© 2004 Foundation for Promotion of Cancer Research
Diffuse Expression of hRFI is Correlated with Blood Vessel Invasion in Gastric Carcinoma
Department of Surgical Oncology, The University of Tokyo, Tokyo, Japan
For reprints and all correspondence: Shin Sasaki, Department of Surgical Oncology, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-8655, Japan. E-mail: SASAKI-1SU{at}h.u-tokyo.ac.jp
Received March 25, 2004; accepted July 11, 2004
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Background: hRFI, which has a relatively high homology to XIAP, is preferentially expressed in esophageal and colorectal carcinomas, and is involved in the initial tumor formation in the colorectal adenoma-carcinoma sequence. Furthermore, its diffuse expression is associated with colorectal carcinogenesis. However, hRFI expression in gastric carcinomas has not been evaluated so far.
Methods: We performed immunohistochemical staining on 76 gastric carcinoma samples using the antibody to hRFI and also analyzed the correlation between the staining pattern of hRFI and the clinico-pathological characteristics.
Results: All of the samples were stained focally (31 cases, 40.8%) or diffusely (45 cases, 59.2%) in the cancerous region. On the contrary, most of the normal gastric region showed no staining, except for a few cases that showed slight immunoreactivity in speckles. Furthermore, the proportion of blood vessel involvement was significantly higher in carcinomas with diffuse hRFI expression (28/45, 62.2%) than in carcinomas with focal expression (7/31, 22.6%) (P < 0.001). Liver metastasis was consistently observed in five cases (11.1%) in diffuse, but only one (3.3%) in focal type during the average follow up period of 5 years. However, the 3-year survival rate did not show significant difference between these different staining patterns of hRFI.
Conclusions: These results suggest that the detection of the expression pattern of hRFI in gastric carcinomas can be another useful predictor of liver recurrence, especially when combined with other factors.
Key Words: gastric carcinoma • inhibitor of apoptosis protein • oncogene • vessel invasion
Abbreviations: IAP, inhibitor of apoptosis protein • TNF, tumor necrosis factor • BIR, baculoviral inhibition of apoptosis protein repeat
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Recently, we cloned a novel gene named hRFI, standing for human Ring Finger homologous to Inhibitor of apoptosis protein (IAP), by the two-hybrid yeast screening using a human homolog of l(2)tid, hTID-1L, as bait (1). hTID-1L is one of the molecules that stimulats cell apoptosis through the release of cytochrome c from mitochondria and subsequent activation of caspase 3 (2). By the homology search of amino acid sequence, 22 residues at the C termini of hRFI are identical to the Ring Finger domain spanning 53 residues of the X-chromosome-linked inhibitor of apoptosis protein (XIAP). hRFI itself is not a member of the IAP family because it lacks the baculoviral inhibition of apoptosis protein repeat (BIR) domain (3), which is indispensable for the anti-apoptotic function of IAP (4). However, we found that HeLa cells transfected with hRFI construct exhibited a tendency to resist TNF-
-induced apoptosis and that hRFI protein was shown to be cleaved within the DEDD sequence spanning residues 230–233 by caspase 3 during the apoptotic induction (1). These observations clearly suggest the involvement of hRFI in the cellular apoptotic pathway and its potent anti-apoptotic function. Furthermore, we found that hRFI is highly expressed in esophageal and colorectal carcinomas. The immunohistochemical staining performed on adenoma-carcinoma sequence of the colorectum disclosed that hRFI is involved in the initial tumor formation and its diffuse expression is associated with colorectal carcinogenesis (5). Hence, we analyzed hRFI expression in gastric carcinomas by immunohistochemical staining and searched for correlation between the expression of hRFI and clinicopathological factors. |
MATERIALS AND METHODS |
|---|
|
TOPAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
STUDY SUBJECTS
The subjects of this study were 76 consecutive patients with gastric carcinomas that were surgically resected at the University of Tokyo between November 1997 and April 2000. In every case, 3-micron sections were prepared from theformalin-fixed, paraffin-embedded tissue and mounted onto slides coated with polylysin. All the resected primary tumors were histologically examined by hematoxylin-eosin staining according to the International Union Against Cancer TNM classification (6). Gastric carcinoma was histologically classified into four types: papillary adenocarcinoma, tubular adenocarcinoma, poorly differentiated adenocarcinoma and signet ring cell carcinoma. The clinicopathological data of each case including age, sex, location, histological type, size, depth, lymphatic and blood vessel invasion were obtained from clinical and pathological documents. Tumor location was divided into three regions: upper (U), middle (M) and lower (L). Blood vessel invasion was also evaluated by elastica van Gieson staining.
