Surg Today (2011) 41:688–692DOI 10.1007/s00595-010-4320-2

Reprint requests to: H.-W. LeeReceived: August 27, 2009 / Accepted: January 12, 2010

Original Article

Expression of Bone Morphogenic Protein-4 Is Inversely Related to Prevalence of Lymph Node Metastasis in Gastric Adenocarcinoma

SEONG-GON KIM1, HYE-RIM PARK

2, SOO-KEE MIN2, JE-YONG CHOI

3, SUNG-HOON KOH4, JONG-WAN KIM

5, and HAE-WAN LEE

5

Departments of 1 Oral and Maxillofacial Surgery, 2 Pathology, 4 Plastic and Reconstructive Surgery, and 5 Surgery, Hallym University Sacred Heart Hospital, 896 Pyungchon-dong, Dongan-gu, Anyang-si, Kyungki-do, Seoul 431-070, Republic of Korea3 Department of Biochemistry, School of Medicine, Kyungpook National University, Daegu, Republic of Korea

AbstractPurpose. Bone morphogenic proteins (BMPs) are the largest subfamily of the transforming growth factor-β superfamily. Initially characterized as factors that induce bone and cartilage formation, BMPs have been found to be critical during mesoderm formation, organogen-esis, and cellular differentiation. Bone morphogenic proteins are also known to modulate the morphologic alteration, adhesion, motility, and invasion of carcinoma cells derived from several organs. However, BMP-4 expression in gastric adenocarcinoma has not yet been clarifi ed. We conducted the present study to defi ne the clinical signifi cance of BMP-4 expression in gastric carcinoma.Methods. Using immunohistochemistry, we investigated the expression of BMP-4 in normal mucosae and gastric adenocarcinoma samples from 64 patients with gastric carcinoma.Results. The expression of BMP-4 was signifi cantly higher in the adenocarcinoma than in the normal mucosae. Moreover, increased BMP-4 expression was associated with the presence of Helicobacter pylori infection. By contrast, the BMP-4 expression rate in gastric carcinoma was inversely related to the prevalence of lymph node metastasis and tumor invasiveness.Conclusions. The fi ndings of this study suggest that BMP-4 expression may be a useful prognostic factor for predicting the outcome of patients with gastric carci-noma. Continued investigation to defi ne the pathophys-iologic mechanism underlying the role of BMP-4 in gastric carcinoma is warranted.

Key words Bone morphogenic protein 4 · Helicobacter pylori · p53 · Gastric cancer

Introduction

Gastric cancer is one of most common malignancies in the world. Despite improvements in treatment strate-gies for gastric adenocarcinoma, the survival rate remains low. At present, early diagnosis gives the patient the only chance of cure and is the most effective means of reducing the mortality from this disease. On the other hand, excellent prognoses of early gastric cancer have interested surgeons in promoting quality of life. For example, minimally invasive surgery is now preferred to radical treatment for early gastric cancer, to preserve body functions and reduce postoperative complications; whereas for early proximal gastric cancer, limited resec-tion and lymphadenectomy based on sentinel node navigation has been recommended.1 A knowledge of the etiological factors and pathogenetic mechanisms involved in gastric carcinogenesis is essential to the pre-vention of gastric cancer and the development of new treatment modalities.

Gastric carcinogenesis is a multistep process involv-ing the transformation of gastric mucosa over many years. It is proposed that the process represents the interaction of three major factors: the agent (Helico-bacter pylori), the host, and the environment. It is now recognized that H. pylori infection plays a prominent causative role.2 Exposure of gastric epithelial cells to H. pylori results in an infl ammatory reaction with the sub-sequent generation of reactive oxygen species (ROS) and an increased level of nitric oxide (NO) synthesis. These ROS, with NO, promote cell proliferation. Break-down of a balance between cell proliferation and apop-tosis has been reported to be associated with the early phases of carcinogenesis.3–5

Bone morphogenic proteins are the largest subfamily of the transforming growth factor (TGF)-β superfamily. Initially characterized as factors that induced bone and cartilage formation,6 BMPs have been found to be criti-cal during gastrulation, mesoderm formation, skeletal

S.-G. Kim et al.: BMP-4 Expression in Gastric Cancer 689

and limb development, organogenesis, and cellular dif-ferentiation.7 Recent studies suggest that BMPs play a similar role in epithelial differentiation in the adult gas-trointestinal tract and that deregulation of the activity of these morphogens promotes the development of gas-trointestinal cancers.8,9 More specifi cally, BMP-4 is an important factor for epithelial differentiation and gastric gland cell proliferation;10 however, little is known about the potential role of BMPs in gastric carcinoma. Fur-thermore, to our knowledge, no study has evaluated the BMP-4 expression correlated with clinicopathological parameters of gastric carcinoma.

