Chronic dosing of oltipraz in people at increased risk for colorectal cancer.

The dithiolethione oltipraz is being developed as a chemopreventive agent for many malignancies, including colorectal cancer, on the basis of its in vivo protective activity against chemically induced tumors in a variety of animal models. This protection has been associated with an enhanced capacity to detoxify reactive carcinogens and, more recently, with increased DNA repair. In a previous single-dose study, elevated detoxification gene expression was observed in the days after oltipraz dosing. Now, in this clinical study, we evaluated the effects of oltipraz when given over a 3-month period. Fourteen individuals with increased risk for colorectal cancer were randomly assigned to one of two oral doses (125 or 250 mg/m2) of oltipraz twice weekly for 12 weeks. Two of seven subjects at the 250 mg/m2 dosage required dose reductions, owing to significant fatigue. The 125 mg/m2 dose level was well tolerated by all patients. Blood or colon tissue (or both) for evaluation of glutathione, glutathione S-transferase, DT-diaphorase activity, and DT-diaphorase mRNA expression were obtained prior to treatment and at weeks 6, 12, and 16. No significant modulation of phase II detoxification enzymes was seen at either dose studied during this period. Phase II trials evaluating a tolerable regimen of oltipraz (as demonstrated in this study) and other possible mechanisms that may be responsible for the protective activity of oltipraz should be pursued.

Activation of the PPAR pathway induces apoptosis and COX-2 inhibition in HT-29 human colon cancer cells.

The gamma isoform of the peroxisome proliferator-activated receptor (PPARgamma) is a nuclear receptor that regulates adipocyte differentiation. Recently it has been shown to be expressed in human colonic mucosa and cancer, but its role in colon carcinogenesis and progression is still unclear. We demonstrate that activation of PPARgamma by ciglitazone (cig), a selective PPARgamma ligand, induces HT-29 human colon cancer cells to undergo apoptosis. Treatment with cig also down-regulates expression of cyclooxygenase-2 (COX-2) protein. Simultaneous exposure of cells to cig and 9-cis-retinoic acid (9-cis-RA), a ligand for retinoid X receptor, results in an increased apoptotic effect and increased inhibition of COX-2 expression, compared with cells treated with either cig or 9-cis-RA alone. As COX-2 is overexpressed in human colon cancer and has been implicated in augmenting invasiveness and tumorigenecity, the ability of PPARgamma activation to decrease COX-2 expression and induce apoptosis suggests that the PPARgamma pathway may be considered as a therapeutic target for colon cancer.

Cholinergic receptor up-regulates COX-2 expression and prostaglandin E(2) production in colon cancer cells.

The M(3) muscarinic cholinergic receptor has important physiological functions on normal colonic cells. It is frequently expressed on human colon cancer cells and is biologically active. Although it is mitogenic in certain cell models, the importance of this receptor on colon carcinogenesis is unknown. In the present study we have determined expression of the M(3) receptor on human colon cancer tissue compared with matched normal tissue and examined the downstream effect of receptor activation in the HT-29 human colon carcinoma cell line. Using reverse transcription-PCR, M(3) receptor RNA expression was detected in all matched colon carcinoma and normal specimens from eight patients. Five of the eight (62%) patients showed an up to 8-fold greater level of M(3) receptor expression in cancer compared with the matched normal tissue. Exposure of HT-29 cells to carbachol, a stable receptor agonist, results in a 10-fold increase in cyclooxygenase-2 (COX-2) protein. This induction of COX-2 protein was dose dependent and was inhibited by the cholinergic receptor antagonist N-methylscopolamine (NMS). Carbachol caused a dose-dependent increase in prostaglandin E(2) (PGE(2)), the main product of cyclooxygenase activity. The maximum stimulatory effect (40-fold increase) was noted with 1mM carbachol. The increase in PGE(2) was completely abolished by NMS and by the COX-2 selective inhibitor NS398. This suggests that the M(3) receptor mediates PGE(2) production by a mechanism involving COX-2. As COX-2 and PGE(2) are known promoters of gastrointestinal cancer, these data suggest that M(3) receptor activation may facilitate progression of colon carcinoma, in part by a COX-2-mediated cellular mechanism.

Human colon cancer cell proliferation mediated by the M3 muscarinic cholinergic receptor.

