TOXICOLOGY

AND APPLIED PHARMACOLOGY

106, 5 18-528

(1990)

Hypergastrinemia Is Associated with Decreased Gastric Acid Secretion 2,3,7,8-Tetrachlorodibenzo-p-dioxin-Treated Rats’

in

THOMAS A. MABLY,* H. MICHAEL THEOBALD,*,~ GLYNNIS B. INGALL,? AND RICHARD E. PETERSON*$~ *School

of Pharmacy, School

$.Environmental University

of Medicine,

Received

May

Toxicology Center, and TDepartment of Wisconsin, Madison, Wisconsin

175, 1990; accepted

August

of Pathology, 53706

IS, 1990

Hypergastrinemia Is Associated with Decreased Gastric Acid Secretion in 2,3,7,&Tetrachlorodibenzo-p-dioxin-Treated Rats. MABLY, T. A., THEOBALD. H. M., INGALL, G. B., AND FETERSON, R. E. (1990). Toxicol. Appl. Pharmacol. 106, 518-528. 2,3,7,8-Tetrachlorodibenzo-pdioxin (TCDD) produces a delayed onset hypergastrinemia in rats. The purpose of the present study was to determine if the increased serum gastrin concentrations were caused by decreased gastric acid secretion, decreased plasma clearance of gastrin, and/or decreased gastric emptying. It was found that TCDD treatment decreased gastric acid secretion as determined by decreases in gastric secretory volume, acidity, and total acid output in pylorus-ligated rats. Also, both doseresponse and time-course curves for decreased gastric acid secretion in TCDD-treated rats were similar to those for hypergastrinemia. These findings, as well as a significant inverse correlation between serum gastrin concentrations and total gastric acid output in rats treated with graded doses of TCDD (5- 100 bg/kg), suggest that TCDD-induced decreases in gastric acid production cause elevated serum gastrin concentrations. Neither hypergastrinemia nor decreased gastric acid secretion were observed in pair-fed control rats, demonstrating that neither effect was secondary to undernutrition. The TCDD-induced decrease in gastric acid secretion was not caused by a decrease in the number of acid-secreting parietal cells in the stomach, but rather was associated with a decrease in parietal cell responsiveness to gastrin-elicited acid secretion. This was evidenced by both elevated serum gastrin concentrations and a pharmacological dose of pentagastrin failing to stimulate gastric acid secretion in TCDD-treated rats. The disappearance of iv-administered gas&in- 17 from the serum was not affected by TCDD treatment, suggesting that reduced serum gastrin clearance is not responsible for the TCDD-induced hypergastrinemia. Although a marked decrease in gastric emptying of a 5’Cr-labeled liquid test meal was also observed in TCDD-treated rats, the lowest dose of TCDD required to produce this effect was greater than that needed to cause hypergastrinemia. This suggests that the hypergastrinemic effect of TCDD is not secondary to a decrease in gastric emptying. We conclude that the most probable cause of hypergastrinemia in TCDD-treated rats is decreased gastric acid secretion. 0 1990 Academic PKSS, IIIC.

2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and certain polychlorinated biphenyls (PCBs) cause morphological changes in the gastrointestinal mucosa of animals. The pathology ’ Supported by NIH ’ Current address: Box 413029, Naples. 3 To whom requests 425 N. Charter Street. Wisconsin, Madison, 0041-008x/90

consists of metaplasia of the gastric mucosa in TCDD- and PCB-exposed rhesus monkeys (Allen and Norback, 1973; Becker et al., 1979; Becker and McNulty, 1984; McNulty, 1985) and metaplasia of the intestinal mucosa and adenocarcinomas of the gastric fundic mucosa of PCB-exposed rats (Morgan et al., 1981; Ward, 1985). TCDD also has an “antiatrophy effect” on the gastrointestinal mucosa of rats that is associated with increased circulating

Grant ES-O 1332. Naples Community Hospital, P.0. FL 33941. for reprints should be addressed at School of Pharmacy, University of WI 53706.

$3.00

Copyright 0 1990 by Academic Press. Inc. All rights of reproduction in any form reserved.

518

TCDD-INDUCED

519

HYPERGASTRINEMIA

levels of gastrin (Theobald et al., 1990). This antiatrophy is characterized by the progressive loss of body weight in rats treated with an overtly toxic dose of TCDD, without the accompanying gastric mucosal erosions, ulcerations, or atrophy that typify the gastrointestinal tract of pair-fed control rats that have lost the same amount of body weight (Christian et al., 1986; Theobald et al., 1990). It is possiblethat gastrin might be involved in the antiatrophy action of dioxin since physiological concentrations of gastrin stimulate growth of the gastrointestinal mucosa (Johnson, 1976; 198I), and serum gastrin concentrations are strikingly higher in TCDD-treated rats than in pair-fed and ad lib&m-fed control animals (Theobald et al., 1990). Also, in control rats subjected to severe feed deprivation, serum gastrin concentrations are reduced significantly and gastric mucosal erosions, ulcerations, and atrophy occur which can be reversed by pentagastrin administration (Thaysen and Thaysen, 1949; Pare and Temple, 1973; Johnson et al., 1974). In the present study we sought to determine the mechanism of TCDD-induced hypergastrinemia in rats. Since gastric hypoacidity elevates fasting serum gas&in levels (Peters et al., 1983), it was conceivable that the TCDDinduced hypergastrinemia might be causedby decreasedgastric acid secretion. A decreasein total gastric acid output in TCDD-treated rats could result from a variety of mechanismsincluding a reduction in the actual number of acid-secreting parietal cells in the fundic mucosa of the stomach, or a decreasedability of parietal cells to secreteacid in responseto gastrin stimulation. The former possibility is supported by the finding in rhesus monkeys treated with TCDD or PCBs that almost all of the acid-secreting parietal cells in the stomach are replaced by mucus-secreting cells (McNulty, 1985). Although gastric acid secretion was not measured in these TCDD- and PCBtreated monkeys, such a drastic reduction in parietal cell number would be expected to diminish total gastric acid output. In rats treated with TCDD and related compounds it is not

