TISSUE AND CELL, 1992 24 (5) 625-633 0 1992 Longman Group UK Ltd.
JOHN M. DALY*, PHILLIP S. OATES and REGINALD G.H. MORGAN
THE EFFECT OF FASTING AND CHOLECYSTOKININ STIMULATION ON NORMAL AND NEOPLASTIC TISSUE IN THE RAT PANCREAS Keywords: Atypical acinar cell nodule, secretion, cholecystokinin.
enzymes. azaserine, neoplastic function.
ABSTRACT. Pancreatic nodules were produced in rats by either feeding raw soya flour alone or by injection of azaserine plus raw soya flour feeding. The resulting nodules were studied to determine whether there was any functional difference between this tissue and the relatively normal internodular pancreas. Tissue DNA and trypsin content were significantly elevated in nodules compared to the adjacent tissue. With fasting, protein and enzyme content increased significantly and equally in both nodular and internodular tissues. RNA levels fell significantly and the decrease was more pronounced in nodular tissue. The responsiveness of the multinodular pancreas to cholecystokinin was examined by measuring pancreatic secretion basally and in response to cholecystokinin. Both the volume and protein content secreted by the multinodular pancreas were greatlv elevated above control levels. When corrected for pancreatic weight, the difference remained significant and appeared to be due to increased basal secretion by the nodular pancreas. These studies demonstrate that azaserine-raw soya flour induced nodules are functionally efficient. Furthermore, the secretory response to cholecystokinin of these nodules is equal to or higher than that of normal tissue.
Introduction
Feeding raw soya flour (RSF) continuously to rats results initially in hypersecretion (Lyman and Lepkovsky, 1957; Oates and Morgan, 1982b), followed by hypertrophy (Oates and Morgan, 1984) and then hyperplasia (Oates and Morgan, 1982b; Oates and Morgan, 1984) of the pancreas. This effect is largely due to the presence within RSF of trypsin inhibitors that are known to raise plasma cholecystokinin (CCK) levels (Falsch et al., 1984), probably by preventing trypsin from inactivating a peptide which stimulates CCK release (Lu er al., 1989). CCK is a powerful pancreatic secretagogue (Peikin et al., 1978;
Department of Physiology, University of Western Australia, Nedlands 6009. W.A., Australia. *To whom correspondence should be addressed. Received 23 April 1992. *Abbreviations: AACN, atypical acinar cell nodule; H & E, haematoxylin and eosin; RSF, raw soya flour; CCK. cholecystokinin.
Williams et al., 1978) and a trophic hormone for the pancreas (Dembinski and Johnson, 1980; Morisset, 1980). Prolonged feeding of RSF (greater than 6mo) results in the development of foci of atypical acinar cells in the pancreas (McGuinness et al., 1982; McGuinness et al.. 1984; McGuinness and Wormsley, 1986). The incidence of these foci is greatly increased when rats are injected with the pancreatic carcinogen azaserine, in combination with RSF feeding (McGuinness et al., 1981; Morgan et al., 1977). Unlike the surrounding tissue, these foci continue to grow throughout the life of the animal, forming macroscopic atypical acinar cell nodules (AACN) that eventually replace most of the normal pancreatic structure (Longnecker and Curphy, 1975; Longnecker, 1981). A small proportion of AACN progress to frank carcinoma (Longnecker et al., 1979). AACN induced by azaserine-RSF treatment are morphologically indistinguishable from those induced by RSF alone (Longnecker. 1981)
DALY ET AL.
626
and also progress to carcinoma (Morgan, 1987). Developing AACN often appear as large masses of cells with no regular acinar structure. When adenomas develop, these often have a thick fibrous capsule that would appear to restrict the drainage of ducts from the adenoma. Invasive neoplasms in this model are often cystic and this may indicate that drainage from the carcinoma is defective. These morphological appearances therefore throw some doubt on the functional capability of the neoplastic cells in AACN and any duct system that might drain them. In rats fed RSF for 4 weeks, fasting results in a rapid accumulation of enzymes (Crass et al., 1987), presumably because protein synthesis continues for some time after discharge ceases. Therefore, by studying the effect of fasting on the enzyme content of nodular tissue, insight would be gained into the functional state of this tissue. Furthermore, by studying the secretory response of the multinodular pancreas to exogenous CCK it would be possible to determine whether nodules possess a functional duct system and if so, how responsive this tissue is to CCK.
from 2-8 mm in diameter were easily separated from normal tissue using forceps. For each pancreas, tissue was taken from 3 to 5 nodules and pooled to form a single sample of 50-100 mg. Only a portion of the large nodules was included in each sample whilst smaller nodules were included whole. 100 mg of tissue which was free of visible nodules was also taken from each pancreas for assay of internodular tissue. All tissue was stored separately at -80°C until being assayed, within a week of collection. For each pancreas, samples of both nodular and internodular tissue were removed, fixed in Carnoy’s fixative and processed for paraffin embedding. Six pm sections were then cut and stained with haematoxylin and eosin (H & E). For higher resolution light microscopy, approximately 1 mm cubes of nodular and internodular tissue were immersion fixed in 2.5% glutaraldehyde in O-1 M Cacodylate buffer pH7.4. The tissue was then post-fixed in 1% Osmium tetroxide and processed for embedding in Aralditem. One pm sections were cut and stained with toluidine blue. Measurement of amylase and trypsin activity.
Materials and Methods Study 1: pancreatic response to fasting Animals. Eight male Wistar derived rats were
purchased from the Animal Resources Centre (Murdoch, W.A.). The rats were fed a diet of raw soya flour (Crass and Morgan, 1982) for 61 weeks, beginning at 10 weeks of age. Four rats were killed without fasting to provide pancreatic enzyme (protein, amylase and trypsin) and nucleic acid (DNA and RNA) values for the ‘fed state’. The remaining 4 rats were fasted for 48 hr prior to sacrifice to provide data for the ‘fasted state’. Histology and tissuepreparation. The animals were deeply anaesthetised with ether, the abdomen opened in the midline and the incision extended into the thorax. The heart was then cut. The pancreas was identified, removed in toto, trimmed free of fat and spread out on a weighing tray for the removal of nodules. Macroscopic nodules were observed in all rats fed RSF for 61 weeks. The nodules were readily identified as pale masses usually extending above the curvature of normal tissue. The nodules, which ranged
At the time of assay the pancreatic tissue was thawed, reweighed and homogenised for 30 set in 5 ml of chilled distilled water using an Ultraturrax homogeniser on a setting of 22,OOOrpm (Model T25 Janke and Kunkel IKA@-Labortechnik). The homogenate was then sonicated for 20 set on a continuous setting of 3 (Branson Sonifier, Model B30; Branson Sonic Power Co., Dannbury, Conn). 500 microlitres of the sonicate was diluted with 4.5 ml of distilled water, while the remaining solution was precipitated with an equal volume of 0.8 N perchloric acid and left on ice for 15 min awaiting assay for nucleic acid determination. Amylase activity (E.C. 3.2.1.1) was determined by the method of Dahlqvist (1962) using 50, 100 and 200 ~1 of diluted sonicate made up to 1 ml with distilled water. In this method, one unit of amylase activity (U) is defined as the amount of amylase that liberates 1 pmole of maltose from starch per min. Trypsin was assayed after activating trypsinogen to trypsin as described by Crass et al. (1987). The resulting trypsin activity was measured by the technique of Preiser et al.