We have demonstrated previously cell surface receptors for gastrointestinal peptides on 10 human colon cancer cell lines. Because most of the cells studied bind muscarinic cholinergic agonists, we undertook the determination of the cholinergic receptor subtype expressed by human colon cancer cells, as well as the biological function of these receptors, and more specifically, the effect on cell proliferation. We used radiolabeled ligand binding, PCR, calcium mobilization, and cellular proliferation studies. The present study demonstrates a muscarinic cholinergic receptor having two classes of binding site for carbamylcholine. Analysis demonstrated 2499+/-153 binding sites/cell, of which 75% had a high affinity for carbamylcholine (Kd 55 microM), and 25% had a low affinity (Kd 0.33 mM). N-Methylscopolamine, a receptor antagonist, recognized only one binding site having high affinity (Kd 0.20 nM). The number of muscarinic cholinergic binding sites/cell found on colon cancer cells is 50% of the number of receptors found on guinea pig chief cells in physiological conditions. Specific cholinergic receptor antagonists inhibit binding in the following order of potency: N-methylscopolamine > 4-DAMP >> pirenzipine > AF-DX116. This order of potency pharmacologically classifies the receptor as an M3 subtype. Receptor expression, studied by reverse transcription-PCR, correlates with the binding data. Specifically, cell lines that exhibit binding, abundantly expressed the M3 receptor subtype, whereas cell lines that do not exhibit binding for muscarinic cholinergic agonists did not abundantly express the M3 receptor. Agonist activation of the M3 receptor on these cells resulted in intracellular calcium mobilization. The dose-response curve of calcium mobilization suggests that there are spare receptors on these cells. Signal transduction can be inhibited by receptor antagonists in the same order of potency in which the binding is inhibited. Exogenous agonist added to the cells in culture induces significant cell proliferation. These results demonstrate a muscarinic cholinergic receptor of the M3 subtype on human colon cancer cells. This receptor induces intracellular calcium mobilization and mediates cell proliferation. The data suggest that there are spare receptors present, and that there may be enhanced intracellular signal activation in response to receptor binding.

Preclinical and clinical evaluation of broccoli supplements as inducers of glutathione S-transferase activity.

Previous studies suggest that cruciferous vegetables may provide protection against carcinogen exposure by inducing detoxification enzymes. ICR(Ha) mice were gavaged with broccoli tablets (1 g/kg), and colon tissues were collected after treatment. Glutathione S-transferase (GST) activity was assayed and peaked on days 1 and 2 after treatment, respectively (P = 0.03). Elevations in GST activity were attributed to the increased expression of mu and pi. These data supported a clinical assessment of broccoli supplements. Twenty-nine subjects at increased risk for colorectal cancer were randomized to group 1 (no cruciferous vegetables) or group 2 (broccoli supplements, 3 g/day) for 14 days. Blood samples and colon biopsies were obtained pre- and postintervention. No significant difference was observed between the GST activities of the control and broccoli supplementation groups posttreatment. Mean lymphocyte GST activity was 107% of baseline in the broccoli supplementation group (range, 79-158%) and 102% of baseline in the control group (range, 75-158 percent;). Correlation of the GST activities of blood lymphocytes and colon mucosa taken simultaneously suggested that the GST activity of blood lymphocytes may be used as a biomarker of the responsiveness of colon tissue to chemopreventive regimens. Future clinical studies evaluating cruciferous vegetables should consider using concentrated dietary supplements in subjects with a previous history of colorectal cancer.

Modulation of gene expression in subjects at risk for colorectal cancer by the chemopreventive dithiolethione oltipraz.

Prolonged exposure to mutagenic substances is strongly associated with an individual's risk of developing colorectal cancer. Clinical investigation of oltipraz as a chemopreventive agent is supported by its induction of the expression of detoxication enzymes in various tissues, and its protective activity against the formation of chemically induced colorectal tumors in animals. The goals of the present study were: to determine if oltipraz could induce detoxicating gene expression in human tissues; to identify effective non-toxic doses for more extensive clinical testing; and to establish a relationship between effects in the colon mucosa and those in a more readily available tissue, the peripheral mononuclear cell. 24 evaluable patients at high risk for colorectal cancer were treated in a dose-finding study with oltipraz 125, 250, 500, or 1,000 mg/m2 as a single oral dose. Biochemical analysis of sequential blood samples and colon mucosal biopsies revealed increases in glutathione transferase activity at the lower dose levels. These effects were not observed at the higher doses. More pronounced changes were observed in detoxicating enzyme gene expression in both tissues at all doses. Peripheral mononuclear cell and colon mRNA content for gamma-glutamylcysteine synthetase (gamma-GCS) and DT-diaphorase increased after dosing to reach a peak on day 2-4 after treatment, and declined to baseline in the subsequent 7-10 d. The extent of induction of gene expression in colon mucosa reached a peak of 5.75-fold for gamma-GCS, and a peak of 4.14-fold for DT-diaphorase at 250 mg/m2 ; higher doses were not more effective. Levels of gamma-GCS and DT-diaphorase correlated closely (P < or = 0.001) between peripheral mononuclear cells and colon mucosa both at baseline and at peak. These findings demonstrate that the administration of minimally toxic agents at low doses may modulate the expression of detoxicating genes in the tissues of individuals at high risk for cancer. Furthermore, peripheral mononuclear cells may be used as a noninvasive surrogate endpoint biomarker for the transcriptional response of normal colon mucosa to drug administration.

