SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Background:

Gastrimmune is an immunogenic form of gastrin. It raises in situ antibodies against two proliferative forms of gastrin: amidated and glycine-extended gastrin-17. It has been shown to have a therapeutic action in several in vivo tumour models. Following immunization, due to the complex equilibrium that exists between the antibodies and gastrin, it is not technically feasible to assay for free gastrin.

Aim:

To determine the effect of Gastrimmune-induced antigastrin antibodies on acid secretion.

Method:

A rat gastric fistula model was used. Animals (six per group) were immunized with a control immunogen or ascending doses of Gastrimmune. Acid output was measured following infusion of increasing doses of gastrin-17 and pentagastrin.

Results:

Gastrimmune-induced antibodies significantly reduced gastrin-17-stimulated acid output compared to control animals (Gastrimmune at 200 μg/rat vs. control; acid output following 30 ng gastrin-17, 0.01 vs. 0.16, P < 0.001; following 120 ng gastrin-17, 0.022 vs. 0.29, P < 0.001).

Conclusions:

Gastrimmune significantly inhibits gastrin-17-stimulated acid output. This biological assay suggests that the antigastrin antibodies effectively bind gastrin-17. In addition to its use as an antineoplastic agent, Gastrimmune may have a role as an acid-decreasing agent in oesophagogastric pathology.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Gastrin is a peptide hormone that is important in the regulation of acid secretion and the growth of both normal and malignant gastrointestinal epithelium.1 Inhibition of gastrin has a potential therapeutic use as an acid suppressant and, perhaps more importantly, in gastrointestinal malignancy, especially colorectal and pancreatic cancer.

Immunoneutralization has been used to acutely block the action of gastrointestinal peptides in vivo.2, 3 Chronic immunoneutralization can be achieved with the use of either immunoglobulin G or its antigen binding fragment (Fab); both are passive and suffer from a loss of effective immunoneutralization.4 Active immunoneutralization can be achieved with the antigastrin immunogen, Gastrimmune. Gastrimmune consists of the nine amino-terminal amino acids of human gastrin-17 linked via a peptide spacer to diphtheria toxoid. Diphtheria toxoid acts as the immunogenic carrier, the gastrin-17 sequence acts as a B-cell epitope and the peptide spacer allows the gastrin moiety to be spatially orientated in such a way that B cells recognize the whole sequence and thus raise high-affinity, neutralizing antibodies.5, 6

As the antibodies generated are directed to the amino-terminal of gastrin-17, the normal function of gastrin-34, any smaller C-terminal fragments and cholecystokinin (CCK) can continue unimpeded. Interaction with the amino-terminal fragment of gastrin-17 results in the inhibition of both amidated and glycine-extended gastrin-17. Glycine-extended gastrin-17 is thought to potentiate the acid-secreting ability of amidated gastrin,7, 8 probably by the up-regulation of the H+,K+-ATPase within gastric parietal cells.9 Several studies have also confirmed a role for glycine-extended gastrin-17 in colorectal neoplasia. Using active immunization in a pig model, immunoneutralization of gastrin-17 has been shown to inhibit acid secretion without having any antiproliferative effects on the gastrointestinal mucosa, indicating that the normal trophic effects of other gastrin species, such as gastrin-34, are preserved.10

The efficacy of the antibodies raised by Gastrimmune has been demonstrated in colorectal tumour models. The effect of active Gastrimmune immunization was evaluated on the in vivo growth of the rat colon tumour, DHDK12. The growth of the tumours in Gastrimmune-immunized rats was significantly lower than that in rats treated with the control immunogen.11 A model of colorectal hepatic metastases was produced by the injection of the colorectal tumour cell line, C170HM2, into the peritoneal cavity of nude mice, which selectively invades the liver. Antibodies raised by Gastrimmune were administered to C170HM2 tumour-bearing mice from day 0 and were given by a daily tail vein injection. Tumour take rate was reduced from 100% in the serum control treated animals to 30% in the animals treated with antiserum raised by Gastrimmune (P=0.0001). A lung metastasis model was produced by injection of the human colorectal cancer cell line, AP5LV, into the muscle layer of the abdominal wall. This resulted in spontaneous metastasis to the lung. Administration of Gastrimmune-induced antiserum to mice injected with AP5LV tumour cells reduced spontaneous metastasis, with lung nodule number reduced from a median of 3.5 in the serum control treated group to 1.0 in the group treated with gastrin-17 antiserum (P=0.00001, Mann–Whitney), and the median nodule cross-sectional area reduced from 11.9 to 3.75 mm2 (P=0.0064).12

