Trophic effects of continuous infusion of [Leu15]-gastrin-17 in the rat

GASTROENTRROLOGY

iaao;a8:33-38

Trophic Effects of Continuous Infusion of [Led5]-GastAnin the Rat B. RYBERG, J. AXELSON, R. HAKANSON, F. SUNDLER, and H. MATTSSON Department of Biology,AB Hassle, Gastrointestinal Research, Mtilndal, Sweden; Departments of Pharmacology,

Medical Cell Research, and Surgery, University

This report describes the trophic effects of exogenous gastrin on the digestive tract and pancreas and the effect on the density of enterochromaffinlike cells in the oxyntic mucosa of the stomach. Female rats were given 1.2 or 2.4 nmol/kg - h of synthetic human [Leu”]-gastrin-17 for 28 days (via osmotic minipumps implanted subcutaneously). As a result, measurable plasma gastrin increased from about 230 pg/ml in the controls to about 500 and 800 pg/ml in the low- and high-dose groups, respectively. The trophic effects of gastrin were reflected in increased stomach weight and oxyntic mucosal mass. Gastrin also increased the enterochromaffinlike cell density and associated parameters (histamine concentration and histidine decarboxylase activity) but was without demonstrable effects on other parts of the digestive tract and pancreas. The results show that continuous infusion of exogenous gastrin for 28 days induces trophic changes similar to those seen after a period of hypergastrinemia induced by treatment with effective inhibitors of acid secretion.

esides stimulating acid secretion, gastrin is generally thought to exert a trophic effect on the acidproducing gastric mucosa and to be involved in the regulation of both the function and proliferation of the histamine-storing endocrine cells, the so-called enterochromaffinlike (ECL) cells, in the oxyntic mucosa (1,2). Evidence for such effects has come from studies in rats with chronic hypergastrinemia after antrum exclusion or treatment with effective acid secretion inhibitors such as omeprazole and ranitidine (3,4). Treatment of rats with high doses of antisecretory agents for 10-20 wk increased the weight and thickness of the oxyntic mucosa and the density of the ECL cells (4,5). These effects were prevented by removal of

B

of Lund, Lund, Sweden

the antrum, which is the main source of gastrin. A close correlation was found between the plasma gastrin level and the ECL-cell density (4). The results were interpreted as follows: sustained inhibition of gastric acid secretion results in hypergastrinemia, which with time increases the mucosal mass and the density of ECL cells. Gastrin has been claimed to exert trophic effects not only on the stomach but also on other parts of the digestive tract and the pancreas (6-8). However, 20 wk treatment with high doses of omeprazole or surgical removal of the acid-producing part of the stomach failed to induce trophic effects in the small intestine or pancreas in spite of very high plasma gastrin levels (9,101. The hypothesis that the omeprazole-evoked trophic effects in the stomach reflect the hypergastrinemia has been challenged (11,~). The aim of the present study was to investigate the trophic effects of exogenous gastrin on the digestive tract and pancreas. Human [Leu”]-gastrin-17 (L-G-171 was administered to rats as a continuous subcutaneous infusion from osmotic minipumps in order to ensure a sustained hypergastrinemia for 28 days.

Materials and Methods Thirty female Sprague-Dawley rats [M$llegaard, Skensved, Denmark], weighing 219-267 g, were used (10 rats/group]. One group received the vehicle (1% albumin], and the other two groups received L-G-17,1.2 nmol/kg - h - h (5.0 pg/kg - h), (2.5 pg/kg . h) and 2.4 nmol/kg respectively.