known if the total number of parietal cells in the fundic mucosa of the stomach is decreased or if gas&in-stimulated gastric acid secretion is reduced. A decreasein gastric acid secretion associatedwith either effect could cause hypergastrinemia. To gain further insight into the causeof the elevated serum gas&in concentrations in TCDD-treated rats (Theobald et al., 1990) we also sought to determine if this effect was due to a decreasein the plasma clearance of gastrin- 17 (G- 17). To this end, the serum disappearance of immunoreactive gastrin following iv administration of G-17 was compared in TCDD-treated and control rats. Last, sincethe physical presenceof food in the stomach stimulates the release of gas&in from G-cells located in the gastric antral mucosa (Lichtenberger, 1982) we investigated whether TCDD might elevate serum gastrin concentrations by decreasinggastric emptying. Accordingly, the effect of TCDD treatment on the stomach emptying rate of a “Cr-labeled liquid test meal was determined. Results of the present investigation show that TCDD-induced hypergastrinemia in rats is most likely caused by a decreasein gastric acid secretion. Furthermore, the decreasein gastric acid production appearsto be due to a decrease in acid-secretory responsivenessof parietal cells to gastrin stimulation.

METHODS Animals and treatments. Adult male Sprague-Dawley rats (275-320 g) were obtained from Harlan SpragueDawley (Indianapolis, IN). The animals were housed individually in suspended wire-mesh cages at a constant room temperature (2 1 + 1“C) with a 12-hr light/dark cycle (lighted, 0500- 1700 hr). Animals were fed ground Purina Rat Chow No. 5012 (Ralston Purina Co., St. Louis, MO) and given water ad libitum. Prior to initiating an experiment, rats were acclimated to feeding and lighting conditions for 7 to 8 days. Rats were then divided into sets of either triplets or pairs such that all members of a set were of similar body weight. Within a triplet set one rat was dosed orally with 100 pg TCDD/kg (TCDD, 98% purity, Cambridge Isotope Laboratories, Woburn, MA) and fed ad libitum (TCDD-treated), one rat was dosed orally with

520

MABLY

an equivalent volume of vehicle (1 ml/kg, corn oil/acetone, 19: 1, v/v) and fed ad libitum (ad libitum-fed control), and one rat was similarly dosed with vehicle and pair-fed to its TCDD-treated counterpart (pair-fed control). Pair-fed rats were dosed (and killed) I day after their TCDD-treated partners. Half of the ad libitum-fed control rats were killed with the TCDD-treated rats and the other half with the pair-fed control rats. Unless otherwise stated, all rats were fasted for 24 hr prior to termination. To ensure that each pair-fed rat was not fasted longer than its ad libitum-fed control or TCDD-treated counterpart, each pair-fed animal was given an additional amount of feed at the start of the fasting period which was equal to the amount that its TCDD-treated counterpart ate on the last day of feeding. The pair-fed rats generally consumed this small amount of feed (2 to 9 g) within 1 hr after it was offered and fasting began immediately thereafter. 23 hr prior to termination. In dose-response studies, groups of rats were administered graded doses of TCDD or vehicle po and were provided feed and water ad libitum. All rats were sacrificed following a 24-hr fast on Day 14 post-treatment. Gastric acid secretion. To determine gastric acid secretion the technique of Shay et al. (1954) was used. Rats were anesthetized with diethyl ether, a midline laparotomy was performed, the pylorus region of the stomach was ligated, and the abdominal incision was closed with sutures. After allowing gastric secretions to accumulate in the pylorus-ligated stomach for 4 hr. each rat was euthanized by diethyl ether overdose. gastric secretions were collected in a 12-ml graduated centrifuge tube, and blood was drawn from the inferior vena cava and placed on ice. After clotting, the blood was centrifuged at 0-4°C. The serum was separated and stored frozen at -76°C for subsequent determination of gastrin concentration. The gastric secretions collected from each rat were centrifuged at 1OOOg for 10 min. The pH and volume of the supematant were quantitated, and an aliquot was assayed for titratable acid concentration by titration with 0.01 N NaOH to pH 7 using a pH meter. Total gastric acid output was calculated for each rat by multiplying the titratable acid concentration of the gastric supematant by the volume of supernatant. To assess gastric secretory responsiveness to pentagasttin stimulation, triplet sets of pylorus-ligated rats (ad libitzrmfed control, pair-fed control, and TCDD-treated) were administered, on Day 14 after TCDD or vehicle treatment. either pentagastrin (50 pg/mlO.9% sodium chloride, 5 ml/ kg, ip; Peninsula Laboratories, Belmont, CA) or 0.9% sodium chloride (5 ml/kg, ip). Ninety minutes later rats were euthanized, gastric secretions collected, and gastric secretory volume, gastric acidity, and gastric acid content of each sample determined. Morphometric analysis of ratjimdic mucosu. The stomach of each rat was removed and prepared for histological examination by the method of Crean et al. (1969). The stomach was opened along the greater curvature and the gastric contents were removed. Immediately prior to fixation the stomach was pinned flat (serosal surface down)