In these animal studies and in the initial phase 1 studies in humans,13 excess high-affinity antigastrin antibodies have been produced. The antibody response to Gastrimmune is polyclonal and likely to lead to a spectrum of antibody affinity in vivo (G. Dockray, personal communication, 1998). Serum manipulation is likely to affect the equilibrium, potentially causing the release of gastrin from low-affinity antibodies. This precludes an accurate assay for free gastrin-17. In order to determine the effectiveness of the antigastrin-17 antibodies in the immunoneutralization of gastrin-17 and in the blockade of its action, the effect on acid secretion can be measured. If the antibodies are 100% effective, infusion of gastrin-17 should not produce a rise in gastric acid secretion. This provides a ‘biological assay’ that demonstrates the effectiveness of chronic immunoneutralization by Gastrimmune. Secondly, it allows the evaluation of the potential application of Gastrimmune as an acid inhibitory agent.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Home Office approval

The following procedures were performed under a project licence No. 40/1505 held by Dr S. A. Watson. Training in fistula insertion and peptone-stimulated acid secretion was obtained in Professor Dockray’s unit at the University of Liverpool. An initial pilot study to assess the technique of fistula insertion was performed successfully before the main study commenced.

Immunization protocol

Eight- to 10-week-old male and female Wistar rats, bred within the Biomedical Services Unit at the Queens Medical Centre, Nottingham, were used. Rats were immunized by subcutaneous injection with rat Gastrimmune (rat Gastrimmune differs from human Gastrimmune by a single amino acid substitution) at 3-week intervals. Gastric fistulas were inserted between the second and third immunizations, at approximately week 5. After four immunizations, acid secretion studies were performed. Further immunizations were performed at 3-weekly intervals. Four groups of six Wistar rats were used. A control group was immunized with diphtheria toxoid. There were three rat Gastrimmune dose groups: 10 μg, 100 μg and 200 μg.

Gastric fistula insertion

The technique for insertion of the gastric fistula was based on a method reported by Dimaline et al.14 The day before fistula insertion, food was withdrawn, but the animals were allowed free access to water.

The rat’s abdomens were shaved and Hypnorm/Hypnoval (1:1:2) dilution in water (Janssen/Roche, 3 mL/kg) was administered intraperitoneally. An adequate depth of anaesthesia was assessed by the loss of the foot pinch reflex. Animals received supplemental oxygen throughout the procedure.

A right paramedian incision was made through which the stomach was delivered. A stainless steel Gregory cannula (Medical School Workshop, University of Nottingham) was then inserted into the body of the stomach along the greater curve using vicryl purse string sutures. Three interrupted sutures were placed in the ‘neck’ of the cannula to prevent rotation. The cannula was exteriorized via a stab incision made in the midline. The abdominal cavity was closed in layers with interrupted chromic catgut. Local anaesthetic (1% lignocaine) was infiltrated over the incision. The cannula was closed with a threaded brass cap, lubricated with silicone grease.

In the immediate post-operative period, animals were given prophylactic antibiotics (Penbritin, 80 mg/kg intramuscularly) and 10 mL of phosphate-buffered saline subcutaneously at two sites. The animals were placed in a warming cabinet and observed at regular intervals. Insertion of the cannula appeared not to cause the animals distress and no further analgesia was required.

In the post-operative period, food was withdrawn over the first 24 h; however, water was freely available. The subcutaneously administered phosphate-buffered saline, together with antibiotics, was continued over the following 7 days. Rats were weighed daily to ensure that no greater than 20% of their body weight was lost. Weight gain recommenced approximately 5–7 days following surgery.

Following insertion of the cannula, rats were housed separately in 9.5 in × 18 in high plastic cages with stainless steel tops. They were fed standard pelleted diet and given normal tap water to drink. Room temperature was maintained at 19–23 °C, with the animals exposed to 12 h light and 12 h dark cycles.

Pharmacological stimulation of acid secretion in terminally anaesthetized rats

Twenty-four hours prior to the procedure, the stomach of each rat was washed out with warm saline via the gastric fistula. The rats were then transferred to a cage with a wire bottom and fasted, although water was freely available. The animals were weighed on the day of the study.

Intraperitoneal anaesthetic was administered (Hypnorm/Hypnoval (1:1:2) dilution in water, 3 mL/kg, Janssen/Roche) and the rats were transferred to a warming cabinet in order to dilate the tail vein. If necessary, additional intraperitoneal anaesthetic was given during the experiment until the rat no longer exhibited a foot pinch reflex.

