NOS isoforms in human stomach mucosa has not been done than the .... Antral mucosa with intestinal metaplasia and mucosa infected by Helicobacter pylori ...
Carcinogenesis vol.18 no.9 pp.1841–1845, 1997
SHORT COMMUNICATION
Expression of differential nitric oxide synthase isoforms in human normal gastric mucosa and gastric cancer tissue
Andrea Rajnakova, Peter M.Y.Goh, Steven T.F.Chan, Sing Shang Ngoi, Ahmet Alponat, Shabbir Moochhala1 Department of Surgery, National University Hospital, Singapore 1To
whom correspondence should be addressed at: Department of Surgery, National University Hospital, 5 Lower Kent Ridge Road, Singapore 119074
The present study investigated the expression and distribution of three isoforms of nitric oxide synthase (NOS) in different anatomical regions of the human stomach and in gastric neoplastic tissues by immunohistochemistry using specific antibodies. Intracellular localization of individual isoenzymes of NOS was detected in normal gastric mucosa. Gastric cancer tissues had a marked reduction of all three NOS isoforms expression. The expression of the endothelial NOS, neuronal NOS and inducible NOS in the tumor tissue was significantly lower than in normal gastric mucosa (P J 0.01, P J 0.02, P < 0.01, respectively). In the tumor tissue the expression of inducible NOS was significantly lower than the expression of both constitutive forms of NOS (P < 0.01). There was a tendency to higher expression of both constitutive forms of NOS in earlier stages T2 of the tumor compared to advanced T4 tumor. In contrast, the expression of inducible NOS was higher than in the advanced T4 tumor than in the earlier stages T2 of the tumor. The mapping of the expression of endothelial NOS, neuronal NOS and inducible NOS in human stomach showed higher expression of NOS isoforms in the distal third than in the proximal third of the stomach (P J 0.03, P J 0.04, P J 0.01, respectively). We conclude that there is greater expression of NOS in the stomach corpus and in antrum than in the proximal third of the normal human stomach mirroring the anatomical predilection of common pathological changes in this part of the human stomach. Furthermore, there was loss of the expression of individual isoenzymes in gastric neoplasms.
Nitric oxide (NO*) is a messenger molecule serving a variety of functions in many different tissues (1). This highly reactive and short-lived molecule is involved in a wide variety of unrelated processes and contributes to many physiological functions and pathophysiological actions. NO is synthesized in mammalian cells from L-arginine by the enzyme nitric oxide synthase (NOS). Three forms of this enzyme, namely the endothelial NOS, neuronal NOS and inducible NOS are expressed in human tissues and vary considerably in their distribution, regulation and function. The constitutive form of NOS, which includes endothelial and neuronal isoforms, produces a small amount of NO (picomoles) which plays a role in signal transmission and results in physiological effects *Abbreviations: NOS, nitric oxide synthase; NO, nitric oxide; iNOS, inducible NOS; ecNOS, endothelial NOS; bNOS, neuronal NOS; AN, adjacent normal mucosa; RN, remote normal mucosa. © Oxford University Press
(2). In contrast, the much larger amounts of NO (nanomoles) released by cells in response to cytokines and endotoxins is synthesized by inducible form of NOS and mediates some of the cytotoxic and cytostatic effects of the immune system (3). In the gastrointestinal tract NO plays a role as a nonadrenergic, noncholinergic neurotransmitter, vasodilator and participates in sustaining the mucosal immunity and integrity of the gut (2,4). It participates in gastrointestinal mucosal protection by regulation of mucosal blood flow through vasodilation and inhibition of platelet aggregation in gastric microvasculature and may also modulate the epithelial cell integrity by secretion of mucus or bicarbonate. Together with its direct microbicidal properties, NO may act as a first line of mucosal defence against luminal aggressors in the stomach (5). The present study investigated the presence and distribution of three isoenzymes of NOS in the human stomach using immunohistochemical staining. Mapping of the expression of NOS isoforms in human stomach mucosa has not been done previously. In our study we also analyzed the expression of NOS isoforms in gastric cancer tissue and in normal adjacent