ET

AL.

to a paraffin block. The degree of stretching was just sufficient to eliminate mucosal folds. The stretched specimen was fixed chemically by floating the paraffin block with the specimen side down in Bouin’s solution for 48 hr. The specimen was then removed from its paraffin block support and transferred to neutral buffered formalin until it was processed for histology. Subsequently, the flattened specimen was removed and trimmed carefully to remove the fundus from both the squamous and the antral portions of the stomach, blotted dry, and weighed. The perimeter of the fundus was traced on white paper, and the area determined by three consecutive readings with a planimeter. Three full thickness samples of the fixed gastric wall (4 X 8 mm) were cut across the greatest hemidiameter of the fundic area and routinely processed in paraffin for light microscopic examination. Parietal cells in 5-flrn sections were stained and counted by the procedure of Marks and Drysdale (1957). Using a calibrated ocular grid, parietal cell nuclei in a 0.1 -mm-wide column (full thickness) were counted. Five counts on randomly selected fields were made on each section. To eliminate bias. the person doing the counting was unaware of which treatment group was the source of each tissue. Mucosal thickness was determined from 10 readings on each section with a calibrated ocular micrometer. Final calculations were made according to the method of Card and Marks (1960). Serum disappearance ofgastrin. Rats were anesthetized with pentobarbital sodium (50 mg/kg, ip). and the femoral artery and femoral vein cannulated (PE-50) for blood sampling and G-17 infusion, respectively. After surgery, a thermistor probe was inserted into the colon and connected to a Tele-Thermometer (Yellow Springs Instrument Co., Yellow Springs, OH). Body temperature was maintained at 37.0 f 05°C by an incandescent lamp placed near the animal. An initial arterial blood sample (0.5 ml) was obtained prior to the infusion of G- 17 to determine the basal serum gastrin concentration. Synthetic human gastrin I (G-17-1; Sigma Chemical Co., St. Louis, MO), dissolved in 0.9% sodium chloride, was infused iv at 800 ng/kg/hr for 2 hr at 0.65 ml/hr. For the first 5 min the infusion was at five times this rate. Blood samples were obtained at 30, 60, 90, and 120 min of infusion and at 2, 4, 8, 16, and 32 min postinfusion. To prevent hypovolemia, the volume of blood withdrawn was replaced with an equal volume of 0.9% sodium chloride. Serum was obtained and stored at -76°C until assayed for gastrin by radioimmunoassay. Serum disappearance of immunoreactive gastrin was calculated from the apparent biexponential decline in serum gas&in concentration upon cessation of the G-17 infusion. Half-lives were calculated using tf = 0.693/a or ff = 0.693/p, where LY and @ are estimates of the slopes of the initial and final components of serum immunoreactive gastrin disappearance, respectively. Gastrointestinal transit. To measure gastric emptying and small intestinal transit of a liquid test meal in ad li-

TCDD-INDUCED

521

HYPERGASTRINEMIA

bitrlm-fed control, pair-fed control, and TCDD-treated rats, a solution containing the nonabsorbable radioactive marker sodium “Cr (0.5 PCi 5’Cr in 3 ml of 0.9% sodium chloride, pH 7.4; New England Nuclear, Boston, MA) was administered po (Derblom et nl., 1966). Thirty minutes later the animal was euthanized and serum collected for gastrin analysis. Ligations were secured at the esophogus, pylorus, and ileocaecal junction. The stomach and small bowel were carefully removed and the small intestine was subsequently divided into four consecutive segments of equal length. The amount of ‘ICr-derived radioactivity in the stomach and each intestinal segment was determined by counting in a gamma counter (Packard Model 5236). Gash radioimmunoassay. A radioimmunoassay kit for determining gastrin concentrations in human serum (Becton-Dickinson Immunodiagnostics, Orangeburg, NY) was used to measure gas&in concentrations in rat serum. Antiserum in the kit cross-reacts with hG- 17-1, 100%; hG17-B. 67%; hG-34-1, 64%; CCK-PZ, 1.9%; and CCK-8, 3.1%. Parallelism with the kit was demonstrated for rat serum. While it was necessary to terminate rats and obtain serum samples on different calendar days, the samples were routinely stored until a given experiment was completed. Radioimmunoassays were then performed so that all serum samples from a given experiment were assayed at the same time. Therefore, results of each experiment were not affected by interassay variation in the gastrin radioimmunoassay. Statistical anal.vsis. Data were analyzed by one-way analysis of variance followed by least significant difference tests (Steel and Torrie, 1980). Where appropriate, data were analyzed by Student’s t test for unpaired observations. When variances were heterogeneous by Bartlett’s tests or F tests, the data were natural log- or square root-transformed prior to analysis. For data that did not display homogeneity of variance, even when transformed (serum gastrin concentrations in Fig. 8). the Kruskal-Wallis nonparametric one-way analysis of variance followed by the distribution-free multiple comparison test was used to determine treatment-related effects (Gad and Weil, 1988). Significance for all analyses wasp < 0.05.