In this study, we analyzed the expression of BMP-4 in gastric carcinoma and background mucosa tissues from 64 patients. We are confi dent that this is the fi rst report to describe the relationship between BMP-4 expression and clinicopathological variables in gastric carcinoma.

Patients and Methods

This was a retrospective study of 64 patients from the tumor registry of patients who underwent surgery for gastric carcinoma at the Department of Surgery of Hallym University Sacred Heart Hospital, between 2004 and 2005. Variable factors; namely, age, gender, pathologic staging of the TNM system, p53 expression, and H. pylori infection, were analyzed. Tumor staging was assessed using the 5th edition of the TNM system according to the Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC).11 The patients were divided into three groups according to the classifi cation of their tumor histology: a well-differentiated group, a moderately differentiated group, and a poorly differentiated group, which included poorly or undifferentiated adenocarcinoma, mucinous, and signet-ring cell carcinoma. Table 2 summarizes the clinicopathological features of each patient. We ana-lyzed BMP-4 expression of carcinoma and surrounding normal gastric mucosa from each patient.

Immunohistochemistry

Tumors and corresponding normal tissues were fi xed in 3.7% formaldehyde and embedded in paraffi n. The paraffi n-embedded tissues were cut into 4-μm thick sec-

tions. Sections designed for single staining were immu-nohistochemically stained by the avidin–biotin-complex method. Paraffi n sections were dewaxed and rehy-drated. The primary antibodies were mouse monoclonal antibody against recombinant human BMP-4 (NCL-BMP-4) (Novocastra Laboratories, Newcastle, UK) and monoclonal antibody against recombinant human p53 (Zymed, San Francisco, CA, USA). Subsequent proce-dures were performed according to the manufacturer’s protocol. The immunohistochemical staining of slides was done simultaneously to reduce technical errors, by NexES IHC system (Ventana, Tucson, AZ, USA). The sections were counterstained with Mayer hematoxylin. The dilution rates were 1:20 for BMP-4 and 1:50 for p53. Prostate adenocarcinoma and oral lichen planus were used as the positive control for BMP-4 and p53 (data not shown), respectively. The negative controls were run with a primary antibody using the same tissue blocks.

Scoring of Immunostaining

For BMP-4 expression, the immunohistochemical reac-tion in the cytoplasm was evaluated using a light micro-scope. All specimens were reviewed independently in at least fi ve areas at 200 × magnifi cation. We calculated the immunohistochemical score by estimating the percent-age of immunoreactive cells: 0, no positive cells or essentially none (<1%); 1, some positive cells (1%–9%); 2, well-defi ned areas of positive cells (10%–50%); and 3, extensive areas of positive cells (51%–100%). The slides were examined by two investigators, who were blinded to the corresponding clinicopathologic data.

Statistical Analysis

The signifi cance of BMP-4 expression correlated with clinicopathological parameters was calculated using the Wilcoxon rank sum test. P values of less than 0.05 were considered signifi cant. All statistical analyses were con-ducted using SAS EG ver. 4 (SAS, Cary, NC, USA).

Results

Table 1 shows the overall immunohistochemical BMP-4 expressions in normal gastric mucosae. Of the 64 samples

Table 1. Reactivity of bone morphogenic protein (BMP)-4 antigens in normal mucosa against the background of gastric cancer

Variables n

BMP-4 reactivity

0 1+ 2+ 3+

Normal mucosa 64 59 (92.2) 5 (7.8) 0 (0) 0 (0)Adenocarcinoma 64 9 (15.0) 17 (26.6) 24 (37.5) 14 (21.9)

690 S.-G. Kim et al.: BMP-4 Expression in Gastric Cancer

of normal mucosa, expression of BMP-4 was detected in only 5 (7.8%), all of which had BMP-4 expression categories (+). On the other hand, of the 64 samples of adenocarcinoma, expression of BMP-4 was detected in 55 (85%). Importantly, the proportion of categories [(++) and (+++)] of BMP-4 expression was higher in the adenocarcinoma group. (Fig. 1).