mucosa obtained at a distance from the tumor. All patients included in this study underwent a total or subtotal gastrectomy. We analyzed samples from 25 patients (12 males, 13 females) with a median age of 61 years (range 39–77 years). Paired samples of the gastric mucosa from the resected stomach were taken from the edge of tumor and surrounding non-neoplastic mucosa 2 cm (adjacent normal, AN) and 5–10 cm (remote normal, RN) away from the tumor. For the detection of NOS isoforms in the tumor tissue we used both frozen sections and paraffin sections. Tissue samples for paraffin sections were fixed in formalin for not more than 24 h and processed in an automatic tissue processor using graded alcohols, xylene and paraffin wax. The tissue samples for frozen sections were snap frozen in liquid nitrogen and stored at 270°C until analyzed. For the mapping of the expression of NOS in the stomach, samples of the mucosa were taken from different anatomical regions of the stomach (cardia, fundus, lesser curvature, greater curvature, antrum, pylorus), snap frozen in liquid nitrogen and stored at 270°C. Tissue samples used for the mapping of NOS isoforms in the stomach were taken from the macroscopically and histologically intact parts of the stomach, minimum 10 cm away from the tumor. Immunohistochemistry was used to localise the inducible NOS (iNOS), endothelial NOS (ecNOS) and neuronal NOS (bNOS) expression in separate slides. Cryostat and paraffin sections were taken at 4 µm and mounted on pretreated slides. After equilibrating with room temperature for 30 min, the frozen section slides were fixed in acetone for 5 min. Prior to staining paraffin sections were deparaffinized and rehydrated and 0.9% H2O2 was applied for 15 min to quench endogenous peroxidase activity. All slides were then incubated in normal horse serum (Vector Lab, USA) followed by incubation with avidin and then biotin (Vector Lab, USA) for 15 min. The primary antibodies were added to the slides 1841
A.Rajnakova et al.
and incubated for 30 min followed by incubation with the secondary antibody for 60 min. The frozen section slides were then incubated in 0.9% H2O2 for 10 min and all slides were stained using the Vectastain elite ABC kit (biotinylated antibody, Vector Lab, USA) for 60 min. The final colouring agent used was 3,39-diaminobenzidine tetrahydrochloride. Tissue was counter stained with haematoxylin, rinsed well with buffer, dehydrated and mounted prior to examination under the microscope. H&E staining was used to verify the pathology. For the detection of ecNOS in the epithelium of the human stomach, anti EC-NOS IgG mouse monoclonal was used (Transduction Laboratories, USA). For the detection of the iNOS, the anti Mac-NOS IgG mouse monoclonal was used (Transduction Laboratories, USA). For the detection of the bNOS, the anti b-NOS IgG mouse monoclonal was used (Transduction Laboratories, USA). The antibodies showed specificity to human ecNOS, bNOS and iNOS. We performed Western Blotting to confirm the presence of specific protein fragments corresponding to ecNOS, bNOS and iNOS respectively (Transduction Laboratories, USA). We used Boeringer Mannheim Chemiluminiscence Western Blotting Kit. With standard chemiluminiscent substrate provided by kit we did not detect the cross-reaction between bNOS and iNOS antibodies. When we used Super Signal Substrate Western Blotting ultra chemiluminiscent with femtogram sensitivity we detected the possible cross-reaction of bNOS antibody with iNOS, but not of iNOS antibody with bNOS. Previous work in our laboratory has shown positive staining of normal human colon mucosa which served as the positive control for the NOS staining characteristics (6). The cancer tissue with the corresponding peritumoral mucosa and the normal mucosa was mounted on the same slide which served as internal controls. A negative control was made by using non-specific horse serum instead of the primary antibody. The slides were evaluated independently by two of the observers who were ‘blinded’ to histological states of the specimens and the scores were then compared to verify the staining intensity of the tissue. Samples were rated according to a score which was calculated by adding the scale of intensity of staining to the area of the staining (score 5 intensity 1 area). The intensity of staining has the following scale; 0, no staining of the mucosal epithelial cells; 11, mild staining; 