Glutathione S-transferase activity and glutathione S-transferase mu expression in subjects with risk for colorectal cancer.

The glutathione S-transferases (alpha, mu, and pi), a family of Phase II detoxication enzymes, play a critical role in protecting the colon mucosa by catalyzing the conjugation of dietary carcinogens with glutathione. We investigated the efficacy of using the glutathione S-transferase (GST) activity of blood lymphocytes and GST-mu expression as biomarkers of risk for colorectal cancer. GST activity was measured in the blood lymphocytes of control individuals (n = 67) and in the blood lymphocytes (n = 60) and colon tissue (n = 34) of individuals at increased risk for colon cancer. Total GST activity was determined spectrophotometrically with the use of 1-chloro-2,4-dinitrobenzene as a substrate. The ability to express the um subclass of GST was determined with the use of an ELISA. Although interindividual variability in the GST activity of blood lymphocytes was greater than 8-fold (range, 16.7-146.8 nmol/min/mg), the GST activity of blood lymphocytes and colon tissue within an individual was constant over time and was unrelated to sex, age, or race. The GST activity of blood lymphocytes from high-risk individuals was significantly lower than that of blood lymphocytes from control individuals (P < or = 0.004). No association was observed between the frequency of GST-mu phenotype and risk for colorectal cancer. Blood lymphocytes from high-risk individuals unable to express GST-mu had lower levels of GST activity than did those from control subjects with the GST-mu null phenotype; however, this difference was significant in male subjects only (P < or = 0.006). Analysis of paired samples of blood lymphocytes and colon tissue indicated a strong correlation between the GST activity of the two tissue types (Spearman's rank correlation, r = 0.87; P < or = 0.0001). The GST activity of blood lymphocytes may be used to identify high-risk individuals with decreased protection from this Phase II detoxication enzyme who may benefit from clinical trials evaluating GST modulators as chemopreventive agents for colorectal cancer. The GST activity of blood lymphocytes may also be used in colorectal cancer chemoprevention trials to monitor the responsiveness of colon tissue to regimens that modify Phase II detoxication enzymes.

Palliative Treatment of Esophageal Cancer.

Palliative interventions for advanced esophageal cancer include surgery, radiation therapy, chemotherapy, chemoradiation, endoscopic procedures, and combinations of the above. Palliative esophagectomy or bypass procedures are difficult to justify in these patients because their life expectancy is so short. Palliative external beam radiation to doses of 50 to 60 Gy is successful in 50% to 70% of patients. The addition of brachytherapy may improve these results. One third to one half of patients treated with radiation develop benign or maglinant stricture. Although response rates to combination chemotherapy are only 50% at best, the majority of patients do have improvement of dysphagia. These regimens are commonly used as part of a multidisciplinary approach with radiation andøor surgery, rather than as a sole modality of treatment. Chemoradiation regimens results in better survival than treatment with radiation alone, and provide palliation of dysphagia in up to 90% of patients. Although acute toxicity of chemoradiation is more severe than radiation alone, this is of limited duration. Chemoradiation may be the treatment of choice for the majority of patients with locally advanced esophageal cancer. Endoscopic techniques are available that provide palliation of dysphagia. The most commonly used technique is esophageal dilatation, either alone or before performing other palliative procedures such as laser therapy or stent placement. The most significant limitation of dilatation alone is that palliation is short-lived and most patients require repeat dilatations. Esophageal stents offer a high degree of palliation, but procedure-related morbidity and mortality rates are not insignificant. Expandable metal stents are associated with few complications but tumor ingrowth through the metallic mesh is frequent. Conventional plastic stents are not affected by tumor ingrowth but can migrate. Endoscopic laser therapy also provides symptoms relief and complication rates are relatively low. It is possible that a combination of laser therapy and external beam or intraluminal radiation will provide more durable palliation than laser treatment alone. BICAP tumor probes, (Circon-ACMI, Stamford, CT), which provide direct application of electrical current, are limited to treatment of tumors that are circumferential. Photodynamic therapy (PDT), which applies laser light along with a photosensitizing agent, has resulted in a high rate of palliation. Limitations of PDT include skin photosensitization requiring patients to stay out of the sun for at least 1 month following treatment, high cost of required equipment, and limited efficacy because of the shallow depth of light penetration. A variety of treatment options exist for the management of tracheoesophageal fistulae (TEF), but only radiation therapy or bypass surgeyr appear to prolong survival. Radiation therapy does not appear to worsen the TEF as was commonly thought in the past, and it is likely applicable in more patients than is surgery. The challenge for the physician in palliating patients with esophageal cancer is to select therapy appropriate for a given patient, taking into account the patient's disease, coexisting medical problems, performance status, and the patient's desires.