The relationship between expression of BMP-4 in gastric carcinoma and clinicopathological variables is summarized in Table 2. BMP-4 expression was signifi -cantly associated with depth of tumor invasion, lymph node metastasis, and H. pylori infection. Gastric carci-nomas with BMP-4 expression tended to have a lower prevalence of lymph node metastasis (P = 0.015). There was a tendency for higher expression of BMP-4 in the presence of H. pylori infection. However, the expression of BMP-4 did not correlate with other factors, such as gender, age, histology, and expression of p53.

Discussion

The epithelial layers of the gastrointestinal tract main-tain their microarchitecture with remarkable accuracy despite years of recurrent damage. To maintain their integrity, the epithelia of the gastrointestinal tract are in a state of constant renewal. Nonetheless, metaplasia and transdifferentiation, i.e., change in tissue structure without cell dysplasia, do occur, suggesting the possibil-ity of loss of control over the microarchitecture. Renewal of adult epithelium, termed “morphostasis,” has several interesting similarities, with patterning events during embryonic development or morphogenesis.12 Molecular controls of morphogenesis are relatively well character-ized, including the contribution of BMPs, which have been shown to control proliferation, differentiation, and apoptosis in the homeostatic system of gastrointestinal epithelial renewal.8,9,13

A B

C

E

D

Fig. 1A–D. Immunohistochemical analysis of bone morphogenic protein (BMP)-4 in gastric cancer and adjacent normal tissue. A Staining for BMP-4 was weakly positive in normal gastric tissue (×200). B Staining for BMP-4 was negative in signet-ring cell type adenocarcinoma and weakly posi-tive in normal glands (×200). C Staining for BMP-4 was weakly positive in poorly differentiated adenocarcinoma with positive infl ammatory cells (×200). D Stain-ing for BMP-4 was intermediately positive in moderately differenti-ated adenocarcinoma (×200). E Staining for BMP-4 was strongly positive in well-differentiated ade-nocarcinoma (×200)

S.-G. Kim et al.: BMP-4 Expression in Gastric Cancer 691

This study supports existing evidence that BMP-4 expression is increased in adenocarcinoma. Bleuming et al. reported on the expression of BMP-2 in histologi-cally normal gastric mucosa.14 In contrast to BMP-2 expression, BMP-4 expression was not detected in the histologically normal antrum of the human stomach. In our study, BMP-4 expression was not observed in most normal mucosae (Table 1). These fi ndings are compati-ble with those of another report of BMP-4 expression in normal gastric mucosa. However, the aberrations in BMP signaling may play an important role in the patho-genesis of human cancer. With respect to tumor cell proliferation, BMPs either stimulate or inhibit cell growth, depending on the target cell type. BMP-4 has been shown to positively modulate the morphologic alteration, adhesion, motility, and invasion of carcinoma cells derived from organs such as the pancreas,15 ovary,16 liver,17 and colon.18 On the other hand, it inhibits prolif-eration and induces apoptosis of multiple myeloma cells.19 It also induces apoptosis, and inhibits invasion

and metastatic potential in lung cancer and breast cancer.20,21

Katoh and Terada examined the expression of BMP-4 mRNA in several gastric cell lines.22 BMP-4 mRNA is preferentially overexpressed in the poorly differenti-ated type of gastric cancer. In this study, no signifi cant association was found between the histologic type and immunoreactivity for BMP-4 in gastric cancer tissue. Sometimes cancer cell lines did not refl ect the condition of cancer tissues completely.

Few studies have investigated the association between BMP-4 expression and the clinicopathological features of gastric carcinoma. To our knowledge, this is the fi rst study to demonstrate a clear correlation between the levels of BMP-4 expression and depth of invasion and lymph node metastasis. Depth of invasion and lymph node metastasis are the most powerful prognostic factors of gastric cancer; lymph node metastasis also being the most important prognostic factor of early gastric cancer.23,24

Table 2. Reactivity of bone morphogenic protein (BMP)-4 antigens in gastric cancer