RESULTS

Serum gastrin concentrations and gastric acid secretion. Figure 1 shows the effects of TCDD treatment and paired-feed restriction on body weight and serum gastrin concentrations in pylorus-ligated rats at the designated times post-treatment. Both TCDD treatment and paired-feed restriction produced comparable decreases in body weight at 6 and 13 days post-treatment, respectively. On Day 7, serum gas&in concentration was similar in all

DAY 6

DAY 13

TREATMENTGROUP

FIG. 1. Effects of TCDD treatment or paired-feed restriction on body weight and serum gastrin concentration in pylorus-ligated rats that were vehicle-treated [ad libitumfed control (AL) and pair-fed control (PF)] or TCDDtreated (TCDD, 100 pg/kg PO). Body weight was determined 6 and 13 days post-treatment. On Days 7 and 14 (following a 24-hr fast) the pylorus was ligated for gastric acid secretion determinations. After 4 hr of pylorus ligation, rats were euthanized and blood and gastric secretions obtained. Values are means i SE of 14-20 rats. An “a” indicates a significant difference from ad libitum-fed control and a “p” a significant difference from pair-fed control (p < 0.05).

groups, but on Day 14 there was a marked increase in the serum gastrin concentration in the TCDD group compared to the ad libitumfed and pair-fed controls. Figure 2 shows the effects of TCDD treatment on total gastric secretory volume (top panel), pH of gastric secretions (middle panel), and total gastric acid secretory output (bottom panel). On Day 7 post-treatment there was a decrease in total gastric secretory volume in TCDD-treated rats compared to ad Zibitumfed controls. On Day 14 all indices of gastric secretion in ad libitum-fed and pair-fed control groups were similar, whereas TCDD-treated rats displayed significant decreases in gastric secretory volume (top) and gastric acid secretory output (bottom) and an increase in the

522

MABLY

ET

AL.

Parietal cell number and responsiveness to pentagastrin-stimulated gastric acid secretion. Table 1 shows that there was a significant loss of body weight in TCDD-treated rats 14 days post-treatment. Yet there were no detectable differences in stomach wet weight, fundic surface area, fundic mucosal height, or fundic mucosal volume. Most notably there was no effect of TCDD treatment on either the number of parietal cells per unit area of fundus or the total number of parietal cells in the entire fundus. Pentagastrin stimulates parietal cells located in the fundic mucosa of the stomach to secrete acid. Figure 5 shows that in ad libitum-fed and pair-fed control rats, pentagastrin administration increased both gastric secretory volume

s 0.60 2zz e? gw” E 0.30 JF P

0 AL PFTCDD

TREATMENT

AL PF TCDD

GROUP

FIG. 2. Effects of TCDD treatment or paired-feed restriction on total gastric secretory volume, acidity of gastric secretions. and total gastric acid secretory output in pylorus-ligated rats. All other conditions as in Fig. 1.

pH of gastric secretions (middle). The amount of solid material collected after centrifugation of the gastric secretions was not significantly different among the three treatment groups at either Day 7 or 14 (results not shown). Dose-response relationships for the loss of body weight, increase in serum gastrin concentration, and decrease in gastric acid secretion 14 days after TCDD dosing are shown in Fig. 3. TCDD doses ranging from 27 to 100 pg/kg decreased body weight, whereas doses ranging from 65 to 100 pg/kg increased serum gas&in concentration and decreased total gastric acid output. The inverse relationship between serum gas&in concentration and total gastric acid secretory output is shown in Fig. 4. In ad libiturn-fed rats treated with either vehicle (control) or graded doses of TCDD (5-100 pg/kg) there was a significant inverse correlation between serum gastrin concentrations and total gastric acid secretion.

P

ok+--?--0

TCDD

DOSE

50

loo

tug/kg)

FIG. 3. Dose-response effects of TCDD treatment on body weight, serum gastrin concentration, and total gastric acid secretory output in pylorus-Iigated rats. Body weight was determined 13 days post-treatment. On Day 14 (following a 24-hr fast) the pylorus was ligated, and after 4 hr rats were euthanized and their serum gastrin concentration and total acid output determined. Values are means t- SE of 6-8 rats. An asterisk indicates a significant difference from ad libitnm-fed control (p < 0.05).

TCDD-INDUCED

523

HYPERGASTRINEMIA

FIG. 4. Relationship between serum gastrin concentration and total acid secretory output in ad lib&m-fed control rats (0) and rats treated with 5-100 pg/kg of TCDD (0). Each value is the mean of 6-8 rats. All other conditions as in Fig. 3.

(top panel) and gastric acid secretory output (bottom panel), and decreased pH of gastric secretions (middle panel). In contrast, pentagastrin administration had no significant effect on any parameter of gastric secretion in TCDD-treated rats. Serum gastrin disappearance. The effect of TCDD treatment on serum disappearance of immunoreactive gastrin after a 120-min iv infusion of G-17 is shown in Fig. 6. Prior to starting the G- 17 infusion, ad-libitum-fed control and TCDD-treated rats had serum

TABLE

gastrin concentrations of 49 -t 6 and 22 1 1 35 pg/ml (means +- SE), respectively. In both treatment groups, serum gastrin concentrations plateaued 30 min after starting the G- 17 infusion and remained constant until the end of the 120-min infusion when the concentration in ad libitum-fed control rats was 304 +- 30 pg/ml and in TCDD-treated rats, 490 + 84 pg/ml. Figure 6 showsthat when the G- 17 infusion was stopped (time 0), serum gas&in concentrations decreasedin a biexponential fashion in both treatment groups. The calculated t4 values for the initial and final components of the serum disappearance of immunoreactive gastrin in control and TCDD-treated rats, respectively, were similar (Fig. 6, inset). Gastric emptying. To determine if TCDD delays stomach emptying, effects of pair-feeding or TCDD treatment on the gastrointestinal distribution of a 5’Cr-labeled liquid test meal (30 min after po administration) was examined in rats on Days 7 and 14 post-treatment (Fig. 7). At Day 7 there was an increase in the percent oral dose of 51Cr in the stomach of pair-fed control rats compared to ad libitumfed controls and a decreasein the percent oral

1

Ad libitum-fed control Body weight (g) Initial Final b Stomach wet weight (mg)d Fundic surface area (cm2)d Fundic mucosal height (Gm)d Fundic mucosal volume (mm3)d Parietal cell number per unit area (5 10 km2)d Total parietal cell population (X 106)d

305 332 1003 10.67 454 483 24.0 51.4

a Rats were treated with either vehicle (ad libitum-fed control) or TCDD for four rats. b Determined on Day 13. L1Significantly different from ad libitum-fed control group (p < 0.05). d Determined after a 24-hr fast on Day 14.

