- Email: [email protected]
Thioproline prevents carcinogenesis in the remnant stomach induced by duodenal reflux
Cancer Letters 237 (2006) 256–262 www.elsevier.com/locate/canlet
Thioproline prevents carcinogenesis in the remnant stomach induced by duodenal reflux Masashi Suo, Ken-ichi Mukaisho, Akihiko Shimomura, Hiroyuki Sugihara, Takanori Hattori* Department of Pathology, Shiga University of Medical Science, Seta-tsukinowa-cho, Ohtsu, Shiga 520-2192, Japan Received 12 April 2005; received in revised form 30 May 2005; accepted 6 June 2005
Abstract An excessive duodenal reflux induced by surgery has been widely accepted to cause gastric carcinogenesis in the remnant stomach. As one of causative factors for malignancy, N-nitroso compounds produced by enteric bacteria have been postulated. However, there is no concrete information to prove this hypothesis. This study was undertaken to elucidate the factors underlying the remnant stomach carcinogenesis, by giving thiazolidine-4-carboxylic acid (thioproline; TPRO) to the rats with duodenal reflux as a nitrite scavenger. Operated 39 animals were used, divided into 2 groups; one with a diet containing 0.5% TPRO (nZ18), and the other with a diet without TPRO (nZ21). Adenocarcinoma developed in 16 rats out of 21 (76.2%) of untreated rats, whereas adenocarcinoma was detected in 1 rat of the TPRO-treated rats (5.6%). TPRO thus prevented the development of gastric cancer in the remnant stomach, thereby suggesting a concern of nitroso compounds to the carcinogenesis. q 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Remnant gastric cancer; Duodenal reflux; Bile acid; Thioproline; Nitroso compound
1. Introduction Epidemiological cohort studies of gastric cancers in the remnant stomach have established that a key factor in carcinogenesis is excessive duodenogastric reflux induced by surgery [1–4]. Recent attention has focused on the mechanism by which reflux of duodenal contents induces gastric carcinogenesis. The duodenal reflux consists of bile, pancreatic juice * Corresponding author. Tel.:C81 77 548 2166; fax: C81 77 543 9880. E-mail address: [email protected] (T. Hattori).
and duodenal secretes. It has been recently reported that pancreaticoduodenal reflux containing bile but not pancreatic juice alone, is responsible for gastric carcinogensis [5], and bile acids are usually cited as promoters of both gastric and colonic carcinogenesis [6–9]. However, bile acids alone do not seem to be carcinogenic, because they are not mutagenic in bacterial studies [10]. As to the carcinogenesis by bile, the following hypothesis has been proposed. Bile acids damage gastric mucosa and result in HCback diffusion through the mucosal surface, which allows potential carcinogens to easily penetrate the mucosal barrier to
0304-3835/$ - see front matter q 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2005.06.019
M. Suo et al. / Cancer Letters 237 (2006) 256–262
Thioproline N-nitrosothioproline
S
S N H
COOH N H
+ NO2
NO3
COOH
H+ (In the stomach)
NO
urine
Fig. 1. The schema of denitrosation in vivo by TPRO. TPRO is trapping nitrite in the stomach and then being excreted as N-nitrosothioproline in the urine.
act on mucosal stem cells. Normally, the gastric contents are sterile, but reflux of duodenal juice and diminished acid secretion lead to bacterial colonization in the stomach. Higher pH values, nitrite levels, and bacterial counts with nitrate reductase activity are observed in the gastric juice of the remnant stomach. The existence of N-nitroso compounds in the environment and their implications for human cancer risk have been well documented [11,12]. Although much of the research in this area has been directed towards formation and biological activity of the N-nitrosamines, less effort has focused upon the N-nitrosamides which are known to be potent directacting mutagens and carcinogens. A major source of nitrosamides in the human environment is the conjugated bile acids that are secreted in bile. The bile acids are normally present as cholic acid amides of glycine or taurine [13,14]. These amides would be N-nitrosated by enteric bacteria in the remnant stomach, possibly causing gastric carcinogenesis. Several reports referred to the presence of bacterial flora in duodenal juice that are capable of catalyzing endogenous reactions to produce nitroso compounds [15,16]. However, there is no direct proof that shows gastric carcinogenesis by nitroso compounds in the duodenal reflux. This may be due to the fact that nitroso compounds are so small in amount and unstable that we could not identify them chemically. Thiazolidine-4-carboxylic acid {thioproline (TPRO)} is a cyclic sulfur-containing amino acid, and is a condensation product of cysteine and formaldehyde [17,18]. And it is reported that TPRO
257
is a trapping agent in vivo for nitrosating species originating from N-nitroso compounds which are denitrosated non-enzymatically in stomach acidic conditions, and it is a most sensitive probe for evaluating nitrosating capacity [19–22], and N-nitrosothioproline is excreted into the urine without being metabolized [20] (Fig. 1). If TPRO traps nitroso compounds, it’s administration in animals seems to decrease the development of cancers caused by nitroso compounds. In the present study, we studies the effect of thioproline administration on remnant stomach carcinogenesis induced by duodenal reflux in rats.
2. Materials and methods 2.1. Duodenal reflux model Male Wistar rats weighing approximately 250 g were used in this experiment. After 24 h fasting, a midline laparotomy incision was made under inhalation anesthesia with diethyl ether, and reflux of duodenal contents was induced according to previously reported procedure [23]. The jejunum was transected from about 2 cm anal to the origin, and the proximal cut end was anastomosed end-to-side to the greater curvature of the gastric corpus. The duodenum was then divided at the supra-papillary portion, and the distal cut end was closed. The proximal cut end of the duodenum was connected with the distal cut end of the jejunum by end-to-end anastomosis to restore intestinal continuity (Fig. 2). All sutures were performed with interrupted 7–0 nylon sutures. Animals were allowed access to water 12 h postoperatively and food 36 h later. Animals were not treated with any known carcinogens. Post-operatively, survived animals were divided into two groups according to diet. Animals belonging to the untreated group were given normal diet (CRF-1) as a control, while the TPRO-treated group was provided food containing 0.5% TPRO. Animals were sacrificed using overdose of diethyl ether at 80th week after surgery, and morphological changes of their stomachs with about 2 cm jejunum, connected with stomach, were studied.