Variables n

BMP-4 reactivity

P value0 1+ 2+ 3+

Sex 0.195 Male 46 7 (15.2) 10 (21.7) 16 (34.8) 13 (28.3) Female 18 2 (11.1) 7 (38.9) 8 (44.4) 1 (5.6)Age (years) 0.448 <60 30 4 (13.3) 7 (23.3) 11 (36.7) 8 (26.7) ≥60 34 5 (14.7) 10 (29.4) 13 (38.2) 6 (17.7)EGC/AGC 0.009 EGC 31 2 (6.5) 6 (19.4) 13 (41.9) 10 (32.3) AGC 33 7 (21.2) 11 (33.3) 11 (33.3) 4 (12.1)Tumor depth 0.008 T1 & T2 46 5 (10.9) 9 (19.6) 19 (41.3) 13 (28.3) T3 & T4 18 4 (22.2) 8 (44.4) 5 (27.8) 1 (5.6)Lymph node metastasis 0.015 Absent 37 4 (10.8) 7 (18.9) 14 (37.8) 12 (32.4) Present 27 5 (18.5) 10 (37.0) 10 (37.0) 2 (7.4)pTNM stage 0.107 I & II 51 7 (13.7) 11 (21.6) 20 (39.2) 13 (25.5) III & IV 13 2 (15.4) 6 (46.2) 4 (30.8) 1 (7.7)Mucosa/Submucosa 0.441 Mucosa 21 1 (4.8) 5 (23.8) 10 (47.6) 5 (23.8) Submucosa 10 1 (10.0) 2 (20.0) 3 (30.0) 4 (40.0)Histology, differentiated: 0.076 Well 18 2 (11.1) 6 (33.3) 6 (33.3) 4 (22.2) Moderately 19 2 (10.5) 3 (15.8) 6 (31.6) 8 (42.1) Poorly 27 5 (18.5) 8 (29.6) 12 (44.4) 2 (7.4)p53 0.894 Negative 35 5 (14.3) 9 (25.7) 13 (37.1) 8 (22.8) Positive 29 4 (13.8) 8 (27.6) 11 (37.9) 6 (20.7)Helicobacter pylori 0.015 Negative 26 7 (26.9) 11 (42.3) 6 (23.1) 2 (9.7) Positive 9 1 (11.1) 0 (0) 6 (66.7) 2 (22.2)

EGC, early gastric cancer (tumor invasion confi ned to the mucosa and submucosa, irrespective of the presence of a lymph node metastasis); AGC, advanced gastric cancer

692 S.-G. Kim et al.: BMP-4 Expression in Gastric Cancer

Infl ammation of the gastric mucosa changes the dynamics of epithelial cell turnover. As infl ammatory cells produce the morphogens that regulate epithelial cell fate,25,26 the infl ux of infl ammatory cells may disturb the normal expression gradient of morphogens in gastric epithelia. H. pylori infection causes chronic infl amma-tion of the gastric mucosa with an infl ammatory infi l-trate. This infl ammatory process may be associated with changes in epithelial turnover and the development of intestinal metaplasia and dysplasia.27 In the present study, we noted a tendency for higher expression of BMP-4 in the presence of H. pylori infection (P = 0.015) and observed focal infi ltrates of BMP-4 expressing cells in H. pylori-infected patients.14 The infl ux of morpho-gen-producing infl ammatory cells induced by H. pylori infection would be an alternative way to disturb mor-phogen signaling in the gastric epithelium. Such deregu-lation may affect the tumor cell biology.

In conclusion, our data demonstrate that the expres-sion of BMP-4 was relatively higher in adenocarcino-mas than in normal gastric mucosa and was signifi cantly associated with depth of tumor invasion and lymph node metastasis. Increased BMP-4 expression in gastric adenocarcinoma is inversely related to lymph node metastasis, and should thus continue to be regarded as a favorable indicator. Further investigations of the role of BMP-4 in gastric carcinoma are warranted.

Acknowledgment. This research was supported by a grant (No. 01-2002-04) from the Hallym University Medical Cancer Research Fund.

References

1. Ohdaira H, Nimura H, Takahashi N, Mitsumori N, Kashiwagi H, Narimiya N, et al. The possibility of performing a limited resection and a lymphadenectomy for proximal gastric carcinoma based on sentinel node navigation. Surg Today 2009;39:1026–31.

2. Parsonnet J, Friedmann GD, Vandersteen DP, Chang Y, Vogelman JH, Orentreich N, et al. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991;325:1127–31.

3. Sepulveda AR. Molecular testing of Helicobacter pylori-associated chronic gastritis and premalignant gastric lesions: clini-cal implications. J Clin Gastroenterol 2001;32:377–82.

4. Moss SF. Cellular markers in the gastric precancerous process. Aliment Pharmacol Ther 1998;12:91–109.

5. Li CQ, Pignatelli B, Ohshima H. Increased oxidative and nitrative stress in human stomach associated with cagA+ Helicobacter pylori infection and infl ammation. Dig Dis Sci 2001;46:836–44.

6. Wozney JM, Rosen V, Celeste AJ, Mitsock LM, Whitters MJ, Kriz RW, et al. Novel regulators of bone formation: molecular clones and activities. Science 1988;242:1528–34.