Clinical, pharmacokinetic and biological studies of topotecan.

The topoisomerase I inhibitor topotecan is a potent water-soluble camptothecin derivative with activity in a wide variety of preclinical models. Topotecan exhibits schedule dependency in vivo, with the greatest activity being observed on repeated dose schedules. On the basis of the initial clinical studies that showed a short plasma half-life, we attempted to prolong drug exposure by giving topotecan as a 24-h infusion weekly. In a phase I trial, we treated 32 patients at doses ranging from 1.0 to 2.0 mg/m2. The patient population had not been heavily pretreated with chemotherapy and was of good performance status. The incidence of neutropenia, which was dose-limiting, increased sharply with relatively small increments in dose. Doses greater than 1.5 mg/m2 were associated with nadirs that developed after one to three weekly treatments. A patient with metastatic colorectal cancer had a prolonged partial response. The plasma pharmacokinetics of topotecan (lactone and open forms) was characterized in 21 patients. Mean plasma steady-state drug levels were proportional to the dose and were within the range required to exert cytotoxicity in preclinical models. Plasma elimination curves were fit to a one-compartment model, in which the harmonic mean half-life of topotecan was 3.5 h. The ratio of the lactone to the total drug concentrations was constant throughout, which suggests that for this schedule the total drug concentration may be used as a measure of active lactone exposure. This conclusion is supported by the pharmacodynamic analysis, which revealed a positive correlation of both lactone and total drug steady-state concentrations with bone marrow toxicity. The further investigation of this and other infusional schedules in phase II trials will be conducted. The steady-state concentrations of total drug will be measured in several of these trials to establish its potential role in adaptive dosing using this schedule. Such a strategy is justified by the interpatient variability in toxicity and the steep dose-response curve observed in this study. Preliminary evidence of interpatient variability in the mRNA expression of topoisomerase I in the peripheral mononuclear cells and colon mucosa is presented. Trials are under way using biological endpoints for further selection of patients in whom the use of topoisomerase inhibitors may be therapeutically beneficial.

Characterization of bombesin receptors using a novel, potent, radiolabeled antagonist that distinguishes bombesin receptor subtypes.

Bombesin (Bn)-related peptides affect numerous cell functions; however, receptor characterization by radiolabeled ligands is limited because only radiolabeled agonists exist. In the present study we demonstrate that [D-Tyr6]Bn(6-13)methyl ester [[D-Tyr6]Bn(6-13)ME] functions as a Bn receptor antagonist with high affinity. The binding of both the radiolabeled agonist 125I-[Tyr4]Bn and the radiolabeled antagonist 125I-[D-Tyr6]Bn(6-13)ME to AR42J cells, murine 3T3 cells, and dispersed guinea pig pancreatic acini was time and temperature dependent, saturable, and reversible. Binding of the antagonist more rapidly reached equilibrium and was more rapidly reversible. Guanine nucleotides did not affect binding of the radiolabeled antagonist, whereas guanosine-5'-(beta,gamma-imido)triphosphate decreased agonist binding by decreasing Bn receptor affinity. Acid stripping studies demonstrated that the radiolabeled agonist, but not the antagonist, was internalized in each cell system. Bn receptor affinities for various Bn receptor agonists or antagonists in each cell system were identical when computed from an analysis of inhibition curves for binding of radiolabeled agonist or antagonist. However, with AR42J cells and 3T3 cells the radiolabeled agonist demonstrated a > 2-fold higher number of Bn receptors than did the radiolabeled antagonist. Binding studies using cell membranes, in contrast to cells, showed equal numbers of Bn receptors with either radiolabeled ligand. The radiolabeled agonist demonstrated high affinity binding to both rat pancreatic acinar and esophageal muscularis mucosa membranes, whereas the radiolabeled antagonist interacted with high affinity only with the gastrin-releasing peptide-preferring subtype of Bn receptors on pancreatic tissue. These results demonstrate that 125I-[D-Tyr6]Bn(6-13)ME is a high affinity radiolabeled antagonist that interacts specifically with Bn receptors. In contrast to the radiolabeled agonist, binding of the antagonist is not affected by guanine nucleotides and it is not internalized, which allows quantitation of only Bn cell surface receptors. Furthermore, the radiolabeled antagonist can distinguish Bn receptor subtypes, whereas the radiolabeled agonist does not. This ligand should prove useful for characterizing Bn receptors as well as studying their regulation.