258
M. Suo et al. / Cancer Letters 237 (2006) 256–262
according to the previous literature [24]. Among these lesions, adenomatous growth with both atypia and invasion was classified as adenocarcinoma; other adenomatous growths were defined as adenoma. If some lesions mentioned above were detected partially, we counted them as positive cases. 2.3. Statistical analyses Statistical evaluation was performed using Fisher’s exact test. Values of P!0.05 were considered statistically significant. Fig. 2. Operative procedure. B, bile duct; D, duodenum; F, fore stomach; G, glandular stomach; J, upper jejunum. All duodenal contents flowed back into the glandular stomach through the gastrojejunal stoma.
2.2. Histological examination and definition of gastric lesions The removed stomachs were fixed in 10% formalin in PBS for 4 h and embedded in paraffin. We made serial 4 mm sections and stained with hematoxylin and eosin (H&E). We examined the glandular stomach and intestine around the site of anastomosis in all cases. Definitions of lesions included adenocarcinoma, adenoma, cystic dilatation, submucosal adenocystic proliferation and pyloric metaplasia are
3. Results The thirty-nine animals survived for the 80-week experimental period, and comprised the untreated group (nZ21) and TPRO-treated group (nZ18). There was no significant difference in body weight of rats between the untreated and the TPRO-treated group. The significant observations were all present around the site of gastrojejunal anastomosis, and between animals were broadly similar overall in the same group. Macroscopically, an uneven surface with erosion and polypoid lesion around the site of anastomosis was detected in all rats of the untreated group (Fig. 3(A)). Most animals of the TPRO-treated
Fig. 3. Macroscopic findings in both groups. (A) the untreated group, (B) the TPRO treated group. D, duodenum; G, glandular stomach; T, tumor. All animals have tumors (polypoid lesions) in the anastomosis in the untreated group (arrow). On the other hand, there are only slight protruded lesions, which have adenocystic proliferation of mucous glands in the submucosal layer, in most animals (14 out of 18) of the TPROtreated group.
M. Suo et al. / Cancer Letters 237 (2006) 256–262 Table 1 Incidence of various lesions in the anastomosis of each groups.
Adenocarcinoma Adenoma Submucosal adenocytic proliferation Pyloric metaplasia
Control (nZ21)
TPRO (nZ18)
16 (76.2%) 21 (100%) 21 (100%)
1 (5.6%) 4 (22.2%) 16 (88.9%)
21 (100%)
18 (100%)
259
21 rats (76.2%) of the untreated group, whereas 1 out of 18 (5.6%) adenocarcinoma was detected in the TPRO-treated rats (Fisher’s exact test, P!0.05).
4. Discussion
group, only mucosal redness, mild erosion and slight protruded lesion were detected (Fig. 3(B)). Although 4 out of 18 rats in the TPRO-treated group displayed similar mucosal changes described above in the control group, mucosal changes in these animals of the TPRO-treated group were mild compared to those in the untreated group. Histological findings in this study were summarized in Table 1. Loupe image of typical lesion in the untreated group was showed in Fig. 4. Stomal gastritis with foveolar hyperplasia, erosion, pyloric metaplasia and adenocystic proliferation of mucous glands were observed in the mucosa adjacent to the anastomosis in both groups. The histological feature was similar to gastritis cystica profunda or gastritis cystica polyposa (GCP) in humans [25,26]. These findings were more marked in rats of the untreated group than those in the TPRO-treated group. Adenomas and adenocarcinomas were detected in GCP like lesions or in the erosive portion near these lesions in both groups. Adenomas (Fig. 5) were detected in all samples (100%) of control groups and in 4 out of 18 rats (22.2%) in the TPRO group, respectively (Fisher’s exact test, P!0.05). Adenocarcinomas (Fig. 6) developed in 16 out of
Remnant gastric cancer is recognized as a carcinoma developing in the remnant stomach after surgery. Excessive reflux of duodenal contents induced by surgery is widely accepted to play a crucial role in carcinogenesis. The cancer in the remnant stomach is occasionally classified into three categories: (1) cancer newly developed in the remnant stomach (CRS), (2) cancer remaining in the remnant stomach at the initial gastric surgery, and (3) recurrent cancer in the remnant stomach [27]. The reflux of duodenal juice into the stomach could be associated with the development of CRS [28]. The present study has confirmed the conclusion in the previous reports that prolonged reflux of duodenal contents alone could induce carcinoma in the stomach [29–33]. So, the duodenal reflux model of this study is a useful model for understanding mechanisms that underly CRS carcinogenesis. CRSs usually appeared at the site of the gastrojejunostomy and showed adenocarcinoma-type histology with GCP. Many investigators believe that GCP is associated with cancer development [34]. In the present study, pyloric metaplasia and adenocystic proliferation, which are the component of GCP, were detected in not only all the animals of the untreated group but also most animals in the TPRO-treated
Fig. 4. Loupe image of typical polypoid lesion in the untreated group. The polypoid lesion is similar to GCP in humans.