Abbreviations used in this paper: ECL, enterochcomaflinlike; HDC, histidlne decarboxylase; J.-G17, &eu”]-g&rim17. 0 laao by the American Gastroenterological Association OOlS-5085/BO/g3.00

34 RYBERG ET AL.

Human L-G-17 and albumin were purchased from Sigma [St. Louis, MO.). [Leu”]-gastrin-17 (dissolved in 1% albumin) or the vehicle was administered with osmotic minipumps (ALZET 2 ML 1, ALZA Corp., Palo Alto, Calif.), implanted subcutaneously in the neck of the rats under Brietal anesthesia (1.5 ml/kg intraperitoneally, 30 mg/ml; Eli Lilly, Indianapolis, Ind.). The animals were treated for 28 days and the minipumps were changed every seventh day. Doktacillin (0.1 ml/rat, 250 mg/ml; Astra, Sweden) was administered intramuscularly on days 7, 10. 14, 17, and 21 to prevent bacterial growth in the area where the minipumps were implanted. Plasma gastrin was determined twice a week, on the third and seventh day after implantation of the minipump, i.e., one sampling when the pump was working at maximal capacity (the third day) and one immediately before the pump was replaced (the seventh day). During the drug treatment, blood (300 ~1) was drawn from the tail and collected in small Eppendorf tubes (Microvette, Sarstedt, West Germany] containing EDTA. When treatment was terminated (day 28) the rats were bled from the abdominal aorta under ether anesthesia (with the minipumps still in place), and the blood was collected in tubes containing EDTA (5 ml Venoject; Teruma, Leuven, Belgium]. The tubes were kept on ice. The blood was centrifuged and the ensuing plasma stored at - 30°C. Gastrin was determined in 100 ~1 plasma using a double antibody liquid phase ‘251-radioimmunoassay (Diagnostic Products Corporation, Los Angeles, Calif.). Results are expressed as picogram equivalents of synthetic human gastrin 17 per milliliter plasma. The pancreas was dissected out in toto and weighed. The stomach was removed, opened along the major curvature, rinsed with iced 0.9% saline, and weighed. Small tissue specimens were taken for histology from the oxyntic gland area (for details see reference 10) along the major curvature. The mucosa of the oxyntic gland area was scraped off, weighed, and homogenized in ice-cold 0.01 M sodium phosphate buffer (100 mg mucosa/ml) for determination of histamine and histidine decarboxylase (HDC) activity. Two specimens (each 1 cm long) were taken from the duodenum, one immediately below the duodenal bulb (proximal duodenum) and one 5 cm distally (distal duodenum). They were rinsed in iced 0.9% NaCl solution, gently blotted, weighed, and subjected to histology. The small and large intestines were removed, and specimens (1 and 5 cm in length) were excised from the mid small and mid large intestines. The l-cm specimens were weighed and then subjected to histology. The 5-cm specimens were gently flushed with iced 0.9% saline. The mucosa was scraped off, and the mucosal scrapings were weighed. The tissue specimens that were taken for histology were fixed in ice-cold 4% buffered formalin (pH 7.2) and embedded in paraffin. Sections (8-10 pm thick] were cut perpendicular to the mucosal surface and stained with H&E. At least 10 determinations of mucosal thickness (gland depth or villus height and crypt depth] were made on at least 2 sections from each specimen. Examination was performed with eyepiece x8 and objective x6.3 (visual field diameter, 2.5 mm).