Ligand binding, internalization, degradation and regulation by guanine nucleotides of bombesin receptor subtypes: a comparative study.

Recent cloning studies confirm two subtypes of Bn receptors exist, a neuromedin B-preferring receptor (NMB-R) and a gastrin-releasing peptide-preferring receptor (GRP-R). Both subtypes occur widely in GI tract and the CNS; however, in contrast to the GRP-R subtype little is known about the ligand-receptor interactions for the NMB-R. Therefore, in the present study we explored the ligand-receptor interactions including kinetics, stoichiometry, internalization, degradation and regulation by guanine nucleotide binding proteins with the NMB-R and compared it to the GRP-R. The rat glioblastoma C-6 cell line which possess functional NMB-R and 3T3 cells which possess functional GRP-R were used. 125I-[D-Tyr0]NMB and 125I-[Tyr4]Bn were prepared using Iodogen and purified on HPLC. At 37 degrees C binding of 125I-[D-Tyr0]NMB to NMB-R or 125I-[Tyr4]Bn to GRP-R was maximal by 5-15 min and decreased to 60-70% after 60 min. HPLC analysis of the 60 min supernatant showed that > 80% of each tracer was degraded. Addition of proteinase inhibitors had a varied inhibitory effect on degradation with the relative order of potency in C-6 cells being leupeptin > bacitracin < chymostatin > phosphoramidon >> bestatin and amastatin and 3T3 cells being bacitracin = phosphoramidon > leupeptin = bestatin > chymostatin > amastatin in 3T3 cells. By HPLC analysis addition of bacitracin prevented the degradation in both cell types. With both receptor subtypes dissociation of bound radioligands was slow, with 70-80% of either 125I-[D-Tyr0]NMB or 125I-[Tyr4]Bn remained cell-associated after 60 min suggesting possible peptide internalization. With an acid wash procedure to remove surface bound radioligands, 60% of the C-6 cell-associated 125I-[D-Tyr0]NMB and 52% of the 3T3 cell-associated 125I-[Tyr4]Bn were internalized after 30 min at 37 degrees C. With membranes from cells possessing either receptor subtype, the stable guanine nucleotide GPP(NH)P inhibited in a dose-dependent fashion binding of ligands. Computer analysis demonstrated that GPP(NH)P decreased receptor affinity for ligands to both receptor subtypes. These results demonstrated that NMB receptors, similar to GRP receptors and rapidly internalize bound agonists and rapidly degrade agonists. The ligand-receptor interaction is regulated by a guanine nucleotide binding protein for both Bn receptor subtypes.

Characterization of functional receptors for gastrointestinal hormones on human colon cancer cells.

Studies demonstrate that some colon cancers possess receptors for various gastrointestinal hormones or neurotransmitters, the occupation of which can affect growth. These results are limited because frequently only a small number of tumors are studied, only 1 or 2 receptors are sought, and the effect on cell function is not investigated. In the present study, 10 recently characterized human colon cancer cell lines were studied to determine whether they possess receptors for any of 12 different gastrointestinal hormones or neurotransmitters and to determine whether these receptors mediate changes in cellular function. Each of the cell lines exhibited receptors for at least one radioligand. Receptors for vasoactive intestinal peptide (VIP) and muscarinic cholinergic agents occurred on 60%, bombesin and gastrin on 30%, beta-adrenergic agents and gastrin-releasing peptide (GRP) on 20%, and somatostatin, opiates, neuromedin B, and substance P on 10%. Analysis of [3H]N-methylscopolamine binding revealed a Kd of 0.2 nM for N-methylscopolamine with a binding capacity of 2500 sites/cell. With the agonist carbamylcholine, the receptor exhibited 2 classes of binding sites: one of high affinity (Kd 55 microM) representing 75% of the binding sites and one of low affinity (Kd 0.3 mM) representing 25% of the binding sites. Analysis of 125I-[Tyr4]bombesin binding revealed a receptor of high affinity (Kd 2.1 microM) with a binding capacity of 3300 sites/cell. Inhibition of binding by agonists revealed relative potencies of 125I-[Tyr4]bombesin greater than GRP much greater than neuromedin B, and two recently described antagonists were similar in potency to GRP. Analysis of 125I-VIP binding revealed a receptor having 2 classes of binding sites: one of high affinity (Kd 3.6 nM) and one of low affinity (Kd 1.7 microM) which represented the majority of the 5.5 x 10(6) binding sites/cell. The relative potencies of agonists were VIP greater than helodermin greater than peptide histidine methionine greater than secretin. Evaluation of biological activity mediated by the muscarinic cholinergic and bombesin receptors revealed an increase of intracellular calcium and of inositol triphosphate by specific receptor agonists. The presence or absence of receptors detected by binding correlated closely with the ability of selective receptor agonists to alter cell function. These results demonstrate the presence of several different receptors for gastrointestinal hormones or neurotransmitters, some described for the first time, on human colon cancer cell lines, including bombesin-related peptides, VIP, somatostatin, substance P, beta-adrenergic agents, calcitonin gene-related peptide, gastrin, muscarinic cholinergic agents, and opiates.(ABSTRACT TRUNCATED AT 400 WORDS)

Covalently cyclized agonist and antagonist analogues of bombesin and related peptides.