7. Hogan BL. Bone morphogenetic proteins: multifunctional regula-tors of vertebrate development. Genes Dev 1996;10:1580–94.

8. van den Brink GR. Linking pathways in colorectal cancer. Nat Genet 2004;36:1038–9.

9. Bienz M, Clevers H. Linking colorectal cancer to Wnt signaling. Cell 2000;103:311–20.

10. van den Brink GR, Hardwick JC, Tytgat GN, Brink MA, Ten Kate FJ, Van Deventer SI, et al. Sonic hedgehog regulates gastric gland morphogenesis in man and mouse. Gastroenterology 2001;121:317–28.

11. Sobin LH, Wittekind C. TNM classifi cation of malignant tumors, 5th ed. New York: Wiley-Liss; 1997.

12. Potter JD. Morphostasis: a missing concept in cancer biology. Cancer Epidemiol Biomarkers Prev 2001;10:161–70.

13. de Santa Barbara P, van den Brink GR, Roberts DJ. Development and differentiation of the intestinal epithelium. Cell Mol Life Sci 2003;60:1322–32.

14. Bleuming SA, Kodach LL, Garcia Leon MJ, Richel DJ, Peppelen-bosch MP, Hardwick JC, et al. Altered bone morphogenetic protein signaling in the Helicobacter pylori-infected stomach. J Pathol 2006;209:190–7.

15. Hamada S, Satoh K, Hirota M, Kimura K, Kanno A, Masamune A, et al. Bone morphogenetic protein 4 induces epithelial-mesen-chymal transition through MSX2 induction on pancreatic cancer cell line. J Cell Physiol 2007;213:768–74.

16. Theriault BL, Shepherd TG, Mujoomdar ML, Nachtigal MW. BMP4 induces EMT and Rho GTPase activation in human ovarian cancer cells. Carcinogenesis 2007;28:1153–62.

17. Maegdefrau U, Amann T, Winklmeier A, Braig S, Schubert T, Weiss TS, et al. Bone morphogenetic protein 4 is induced in hepa-tocellular carcinoma by hypoxia and promotes tumour progres-sion. J Pathol 2009;218:520–9.

18. Deng H, Makizumi R, Ravikumar TS, Dong H, Yang W, Yang WL. Bone morphogenetic protein-4 is overexpressed in colonic adeno-carcinomas and promotes migration and invasion of HCT116 cells. Exp Cell Res 2007;313:1033–44.

19. Hjertner O, Hjorth-Hansen H, Borset M, Siedel C, Waage A, Sundan A. Bone morphogenetic protein-4 inhibits proliferation and induces apoptosis of multiple myeloma cells. Blood 2001;267:182–8.

20. Buckley S, Shi W, Driscoll B, Ferrario A, Anderson K, Warburton D. BMP4 signaling induces senescence and modulates the onco-genic phenotype of A549 lung adenocarcinoma cells. Am J Physiol Lung Cell Mol Physiol 2004;286:L81–6.

21. Shon SK, Kim A, Kim JY, Kim KI, Yang Y, Lim JS. Bone morpho-genetic protein-4 induced by NDRG2 expression inhibits MMP-9 activity in breast cancer cells. Biochem Biophys Res Commun 2009;385:198–203.

22. Katoh M, Terada M. Overexpression of bone morphogenic protein (BMP)-4 mRNA in gastric cancer cell lines of poorly differenti-ated type. J Gastroenterol 1996;31:137–9.

23. Kunisaki C, Shimada H, Takahashi M, Ookubo K, Moriwaki Y, Akiyama H, et al. Prognostic factors in early gastric cancer. Hepatogastroenterology 2001;48:294–8.

24. Kim JP, Kim YW, Yang HK, Noh DY. Signifi cant prognostic factors by multivariate analysis of 3926 gastric cancer patients. World J Surg 1994;18:872–7.

25. Varas A, Harger-Theodorides AL, Sacedon R, Vicente A, Zapata AG, Crompton T. The role of morphogens in T-cell development. Trends Immunol 2003;24:197–206.

26. Lowrey JA, Stewart GA, Lindey S, Hoyne GF, Dallman MJ, Howie SE, et al. Sonic hedgehog promotes cell cycle progression in activated peripheral CD4(+) T lymphocytes. J Immunol 2002;169:1869–75.

27. Peek RM Jr, Blaser MJ. Helicobacter pylori and gastrointestinal tract adenocarcinomas. Nature Rev Cancer 2002;2:28–37.