SECRETORY
FUNCTION IN NEOPLASTIC
627
PANCREAS
(1975) using benzoyl-DL-arginine-p-nitroanilide (BAPNA) as substrate. One unit of trypsin activity (U) is defined as the amount of trypsin that liberates 1 pmole of p-nitroanilide from BAPNA per min. Measurement of nucleic acid and protein content. DNA, RNA and protein were extracted
as described previously (Oates and Morgan, 1982a). Protein content of the tissue was measured using the method of Schacterle and Pollack (1973), and DNA using the method of Burton (1956). Study 2: Pancreatic secretory response to exogenous CCK Animals.
Three rats were injected with a single dose of azaserine (30 mg/kg i.p.) at 19 days of age and then fed RSF for 71 weeks. Three rats injected with saline and fed chow for 71 weeks acted as the controls. Collection of pancreatic secretions. After a 24 hr fast, the rats were anaesthetised with urethane (1.25 g/kg i.p.), made up as a 25% solution in normal saline. 75% of the dose was given initially with the remainder given 30 min later. After the onset of anaesthesia, the external jugular vein was cannulated with S.P. 31 polyethylene tubing (Dural Plastics, i.d. 0.5 mm, o.d. 0.8 mm). The abdomen was then opened and the bile duct and pylorus were ligated. The pancreatic duct at the ampulla of Vater was identified and cannulated with S.P. 10 polyethylene tubing (id. 0.28mm, o.d. 0.61 mm). After the cannula had cleared of bile, a pre-weighed Eppendorf tube was attached to the tubing and basal secretions were collected over a 30-min period. The gland was then stimulated at 30min intervals with bolus injections of CCK (Boots, Nottingham, England) ranging from 0.625 to 10.00 Crick, Harper, Raper Units
(CHRU). The stimulation of protein output and flow was complete within 30 min of injection. Pancreatic secretions were collected for 30 min following each dose of CCK and the volume and protein output measured. For details of this procedure see Oates and Morgan (1982). Due to the copious flow of pancreatic secretions in azaserine-RSF treated rats (4-7 ml/3 hr), a volume of isotonic saline equivalent to the volume secreted, was injected intravenously at each 30-min interval in order to maintain blood volume. Statistics. Results from Study 1 were evalu-
ated by Student’s t-test (Systat 5.0, Systat Inc, Evanston, IL, USA). The secretory response to CCK of azaserine-RSF treated rats and controls were compared statistically by a 2 way analysis of variance using Systat 5.0. Significance was considered to be p < 0.05. Results are expressed as mean 2 standard error of the mean. Results Pancreatic Weight. The mean body weight
and pancreatic weight of each group of rats is shown in Table 1. Study I: pancreatic response to fasting Histology. The size and distribution
of nodules varied within and between pancreases of the eight animals fed RSF for 61 weeks. The smallest nodule that could be removed free of surrounding tissue was about 2 mm in diameter, while the largest was about 8 mm in diameter. In both the fed and fasted conditions, the most obvious feature of nodules, compared to surrounding tissue was the presence of increased nuclear staining. The nodule could be clearly distinguished from adjacent cells by a defined interstitial septum and in many
Table 1. Body and pancreatic weight (g) in rats fed RSF for 61 weeks, rats fed chow for 71 weeks and rats injected with a single dose of azaserine (30 mg/kg i.p.) at 19 days of age and then fed RSF for 71 weeks. Results = mean 2 SEM. n = number or rats.
Treatment
n
Body Weight
RSF (61 weeks-Fed) RSF (61 weeks-Fasted) Chow (71 weeks) RSF (71 weeks + Azaserine)
4 4 3 3
398 384 423 384
? ? 2 2
10 38 61 48
(g)
Pancreatic 2.22 2.55 1.21 12.53
Weight t t 2 t
0.15 0.20 0.16 1.89
(g)
DALY ET AL.
Figs 1 and 2. Normal tissue (Fig, 1) and nodular tissue (Fig. 2) in the fed (Figs la, 2a) and fasted state (Figs lb, 2b). Zymogen granule density of both tissues is higher in the fasted state. One pm Araldite sections stained with toluidine blue. x250.
cases the nodules extended beyond the curvature of the surrounding pancreatic tissue. No nodules or atypical foci were detected histologically in random samples of internodular pancreas. Some of the nodules used for assay of enzyme content did meet the suggested size criteria for classification as adenomas (>3mm diam) (Longnecker, 1983). However, the presence of a thickened connective tissue capsule was not seen in any of the nodules examined. In the fasted state, both nodular and internodular pancreas exhibited significantly greater zymogen granule density than in their corresponding fed
condition (Compare Figs la and 2a with Figs lb and 2b). Nucleic acid and protein content. To allow
meaningful comparison between normal and nodular tissue, data were expressed per 100 mg tissue. Fed State: Both DNA content (Fig. 3) and trypsin content (Fig. 4) were significantly elevated in nodules compared to the adjacent internodular tissue. Nodules, however, did not differ significantly from internodular tissue with
SECRETORY
FUNCTION IN NEOPLASTIC
INTERNODULAR
Fig. 3. Pancreatic DNA were fasted for 48 hr prior killing 0. Nodules and Mean 5 SEM of 4 rats in
PANCREAS
NODULAR
INTERNODULAR
NOD”L*R
(A) and RNA (B) in rats fed RSF for 61 weeks. In one group. rat\ to death Ha, while in the other group rats were fed until the time of internodular tissue were obtained from the same pancrcascs each group.
Fig. 4. Protein (a), amylase (b) and trypsin (c) in internodular and nodular pancreatic tissue from rats fed RSF for 61 weeks. In one group. rats were fasted for 48 hr prior to death 9. while in the other group rats were fed until the time of killing 0. Mean 2 SEM of 4 rats in each group
respect to RNA, protein or amylase content (Figs 3, 4). Fasted State: In the internodular tissue, fasting produced a significant decrease in RNA content (Fig. 3) and increases in protein, amylase and trypsin content (Fig. 4). Nodular tissue responded similarly, showing a significant decrease in RNA, and increases in protein, amylase and trypsin. However, the decrease in RNA was more pronounced in nodular tissue. Fasting did not affect DNA levels in either tissue (Fig. 3). Study 2: pancreatic secretory response to exogenous CCK Histology. No nodules or atypical foci were
detected in control rats. Rats treated with azaserine-RSF had grossly enlarged pancreases containing hundreds of macroscopic nodules and adenomas ranging from 210mm in diameter. It was estimated that less than 10% of each pancreas was composed of normal acinar structure. Some ‘Car-
cinemas in situ’ as described by Longnecker (1983) were detected histologically, but these constituted only a small proportion of pancreatic tissue. Basal and stimulated pancreatic secretions. In azaserine-RSF treated rats, there was a 50fold increase in basal volume output (Fig. 5a) and a 30-fold increase in basal protein output (Fig. 5b) compared to controls and this difference was highly significant. When the data were corrected for the increased pancreatic weight of the azaserine-RSF treated rats, the basal volume output was 5fold higher (Fig. SC) and basal protein output was 3-fold higher (Fig. 5d) than controls. In both controls and rats treated with azaserine-RSF, stimulation with CCK from 0.625 to 10 CHRU, resulted in increased volume (Fig. 5a) and protein output (Fig. 5b) that was linearly related to the log of the dose of CCK. For each dose of CCK, the volume and protein output from the nodular pancreas was significantly greater than that of
630
DALY ET AL.