21, moderate staining; 31, marked staining (comparable with that of the control mucosa). The area of staining was evaluated differently using the following scale: 0, no staining of the mucosal epithelial cells in any field; 11, ,25% of the epithelium stain positive; 21, 25–50% stain positive; 31, 50–75% stain positive; 41, .75% stain positive. The maximum score when added was 7 and the minimum 0. The presence of Helicobacter pylori was detected by H&E staining and immunohistochemical staining using the DAKO LSAB1Kit, Peroxidase K679. The non-parametric Cochran–Mantel–Haenszel test was used to compare the expression of ecNOS, bNOS and iNOS in tumor vs peritumoral sites and normal mucosa. The nonparametric Mann–Whitney U-test was used to compare the expression of individual isoenzymes in the different groups of tumor grade. The Kruskal–Wallis test was used to compare NOS isoenzyme expression in different tumor stages. The Wilcoxon Signed-Ranks test was used for the mapping of the expression of ecNOS, bNOS and iNOS in different anatomical regions of the stomach A P-value of ,0.05 was considered significant. All tumors originated in the mucosal layer and were classified 1842
as adenocarcinoma. The grade of the tumor in 16 cases was classified as poorly differentiated adenocarcinoma and in 9 cases as moderately differentiated adenocarcinoma (Table I). Immunohistochemistry was used to detect the presence and the distribution of three different isoforms of NOS in the human stomach and human gastric cancer tissue. The expression of ecNOS, bNOS and iNOS in the tumor tissue was significantly lower (P 5 0.01, P 5 0.02, P , 0.01) than in the adjacent normal mucosa (AN) and remote normal mucosa (RN) (n 5 11, Table II) (Figure 1). In the tumor tissues (n 5 25) the expression of iNOS was significantly lower than the expression of both constitutive forms ecNOS and bNOS (P , 0.01) (Table I). In the tumor tissue iNOS was undetectable in 80% of all tumors, reduced in 20%; ecNOS was undetectable in 28% of tumors, reduced in 32% and normal in 40%; bNOS was undetectable in 32% of tumors, reduced in 56% and normal in 12%. There were no significant differences in the expression of individual isoforms in the different stages of tumor, but there was a tendency to higher expression of both constitutive forms of NOS in earlier stages of the tumor (T2) compared to advanced tumor (T4). The expression of iNOS had a tendency to be higher in the advanced tumor (T4) than in the earlier stages of tumor (T2). The distribution of the constitutive forms of NOS in the epithelial cells was cytoplasmic with a slightly basal localisation of the stain. The localisation of the inducible form of NOS was mostly in the apical part of the epithelial cells. Antral mucosa with intestinal metaplasia and mucosa infected by Helicobacter pylori was stained markedly with all three types of NOS antibodies. Expression of the constitutive forms of NOS when compared with the inducible form was stronger in the non-neoplastic mucosa of the stomach. Constitutive forms of NOS were localized in the cytoplasm of the epithelial cells, and also was detectable in the vascular endothelium. The inducible form in the neoplastic tissue was expressed in only five patients, all of them with advanced stage of tumor (T3 in two patients, T4 in three patients). iNOS staining was localized in the cytoplasm of the tumor cells and also in the endothelial cells of the neovasculature. The production of tumor necrosis factor stimulates this isoenzyme production (7). The development of neovasculature is essential for sustaining tumor growth and inducible NOS activity was detected in the endothelial cells of tumor capillaries and vascular smooth muscles in well-established tumors, but not at earlier stages of tumor development (8). Although resting unstimulated cells do not express iNOS, the capacity to express this enzyme exists in nearly every tissue in the body (9). Tumors in humans have been shown to be variably infiltrated by lymphocytes that include host T-cells and macrophages. In addition, the antral mucosa in 52% of our gastric cancer patients was colonised by Helicobacter pylori. Heliobacter pylori colonisation of antral mucosa is usually accompanied by mononuclear infiltration. Cells outside of the classic immune system participate in a programmed manner during inflammation and respond to cytokine stimulation by expressing iNOS and associated genes (7). Tumor cells of the early stages of gastric cancer lack iNOS generating capacity and NO could be produced by activated macrophages and inflammatory cells as a result of the host response. Tumor necrosis factor increases the expression of the high output iNOS (7) thus this isoenzyme may be expressed in greater