+ + f f i iIt *

TCDD

304 234 968 10.13 497 502 23.6 47.5

1

7 35 0.41 19 9 0.7 2.0

(100 pg/kg,

PO). Values

2 -r + + i + + *

I 4’ 12 0.39 17 7 1.2 0.7

are the means + SE

MABLY

524

ET AL.

concentration and decrease in gastric emptying 14 days post-treatment are shown in Fig. 8. Increases in serum gastrin concentrations occurred at TCDD doses of 65-100 pg/kg while decreased gastric emptying was observed at 100 pg/kg. DISCUSSION

AL

PF

TREATMENT

TCDD

GROUP

FIG. 5. Effects of TCDD on pentagastrin-elicited gastric acid secretion. Gastric secretory volume, acidity of gastric contents, and total gastric acid secretory output is shown in response to an ip injection of saline (open bars) or pentagastrin (solid bars) in pylorus-ligated rats pretreated with vehicle [ad libitum-fed control (AL) and pair-fed control (PF)] or TCDD (100 pg/kg, PO). Rats were administered TCDD (100 wg/kg, po) or vehicle on Day 0. After a 24hr fast, on Day 14. the pylorus was ligated and the animal injected ip with either pentagastrin (250 pg/kg) or saline. After 90 min rats were euthanized and gastric secretions collected and assayed for pH and total acid content. Values are means t SE of 6-8 rats. An asterisk indicates a significant difference (p < 0.05) from saline-injected rats within the same treatment group (AL, PF, or TCDD).

dose of “Cr in the fourth quarter of the small intestine of TCDD-treated rats compared to pair-fed controls. At Day 14 there were no differences among treatment groups in the percentage oral dose of 51Cr in any of the intestinal segments. However, the percentage oral dose of “Cr in the stomach of TCDD-treated rats was 2.5 times greater than that in ad libitumfed controls and 6 times greater than that in pair-fed controls. Dose-response relationships in TCDDtreated rats for the increase in serum gas&in

TCDDGnduced hypergastrinemia. Hypergastrinemia occurred in TCDD-treated rats 2 weeks after an overtly toxic dose of TCDD. However, the hypophagia-induced weight loss that is associated with overt TCDD toxicity was not responsible for the hypergastrinemic response since pair-fed control rats that lost the same amount of body weight as their TCDD-treated partners were not hypergastrinemic. Also, it is well established that food deprivation in rats decreases the serum gastrin concentration (Lichtenberger et al., 1975) rather than increasing it as was seen in rats treated with TCDD. An alternative explanation for the hypergastrinemia is that it is caused by a decrease in the rate at which gastrin is cleared from the blood. The major circulating forms of gastrin are heptadecapeptide gastrin (G-l 7) and big gastrin (G-34) (Strunz et al., 1978; Doyle et al., 1984). If hypergastrinemia was due to de-

TIME AFTER STOPPING G-17

INFUSION (mini

FIG. 6. Effect of TCDD treatment on the serum disappearance of immunoreactive gastrin following an iv infusion of G-17. Serum disappearance of gastrin was determined after a 24-hr fast on Day 14 post-treatment. Ad libitum-fed control group is designated by (0), and TCDDtreated group by (0). Values are means t- SE of 5 rats.

TCDD-INDUCED

QL

PF TREATMENT

525

HYPERGASTRINEMIA

TCOO GROUP

FIG. 7. Effects of TCDD treatment or paired-feed restriction on the gastrointestinal distribution of a “Cr-labeled liquid test meal. 30 min after po administration. The test meal was administered after a 36-hr fast on Day 7 or 14 post-treatment to vehicle-treated rats [ad libitumfed control (AL) or pair-fed control (PF)] or TCDD-treated rats (100 pg/kg, PO). The stomach is designated by (W) and small intestine segments by: (a) first quarter, (0) second quarter, (N) third quarter, and @) fourth quarter. Values are means f SE of 5 rats. An “a” indicates a significant difference from ad libitum-fed control and a “p” a significant difference from pair-fed control (p < 0.05).