The tail vein was cannulated using a catheter (made from silastic tubing, 0.02 in × 0.37 in, and 23 gauge needle with hub removed), which was placed in the tail vein and flushed with phosphate-buffered saline. This catheter was used for the injection of all test reagents. A 5 cm midline incision was made through the skin and peritoneal wall of the rat, extending from just below the xiphoid process and around the indwelling midline fistula.

The stomach was mobilized, as some rats had developed marked adhesions around the fistula, often involving the liver. A transverse incision was made 0.5 cm above the oesophagogastric junction. A 0.05 in × 0.09 in silastic tube was passed down the oesophagus, until it could be felt in the stomach, and was held in place by two 4/0 vicryl ties; this also prevented back flushing. A vicryl tie was also placed distal to the pyloric junction to block flow into the duodenum. The abdominal wall was loosely closed with a single suture and a piece of phosphate-buffered saline soaked gauze was placed over the incision.

The cap from the previously inserted gastric fistula was removed and connected to tubing to allow drainage and collection of gastric contents. The stomach was irrigated to remove food particles by flushing with 0.9% saline at 35 °C. The oesophageal tubing was connected to a pump and 0.9% saline, maintained at 35 °C in a constant water bath, was infused at a constant rate (0.5 mL/min). The body temperature of the rat was monitored by a thermometer attached to the chest wall. Normal body temperature was maintained by placing the rat on a heating pad with insulation.

Approximately 30 min prior to the administration of any test solution, 10.8 μg of atropine sulphate was injected to stabilize the acid secretion baseline. Once regular baseline readings had been established, infusion of acid stimulants was performed. Rat amidated gastrin-17 was dissolved in phosphate-buffered saline to a concentration of 120 ng/mL. In ascending dose order, gastrin was administered at 30, 60 and 120 ng over a 5-min period.

Pentagastrin was used as a positive control as it is composed of the 5-carboxy-terminal amino acids of gastrin-17 and is therefore unaffected by Gastrimmune, which is directed against the amino-terminal of gastrin-17. Pentagastrin was dissolved in phosphate-buffered saline at a concentration of 600 ng/mL; 1 mL was infused slowly over a period of 5 min.

Acid was collected at 5-min intervals and the volume of each sample was measured gravimetrically. A 1-mL aliquot of each perfusate sample was titrated with 0.01 M NaOH to determine the micromoles of acid per aliquot. The total number of micromoles of acid per sample was then calculated from the total volume of the perfusate sample (micromoles per 1-ml aliquot × number of millilitres in sample=total micromoles).

Terminal antigastrin antibody measurement

Following completion of the acid secretion tests, the animals were terminated. A terminal cardiac puncture was performed to obtain serum for antigastrin antibody measurement. The gastrin antibody assay has been described previously.11

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

A typical acid output profile for a control animal is shown in Figure 1. Pentagastrin (600 ng) was administered to generate maximal acid output and, as it is unaffected by Gastrimmune, acts as an internal positive control. The acid output was calculated every 5 min for a period of 20 min following each intravenous injection of peptide. Acid output stimulated by rat amidated gastrin-17 can be expressed as a percentage of the maximal pentagastrin-stimulated acid output. Infusion of the highest dose of gastrin-17 (120 ng) led to 91% maximal acid output in control animals. The effect of antigastrin antibodies, induced by Gastrimmune, on acid secretion after infusion of amidated gastrin-17 can be seen in Figures 2 and 3. Immunization with Gastrimmune at 200 μg reduced the acid output following infusion of gastrin-17 (120 ng) to 6.5% (P < 0.001) of the maximal pentagastrin-stimulated acid output. Following immunization with Gastrimmune at 100 μg and 10 μg, the effect on gastrin- 17-stimulated acid output was less pronounced: 65% (P=0.01) and 44% (P=0.78), respectively.

image

Figure 1.  Gastrin-17 (G17)- and pentagastrin (PG)-stimulated acid output in a control (diphtheria toxoid) animal.

Download figure to PowerPoint

image

Figure 2.  Gastrin-17 (G17)- and pentagastrin (PG)-stimulated acid output in a rat administered Gastrimmune at 200 μg.

Download figure to PowerPoint

image

Figure 3.  Effect of gastrin-17 (G17) and pentagastrin (PG) on acid output in rats immunized with Gastrimmune at 10 μg.

Download figure to PowerPoint

Table 1 summarizes the acid output following each infusion of secretagogue for each immunogen group. Any difference in acid output between dose groups was then compared using the Mann–Whitney U-test. As expected, the presence of antigastrin antibodies did not affect acid output after infusion of pentagastrin. Increasing the dose of rat amidated gastrin-17 in the control animals led to an increase in acid output.