Expression of differential nitric oxide synthase isoforms
Table I. The expression of ecNOS, bNOS and iNOS in the tumor Patient age/sex
Site of the tumor
Grade of the tumor
Stage of the tumor
ecNOS in tumor (intensity/area)
bNOS in tumor (intensity/area)
iNOS in tumor (intensity/area)
Type of tissue processing
61/M 61/F 70/F 76/F 39/F 61/M 75/M 67/M 61/F 74/M 57/F 67/M 56/M 67/M 60/F 64/M 68/F 60/M 61/M 70/M 58/F 47/F 51/F 77/F 74/F
cardia antrum antrum corpus antrum corpus antrum cardia corpus antrum corpus corpus corpus corpus antrum antrum cardia cardia cardia cardia corpus antrum antrum antrum antrum
moderate moderate moderate poor poor moderate moderate moderate moderate moderate poor poor poor poor poor poor moderate poor poor poor poor poor poor poor poor
T1 T1 T1 T1 T1 T2 T2 T3 T3 T3 T3 T3 T3 T3 T3 T3 T4 T4 T4 T4 T4 T4 T4 T4 T4
2/2 3/4 3/4 0 0 3/4 1/2 0 2/3 3/4 2/4 3/2 1/2 3/4 0 2/1 0 0 3/4 3/4 0 2/3 3/4 1/2 3/4
3/2 1/2 2/3 0 0 1/2 3/1 0 1/2 3/3 1/1 3/2 2/3 1/1 0 1/1 0 0 1/2 1/3 0 1/1 1/1 0 1/1
0 0 0 0 0 0 0 0 0 1/1 0 2/1 0 0 0 0 0 0 1/1 0 0 1/1 0 0 1/1
paraffin paraffin paraffin frozen frozen paraffin frozen frozen paraffin paraffin paraffin frozen frozen frozen frozen paraffin frozen frozen paraffin paraffin frozen frozen paraffin paraffin paraffin
Table II. The expression of ecNOS, bNOS and iNOS in the tumor tissue (T), adjacent normal mucosa (AN) and remote normal mucosa (RN). All samples in this table were processed as frozen section. The scoring system (intensity/area) is defined in the text Patient age/sex
76/F 39/F 75/M 67/M 56/M 67/M 60/F 68/F 60/M 58/F 47/F
Site of the tumor
corpus antrum antrum cardia corpus corpus antrum cardia cardia corpus antrum
Grade of the tumor
poor poor moderate moderate poor poor poor moderate poor poor poor
Stage of the tumor
T1 T1 T2 T3 T3 T3 T3 T4 T4 T4 T4
amounts in the advanced stages of the tumor. Nitric oxide with its tumoricidal activity may contribute to the process of necrosis, a feature frequently observed in solid tumors. Normal gastric mucosa at the cardia and fundus does not express the iNOS, but antral mucosa and corporal mucosa in all individuals infected by Helicobacter pylori expressed iNOS. The tissue at tumor adjacent or remote mucosa which expressed iNOS was taken from gastric mucosa which had been colonized by Helicobacter pylori in antrum or in the stomach corpus. Thus this could be one of the possible explanations for the higher expression of iNOS in the distal half of the stomach. Expression of ecNOS, bNOS and iNOS isoenzymes were greater in the lower part of the body of stomach than in the proximal part of the stomach (cardia) (P 5 0.03, 0.04 and 0.01, respectively). This could be explained either by: (i)
H.pylori presence negative positive positive positive positive positive positive positive negative negative positive
ecNOS
bNOS
iNOS
T
AN RN
T
AN RN
T
AN RN
0 0 1/2 0 1/2 3/2 0 0 0 0 2/3
0 0 1/3 1/1 1/2 2/1 2/1 0 0 0 3
0 0 3/1 0 2/3 3/2 0 0 0 0 1/1
1/1 1/3 1/2 1/3 2/2 2/2 3/1 0 0 2/1 2/1
0 0 0 0 0 2/1 0 0 0 0 1/1
0 0 1/3 0 0 1/2 3/1 1/1 0 0 2/1
0 2/1 2/4 3/3 2/1 2/2 3/3 0 0 1/1 2/3
1/1 3/1 2/3 3/3 2/1 3/2 3/4 1/1 0 2/2 2/1
0 2/1 2/2 2/3 1/1 2/1 2/2 1/2 0 0 2/1
Helicobacter pylori colonises the antral mucosa and releases products that result in tissue inflammation (10). This bacteria was present in 52% of our patients. Helicobacter pylori colonisation of gastric antral mucosa is usually accompanied by mononuclear cell infiltrates that are capable of releasing potent mediators including NO in response to luminal antigens or (ii) intestinal metaplasia represents a change of mucosal pattern from gastric type to small bowel and/or colonic type. It is a common pattern in persons over 60 years of age and is often associated with multifocal atrophic gastritis. It can affect any area of the stomach but the most common is in the antrum. We found intestinal metaplasia in 68% of our patients and the metaplastic epithelium was markedly stained by all three different antibodies. It has been shown that human colonic epithelial cells express both iNOS and ecNOS (6). 1843
A.Rajnakova et al.
