Outbred female Syrian golden hamsters 16 weeks old were used. They were housed in plastic cages and kept under standard laboratory conditions (room.
American Journal of Pathology, Vol. 147, No. 5, November 1995 Copyright C) American Societyfor Investigative Pathology
The Role of Pancreatic Islets in Experimental Pancreatic Carcinogenicity
Osamu Ishikawa,* Hiroaki Ohigashi,* Shingi Imaoka,* Ichiro Nakai,t Manabu Mitsuo,t Lamont Weide,* and Parviz M. Pour§ From the Department of Surgery,* The Center for Adult Diseases, Osaka, and Second Department of Sugery,t Kyoto Prefectural University of Medicine, Kyoto, Japan; and Department ofInternal Medicinc4 and Department of Pathology and Microbiology, and the Eppley Institute for Research in Cancer and Allied Diseases,§ University of Nebraska Medical Center, Omaha, Nebraska
Our previous studies have suggested that the presence ofintact islets is essentialfor the induction ofpancreatic exocrine tumors in the Syrian hamster modeL To validate this, we investigated the effect of the carcinogen, N-nitrosobis(2-oxopropyl)amine (BOP) in hamsters, in which homologous isolated intact islets were transplanted into the submandibulargland (SMG). Freshly isolated pure islets from hamster donors were transplanted into the left SMG of 20 female host hamsters. Ten of these hamsters (group 1) received BOP (40 mg/kg) weekly for 3 weeks. Another 10 hamsters (group 2) were kept untreated. In groups 3 and 4 (10 hamsters each) the salt solution or isolated pancreatic ductal ceUs, respectively, was injected into the gland. In other groups (10 hamsters each) islets were transplanted into the peri-SMG connective tissue (group 5) or into the renal subcapsular space (group 6). Hamsters of groups 3 to 6 were treated with BOP as in group 1 (40 mg/kg, weekly for 3 weeks) as were group 7 hamsters, which served as BOP-treated controls. AU BOPtreated hamsters developed pancreatic lesions. Similar hyperplastic and atypical ductal/ductular proliferation and in situ carcinoma were found in the SMG of many group I hamsters. No such lesions werefound in the SMG, peri-SMG, or renal subcapsular space of the other groups. Islets appear to be involved in carcinogenicity of BOP. The mechanism is obscure. (Am J Pathol 1995, 147:1456-1464)
1456
Pancreatic tumors induced in Syrian golden hamsters resemble the human disease morphologically, biologically, antigenetically, and in terms of molecular biology.1`4 In this model, streptozotocin- and alloxan-induced diabetes inhibits pancreatic carcinogenicity.5-a The inhibitory effect of diabetes on tumor induction was restricted to the pancreas but not to other target tissues of the carcinogen N-nitrosobis(2oxopropyl)amine (BOP), indicating that intact islets are important for pancreatic tumor induction. This view was confirmed in a subsequent study, which showed that genetically diabetic hamsters with atrophic islets were resistant to pancreatic carcinogenicity of BOP, whereas the genetically nondiabetic hamsters with normally functioning islets were not.9 A study on homologous whole pancreas transplantation10 could not clarify whether the damaged islets or the diabetic metabolism alterations play a role in pancreatic carcinogenicity of BOP. To clarify this issue, we examined the effects of BOP in hamsters, in which intact homologous intact islets were transplanted in the submandibular glands (SMGs). This gland, which morphologically and physiologically is similar to the pancreas,11 is not a target tissue of BOP. However, in the hamster SMG, BOP causes the same DNA damage (06-methylguanin) as in the pancreas.12'13 Moreover, in both tissues, the DNA repair is much slower than in other hamster tissues.12,13 Consequently, the inability of BOP to induce tumors in the hamster SMG could be due to the lack of islets in this tissue. Our previous experiment showed that SMG is a suitable site for islet transplantation and that islets retain their function for the duration of an 8-week experiment. This study was undertaken to examine whether transplanSupported by a Grant in Aid for Scientific Research on Priority Areas, Cancer-Bioscience, from the Ministry of Education, Science and Culture, Japan (#9-A-01) Accepted for publication July 6, 1995. Address reprint requests to Parviz M. Pour, M.D., The Eppley Institute for Research in Cancer and Allied Diseases, University of Nebraska Medical Center, 600 South 42nd Street, Omaha, NE 68198-6805.