After discharge, all the patients were followed up at 3-month intervals during 2 years after surgery and 6-month intervals thereafter with physical examination and serum level of CEA and CA19-9 in the peripheral blood. Further, in order to check the intra-abdominal tumor recurrence, upper gastrointestinal endoscopy as well as CT scan or ultrasonography imaging were performed every 6 months within 2 years after surgery and every 12 months thereafter. The average duration of follow up was 5 years.
IMMUNOHISTOCHEMISTRY
In our previous study, we generated the polyclonal rabbit anti-human RFI antibody (1). The method of generation of the antibody was as described previously (1). After incubating the sections with the primary antibody, detection of the antigen-antibody complex was performed by the avidin/biotin method according to the manufacturer's recommendation. In brief, after deparaffinization with xylene for 15 min, sections were washed in a graded series of alcohol and then washed three times in phosphate-buffered saline (PBS). Sections were then microwaved for 21 min, and again washed in PBS. Endogenous peroxidase activity was quenched with 1.5% H2O2 in methanol for 25 min and the sections were washed in PBS. Non-specific binding sites were blocked with block buffer HISTOFINE SAB-PO(R) kit (Nichirei), and then washed again in PBS (3 x 5 min). Thereafter, the sections were incubated with primary antibody against hRFI at a concentration of 0.5 µg/ml at 4°C for 24 h. After several washes with PBS, detection of the primary antibody was performed with the same kit. After the removal of the secondary antibody complex with PBS, counterstaining with Mayer's hematoxylin was performed at room temperature. Thereafter, the slides were washed, dehydrated and covered with xylene gelatine.
EVALUATION AND STATISTICAL ANALYSIS
The immunostained specimens were evaluated by two different investigators without the knowledge of the clinicopathological features. When almost all the tumor cells were homogenously stained, they were classified as diffuse type. In other cases, the staining was heterogenous among each tumor cells and positive staining was detected in less than 50% of the tumor cells that were classified as focal type. Statistical analysis was performed using the likelihood chi-squared analysis, Fisher's exact test or Mann–Whitney U test. Probability (P) values that were <0.05 were considered statistically significant.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Of the 76 gastric carcinomas examined by the immunohistochemical staining, all were stained focally (31 cases, 40.8%) or diffusely (45 cases, 59.2%) in the cancerous region (Fig. 1A and B). In most of the stained cases, hRFI was dominantly expressed in the cytoplasm of the cancer cells. On the contrary, most of the normal gastric region showed no staining, except for a few cases that showed slight immunoreactivity in speckles (Fig. 2). The case shown in Fig. 1C–E exhibited vessel invasion, and tumor cells stained with anti-hRFI antibody were visible. The relation with clinicopathological features is summarized in Table 1. The staining pattern of hRFI did not show any significant correlations with age, sex, depth of invasion, histology, lymphatic invasion or lymph node metastasis. However, in the middle third of the stomach (M region), the proportion of cases with focal staining (18/30, 60.0%) was significantly higher than that of cases with diffuse staining (12/30, 40.0%) (P = 0.017). Furthermore, blood vessel involvement was detected more frequently in diffuse (28/45, 62.2%) than in focal (7/31, 22.6%) staining pattern, and the difference was statistically significant (P < 0.001). Distant metastasis was observed in five cases (11.1%) in diffuse type, but only one case (3.3%) in focal type during the average follow up period of 5 years. In the diffuse type, all metastatic lesions were observed in the liver. The liver metastasis was synchronously detected in one case, while the other two cases developed metastasis within 1 year after surgery. In the focal type, the metastatic sites in the single case were the liver and the skin. However, the 3-year survival rate did not show a significant difference in the staining pattern of hRFI.
|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
In our previous studies, hRFI homologous to a member of IAP was highly expressed in esophageal and colorectal carcinomas by the Northern blotting and/or immunohistochemical staining (1,5). For the colorectal carcinomas, the expression of hRFI was specific in the cancerous regions, and none of the normal colorectal glands exhibited any staining. Furthermore, we found that the expression of hRFI was an early event in the adenoma-carcinoma sequence of the colorectum and that the expression pattern changed from focal to diffuse as the degree of dysplasia of adenomas progressed. To clarify the expression profile of this novel gene in gastric carcinomas, we performed immunohistochemical staining with the anti-hRFI antibody. The distinction of the staining pattern was the same as the previous study. Like the colorectal carcinomas, all specimens were stained focally or diffusely in the cancerous region, and most of the normal glands showed negative staining. The preferential expression of hRFI in malignant cells suggests that the hRFI products affect carcinogenesis in a variety of organs in the digestive tract. However, unlike in the colorectal specimen, we did find that a few normal gastric glands were stained in speckles. Presently, we cannot explain any biological role of the spotty expression of hRFI seen in these normal gastric glands.