260
M. Suo et al. / Cancer Letters 237 (2006) 256–262
Fig. 5. Tublar adenoma developed in the untreated group. (A) HE!100, (B) higher magnification of Fig. 5(A) (HE!400). Tubular adenoma consists of closely packed tubules (A). The tubules are relatively regular, and the nuclei of adenoma cells have low grade atypia (B).
group. However, the mucosal changes were milder in the TPRO-treated group than those of the untreated rats. As one of causative factors for malignancies of the digestive organs, N-nitroso compounds have been postulated. Also in the remnant stomach carcinogenesis, it is probable that N-nitroso compounds play a role, whereas Busby et al. have shown that N-nitroso cholic and taurocholic acids caused gastric cancers in non-operated rats [35]. In this, a question is if bile acids may be nitrosated, and another problem may be if other N-nitroso compounds are formed in a condition of duodenal reflux. Correa et al. hypothesized the following mechanism of carcinogenesis by nitroso compounds in nonoperated human stomach. A hypoacidy state is induced by atrophic gastritis, which allows the overgrowth of bacterial microflora with nitrate-reducing activity,
which convert dietary nitrate to nitrite, and it seems also probable that nitrite is produced by bacterias in the duodenum with reflux operation, and it combines with amines or a bile acid as the amides, which is present in the duodenal juice, to form N-nitroso compounds [36]. Matsumoto observed nitrate-reducing bacteria was increased in duodenal contents of rats with reflux operation [37]. However, he could not demonstrate the presence of any N-nitroso-TCA or N-nitroso-GCA, and other N-nitroso derivatives in any sample of duodenal contents of rats with reflux operation [38]. In previous studies, it was difficult to detect N-nitroso compounds directly, because they are so small in amount and may be instable chemically. The present study can speculate N-nitroso compounds play an important role as the cause of neoplastic development in the remnant stomach.
Fig. 6. Tublar adenocarcinoma developed in the untreated group. (A) HE!100, (B) higher magnification of Fig. 6(A) (HE!400). Atypical glands invade to the stromal tissue. The tumor contains branching tubles with structural atypia (A). The carcinoma cells also have moderate nuclear pleomorphism, and mitoses (arrows) are sometimes detected.
M. Suo et al. / Cancer Letters 237 (2006) 256–262
In this study, we evaluated the inhibitory effect of TPRO as a nitrite scavenger and as a most sensitive probe to detect N-nitroso compounds. We found TPRO prevented the remnant stomach carcinogenesis, thereby indicating the possible relevance of N-nitroso compounds to the carcinogenesis. Only1 rat out of 18 (5.6%) developed gastric cancer in TPRO-treated group, whereas 16 rats out of 21 (76.2%) of untreated group developed gastric cancer. In the TPRO-treated rats, it was interesting that the mucosal changes were less severe than in the untreated rats and the occurrence of adenoma, a cancer-associated lesion, was less frequent than those of the untreated rats. Recently, we also applied TPRO to Barrett’s esophagus carcinogenesis in rats with duodenogastro-esophageal reflux, and found that TPRO inhibited the development of esophageal adenocarcinomas [39]. A situation in the remnant stomach carcinogenesis appears to be analogous to that of Barrett’s esophagus, and N-nitroso compounds seemed to play an important role in the reflux carcinogenesis.
Acknowledgements This work was supported in part by Grant-in-aid for Cancer Research from the Ministry of Health, Labour and Welfare.
References [1] P. Schlag, R. Bockler, H. Ulrich, M. Peter, P. Merkle, C. Hefarth:, Are nitrite and N-nitroso compounds in gastric juice risk factors for carcinoma in the operated stomach?, Lancet 1 (1980) 727–729. [2] R. Orlando 3rd, J.P. Welch, Carcinoma of the stomach after gastric operation, Am. J. Surg. 141 (1981) 487–491. [3] C.P. Caygill, M.J. Hill, J.S. Kirkham, T.C. Northfield, Mortality from gastric cancer following gastric surgery for peptic ulcer, Lancet 1 (1986) 929–931. [4] C. Toftgaard, Gastric cancer after peptic ulcer surgery. A historic prospective cohort investigation, Ann Surg 210 (1989) 159–164. [5] K. Miwa, T. Fujimura, H. Hasegawa, T. Kosaka, R. Miyata, I. Miyazaki, T. Hattori, Is bile or are pancreaticoduodenal secretions related to gastric carcinogenesis in rats with reflux through the pylorus?, J. Cancer. Res. Clin. Oncol. 118 (1992) 570–574.