During a search for possible cyclization points in shortened, potent bombesin agonists and antagonists, it was found that the joining of amino acid residues in positions 6 and 14 by various means resulted in retention of significant binding affinity for rat pancreatic acini and murine Swiss 3T3 cells. In one series of analogues, Cys residues in these positions were used for bridging via a disulfide bond. (D)-C-Q-W-A-V-G-H-L-C-NH2 retained significant binding affinity for rat pancreatic acini cells and was a full amylase releasing agonist (EC50 187 nM). Potency was markedly increased by substituting D-Ala for Gly (EC50 67 nM compared to 10 nM for its linear counterpart) and was decreased by substituting L-Cys for D-Cys in this analogue (EC50 214 nM), thus strongly suggesting stabilization of peptide folding by the D residues. Elimination of the COOH-terminal amino acid produces competitive antagonists in the linear analogues; however, (D)-C-Q-W-A-V-G-H-C-NH2 was devoid of activity. Likewise, cyclization to position 13 with the 14 amino acids intact to give (D)-C-Q-W-A-V-G-H-C-L-NH2 resulted in an almost inactive peptide. On the other hand, as in the linear series, the reduced peptide bond analogue, (D)-C-Q-W-A-V-(D)-A-H-L-psi (CH2NH)-C-NH2, was a receptor antagonist (IC50 5.7 mM), albeit much weaker than the corresponding linear analogues, but with no residual agonist activity. Direct head-to-tail cyclization was also tried. Both cyclo[(D)-F-Q-W-A-V-G-H-L-L] (EC50 346 nM) and the shorter cyclo [Q-W-A-V-G-H-L-L] (EC50 1236 nM) were full agonists. Elimination of the COOH-terminal residue in cyclo[(D)-p-Cl-F-Q-W-A-V-(D)-A-H-L] produced an agonist (EC50 716 nM) rather than an antagonist. These results provide support for the proposal that both bombesin agonists and antagonists adopt a folded conformation at their receptor(s). Furthermore, the retention of appreciable potencies using several cyclization strategies and chain lengths suggests that further optimization of these structures both in terms of potency and ring size is possible. Since these peptides have increased conformational restriction, they should begin to serve as useful substrates for NMR and molecular modeling studies aimed at comparing the obviously subtle differences between agonist and antagonist structures.

Pharmacokinetics of oral and intravenous omeprazole in patients with the Zollinger-Ellison syndrome.

The pharmacokinetics and pharmacodynamics of oral and IV omeprazole after a single dose were studied in 9 patients with the Zollinger-Ellison syndrome to determine whether the increased dose required to control gastric acid hypersecretion could be explained on the basis of altered pharmacokinetics. Each patient was studied both after receiving a single IV bolus of omeprazole (40 mg) and after receiving a single oral dose of omeprazole (80 mg). Intravenous and oral omeprazole doses were administered 1 week apart. Gastric acid secretion and plasma concentrations of omeprazole after drug administration were determined in each patient. The area under the plasma concentration curve, clearance, and volume of distribution after IV omeprazole administration and the area under the plasma concentration curve, peak plasma concentration, and time required to reach the peak after oral omeprazole administration were not different from those reported previously for normal subjects and patients with peptic ulcer disease. Mean (+/- SEM) bioavailability of oral omeprazole for all patients was 68% +/- 16%, which was similar to the bioavailability reported previously for normal subjects. Three patients had a significantly lower bioavailability reported previously for normal subjects. Three patients had a significantly lower bioavailability (20% +/- 8%) than the others, and their basal acid outputs were significantly higher than those of the other 7 patients. For all patients there was an inverse correlation between bioavailability and basal acid output (r = 0.76; P less than 0.02). The mean (+/- SEM) elimination half-lives of IV and oral omeprazole were not different (2.3 +/- 0.4 vs. 2.4 +/- 0.5 hours) but were significantly longer than those reported previously for normal subjects (P less than 0.02). The duration of action correlated with the elimination half-life of the drug (r = 0.87; P less than 0.003) and area under the plasma concentration curve (r = 0.72; P less than 0.03). The mean durations of action of IV and oral omeprazole were not significantly different (34 +/- 7.2 vs. 35 +/- 6.2 hours). It was concluded that altered pharmacokinetics do not account for the increased drug requirement of omeprazole in patients with the Zollinger-Ellison syndrome. In contrast to a previous study, the oral and IV omeprazole had the same duration of action, suggesting that intermittent bolus administration of parenteral omeprazole will obviate the need for continuous infusion of histamine H2-receptor antagonists in patients requiring parenteral antisecretory drugs. Furthermore, an IV dose every 12 hours controlled acid secretion in all patients, suggesting this as the recommended dose interval in patients requiring parenteral drug therapy.