We examined the correlation between the staining pattern of hRFI and several clinicopathological factors, and found that the proportion of blood vessel involvement was significantly higher in carcinomas with diffuse hRFI expression than in carcinomas with focal expressions (P < 0.001). Consistently, the liver metastasis was observed in five cases with the diffuse type, but in only one with the focal type, although the survival did not show a significant difference between the two staining patterns. This is supposed to be caused by the fact that the peritoneal or lymph node recurrences are major factors to determine the prognosis of gastric cancer patients. However, from our data, we can hypothesize that the progression of hRFI expression from focal to diffuse by reasons such as the acquisition of clonality may play a potential role in vessel invasion and metastasis. In this study, we also found that in the middle third of the stomach (M region), the proportion of cases with focal staining was significantly higher than that of cases with diffuse staining (P = 0.017); however, the reason for this is unclear.
Various factors such as the tumor size, Bormann type, differentiation histology as well as serum tumor marker have been correlated with the recurrence of gastric cancer in the liver(7–9). In addition, recent reports have shown that the expressions of certain molecules such as IL-1alpha (10), c-Met (11), c-erbB2 and beta catenin (12) in the primary tumors have a positive relation with liver metastasis. Our data suggest that the detection of the expression pattern of RFI in gastric cancer can be another useful predictor of recurrence in the liver, especially when combined with these factors.
|
Acknowledgments |
|---|
We are grateful to Ms Amitani for her assistance in material preparation. This study was supported in part by a Grant-in-Aid from the Ministry of Education, Science, Sports, Culture and Technology of Japan, in part by a Grant-in-Aid from the Ministry of Health, Labour and Welfare of Japan and in part by the Public Trust of Surgery Research Fund.
|
References |
|---|
|
TOPAbstractINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
1 Sasaki S, Nakamura T, Arakawa H, Mori M, Watanebe T, Nagawa H, et al. Isolation and characterization of a novel gene, hRFI, preferentially expressed in esophageal cancer. Oncogene 2002;21:5024–30.[CrossRef][Web of Science][Medline]
2 Syken J, De-Medina T, Munger K. TID1, a human homolog of the Drosophila tumor suppressor l(2)tid, encodes two mitochondrial modulators of apoptosis with opposing functions. Proc Natl Acad Sci USA 1999;96:8499–504.
3 Birnbaum MJ, Clem RJ, Miller LK. An apoptosis-inhibiting gene from a nuclear polyhedrosis virus encoding a polypeptide with Cys/His sequence motifs. J Virol 1994;68:2521–8.
4 Takahashi R, Deveraux Q, Tamm I, Welsh K, Assa-Munt N, Salvesen GS, et al. A single BIR domain of XIAP sufficient for inhibiting caspases.J Biol Chem 1998;273:7787–90.
5 Sasaki S, Watanabe T, Konishi T, Kitayama J, Nagawa H. Effects of expression of hRFI on adenoma formation and tumor progression in colorectal adenoma-carcinoma sequence. J Exp Clin Cancer Res (in press).
6 Hermanek P, Sobin LH, editors. UICC TNM Classification of Malignant Tumors. 4th edition. Berlin: Springer 1987.
7 Koga S, Takebayashi M, Kaibara N, Nishidoi H, Kimura O, Kawasumi H, et al. Pathological characteristics of gastric cancer that develop hematogenous recurrence, with special reference to the site of recurrence. J Surg Oncol 1987;36:239–42.[Web of Science][Medline]
8 Yoo CH, Noh SH, Shin DW, Choi SH, Min JS. Recurrence following curative resection for gastric carcinoma. Br J Surg 2000;87:236–42.[CrossRef][Web of Science][Medline]
9 Ohno S, Fujii T, Ueda S, Nakamoto T, Kinugasa S, Yoshimura H, et al. Predictive factors and timing for liver recurrence after curative resection of gastric carcinoma. Am J Surg 2003;185:258–63.[CrossRef][Web of Science][Medline]
10 Tomimatsu S, Ichikura T, Mochizuki H. Significant correlation between expression of interleukin-1 alpha and liver metastasis in gastric carcinoma. Cancer 2001;91:1272–6.[CrossRef][Web of Science][Medline]
11 Amemiya H, Kono K, Itakura J, Tang RF, Takahashi A, An FQ, et al. c-Met expression in gastric cancer with liver metastasis. Oncology 2002;63:286–96.[CrossRef][Web of Science][Medline]
12 Ougolkov A, Yamashita K, Bilim V, Takahashi Y, Mai M, Minamoto T. Abnormal expression of E-cadherin, beta-catenin, and c-erbB-2 in advanced gastric cancer: its association with liver metastasis. Int J Colorectal Dis 2003;18:160–6.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Konishi, S. Sasaki, T. Watanabe, J. Kitayama, and H. Nagawa Overexpression of hRFI (human ring finger homologous to inhibitor of apoptosis protein type) inhibits death receptor-mediated apoptosis in colorectal cancer cells Mol. Cancer Ther., May 1, 2005; 4(5): 743 - 750. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||