261
[6] A. Kuwahara, T. Saito, M. Kobayashi, Bile acids promote carcinogenesis in the remnant stomach of rats, J. Cancer Res. Clin. Oncol. 115 (1989) 423–428. [7] K. Kaibara, M. Takebayashi, H. Koga, Deoxycholic acids as a promoter in the development of gastric carcinoma induced by N-methyl-N 0 -nitro-N-nitrosoguanidine in rats, Cancer J. 2 (1988) 14–16. [8] P.Y. Cheah, H. Bernstein, Modification of DNA by bile acids: a possible factor in the etiology of colon cancer, Cancer Lett. 49 (1990) 207–210. [9] D. Qiao, W. Chen, E.D. Stratagoules, J.D. Martinez, Bile acid-induced activation of activator protein-1 requires both extracellular signal-regulated kinase and protein kinase C signaling, J. Biol. Chem. 275 (2000) 15090– 15098. [10] S. Venitt, D. Bosworth, D.F. Easton, Lack of mutagenic activity of bile acids in bacterial fluctuation tests, Mutat. Res. 190 (1987) 191–196. [11] P. Schlag, R. Bockler, M. Peter, C. Herfarth, Nitrite and N-nitroso compounds in the operated stomach, Scand. J. Gastroenterol. 67 (Suppl.) (1981) 63–69. [12] P.N. Magee, J.M. Barnes, Carcinogenic nitroso compounds, Adv. Cancer Res. 10 (1967) 163–246. [13] J.B. Carey Jr., in: P.B. Nair, D. Kritchevsky (Eds.), The Bile Acids, Plenum Publishing, New York, 1973, pp. 58–82. [14] T.A. Miettinen, in: P.B. Nair, D. Kritchevsky (Eds.), The Bile Acids vol. 2, Plenum Publishing Corp., New York, 1973, pp. 191–247. [15] L. Domellof, B.S. Reddy, J.H. Weisburger, Microflora and deconjugation of bile acids in alkaline reflux after partial gastrectomy, Am. J. Surg. 140 (1980) 291–295. [16] T.J. Muscroft, S.A. Deane, D. Youngs, D.W. Burdon, M.R. Keighley, The microflora of the postoperative stomach, Br. J. Surg. 68 (1981) 560–564. [17] M.P. Schubert, Compounds of thiol acids with aldehydes, J. Biol. Chem. 114 (1936) 341–350. [18] S. Ratner, H.T. Clarke, The action of formaldehyde upon cystein, J. Am. Chem. Soc. 59 (1937) 200–206. [19] T. Tahira, M. Suda, K. Wakabayashi, M. Nagao, T. Sugimura, Kinetics of nitrosation of thioproline, the precursor of a major nitroso compound in human urine, and its role as a nitrite scavenger, Gann. 75 (1984) 889–894. [20] M. Tsuda, Y. Kurashima, Nitrite-trapping capacity of thioproline in the human body, IARC Sci. Publ. 105 (1991) 123–128. [21] N. Frank, M. Tsuda, H. Ohgaki, E. Frei, T. Kato, S. Sato, Detoxifying potential of thioproline against N-nitroso compounds, N-nitrosodimethylamine and N-nitrosocimetidine, Cancer Lett. 50 (1990) 167–172. [22] M. Tsuda, N. Frank, S. Sato, T. Sugimura, Marked increase in the urinary level of N-nitrosothioproline after ingestion of cod with vegetables, Cancer Res. 48 (1988) 4049–4052. [23] K. Mukaisho, K. Miwa, H. Kumagai, M. Bamba, H. Sugihara, T. Hattori, Gastric carcinogenesis by duodenal reflux through gut regenerative cell lineage, Dig. Dis. Sci. 48 (2003) 2153–2158.
262
M. Suo et al. / Cancer Letters 237 (2006) 256–262
[24] K. Miwa, T. Kamata, I. Miyazaki, T. Hattori, Kinetic changes and experimental carcinogenesis after Billroth I and II gastrectomy, Br. J. Surg. 80 (1993) 893–896. [25] E.R. Littler, E. Gleibermann, Gastritis cystica polyposa (gastric mucosal prolapse at gastroenterostomy site, with cystic and infiltrative epithelial hyperplasia), Cancer 29 (1972) 205–209. [26] G. Franzin, P. Novelli, Gastritis cystica profunda, Histopathology 5 (1981) 535–547. [27] T. Kidokoro, Long term survival of patients who underwent resection of cancer in the remnant stomach, Jpn J. Clin. Oncol. 17 (1982) 2029–2034. [28] N. Tanigawa, E. Nomura, M. Niki, H. Shinohara, K. Nishiguchi, M. Okuzawa, M. Toyoda, S. Morita, et al., Clinical study to identify specific characteristics of cancer newly developed in the remnant stomach, Gastric cancer 5 (2002) 23–28. [29] P. Langhans, R.A. Heger, J. Hohenstein, W. Schlake, H. Bunte, Operation-sequelae of carcinoma of the stomach, Experimental studies of surgical techniques with or without resection, World J. Surg. 5 (1981) 595–605. [30] K. Kondo, H. Suzuki, T. Nagayo, The influence of gastrojejunal anastomosis on gastric carcinogenesis in rats, Jpn J. Cancer Res. 75 (1984) 362–369. [31] F. Theuring, S. Dittrich, F.H. Wolter, On the varying degrees of carcinogenicity of modified gasrtroentero-anastomoses, Exp. Pathol. 27 (1985) 179–184. [32] R.C. Mason, Duodenogastric reflux in rat gastric carcinoma, Br. J. Surg. 73 (1986) 801–803.
[33] K. Miwa, H. Hasegawa, T. Fujimura, H. Matsumoto, R. Miyata, T. Kosaka, I. Miyazaki, T. Hattori, et al., Duodenal reflux through the pylorus induces gastric adenocarcinoma in the rat, Carcinogenesis 13 (1992) 2313–2316. [34] H. Mitomi, K. Iwabuchi, A. Amemiya, G. Kaneda, K. Adachi, T. Asao, Immunohistochemical analysis of a case of gastritis cystica profunda associated with carcinoma development, Scand. J. Gastroenterol. 33 (1998) 1226–1229. [35] W.F. Busby Jr., D.E. Shuker, G. Charnley, P.M. Newberne, S. R. Tannenbaum, G.N. Wogan, Carcinogenicity in rats of the nitrosated bile acid conjugates N-nitrosoglycocholic acid and N-nitrosotaurocholic acid, Cancer Res. 45 (1985) 1367–1371. [36] P. Correa, W. Haenszel, C. Cuello, S. Tannenbaum, M. Archer, A model for gastric cancer epidemiology, Lancet 12 (1975) 58–60. [37] H. Matsumoto, Mutagenicity of gastric and duodenal juice on gastric carcinogenesis in rats with duodenal reflux, Nippon Geka Gakkai Zasshi 95 (1994) 368–375 [Article in Japanese, English abstract]. [38] M. Fein, K.H. Fuchs, H. Stopper, S. Diem, M. Herderich, Duodenogastric reflux and foregut carcinogenesis: analysis of duodenal juice in a rodent model of cancer, Carcinogenesis 21 (2000) 2079–2084. [39] H. Kumagai, K. Mukaisho, H. Sugihara, K. Miwa, G. Yamamoto, T. Hattori, Thioproline inhibits development of esophageal adenocarcinoma induced by gastroduodenal reflux in rats, Carcinogenesis 25 (2004) 723–727.