Islet Cells and Exocrine
Carcinogenesis
1457
AJP November 1995, Vol. 147, No. 5
Isolation and Treatment of Pancreatic Ductal Cells Hamster pancreatic ductal cells were isolated as
reported.17
Figure 1. Isolated islet cells prepared for transplantation (dithizone stain, X800). There was no contamination with other cells, such as acinar, ductular, or ductal cells. The size of islet cells varied between 100 and 400 ,um.
tation of intact islets into the SMG leads to local tumor formation.
Materials and Methods Animals Outbred female Syrian golden hamsters 16 weeks old were used. They were housed in plastic cages and kept under standard laboratory conditions (room temperature, 21 ± 3°C; light/dark cycle, 12 hours/12 hours; and relative humidity, 40 ± 5%). They received water and pelleted diet (Oriental Kobo Co. Ltd. Tokyo, Japan) ad libitum. Twenty hamsters served as islet donors.
Islet Transplantation Under pentobarbital anesthesia (75 mg/kg body weight), the papilla of Vater was ligated, and a thin catheter was inserted into the common bile duct. Through this catheter, collagenase (1500 U/ml, Type Xl, Sigma Chemical Co., St. Louis, MO) dissolved in Hanks' balanced salt solution (HBSS) was carefully infused into the pancreas, which was digested in HBSS buffer at 370C for 17 minutes. The islets were collected by the dextran gradient method with a slight modification14 and hand-picked under a dissecting microscope. Using the dithizone staining method (Diphenylthicarbazone, Sigma Chemical Co.),15 we confirmed that there were no other cells (ductular, ductal, or acinar cells) included in the collected specimen (Figure 1). Two hundred islets in 0.2 ml HBSS buffer were prepared for each recipient animal. They were injected into the inferior pole of the left SMG,16 into peri-SMG connective tissue, or under the capsule of the left kidney.
Islets were transplanted into the inferior pole (area) of the left SMG of 20 hamsters. Ten of these hamsters (group 1) received BOP as described below, whereas the other 10 hamsters (group 2) were kept untreated. In 10 hamsters (group 3), 0.2 ml of HBSS buffer without islets, and in 10 hamsters (group 4) freshly isolated pancreatic ductal cells were injected into the inferior pole of the left SMG. In group 5 (10 hamsters), islets were transplanted into the peri-SMG connective tissue of the left SMG, and in group 6 islets were transplanted under the left renal capsule of 10 hamsters. Group 7 (10 hamsters) received BOP only. Two separate groups (5 hamsters each) received digested fragments of homologous thyroid glands into the left SMG and about 1 x 106 immortal pancreatic ductal cells18 into the right SMG (group 8). Group 9 received digested fragments of homologous heart muscles into the left and starch powder in physiological saline into the right SMG. Three days after islet transplantation, all hamsters of groups 1 and 3 to 9 were treated subcutaneously with 40 mg/kg body weight of BOP weekly for 3 weeks. From groups 1 to 7, five hamsters were sacrificed 10 weeks after BOP treatment, and the remaining hamsters, including those in groups 8 and 9, 20 weeks after BOP treatment. Table 1 summarizes the experimental design.
