates that SCCL cells express not only vasopressin V1a receptors but also mRNAs and proteins representing normal V1b receptors and V2 receptors. They were also shown to express mRNA for a human form of the putative receptor rabbit vasopressin-activated calcium-mobilizing receptor (VACM-1). Additionally, SCCL tumor cells were found to express mRNA and protein representing a possible nonfunctional, shortened, "diabetic" form of the vasopressin V2 receptor that is the product of incomplete posttranscriptional splicing. At least four of these five vasopressin receptors were produced by cell lines exemplifying classical and variant forms of SCCL. No differences in the sequences for the V1 receptors between classical and variant SCCL were found. However, although the nature and expression of both vasopressin V1 receptors and human VACM are apparently unaffected by dedifferentiation in SCCL, only the abnormal (and probably nonfunctional) form of the V2 receptor could be demonstrated in variant cell line NCI H82. Functional engagement of vasopressin V2 receptors is reported to produce rises in cAMP and activation of protein kinase A, whereas stimulation of V1b receptors is believed to produce similar changes to those produced by V1a receptors, i.e., activation of phospholipases and of protein kinase C. Stimulation of VACM receptors raises intracellular free Ca2+ through currently unknown but phosphoinositide-independent mechanisms. The presence of all known vasopressin receptors that are, together, potentially capable of inducing several different transduction cascades in small cell tumor cells suggests that this peptide serves a multifaceted role in tumor physiology.

Small-cell carcinoma of the lung (SCCL) is a neuroendocrine tumor characterized by having the capacity to produce and secrete a number of small neuropeptides. These peptides serve the tumor as autocrine growth factors. SCCL is known to undergo a process of dedifferentiation to a variant (drug-resistant) form, and this process is associated with loss of marker enzymes such as neuron-specific enolase (NSE) and dopa decarboxylase (DDC). The current study was designed to discover if variant SCCL, represented by cell line NCI H82, retains some capacity to generate active neuropeptides (like vasopressin) from their precursors by continuing to express the three key classes of enzymes necessary for such conversions, namely prohormone convertases (PCs), carboxypeptidases (CPs), and peptidylglycine a-amidating monooxygenase (PAM). RT-PCR for mRNAs representing PC1, PC2, CPE, and PAM was performed on total RNA extracted from NCI H82. The primers selected for PCR and partial sequencing were synthetic 20, 21, 22, and 24 oligomers designed to yield products of 533, 880, 405, and 560 base pairs (bp) for PC1, PC2, CPE, and PAM, respectively. For the conditions used, we were able to demonstrate products for all four enzymes. Each of the four products generated were of the expected size. Cloning and sequencing of these products revealed that each had a structure identical to that published for the human form of the respective enzyme. Western analysis with antibodies against PC1, PC2, CPE, and PAM, provided evidence that mRNAs for the four enzymes are translated into proteins that could represent functional forms. Our findings therefore demonstrate that key enzymes involved in the generation of active neuropeptides, unlike the marker enzymes NSE and DDC, continue to be expressed by variant SCCL.

Vasopressin and other neuropeptides are believed to serve as autocrine growth factors for small-cell carcinoma of the lung (SCCL), and these mitogenic influences are reported to involve increases in intracellular Ca2+. Of the classical and variant forms of SCCL, the latter is not only more drug-resistant but also refractory to vasopressin, and other peptides, with respect to changes in intracellular Ca2+. It is currently unclear if this refractiveness of variant SCCL is due to the absence of involved peptide receptors, to the production of abnormal receptors, or to abnormalities in components of induced transduction cascades. In this study, the presence of structurally-normal and functional vasopressin V1a receptors, was examined in a classical SCCL cell line (NCI H345) that is Ca(2+)-responsive to vasopressin, and a variant SCCL cell line (NCI H82) that is unresponsive in this regard to the peptide. Both cell lines were shown to express an mRNA of 1.9 Kb for the vasopressin V1a receptor. RT-PCR, cloning, and DNA sequencing revealed the structure of the mRNA was identical for both cell lines, and, in turn, identical to the mRNA expressed for this receptor by human liver cells. In both cell lines and liver, this mRNA was shown by Western analysis and RIA to generate major protein products of approximately 70,000 and 43,000 daltons. Vasopressin action on NCI H82 cells resulted in a substantial rise in the levels of total inositol phosphates. However, it was reaffirmed that these changes in inositol phosphates were not accompanied by a rise in Ca2+ levels. All of these data indicate that variant SCCL, as well as classical SCCL, expresses structurally-normal and functional vasopressin V1a receptors, but their activation in variant SCCL raises IP3 levels without a corresponding rise in intracellular Ca2+. This difference between the two SCCL sub-types therefore involves either steps in the inositol triphosphate cascade beyond the activation of phospholipase C, or alternatively, components of other transduction events that might be involved with changes in intracellular Ca2+.