”
m 06
1
9
10
CCK (CHRU)
Fig. 5. Pancreatic outputmeasured 1s volume (a), mg protein (b), volume/g (c) and mg protein/g (d) in azaserine-RSF treated rats -W and controls -II-. Mean + SEM of 3 rats in each group.
controls. When this data was corrected for the much greater weight of the multinodular pancreas, the difference between the 2 groups was reduced. However, 2 way analysis of variance revealed that the dose response curve of the multinodular pancreas (dose versus volume/g or dose versus mg/g) had an elevation which was significantly higher than that of controls. Thus both the volume and protein output per g tissue were significantly higher in the multinodular pancreas. The slope of these dose response curves, however, did not differ significantly, suggesting that the higher output of the multinodular pancreas was due to differences in basal output only. Discussion In this study, nodules were produced either by RSF alone or by giving a single injection of azaserine, followed by RSF feeding. The resulting nodules were then studied to determine whether there were any functional differences between this tissue and the adjacent
internodular tissue concerning the response to fasting and exogenous CCK. In the first study, an important consideration was the ability to harvest nodules devoid of surrounding tissue. From histological examination of the pancreas, the nodules were seen to be separated from the adjacent tissue by a defined interstitial septum. Although the three dimensional structure of nodules and pancreatic adenoma has not been studied, presumably they are attached to adjacent tissue by a stalk containing blood vessels and interlobular ducts. This attachment, however, is weak and the nodules can easily be removed by blunt dissection with forceps (Morgan et al., 1986). The nodules showed increased nuclear staining per field compared to the adjacent tissue and the DNA content of nodules was significantly elevated in both the fed and fasted states. In view of these findings, it was concluded that the technique for removal of nodules from surrounding tissue was valid and allowed adequate partitioning for biochemical study.
SECRETORY
FUNCTION
IN NEOPLASTIC
PANCREAS
It was previously shown that RSF or trypsin inhibitor stimulate protein synthesis, probably because of increased enzyme discharge (Konijin er al., 1970; Rausch et al., 1987). This effect is most likely due to increased release of endogenous CCK. With fasting, the drive on protein synthesis and discharge would be lost and this is reflected in the present study by a reduction in tissue RNA content. However, it appears that protein synthesis persists for some time after discharge ceases, since protein and enzymes accumulate. This was shown by Crass et al. (1987) using rats that had been fed RSF for 4 weeks before fasting. In the present study with rats fed RSF for 61 weeks we also report that fasting leads to an accumulation of enzymes. Importantly, this accumulation of enzymes was similar in both nodular and internodular tissue, suggesting that nodular pancreas is as sensitive to the drive on protein synthesis and discharge as is the adjacent, relatively normal tissue. In the fed state when plasma CCK levels would be highest, nodules contained as much amylase and more trypsin than internodular tissue. This is in contrast to acinar cell carcinoma in the rat which contains reduced amounts of trypsin and amylase (Herzig et al., 1991). Tissue RNA levels were similar or slightly elevated in nodular tissue and decreased sharply with fasting. This contrasts with the azaserineinduced transplantable acinar cell carcinoma which contains reduced amounts of RNA (Aubin et al., 1987). The results suggest that, unlike carcinomas, nodules synthesize enzymes at a rate that is at least equal to normal. In the second study, protein output from the multinodular pancreas was measured. Unlike RSF alone, which produced only 520 macroscopic nodules per pancreas, azaserine-RSF treatment produced several hundred nodules. After 71 weeks, it was estimated that >90% of the pancreas consisted of nodular tissue (including adenomas). Carcinomas were not visible macroscopically and carcinomas in situ represented only a small fraction of each pancreas. This is in contrast to the study by Herzig et al. (1991) in which carcinomas predominated over normal and nodular tissue and may be due to the differences in diet, the number of azaserine injections and the strain of rat used. Compared to controls, the multinodular
hil
pancreas secreted a significantly greater volume as well as more protein. When expressed per g of pancreas, basal and stimulated secretions were still higher in the multinodular pancreas. Since only small amounts of normal tissue remained in the RSF-azaserine treated rats, it is presumed that the protein output in these rats largely reflects secretion from nodular tissue. Therefore, it can be inferred that these nodules do possess a functional duct system and can secrete into the main pancreatic duct. The secretory response of nodular tissue to exogenous CCK appears to be at least equal to that of normal pancreatic tissue and basal secretion in nodules is significantly higher than controls. This may reflect increased responsiveness of nodular tissue to basal levels of endogenous CCK. It has recently been shown that azaserineinduced nodular acinar tissue in the rat pancreas overexpresses CCK receptors (Bell et al., 1992), and an increased responsiveness to CCK would therefore be expected. This finding also supports the other study reported here on the response of nodules to fasting, since it predicts that the secretory response to CCK of nodular tissue would be at least equal to that of normal tissue. While this manuscript was in preparation the report of Herzig et al. (1991) was published. Herzig et al. (1991) showed an increase in the volume and trypsin content of basal secretion from azaserine-induced pancreatic carcinoma. These authors also reported decreased responsiveness of carcinoma tissue to exogenous CCK. The pathological changes that lead to reduced secretion in response to exogenous CCK must take place subsequent to the development of macroscopic nodules and adenoma since we have demonstrated normal or increased responsiveness to exogenous CCK in these precursor tissues. The differences between the results of the present study and those reported by Herzig et al. (1991) probably reflect differences in the tissue examined, i.e. predominantly nodule tissue versus predominantly carcinoma. A pattern of increased responsiveness during the early stages of neoplasia with reduced responsiveness after the development of frank carcinoma, would fit the concept that one of the effects of azaserine treatment is to increase the expression of CCK receptor on the surface of the affected cells. This
DALY ET AL.
632
would result in an increased responsiveness to CCK that would confer a ‘growth advantage’ on altered foci and AACN. These neoplastic cells would therefore divide and grow faster than the surrounding tissue. During the final stage of r,eoplasia, the cells may become autonomous (perhaps developing receptors for autocrine growth factors). The disorganised growth of such carcinomas may impede drainage through the duct system,
this reducing the secretory response to CCK.