creased G- 17 catabolism and/or excretion it might be reflected in a slower serum disappearance of immunoreactive gastrin in TCDD-treated rats infused with G- 17. Yet the tt for disappearance of immunoreactive gastrin from serum of control and TCDD-treated rats was similar. While this finding does not rule out the possibility of TCDD treatment affecting the serum disappearance of G-34, it seems unlikely that a decrease in serum gastrin clearance is a major cause of the hypergastrinemia in TCDD-treated rats. If TCDD-induced hypergastrinemia cannot be attributed to a decrease in the serum disappearance of gastrin, it must result from an increase in the rate at which gastrin is secreted into the circulation. An important physiological stimulator of gastrin release is the ingestion

of food. It has been shown in rats that serum gastrin levels are directly related to the quantity of food consumed (Johnson et al., 1975; Lichtenberger et al., 1975). It is believed that food components, such as small peptide and amino acid fragments from the digestion of proteins, stimulate gastrin secretion by contacting the antral mucosa of the stomach (McArthur et al., 1983). The decreased gastric emptying rate of a liquid test meal in TCDDtreated rats suggests that food components might remain in contact with the antral mucosa longer than in control rats and potentially cause hypergastrinemia. However, additional results of the present study argue against this mechanism. Most notably, hypergastrinemia occurs in TCDD-treated rats following a 24hr fast at a time when there is the same amount of undigested feed in the stomach of TCDDtreated animals as in pair-fed and ad libitumfed controls. Furthermore, the lowest dose of TCDD that caused hypergastrinemia did not

3 so~L+--p& 0

TCDD WSE

lug/kg)

FIG. 8. Dose-response effects of TCDD treatment on serum gastrin concentration and gastric emptying of a “Crlabeled liquid test meal on Day 14 post-treatment. Rats were fasted 24 hr prior to determination of serum gastrin concentrations and gastric emptying. Values are means + SE of 6- 10 rats. An asterisk indicates a significant difference from ad libitum-fed control (p < 0.05).

526

MABLY

decreasegastric emptying. We conclude that delayed gastric emptying is not the primary cause of hypergastrinemia in TCDD-treated rats. Somatostatin, releasedfrom antral D-cells, exerts an inhibitory effect on gastrin secretion from antral G-cells (Larsson et al., 1979). Therefore, the hypergastrinemia in TCDDtreated rats could be due to a diminished effect of the somatostatin restraint that is normally exerted on antral gastrin secretion. The finding that TCDD treatment decreasesantral content of somatostatin in rats (Potter et al., 1983; Theobald et al., 1988) supports this potential mechanism. Furthermore, the TCDD doseresponse relationship for decreasing antral content of somatostatin is shifted to the left of that for hypergastrinemia, and the decreasein antral somatostatin content in TCDD-treated rats can be detected 1 week earlier than hypergastrinemia (Theobald, 1988). We suggest that a lessening of somatostatin restraint on antral gastrin secretion is a potential causeof the hypergastrinemia in TCDD-treated rats. Acid secretion by parietal cells located in the fundic mucosa of the stomach is involved in regulating gastrin secretion by G-cells in the antral mucosa. More specifically, gastrin secretion is inhibited by acidification of the antral mucosa (Walsh et al., 1975) and is stimulated by alkalinization (Peters et a/., 1983; Mohammed et al., 1983). A major finding of the present study was the close association between hypergastrinemia and decreasedgastric acid secretion in TCDD-treated rats. Both the dose responseand the time course for hypergastrinemia in TCDD-treated rats were similar to those for decreased gastric acid secretion, and there was a significant inverse correlation between serum gastrin concentration and total gastric acid output in rats treated with graded dosesof TCDD. We conclude that decreased gastric acid secretion, along with a lessening of somatostatin restraint on antral gas&in secretion, is a probable cause of the hypergastrinemia in TCDD-treated rats. In addition, neither hypergastrinemia nor decreasedgastric acid secretion was observed

ET AL.

in pair-fed control rats that lost the same amount of body weight asTCDD-treated rats. This demonstratesthat neither effect of TCDD was secondary to undernutrition.

TCDD diminishes parietal cell responsiveness to gastrin-stimulated acid secretion. The decreased gastric acid secretion in TCDDtreated rats could be caused potentially by a decrease in parietal cell mass of the fundic mucosa and/or a decreasein parietal cell responsivenessto gastrin-elicited acid secretion. In rhesus monkeys, parietal cells are replaced by mucous-secreting cells as early as 12 days after dietary PCB exposure (Becker et al., 1979). Becker and McNulty (1984) suggested that the PCB-induced lossof parietal cells was the result of either a block in the differentiation of stem cells into acid-secreting parietal cells or, considering the long half-life of parietal cells (Johnson, 1981) conversion of mature parietal cells into mucous-secreting cells. In contrast to theseprevious findings in PCB- and TCDD-treated rhesus monkeys (McNulty, 1985), we found no decrease in the number of parietal cells in the fundic mucosa of rats 14 days after TCDD treatment. This may have occurred becauseit takes longer after exposure to TCDD and related compounds for parietal cell lossto develop in rats than in monkeys. Since the population of parietal cells in the fundic mucosa of rats is not reduced by TCDD treatment, decreasedparietal cell responsiveness to gastrin stimulation most likely accounts for the decreasedgastric acid secretion. In support of this interpretation, the administration of pentagastrin at a dose which produced maximal gastric acid secretion in control rats (Barrett, 1966) had no stimulatory effect on gastric acid secretion in TCDD-treated rats. Also, in TCDD-treated rats that were markedly hypergastrinemic, gastric acid secretion wasreduced. We conclude that parietal cells in TCDD-treated rats are lessresponsive to gas&in-stimulated acid secretion. It is also possiblethat decreasedgastric acid secretion in TCDD-treated rats could be due to the secretion of biologically inactive gastrin precursors or lesspotent gastrins. Post-trans-

TCDD-INDUCED

HYPERGASTRINEMIA

lational modification of gastrin precursors within the G-cell can result in the formation of serum gas&ins that differ in number of amino acids (Rehfeld, 1972) and in potency to stimulate gastric acid secretion (Walsh and Grossman, 1975; Walsh and Lam, 1980). On a molar basis G- 17 is approximately six times more potent at stimulating gastric acid secretion than G-34. However, in hypergastrinemia of antral origin the percentage of gastrins with longer chains and lower biologic potency is increased (Anderson et al., 1983; Anderson et al., 1985). Thus, if TCDD caused less potent forms of immunoreactive gastrin to be secreted, it could explain why the high serum gastrin concentrations in TCDD-treated rats were unable to stimulate gastric acid secretion. Even if the circulating forms of gastrin in TCDD-treated rats were found to be less potent in stimulating gastric acid secretion, this would not negate our conclusion that TCDD diminishes parietal cell responsiveness to gastrin. This was clearly demonstrated by a pharmacologic dose of pentagastrin failing to enhance gastric acid secretion in TCDD-treated rats.