Expression of the vasopressin gene appears to be a property common to all small-cell lung tumours. For some cultures of small-cell lung carcinoma (SCCL), Northern and Western Blot analyses have revealed that expression of this gene and its protein products are regulated by cAMP and glucocorticoids. In this study, these evaluations have been extended by examining the production of vasopressin-associated human neurophysin (VP-HNP) by computer-enhanced quantitative immunocytochemistry in a classical cell-line (H69) of SCCL, and defining the amount of protein in cells by area of positive staining above an arbitrarily set threshold. Intracellular cAMP was raised by incubating cells with either 8,Br-cAMP (0.5 mM) and IBMX (0.5 mM), or with forskolin (25 microM) and IBMX (0.5 mM). Both of these treatments caused a significant increase in the amount of positive VP-HNP immunoreactivity in the cells, an increase that was further enhanced by simultaneous administration of dexamethasone (0.1 microM). Addition of dexamethasone alone, however, caused a significant decrease in VP-HNP levels. Results confirm earlier findings from Western Blot analysis revealing the influence these agents have on production of vasopressin gene-related proteins by H69 cells, and indicate that computer-enhanced quantitative immunocytochemistry can be effectively used to provide a suitable index of this production.

Four classical and three variant small-cell carcinoma of the lung (SCCL) cell lines were examined for vasopressin and vasopressin V1a-receptor immunoreactivity. One of these classical cell lines, NCI-H345, and one variant cell line, NCI-H82, were further investigated for binding of V1 and V2 vasopressin-receptor antagonists, vasopressin-induced calcium mobilization, and vasopressin-induced thymidine uptake. All classical and variant SCCL cell lines examined contained vasopressin and vasopressin-receptors as determined by immunocytochemistry. Both NCI-H82 and NCI-H345 demonstrated similar binding patterns with the V1 and V2 vasopressin-receptor antagonists, indicating the presence of both receptor subtypes. For the classical cell line (NCI-H345), vasopressin (1 microM) induced an increase in cytosolic free calcium, while the peptide was ineffective at increasing cytosolic calcium in the variant cell line (NCI-H82). However, vasopressin (0.1 or 1 microM) was unable to stimulate thymidine uptake in the classical (NCI-H345) or variant (NCI-H82) cell lines for the conditions used. These results indicate that both classical and variant SCCL produce vasopressin, and vasopressin V1a and V2 receptors. In the variant cell line, there appears to be a disruption in the activation cascade for V1a receptors as indicated by the lack of vasopressin-induced calcium mobilization.

Small-cell neuroendocrine carcinoma of the lung is known to express products related to the vasopressin gene, although these products have been reported to sometimes differ from those generated by neurones of the hypothalamo-neurohypophyseal system. To further investigate vasopressin gene expression in neuroendocrine carcinomas, we performed immunohistochemistry on 24 histologically classified small-cell carcinomas using antibodies directed against different regions of the vasopressin precursor. All of the tumours examined contained at least two parts of the vasopressin precursor, suggesting that vasopressin might have a biological role in these tumours and indicating a role for these products in tumour diagnosis and treatment. Sixty-seven per cent of the tumours contained immunoreactivity for all major regions of the precursor: vasopressin, vasopressin-associated human neurophysin, the bridging region between the hormone and the neurophysin, and vasopressin-associated human glycopeptide. However, 33% of the tumours examined appeared to express only part of the vasopressin precursor, as evidenced by the absence of immunoreactivity for the neurophysin and/or the glycopeptide. These results support the proposition that both normal and abnormal vasopressin gene expression occurs in small-cell carcinoma of the lung.