Acknowledgements This work was supported by grants from the Australian Research Council, The Medical Faculty of The University of Western Australia and the Cancer Foundation of Western Australia.
References Aubin, R. .I., Malouin, F., Leduc, Y., Taton, G., Delhaye, M., Larose, L., Longnecker, D. S., Poirier, G. G. and Morisset, J. A. 1987. Characterization of a transplantable acinar cell carcinoma of the rat pancreas: biochemical studies. In Experimental pancreatic carcinogenesis. (eds. D. G. Scarpelli, J. K. Reddy and D. S. Longnecker), pp. 277-288. CRC Press. Boca Raton. Bell, R. H., Kuhlmann, E. T., Jensen, R. T. and Longnecker, D. S. 1992. Overexpressionof cholecystokinin receptors in azaserine-induced neoplasms of the rat pancreas. Cancer Res., 52, 3295-3299. Burton, K. A. 1956. A study of the conditions and mechanism of the diphenylamine reaction for the calorimetric estimation of deoxyribonucleic acid. Biochemistry Journal, 62, 315-323. Crass, R. A. and Morgan, R. G. H. 1982. The effect of long term feeding of soya-bean flour diets on pancreatic growth in the rat. Bn’t. J. Nun., 47, 119-129. Crass, R. A., Oates, P. S. and Morgan, R. G. H. 1987. The effect of fasting on enzyme levels in the enlarged and involuting rat pancreas. &it. J. Nutr., 58,427-436. Dahlqvist, A. 1962. A method for the determination of amylase in intestinal contents. &and. J. Clin. Lab. Inuest., 14, 145-151. Dembinski, A. B. and Johnson, L. R. 1980. Stimulation of pancreatic growth by secretin, caerulein and pentagastrin. Endocrinology,
106, 323-328.
Folsch, U. R., Mustroph, D., Schafmayer, A., Becker, H. D. and Creutzfeldt, W. 1984. Elevated CCK plasma concentrations during acute and chronic feeding of soybean flour. Digestion, 30, 88. Herzig, K. H., Creutzfeldt, W. and Folsch, U. R. 1991. Secretagogue response of azaserine-induced rat pancreatic acinar tumors in viva. Gastroenterology, 101,220-227. Konijin, A., M., Birk, Y. and Guggenheim, K. 1970. In vitro synthesis of pancreatic enzymes: effects of soybean trypsin inhibitor. Am. J. Physiol., 218, 11131117. Longnecker, D. S. 1981. Carcinoma of the pancreas in azaserine-treated rats. Am. J. Pafhol., 94, 105. Longnecker, D. S. 1983. Early morphologic markers of carcinogenicity in rat pancreas. In Applicafion of biological markers to carcinogen testing. (eds. H. A. Milman and S. Sell), pp. 4360. Plenum Press, New York. Longnecker, D. S. and Curphy, T. J. 1935. Adenocarcinoma of the pancreas in azaserine-treated rats. Cancer Ref., 35, 2249-2251. Longnecker, D. S., Lilja, H. S., French, J. I., Kuhlmann, E. and Nell, W. 1979. Transplantation of azaserine-induced carcinomas of pancreas in rats. Cancer Len., 7, 197-202. Lu, L., Louie, D. and Owyang, C. 1989. A Cholecystokinin Releasing Peptide Mediates Feedback Regulation of Pancreatic Secretion. Am. J. Physiol., 256 G43O-G435. Lyman, R. L. and Lepkovsky, S. 1957. The effect of raw soybean meal and trypsin inhibitor diets on pancreatic enzyme secretion in the rat. J. N&r., 62, 269-284. McGuinness, E. E., Hopwood, D. and Wormsley, K. G. 1982. Further studies of the effects of raw soya flour on the rat pancreas. Stand. J. Gastroenterol., 17, 273-277. McGuinness, E. E., Morgan, R. G. H., Levison, D. A., Hopwood, D. and Wormsley, K. G. 1981. Interaction of azaserine and raw soya flour on the rat pancreas. Stand. J. Gastroenterol., 16, 49-56. McGuinness, E. E., Morgan, R. G. H. and Wormsley, K. G. 1984. Effects of soybean flour on the pancreas of rats. Environ. Health Persp., 56, 205212.
McGuinness, E. E. and Wormsley, K. G. 1986. Effects of feeding partial and intermittant raw soya flour diets on the rat pancreas. Cancer Lett., 32, 73-81. Morgan, R. G. H. 1987. Raw soya flour and pancreatic cancer in experimental animals. In Experimental Pancreatic Carcinogenesb. (eds. D. G. Scarpelli, J. K. Reddy and D. S. Longnecker), pp. 159-174. CRC Press, Boca Raton. Morgan, R. G. H., Levinson, D. A., Hopwood, D., Saunders, J. H. B. and Wormsley, K. G. 1977. Potentiation of the action of azaserine on the rat pancreas by raw soya bean flour. Cancer Len., 3, 87-90.
SECRETORY
FUNCTION IN NEOPLASTIC
PANCREAS
633
Morgan, R. Cl. H., Schaeffer, B. K. and Longnecker, D. S. 1986. Size and number of nuclei differ in normal and neoplastic acinar cells from rat pancreas. Pancreas, 1, 37-43. Morisset, J. 1980. Stimulation of pancreatic growth by secretin and caerulein in suckling rats, Eiomed. Res., 10, I I22. Oates, P. S. and Morgan, R. G. H. 1982a. Pancreatic growth and cell turnover in the raw soya flour fed rat. Am. J. Pathol., 108, 217-224. Oates, P. S. and Morgan, R. G. H. 1982b. Pancreatic response to bolus injection of cholecystokinin-pancreozynun in anaesthetized and conscious rats fed raw soya flour. Aust. 1. Biol. Sci., 35, 505-515. Oates. P. S. and Morgan, R. G. H. 1984. Short term effects of feeding raw soya flour on pancreatic cell turnover in the rat. Am. J. Physiol., 241, G667-G673. P&kin, S. R., Rottman, A. J., Batzri, S. and Gardner, J. D. 1978. Kinetics of amylase release by dispersed acini prepared from guinea pig pancreas. Am. J. Physiol., 235, E743-E749. Preiser, H., Schmida, J., Maestracci, D. and Crane, R. K. 1975. Modification of an assay for trypsin and its application for the estimation of enteropeptidase. Clin. Chim. Acta., 59, 169175. Rausch, U., Weidenbach, H., Adler, G. and Kern, H. F. 1987. Stimulation of pancreatic secretory process in the rat by low-molecular weight proteinase inhibitor. II. Regulation of total protein and individual enzyme biosynthesis Cell Tissue Res., 249, 63-67. Schacterle, G. and Pollack, R. 1973. A simplified method for the quantitative assay of small amounts of protein in biologic material. Anal. Biochem., 51,6.54-655. Williams, J. A.. Kork, M. and Dormer, R. L. 1978. Action of secretagogues on a new preparation of functionall!. intact. isolated pancreatic acini. Am. J. fhysiol., 235, E517-E524.