Thioproline prevents carcinogenesis in the remnant stomach induced by duodenal reflux Masashi Suo, Ken-ichi Mukaisho, Akihiko Shimomura, Hiroyuki Sugihara, Takanori Hattori* Department of Pathology, Shiga University of Medical Science, Seta-tsukinowa-cho, Ohtsu, Shiga 520-2192, Japan Received 12 April 2005; received in revised form 30 May 2005; accepted 6 June 2005
Abstract An excessive duodenal reflux induced by surgery has been widely accepted to cause gastric carcinogenesis in the remnant stomach. As one of causative factors for malignancy, N-nitroso compounds produced by enteric bacteria have been postulated. However, there is no concrete information to prove this hypothesis. This study was undertaken to elucidate the factors underlying the remnant stomach carcinogenesis, by giving thiazolidine-4-carboxylic acid (thioproline; TPRO) to the rats with duodenal reflux as a nitrite scavenger. Operated 39 animals were used, divided into 2 groups; one with a diet containing 0.5% TPRO (nZ18), and the other with a diet without TPRO (nZ21). Adenocarcinoma developed in 16 rats out of 21 (76.2%) of untreated rats, whereas adenocarcinoma was detected in 1 rat of the TPRO-treated rats (5.6%). TPRO thus prevented the development of gastric cancer in the remnant stomach, thereby suggesting a concern of nitroso compounds to the carcinogenesis. q 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Remnant gastric cancer; Duodenal reflux; Bile acid; Thioproline; Nitroso compound
1. Introduction Epidemiological cohort studies of gastric cancers in the remnant stomach have established that a key factor in carcinogenesis is excessive duodenogastric reflux induced by surgery [1–4]. Recent attention has focused on the mechanism by which reflux of duodenal contents induces gastric carcinogenesis. The duodenal reflux consists of bile, pancreatic juice * Corresponding author. Tel.:C81 77 548 2166; fax: C81 77 543 9880. E-mail address: [email protected] (T. Hattori).
and duodenal secretes. It has been recently reported that pancreaticoduodenal reflux containing bile but not pancreatic juice alone, is responsible for gastric carcinogensis [5], and bile acids are usually cited as promoters of both gastric and colonic carcinogenesis [6–9]. However, bile acids alone do not seem to be carcinogenic, because they are not mutagenic in bacterial studies [10]. As to the carcinogenesis by bile, the following hypothesis has been proposed. Bile acids damage gastric mucosa and result in HCback diffusion through the mucosal surface, which allows potential carcinogens to easily penetrate the mucosal barrier to
0304-3835/$ - see front matter q 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2005.06.019
M. Suo et al. / Cancer Letters 237 (2006) 256–262
Thioproline N-nitrosothioproline
S
S N H
COOH N H
+ NO2
NO3
COOH
H+ (In the stomach)
NO
urine
Fig. 1. The schema of denitrosation in vivo by TPRO. TPRO is trapping nitrite in the stomach and then being excreted as N-nitrosothioproline in the urine.
act on mucosal stem cells. Normally, the gastric contents are sterile, but reflux of duodenal juice and diminished acid secretion lead to bacterial colonization in the stomach. Higher pH values, nitrite levels, and bacterial counts with nitrate reductase activity are observed in the gastric juice of the remnant stomach. The existence of N-nitroso compounds in the environment and their implications for human cancer risk have been well documented [11,12]. Although much of the research in this area has been directed towards formation and biological activity of the N-nitrosamines, less effort has focused upon the N-nitrosamides which are known to be potent directacting mutagens and carcinogens. A major source of nitrosamides in the human environment is the conjugated bile acids that are secreted in bile. The bile acids are normally present as cholic acid amides of glycine or taurine [13,14]. These amides would be N-nitrosated by enteric bacteria in the remnant stomach, possibly causing gastric carcinogenesis. Several reports referred to the presence of bacterial flora in duodenal juice that are capable of catalyzing endogenous reactions to produce nitroso compounds [15,16]. However, there is no direct proof that shows gastric carcinogenesis by nitroso compounds in the duodenal reflux. This may be due to the fact that nitroso compounds are so small in amount and unstable that we could not identify them chemically. Thiazolidine-4-carboxylic acid {thioproline (TPRO)} is a cyclic sulfur-containing amino acid, and is a condensation product of cysteine and formaldehyde [17,18]. And it is reported that TPRO
257
is a trapping agent in vivo for nitrosating species originating from N-nitroso compounds which are denitrosated non-enzymatically in stomach acidic conditions, and it is a most sensitive probe for evaluating nitrosating capacity [19–22], and N-nitrosothioproline is excreted into the urine without being metabolized [20] (Fig. 1). If TPRO traps nitroso compounds, it’s administration in animals seems to decrease the development of cancers caused by nitroso compounds. In the present study, we studies the effect of thioproline administration on remnant stomach carcinogenesis induced by duodenal reflux in rats.
2. Materials and methods 2.1. Duodenal reflux model Male Wistar rats weighing approximately 250 g were used in this experiment. After 24 h fasting, a midline laparotomy incision was made under inhalation anesthesia with diethyl ether, and reflux of duodenal contents was induced according to previously reported procedure [23]. The jejunum was transected from about 2 cm anal to the origin, and the proximal cut end was anastomosed end-to-side to the greater curvature of the gastric corpus. The duodenum was then divided at the supra-papillary portion, and the distal cut end was closed. The proximal cut end of the duodenum was connected with the distal cut end of the jejunum by end-to-end anastomosis to restore intestinal continuity (Fig. 2). All sutures were performed with interrupted 7–0 nylon sutures. Animals were allowed access to water 12 h postoperatively and food 36 h later. Animals were not treated with any known carcinogens. Post-operatively, survived animals were divided into two groups according to diet. Animals belonging to the untreated group were given normal diet (CRF-1) as a control, while the TPRO-treated group was provided food containing 0.5% TPRO. Animals were sacrificed using overdose of diethyl ether at 80th week after surgery, and morphological changes of their stomachs with about 2 cm jejunum, connected with stomach, were studied.