In previous studies we have demonstrated the high incidence of vasopressin gene expression as a characteristic feature of small-cell carcinoma of the lung. In the present study we examined expression of this gene in non-neuroendocrine tumors to determine if vasopressin production is a common feature of all lung tumors. We carried out the immunohistochemical evaluation of 22 non-neuroendocrine tumors (12 adenocarcinomas and 10 squamous-cell carcinomas) with antibodies to vasopressin, to oxytocin, and to their related neurophysins. The antibody preparations directed against vasopressin, oxytocin, or oxytocin-associated human neurophysin did not react with any of the tumors examined. Of two monoclonal antibodies to vasopressin-associated human neurophysin used, one did not react with any of the tumors, while the other stained neoplastic cells in only one adenocarcinoma and one squamous-cell carcinoma. These findings, taken with previous reports, indicate that among lung carcinomas, a high incidence of vasopressin/oxytocin gene expression is confined to neuroendocrine tumors.

Production by small-cell carcinoma (SCCL) of neurophysins (HNPs) and neurophysin-related cell-surface antigen (NRSA) was examined for two cell lines, for mouse xenografts, and for a resected human tumor, using polyclonal and monoclonal antibodies to vasopressin-associated human neurophysin (VP-HNP) and polyclonal antibodies to vasopressin (VP). The nature of the mRNA responsible for giving rise to these neurophysin-related products was investigated by performing Northern analysis on preparations of poly A+RNA and cDNA probes complimentary to portions of the exon A, exon B, and exon C regions of the human VP gene. SDS-electrophoresis and Western analysis revealed two prominent proteins of 42,000 and 20,000 Da in acid extracts from all SCCL sources when the monoclonal anti-HNP or one of the two polyclonal anti-HNP preparations were used. These antibodies also disclosed the presence of a minor component of 10,000 Da. A second polyclonal anti-HNP preparation reacted with one prominent protein of 30,000 Da and, for one cell line and mouse xenografts, another protein of 32,000 Da. Both of two anti-VP preparations reacted with proteins of 42,000, 30,000, 25,000, and 20,000 Da in extracts from all SCCL source material. The immunoreactive proteins of 42,000, 30,000, and 20,000 Da were all components of a membrane fraction from SCCL cells and tissues. In Northern analysis, a single RNA of about 900 bases hybridized with exon A and exon B probes, but not with the cDNA probe complimentary to exon C of the VP gene.(ABSTRACT TRUNCATED AT 250 WORDS)

Vasopressin (VP) and oxytocin (OT) were evaluated as tumor markers for small cell carcinoma of the lung by measuring the concentrations of these hormones in plasma samples obtained from patients at the onset of therapy and during treatment. Patient levels of VP before treatment ranged from 0.9-116 pmol/L, and this hormone was elevated (greater than 2.4 times) in 37 of 80 patients (46%) when values were compared to those of 25 healthy volunteers (normal mean, 2.13 +/- 0.15 pmol/L). Seventeen patients with elevated arginine VP displayed symptoms of the syndrome of inappropriate secretion of antidiuretic hormone. Patient levels of OT ranged from 0.3-124 pmol/L, and OT was elevated (greater than 2.4 times) in 14 of 72 patients (19%) compared with values in normal subjects (normal mean, 2.23 +/- 0.34 pmol/L). Both hormones were elevated in 6 patients. A positive response to treatment (partial or complete remission) was associated with reductions of elevated VP to 34.6 +/- 4.0% and of elevated OT to 34.7 +/- 7.5%, of values before treatment. Relapse was associated with an increase to 334 +/- 93% of remission values for VP (6 patients) and to 307% for OT (1 patient). These results indicate that VP and OT may be suitable plasma markers for a majority of small cell tumors. In most cases, an elevated concentration of hormone was associated with an elevation of the biosynthetically related neurophysin and vice versa. However, there were a number of exceptions, so that an elevated plasma concentration of VP, OT, or a neurophysin was found for 88% of patients with extensive disease and 70% of patients with limited disease.