Other halogenated aromatic hydrocarbons. Persistent coplanar congeners of various halogenated aromatic hydrocarbons such as halogenated biphenyls, dibenzo-p-dioxins, dibenzofurans, azobenzenes, and azoxybenzenes are structurally similar to TCDD and produce a TCDD-like pattern of biological responses (Poland and Knutson, 1982). Inasmuch as these congeners act by the same receptor-mediated mechanism as TCDD in other organs, we would predict that they will also elicit decreased gastric acid secretion and hypergastrinemia in rats. However, the Ah receptor has yet to be characterized in acid-secreting parieta1 cells of the rat stomach and structure-dependent effects of TCDD and related compounds on these two responses remain to be determined.

Possible role of decreased gastric acid secretion and hypergastrinemia in TCDD toxicity. Although decreased gastric acid secretion and hypergastrinemia

are not primary

effects

527

of TCDD treatment they may contribute to TCDD toxicity. Decreased gastric acid secretion might affect the gastric phase of digestion in addition to causing hypergastrinemia. Also, since gas&in has a trophic effect on the fundic mucosa of the ral stomach (Johnson, 198 1) the hypergastrinemia in TCDD-treated rats might explain why the fundic mucosa of these animals does not undergo atrophy as is observed in pair-fed control rats during later stages of the wasting syndrome (Theobald et al., 1990).

ACKNOWLEDGMENTS The authors are grateful to Dr. Laurence B. Katz of the R. W. Johnson Pharmaceutical Research Institute, Raritan, New Jersey 08869, for his expert advice on gastric acid secretion measurements and for critically reviewing this manuscript prior to its submission for publication.

REFERENCES ALLEN, J. R., AND NORBACK, D. H. (1973). Polychlorinated biphenyl- and triphenyl-induced gastric mucosal hyperplasia in primates. Science 179,498-499. ANDERSON,B. N., PETERSON,B., AND BORCH, K. (1985). Decreased sulfation of serum and tissue gastrin in hypergastrinemia of antral origin. Digestion 31, 17-24. ANDERSON, B. N., PETERSON,B., BORCH, K., AND REHFELD, J. F. (1983). Variations in the sulfation of circulating gastrins in gastrointestinal diseases. Stand. J. Gastroenterol.

18, 556-569.

BARRETT, A. M. (1966). Specific stimulation of gastric acid secretion by a pentapeptide derivative of gastrin. J. Pharm.

Pharmacol.

I&633-639.

BECKER, G. M., AND MCNULTY, W. P. (1984). Gastric epithefial cell proliferation in monkeys fed 3,4,3’,4’-tetrachlorobiphenyl. J. Pathol. 143, 267-274. BECKER,G. M., MCNULTY, W. P., AND BELL, M. (1979). Polychlorinated biphenyl-induced morphologic changes in the gastric mucosa of the rhesus monkey. Lab. Invest. 40, 373-383.

CARD, W. I., AND MARKS, I. N. (1960). The relationship between the acid output of the stomach following “maximal” histamine stimulation and parietal cell mass. Clin. Sci. 19, 147-163. CHRJSTIAN,B. J., INHORN, S. L., AND PETERSON, R. E. (1986). Relationship of the wasting syndrome to lethality in rats treated with 2,3,7,8-tetrachlorodibenzo-g-dioxin. Toxicol.

Appl. Pharmacol.

82, 239-255.

528

MABLY

ET

AL.