JOHN M. DALY*, PHILLIP S. OATES and REGINALD G.H. MORGAN
THE EFFECT OF FASTING AND CHOLECYSTOKININ STIMULATION ON NORMAL AND NEOPLASTIC TISSUE IN THE RAT PANCREAS Keywords: Atypical acinar cell nodule, secretion, cholecystokinin.
enzymes. azaserine, neoplastic function.
ABSTRACT. Pancreatic nodules were produced in rats by either feeding raw soya flour alone or by injection of azaserine plus raw soya flour feeding. The resulting nodules were studied to determine whether there was any functional difference between this tissue and the relatively normal internodular pancreas. Tissue DNA and trypsin content were significantly elevated in nodules compared to the adjacent tissue. With fasting, protein and enzyme content increased significantly and equally in both nodular and internodular tissues. RNA levels fell significantly and the decrease was more pronounced in nodular tissue. The responsiveness of the multinodular pancreas to cholecystokinin was examined by measuring pancreatic secretion basally and in response to cholecystokinin. Both the volume and protein content secreted by the multinodular pancreas were greatlv elevated above control levels. When corrected for pancreatic weight, the difference remained significant and appeared to be due to increased basal secretion by the nodular pancreas. These studies demonstrate that azaserine-raw soya flour induced nodules are functionally efficient. Furthermore, the secretory response to cholecystokinin of these nodules is equal to or higher than that of normal tissue.
Introduction
Feeding raw soya flour (RSF) continuously to rats results initially in hypersecretion (Lyman and Lepkovsky, 1957; Oates and Morgan, 1982b), followed by hypertrophy (Oates and Morgan, 1984) and then hyperplasia (Oates and Morgan, 1982b; Oates and Morgan, 1984) of the pancreas. This effect is largely due to the presence within RSF of trypsin inhibitors that are known to raise plasma cholecystokinin (CCK) levels (Falsch et al., 1984), probably by preventing trypsin from inactivating a peptide which stimulates CCK release (Lu er al., 1989). CCK is a powerful pancreatic secretagogue (Peikin et al., 1978;
Department of Physiology, University of Western Australia, Nedlands 6009. W.A., Australia. *To whom correspondence should be addressed. Received 23 April 1992. *Abbreviations: AACN, atypical acinar cell nodule; H & E, haematoxylin and eosin; RSF, raw soya flour; CCK. cholecystokinin.
Williams et al., 1978) and a trophic hormone for the pancreas (Dembinski and Johnson, 1980; Morisset, 1980). Prolonged feeding of RSF (greater than 6mo) results in the development of foci of atypical acinar cells in the pancreas (McGuinness et al., 1982; McGuinness et al.. 1984; McGuinness and Wormsley, 1986). The incidence of these foci is greatly increased when rats are injected with the pancreatic carcinogen azaserine, in combination with RSF feeding (McGuinness et al., 1981; Morgan et al., 1977). Unlike the surrounding tissue, these foci continue to grow throughout the life of the animal, forming macroscopic atypical acinar cell nodules (AACN) that eventually replace most of the normal pancreatic structure (Longnecker and Curphy, 1975; Longnecker, 1981). A small proportion of AACN progress to frank carcinoma (Longnecker et al., 1979). AACN induced by azaserine-RSF treatment are morphologically indistinguishable from those induced by RSF alone (Longnecker. 1981)
DALY ET AL.
626
and also progress to carcinoma (Morgan, 1987). Developing AACN often appear as large masses of cells with no regular acinar structure. When adenomas develop, these often have a thick fibrous capsule that would appear to restrict the drainage of ducts from the adenoma. Invasive neoplasms in this model are often cystic and this may indicate that drainage from the carcinoma is defective. These morphological appearances therefore throw some doubt on the functional capability of the neoplastic cells in AACN and any duct system that might drain them. In rats fed RSF for 4 weeks, fasting results in a rapid accumulation of enzymes (Crass et al., 1987), presumably because protein synthesis continues for some time after discharge ceases. Therefore, by studying the effect of fasting on the enzyme content of nodular tissue, insight would be gained into the functional state of this tissue. Furthermore, by studying the secretory response of the multinodular pancreas to exogenous CCK it would be possible to determine whether nodules possess a functional duct system and if so, how responsive this tissue is to CCK.
from 2-8 mm in diameter were easily separated from normal tissue using forceps. For each pancreas, tissue was taken from 3 to 5 nodules and pooled to form a single sample of 50-100 mg. Only a portion of the large nodules was included in each sample whilst smaller nodules were included whole. 100 mg of tissue which was free of visible nodules was also taken from each pancreas for assay of internodular tissue. All tissue was stored separately at -80°C until being assayed, within a week of collection. For each pancreas, samples of both nodular and internodular tissue were removed, fixed in Carnoy’s fixative and processed for paraffin embedding. Six pm sections were then cut and stained with haematoxylin and eosin (H & E). For higher resolution light microscopy, approximately 1 mm cubes of nodular and internodular tissue were immersion fixed in 2.5% glutaraldehyde in O-1 M Cacodylate buffer pH7.4. The tissue was then post-fixed in 1% Osmium tetroxide and processed for embedding in Aralditem. One pm sections were cut and stained with toluidine blue. Measurement of amylase and trypsin activity.
Materials and Methods Study 1: pancreatic response to fasting Animals. Eight male Wistar derived rats were
purchased from the Animal Resources Centre (Murdoch, W.A.). The rats were fed a diet of raw soya flour (Crass and Morgan, 1982) for 61 weeks, beginning at 10 weeks of age. Four rats were killed without fasting to provide pancreatic enzyme (protein, amylase and trypsin) and nucleic acid (DNA and RNA) values for the ‘fed state’. The remaining 4 rats were fasted for 48 hr prior to sacrifice to provide data for the ‘fasted state’. Histology and tissuepreparation. The animals were deeply anaesthetised with ether, the abdomen opened in the midline and the incision extended into the thorax. The heart was then cut. The pancreas was identified, removed in toto, trimmed free of fat and spread out on a weighing tray for the removal of nodules. Macroscopic nodules were observed in all rats fed RSF for 61 weeks. The nodules were readily identified as pale masses usually extending above the curvature of normal tissue. The nodules, which ranged
At the time of assay the pancreatic tissue was thawed, reweighed and homogenised for 30 set in 5 ml of chilled distilled water using an Ultraturrax homogeniser on a setting of 22,OOOrpm (Model T25 Janke and Kunkel IKA@-Labortechnik). The homogenate was then sonicated for 20 set on a continuous setting of 3 (Branson Sonifier, Model B30; Branson Sonic Power Co., Dannbury, Conn). 500 microlitres of the sonicate was diluted with 4.5 ml of distilled water, while the remaining solution was precipitated with an equal volume of 0.8 N perchloric acid and left on ice for 15 min awaiting assay for nucleic acid determination. Amylase activity (E.C. 3.2.1.1) was determined by the method of Dahlqvist (1962) using 50, 100 and 200 ~1 of diluted sonicate made up to 1 ml with distilled water. In this method, one unit of amylase activity (U) is defined as the amount of amylase that liberates 1 pmole of maltose from starch per min. Trypsin was assayed after activating trypsinogen to trypsin as described by Crass et al. (1987). The resulting trypsin activity was measured by the technique of Preiser et al.