GASTROENTEROLOGY

Vol. 98. No. 1

For the demonstration of ECL cells we used the silver impregnation method of Sevier and Munger and immunostaining using antibodies against the histamine-forming staining, thought to be enzyme HDC (4,131. Sevier-Munger specific for ECL cells in the rat oxyntic mucosa (141, was performed on formalin-fixed, paraffin-embedded material. For the immunocytochemical demonstration of HDC, the specimens were immersed in 4% buffered formaldehyde overnight, washed in sucrose-enriched buffer, frozen on solid CO,, and sectioned at -25’C (8-10 pm). Sections were cut perpendicular to the mucosal surface and stained for HDC using the immunofluorescence method (4). At least 5 determinations of the ECL-cell density (number of cells per visual field) were made on at least 2 transverse sections from each specimen. Basically, the fields were randomly selected. The only inclusion criterion was that they should cover the entire thickness of the mucosa. Only ECL cells with visible nuclei were counted. Examination was performed with eyepiece x 12.5 and objective x 10 (visual field diameter, 1.2 mm]. Morphometric analysis was performed without knowledge of the group to which the specimens belonged. For determination of histamine, the oxyntic mucosal homogenate was diluted 1:lO with phosphate buffer and heated in boiling water for 10 min to release bound histamine. The homogenate was then centrifuged at 6000g for 20 min, and histamine was determined in the supernatant as described previously (4) For determination of HDC activity, 80 ~1 oxyntic mucosal homogenate was incubated with (l-‘4C]L-histidine (20 nCi and 0.36 nmol; New England Nuclear, Boston, Mass.], 5 x 10e4 M L-histidine, and lo-’ M pyridoxal-5-phosphate in a total volume of 160 ~1 at 37°C for 60 min, as described previously (4). The results are expressed as means t SEM. A

1

1

$ 2

ooo--

HOO--

600--

C'l-l c u : 400-VI 200--

01

0

3

7

10

Time,

14

17

21

24

26

days

Figure I. Plasma levels of gastrin in rats treated with vehicle (o), 1.2 nmol/kg . h L-G-17 (A), or 2.4 nmol/kg . h L-G-17 (0) (mean + SEM; n = 10).

TROPHIC EFFECTS OF EXOGENOUS

January 1990

Table 1. Effects of Exogenous

Gastrin

35

GASTRIN INFUSION

on the Wet Weight of the Digestive Tract and the Pancreas in the Rat Wet weight L-G-17 1.2 nmol/kg

Controls

Tissue

Stomach Whole wall Oxyntic mucosa Pancreas Intestine, whole wall (l-cm specimens) Proximal duodenum Distal duodenum Mid small intestine Mid large intestine Mucosal scrapings (5-cm specimens] Mid small intestine Mid large intestine

112 59 52 71

f f + +

L-G-17

ah

1530 * 30 452 f 15 780 f 40

1420 + 30 367 * 20 810 f 40

124 58 53 63

5 4 2 7

+ f f +

6 4 4 6

P

2.4 nmol/kg . h

P

to.05 to.01 NS

1620 + 40 528 + 21 740 * 70

to.01 to.01

123 62 51 77

+ + + +

NS

7 7 1 6

NS NS NS

NS

91 f 9

NS

81 + 5

NS NS

NS NS NS NS

104 zt 8 68 f 7

100 zt 8 71 * 11

fmg]

NS

Results are expressed as mean * SEM. Ten rats in each group.

the low and the high doses of L-G-17, respectively (Figure 1).

nonparametric Kruskal-Wallis test, a one-way analysis of variance by ranks, was used for assessment of the results; p =z 0.05 was considered significant.

Pancreas

Plasma Gastrin

The stomach weight increased by 9% and 14% following treatment with 1.2 and 2.4 nmol/kg - h L-G-17 (Table 1). The increase in stomach weight reflected the increased weight of the oxyntic mucosa; the weight of the mucosal scrapings increased by 25% and 4470, respectively, compared with control values. The thickness of the oxyntic mucosa was increased by the high dose of gastrin [Table 2). The treatment did not increase the weight or the mucosal thickness of other parts of the digestive tract or the weight of the pancreas [Tables 1 and 2). The ECL-cell density according to HDC immunostaining was 177 k 8 in the controls and 186 + 14 (NS) and 251 + 16 (p < 0.01) in the rats treated with 1.2 and

Plasma gastrin was determined during treatment [day 3-day 24) and at termination of treatment (day 28) (Figure 1). In control rats, plasma gastrin varied from 168-335 pg/ml. In animals treated with L-G-17, the plasma gastrin levels were elevated compared with the control values. The increase in the low-dose group (1.2 nmol/kg - h] amounted to 5O%285% (mean, 130%) on all sampling occasions (except on day 21). In the high-dose group (2.4 nmol/kg . h), the gastrin levels were increased by 160%-500% [mean, 260%). The mean plasma gastrin levels during the 4 wk of treatment were about 230 pg/ml in the controls and about 500 and 800 pg/ml in the rats given