Table 1.   Summary of amidated gastrin-17 (G17-NH2)- and pentagastrin (PG)-stimulated acid output in control and Gastrimmune (Gast)-treated rats Thumbnail image of

Gastrimmune administered at 200 μg and 100 μg significantly reduced the acid output seen after rat amidated gastrin-17 was infused at 30, 60 and 90 pg. The lowest dose of Gastrimmune, 10 μg, did not significantly inhibit acid secretion when rat amidated gastrin-17 was infused at 120 pg, but did inhibit acid secretion at lower doses. The acid inhibition produced by immunization with Gastrimmune at 200 μg was significantly greater than in the 100 μg immunized group at the maximal infusion of gastrin-17 (P=0.001), and Gastrimmune at 100 μg was significantly more effective than 10 μg at all gastrin-17 infusion doses. A correlation of acid response to gastrin-17 dose for control and treatment groups is shown in Figure 4.

image

Figure 4.  Correlation of acid response to gastrin-17 (G17) dose for control (DT, diphtheria toxoid) and treatment groups.

Download figure to PowerPoint

The excess antigastrin antibody titres produced following Gastrimmune immunization were measured at the end of the study The median antibody titre for Gastrimmune administered at 200 μg was 0.5 (range, 0.38–0.55), at 100 μg was 0.46 (0.32–0.51) and at 10 μg was 0.28 (0.10–0.4). There was no significant difference between the groups.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

Direct infusion of rat amidated gastrin-17 into anaesthetized rats and its effect on acid secretion were assessed following immunization with the gastrin immunogen. The presence of antigastrin-17 antibodies led to complete inhibition of amidated gastrin- 17-stimulated acid secretion, confirming efficacy. The degree of acid suppression was significantly related to Gastrimmune dose. This study confirms that the antibodies raised by Gastrimmune are effective in neutralizing gastrin-17 and supports the positive therapeutic effects seen in the in vivo tumour models. 11, 15, 16 Even at high doses of infused gastrin-17, acid output could be effectively suppressed. This biological assay suggests that there is little free gastrin, if any, circulating in the presence of the antigastrin antibodies. The radioimmunoassay measured free antibodies of increasing levels in the three groups which, due to the nature of the assay, are in excess of those required to neutralize gastrin-17. The requirement for an excess of antigastrin antibodies to effectively neutralize gastrin has also been demonstrated by Ohning et al.17 Using a monoclonal antigastrin antibody that was passively infused, they determined that at least a two-fold excess of antigastrin antibodies was required, and the efficacy was increased if the excess was raised 10-fold. The absence of a significant difference in antibody titres between groups is most likely to be due to the small numbers in each group and the fact that excess antibody titres were measured as opposed to total antibody titres.

The significant acid inhibition seen suggests that Gastrimmune may have a role as an acid suppressant in oesophagogastric disease. In humans, following a course of three injections, significant excess antibody titres have been observed at 6 months; thereafter, the antibody titres decline.18 Long-term acid suppression, which is likely to be reversible, could be achieved without the concurrent hazard of hypergastrinaemia. Recently, two well-performed animal studies using transgenic hypergastrinaemic mice have shown that there is significantly increased proliferation of the colonic mucosa in the presence of hypergastrinaemia compared to controls.19, 20 These findings are reflected in a temporal study in humans by Thorburn et al.21 They demonstrated that elevated gastrin levels above normal were associated with an increased risk of colorectal malignancy (odds ratio, 3.9; 95% confidence interval, 1.5–9.8). A significantly greater proportion of colorectal cancer cases had gastrin levels above the normal threshold (90 pg/mL), with 25 cancer cases (10.7%) but only nine controls (3.9%, P < 0.005). Thus the significance of the adverse effects of hypergastrinaemia is emerging. Gastrimmune, in addition to being an antineoplastic agent, may offer an alternative form of acid suppression.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References