SECRETORY
FUNCTION IN NEOPLASTIC
627
PANCREAS
(1975) using benzoyl-DL-arginine-p-nitroanilide (BAPNA) as substrate. One unit of trypsin activity (U) is defined as the amount of trypsin that liberates 1 pmole of p-nitroanilide from BAPNA per min. Measurement of nucleic acid and protein content. DNA, RNA and protein were extracted
as described previously (Oates and Morgan, 1982a). Protein content of the tissue was measured using the method of Schacterle and Pollack (1973), and DNA using the method of Burton (1956). Study 2: Pancreatic secretory response to exogenous CCK Animals.
Three rats were injected with a single dose of azaserine (30 mg/kg i.p.) at 19 days of age and then fed RSF for 71 weeks. Three rats injected with saline and fed chow for 71 weeks acted as the controls. Collection of pancreatic secretions. After a 24 hr fast, the rats were anaesthetised with urethane (1.25 g/kg i.p.), made up as a 25% solution in normal saline. 75% of the dose was given initially with the remainder given 30 min later. After the onset of anaesthesia, the external jugular vein was cannulated with S.P. 31 polyethylene tubing (Dural Plastics, i.d. 0.5 mm, o.d. 0.8 mm). The abdomen was then opened and the bile duct and pylorus were ligated. The pancreatic duct at the ampulla of Vater was identified and cannulated with S.P. 10 polyethylene tubing (id. 0.28mm, o.d. 0.61 mm). After the cannula had cleared of bile, a pre-weighed Eppendorf tube was attached to the tubing and basal secretions were collected over a 30-min period. The gland was then stimulated at 30min intervals with bolus injections of CCK (Boots, Nottingham, England) ranging from 0.625 to 10.00 Crick, Harper, Raper Units
(CHRU). The stimulation of protein output and flow was complete within 30 min of injection. Pancreatic secretions were collected for 30 min following each dose of CCK and the volume and protein output measured. For details of this procedure see Oates and Morgan (1982). Due to the copious flow of pancreatic secretions in azaserine-RSF treated rats (4-7 ml/3 hr), a volume of isotonic saline equivalent to the volume secreted, was injected intravenously at each 30-min interval in order to maintain blood volume. Statistics. Results from Study 1 were evalu-
ated by Student’s t-test (Systat 5.0, Systat Inc, Evanston, IL, USA). The secretory response to CCK of azaserine-RSF treated rats and controls were compared statistically by a 2 way analysis of variance using Systat 5.0. Significance was considered to be p < 0.05. Results are expressed as mean 2 standard error of the mean. Results Pancreatic Weight. The mean body weight
and pancreatic weight of each group of rats is shown in Table 1. Study I: pancreatic response to fasting Histology. The size and distribution
of nodules varied within and between pancreases of the eight animals fed RSF for 61 weeks. The smallest nodule that could be removed free of surrounding tissue was about 2 mm in diameter, while the largest was about 8 mm in diameter. In both the fed and fasted conditions, the most obvious feature of nodules, compared to surrounding tissue was the presence of increased nuclear staining. The nodule could be clearly distinguished from adjacent cells by a defined interstitial septum and in many
Table 1. Body and pancreatic weight (g) in rats fed RSF for 61 weeks, rats fed chow for 71 weeks and rats injected with a single dose of azaserine (30 mg/kg i.p.) at 19 days of age and then fed RSF for 71 weeks. Results = mean 2 SEM. n = number or rats.
Treatment
n
Body Weight
RSF (61 weeks-Fed) RSF (61 weeks-Fasted) Chow (71 weeks) RSF (71 weeks + Azaserine)
4 4 3 3
398 384 423 384
? ? 2 2
10 38 61 48
(g)
Pancreatic 2.22 2.55 1.21 12.53
Weight t t 2 t
0.15 0.20 0.16 1.89
(g)
DALY ET AL.
Figs 1 and 2. Normal tissue (Fig, 1) and nodular tissue (Fig. 2) in the fed (Figs la, 2a) and fasted state (Figs lb, 2b). Zymogen granule density of both tissues is higher in the fasted state. One pm Araldite sections stained with toluidine blue. x250.
cases the nodules extended beyond the curvature of the surrounding pancreatic tissue. No nodules or atypical foci were detected histologically in random samples of internodular pancreas. Some of the nodules used for assay of enzyme content did meet the suggested size criteria for classification as adenomas (>3mm diam) (Longnecker, 1983). However, the presence of a thickened connective tissue capsule was not seen in any of the nodules examined. In the fasted state, both nodular and internodular pancreas exhibited significantly greater zymogen granule density than in their corresponding fed
condition (Compare Figs la and 2a with Figs lb and 2b). Nucleic acid and protein content. To allow
meaningful comparison between normal and nodular tissue, data were expressed per 100 mg tissue. Fed State: Both DNA content (Fig. 3) and trypsin content (Fig. 4) were significantly elevated in nodules compared to the adjacent internodular tissue. Nodules, however, did not differ significantly from internodular tissue with
SECRETORY
FUNCTION IN NEOPLASTIC
INTERNODULAR
Fig. 3. Pancreatic DNA were fasted for 48 hr prior killing 0. Nodules and Mean 5 SEM of 4 rats in
PANCREAS
NODULAR
INTERNODULAR
NOD”L*R
(A) and RNA (B) in rats fed RSF for 61 weeks. In one group. rat\ to death Ha, while in the other group rats were fed until the time of internodular tissue were obtained from the same pancrcascs each group.
Fig. 4. Protein (a), amylase (b) and trypsin (c) in internodular and nodular pancreatic tissue from rats fed RSF for 61 weeks. In one group. rats were fasted for 48 hr prior to death 9. while in the other group rats were fed until the time of killing 0. Mean 2 SEM of 4 rats in each group
respect to RNA, protein or amylase content (Figs 3, 4). Fasted State: In the internodular tissue, fasting produced a significant decrease in RNA content (Fig. 3) and increases in protein, amylase and trypsin content (Fig. 4). Nodular tissue responded similarly, showing a significant decrease in RNA, and increases in protein, amylase and trypsin. However, the decrease in RNA was more pronounced in nodular tissue. Fasting did not affect DNA levels in either tissue (Fig. 3). Study 2: pancreatic secretory response to exogenous CCK Histology. No nodules or atypical foci were
detected in control rats. Rats treated with azaserine-RSF had grossly enlarged pancreases containing hundreds of macroscopic nodules and adenomas ranging from 210mm in diameter. It was estimated that less than 10% of each pancreas was composed of normal acinar structure. Some ‘Car-
cinemas in situ’ as described by Longnecker (1983) were detected histologically, but these constituted only a small proportion of pancreatic tissue. Basal and stimulated pancreatic secretions. In azaserine-RSF treated rats, there was a 50fold increase in basal volume output (Fig. 5a) and a 30-fold increase in basal protein output (Fig. 5b) compared to controls and this difference was highly significant. When the data were corrected for the increased pancreatic weight of the azaserine-RSF treated rats, the basal volume output was 5fold higher (Fig. SC) and basal protein output was 3-fold higher (Fig. 5d) than controls. In both controls and rats treated with azaserine-RSF, stimulation with CCK from 0.625 to 10 CHRU, resulted in increased volume (Fig. 5a) and protein output (Fig. 5b) that was linearly related to the log of the dose of CCK. For each dose of CCK, the volume and protein output from the nodular pancreas was significantly greater than that of
630
DALY ET AL.