Preliminary studies are reported on the use of 131I-labelled antibodies against vasopressin associated human neurophysin to image tumors in patients with small cell carcinoma of the lung (SCCL). The rabbit polyclonal antibodies used in these studies were affinity purified on columns of neurophysin-Sepharose. Patients were pre-screened for the presence of neurophysin producing tumors by plasma RIA. Six patients who screened positive received approximately 1 mCi/70 kg body weight of radioiodinated antibody preparation, and scintigraphy was subsequently performed at 24 h, 48 h, and 72 h using a subtraction technique based on simultaneous imaging with radiopharmaceutical agents labelled with technetium-99m. Tumor was clearly demonstrated at 72 h in all five patients who had measurable lesions at the time of study, while no positive image was noted for one patient in complete clinical remission. Since approximately 70% of all SCCL patients have neurophysin-producing tumors, our findings suggest that neurophysin antibodies may be effectively used to scan tumors in a majority of patients with SCCL.

A study of 61 patients with small cell carcinoma of the lung using specific RIAs for 2 human neurophysins (HNPs) has revealed that plasma levels of 1 or both HNPs are substantially elevated (> 3 times) in 62% of the patients before the commencement of therapy. These elevated HNPs may be a consequence of production/release by tumor. Eighteen patients with elevated plasma HNPs and 14 with normal values were followed during therapy. All of the patients with normal pre-therapy levels maintained these normal levels regardless of the course of their disease. For all patients with elevated HNP levels before therapy, there was good agreement between changes in these elevated values and clinically assessed responses. Partial or complete remission (12 patients) was associated with a 2- to 30-fold reduction in HNP levels, progressive disease (6 patients) was associated with a rise in HNP levels, and relapse after a previous objective response (6 patients) was associated with an increase in plasma HNPs over values found during remission. For many of the patients in clinical remission, HNPs remained elevated above normal values, and RIA data seem to forecast recurrent disease several weeks before clinical recognition. These data provide good evidence that our RIAs for HNPs can provide a valuable guide to therapeutic management of small cell carcinoma of the lung.

At diagnosis, 65% of 103 patients with small cell carcinoma of the lung were found to have elevated plasma concentrations of vasopressin-associated human neurophysin (VP-HNP), oxytocin-associated human neurophysin (OT-HNP), or both, which were thought to be related to tumor secretion of these proteins. The remainder of patients were designated as nonsecretors (24%) or possible secretors (11%), depending upon plasma concentration of the neurophysins prior to therapy. There was a significantly higher percentage of secretors among patients with extensive disease (82%) than among those with limited disease (40%) (P = 0.001). However, within each stage group, there was no correlation between secretory status and response to therapy, survival, or histologic subtype. In addition, patients who initially were nonsecretors or possible secretors maintained this status throughout the course of disease remission and subsequent relapse. These findings suggest the possibility of biochemical differences between tumors which present as limited disease and those which present as extensive disease. The syndrome of inappropriate antidiuretic hormone secretion (SIADH) was infrequent in limited disease but was present in 33% of patients with extensive disease. SIADH was not seen without VP-HNP elevation; however, with extensive disease, 49% of patients with elevated VP-HNP had SIADH. In contrast, elevated plasma concentrations of the neurophysins were seen in only 19.6% of 56 patients with non-small cell carcinoma of the lung. The levels were in general lower than those in patients with small cell carcinoma and were seen at approximately equal frequencies in each major cellular subtype.

Plasma human neurophysins (HNPs) were evaluated as tumor markers for patients with small cell carcinoma of the lung (SCCL) who were entered on limited disease and extensive disease treatment trials conducted by Cancer and Acute Leukemia Group B (CALGB). HNP values obtained before treatment showed 44% of tumors secreting vasopressin-associated HNP (VP-HNP), 14% secreting oxytocin-associated HNP (OT-HNP), and 11% producing both HNPs. There was a significantly higher incidence of HNP-secreting tumors for patients with extensive disease and two or more metastatic lesions than for patients with limited disease. There were no clear differences in response to treatment or in survival between patients with HNP-secreting tumors and those with nonsecreting tumors. Response to treatment evaluated by the change in plasma HNP, gave a 91% agreement with independently derived clinical impressions. Our data indicates that HNP evaluations provide sufficient sensitivity to forecast clinical response when it cannot be clearly assessed by conventional methods.