Table 2. Effects of Exogenous

Gastrin on Mucosal Thickness Controls

Oxyntic mucosa Antral mucosa Proximal duodenum Villus height CIyptus depth Distal duodenum Villus height Crypt depth Jejunum Villus height Crypt depth Colon

Effects in Stomach, Intestines, and

Trophic

Results

in the Digestive Tract, pm

L-G-17 1.2 nmol/kg

.h

P

L-G-17 2.4 nmol/kg

.h

P

460 * 12 [lo] 218 A 9 (61

450 + 13 (lo] 226 f 7 (6)

NS NS

510 zt 18 (10) 212 * 9 (9)

<0.05 NS

604 f 21(9) 188 * 7 (71

571 * 24 (9) 185 f 4 (8)

NS NS

544 * 24 (7) 195 + 7 (8)

NS NS

457 f 44 (91 151 * 7 (9)

493 * 21(10) 162 k 4 (10)

NS NS

500 * 29 (9) 159 f 7 (91

NS NS

332 f 19 (8) 135 + 8 (8) 244 zt 41(8)

371 + 21(9) 146 + 6 (10) 225 + 4 (10)

NS NS NS

369 * 22 (7) 150 k 7 (8) 222 f 12 (6)

NS NS NS

36 RYBERG ET AL.

GASTROENTEROLOGY

Vol. 98, No. 1

+42X

H

L-G-17 1.2 nmol/kg*h

m

L-G-17 2.4 nmol/kg*h

Figure 2. Density of ECL cells in the oxyntic mucosa after 26 days of treatment with vehicle, 1.2 nmolikg . h L-G-17, or 2.4 nmolbg . h L-G-17 using (A) HDC immunostaining and (II)Sevier-Munger impregnation methods (mean + SEMJ

B 2.4 nmol/kg - h L-G-17, respectively (Figure 2A). The corresponding figures according to Sevier-Munger staining were 160 + 9 [controls], 173 + 9 (NS), and 243 f 18 (p < 0.01) [Figure 2B). The activity of oxyntic mucosal HDC is shown in Figure 3. The HDC activity was elevated from 44.9 f 6.77 pmol CO,/mg wet weight - h in the controls to 126.9 + 16.98 pmol CO,/mg wet weight - h in the animals treated with the high dose of L-G-17. The rats receiving the low dose of L-G-17 did not respond. The histamine concentration in the oxyntic mucosa of the various groups is shown in Figure 4. The concentration increased from 53.0 + 4.06 pg/g wet weight in the controls to 71.5 + 4.76 (+35%) (NS) and 109.8 k 5.45 pg/g wet weight (+107%) in the rats treated with 1.2 and 2.4 nmol/kg - h L-G-17, respectively.

n B

Control

L-G-17 1.2 nmol/kg*

Discussion Continuous infusion of L-G-17 via osmotic minipumps increased the plasma levels of gastrin. [Leu”]-gastrin-17 is a synthetic analogue of human gastrin-17 and is considered more stable than natural human gastrin [Met15]-G-17 while having the same bioactivity (15,16). The plasma gastrin levels varied during the infusion of L-G-17. Comparatively low levels were observed just before the pumps were changed, indicating that the capacity of the pumps had declined. However, the gastrin levels were increased compared with control values. The mean gastrin levels during the whole treatment period were 230 pg/ml in the controls and about 500 and 800 pg/ml in the rats given L-G-17 1.2 and 2.4 nmol/kg . h. Gastrin has a number of actions. Apart from being a potent gastric acid secretagogue, it is trophic for the oxyntic mucosa (1,9,17), including the ECL cells (39).