”
m 06
1
9
10
CCK (CHRU)
Fig. 5. Pancreatic outputmeasured 1s volume (a), mg protein (b), volume/g (c) and mg protein/g (d) in azaserine-RSF treated rats -W and controls -II-. Mean + SEM of 3 rats in each group.
controls. When this data was corrected for the much greater weight of the multinodular pancreas, the difference between the 2 groups was reduced. However, 2 way analysis of variance revealed that the dose response curve of the multinodular pancreas (dose versus volume/g or dose versus mg/g) had an elevation which was significantly higher than that of controls. Thus both the volume and protein output per g tissue were significantly higher in the multinodular pancreas. The slope of these dose response curves, however, did not differ significantly, suggesting that the higher output of the multinodular pancreas was due to differences in basal output only. Discussion In this study, nodules were produced either by RSF alone or by giving a single injection of azaserine, followed by RSF feeding. The resulting nodules were then studied to determine whether there were any functional differences between this tissue and the adjacent
internodular tissue concerning the response to fasting and exogenous CCK. In the first study, an important consideration was the ability to harvest nodules devoid of surrounding tissue. From histological examination of the pancreas, the nodules were seen to be separated from the adjacent tissue by a defined interstitial septum. Although the three dimensional structure of nodules and pancreatic adenoma has not been studied, presumably they are attached to adjacent tissue by a stalk containing blood vessels and interlobular ducts. This attachment, however, is weak and the nodules can easily be removed by blunt dissection with forceps (Morgan et al., 1986). The nodules showed increased nuclear staining per field compared to the adjacent tissue and the DNA content of nodules was significantly elevated in both the fed and fasted states. In view of these findings, it was concluded that the technique for removal of nodules from surrounding tissue was valid and allowed adequate partitioning for biochemical study.
SECRETORY
FUNCTION
IN NEOPLASTIC
PANCREAS
It was previously shown that RSF or trypsin inhibitor stimulate protein synthesis, probably because of increased enzyme discharge (Konijin er al., 1970; Rausch et al., 1987). This effect is most likely due to increased release of endogenous CCK. With fasting, the drive on protein synthesis and discharge would be lost and this is reflected in the present study by a reduction in tissue RNA content. However, it appears that protein synthesis persists for some time after discharge ceases, since protein and enzymes accumulate. This was shown by Crass et al. (1987) using rats that had been fed RSF for 4 weeks before fasting. In the present study with rats fed RSF for 61 weeks we also report that fasting leads to an accumulation of enzymes. Importantly, this accumulation of enzymes was similar in both nodular and internodular tissue, suggesting that nodular pancreas is as sensitive to the drive on protein synthesis and discharge as is the adjacent, relatively normal tissue. In the fed state when plasma CCK levels would be highest, nodules contained as much amylase and more trypsin than internodular tissue. This is in contrast to acinar cell carcinoma in the rat which contains reduced amounts of trypsin and amylase (Herzig et al., 1991). Tissue RNA levels were similar or slightly elevated in nodular tissue and decreased sharply with fasting. This contrasts with the azaserineinduced transplantable acinar cell carcinoma which contains reduced amounts of RNA (Aubin et al., 1987). The results suggest that, unlike carcinomas, nodules synthesize enzymes at a rate that is at least equal to normal. In the second study, protein output from the multinodular pancreas was measured. Unlike RSF alone, which produced only 520 macroscopic nodules per pancreas, azaserine-RSF treatment produced several hundred nodules. After 71 weeks, it was estimated that >90% of the pancreas consisted of nodular tissue (including adenomas). Carcinomas were not visible macroscopically and carcinomas in situ represented only a small fraction of each pancreas. This is in contrast to the study by Herzig et al. (1991) in which carcinomas predominated over normal and nodular tissue and may be due to the differences in diet, the number of azaserine injections and the strain of rat used. Compared to controls, the multinodular
hil
pancreas secreted a significantly greater volume as well as more protein. When expressed per g of pancreas, basal and stimulated secretions were still higher in the multinodular pancreas. Since only small amounts of normal tissue remained in the RSF-azaserine treated rats, it is presumed that the protein output in these rats largely reflects secretion from nodular tissue. Therefore, it can be inferred that these nodules do possess a functional duct system and can secrete into the main pancreatic duct. The secretory response of nodular tissue to exogenous CCK appears to be at least equal to that of normal pancreatic tissue and basal secretion in nodules is significantly higher than controls. This may reflect increased responsiveness of nodular tissue to basal levels of endogenous CCK. It has recently been shown that azaserineinduced nodular acinar tissue in the rat pancreas overexpresses CCK receptors (Bell et al., 1992), and an increased responsiveness to CCK would therefore be expected. This finding also supports the other study reported here on the response of nodules to fasting, since it predicts that the secretory response to CCK of nodular tissue would be at least equal to that of normal tissue. While this manuscript was in preparation the report of Herzig et al. (1991) was published. Herzig et al. (1991) showed an increase in the volume and trypsin content of basal secretion from azaserine-induced pancreatic carcinoma. These authors also reported decreased responsiveness of carcinoma tissue to exogenous CCK. The pathological changes that lead to reduced secretion in response to exogenous CCK must take place subsequent to the development of macroscopic nodules and adenoma since we have demonstrated normal or increased responsiveness to exogenous CCK in these precursor tissues. The differences between the results of the present study and those reported by Herzig et al. (1991) probably reflect differences in the tissue examined, i.e. predominantly nodule tissue versus predominantly carcinoma. A pattern of increased responsiveness during the early stages of neoplasia with reduced responsiveness after the development of frank carcinoma, would fit the concept that one of the effects of azaserine treatment is to increase the expression of CCK receptor on the surface of the affected cells. This
DALY ET AL.
632
would result in an increased responsiveness to CCK that would confer a ‘growth advantage’ on altered foci and AACN. These neoplastic cells would therefore divide and grow faster than the surrounding tissue. During the final stage of r,eoplasia, the cells may become autonomous (perhaps developing receptors for autocrine growth factors). The disorganised growth of such carcinomas may impede drainage through the duct system,
this reducing the secretory response to CCK.