Fig. 1. Immunohistochemistry of NOS isoenzymes in normal stomach mucosa and gastric cancer tissue. The dilution for each NOS antibodies was 1:300. A2F represent normal and tumor tissue taken from the same patient and processed in the same way as frozen sections. (A) ecNOS in normal gastric mucosa showing the staining of the gastric glands (20x), taken from the antrum, (B) iNOS in normal gastric mucosa (20x), taken from the antrum of the stomach, (C) bNOS in normal gastric mucosa (20x), taken from the antrum of the stomach, (D) ecNOS in gastric cancer tissue, the absence of the stain in the tumor glands (20x), (E) iNOS in gastric cancer tissue, the absence of the stain in the tumor glands (20x), (F) bNOS in gastric cancer tissue, the absence of the stain in the tumor glands (20x).
The function of NO in tumor development, promotion and progression is not clear. One can speculate that NO plays both beneficial and detrimental role in patients with gastric cancer. Its beneficial tumoricidal effect which results in tumor cell necrosis could help in the defensive role of the immune system of the body. The inverse relationship between NO production and the metastatic potential of the tumor has been proposed 1844
(11). Nitric oxide has been shown to be responsible for the growth arrest and terminal differentiation of cells. A loss of nitric oxide in the tumor could help tumor cells escape programmed cell death and retain the ability to multiply. On the other hand NO could promote tumor growth through creating neovasculature. Further studies are necessary to understand the possible
Expression of differential nitric oxide synthase isoforms
correlations between the stage of the tumor and the presence of tumor metastases and NOS expression and subsequent NO generating capacity. There may be important clinical implications if we can establish the point at which NO ceases to be beneficial for the body by tumoricidal and microbicidal activity and becomes harmful by supporting tumor promotion/ progression. Acknowledgements The authors gratefully acknowledge the technical assistance of Ms Tan Li Li and statistical analysis of the data by Mrs Bee Choo Koh. This work was supported by the Singapore Cancer Society and The Rueben Meyer Trust Fund.
References 1. Moncada,S., Palmer,M.J. and Higgs,E.A. (1991) Nitric oxide: physiology, pathophysiology and pharmacology. Pharmacol. Rev., 43, 109–139. 2. Stark,M.E. and Szurszewski,J.H. (1992) Role of nitric oxide in gastrointestinal and hepatic function and disease. Gastroenterology, 103, 1928– 1949. 3. Assurey,J., Canha,F.Q., Liew,F.Y. and Moncada,S. (1993) Feedback inhibition of nitric oxide synthase activity by nitric oxide. Br. J. Pharmacol., 108, 833–837. 4. Brown,J.F., Tepperman,B.L., Hanson,P.J., Whittle,B.J.R. and Moncada,S. (1992) Differential distribution of nitric oxide synthase between cell fractions isolated from the rat gastric mucosa. Biochem. Biophys. Res. Commun., 184, 680–685. 5. Desai,K.M., Sessa,W.C. and Vane,J.R. (1991) Involvement of nitric oxide in the reflux relaxation of the stomach to accommodate food or fluid. Nature, 351, 477–479. 6. Moochhala,S., Chhatwal,V.J.S., Chan,S.T.F., Ngoi,S.S., Chia,Y.W. and Abu Rauff. (1996) Nitric oxide synthase activity and expression in human colorectal cancer. Carcinogenesis, 17(5), 1171–1174. 7. Billiar,T.R. (1995) Nitric oxide—novel biology with clinical relevance. Ann. Surg., 221, 339–349. 8. Buttery,L.D.K., Springall,D.R., Andrade,S.P., Riveros-Moreno,V., Hart,I., Piper,P.J. and Polak,J.M. (1993) Induction of nitric oxide synthase in the neo-vasculature of experimental tumours in mice. J. Pathol., 171, 311–319. 9. Morris,S.M. Jr. and Billiar,R.R. (1994) New insights into the regulation of inducible nitric oxide synthesis. Am. J. Physiol., 266, E8292E839. 10. Blaser,M.J. (1992) Hypotheses on the pathogenesis and natural history of Helicobacter pylori-induced inflammation. Gastroenterology, 102, 720– 727. 11. Yim,C., Bastian,N.R., Smith,J.C., Hibbs,J.B. Jr and Samlowski,W.E. (1993) Macrophage nitric oxide synthesis delays progression of ultraviolet lightinduced murine skin cancers. Cancer Res., 53, 5507–5511. Received on January 27, 1997; revised on May 30, 1997; accepted on June 6, 1997
1845