258
M. Suo et al. / Cancer Letters 237 (2006) 256–262
according to the previous literature [24]. Among these lesions, adenomatous growth with both atypia and invasion was classified as adenocarcinoma; other adenomatous growths were defined as adenoma. If some lesions mentioned above were detected partially, we counted them as positive cases. 2.3. Statistical analyses Statistical evaluation was performed using Fisher’s exact test. Values of P!0.05 were considered statistically significant. Fig. 2. Operative procedure. B, bile duct; D, duodenum; F, fore stomach; G, glandular stomach; J, upper jejunum. All duodenal contents flowed back into the glandular stomach through the gastrojejunal stoma.
2.2. Histological examination and definition of gastric lesions The removed stomachs were fixed in 10% formalin in PBS for 4 h and embedded in paraffin. We made serial 4 mm sections and stained with hematoxylin and eosin (H&E). We examined the glandular stomach and intestine around the site of anastomosis in all cases. Definitions of lesions included adenocarcinoma, adenoma, cystic dilatation, submucosal adenocystic proliferation and pyloric metaplasia are
3. Results The thirty-nine animals survived for the 80-week experimental period, and comprised the untreated group (nZ21) and TPRO-treated group (nZ18). There was no significant difference in body weight of rats between the untreated and the TPRO-treated group. The significant observations were all present around the site of gastrojejunal anastomosis, and between animals were broadly similar overall in the same group. Macroscopically, an uneven surface with erosion and polypoid lesion around the site of anastomosis was detected in all rats of the untreated group (Fig. 3(A)). Most animals of the TPRO-treated
Fig. 3. Macroscopic findings in both groups. (A) the untreated group, (B) the TPRO treated group. D, duodenum; G, glandular stomach; T, tumor. All animals have tumors (polypoid lesions) in the anastomosis in the untreated group (arrow). On the other hand, there are only slight protruded lesions, which have adenocystic proliferation of mucous glands in the submucosal layer, in most animals (14 out of 18) of the TPROtreated group.
M. Suo et al. / Cancer Letters 237 (2006) 256–262 Table 1 Incidence of various lesions in the anastomosis of each groups.
Adenocarcinoma Adenoma Submucosal adenocytic proliferation Pyloric metaplasia
Control (nZ21)
TPRO (nZ18)
16 (76.2%) 21 (100%) 21 (100%)
1 (5.6%) 4 (22.2%) 16 (88.9%)
21 (100%)
18 (100%)
259
21 rats (76.2%) of the untreated group, whereas 1 out of 18 (5.6%) adenocarcinoma was detected in the TPRO-treated rats (Fisher’s exact test, P!0.05).
4. Discussion
group, only mucosal redness, mild erosion and slight protruded lesion were detected (Fig. 3(B)). Although 4 out of 18 rats in the TPRO-treated group displayed similar mucosal changes described above in the control group, mucosal changes in these animals of the TPRO-treated group were mild compared to those in the untreated group. Histological findings in this study were summarized in Table 1. Loupe image of typical lesion in the untreated group was showed in Fig. 4. Stomal gastritis with foveolar hyperplasia, erosion, pyloric metaplasia and adenocystic proliferation of mucous glands were observed in the mucosa adjacent to the anastomosis in both groups. The histological feature was similar to gastritis cystica profunda or gastritis cystica polyposa (GCP) in humans [25,26]. These findings were more marked in rats of the untreated group than those in the TPRO-treated group. Adenomas and adenocarcinomas were detected in GCP like lesions or in the erosive portion near these lesions in both groups. Adenomas (Fig. 5) were detected in all samples (100%) of control groups and in 4 out of 18 rats (22.2%) in the TPRO group, respectively (Fisher’s exact test, P!0.05). Adenocarcinomas (Fig. 6) developed in 16 out of
Remnant gastric cancer is recognized as a carcinoma developing in the remnant stomach after surgery. Excessive reflux of duodenal contents induced by surgery is widely accepted to play a crucial role in carcinogenesis. The cancer in the remnant stomach is occasionally classified into three categories: (1) cancer newly developed in the remnant stomach (CRS), (2) cancer remaining in the remnant stomach at the initial gastric surgery, and (3) recurrent cancer in the remnant stomach [27]. The reflux of duodenal juice into the stomach could be associated with the development of CRS [28]. The present study has confirmed the conclusion in the previous reports that prolonged reflux of duodenal contents alone could induce carcinoma in the stomach [29–33]. So, the duodenal reflux model of this study is a useful model for understanding mechanisms that underly CRS carcinogenesis. CRSs usually appeared at the site of the gastrojejunostomy and showed adenocarcinoma-type histology with GCP. Many investigators believe that GCP is associated with cancer development [34]. In the present study, pyloric metaplasia and adenocystic proliferation, which are the component of GCP, were detected in not only all the animals of the untreated group but also most animals in the TPRO-treated
Fig. 4. Loupe image of typical polypoid lesion in the untreated group. The polypoid lesion is similar to GCP in humans.