Histological and Immunohistochemical Examination From all hamsters, the pancreas, the right and left SMG and the neighboring soft tissue were removed, fixed immediately in 10% buffered formalin and processed for histology according to conventional methods. In group 5, the left kidney was also removed and treated similarly. From each block, 5 to 10 step sections, each 5 ,tm thick, were cut and either stained with hematoxylin and eosin (H&E) or Grimelius or processed by a multilabeling technique19 with anti-insulin, anti-somatostatin, and antiglucagon antibodies (Dako, Tokyo, Japan), as reported.19 Sections containing induced lesions were also immunostained with an antibody against the blood group A antigen (Dako, Santa Barbara, CA) as described 20.21
Ishikawa et al
1458
AJP November 1995, Vol. 14 7, No. 5
Table 1. Incidence and Type of Pancreatic Lesions Induced in Syrian Hamsters*
Group 1
2 3 4
5 6 7 8 9
Material and site of transplants Islet, SMG Islet, SMG HBBS, SMG Isolated pancreatic ductal cells, SMG Islet, peri-SMG Islet, renal subcapsule
Thyroid, left SMG; TAKA-1 cells, right SMG Heart muscle, left SMG; and starch, right SMG
BOP
A
10 weeks B
C
A
20 weeks B
Yes No Yes Yes
4/5
1/5
0/5
4/4
3/4
0/4 3/5 4/5
0/4 1/5 0/5
0/4 0/5 0/5
0/4 4/5 2/5
0/4 3/5 2/5
1/4 0/4 1/5 1/5
Yes Yes Yes Yes
4/5 5/5 5/5 NE
1/5 1/5 2/5 NE
0/5 0/5 0/5 NE
5/5 5/5 5/5 5/5
4/5 4/5 4/5 4/5
2/5 1/5 2/5 3/5
Yes
NE
NE
NE
4/5
5/5
2/5
C
*A, atypical ductular proliferation; B, ductal/ductular in situ carcinoma; C, invasive ductal/ductular carcinoma. NE, not examined. All hamsters treated with BOP had multiple lesions distributed throughout the entire pancreas but predominantly located around or within the islets. The incidence of each lesion between the BOP-treated hamsters did not differ significantly.
Electron Microscopical Examination Three blocks, in which the presence of neoplastic lesions had been confirmed histologically, were prepared for electron microscopy. In brief, the blocks were dewaxed, fixed in 2.5% glutaraldehyde and then in 2% osmium tetroxide, and embedded in epoxy resin. The ultrathin (700 A) sections were used for uranium-lead stain and observed by electron microscope (JEM-100 CX, Nihon Denshi Co. Ltd, Tokyo, Japan). Histological diagnosis of pancreatic lesions was in accordance with our previously reported terminology.22-24
Results Only one hamster in group 1 and two hamsters in group 2 died during the experiment. Because of advanced autolytic changes, these hamsters were excluded from the study. All BOP-treated hamsters gained less weight than the untreated hamsters. However, no differences were found between the BOP-treated hamsters in the pattern of food intake and weight gain. There were no complications after islet transplantation.
Pancreatic Lesions All BOP-treated hamsters developed pancreatic lesions (Table 1). The incidence of these usually multiple lesions was higher in hamsters examined after 20 weeks than in those studied 10 weeks after BOP. Invasive carcinomas were found only after 20 weeks.
The incidence of each type of lesion did not differ significantly between the groups. As in our previous nue ductular lesions were studies, 22 24 most induced around or within the islets, which were partially or mostly destroyed by atypical or malignant ductular cells (Figure 2). In many lesions, the remains of islet cells could be detected immunohistochemically. As in our previous studies,21,21 all preneoplastic and neoplastic pancreatic lesions were immunoreactive with anti-A antibody.
SMG Lesions Four out of five group 1 hamsters, which were examined at week 10, and all of those examined at week 20, had lesions in the left SMG (Table 2, Figure 3). These lesions were solitary in six hamsters and multiple (two or three) in two hamsters. They consisted of abnormal and atypical ductular structures, which were morphologically identical to those induced in the pancreas. Consequently, they were diagnosed as atypical ductular proliferation and ductular in situ carcinoma. Contrary to the normal SMG ducts, many cells of these lesions contained PAS-positive mucin. Like the pancreatic lesions, many of those in the SMG showed mild to moderate inflammatory and fibrotic changes. All showed immunoreactivity with anti-A antibody (Figure 3e). The normal SMG acini and ductules did not express A antigen. In a hamster examined at 20 weeks, the patterns of the lesions showing back-to-back arrangements of irregularly shaped glands with cystic distention warranted the diagnosis of an in situ carcinoma. There were no islet
Islet Cells and Exocrine Carcinogenesis
1459
AJP November 1995, Vol. 14 7, No. 5
.i
Figure 2. Induction ofperi- and intra-insular ductular lesions in hamsters after 10 weeks (a and b) and 20 weeks (c). In a and b, the hypercellular and partially distended ductules occupy a part of the islet and show no connection to the surrounding exocrine ductules. In c, part of the islet is replaced by bvperplastic, atypical, and branched ductular structures. H&E, X 210.