Use of omeprazole in patients with Zollinger-Ellison syndrome.

Omeprazole, a substituted benzimidazole, has been shown to be a potent inhibitor of gastric acid secretion in patients with Zollinger-Ellison syndrome (ZES). We review our experience, as well as the published data on 210 patients with ZES who have required omeprazole for control of gastric acid hypersecretion over the past seven years. The dose of omeprazole required in individual patients ranged from 10 to 180 mg/24 hr with 20-60% requiring a split dosage regimen. Omeprazole was effective in approximately 99% of the patients over a period ranging from 0.5 to 54 months. Twenty-four percent of patients required an increase in omeprazole dose, while 26% required a decrease in dose. Adverse effects attributable to omeprazole were reported in 2% of patients, and in all cases, they were mild (ie, rash, constipation, headache). There was no effect of omeprazole on serum gastrin concentration or on gastric endocrine cells in three studies. Although one patient with multiple endocrine neoplasia, type-I syndrome (MEN-I) in this series developed a gastric carcinoid while taking omeprazole, evidence is presented that suggests the presence of MEN-I per se may be important in determining the development of gastric carcinoid in patients with ZES. It is concluded that omeprazole is safe and effective in patients with ZES, and in these patients, it is the drug of choice for the management of gastric acid hypersecretion. However, yearly assessment is indicated to clearly evaluate the long-term risk of gastric carcinoid as well as therapy directed at the gastrinoma itself.

Zollinger-Ellison syndrome: prospective assessment of abdominal US in the localization of gastrinomas.

The ability of abdominal ultrasound (US) to help localize gastrinomas was prospectively studied in 79 patients with Zollinger-Ellison syndrome. The results were assessed by means of laparotomy, autopsy, or percutaneous liver biopsy. For hepatic gastrinoma, US had a sensitivity of 63% and a specificity of 100%, with a positive predictive value of 100% and a negative predictive value of 89%. US was slightly less sensitive for detecting gastrinoma in the liver than were computed tomography (CT) (66%) and selective angiography (78%). For detection of extrahepatic gastrinoma, US had a sensitivity of 30%, a specificity of 94%, a positive predictive value of 100%, and a negative predictive value of 25%. US enabled detection of tumor in eight cases not detected with CT and in four not detected with angiography. Specificity for extrahepatic gastrinoma was similar for all three modalities (89%-95%). CT and US were equally effective for the detection of extrahepatic gastrinoma, and angiography was significantly more effective than both US and CT (P less than .01). The authors conclude that US, although of low sensitivity, remains useful as the initial imaging modality in patients with Zollinger-Ellison syndrome.

Cholera toxin inhibits signal transduction by several mitogens and the in vitro growth of human small-cell lung cancer.

Cholera toxin (CT) inhibited the in vitro growth of three of four human small-cell lung carcinoma (SCLC) cell lines with a 50% inhibitory concentration of 27-242 ng/ml. Loss of surface membrane ruffling and the capacity of [Tyr4]-bombesin, vasopressin, and fetal calf serum to stimulate increases in intracellular free calcium clearly preceded effects on cellular metabolic activity and cell growth. 125I-[Tyr4]-bombesin binding was unaffected by CT treatment but [Tyr4]-bombesin stimulated phospholipase C activity was decreased in membranes from CT-treated SCLC cells. CT stimulated a rapid but transient increase in intracellular cyclic AMP ([cAMP]i) in SCLC. The effects of CT on susceptible SCLC were not reproduced by elevations of [cAMP]i induced by forskolin or cyclic AMP analogues. GM1 ganglioside, the cellular binding site for CT, was highly expressed in the CT-sensitive but not the CT-resistant SCLC cell lines. In contrast, expression of guanine nucleotide binding protein substrates for ADP-ribosylation by CT was similar. These data demonstrate the existence of a CT-sensitive growth inhibitory pathway in SCLC-bearing GM1 ganglioside. Addition of CT results in decreased responsiveness to several mitogenic stimuli. These results suggest novel therapeutic approaches to human SCLC.