260
M. Suo et al. / Cancer Letters 237 (2006) 256–262
Fig. 5. Tublar adenoma developed in the untreated group. (A) HE!100, (B) higher magnification of Fig. 5(A) (HE!400). Tubular adenoma consists of closely packed tubules (A). The tubules are relatively regular, and the nuclei of adenoma cells have low grade atypia (B).
group. However, the mucosal changes were milder in the TPRO-treated group than those of the untreated rats. As one of causative factors for malignancies of the digestive organs, N-nitroso compounds have been postulated. Also in the remnant stomach carcinogenesis, it is probable that N-nitroso compounds play a role, whereas Busby et al. have shown that N-nitroso cholic and taurocholic acids caused gastric cancers in non-operated rats [35]. In this, a question is if bile acids may be nitrosated, and another problem may be if other N-nitroso compounds are formed in a condition of duodenal reflux. Correa et al. hypothesized the following mechanism of carcinogenesis by nitroso compounds in nonoperated human stomach. A hypoacidy state is induced by atrophic gastritis, which allows the overgrowth of bacterial microflora with nitrate-reducing activity,
which convert dietary nitrate to nitrite, and it seems also probable that nitrite is produced by bacterias in the duodenum with reflux operation, and it combines with amines or a bile acid as the amides, which is present in the duodenal juice, to form N-nitroso compounds [36]. Matsumoto observed nitrate-reducing bacteria was increased in duodenal contents of rats with reflux operation [37]. However, he could not demonstrate the presence of any N-nitroso-TCA or N-nitroso-GCA, and other N-nitroso derivatives in any sample of duodenal contents of rats with reflux operation [38]. In previous studies, it was difficult to detect N-nitroso compounds directly, because they are so small in amount and may be instable chemically. The present study can speculate N-nitroso compounds play an important role as the cause of neoplastic development in the remnant stomach.
Fig. 6. Tublar adenocarcinoma developed in the untreated group. (A) HE!100, (B) higher magnification of Fig. 6(A) (HE!400). Atypical glands invade to the stromal tissue. The tumor contains branching tubles with structural atypia (A). The carcinoma cells also have moderate nuclear pleomorphism, and mitoses (arrows) are sometimes detected.
M. Suo et al. / Cancer Letters 237 (2006) 256–262
In this study, we evaluated the inhibitory effect of TPRO as a nitrite scavenger and as a most sensitive probe to detect N-nitroso compounds. We found TPRO prevented the remnant stomach carcinogenesis, thereby indicating the possible relevance of N-nitroso compounds to the carcinogenesis. Only1 rat out of 18 (5.6%) developed gastric cancer in TPRO-treated group, whereas 16 rats out of 21 (76.2%) of untreated group developed gastric cancer. In the TPRO-treated rats, it was interesting that the mucosal changes were less severe than in the untreated rats and the occurrence of adenoma, a cancer-associated lesion, was less frequent than those of the untreated rats. Recently, we also applied TPRO to Barrett’s esophagus carcinogenesis in rats with duodenogastro-esophageal reflux, and found that TPRO inhibited the development of esophageal adenocarcinomas [39]. A situation in the remnant stomach carcinogenesis appears to be analogous to that of Barrett’s esophagus, and N-nitroso compounds seemed to play an important role in the reflux carcinogenesis.
Acknowledgements This work was supported in part by Grant-in-aid for Cancer Research from the Ministry of Health, Labour and Welfare.
References [1] P. Schlag, R. Bockler, H. Ulrich, M. Peter, P. Merkle, C. Hefarth:, Are nitrite and N-nitroso compounds in gastric juice risk factors for carcinoma in the operated stomach?, Lancet 1 (1980) 727–729. [2] R. Orlando 3rd, J.P. Welch, Carcinoma of the stomach after gastric operation, Am. J. Surg. 141 (1981) 487–491. [3] C.P. Caygill, M.J. Hill, J.S. Kirkham, T.C. Northfield, Mortality from gastric cancer following gastric surgery for peptic ulcer, Lancet 1 (1986) 929–931. [4] C. Toftgaard, Gastric cancer after peptic ulcer surgery. A historic prospective cohort investigation, Ann Surg 210 (1989) 159–164. [5] K. Miwa, T. Fujimura, H. Hasegawa, T. Kosaka, R. Miyata, I. Miyazaki, T. Hattori, Is bile or are pancreaticoduodenal secretions related to gastric carcinogenesis in rats with reflux through the pylorus?, J. Cancer. Res. Clin. Oncol. 118 (1992) 570–574.