Acknowledgements This work was supported by grants from the Australian Research Council, The Medical Faculty of The University of Western Australia and the Cancer Foundation of Western Australia.
References Aubin, R. .I., Malouin, F., Leduc, Y., Taton, G., Delhaye, M., Larose, L., Longnecker, D. S., Poirier, G. G. and Morisset, J. A. 1987. Characterization of a transplantable acinar cell carcinoma of the rat pancreas: biochemical studies. In Experimental pancreatic carcinogenesis. (eds. D. G. Scarpelli, J. K. Reddy and D. S. Longnecker), pp. 277-288. CRC Press. Boca Raton. Bell, R. H., Kuhlmann, E. T., Jensen, R. T. and Longnecker, D. S. 1992. Overexpressionof cholecystokinin receptors in azaserine-induced neoplasms of the rat pancreas. Cancer Res., 52, 3295-3299. Burton, K. A. 1956. A study of the conditions and mechanism of the diphenylamine reaction for the calorimetric estimation of deoxyribonucleic acid. Biochemistry Journal, 62, 315-323. Crass, R. A. and Morgan, R. G. H. 1982. The effect of long term feeding of soya-bean flour diets on pancreatic growth in the rat. Bn’t. J. Nun., 47, 119-129. Crass, R. A., Oates, P. S. and Morgan, R. G. H. 1987. The effect of fasting on enzyme levels in the enlarged and involuting rat pancreas. &it. J. Nutr., 58,427-436. Dahlqvist, A. 1962. A method for the determination of amylase in intestinal contents. &and. J. Clin. Lab. Inuest., 14, 145-151. Dembinski, A. B. and Johnson, L. R. 1980. Stimulation of pancreatic growth by secretin, caerulein and pentagastrin. Endocrinology,
106, 323-328.
Folsch, U. R., Mustroph, D., Schafmayer, A., Becker, H. D. and Creutzfeldt, W. 1984. Elevated CCK plasma concentrations during acute and chronic feeding of soybean flour. Digestion, 30, 88. Herzig, K. H., Creutzfeldt, W. and Folsch, U. R. 1991. Secretagogue response of azaserine-induced rat pancreatic acinar tumors in viva. Gastroenterology, 101,220-227. Konijin, A., M., Birk, Y. and Guggenheim, K. 1970. In vitro synthesis of pancreatic enzymes: effects of soybean trypsin inhibitor. Am. J. Physiol., 218, 11131117. Longnecker, D. S. 1981. Carcinoma of the pancreas in azaserine-treated rats. Am. J. Pafhol., 94, 105. Longnecker, D. S. 1983. Early morphologic markers of carcinogenicity in rat pancreas. In Applicafion of biological markers to carcinogen testing. (eds. H. A. Milman and S. Sell), pp. 4360. Plenum Press, New York. Longnecker, D. S. and Curphy, T. J. 1935. Adenocarcinoma of the pancreas in azaserine-treated rats. Cancer Ref., 35, 2249-2251. Longnecker, D. S., Lilja, H. S., French, J. I., Kuhlmann, E. and Nell, W. 1979. Transplantation of azaserine-induced carcinomas of pancreas in rats. Cancer Len., 7, 197-202. Lu, L., Louie, D. and Owyang, C. 1989. A Cholecystokinin Releasing Peptide Mediates Feedback Regulation of Pancreatic Secretion. Am. J. Physiol., 256 G43O-G435. Lyman, R. L. and Lepkovsky, S. 1957. The effect of raw soybean meal and trypsin inhibitor diets on pancreatic enzyme secretion in the rat. J. N&r., 62, 269-284. McGuinness, E. E., Hopwood, D. and Wormsley, K. G. 1982. Further studies of the effects of raw soya flour on the rat pancreas. Stand. J. Gastroenterol., 17, 273-277. McGuinness, E. E., Morgan, R. G. H., Levison, D. A., Hopwood, D. and Wormsley, K. G. 1981. Interaction of azaserine and raw soya flour on the rat pancreas. Stand. J. Gastroenterol., 16, 49-56. McGuinness, E. E., Morgan, R. G. H. and Wormsley, K. G. 1984. Effects of soybean flour on the pancreas of rats. Environ. Health Persp., 56, 205212.
McGuinness, E. E. and Wormsley, K. G. 1986. Effects of feeding partial and intermittant raw soya flour diets on the rat pancreas. Cancer Lett., 32, 73-81. Morgan, R. G. H. 1987. Raw soya flour and pancreatic cancer in experimental animals. In Experimental Pancreatic Carcinogenesb. (eds. D. G. Scarpelli, J. K. Reddy and D. S. Longnecker), pp. 159-174. CRC Press, Boca Raton. Morgan, R. G. H., Levinson, D. A., Hopwood, D., Saunders, J. H. B. and Wormsley, K. G. 1977. Potentiation of the action of azaserine on the rat pancreas by raw soya bean flour. Cancer Len., 3, 87-90.
SECRETORY
FUNCTION IN NEOPLASTIC
PANCREAS
633
Morgan, R. Cl. H., Schaeffer, B. K. and Longnecker, D. S. 1986. Size and number of nuclei differ in normal and neoplastic acinar cells from rat pancreas. Pancreas, 1, 37-43. Morisset, J. 1980. Stimulation of pancreatic growth by secretin and caerulein in suckling rats, Eiomed. Res., 10, I I22. Oates, P. S. and Morgan, R. G. H. 1982a. Pancreatic growth and cell turnover in the raw soya flour fed rat. Am. J. Pathol., 108, 217-224. Oates, P. S. and Morgan, R. G. H. 1982b. Pancreatic response to bolus injection of cholecystokinin-pancreozynun in anaesthetized and conscious rats fed raw soya flour. Aust. 1. Biol. Sci., 35, 505-515. Oates. P. S. and Morgan, R. G. H. 1984. Short term effects of feeding raw soya flour on pancreatic cell turnover in the rat. Am. J. Physiol., 241, G667-G673. P&kin, S. R., Rottman, A. J., Batzri, S. and Gardner, J. D. 1978. Kinetics of amylase release by dispersed acini prepared from guinea pig pancreas. Am. J. Physiol., 235, E743-E749. Preiser, H., Schmida, J., Maestracci, D. and Crane, R. K. 1975. Modification of an assay for trypsin and its application for the estimation of enteropeptidase. Clin. Chim. Acta., 59, 169175. Rausch, U., Weidenbach, H., Adler, G. and Kern, H. F. 1987. Stimulation of pancreatic secretory process in the rat by low-molecular weight proteinase inhibitor. II. Regulation of total protein and individual enzyme biosynthesis Cell Tissue Res., 249, 63-67. Schacterle, G. and Pollack, R. 1973. A simplified method for the quantitative assay of small amounts of protein in biologic material. Anal. Biochem., 51,6.54-655. Williams, J. A.. Kork, M. and Dormer, R. L. 1978. Action of secretagogues on a new preparation of functionall!. intact. isolated pancreatic acini. Am. J. fhysiol., 235, E517-E524.