Table 2. Incidence and Type of Lesions Found in the Left SMG*
Group 1 2 3 4
5 6 7 8 9
Material and site of transplants Islet, SMG Islet, SMG HBBS, SMG Isolated pancreatic ductal cells, SMG Islet, peri-SMG Islet, renal subcapsule
Thyroid, left SMG; TAKA-1 cells, right SMG Heart, left SMG; starch, right SMG
BOP
A
10 weeks B
C
A
20 weeks B
C
Yes No Yes Yes
4/5 0/4 0/5 0/5
0/5 0/4 0/5 0/5
0/5 0/4 0/5 0/5
1/3 0/4 0/5 0/5
1/4 0/4 0/5 0/5
0/4 0/4 0/5 0/5
Yes Yes Yes Yes
0/5 0/5 0/5
0/5 0/5 0/5
0/5 0/5 0/5
NE
NE
NE
0/5 0/5 0/5 0/5
0/5 0/5 0/5 0/5
0/5 0/5 0/5 0/5
Yes
NE
NE
NE
0/5
0/5
0/5
*A, atypical ductular proliferation; B, ductal/ductular in situ carcinoma; C, invasive ductal/ductular carcinoma. NE, not examined. The data between group 1 hamsters were significant (P < 0.05) compared with the data in the remaining group (X2 method). aO.2 ml of HBBS was injected. bFreshly isolated pancreatic ductal cells were injected.
cells but many hemosiderin deposits within or around these ductular lesions, all of which were confined to lower pole of the left SMG, where the islets were transplanted. With the electron microscope, the SMG lesions showed patterns that were found in induced pancreatic cancer.2" The cells had irregular-shaped nuclei, poorly developed rough endoplasmic reticulum, and electron-dense bodies resembling mucigen. There were no zymogen or endocrine granules (Figure 4). Microvilli were present at the luminal poles of the cells.
Such SMG lesions were not found in the right SMG of any hamsters, nor were there any lesions in the left SMG of hamsters in group 3 or 4, in which HBSS or freshly isolated pancreatic ductal cells, respectively, were injected. There were also no lesions in the peri-SMG of the group 5 hamsters. In all 10 hamsters of group 2, small or large aggregates of islets were found. The cellular composition of these islets (a, f3, and 8 cells) was similar to pancreatic islets. Most of these islets were arranged around the SMG excretory ducts (Figure 5). Hemosiderin deposits and minimal or
1460
Ishikawa et al
AJP November 1995, Vol. 147, No. 5
Figure 3. Atypical ductularproliferation in the Iclft 5MtIG of bamsters 10 weeks after BOP treatment. (a) Lesion in the inferior pole of the gland (center). Inflammatory cells infiltrating around the lesion can be seen at this magnification (H&E, X 42). (b) Higher power view of the lesion illustrated in a. 7he atypical ductules are surrounde-d by inflammatory cells. The larer ductule in the center is branched and contains inflammatory and sbed cels- in the lumen. Similar changes werefrequently seen in the induced pancreatic lesions (H&E, X 210). (c) Lesions in the inferior pole of another ha mster. iVote the presence of ductules of various size and cell tvpe. Some glands are filled wvith mucus (center), which uwas PAS-positive. (H&E, X 240). (d) Lesion in the lcft SMG of a bamster at 20 weeks. The pattern (o this lesion was consistent with in situ carcinioma. Fibrotic and inflammatory' reaction are present. No islet cells could he identified immunohistochemically. (x210). (e) Immunohistochemical staininzg with anti-A antibody shows a strong immunoreactivity of the induced 5MG lesion. (Group 1, at 10 weeks; avidin-hiotin complex,