0

Control

B

L-G-17 1.2 nmol/kg*h

m

L-G-17 2.4 nmol/kg.h

h

Figure 3. Histidine decarboxylase activity in oxyntic mucosa after 28 days’ treatment with vehicle, 1.2 nmol/kg - h L-G-17, or 2.4 nmol/kg . h L-G-17 (mean f SEM; n = 10).

Figure 4. Histamine concentration in oxyntic mucosa after 28 days’ treatment with vehicle, 1.2 nmol/kg . h L-G-17 or 2.4 nmol/kg . h L-G-17(mean + SEM; n = 10).

TROPHIC EFFECTS OF EXOGENOUS

January 1990

Johnson et al. (6-8) claimed that gastrin also evokes trophic effects in other parts of the digestive tract (duodenum, colon, and pancreas). The view that gastrin is trophic for the duodenum, colon, and pancreas is based on the results of experiments in which pentagastrin, not gastrin-17, was administered (6-8). Pentagastrin is likely to stimulate not only gastrin receptors but also cholecystokinin receptors (1). Consequently, pentagastrin might be expected to reproduce the combined effects of gastrin and cholecystokinin. However, in a previous study in which hypergastrinemia was induced in rats by gastric surgery, it was possible to demonstrate trophic effects in the oxyntic mucosa only (10). Similar observations were made in rats and several other species treated with a high dose of omeprazole for 10-20 wk (9). In the present study in which L-G-17 was administered, the trophic effect was confined to the stomach and reflected in an increase in stomach weight, oxyntic mucosal weight, and oxyntic mucosal thickness. These effects were obvious with the high dose: the low dose produced only marginal effects. The rest of the digestive tract (duodenum, small and large intestine) and the pancreas were unaffected. [Leu’5]-gastrin-17 increased the ECL-cell density and the mucosal histamine concentration and HDC activity. Thus, both the number of ECL cells and their capacity to produce histamine had been increased. The increase in ECL-cell density and HDC activity was statistically significant only in the rats given the high dose of L-G-17, which resulted in a plasma gastrin level of about 800 pg/ml (controls, 230 pg/ml). Thus, the plasma gastrin level obtained by the low dose of L-G-17 over a 4-wk period was not enough to produce statistically significant increase in the ECL cell number. A previous study showed a close correlation of the plasma gastrin concentration and ECL-cell density and the histamine concentration (4). In the

0

Figure 5. Correlation between (A) plasma gastrin and oxyntic mucosal histamine concentration and (B)ECL-cell density after 2g days’ treatment with 1.2 nmol/kg . h L-G-17 or 2.4 runol/kg . h L-G-17. Individual data: A. r = 0.76 @ = 0.0001); B, r = 0.70 (p = 0.0001).

GASTRIN INFUSION

37

present study, exogenous gastrin produced trophic effects in the oxyntic mucosa similar to those observed after long-term treatment with the two antisecretagogues omeprazole and ranitidine (4). The effects on the ECL cells were particularly notable and suggested a direct relationship between plasma gastrin and ECL-cell density (or oxyntic mucosal histamine concentration) (Figure 5A and B). The present results support the view that the ECL-cell hyperplasia after treatment with high doses of omeprazole and ranitidine is caused by the increased plasma gastrin levels and not the inhibition of acid secretion per se. In parallel experiments, infusion of human gastrin-17 (2.4 nmol/kg - h) in rats with chronic gastric fistula stimulated acid secretion by about 200% (Ryberg et al., unpublished observation, October 1988). Thus, gastrin can evoke ECL-cell hyperplasia under conditions of stimulated acid secretion. It can be concluded that the trophic effect of exogenous gastrin-17 is confined to the oxyntic mucosa. Other parts of the gastrointestinal tract and the pancreas were unaffected. Like endogenous hypergastrinemia, exogenous gastrin promotes ECL cell hyperplasia. The results support the view that increased plasma gastrin levels rather than achlorhydria per se causes proliferation of ECL cells during treatment with antisecretagogues.