CREAN, G. P., MARSHALL, M. W., AND RUMSEY, MORGAN, R. W., WARD, J. M., AND HARTMAN, P. E. R. D. E. (1969). Parietal cell hyperpiasia induced by the (198 1). Arocior 1254-induced intestinal metaplasia and administration of pentagastrin (ICISO, 123) to rats. Gusadenocarcinoma in the glandular stomach of F344 rats. Cancer Res. 41, 5052-5059. troenterology 57, 147- 155. DERBLOM, H., JOHANSSON, H., AND NYLANDER, G. PARE, W. P., AND TEMPLE, L. J. (1973). Food deprivatioti, (1966). A simple method of recording quantitatively shock stress and stomach lesions in the rat. Physiol. certain gastrointestinal motility functions in the rat. Actu Behav. 8, 371-375. PETERS, M., FELDMAN, M., WALSH, J., AND RICHARDSON, Chir. Scar& 132, 154-165. C. (1983). Effect of gastric alkalinization on serum gasDOYLE, J. W., WOLFE, M. M., AND MCGUIGAN, J. E. trin concentrations in humans. Gastroenterology 85,35(1984). Hepatic clearance of gastrin and cholecystokinin 39. peptides. Gastroenterology 87, 60-68. POLAND, A., AND KNUTSON, 5. C. (1982). 2,3.7,8-TetraGAD, S. C., AND WEIL, C. S. (1988). Statistics and Exchlorodibenzo-p-dioxin and related halogenated aroperimental Design Jbr Toxicologists. Telford Press, matic hydrocarbons: Examination of the mechanism of Caldwell, NJ. toxicity. Annu. Rev. Pharmacol. Toxicol. 22, 5 17-554. JOHNSON, L. R. (1976). The trophic action of gastroinPOTTER, C. L., SIPES, I. G., AND RUSSELL, D. H. (1983). testinal hormones. Gustroenterology 70, 278-288. Hypothyroxinemia and hypothermia in rats in response JOHNSON, L. R. (198 1). Regulation of gastrointestinal to 2,3,7,8-tetrachlorodibenzo-p-dioxin administration. growth. In Physiology of the Gastrointestinal Tract Toxicol. Appl. Pharmacol. 69, 89-95. (L. R. Johnson, Ed.), pp. 169-198. Raven Press, New REHFELD, J. F. (1972). Three components of gastrin in York. human serum. Gel filtration studies on the molecular JOHNSON, L. R., COPELAND, E. M., DUDRICK, S. J., size of immunoreactive serum gastrins. Biochim. BioLICHTENBERGER, L. M., AND CASTRO, G. A. (1975). phys. Acta 225, 364-312. Structural and hormonal alterations in the gastrointesSHAY, H., SUN, D. C. H., AND GRUENSTEIN, M. (1954). tinal tract of parenterally fed rats. Gastroenterology 68, A quantitative method for measuring spontaneous gas1177-l 183. tric secretion in the rat. Gastroenterology 26, 906-9 13. JOHNSON, L. R., GASTRO, G. A., AND LICHTENBERGER, STEEL, R. G. D., AND TORRIE, J. H. (1980). Principles L. M. (I 974). The significance of the trophic action of and Procedures of Statistics. A Biometrical Approach, gastrin. Gastroenterology 66, 118. 2nd ed. McGraw-Hill, New York. LARSSON, L., GOLTERMANN, N., DE MAGISTRIS, L., STRUNZ, U. T., THOMPSON, M. R., ELASHOFF, J., AND REHFELD, J. F., AND SCHWARTZ, T. W. (1979). SOGROSSMAN, M. I. (1978). Hepatic inactivation of gasmatostatin cell processes as pathways for paracrine setrim of various chain lengths in dogs. Gustroenterology cretion. Science 205, 1393- 1394. 74,550-553. LICHTENBERGER, L. M. (1982). Importance of food in the THAYSEN, E. H., AND THAYSEN, J. H. (1949). Morphoregulation of gastrin release and formation. Amer. J. logical changes in the gastrointestinal tract of the white Physiol. 243, G429-G44 1. rat following inanition. Acta Pathol. Microbial. Stand, LICHTENBERGER, L. M., WELSH, J. D., AND JOHNSON, 26,370-380. L. R. (1975). Relationship between the changes in gastrin THEOBALD, H. M., MABLY, T. A., INGALL, G. B., AND levels and intestinal properties in the starved rat. Amer. PETERSON, R. E. (1988). Antiatrophy effect of 2,3,7,8J. Dig Dis. 21, 33-38. tetrachlorodibenzo-p-dioxin (TCDD) on rat gastric MCARTHUR, K. E., ISENBERG, J., HOGAN, D., AND DRIER, mucosa and its possible relationship to hypergastrinS. (1983). Intravenous infusion of L-isomers of phenylemia. Toxicologist 8, 92. alanine and tryptophan stimulates gastric acid secretion THEOBALD, H. M., MABLY, T. A., INGALL, G. B., AND at physiological plasma concentrations in normal subPETERSON, R. E. (1990). Role of hypergastriuemia in jects and after par&al cell vagotomy. J. Clin. Invest. the antiatrophy effect of 2,3,7,8-tetrachlorodibenzo-p71,1X4-1262. dioxin on oxyntic gland mucosa of the rat stomach. J. MCNULTY, W. P. (1985). Toxicity and fetotoxicity -of Biochem Toxicoi. 5, in press. TCDD, TCDF and PCB isomers in rhesus macaques WALSH, J. H., AND GROSSMAN, M. I. (1975). Gastrin. N. (Macaca mulatta). Environ. Health Perspect. 60, 77Engl. J. Med. 292, 1324-1334. 88. WALSH, J. H., AND LAM, S. K. (1980). Physiology and MARKS, I. N., AND DRYSDALE, K. M. (1957). A modifipathology of gastrin. In Clinics in Gastroenterology (W. cation of Zimmerman’s method for differential staining Creutzfeldt, Ed.). Vol. 9, No. 3, pp. 567-591. W. B. of gastric mucosa. Stain Technol. 32, 48. Saunders Co., Ltd., Philadelphia. MOHAMMED, R.. HOLDEN, R. J., HAEARNS, J. B., WALSH, J., RICHARDSON, C., AND FORDTRAN, J. (1975). MCKIBBEN, B. M., BUCHANAN, K. D., AND CREAN, pH dependence of acid secretion and gastrin release in G. P. (I 983). Effects of eight weeks’ continuous treatnormal and ulcer patients. J. C/in. Imwt. 55, 462-469. ment with oral ranitidine and cimetidine on gastric acid WARD, J. M. (1985). Proliferative Iesions of the glandular secretion, pepsin secretion, and fasting serum gastrin. stomach and liver in F344 rats fed diets containing ArGut 24, 61-66. oclor 1254. Environ. Health Perspect. 60, 89-95.