This study was funded by a generous grant from Aphton Corporation, California, USA.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. References
  • 1
    Walsh J. Gastrin. In: Walsh J, Dockray G, eds. Gut Peptides. New York: Raven Press, 1994: 75–121.
  • 2
    Wong H, Walsh J, Yang H, Tache Y, Buchan A. A monoclonal antibody to somatostain with potent in vivo immunoneutralising activity. Peptides 1990; 11: 70712.
  • 3
    Yang H, Wong H, Walsh JH, Tache Y. Effect of gastrin monoclonal antibody 28.2 on acid response to chemical vagal stimulation in rats. Life Sci 1989; 45(25): 24138.
  • 4
    Levy R & Miller R. Tumour therapy with monoclonal antibodies. Fed Proc 1983; 42: 26506.
  • 5
    Watson SA, Michaeli D, Morris TM, Robinson G, Hardcastle JD. Gastrimmune inhibits the primary and secondary growth of 2 rat colon carcinomas. Cancer Res 1994; 35: 257257.
  • 6
    Makishima R, Larkin D, Michaeli D, Gaginella TS. Active immunisation against gastrin-17 with an N-terminal derived immunogen inhibits gastric and duodenal lesions in rats. Gastroenterology 1995; 106: A824A824(Abstract).
  • 7
    Higashide S, Gomez G, Greeley GH Jr, Townsend CM Jr, Thompson JC. Glycine-extended gastrin potentiates gastrin-stimulated gastric acid secretion in rats. Am J Physiol 1996; 270(1 Part 1): G2204.
  • 8
    Chen D, Zhao C, Dockray G, et al. Glycine-extended gastrin synergizes with gastrin 17 to stimulate acid secretion in gastrin-deficient mice. Gastroenterology 2000; 119(3): 75665.
  • 9
    Kaise M, Muraoka A, Seva C, Takeda H, Dickinson C, Yamada T. Glycine-extended progastrin processing intermediates induce H+,K(+)-ATPase alpha-subunit gene expression through a novel receptor. J Biol Chem 1995; 270(19): 11 15560.
  • 10
    Justin T. The role of gastrin immunoneutralisation in gastrointestinal malignancy and gastric acid disorders [D.M.]. Nottingham: Nottingham University, 1995.
  • 11
    Watson SA, Michaeli D, Grimes S, et al. Gastrimmune raises antibodies that neutralise amidated and glycine-extended gastrin-17 and inhibit the growth of colon cancer. Cancer Res 1996; 56: 8805.
  • 12
    Watson S, Michaeli D, Morris T, et al. Gastrimmune reduces lung metastases in a human colorectal model. Gut 1995; 37 (Suppl 2): A37A37(Abstract).
  • 13
    Smith A, Justin T, Watson SA, Michaeli D, Broome P, Maxwell-Armstrong C. Clinical outcome of advanced colorectal cancer patients treated with the anti-gastrin immunogen, Gastrimmune. Br J Surg 1998; 85: 15561556.
  • 14
    Dimaline R, Carter N, Barnes S. Evidence for reflex adrenergic inhibition of acid secretion in the conscious rat. Am J Physiol 1986; 251 (Gastrointestinal Liver Physiology 14): G61520.
  • 15
    Watson SA, Michaeli D, Grimes S, et al. A comparison of an anti-gastrin antibody and cytotoxic drugs in the therapy of human gastric ascites in SCID mice. Int J Cancer 1999; 81(2): 24854.DOI: 10.1002/(sici)1097-0215(19990412)81:2<248::aid-ijc14>3.0.co;2-g
  • 16
    Watson SA, Michaeli D, Justin TA, et al. Pre-clinical evaluation of the Gastrimmune immunogen alone and in combination with 5-fluorouracil/leucovorin in a rat colorectal cancer model. Int J Cancer 1998; 75(6): 8737.DOI: 10.1002/(sici)1097-0215(19980316)75:6<873::aid-ijc9>3.3.co;2-0
  • 17
    Ohning GV, Wong HC, Walsh JH. Differential kinetics for immunoneutralization of circulating gastrin by gastrin monoclonal antibody and its Fab1 fragment in rats. Peptides 1994; 15(3): 41723.
  • 18
    Smith A, Justin T, Michaeli D, Watson S. Phase I/II study of G17-DT, an anti-gastrin immunogen, in advanced colorectal cancer. Clin Cancer Res 2000; 6: 471924.
  • 19
    Wang TC, Koh TJ, Varro A, et al. Processing and proliferative effects of human progastrin in transgenic mice. J Clin Invest 1996; 98(8): 191829.
  • 20
    Koh TJ, Dockray GJ, Varro A, et al. Overexpression of glycine-extended gastrin in transgenic mice results in increased colonic proliferation. J Clin Invest 1999; 103(8): 111926.
  • 21
    Thorburn CM, Friedman GD, Dickinson CJ, Vogelman JH, Orentreich N, Parsonnet J. Gastrin and colorectal cancer: a prospective study. Gastroenterology 1998; 115(2): 27580.