Detection of duodenal gastrinomas by operative endoscopic transillumination. A prospective study.

The ability of operative endoscopic transillumination of the bowel wall to detect duodenal gastrinoma was evaluated prospectively in 26 patients with the Zollinger-Ellison syndrome. The results were assessed by exploratory laparotomy and compared with the results of other localization techniques. Twelve duodenal gastrinomas were resected from 10 patients. Operative endoscopic transillumination detected 10 of the 12 gastrinomas, a sensitivity of 83%, which was significantly greater (P less than 0.05) than that for either preoperative imaging (25%) or intraoperative ultrasonography and palpation (42%). The sensitivity of operative endoscopic transillumination was a result of the ability to detect focal areas that did not transilluminate on the serosal side of the duodenum, and not the mucosal appearances seen through the endoscope, which were not helpful. Operative endoscopic transillumination detected gastrinomas less than 1 cm in diameter throughout the duodenum. Of the patients in this study, 39% had duodenal gastrinomas, a greater frequency than previously reported. These results indicate that operative endoscopic transillumination is the most sensitive technique yet described for detecting duodenal gastrinomas and should be performed routinely in all patients with the Zollinger-Ellison syndrome who undergo exploratory laparotomy for cure.

des-Met carboxyl-terminally modified analogues of bombesin function as potent bombesin receptor antagonists, partial agonists, or agonists.

In the present study we examined the effect of carboxyl-terminal modifications of des-Met14-bombesin (Bn) on Bn receptor affinity in murine 3T3 cells, rat and guinea pig pancreatic acini, and the ability to initiate biologic responses by synthesizing 18 des-Met14-Bn(6-13) analogues. With guinea pig acini and 3T3 cells, affinity was affected by the chain length of the alkyl moiety (R) added to [D-Phe6]Bn(6-13)NH2R with relative potencies: propyl greater than ethyl greater than butyl = hexyl greater than heptyl greater than free amide, whereas in rat acini affinity was not increased by the chain length. In each cell system the affinity of the alkylamide was not increased by insertion of a phenyl group in the alkyl side chain, by making the analogue more neuromedin B-like or by addition of a reduced peptide bond. The affinity in each cell system was increased by additions of other electron releasing groups to the COOH-terminal carboxyl group such as [D-Phe6]Bn(6-13)ethyl or methyl ester, or hydrazide. In guinea pig pancreas and 3T3 cells, 12 analogues were antagonists, 1 a full and 5 partial agonists. In rat pancreas, 8 were antagonists, 5 full agonists, and 5 partial agonists. Potent antagonists in each cell system were the methyl and ethyl ester, hydrazide, and ethylamide analogues. In 3T3 cells or guinea pig pancreas, agonist activity of the alkylamide was critically dependent on the chain length, whereas with rat pancreatic Bn receptors any alkylamide longer than the ethylamide had agonist activity. In all three cell systems any alteration that made the alkylamide more neuromedin B-like caused agonist activity. These results demonstrate that the nature of the substitution on the carboxyl terminus of des-Met14-Bn analogues is critically important, not only for determining Bn receptor affinity, but also for determining the ability to initiate a biologic response. In contrast to previous studies, the present results demonstrate that the presence of the COOH-terminal amino acid in position 14 of Bn is not essential for initiating a biologic response. Several des-Met14-Bn analogues were potent partial agonists, whereas others such as the hydrazide or ethyl ester are very potent antagonists.

Intravenous omeprazole in patients with Zollinger-Ellison syndrome undergoing surgery.

Twenty patients with Zollinger-Ellison syndrome who were undergoing surgery were studied prospectively to assess the efficacy and safety of IV omeprazole. During the preoperative period, in 19 of 20 patients, omeprazole 60 mg administered as an IV bolus every 12 hours inhibited acid output to less than 5 mEq/h measured in the last hour before the next dose of drug. In one patient, acid output was 25 mEq/h 12 hours after omeprazole, 60 mg, and increasing the dose to 100 mg every 12 hours reduced acid output to less than 5 mEq/h. During the operative and postoperative periods, IV omeprazole controlled gastric acid hypersecretion in all patients for up to 15 days. During this time, all patients received the dose determined preoperatively. No patient developed any clinical, hematological, or biochemical toxicity that could be attributed to omeprazole therapy during the preoperative or postoperative period. The present study demonstrates that omeprazole administered by IV bolus is safe and effective for controlling gastric acid hypersecretion. In contrast to IV histamine H2-receptor antagonists, IV omeprazole has the advantages of not requiring continuous infusion or postoperative dose adjustments. Intravenous omeprazole will become the drug of choice in patients with Zollinger-Ellison syndrome undergoing surgery.