Reference6

1. Hakanson R, Oscarson J, Sundler F. Gastrin and the trophic control of gastric mucosa. Stand J Gastroenterol 1986;21(Suppl 118):18-30. 2. Hakanson R, B6ttcher G, Sundler F, Vallgren S. Activation and hyperplasia of gastrin and enterochromaffin-like cells in the stomach. Digestion 1986;35(Suppl1):23-41. 3. Alumets J, El Munshid HA, Hakanson R, Liedborg G, Oscarson J, Rehfeld JF, Sundler F. Effect of antrum exclusion on endocrine cells of rat stomach. J Physiol1979;268:145-55. 4. Larsson H. Carlsson E, Mattsson H, Lundell L, Sundler F, Sundell G, Wallmark B, Watanabe T, Hakanson R. Plasma gastrin and gastric ECL activation and proliferation. Gastroenterology 1986:90:391-9. 5. Sundler F, Hakanson R, Carlsson E, Larsson H. Mattsson H. Hypergastrinemia after blockade of acid secretion in the rat: trophic effects. Digestion 1986;35(Suppl1):56-69. 6. Johnson LR. Gastrointestinal hormones and their functions, Am Rev Physiol1977:39:135-58. 7. Enochs MR, Johnson LR. Trophic effects of gastrointestinal hormones: physiological implications. Fed Proc 1977;36:1942-7. 8. Johnson LR. Regulation of gastrointestinal mucosal growth. J Surg 1979;3:477-87. 9. Hakanson R, Blom H, Carlsson E, Larsson H. Ryberg B, Sundler F. Hypergastrineamia produces trophic effects in stomach but not in pancreas and intestines. Regul Peptides 1986;13:223-33. 10. Oscarson J. Hakanson R, Liedberg G, Lundqvist G, Sundler F, Thorell J. Variated serum gastrin concentration: trophic effects on the gastrointestinal tract of the rat. Acta Physiol Stand (Suppl) 1979:475:2-27.

38 RYBERG ET AL.

11. Wormsley KG. Is chronic long-term inhibition of gastric secretion really dangerous? Stand J Gastroenterol 1988:23(Suppl 148):X6-74. 12. Penston 1. Wormsley KG. Achlorhydria: hypergastrinaemia: carcinoids-a flawed hypothesis. Gut 1987;28:488-505. 13. Sevier AC, Munger BL. A silver method for paraffin sections of neural tissue. J Neuropathol Exp Neuroll965;24:130-5. 14. Larsson H, Carlsson E, Hakanson R, Mat&on H, Nilsson G, Seensalu R, Wallmark B, Sundler F. Time-course of development and reversal of gastrin, endocrine cell hyperplasia after inhibition of acid secretion. Studies with omeprazole and ranitidine in intact and antrectomized rats. Gastroenterology 1988;95:1477-86. 15. Morley 1s. Structure-activity relations in GI hormones. In: Andersson S (ed). Frontiers in gastrointestinal hormone re-

GASTROENTEROLOGY

Vol. 98, No. 1

search, Nobel symposium 16. Stockholm: Almquist & Wiksell, 1973:143-g. 16. Petersen B. Christiansen J, Rehfeld JF. Acid secretory potency and elimination of the 15-leucine gastrin-17 analogue in man. Stand J Gastroenterol1981;16:437-40. 17. Crean GP, Marshall MW, Ramsay RDE. Parietal cell hyperplasia induced by the administration of pentagastrin (ICI 50,123) to rats. Gastroenterology 1969;57:147-55.

Received December 12,1988. Accepted June 21,1989. Address requests for reprints to: Birgitta Ryberg, Ph.D. Gastrointestinal Research, Department of Biology, AB Hlssle, S-431 83 M6lndal, Sweden.