261
[6] A. Kuwahara, T. Saito, M. Kobayashi, Bile acids promote carcinogenesis in the remnant stomach of rats, J. Cancer Res. Clin. Oncol. 115 (1989) 423–428. [7] K. Kaibara, M. Takebayashi, H. Koga, Deoxycholic acids as a promoter in the development of gastric carcinoma induced by N-methyl-N 0 -nitro-N-nitrosoguanidine in rats, Cancer J. 2 (1988) 14–16. [8] P.Y. Cheah, H. Bernstein, Modification of DNA by bile acids: a possible factor in the etiology of colon cancer, Cancer Lett. 49 (1990) 207–210. [9] D. Qiao, W. Chen, E.D. Stratagoules, J.D. Martinez, Bile acid-induced activation of activator protein-1 requires both extracellular signal-regulated kinase and protein kinase C signaling, J. Biol. Chem. 275 (2000) 15090– 15098. [10] S. Venitt, D. Bosworth, D.F. Easton, Lack of mutagenic activity of bile acids in bacterial fluctuation tests, Mutat. Res. 190 (1987) 191–196. [11] P. Schlag, R. Bockler, M. Peter, C. Herfarth, Nitrite and N-nitroso compounds in the operated stomach, Scand. J. Gastroenterol. 67 (Suppl.) (1981) 63–69. [12] P.N. Magee, J.M. Barnes, Carcinogenic nitroso compounds, Adv. Cancer Res. 10 (1967) 163–246. [13] J.B. Carey Jr., in: P.B. Nair, D. Kritchevsky (Eds.), The Bile Acids, Plenum Publishing, New York, 1973, pp. 58–82. [14] T.A. Miettinen, in: P.B. Nair, D. Kritchevsky (Eds.), The Bile Acids vol. 2, Plenum Publishing Corp., New York, 1973, pp. 191–247. [15] L. Domellof, B.S. Reddy, J.H. Weisburger, Microflora and deconjugation of bile acids in alkaline reflux after partial gastrectomy, Am. J. Surg. 140 (1980) 291–295. [16] T.J. Muscroft, S.A. Deane, D. Youngs, D.W. Burdon, M.R. Keighley, The microflora of the postoperative stomach, Br. J. Surg. 68 (1981) 560–564. [17] M.P. Schubert, Compounds of thiol acids with aldehydes, J. Biol. Chem. 114 (1936) 341–350. [18] S. Ratner, H.T. Clarke, The action of formaldehyde upon cystein, J. Am. Chem. Soc. 59 (1937) 200–206. [19] T. Tahira, M. Suda, K. Wakabayashi, M. Nagao, T. Sugimura, Kinetics of nitrosation of thioproline, the precursor of a major nitroso compound in human urine, and its role as a nitrite scavenger, Gann. 75 (1984) 889–894. [20] M. Tsuda, Y. Kurashima, Nitrite-trapping capacity of thioproline in the human body, IARC Sci. Publ. 105 (1991) 123–128. [21] N. Frank, M. Tsuda, H. Ohgaki, E. Frei, T. Kato, S. Sato, Detoxifying potential of thioproline against N-nitroso compounds, N-nitrosodimethylamine and N-nitrosocimetidine, Cancer Lett. 50 (1990) 167–172. [22] M. Tsuda, N. Frank, S. Sato, T. Sugimura, Marked increase in the urinary level of N-nitrosothioproline after ingestion of cod with vegetables, Cancer Res. 48 (1988) 4049–4052. [23] K. Mukaisho, K. Miwa, H. Kumagai, M. Bamba, H. Sugihara, T. Hattori, Gastric carcinogenesis by duodenal reflux through gut regenerative cell lineage, Dig. Dis. Sci. 48 (2003) 2153–2158.
262
M. Suo et al. / Cancer Letters 237 (2006) 256–262
[24] K. Miwa, T. Kamata, I. Miyazaki, T. Hattori, Kinetic changes and experimental carcinogenesis after Billroth I and II gastrectomy, Br. J. Surg. 80 (1993) 893–896. [25] E.R. Littler, E. Gleibermann, Gastritis cystica polyposa (gastric mucosal prolapse at gastroenterostomy site, with cystic and infiltrative epithelial hyperplasia), Cancer 29 (1972) 205–209. [26] G. Franzin, P. Novelli, Gastritis cystica profunda, Histopathology 5 (1981) 535–547. [27] T. Kidokoro, Long term survival of patients who underwent resection of cancer in the remnant stomach, Jpn J. Clin. Oncol. 17 (1982) 2029–2034. [28] N. Tanigawa, E. Nomura, M. Niki, H. Shinohara, K. Nishiguchi, M. Okuzawa, M. Toyoda, S. Morita, et al., Clinical study to identify specific characteristics of cancer newly developed in the remnant stomach, Gastric cancer 5 (2002) 23–28. [29] P. Langhans, R.A. Heger, J. Hohenstein, W. Schlake, H. Bunte, Operation-sequelae of carcinoma of the stomach, Experimental studies of surgical techniques with or without resection, World J. Surg. 5 (1981) 595–605. [30] K. Kondo, H. Suzuki, T. Nagayo, The influence of gastrojejunal anastomosis on gastric carcinogenesis in rats, Jpn J. Cancer Res. 75 (1984) 362–369. [31] F. Theuring, S. Dittrich, F.H. Wolter, On the varying degrees of carcinogenicity of modified gasrtroentero-anastomoses, Exp. Pathol. 27 (1985) 179–184. [32] R.C. Mason, Duodenogastric reflux in rat gastric carcinoma, Br. J. Surg. 73 (1986) 801–803.
[33] K. Miwa, H. Hasegawa, T. Fujimura, H. Matsumoto, R. Miyata, T. Kosaka, I. Miyazaki, T. Hattori, et al., Duodenal reflux through the pylorus induces gastric adenocarcinoma in the rat, Carcinogenesis 13 (1992) 2313–2316. [34] H. Mitomi, K. Iwabuchi, A. Amemiya, G. Kaneda, K. Adachi, T. Asao, Immunohistochemical analysis of a case of gastritis cystica profunda associated with carcinoma development, Scand. J. Gastroenterol. 33 (1998) 1226–1229. [35] W.F. Busby Jr., D.E. Shuker, G. Charnley, P.M. Newberne, S. R. Tannenbaum, G.N. Wogan, Carcinogenicity in rats of the nitrosated bile acid conjugates N-nitrosoglycocholic acid and N-nitrosotaurocholic acid, Cancer Res. 45 (1985) 1367–1371. [36] P. Correa, W. Haenszel, C. Cuello, S. Tannenbaum, M. Archer, A model for gastric cancer epidemiology, Lancet 12 (1975) 58–60. [37] H. Matsumoto, Mutagenicity of gastric and duodenal juice on gastric carcinogenesis in rats with duodenal reflux, Nippon Geka Gakkai Zasshi 95 (1994) 368–375 [Article in Japanese, English abstract]. [38] M. Fein, K.H. Fuchs, H. Stopper, S. Diem, M. Herderich, Duodenogastric reflux and foregut carcinogenesis: analysis of duodenal juice in a rodent model of cancer, Carcinogenesis 21 (2000) 2079–2084. [39] H. Kumagai, K. Mukaisho, H. Sugihara, K. Miwa, G. Yamamoto, T. Hattori, Thioproline inhibits development of esophageal adenocarcinoma induced by gastroduodenal reflux in rats, Carcinogenesis 25 (2004) 723–727.